State of New York— Department of Agriculture. Twentieth Annual Report OF THE BOARD OK CONXROIv OF THE LIBRARY NEW YORK NEW YORK BOTANICA. QARDEN Agricultural Experiment Station (GENEVA, ONTARIO COUNTY). KoR the: Ye^ar 1901. With Reports of Director and Other Officers. TRANSMITTED TO THE LEGISLATURE JANUARY 13, 1902. ALBANY J. B. LYON COMPANY, STATE PRINTERS 1902 //5 Hoi State of New York. No. 59. IN ASSEMBLY, January 13, 1902. Twentieth Annual Report OF THE Board of Control of the iNew York Agricultural Experiment Station. STATE OF NEW YORK: Departmbsnt of Agriculttirb, Albany, January 13, 1902. To the Assemlhj of the State of New Torh: ■ I have tbe honor to herewith submit the Twentieth Annual Report of the Director and Board of Managers of the New York Agricultural Experiment Station at Geneva, N. Y., in pursuance of the provisions of the Agricultural Law. I am, respectfully yours, CHARLES A. WIETING, (Jommissioner of A(;noulture. 1901. ORGANIZATION OF THE STATION. BOARD OF CONTROL. Governor Benjamin B. Odell, Jr., Albany. Stephen H. Hammond, Geneva. Austin C. Chase, S.yracuse. Frank O. Chamberlain, Canandaigua. Frederick C. Schraub, Lowville. Nicholas Hallock, Qneens. Lyman P. Haviland, Caindeu. Edgar G. Dusenbury, Portville. Oscar H. Hale, North Stockholm. Martin L. Allen, Fayette. OFFICERS OF THE BOARD. Stephen H. Hammond, President. William O'Hanlon, Secretary and Treasurer. EXECUTIVE committee. Stephen H. Hammond, Frederick C. Schraub, Martin L. Allen, Lyman P. Haviland, Frank 0. Chamberlain, Nicholas Hallock. STATION STAFF. Whitman H. Jordan, Sc. D., Director. George W. Churchill, Agriculturist and Superin- tendent of Labor. William P. Wheeler, First Assistant (Animal Industry). Fred C. Stewart, M. S., Bofn)rist. Harry J. Eustace, B. S., Student Assistant in Botany. Lucius L. VanSlyke, Ph.D., Chemist. Christian G. Jenter, Ph.C, •William H. Andrews, B.S., IT J. Arthur LeClerc, B.S., Frederick D. Fuller, B.S., Edwin B. Hart, B.S., * Charles W. Mudge, B.S., Andrew J. Patten, B.S., Assistant Chemists. Harry A. Harding, M.S., Dairy Bacteriologist. Lore A. Rogers, B.S, Assistant Bacteriologist. George A. Smith, Dairy Expert. Frank H. Hall, B.S., Editor and Libiariau. Victor H. Lowe, M.S., tF. Atwood Sirrine, M.S., Entomologist !t. Percival J. Parrott, A.M., Assistant Entomologist. Spencer A. Beach, M.S., Horticulturist. Nathaniel O. Booth, B.Agr., Assistant Horticulturist. Orrin M. Taylor, Foreman in Horticuliuie Frank E. Newton, Jennie Terwilliger, Clerks and Stenographers. Adin H. Horton, Comjyuter. Address all correspondence, not to individual members of the staff, but to the New York Agricultural Experiment Station, Geneva, N. Y. Tli3 Bulletins puhli.shed by the Station will be sent free to any fiirmer applyinji; for them. •Connected with Fertilizer Control. t At Second Judicial Department Branch Station, Jamaica, N. Y, ^ Absent on leave. TABLE OF CONTENTS. UBRAR\ NEW YORK BOTANICAL GARDEN PAGE Treasurer's report 1 Director's report 9 Report of Department of Animal Husbandry: Tlie food source of mills fat with studies on the nutrition of milch cows 29 The immediate effect on milk pix)duction of changes in the ration. 61 Report of the Department of Botany: An epidemic of currant authracnose 123 Notes from the Botanical Department 142 Report of the Chemical Department: A study of enzymes in cheese 165 Conditions affecting weight lost by ciieese in curing 194 Report on Crop Production: Influence of manure upon sugar boots 223 Commercial fertilizers for onions 236 Report of the Department of Entomology: San Jose scale investigations, III 247 Treatment for San Jose scale in orchards: I. Orchard fumigation. 292 Report of the Horticultural Department: Stable manure and nitrogenous commercial fertilizers fnr forcing lettuce 321 Ginseng culture 356 Report of Inspection Work: Inspection of feeding stuffs 301 So-called " Red Albumen " a fraud 390 Report of analyses of commercial fertilizers for the spring and fall of 1901 391 Report of analyses of paris green and other insecticides in 1901. . 3n6 List of periodicals received by the Station 403 Meteorological records 410 Index 423 o cc Q_ < TWENTIETH ANNUAL REPORT OF THE Board of Control of the New York Agricultural Experiment Station. TREASURER'S REPORT. Geneva, N. Y., Octoher 1, 1901. To the Board of Control of the New York Agricultural Experiment Station: As Treasurer of the Board of Control, I respectfully submit the following report for the fiscal year ending September 30, 1901: GsN'ERAii Expense, Appropriation 1900-1901. Receipts. 1900. Oct. 1. To balance on hand $2,742 20 To amount received from Comptroller |18,750 00 Less balance due appro- priation 1899-1900 3,750 00 — 15,000 00 ,742 26 c:: 2 Report of the Treasurer of the ]^9(]Q Expenditures. Oct. 1, By building and repairs $780 57 By chemical supplies 230 34 By contingent expenses 804 45 By feeding stuffs 1,817 86 By fertilizers 84 20 By freight and express 478 G5 By furniture and fixtures 1,094 63 By heat, light and water 2,033 45 By library 068 50 By live stock 8 75 By postage and stationery 884 91 By publications 1,954 90 By scientific apparatus 151 50 By seeds, plants and sundry supplies. . . 1,073 15 By tools, implements and machinery... 406 13 l^y traveling expenses 958 61 114,030 69 By balance, October 1, 1901 3,711 57 $17,742 26 I Salaries, AppRorRiATioN 1900-1901. Receipts. 1900. Oct. 1. To balance |5,G48 06 To amount received from Comptroller $28,750 00 Less balance due appro- priation 1899-1900 5,750 00 ■ 23,000 00 $28,648 66 Hew York Agricultural Experimeint Station. 3 Expenditures. 1901. Oct. 1. By salaries $21,860 07 Bj balance 6,788 59 128,648 66 / Labor. Receipts. 1900. Oct. i. To balance |2,975 71 To amount received from Comptroller |15,000 GO Less balance due appro- priation 1899-1900 3,000 00 — 12,000 00 114,975 71 Expenditures. 1901. Oct. 1. By labor |11,728 90 By balance 3,246 81 $14,975 71 ExTENSB OP Bulletins and Enforcing Provisions of Chapter 955, Laws of 1S9G.— Appropriation 1898-1899. Receipts. 1900. Oct. 1. To balance on baud $6 20 ExpcndUurcs. By postage and stationery $6 20 4 Kbport op the Trbasurer of xnR COMMBRCTAL FERTILIZERS, AprROPRIATION 1900-1901. 1900. Receipts. Oct. 1. To balance $549 48 To amount received from Comptroller $12,000' 00 Le-ss balance due appro- priation 1899-1900 2,000 00 10,000 00 110,549 48 EiDpenditures. Oct. 1. By chemical supplies $507 2.3 By contingent expenses 4 07 By freight and express 48 40 By heat, light and water 530 95 By postage and stationery 4 48 By publications 1,337 52 By salaries 4,992 3() By seeds, plants and sundry supplies. . . 138 01 By tools, implements and machinery. ... 90 By traveling expenses 778 40 By balance 2,207 IG 110,549 48 Concentrated Feeiding Stuffs Inspection, Appropriation 1900-1901. 1900. Receipts. Oct. 1. To balance on hand $329 44 To amount received from Comptroller |3,000 00 Less balance due appro- priation 1899-1900 .500 00 2,"»00 00 $2,829 44 New York Agiucultural Experiment Station. j^QQQ Expenditures. Oct. 1. By contingent expenses By freight and express By postage and stationery By publications By salaries By seeds, plants and sundry supplies. By traveling expenses Balance October 1, 1901. $1 38 10 20 123 70 6G4 76 1,427 58 12 71 318 89 $2,559 22 270 22 $2,829 44 Second Judicial Department, Appropriation 1900-1901. Receipts L900-1901. To amount rece'ived from Comptroller.. $8,355 10 Expenditures, By cliemical supplies $71 80 By contingent expenses 238 50 By fertilizers 12 70 * By freight and express 28 78 By furniture and fixtures 18 42 By beat, light and water 73 27 By labor 702 96 By library 52 10 By postage and stationery 37 87 By publications 1,782 47 By salaries 3,753 96 By scientific apparatus 85 By seeds, plants and sundry supplies. .. 214 50 By tools, implements and machinery. . . . 165 38 By traveling expenses 735 63 By rents 465 91 $8,355 10 6 Kei'oht of the Tkeasuuek of tub Paris Green Law, ArpRorniATioN 1898-18U9. Receipts. 1900-1901. To amount received from Comptroller.. $2.50 SO Expenditures. By publications $250 80 SrECAL Appropriation, Building and Repairs, Appropriation 1898-1899. Receipts. 1900-1901. To amount received from Comptroller.. |117 09 Expenditures. By building and repairs |117 G9 Fertilizer License, Chapter 955, Laws 1896, Amendeid by Chapter G87, Laws 1899. Receipts. 1900-1901. To amount received for fertilizer license. |10,8S0 00 Expenditures. By amount remitted to Treasurer State of New York .^10,880 00 Feeding Stuffs License, Chapteir 338, Laws 1893, Amended by Chapter 510, Laws 1899. Receipts. 1900-1901. To amount received for fccdinj; slnfTs license |3,150 00 Expenditures. By amount remitted to Treasurer State of New York .|!8,150 00 New York Agricultural ExrERiMENT StatioiSI. I Special Api'ropriation 19U0-1U01, Director's Housej. Beceipts. 1900-1901. To amount received from Comptroller. . |6,981 76 Ewpenditures. By construction Director's house |6,981 76 The United States Appropriation, 1900-1901. Dr. To receipts from the Treasurer of the United States as per appropriation for fiscal 3 ear ending June 30, 1901, as per act of Congress approved March 2, 1887 11,500 00 Cr. By publications |803 15 By postage and stationery 50 By heat, light, water and power 21 89 By chemical supplies 33 50 By seeds, plants and sundry supplies. . . 147 17 By feeding stuffs 82 09 By library 62 09 Bv furniture and fixtures 75 00 By scientific apparatus 69 44 By traveling expenses 126 04 By contingent expenses 79 13 ,•) 00 OU All expenditures are supported by vouchers approved by thiC Auditing Committee of the Board of Control and have been furnished the Comptroller of the State of New York. William O'Hanlon, Trrdsiifer. DIRECTOR'S REPORT FOR 1901.' To the Honorable Board of Control of the New York Agrioultural Ea:periment Htation: Gentlemen. — I have the honor to present herewith a report for the year 1901 of the institution under your charge. As in former years, this report, outside of the matter dealing with the various lines of inspection, is made up chiefly of the results of investigations and experiments of a scientific or semi-scientiflc character. In other words, it is mainly a presentation of the outcome of efforts to study problems or conditions important to the practice of agriculture and is not intended, for the most part, to convey information of a common or general character. This is in accordance with the well established policy of holding the Station to the work of investigation rather than of instruc- tion, a policy entirely harmonious with fundamental con- ceptions and the legal provisions applying to this institution. The contents of this report make it very evident also that, excepting the inspection work, the members of the Station staff are dealing largely with problems particularly alTecting the dairy and horticultural interests, a condition of things quite consistent with the status and demajids of the agricultural industries of New York. Dairying is predominant in the stock husbandry of the State and the commanding importance of our gardening and fruit interests cannot be denied by any one familiar with the facts. Moreover, in dairying and fruit grow- ing there come to the front certain questions of a chemical, botanical, bacteriological or entomological character, so specific an;l so w(dl defined, that they offer promising and useful oppor- *A icDrint of Bulletin No. 211. 10 DiRECTou's Ebport of the luinties for rosoarcli. In addUiun to the above considerations, the dairymen and fruit growers are well organized for discus- sion and for the insistent presentation of their needs and so are likely to receive their full share of attention at the hands of this or any other t^tate institution which is concerned with their interests. STATION STAFF. Several changes have occurred in the Station staff during the past year, Heinrich Hasselbring, B.S.A., Assistant in Horti- culture, was called, at an increased salary, to the position of Assistant Botanist in the agricultural dejiartment of the Uni- versity of Illinois. His place has been filled by the election of Nathaniel O. Booth, B. Agr., who previously occupied a similar position in the University of Missouri. Mr. Booth is a graduate from the University of Missouri in the course in agriculture, and before coming to New York had shown himself capable of successful work in experimental horticulture. Am'asa D. Cook, Ph.C, after serving the Station for more than eight years as Assistant Chemist, resigned his place at the end of his year's leave of absence in order to continue his studies at Cornell University. Edwin B. Hart, B.S., riPturned from Europe in August after a year's study with Professor A. Kossel, Marbourg, and at Heidelberg, Germany, w^here he devoted his attention chiefly to the chemistry of the proteids. Harry J. Eustace, B.S., a graduate from the Michigan Agri- cultural College, was selected as student assistant in botany and will spend the larger part of 1902 at the Station, devoting some weeks to special studies at Cornell Uuniversity. It was decided bj' vote of your board to abolish the position of Second Assistant Horticulturist and create a new position to be known as Foreman in Horticulture. After competitive examination Orrin ^M. Taylor was selected for that position, and lias entered upon his duties in immediate snpei'vision of the lU'actical execution of cxiuriinent details in the orchards, gardens and forcing liousrs. New York Agricultural Experiment Station. 11 J. Arthur LeOlerc, U.S., was granted one year's leave of absence for further study, to take etfect September 1, 1001. Mr. LeClerc is now in Eurojie. BUILDINGS AND EQUIPMENT. The completion of a house for the Director of the Station marks another step in the progress of the institution. It is gratifying to note that the legislature of 1901 appropriated $8,500 for the repairs of the original Station building so long jointly occupied by the Director's family and part of the busi- ness offices. It is expected that before another year elapses all the administrative work of the Station will be located in this building in such a way as to greatly increase convenience and efficiency. the mailing list. The mailing list has reached the highest point since the estab- lishment of the Station. Its growth is steady, and because its enlargement is not forced by any special effort, it measures in a general way the rate of development of the influence of the Station. Popular Bulletin List. Residents of New York 34, 100 Residents of other States 1, 150 Newspapers 767 Experinaent stations and their staffs 785 Miscellaneous 131 Total 3G, 933 CojiPLETE Bulletin List. Experiment stations and their staffs 785 Ijibraries, scientists, etc 2G1 Forei.iiii list 115 Individuals 1,390 Miscellaneous 131 Total 2, (182 WORK IN THE SECOND JUDICIAL DEPARTMENT. In 1801 special work was instituted in the Second Judicial Department. This effort was doubtless brought about by the conditions prevailing in the immediate vicinity of New York 12 Director's RioroRT of the City wIk'1'0 louy,' (established and intensive agriculture had come to have serious problems relating to fungoid and insect pests. It was thought best by those administering the afi'airs of the Station at that time to establish a branch office at Jamaica, L. I., as a center from which to work. This was probably a wise arrangement under the conditions then prevailing. Since that time the Station has become more fully organized into well defined departments and it is now clearly good policy to so rearrange the administration of our outside experimental work as to bring the responsibility and details directly to the several dei)artments of the Station. Moreover, there appears to be no good reason for the extra expense attending a branch office because of duplication of men and equipment. Acting in accord- ance with these views your Board voted to discontinue the branch office at Jamaica after June 30, 1902. It is definitely understood that this action is in no way to affect the character or extent of the experiments conducted in Eastern New York unless it have the effect of enlargment and greater efficiency, and any assertions to the contrary by the uninformed should be discredited. INSPECTION WORK. The inspection of fertilizers, feeding stuffs, Babcock glassware and insecticides has come to absorb a generous share of the energy of the Station staff. The data collected for 1901 is briefly summarized in what follows: Inspection of fertllkcrs. — During the year 1901, there were col- lected for analysis 9G3 samples of commercial fertilizers, repre- senting 456 different brands; of these 324 brands were complete fertilizers. The average amounts of plant-food constituents found and guaranteed are as follows: Giiarantoerl Found Nitrogen. 7V)- <■« Available. phosphorlo acid. Per vt. Potash. I\r ct. 1.89 7.67 4.13 2.01 8.80 4.47 In six cases, the nitrogen and phosphoric acid were more than O.n jtcr ct. below guarantee; in 10 cases, the potash was more than 0.5 per ct. below guarantee. New York Agricultural Experiment Station. 13 The retail selling? price of complete fertilizers averages |25.71 a ton, while the retail cost of the separate ingredients, unmixed, averages |19.81, or |5.90 a ton less than the selling price. The average cost of one pound of plant-food in mixed fertilizers to consumers is as follows: nitrogen, 20.8 cents; available phos- phoric acid, 6.3 cents; potash, 5.9 cents. In 1901, 82 manufacturers paid license fees on 550 different brands of fertilizers. The requirement of a license fee has reduced the number of brands offered for sale from 2,268 to 550. Tlic inspection of commercial feeding- stuffs. — The outcome of the inspection of feeding stuffs is given in Bulletin No. 198. It is shown that 92 manufacturers complied with the law by registering the guaranteed composition of 126 brands, and pay- ing the required license therefor. * Sixty-six of these brands were standard feeding stuffs having more or less fixed or definite char- acteristics, while 60 were feeds compounded from various manu- facturing offals, the majority of which contained some inferior ingredient. The analyses of 297 samples taken by representatives of the Station are reported, representing 98 brands inspected in the fall of 1,900 and 101 brands found in the winter of 1901. The unmixed or standard goods were found to be of fairly uniform quality and practically as good as the guarantees, except in a single instance. The discrepancies occurred with the mixed goods, many of which contained oat hulls, as shown by the percentage of crude fibre present. Adulteration of corn meal and other grain products appears to be practiced. On the whole, it can be said with good reason that the compounding of feeds and the use of inferior materials for adulteration is a serious menace to the prosperity of the Ktock keeper if he continues to buy cuttle foods freely. These mixtures are inferior in quality in most instances and are sold at prices relatively too high. Inspection of Paris green and otlicr insocticides. — In forty sam- ples of Paris green examined, the amount of arsenious oxide varied from 56.13 to 62.87 per ct., with an average of 58.1U per 14 Director's Kei'ort of the ft.; the water-soluble arseiiious oxide, from 0.S8 to 2.04 per ct., with an averaj^e of 1.28 per ct.; the copper oxide, from 26.53 to 31.14 per ct., with an average of 29.88 per ct., and the arsenious oxide in combination wuth copper, from 4J).70 to 57.72 per ct., with an average of 55.98 per ct. These results indicate that the I'aris green in the market during 1901 was of good quality in every respect. Inspection of Bahcock glassware. — In 1901 the Station tested glassware for seventy-seven cheese factories and creameries, including 3,473 milk test bottles, 56 cream bottles and 97 pipettes. Of these 119 were found incorrect and rejected. The Station is not required to inspect cheese factories and creameries to determine whether they are complying with the law as to Babcock glassware. • The responsibility in respect to this compliance rests entirely with those having the manage- ment of the factories and creameries. ANIMAL husbandry. The food sovrcc of milk fat. — The results reported in Bulletin No. 107, relating to the food source of milk fats, were in con- tinuation of the investigations discussed in Bulletin No. 132. The conclusion reached in the former experiment that part, at least, of the milk fat comes from the carbohydrates is con- firmed; and other facts relating to metabolism and the utiliza- tion of food by milch cows are brought out. Three cows were used: Cow 12 fed a fat-poor ration in which the protein sui)ply was gradually decreased from 2.6 lbs. daily to 1.6 lbs. and then gradually restored to the maximum, with accompanying increase and decrease in carbohydrates so that the digestible dry matter of the ration was kept fairly uniform; (\)w 10 fed a ration with normal supply of fat at first which was gradually increased to 1.4 lbs. daily, then gradually restored to the uiu'mal; Cow 2 fed the herd ration having a nutritive ratio about 1:5.6. These rations w^ere quite varied in character and contained some fat-extracted foods; yet they shoAved a quite uni- form digestibility of about 70 per ct. of (he dry matter. It is JS'ew York Aguicultukal Experiment Station. 15 believed that this figure represents fairly the digestibility of rations made up in part of silage and containing a fair propor- tion of high-class grains. A widening of the nutritive ratios ajtpeared to render rations less digestible, especially the protein. The marked changes in protein content and in fat content of rations did not produce noticeable changes in the character or composition of the milk. In the former test, during 59 days, 18.4 lbs. of fat was formed in the milk which could not have had its source in food fat or food protein and could hardly have been drawn from the cow's body fat as she increased in weight 33 lbs; in the same time. In the second test Cow 12 in 74 days produced 39 lbs. of fat similarly unaccounted for, with a body gain of 15 lbs., and Cow 2 in 4 days, 1^ lbs. These amounts of fat must have come from the carbohydrates in the food. A lessening of protein supply in the food did not produce a corresponding decrease of protein in the milk solids, but caused --a marked lessening of protein decomposition in the body. Calo- rimeter determinations show that the heat value of urine bears no constant relation to its nitrogen content, and also prove that the formula used in computing heat energy of urine, Nx5.343 Cal., is greatly in error, actual results being from 3 to 4 times as large as calculated by this formula. The energy value of nutrients as given by Rubner — protein and carbohydrates each 4.1 Cal. and fats 9.3 Cal.^ — appear to be fully high enough for herbivora, even when the loss due to escape of unoxidized gases, methane chielly, is not considered. Over 40 per ct. of the available energy value of the rations was used for maintenance, over 30 per ct. reappeared in the milk solids, leaving a balance of from one-fifth to one-fourth of the ration. The logical conclusion is that this balance, in jjart at least, sustains the work of milk secretion. 77/c immediate effect on milk foiv of chauf/cs in the composition of the ration. — A large number (nearly 1,000) of the individual records from a daily herd have been averaged according to dilferent relations in the constituents of the food to show the 16 DiKECTou's Rei-okt of tub goncral tendoncj of certain cliangos to affoct tlio milk flow, Observations were made in this case only in regard to tlie im- mediate effect of tliese changes. Only rations which approximated those of the common feed- ing standards were considered. Within these limits changes in the amount of total digestible organic matter showed a greater and more constant influence than any other. An increase^ in amount of the total nutrients had a generally favorable effect on the milk yield, and a reduction an unfavorable one, either when the amount was more or less than the 15.5 lbs. jier day for each 1,000 lbs. live weight. Changes in the fuel value of the ration showed effects cor- responding to those in amount of total nutrients both above and below the value of 30,000 Calories. Changes in the protein content of the ration within the ordi- nary limits showed less effect than changes in the amount of nutrients. In general an increase in the amount of protein up to 2.5 lbs. per day for each 1,000 lbs. live weight affected the milk flow favorably. Above that amount, for ordinary cows, a reduction had a favorable effect. The effects of changes in the nutritive ratio corresponded in a general way to those following changes in the protein content. DBPARTMEXT OF BACTERIOLOGY. Much of the work performed in the department was a union of effort with the chemical department in studying the factors which are operative in the curing of cheese and so far as reported this is summarized in what is presented from the chemical department. Study has also been given to certain cheese troubles, a report of which will be made after the accu- mulation of further data. DEPARTMENT OP BOTANY. Currant anfhracnose. — In the Hudson Valley there has been an epidemic of currant anthracnose, a fungous disease which causes the leaves to fall prematurely. Much damage was done. New Yokk Agkicultukal Experiment Station. 17 In some eases the yield of fruit was reduced one-half. This unusual outbreak furnished an excellent opportunity for the study of the disease. It has been discovered that the fungus attacks not only the foliage, but also the fruit, fruit stems and canes; that some varieties, notably Wilder and Prince Albert, are very resistant to the disea^se; and that plants in high situa- tions on dry soil are more affected than those growing in low situations on moist soil. There is no cause for alarm. It is improbable that the disease will continue to be destructive, but in case it should do so it can probably be controlled by spraying with Bordeaux mixture. Trouble icith pears in a nursery cellar. — The Station Botanist has investigated a case in which pear trees stored in a nursery cellar were severely injured by being thawed too quickly. The sand around the roots of the trees had become frozen and to facilitate the removal of the trees a small wood fire was built to thaw the sand. The tops of 25,000 trees were blackened and killed. Had the trees been thawed very gradually it is prob- able that no injury would have resulted. Cherry shot-Jiole fungus. — Heretofore it has been sui^posed that the common shot-hole fungus of plums and cherries, Cylindro- sporiiim padi, confines its attacks to the leaves; but during the past season the discovery has been made that, on sour cherries, it also attacks the fruit pedicels with great severity. This dis- covery is of scientific interest chiefly and has no important bear- ing on the treatment of the disease. Anthracnose of cultivated snapdragon. — Our last report con- tained an account of a destructive anthracnose affecting tlie Antirrhinum or cultivated snapdragon. Recently it has been discovered that the same disease attacks the yellow toad-flax, Linarla vulgaris, a common weed closely related to the Antir- rhinum. This fact makes the prevention of the disease some- what more difficult than we have supposed it to be. Imperfect fertilization of peaches. — Througli imi)erfect fertiliza- tion of peach blossoms there may come about a condition some- what resembling the dreaded " little peach " disease. However, 18 DiKKCTou's Kei'okt of tub the two troubles may be readily distinjjjuished by the fuet that imperfectly fertilized peaelies have undersized pits eoutaiuiuj^ no kernel or else only a partially developed one; whereas, in the " little i)each " disease the pit is of normal size and contains a well developed kernel. Tile drain clogged hg fungus. — At Milton, N. Y., the three-inch « tile drain to a vinegar cellar became completely clogged by an unusually luxuriant growth of the fungus Lcptoniitus lacteus. The obstruction was readily removed by placing a handful ('f copper sulphate crystals in the upper end of the drain. Fungus in rrfrigorators. — The water pipes to refrigerators often become clogged with a dark-gray, slimy substance. The principal part of this slime consists of a fungus which is a vege- table growth and not an accumulation of matter from the ice. It ma}' be removed by occasionally pouring boiling water through the waste pipe. DEPARTMENT OF CHEMISTRY. Conditions affecting cJiasc curing. (1) Conditions affecting loss of tceight. — Loss of weight in cheese during ripening is due mainly to evaporation of moisture from cheese and, at long-con- tinued temperatures above 70° F., to leakage of fat. Loss of weight varies with following conditions: (1) Amount of moisture originally in cheese; the greater the percentage of moisture in the ch<'ese the more rapid and greater the loss of moisture. (2) Temperature of curing-room; the higher the temperature the greater and more rapid the loss of moisture. (3) The degree of saturation w^ith moisture in air of curing-room; the more moist the air the less rapid the loss of weight. (4) The size and shape of cheese; increase of height or diameter of cheese decreases the rapidity of relative loss of weight. (5) The tex- ture of cheese; the closer and more solid the texture, the less rapid the loss of moisture. These results point conclusively to the necessity of providing curing-rooms in which the conditions of moisture and tempera- ture can be controlled. Lower temperatures with proper New Youk Auuicultuual ExrERiMENT Station. 19 amoimt of moisture in air result in larger amounts of clieese to sell and at the same time cheese of better quality. (2) A study of enzjfmcs in cheese. — Methods of making and curing cheese improve slowly, because we do not yet know with cer- tainty what agent or agents are the causes of cheese ripening. During the past three years the chemical and bacteriological departments have been making a careful study of the factors that are commonly regarded as the active ones in producing ripening of cheese. The results, as far as published, appear to indicate that neither the enzymes secreted in cows' milk nor those produced by bacteria in the milk previous to its being made into cheese are to be regarded as the most prominent factors in normal cheese ripening. DEPARTMENT OF ENTOMOLOGY. f^praijing experiments with crude petroleum. — ^Series I of these tests included the experiments to determine the effect of crude petroleum upon normal trees, and Series II tlie experiments to determine the percentage of petroleum in an emulsion with water required to kill the San Jos6 scale. Three hundred and twenty-one fruit trees were included in these experiments, con- sisting of apples, cherries, pears, peaches and plums. The results were fairly uniform. In the experiments of Series I no injury was caused by the 25 per ct. emulsion except to peach trees, but in every case 40 per ct. and higher i^yc^'centages caused serious injur}- to European plum trees, and to apple trees when the emulsion was applied during the fall or winter. Early spring applications of the 40 per ct. emulsion did not injure apple trees. Pear and cherry trees were not harmed by the emulsion or undiluted petroleum even when applied during the fall or winter. The experiments to ascertain the percentage of petroleum required to kill the hibernating scales also gave uniform results. The 25 per ct. emulsion failed to affect the scales materially v/hile the 40 per ct. and higher percentages killed them in every instance. 20 DiitKCTou's Kki'out of the Taken as a v/hole these experiments indicate the following: 1. Vigorous trees are probably less liable to injury by crude petroleum than weak ones. 2. Peach and plum trees are more sensitive to crude petro- leum than apples, cherries or pears. 3. There is less danger of injury if trees are sprayed in early spring than during the fall or winter. 4. The 25 per ct. emulsion of crude petroleum and water can- not be depended upon to kill the hibernating scales in the latitude of western New York while the 40 per ct. has proven efficient. 5. Much pains should be taken to avoid over-drenching the trees. Only enough of the emulsion should be applied to wet the bark evenly and thoroughly. Washes. — The resin-lime mixture and government whitewash did not adhere to the trees well and apparently had but little effect on the scales. Fumigation. — The fumigation experiments in western New York with hydrocyanic acid gas were also divided into two series. Series I included the experiments to determine the effect of the gas upon bud sticks for budding purposes, and Series II the strength of the gas required to kill the hibernating scales. In both series the gas was used at strengths varying from .18 to .3 gram of cyanide of potassium per cubic foot of air space. The exposure of the buds to the gas varied from one- half hour to one hour. The experiments with buds, while not entirely satisfactory owing to the somewhat unfavorable conditions surrounding the treated buds, gave sufficiently uniform results to indicate clearly that the gas is harmless except in the case of the peaches, which were evidently injured slightly by the gas at .3 gram of cyanide. There was but little difference in the percentage of treated buds that set and the checks. In all 4,483 buds were treated, 78 per ct. of which set. The checks numbered 4,804, of which 85.5 per ct> set, making but a slight diiference in their favor. This differ- New York Agricultural Expeirimbnt Station. 21 ence was probably due in large part to imperfect protection and accidents to the treated buds after setting. Tlie experiments of Series II resulted in a failure to kill the scales during the winter with gas of less strength than .3 gram of cyanide. The spring treatment gave different results. The gas at a little more than half the strength (.18 gram) killed the scales in every case and did not injure the foliage. In tests made on Long Island the conclusion was reached that it is possible to exterminate the scale in small, isolated orchartTs of small trees by fumigation. Under favorable circumstances the gas from .15 gram of cyanide per cubic foot of space sufQced to kill the scales; but where the fumigation is done over damp soil, or when the trees are wet, it is best to use twice this amount as the gas is rapidly absorbed by water, thus reducing the percentage in the air. It is safe to use gas of this strength (.3 gram of cyanide per cubic foot) for frojn 30 to 60 minutes upon all dormant orchard trees. Trials of different proportions of cyanide, acid and water in the formula for generating the gas in fumigation showed that 1 part of lump cyanide by weight, \\ times as much acid by volume and 3 times as much water by volume gave complete, rapid and not too violent chemical action. This formula differs but slightly from the commonly used formula (1-1^-2:^); so that the latter may be followed if preferred, using a little more water if the action seems too violent. Promising inseotioides. — Certain insecticides which were tried as most promising remedies for the San Job6 scale, but which require further tests to demonstrate their value are whale-oil soai) and crude petroleum compound, the lime-sulphur-and-salt wash and a kerosene-lime emulsion. Modlfloation of the Station fumigatm: — This consists of a new method of holding the door in place. Instead of buttons, four strips extend across the front of the door and project about three inches on each side. The projecting ends are cut on a bevel and fit against corresponding surfaces of blocks fastened to the sides of the fumigator. As the door is pressed down it is forced securely into place. 22 Diuector's Eeport of the Hexagonal foklimj funiUjalor. — For the work on Long Island a new form of fnmigator was devised, wliicli possesses some advantages over all other forms. This is hexagonal in form, with sides hinged to allow of folding into compact form for transportation and storage, and with removable folding top. In operation the box is held rigid by the top and by braces at the bottom. Two sides and part of the top swing back easily to allow of placing the fumigator about the tree to be treated. The hexagonal form avoids waste space about the tree. DEPARTMENT OF HORTICULTURE. The forcing of lettuce has come to be one of the important industries connected with market gardening in this and adjacent States. In 1895 a line of- experiments was undertaken at this Station bearing upon practical problems which are to be met in the business of forcing lettuce. The first report on this line of work was given in 1898, in Bulletin 14G, and also in the Station's Annual Report for that year. This report treated of " Soil Mix- ture for Forcing Lettuce," and " The Use of Commercial Fer- tilizers in Forcing Head Lettuce." In the conclusions therein set forth it was stated that when the soil was fertilized with heavy applications of stable manure no advantage seemed to fol- low the addition of either sulphate of potash, acid phosphate or nitrate of soda. On the clay loam mixed with 15^ per ct. stable manure by weight a slight increase in growth followed the addi- tion of nitrate of soda. Since 1898 the investigations have been continued each year for the purpose of gaining further informa- tion on the economical use of commercial fertilizers in forcing lettuce either when used alone or in combination with stable manure. Nitrogenous commercial fertilizers were tried alone and in combination with various percentages of manure. The tests were made with loose lettuce and head lettuce both on a medium clay loam and a light sandy loam. The nitrogenous commercial fertilizers which were com])ared were nitrate of soda, at the rate of 600 lbs. i)er acre, sulphate of ammonia 480 lbs. per acre, dried blood 1,000 lbs. per acre, and a combination New York Agricultural Experiment Station. 23 of 850 lbs. of dried blood and 100 lbs. nitrate of soda per acre. The aiiioniit of nitrogen thus applied was approximately the same in each case. The use of these commercial fertilizers with no manure was followed by a much better yield of lettuce than that produced by similar soil not fertilized. On the clay loam the use of the nitrate of soda without manure -was followed by a better yield than followed the use of either sulphate of ammonia or dried blood without manure. On the sandy soil without manure dried blood generally gave better results than either the sulphate of ammonia or the nitrate of soda. With sulphate of ammonia and no manure the yields were very variable. These nitrogen- ous fertilizers alone, in the amounts applied, proved inadequate for forcing lettuce in a sufficiently short time to be profitable. Very much better crops were obtained when stable manure was added. The higher percentages of manure when combined with the jiitrogenous commercial fertilizers above named obscured the action of the latter so that it was not possible to decide that any advantage was obtained from adding them with the manure. With the smaller percentages of manure (5 per ct. and 10 per ct.) the addition of dried blood gave in the aggregate better results than either nitrate of soda ov sulphate of ammonia simi- larly combined. When 5 per ct. of manure was added to the soil with the com- mercial fertilizers referred to, the yields were invariably very much increased over those obtained with the same fertilizers and no manure. Double, triple and quadruple portions of manure increased the yield of the first crop but not to a cor- responding extent. With succeeding crops the cumulative effect of successive heavy applications of manure was seen in the actual decrease of the yield below that obtained with more mod- erate applications of manure. In forcing lettuce it is not uncommon for gardeners to use from 5 per ct. to 20 per ct. of manure. The amount which they use doubtless most often approaches the 20 per ct. rate, lu 24 Diuector's Rri'ort of tub tlicso experiments repeated applieatious at tlie rate of 15 per ct. to 20 per ct. of manure proved not only wasteful of manure but also lessened the yield. As in the previous experimentfi reported in Bulletin 14G tli(^ clay loam gave better crojis of lettuce than the sandy loam when both were given equal amounts of stable manure. The amount of manure which it is economical to use in forcing letfuce necessarily varies with the character of the manure and of the soil. It a'. so would vary to some extent with the differ- ence between the prices received for fancy lettuce and those received for the ordinary grades. For these rasons definite amounts cannot be recommended. CROP PRODUCTION. Commercuil fertUkcrs in onion ^roirt^^r.— Experiments in the use of diff(_'reut quantities of a complete fertilizer in growing onions were conducted at Florida, Orange Co., N. Y., for four years on the same field and for one year on a field of another farm. • The quantities of fertilizer used per acre were none, 500 lbs., 1000 lbs., 1500 lbs. and 2000 lbs. On the Purdy field (4 years), when only 500 lbs. of fertilizer was used the manure cost of the increase of crop was 16.6 cts. per barrel; with 1000 lbs., 79.3 cts., with 1500 lbs., 80.4 cts., and with 2000 lbs., 227.8 cts. The profit from using the fertilizer came mostly from the first 500 lbs. applied, averaging |35.84 per acre. With onions at 11.25 per barrel the profit was slightly larger (about |3 per acre), with both the 1000 lbs. and 1500 lbs. of fertilizer per acre; but 2000 lbs. was used at a loss. On the Mars field one experiment was conducted which showed no increase of yield from applying commercial fertilizer even in the larger quantities. The results of these ex[)eriment8 show clearly that the crops were limited more by other condilions than by the extent of the plant food supply. With the best conditions of season and water New Yokk Agricultural Experiment Station. 25 supply the smallest amount of fertilizer supported the maximum crop. Considering the varying market price of onions from one year to another and the various vicissitudes to which the crop is sub- jected, the use of the larger quantities of fertilizer (above 500 lbs.) was attended by danger of financial loss. Effect of manures on sugar heets. — These experiments were undertaken to tesl: the accuracy of the statement that sugar beets are of an inferior quality when grown on land to which stable manure is applied in the spring. The experiments have been conducted during four consecutive years, mostly on the Station farm. Comparisons have been made of the quality of beets not manured, those grown with commer- cial fertilizer, mostly 1000 lbs. per acre, and those grown on land receiving in the spring, before planting the beets, from 40,000 lbs. to 80,000 lbs. stable manure per acre. Beets from at least six varieties of seed were grown during the four years. The results are almost unanimous in one direction. The beets have been of high quality with all three methods of treatment, averaging somewhat better with the farm manure than with no manure or with commercial fertilizers. BULLErriNS PUBLISHE5D IN 1901. No. 197. October. — The food source of milk fat; with studies on the nutrition of milch cows. W. H. Jordan, C. G. Jenter and F. D. Fuller. Pages 32. No. 198. November. — Inspection of feeding stuffs, 1900-1901. W. n. Jordan and C. G. Jenter. Pages 29. No. 199. November. — An epidemic of currant anthracnose. F. C. Stewart and H. J. Eustace. Pages 18, plate 1. No. 200. November. — Notes from the Botanical Department: Trouble with pears in a nursery cellar; shot-hole fungus on cherry fruit pedicels; anthracnose of yel- low toad-flax; imperfect fertilization of peaches; tile drain clogged by fungus; a fungus in refrigera- tors. F. C. Stewart and H. J Eustace. Pages 21, plates 4. 26 DiKECTOU'S Rk1'(JIIT. No. 201, December. — Report of analyses of commercial ferti- izers for the sprinj? and fall of 1901. L. L. Van Slyke and W. H. Andrews. Pages 60. No. 202. December. — San Jos6 scale investigations, III: Spray- ing experiments with crude petroleum and other insecticides; fumigation experiments with hydro- cyanic acid gas; other promising insecticides; a modification of the Station fumigator. V. H. Lowe and P. J. Parrott. Pages 40, plates 2, figure 1. No. 203. December. — A study of enzymes in cheese. L. L. Van Slyke, H. A. Harding and E. B. Hart. Pages 30. No. 204. December. — Inspection of Paris green and other in- secticides, 1901, L. L, Van Slyke and W. H, An- drews, Pages 6. No, 205, December, — Influence of manure on sugar beets, W. H, Jordan and G. W, Churchill. Pages 14. No. 200. December. — Commercial fertilizers for onions. W. H. Jordan and F. A. Sirrine. Pages 10. No. 207. December. — Conditions affecting weight lost by cheese in curing. L, L, Van Slyke, Pages 30, fig- ures 6. No. 208. December. — Stable manure and nitrogenous chemical fertilizers for forcing lettuce. S. A. Beach and H. Hasselbring. Pages 30, plates 10. No. 209. December. — Treatment of San Jos6 scale in orchards, I: Orchard fumigation, F. A. Sirrine, Pages 29, plates 10. No. 210, December, — Effect on milk production of change of rations, W, P, Wheeler, Pages 63. No. 211. December. — Director's report for 1901. W. H. Jordan. Pages 19. W. n. Jordan, Director. New York Agricultural Experinicnt Station, Gfueva, N. Y., Dec. 31, 1901. REPORT OF THE Department of Animal Husbandry. W. H. Jordan, Director. W. P. Wheeler, First Assistant. C. G. Jbnter, Assistant Chemist. F. U. Fuller, Assistant (Jlwmist. Table of Contents. I. The food source of milk fat, with studies on the nutrition of mikli cows. 11. Elfect on milk production of changes in the ration. REPORT OF THE DEPARTMENT OF ANIMAL HUSBANDRY. THE FOOD SOURCE OF MILK FAT; WTIH STUDIES ON THE NUTRITION OF MILCH COWS.* W. H. JORDAN, C. G. JENTER AND F. D. FULLER. The tests herein reported are in continuation of one given in Bulletin 132 relating- to the food source of milk fats. The conclusion reached in that experiment, that part, at least, of the milk fat comes from the carbohj'drates, is confirmed; and other facts relating to metabolism and utilization of food by milch cows are brought out. Three cows were used: Cow 12 fed a fat-poor ration in which the protein supply was gradually decreased from 2.G lbs. daily to l.G lbs. and then gradually restored to the maximum, with accompanying increase and decrease in carbohydrates so that the digestible dry matter of the ration was kept fairly uniform; Cow 10 fed a ration with normal supply of fat at first which was gradually increased to 1.4 lbs. daily, then gradually restored to the normal; Cow 2 fed the herd ration having a nutritive ratio about 1:5.6. These rations were quite varied in character and contained some fat-extracted foods; yet showed a quite uniform digestibility of about 70 per ct. of the dry matter. It is believed that this figure represents fairly the digestibility of rations made up in part of silage and containing a fair propor- tion of high class grains. A widening of the nutritive ratios appeared to render rations less digestible, especially the pro- tein. The marked changes in protein content and in fat content of rations did not produce noticeable changes in the character ^\ lepriut of Bulletin No. 197. 80 Rrport of Department of Animal TTusBANDRt of the or composition of tlie milk. In the former test, during 59 days, 18.4 lbs. of fat was formed in the milk which could not have had its source in food fat or food protein and could hardly have been drawn from the cow's body fat as she increased in weight 33 lbs. in the same time. In this test Cow 12 in 74 days pro- duced 39 lbs. of fat similarly unaccounted for, with a body gain of 15 lbs.; and Cow 2, in 4 days, 1^ lbs. These amounts of fat must have come from the carbohydrates in the food. A lessening of protein supply in the food did not produce a corresponding decrease of protein in the milk solids, but caused a marked lessening of protein decomposition in the body. Calorimeter determinations sl>ow that the heat value of urine bears no constant relation to its nitrogen content, and also prove that the formula used in computing heat energy of urine, Nx5.343 Cal., is greatly in error, actual results being from 3 to 4 times as large as calculated by this formula. The energy values of nutrients as given by Rubner, — protein and carbohy- drates each 4.1 Cal. and fats 9.3 Cal. appear to be fully high enough for herbivora, even when the loss due to escape of unoxidized gases, methane chietly, is not considered. Over 49 per ct. of the available energy value of the rations was used for maintenance, over 30 per ct. reappeared in the milk solids, leaving a balance of from one-fifth to one-fourth of the ration. The logical conclusion is that this balance, in part at least, sustains the work of milk secretion. INTRODUCTION. Bulletin No. 132 of the New York Agricultural Experiment Station presented the results of an. experiment to determine the sources of milk fa't as related to the food suppl3\ The main con- clusion therein stated was that milk fat can be formed in part, at least, from carbohydrates, the data of the experiment point- ing to this conclusion in a most convincing way. It was felt, however, that so import aut a generalization, if correct, should be sup])orted by results secured with more than one animal. An ()p]K»i-tuiiity was d-sircd also for enlarging the scope of the observations. KxiteriiKculs have been conducted, therefore, New York AoRrcuLTURAL Experimext Station. 31 with three other cows and the data thus obtained not only fur- nish additional evidence concerning the main question of the source of milk fat but have been used in studying other ques- tions relating to the metabolism of the milch cow. The publica- tion of the results secured has been much delayed, largely because of the great amount of work involved in the investigations. THE PLAN, MATERIALS AND PROCEDURE OF THE EXPERIMENTS. THE FORMER EXPERIMENT. In the experiment previously reported the cow was fed a nor- mal ration for a time; then, for 95 days, she was given foods that, because of extraction, contained very small proportions of ether extract. During 66 days the solid and liquid excreta were collected for analysis and daily analyses of the milk were made for a longer period. The quantity of the ration and the nutritive ratio were varied in a way calculated to show the influence of the protein supply upon fat secretion. The experiment was so planned as to make it impossible for the milk fat to have its source wholly in the protein and ether extract of the food and so long continued that any material draft upon the body of the cow for milk pro- duction would produce a marked change in the condition and weight of the animal. (For details see Bulletin 132.) THE NEW EXPERIMENT®. The later experiments involved the use of three cows quite unlike in their characteristics and productive capacity, each of which received a ration distinctly different from the rations given the other two. (1) Cow 12, a grade Shorthorn weighing about 1,200 pounds, fresh in milk at the beginning of the experiment and not preg- nant, was fed for 88 days a ration containing little ether extract, the nutritive ratio being very gradually varied from narrow to wide and back again, the total quantity of digestible dry matter consumed daily being maintained at a fairly uniform quantity. 32 IvKrouT OF Dei'aktmen't op Animal Husbandry of the A record of the amount and composition of tlie food, milk and excreta was kept for 74 days. (2) Cow 10, a grade Jersey weighing about 750 pounds, fresh in milk at first and not pregnant, was fed a mixture of normal feeding stuffs for G8 days, the fat in which was varied from a proportion fully as large as is ever found in practice to a quan- tity somewhat excessive, the daily supply of digestible matter remaining quite constant. A record of the amount and com- position of the food, milk and excreta was kept for 54 days. (3) Cow 2, a full blood Jersey, of very large productive capacity and in full milk flow, weighing about 780 pounds, was fed the usual herd ration. The amount of food was accurately weighed for 20 days and the milk and excreta were weighed and analyzed for the last four. THE FOODS. The feeding stuffs used in these experiments, some of them normal and some having been submitted to a petroleum extrac- tion, had the following composition: Composition of Feeding Stuffs. Fat Lab. Water. Ash. Nitro- Pro- (Pt-trolcum No. gen. telu.*. extract.) Per ct. Per ct. Per ct. Per ct. Per ct. 712 Mixed hay, Cow 10 14.5 6. 30 1.G2 10.12 1.30 717 Alfalfa hay, Cows 2 and 12. . 10. GO 0.92 2.39 14.94 1.44 710 Oat straw, Cow 12 10.75 5.27 0.33 2.00 1.44 713 Corn meal, Cow 10 22.12 1.28 1.24 7.75 2.27 730 Corn meal, Cow 10 20.20 1.55 1.35 8.44 1.57 720 Corn meal, extracted, Cow 12 12.30 1.10 1.50 9.73 0.37 740 Corn meal, extracted, Cow 12 11.05 1.14 1.53 9.50 0.37 718 Rice meal. Cow 12 12.07 0.31 1.53 9.18 0.11 719 Rice meal, Cow 12 11.87 0.34 1.53 9.18 0.09 727 Rice meal, Cow 12 13.45 0.00 1.13 0.78 0.:U 741 Rice meal, Cow 12 12.79 0.38 1.50 9.O0 O.os 721 Oats, extracted, CoAv 12 11.18 3.00 2.20 13.20 0.79 739 Oats, extracted, Cow 12 12.07 2.98 2. 12. 0.71 745 Wheat bran. Cow 2 11.21 1.29 2.44 13.91 4.33 740 INIalt sprouts. Cow^ 2 10.40 5.38 4.22 25.-32 1.07 714 Linseed meal, Cow 10 7.05 4.75 5.9(i 32.78 0.71 737 Linseed meal. Cow 10 9.48 4.78 5. 84 .32.12 7.07 747 Linseed meal. Cow 2 9.52 4.04 5.85 .32.17 0.94 715 Flaxseed, ground, Cow 10. . . 9.75 2.91 3.34 18.37 37.2!) 7.38 Flaxseed, ivround. Cow 10... 4.04 3.00 3.42 18.81 38. S4 722 Wheat gluten, Cow 12 7.10 0.33 13.08 75. .54 0.71 723 Wheat gluten. Cow 12 9.10 0.43 12.04 71.01 o..-)3 742 Wheat gluten, Cow 12 6.38 0.72 12.80 72.90 0.94 Sugar beetst ♦With hay, oat straw and corn imvil, profcin = N x (! 25 ; with r:ce moal, oats atul malt spmiits, protein = N xfi; with wheat, I>ran amt wheat gluten, protein = N x 5.7 ; with linseed meal and grounil flaxseed, protein - .'^ x 5.5. t n.iily analyses of sii^ar beets were nndo. New York Agricultural Ex^brimbnt Station. 33 the rations. Coic 12. — The daily ration of this animal as established at first was as follows: Ration of Cow 12. Alfalfa hay 8 lbs. Rice meal 3% lbs. Oat straw 8 lbs. Oats (extracted) 3 lbs. Sugar beets 27 lbs. Corn meal (extracted) 2 lbs. Wheat gluten I14 lbs. After three w^eeks, including one week of preliminary feeding, the wheat gluten was diminished one ounce per day and the rice meal was increased by a like amount. This change was con- tinued until no wheat gluten was fed. After seven days of absence of wheat gluten from the ration it was again intro- duced, one ounce the first day, two ounces the second and so on, the rice meal being diminished to the same extent until the original ration was reached. The feeding was then continued on this basis. It should be said that a uniform addition of 1 pound of rice meal w^as made to the ration at the end of the first month's feeding. This method of varying the ration allowed a very gradual change in the protein supply in both. directions with no lessen- ing of the supply of digestible organic matter, thus making it possible to study the relation of protein to milk secretion with- out the disturbing influence of sudden changes in the character of the ration or of a deficiency of carbohydrates. Cow 10. — The basal ration of this cow consisted wholly of normal foods and was designed to supply a generous amount of vegetable oils. Ration of Cow 10. Mixed hay 12 lbs. Linseed meal , . 2 lbs. Sugar beets 27 lbs. Ground flaxseed 1 lb. Corn meal 4 lbs. The ether extract in this ration at first was about .8 lb. per day and it was maintained at this amount for about one month when it was gradually increased by a substitution of ground flaxseed for a like quantity of linseed meal at the rate of ^ lb. per day until the ether extract ingested daily was 1.4 lbs. 3 34 RBroHT OF Department of Animal Husbandry of tue After feeding at this rate^for a week tbe fat was diminished at the same rate as it had been increased by gradually substitut- ing rice meal and extracted corn meal for normal corn meal and the ground flaxseed. The purpose of these changes was to note the effect of the supply of fat upon milk secretion and pro- tein exchange. Cow 2. — ^This cow, whose rate of yield was over 2 lbs. of but- ter per day, was fed essentially the same mixture as was given to the whole Station herd. Ration of Cow 2. Alfalfa hny 6 lbs. Wheat bran 4i/ilbs. Corn silase 40 lbs. Malt sprouts 2y^ lbs. Sugar beets 10 lbs. Liuseed meal 2^/4 lbs. This ration was accurately weighed from INIarch 29th to April 18th inclusive, and during the last four days of this time the necessary observations were made in the collection of excreta and weighing of milk. The quantities of each constituent of the rations which were consumed during the various periods, expressed in grams and so arranged as to easily trace the changes, are shown in the succeeding table: New York Agricultural Experiment Station. 35 a) — m CQ •a Ji?3 s 3 5 © _ ■* 00 tH CQt^ C l> C t- O »0C3-t-^-tiO-f+^0 Tt( lO C^ CO "M ^- CO o «o O 05 +^ '-' C3 I- r-< o a bl) 05 Q O M a P-. m O |-^ o O 5 «»3 ^ 3 =ooTi-;i?j or. s) do 3 to OO CO ^ '?^ «3 o CO CO CO CO CO o o ro o o CO o o CO c% o\ CO (M O CD 03 Ol CO CD O oj t^j >M 1-1 T-l (M S^ 'J-« ■S 2 O I CO get C5 CI I- CI cf .00 00 Sco CCI 5>i0 00 CI o CO CO 00 CI o CO CI CI 1^1 I- t- t- -ti -tn ;i; CI CA ^1^ cT cf ^\ CO CO CI o CO 00 CI o 00 00 GO CO CI o CO CO CI o CO 00 CI o CO ^CO 00 oo 00 00 QO CI CO CO CI I- -f CI cf >.CI C3 • •=co a, ^ CI CI cF CO ■^ ^ T-H O CO "-I O +^ lO t-l I-i I-l T-( ^ 00 CO o d '-H*-' O 1H CO CO 2 d T-H +^ O ?H 00 M CI CI CI CO t- CI cl M CO iro C5 oi ^^ g a) to 0) c3 " t- 9 '1' 3 •^^ 7j l-t li O yj CO +^ »-: 3 •v o n ^ 03 r-l CI a. !«, a +-1 CO o r3 CI CI tH <5 o o 1^ o C3 o o 5=; ;3 oo t- fO CJ /= J3 OJ »!H ft, >i^5 O .a GO CI -t-i \o 03 CI C3 o C3 fc. s 3 ^ 2 3 ©■i So 0) 36 RfiroRT OF Department of Animal Husbandry of the METHODS OF SAMPLING AND ANALYSIS. The rations were weighed out at several different times during the course of the experiment, and each time this was done samples were taken of the various foods. The milk, urine and feces were taken directly to the labora- tory and immediately weighed and sampled, excepting that the night's milk was kept in an ice box until morning, when it was mixed with the morning's milk and a sample was then drawn from the mixture. The feces were thoroughly stirred and samples (4 lbs.) of the fresh material were taken for drying. These samples were dried over steam coils at a temperature ture not exceeding 60° C. In general the methods of the A. O. A. C. were followed in the analyses, the only exception being that petroleum ether was used instead of sulphuric ether in extracting the fats from the foods and feces. (See Bulletin 132.) Nitrogen was determined directly in fresh samples of the urine and feces. The drying of the feces at a temperature vary- ing from 50° to 60° C. caused a material loss of nitrogen, as previous results clearly show. (See Bulletin 132.) THE RESULTS OF THE EXrEEIMENTS. The results of these experiments are presented without a full statement of the data involved. The omissions are the daily weights and daily composition of the feces, urine and milk, the figures for which would occupy many pages and would be of use only to those who wish to study the data from some standpoint not considered by the authors. The points that will be discussed are the following: (1) The digestibility of the rations, with some reference to the influence upon digestibility of the proportions of nutrients in the case of Cow 12. (2) The influence of the composition of the ration upon the quantity and composition of the milk and upon the composition of the milk fat. New York Agricultural Experiment Station. 37 (3) The source of milk fat and the relation of milk fat forma- tion (a) to protein exchange and (b) to the supply of food fat. (4) The metabolism of nutrients as affected by the composi- tion of the ration. (5) The energy value of the digestible portion of a ration. (6) The distribution of the food energy in maintaining a mikh cow. THE digestibility OF THE RATIONS. The figures which immediateh^ follow show the extent to which the rations were digested. Digestibility of Rations. Dry sub- stance. Organic matter. P-te.„. C-^^,\y- Ash. Coir 12. during 1st Qrams. Grams. Grama. Grams. Grama. period. 7 days (Jan. 30-Feb. G) In foods 909S5.1 8G594.5 4390.6 12371.9 73340 In feces 27698.6 24894.7 2803.9 4210.6 19981. S Digested 63286.5 61699.8 1586.7 8161.3 53358.2 Percentage di- gested 69.6 71.3 36.1 66.0 72.8 Eaten daily, lbs. 19.9 19.4 0.5 2.6 16.8 Nutritive ratio 1:6.3 Cow 12, during 2d period, 7 days (Feb. 19 to 26n In foods 92657 . 8 88259 . 6 4398 . 2 11261 . 9 76105 . 2 Id feces 28528 . 1 25935 . 1 2593 . 4409 . 4 20850 . 1 Digested 64129.7 62324.5 1805.2 68.52.5 55255.1 Percentage di- gested 69.2 70.6 41.0 60.9 72.6 Eaten daily. lbs. 20.2 19.6 0.6 2.2 17.4 Nutritive ratio 1:8.0 Cow 12, during 3d period, 8 days March 8 to 16). lu foods 105344 . 1 100308 . 7 5035 . 4 10538 . 1 88776 . 6 In feces 34.509.2 31587.6 2921.6 4774.4 26077.7 Digested 708.34.9 68721.1 2113.8 5763.7 62698.9 Percentage di- gested 67.2 68.5 42.0 54.7 70.6 Eaten daily, lbs. 19.5 18.9 0.6 1.6 17.3 Nutritive ratio 1:10.9 Ether extract. Grams. 882.6 702.3 180.3 20.4 0.00 892.5 675.6 216.9 24.3 0.05 9W.0 735.5 258.5 26.0 0.07 S8 Report of Department of Animal Husbandry of thb Dry sub- Organic .:», PrntPln Carbohy- Ether stance. matter. x-roxem. drates. extract Cow 10, period 10 Grams. -Orams. Grains. Grams, Grams, Grama. (lays (Feb. 4th- 14ih). In food 102.-188.5 96887.2 5501.3 13178.7 80078.0 3630.5 In feces 28023.3 25473.1 2550.2 5114.4 19632.3 726.4 Digested 74365.2 71414.1 2951.1 8064. 3 60445.7 2904.1 Percentage di- gested 72.6 73.7 53.6 61.2 75.5 80.0 Eaten daily, lbs. 16.4 15.7 0.7 1.8 13.3 0.6 Kutritive ratio 1:7.5 Cow 2, period 4 days (April 14th-18th). In food 48546.0 462S4.2 2261.8 7768.1 36982.2 1533.9 In feces 15128.5 13730.5 1398.0 2424.5 11275.9 30.1 Digested 33417.5 32553.7 863.8 5343.6 25706.3 1503.8 Percentage di- gested 68.8 70.3 38.2 68.8 69.5 98.0 Eaten daily, lbs. 18.4 17.9 0.5 2.9 14.2 0.8 Nutritive ratio 1:5.6 The several rations fed to these cows exhibit a somewhat note- worthy similarity of digestibility. It appears from these and other observations that when a ration is made in part of silage and contains a moderately large proportion of the high class grains not far from 70 per ct. of the dry matter is digested. In the case Cow 12 an opportunity is given to note the influence upon digestibility of varying the proportion of nutri- ents. The data which it is essential to consider are the following: Effect of Varying PnoroRxiONS of Nutjjients on Digestibility. bleorgiinlc H,'5nr*i Nutri- Organic „. ,„,„ Period. CHANGES IX FOOD, matter .if„iT,™ live matter i!'<"^'" eatea ^Zn% '''"°- digested. •^'«^^"^<1- dally. '^^"y- Jan. 30 to Feb. 6... U^ lbs. wheat Lbs. zbs. r,;rt. rer ^^ gluten fed. 19.4 2.6 1:6.5 71.3 66Vo Feb. 19 to 26 Wheat gluten partly re- placed by rice meal.. 19.6 2.2 1:8.0 70,6 GO. 9 Mar. 8 to 10 Wheat gluten wholly re- placed by rice meal.. IS. 9 l.G 1:10.9 68.5 54.7 New York Agricultural Experiment Station. 39 Much stress has been laid in the past upon the necessity of avoiding too wide a nutritive ratio, that is, too large a propor- tion of carbohydrates, because of a depression of the digestibil- ity of the ration, especially of the protein. In the case under consideration there appears to be a gradual decrease in the pro- portion of total organic matter digested from the first to the third periods, and it is logical to conclude that this v.'as caused by a widening of the nutritive ratio, because other conditions remained the same. The protein is also apparently consider- ably less digestible after the withdrawal of the wheat gluten. This is due in part at least to the high rate of digestibility of the protein in wheat gluten as compared with that in other parts of the ration. Moreover, under these conditions the presence of certain metabolic products would cause an apparent rather than a real decrease in digestibility. The extent of the influence of an increase in the proportion of carbohydrates cannot be seen clearly in this instance, although the evidence in favor of a depression of digestibility is as valid as that from which former conclusions have been drawn. The foregoing figures make it plain that the several rations furnished abundant and not unusual nutrition to the cows eat- ing them, excepting, of course, the very small amount of fat supplied to Cow 12 and the abnormal supply of fat in the ration of Cow 10 for a short period. THE INFLUENCE OF THE OOMPOSITION OF THE RATION UPON THE QUA:S'TITY AND COMPOSITION OF THE MILK AND MILX FAT. Before discussing the questions to which these investigations have more especial reference, it is important to inquire whether the quantity and character of the cow's product were in any- way modified by the unusual character or variations of the rations. This inquiry is all the more pertinent because of the prevail- ing notion, not yet justified by any researches whatever, that the composition of the ration determines to a large extent the character and composition of the milk. 13.47 4.01 13. G5 4.05 13.73 4.11 13.78 4. OS 40 Eeport of Department of Ammal Husbanduy of tjib Effect of Variations ix Ratiox upon the Milk, Cow 12. Variations in Protein Supply. Jlllk Solids Fat Period. CHANGES IN RATION. yi.ld in In dully. milk. milk. Jan. 30 to Feb. G... Maximuiu protein fed (2.G lbs. Lbs. Pent, p-rct. daily) 35.1 12.02 3.72 Feb. 6 to IG Maximum protein fed 32.2 13.04 3.G8 Feb. 16 to 2G Protein diminishing, carbohy- drates increasing 30.1 13. 3G 3.92 Feb. 26 to Mar. 8... Protein still diminishing, carbo- hydrates still increasing 28.4 13.37 3.87 Mar. 8 to 18 Protein at minimum (1.6 lbs. daily) 26.0 Mar. 18 to 28 Protein increasing, carbohydrates diminishing 26 . 1 Mar. 28 to Apr. 7 . . Protein still increasing, carbo- hydrates still diminishing 26.5 Apr. 7 to 14 Protein at maximum (2.6 lbs. daily) 26.1 Cow 10. yariations in Food Fat Supply. Jan. 30 to Feb. G. . . Normal ration (fat fed daily, Lbs. Perct. Perct. .8 1b.) 22.9 14.31 4.74 Feb. 6 to 13 Ration unchanged 22.8 14.20 4.74 Feb. 13 to 20 Ration unchanged 22.8 14.20 4.75 Feb. 20 to 27 Ration unchanged 23.5 13.90 4.46 Feb. 27 to Mar. 6.. . Food fat increasing 23.4 14.00 4.60 Mar. 6 to 13 Food fat at maximum (1.4 lbs. daily) 23.7 14.17 4.7G Mar. 13 to 20 Food fajt dimini.shiug 24.6 13.81 4.44 Thre is nothing in these data to warrant the conclusion that supplying more or less protein or more or less fat to a milch cow causes material changes in the milk. In the case of Cow 12 her milk suffered a gradual and quite constant increase in its proportion of solids and of fat, but this change was in no way disturbed in its progress by the fall or rise in the proportion of protein in the food. With Cow 10, the increase of the food fat to 1.4 lbs. daily, a most abnormal quantity, did not raise the milk fat above what appeared to be the normal proportion. These results stand in accord with the outcome of many other carefully conducted investigations. The question whether entirely normal milk fat was produced with a fat-free ration, or nearly so, is an interesting one. The Isnw York Agfjcultural Experiment Station. 41 only evidence whicli these experiments supply along this line was obtained by the partial analysis of the milk fat from the cow which w^as the subject of the experiments detailed in Bul- letin 132, In this experiment the cow was fed a normal ration during a portion of the time she was under observation, so that it is possible to compare the fat produced before and after the food fat was almost entirely withdrawn. Partial Composition of Milk ^vitii Normal and Extracted Foods. Periods represente by diff.Tt-nt lou of milk fat. April 19 to 26.. May 3 to 10... a KIND OF RATION. . . Normal food. Food fat daily Extracted food. Food fat daily .63 lb. . Average daily yield milk fat. .. .96 Reichert number grams of fat. 16.3 .125 lb. .. .74 13.6 May 10 to 17.. . Extracted food. Food fat daily .125 lb. .. .76 14.1 May 17 to 21:. . Extracted food. Food fat daily .125 lb. .75 14.1 May 24 to 31.. . Extracted food. Food fat daily .125 lb. .. .65 14.5 May 31 to June 7 Extracted food. Food fat daily .125 lb. .. .60 14.9 June 7 to 14 . .. Extracted food. Food fat daily .125 lb. ... .60 14.9 The Reichert number shows the relative proportion of volatile acids in milk fat. As an unmistakable and permanent drop occurred in this number immediately following the substitution of the fat-poor ration for the normal, it is fair to attribute to the food fat some influence upon the milk fat, though a single observation of this kind should not be taken as conclusive evi- dence. At the same time, the milk fat produced while the extracted foods were being fed contained a proportion of vola- tile acids larger than often found with normal rations. That the glvcerides of these acids were formed in the cow and were not derived as such from the food is so evident as to render dis- cussion unnecessary. THE SOURCE OF MILK FAT. The main question involved in this investigation is the source of milk fat. The importance of the question and the reasons why it seemed to demand further investigation are set fo'^th in Bulletin 132. It is sufficient to state in this connection that the inquiry is concerned with the relation of the several nutrients to 42 Report of DErARXMENT of Animal Husbandry of the milk fat secretion, whetlier this fat is derived wholly from fats previously formed in the plant or from protein, or whether car- bohydrates may support its formation either directly or through the previous formation of body fat. The tables which immediately follow give the daily nitr n and fat balance for the three cows. From these tables various summaries are derived which show the bearing of the evidence secured. New York Agricultural Experiment Station. 43 M j^ _ CO I- 05 -^ CO t- Irt t- iM LO t- t- CO lO I- CO O tH o CO 00 t- 05 t- lO I- L- O - ■^ ^ I L-^ lo irri th CD i35 oD O o -^i -+i t-- b- CO iH c-i th oti ci lo 00 lO CO (ri c;3 cc cr' 70 S rt ^ - - -^ - . riTHOOKtlMCI5CO<35U'OCCCOlOOOTt-«tlr-lTtiOT-(.-(b--^OOoO'-^C^1-fcO'-t^'-tfOO'*iT-iOcoqir-cDt-r-ico(MC4!rocc^c« b<^t-l, COOCOCOl-t-CDCOCOCDlOCOCOCD^OCDcOCDCO^'XiCOCOlfJOC) KOi5t:-QOaOiMOmcO^'HC^1Oi35C005«0»0-^iNMCDt-0''-0 03 00c \^\ •'MOOOOOOOS'PCOi-Mt— OS'iPO-tiOt—OSlOiHh-QOCOCZ) irHrsjsvicJsoOGOoocai-OCTococaQCCsoosaso^oi .i tS ^QOr^t-rt^^-.co•<*^(^^05<^^co'-|O0i'>*lTtiiM ScOoOKOCOlQlftTtiCOGO'^O'OOOcOCZJt-r-COCOt-l-OOQOoOt-CDCOCO s- CJ rj !>« I^ M Cl ^4 iM CM C^l (M (M CI c^^ Ol C^ C^l (M (M o| is| C^l > Oj OO CO 00 O iH oiiM' Cl o 2 1- W O 05 b- O O'^J<^t- tH ITS tH c do' r-I co" CO r-l 05 Cjco CI CO >-< oicoOcococorHOjcicico CI C^ CI rH tH CJ ,-1 r-l 00 C^|05 CO _• ^^005MTj''-'^fOr-(OW 05 3 o Eh CO t- I- GO O I'+l lO ^ CO GO ' CI C4 CI C< CI ?o Tt< CO CO J2 00 >2 JS CO r-i Jr '^ c o rH c-J t- 00 lO IQ CO t^ O JO Jo t- «3 I- CO 1- CD '- M CM C^ CO'Hl-ini05t-05rH©WcOeOCO (S<»' I CO ( I GO : O 1 O < i« O 05( ' O 05 00 ! il>t:rHr-lQ00-tiOl0"*'*'C£'Q0O205CD30 0005COOC»rHooocfeoa505iHgo50505 aJ aOOOCMcoOCO-f'^TticOlOTHTtiCOOOi-IGOCMTtit-'stit-O'^OO'*'-' dg •••• •E CO-!tit-Ofc-'^CJQlOi005t:-TtiT-tcOC005CICD0005-.t-0Oi>t-U) J 2 0500»OlrtC'1OC.0005h-|— t— rft--QO X 00 I- 00 00 I- !» 00 00 CO I- I- I- I- L- t~ I- b- I- t- t- .'-llO05i-liHC0lrt'HC0C0IOC0lO'^ClTt-(_)-l_l+-J■t-'-^J-4_l-^^-)^_)+-'+J+J-^_^+J^->^_(+J+J^_>.^-t-'^^-M-w O_'-ICIC0Ttii0C0|:^X05O'-|CIC0-*i0c0l-G005OT-iciiC-+ii'--O CO J^ tH ,H T-l iH ,H rH ,-i 1-1 tH i-l Cl M CI CI Cl CICI pj a •^* ■^" -^ -=J -2 ^' -O XJ ^' ;0 ^' -O X2 ^' ^ ;d ^' ^ ^ ;d ;0 ;0 -O X5 ;2 ^" 44 Report of Department of -Animal Husbandry of the fc- M -f CO C5 CO 1.1 >;T' 1-1 CO fO O LO LO •* r-l rH Tf :S ^ C C tH T-l tH L'^ r-J 00 •<* c >. . ,; K * 5 C QO C5 C5 "^O O I 1< GO I': I- I- GC fl >C T- '^ CO I- CI -f X C^I fl Lt! 't — ' o o 2 S ' C C CC ir: rH CO 'tl IC 1* -t — ?0 1~ -t X CI -t" -+ CO — tH X I- IS 1-H rH X l-i lO J^^ttL'OlOTflTtiTtlTflTtiTtlTtlTJiTIiTtlTtlCOTj'TtlTP'^Ttl-^Tt'^tl'^TtH-'tl-TlTtl'ii* r . « M Tt< rtn LO 1-1 Tt^ to O X I- t- C5 C: X M M Ci C^l OJ O C I- 01 r-i X CO * iO Ci Of) O ^ '^ — c ^ Z CO Cl o c^ X i>- 1^ o: c c; CO CO '^ f X X o: a W a r- . CO to X CO 'f X I- c o rH -i< o CO t- o 1-1 o -f iH -t< 1-j 1- c; 1- Cl 1- -* '^ is 5 o S 1-1 CO -f< M CO 1- 10 05 -t I- 1- 01 CO O 1- C CO C 1 1* I- 10 -t- 01 )0 1- C C5 !. -t< -t 01 CO ^ CI 01 01 ^ tH O C' O O C5 1-1 O T-t iH rH Cl CO 01 1— CO 1 — r O-i O M CI 01 Cl Cl 01 Cl Cl Cl Cl iH Cl C^l Cl 1-t CI Cl CI Cl CI CI CI CI CI Cl CI CI CA c IS* oi ^O OC5'3i©XCO'^OiHTj0 CO -P Cl x" CJ 10 C © C -t 11^ -t< -t- c. x' oi ft g o 05 o C5 OC' o o c; o X c; o; 05 c; Oi C5 cr. c - c c c X c X ST. o tp »' K ^ M 05 "* iH 05 Cl I- X I- CO X LO Cl X C ■* 01 10 f- X 0? © LO' Tf CO Tt Cl r/j z 5 C CO -MO CO 1- CO CO CO 1- 1- 01 © to 1-- e >0 C I- 1- to 10 O' CO 1- Cl ^ 1-1 © Qj '-^ CO O LO T)< LO lO LO tP Tti -f ^ TjH Tti Tti ■* •* Tfi r)i -f in CO LO LO CO to I- to O V 2'~'Tt<©e0f0r-l©©C^C0l"~'-'iHL0©t— liOClC<|l^©NlO©Tt<05t-.l-» -< .i > 10 in •*' co' Cl in ^t r' <^^ ci '— © -i^' t -•" — - ^-' -■ Ir ©' i- C c; c: J? ©' tr tS I- i^ t- I- I- I- 1- t- CO © © -O CO © © CO CO CO CO © I- I- I- I- CO «o it: 1© S 2 . © © -H LO © CO »n "+ X CO CO X © 1-1 Cl © -H © © CO LO Cl ->< © t- I- 05 00 tfS « • • • • • . . ?3 ?i-ii-x^©gC'i-i^oix — ©C'"'i~©©c-fXTJioxr--f-jr-;i-( -'•i rO l*^ CO CO CO Cl 01 01 r^ — — 1-- 1-" -^ r^ c; c 1-^ Oi 01 01 ^t co co -r -^ -f ^ m 2= ^ (M C^ C^ M C^ Cvl Cl Cl :-] 01 Cl Ol Cl Ol ^\ Cl Cl Ol Cl Cl c-l Cl CJ ci in M d Cl CO >^ oi cio •* >o' CO i-" QC ©' © i-* ci ff^ ■-)' >n ©' t-^ CO S5 o T-< c^ CO Tf lo co' •- Cl >^ r-l iH rH rH tH iH iH tH rH rH c^i C^ J^J C^l ^'l C-l f, 71 oooooooooooocooocoooooocooco I- X ,-1 Cl CO -t m © t- X © © i- Cl J.- -tH 10 © I- QC © © 1- Cl r: -t m ^ Cl Cl 1-1 1-1 1- ^ ^ rH ^ ^ i-i ^ Cl Cl Cl Cl M C\ Cl X ~' t.' u tJ ;-<' (J u :-' ~' C ~° ^° C C C C •— il. u u il tj l.' v.* t.' C u '0"o:'-;csc3c:c;nc:T:r:c;c:c:rtnc:c:r:t;rtc;ci!:irti;cicj Kev/ York Agricultural Experiment Station. 45 5 ocoa5-tiTHroin©Or-i>r:(MC'ir-(Ttir-icoo ^ o C) C5 t- QO O t- O tH CO CI C QO Oj QC t- QO '-0 O LO 'X LQ X lO L-S L-. >S O CC t~ C) CO 00 O X Tfi ^ I": ^73 o s o CI o o CI O ii > C32 O H -< O Z o o H ? O H w < a . D be o bo 3 Q « M©GOlOT-l'4 O X to L-O iH X LO X X O CC I- O CJ L- T-l T-( CI CI -*< CD Oi O tH lM05t-C0Cii-li-(XX05050000XCS»OT-IO S OXOI^XC1l-rHCD00t-i-IO0iC5OC4i-l rs O O C^ 35 C5 O O C5 O O X O O O' " X C O t- CI cr X CD l- 35 iH O iH CD X iH CI 35 CI " o CO 35 CD CO CO o C tH I- '^'-fLOt-XC>l»OCD-^Tt-0 -^ lO -O lO t- t- CD X 1- 35 X -fi t- t-; Cl C^ C4 CI d C< CI CJ CI CI CI CI . t- I- «D I- l- CI CI X tH -r L-O LO X X X © 35 b- CI CD L-O o »^ OO 1-1 krt. 00 1-1 © ooS CI«DxOf'CD00XCI*t*35OL'0i-t<^>2-f35 I'O "--O KO S - CD I- I- X X X 20 X X X X X X CI C^l C4 CI CI CI CI CI CI ^1 CI C^< CI CI CI 'X350iHiHC100'tllOcDl"-X350r-iCIOO CI CI CI CO CO iH i-( i-< 1-1 ^ ^ r*. ^ '"rL '-^ '-^ '-^ '-^ '-^ t- tn :-" ^" il, u u u tc 5. a '16 Keport of Department of Animal Husbandry of thd 5lft'^C5«OT}'C5M'-'M^L':'>*MNe'1M©00 J -5 o i CD Tt< r^' C5 :t' c^i ,-H CI ci 1-- ir:' !"! L-^ t-^ d 1-^ LO tJh' 1^ '-' c cc" '^ C5 cJ Tin © CO ■jao g -^il-KOCSc-l-^CDCClMTl-ooOOcCCO^CCiXii— C|T-i05^^C:ca05CO IJ A o < ft iz; o EH fa O H Ed W w * CS V' ^ CO S C5 OO to Ol irs -t< ■+ LO fO O CC ^f O O li^ 'H 05 t- Oi 1- fC I- tH cC fC CO CI o ©C0i-lCCaiO(N'*O^'^"^0ll0"*00'^rt"l-ajl-L-t. CO CO CD lO k^' lO ei5CDia'-lrH0Ci00-*CDM©»AC5COTt<-<*<©TiiTt<©t-l-C0'«* Tti>r3'Tti'*io>ni'<*i'*^-^'^ir5'*'ti'*iOL3Tt<^'*'*'^'*'* OMl-Ciit-t-ai05N0505t-(N05NNMC505M05CTH,-jOOOOOO'-''-'W(M'-(lMC:THi-iO r-(r-(r-liHr-IM CD GO O C5 t- (N O CO' t- rH I- IO0Q0 0000O0O5 0000COO5C0 b-t-Ci'*0©COiroOO©(MOOrJ<00©MM<0000»OTt<(McOOOa)C'*05-^05t-' u !. COcDCDkOK3CD»OCD'^CDCDCDCDCOCDi Cb iCOCO<3500lf5-*rJ<»lOt-lOb-t-lA©(NNC0©©>'5t-f0©b-rHe0 £ ©iH00C5a>©T-l<35©©r-!©T-lT-t©CO-<*T^O5C0CO<^T-JC»eOi-l(502 ft OiHpOOTMiHO'H-r-lrHr-li-lr-lTHiHr-l»HpTHT-li:^T-lO-rHT-;OS c^I^^o^c^^c^c^M(^^|^^(^^®OC)®©*0000000©0000_goOOOoO ■♦J^+J-l^-l-l+J-t->-(->+J-l->-M4J4-l+J+J-H-P+J-t-'+-l+J-'-^-t-'+-'+^+-'-»->-»-' O^T-|lMcO-*lO©t-(»(»OiHC^ = 2 O 5'o O ^ c3 O tH M t35 tH M ec O o c; :o ifi co JO'-' 00 O tH cc o o o CD CO GO -t< ■*' tZ) O rH CO CO ir; fO 0 20 00 rH O iM t-- LO ^ L'o I- in t- t- 00 i-i 1-1 o CI -o O CO CO CO CO CO CO CO POTl<-*-: c^ooco^5»ATt^TtCOiH!MCO-*LnCOt-C0050i-ld §• Cl C^ T-l r-l r-l c; t« r; rj U U ;J C t-' U U tJ* U t^ S^ tj fc* •-' 'S'gj cjcicscicicictcjcjcicsci pi4 t£< rt ^ ^ * ^ CI 1* in CO I- I- Cl CI Cl Cl g « inNN tH o S . . . • 0 CO kr3 CD ^ r-i CO rH «C0 t- t- fO 00 '^ o H O o o H «d «• COlC© 5 » « a ,• • • O " g g CO CD tH u be ^5 6 a^a ft =£ ^ g 0^ r «' c: :c lit' cc 'T C5 © ?J 3! 5S -* in Ttl (TJ I ^ l(t' C' I- S'g tH 1-1 T-l iH T-( tH tH " '^ '^ tn -E ^ 5 ? _ CO >0 CO O CI X "* ^ ig c» 02 ro S -tH C5 rC S5 S «D lo C CO w ■:; -^ ^ ^ ^ Cl si > s> So cs a 15 = i:=1s e — CD *■ tt-^ a a 3 ti c^ s » o o a in 1-0 6 1— I rH iH OS CO fO fO 04 ©10 H3 fO rH o -r I- t- ::? cn c^ fO re CI 00 in C5 00 CO 3;; d la ^ ■'T ^ ^ __. cj cj CO (5 cc" ci C5 o ^ T-l 1-1 r-l C) . 1-' CO ca th e ci d c T-H ^ 05 -tH CC 'ti cr c:: -f Lt CI CI coco CO d t- ■* i^ t~ 00 ■»*< *; =^ ^ Orfi O I- C ZS Cj (X X I- I- X I- I- ■©rf* C5 CO O e C5 C5 O X I— Oi o ( C4 a J; (• I. *^a S' o «0 I- CO T-1 X 10 tH -^ -ti CI X CO -4< o X m c:;' m 1- o ci o X X c « ??= 3 I. - rt a c -r !- C I- I- - a •= r~ X X I- Ci Ci X CI CO O CO ■* ..... (ft 00 cj C5 o ea i* TH CI CO -* ic :c I- X Ci o New York Agricultural Experiment Station. 57 In Period 1 with Cow 12 the amount of digested protein was large, about 2.6 lbs. daily, and in Period 3 it was only 1.6 lbs. The decrease in ingested protein caused a marked decrease in urine nitrogen. In Period 1 the heat value of the urine was equal to 19.1 Cal. for each gram of nitrogen present, and in Period 3, 30.9 Cal. This simply means that the less the nitro- gen compounds present, the larger the relative quantity of other substances. We must conclude, then, that the loss of energy in urine cannot be calculated from any constant factor but must be determined in every particular case. Heat Energy of the Nutrients in Several Periods. Calories in Calories in digestible digestible matter matter minus heat minus the value of Calories In heat value urine and digestible of the of methane matter. urine. lost. Total Cal- Total Cal- ories in dry ories In dry Total Cal- matter matter ories in Periods. consumed. of feces. urine. Coic 12. 1st Period,! days .... 396017.9 129879.6 11623.2 266138.3 254515.1 224S29.3 2d Period, 7 days 399765.8 136032.9 10S81.2 263642.9 252761.7 222024.4 3d Period, 8 days .... 45032G.2 162177.0 105S4.0 288149.2 277565.2 242683.3 Goto 10. Feb. 4-10, 10 days .... 448583.6 127231.3 14525.1 321352.3 306827.2 273199.8 We have now reached a point in the development of our data where we can see what they teach concerning the unit heat values of the nutrients utilized by these two animals. From the figures of the two preceding tables the unit values given in the succeeding table were calculated. Heat Values Based on a Unit gf^One Gram. Heat value dry matter of ration. Heat valuo dry matter feces. Heat value digestible dry matter. Heat value digestible organic matter. Heat value digestible organic matter less heat value urine. Heat value digestible organic matter less less heat values urine and methane. Coio 12. Cal Cal. Cal. Cal. Cal. Cal. 1st Period. 4.35 4.69 4.20 4.31 4.12 8.64 2d Period.. 4.31 4.77 4.11 4.23 4.05 3.56 3d Period.. 4.27 4.70 4.07 4.19 4.04 3.-53 Cow 10 4.38 4.54 4.32 4.49 4.29 3.82 Average .... 4.33 4.67 4.17 4.30 4.12 3.64 58 Report of Department of Animal Husbandry of the It is noteworthy, as lias been pointed out by other observers, that the dry matter of the feces has a unit heat value consider- ably larger than that of the total dry matter of the ration and consequently the heat value of the digested portion should not be assumed on the basis of the proportion of the dry matter digested. In computing the energy value of rations it has been custom- ary to use the figures proposed by Rubner for the several classes of, nutrients, viz.: Protein 4.1 Cal., carbohydrates 4.1 Cal. fats 9.3 Cal. It appears now that these unit values are, if any- thing, fully high enough for herbivora, even if no allowance is made for loss of methane. Let us compare the actual heat values of the digestible matter in the several periods with what the calculated values would be on the Rubner basis. Cow 12. First Period. Protein S161.3 grams x 4.1= 33,4G1.3 Cal. Carbohydrates 52r;.j8.2 grams x 4.1=218, 7G8.6 Cal. Fats ISO. 3 grams X 9.3= l,G76.8Cal. 253,906.7 CaL As determined* =254, 515 . 1 Cal. Seco)id Period. Protein 6852.5 grains x 4.1= 28,095.2 Cal. Carbohydrates 55255.1 grams x 4.1=226,545.0 Cal. Fats 216.9 grams X 9.3= 2,017.2Cal. 256.658.3 Cal. As determined* =252,701.7 Cal. TJiird Period. Potein 5763.7 grains X 4.1= 23,631.2 Cal. Carbohydrates 62098.9 grains x 4.1=258,065.5 CaL Fats 258.5 grains X 9.3= 2.404.0Cal. 284, 100.7 Cal. As determined* =277,565.0 Cal. Cow 10. Protein 8004.3 grams x 4.1= 33.063.6 Cal. Carbohydrates 60445.7 grams x 4.1=247,627.4 Cal. Fats 2904.1 grams x 9.3= 27,008.1 Cal. 807,809.1 Cal. As determined* =306,827.2 Cal. ►No allowance for methane. New York Agricultural Exferiment Station. 59 If no allowance is made for the energy loss in methane the Rubner unit values are not greatly larger than the average of these determinations. If, however, the loss from the escape of hydrocarbons is as large as that observed by Kellner and others then the available energy of the digestible organic matter was 3.64 Cal. per gram. Kellner found it to be 3.5 Cal. for the digestible organic matter of meadow hay, and Armsby, in deter- minations with steers, using timothy hay, found it to be 3.62 Cal. THE DISTRIBUTION OF THE ENERGY OF A RATION. it is interesting and instructive to note the distribution of the energv of a ration. It mav be classified, first as available and non-available. The available energy is applied to the various uses of the animal and in the case of the milch cow is distri- buted among the milk solids, the work of milk production and maintenance needs. These energy relations are best expressed in terms of per- centages without repeating any previous figures. Distribution of Total Energy of Ration. Methane i Available energy "iXS ]No.-aval,abIe energy. !™« Assumed f Available energy methane J ^ Urine Per. 1. Per ct. . 64.3 . 2.0 . o2.S . .JH.S 2.9 energy deducted ..i Xon-available energy. < Methane. 7.5 ^I eces 32.8 Cow 12. Per. 2. Per ct. G.3.2 2.7 .34.0 .5.1.. 5 2.7 7.7 34.0 Per. 3. Per ct. 61.6 2.4 36.0 53.9 2.4 7.7 36.0 Cow 10. Per ct. 68.4 3.2 28.4 60.9 3.2 7.5 28.4 Co^D 12. Period 1. .. Period 2. ., Period 3... Cow 10 , Distribution of Available Energy. Maintenance energy.t Energy of milk solids. Total available energy used daily.* Cal 32118 31718 30335 27320 Calories dail.v. 13S46 13846 33S46 10152 Per ct of total available energy. 43.1 43.6 45.3 Calories daily. 11176 10169 10.547 8451 Per ct. of total available energy. 34.8 32.1 30.4 80.9 Energy balance. Per ct. of total Calories available daily. energy. 7090 7703 5942 6717 22.1 24.3 24.3 32.1 ♦After allovring for loss of energy in methane. t Calculated on basis of 13,000 Cal. for 500 kilo animal. 60 Report of Department of Animal IIu5BA^DKY. The foregoing figures are instructive in showing approxi- mately how the energy of the rations was utilized bv these two milch cow^s. After accounting for the energy of maintenance on the basis of the best knowm data, viz: 13,000 Calories for an animal weighing 500 kilos (1100 lbs.) and the energy of the milk solids, we have a balance amounting on the average to about one-quarter of the available energy of the ration. If the ration were diminished to the extent of this balance it would certainly result in a lessened milk yield, consequently we are justified in concluding that this balance has some necessary function or relation in milk secretion. The most natural and logical conclusion is that in part at least it sustains the work of milk secretion, i. p., the vital activity involved in the metabolic changes occurring in the milk glands or elsewhere in the formation of milk solids. There is, of course,*more work demanded for mastication and digestion than is the case with the much smaller maintenance ration. Nevertheless it is fair to regard the milch cow as a working animal, not in the exercise of mechanical force but in the main- tenance of manufacturing processes which are sustained by the application of the quiet energies of life. THE IMMEDIATE EFFECT ON MILK PRODUC- TION OF CHANGES IN THE RATION * W. p. WHEELER. INTRODUCTOKY ^^OTE BY THE DIRECTOE. The data discussed in this bulletin were mostly secured under a former administration. They are not the result of feeding experiments logically planned for the study of rations, but were incidental to the extensive breed tests that were carried on for several years, and for this reason they cannot be made the basis of so extensive or so safe conclusions as to feeding questions as otherwise might be the case. The tables which Mr. Wheeler presents from nearly one thou- sand feeding periods with different animals may seem to be duplications in some cases, as for instance V, IX. X of the tables as compared with XXVI, XXV and XXVII. It is to be noticed that these tables differ as to the number of periods included, which is due to the fact that in one set the selection of data was made with reference to the effect of varying the quantity of nutrients and in the other set with reference to the influence of varying the protein. It is gratifying to note, however, that different groups of periods, involving practically similar condi- tions, indicate essentially similar conclusions. Those who may study this bulletin should keep clearly in mind that only the immediate and not the permanent effect of rations is considered. Moreover, there should be a proper reser- vation of judgment concerning data logically so imperfect. Nevertheless, in some particulars these records mean much more in a practical way than some offered to the public which involve the use of very few animals during only two or three feeding *A reprint of Bulletin No. 210. 62 Report of Department of Animal Husbandry of the periods. The evidence presented concerning one point is fairly consistent and is important, viz.: that changes in the quantity of nutrients have greatly more influence on the milk yield than proportionally large changes in the amount of protein. These data offer strong evidence that if the available energy of the ration is sufficient and is kept at a uniform point, there may be quite a wide range in the nutritive ratio without materially affecting the milk flow. This emphasizes the necessity of uni- form feeding and the importance of knowing to what extent changes in the materials of a ration cause changes in the nutrients actively available. The evidence relative to the pro- tein supply at least suggests the practicability of successfully feeding dairy cows from a w^ell selected list of crops grown on the farm. W. H. Jordan, Director. SUMMARY. The data published in this bulletin show the changes in milk production which have immediately followed changes in the composition of the ration fed to cows. The eflieiency of differ- ent rations for sustaining milk production over long periods is not shown. Several hundred individual records (981 in all) for limited periods covering different changes in the rations were selected from those kept for a dairy herd during several years. These were grouped in accordance with certain relations which they held to factors of the recognized feeding standards, and aver- ages were made. The average data considered are those which relate to changes in the amount of total digestible organic mat- ter or '' total nutrients," the fuel value, the amount of protein, and the nutritive ratio. Xo rations that would appear in any respect radically deficient were fed. TOTAL NUTRIENTS. In general, the milk flow increased most or diminished least when the greatest increase of total nutrients was made without regard to moderate changes in protein content. The most rapid New York Agricultural Experiment Station. 63 shrinkage of milk flow generally occurred when the percentage reduction of total nutrients was greatest, although this usually was associated with a reduction of protein. On the average for all records when an increase of the total nutrients was made, there was no change in milk production. On the average for all when the amount of nutrients was reduced, the shrinkage in milk flow was at twice the normal rate. An increase in the amount of total nutrients to not more than 15.5 lbs. per day for each 1000 lbs. live weight, with cows giving about 20 lbs. of milk, resulted in a maintenance of the milk yield without diminution. When the nutrients were reduced in corresponding rations, more than the usual shrinkage followed. An increase of the total nutrients from less than 15.5 lbs. to more than that amount, for cows giving about 23 lbs. of milk, resulted in a maintenance of milk yield without diminution. A reduction of the nutrients from above 15.5 lbs. to less than that amount was followed by twice the normal shrinkage in milk yield. An increase of the total nutrients when above 15.5 lbs. for cows giving about 24 lbs. of milk, was followed by less than the usual shrinkage in milk flow. A reduction of the total nutrients to not less than 15.5 lbs. was followed by more than the usual shrinkage. FUEL VALUE. An average of all records when an increase in the fuel value of the ration was made shows a diminution in milk yield about one-fifth as great as would usually occur under unchanged rations. An average of all records when the fuel value was re- duced shows about twice the usual diminution in milk yield. An increase in the fuel value to not more than 30,000 Cal. per day per 1000 lbs. live weight, for cows giving about 20 lbs. of milk, was followed by a slight average increase in milk flow. A reduction of the fuel value when below 30,000 Cal. was followed bv considerablv more than the normal shrinkage in milk. 64 Keport of Department of Animal Husbandry of the An increase in the fuel value from below 30,000 Cal. to more than that amount, for cows giving about 22 lbs. of milk, was followed by about the usual shrinkage in milk flow or less. A reduction of the fuel value from above to below 30,000 Cal. was followed by twice the usual shrinkage. An increase in the fuel value when above 30,000 Cal., for cows giving about 24 lbs. of milk, resulted in a maintenance of the milk yield without shrinkage. A reduction of the fuel value to not less than 30,000 Cal., was followed by about twice the usual shrinkage in milk. PROTEIN. In general, changes in the amount of protAi within ordinary limits produced less effect than changes in the amount of total nutrients. On the whole the diminution in milk flow was less when the amount of protein was increased than when it was re- duced. On the average for all records when the protein was increased, those including also an increase of total nutrients show no fall- ing oft' in milk production, those with but little change in nutrients show a normal diminution or less, those with a reduc- tion of nutrients show a shrinkage greater than usual. On the average for all records when the protein was reduced those with an increase of total nutrients show less than the usual decrease in milk production, those with but little change of nutrients show about the normal shrinkage, those with a reduc- tion of nutrients show a falling off at twice the normal rate. The average of those records where there was an increase of protein without change in amount of total nutrients shows an increased cost of production. There was no increase in the cost of production, on the average, when the protein was reduced without change in amount of nutrients. A reduction in the amount of protein, when the ration con- tained less than 2 lbs. per day per 1000 lbs. live weight, was fol- lowed by about twice the usual shrinkage in milk flow. An in- crease in the amount of protein when less than 2 lbs. was, in general, followed by less than the usual shrinkage. Kew York Agricultural Expeiriment Station. 65 A reduction in the amount of protein when between 2 lbs. and 2.25 lbs. was followed by more than the usual shrinkage in milk flow. An increase of the protein when between 2 and 2.25 lbs. was followed, on the whole, by less than the usual shrinkage. While reduction of the protein when between 2.25 lbs. and 2.5 lbs. was followed by excessive diminution in milk flow, there was an accompanying reduction of total nutrients which would largely account for it. An increase in the amount of protein when between 2.25 and 2.5 lbs. was followed by very much less than the usual shrinkage of milk. A reduction in the amount of protein when above 2.5 lbs. was followed by favorable results on the average for 123 records. The natural shrinkage in milk flow was retarded. The cost of production was not aifected. An increase of the protein when above 2.5 lbs. is shown by only a few records — in these, however, there followed twice the usual shrinkage in milk flow. NUTRITIVE! RATIO. Changes in the nutritive ratio within the ordinary limits had considerably less influence on the milk flow than did changes in the amount of total nutrients. In general, however, a nar- rowing of the ratio had a favorable effect on milk production, while a widening of the ratio tended toward the reverse. When but little change in the amount of total nutrients oc- curred, a narrowing of the ratio was followed by less than the usual decrease in milk yield and a widening of the ratio by more than the usual decrease. With an increase in the amount of total nutrients a narrowing of the ratio was followed by an increase in milk yield. A widen- ing of the ratio was followed by a decrease (to less than the usual extent) although the average increase of total nutrients was nearly a pound greater than when the ratio was narrowed. With a reduction in the amount of total nutrients of 10 or 12 per ct., there followed about the same average shrinkage in milk flow whether the ratio was narrowed or widened. 5 66 Kei'ort of Department of Animal Husbandry of the When the nutritive ratio was narrowed but kept wider than 1:6, no change occurred in the average amount uf digestible di'y matter consumed for each pound of milk solids produced. When the ratio was made wider there was an increase (over 4 per ct.) in amount of digestible dry matter for each pound of milk solids. When rations with a narrower ratio than 1:6 were made si ill narrower there was the same increase in the amount of digesti- ble dry matter required for each pound of milk solids as when corresponding rations were made wider but kept narrower than 1:6. It must be remembered that these summarized results apply only to the immediate effect on milk production of the specified changes in the ration. It is not unreasonable to assume, how- ever, that those modifiations of the ration which at once lead to increase in milk flow point toward the composition of a ration adapted to more permanent advantage, and that those modifl cations which are immediately followed by diminished product point in the direction of a ration more likely to prove inefficient for extended periods. INTRODUCTION. In the feeding of milch cows as well a& that of other animals many difficult problems are involved. The solution of most of these cannot be accomplished without numerous investigations and studies under specially arranged conditions where factors commonly uncontrolled can be directly accounted for and their value considered. Many years must necessarily elapse before much of the positive knowledge sought can be secured. This knowledge will come in time. But every day the animals must have food, and any information relating to the commonly prac- ticed methods of feeding is worth considering. Any carefully collected data should, therefore, repay attention; though they may not light up certain of the complex problems of nutrition, and may show only in a circumstantial way relations between the milk and the food. While the fixing of an absolute standard stated in the terms we now use is not possible, it is still probable that with wider New York Agricultural Experiment Station. 67 information the limitations of the average physiological stand- ard can be made more definite. All records of production under rations constructed In conformity with standard requirements should furnish information of more or less value. The results of departure, on one side or the other of the path marked lout, should be worth recording. The data published in this bulletin show changes in milk pro- duction which have immediately followed changes in the com- position of the ration. Some changes in the ration appear to have no immediate effect on milk production. Others apparently have in a moderate degree either stimulating or depressing effects. For many years rations conforming to certain limitations have been recommended by different investigators. The most conven- ient and practicable way of stating the food requirements of the animal in a concise and general manner has been in terms of digestible nutrients. As ordinarily grouped, these are protein, carbohydrates and fat. The total digestible organic matter is stated as well as the total dry matter. Water is taken for granted, and the mineral nutrients, although important, are not considered because the foods naturally available for supplying other nutrients would contain abundant ash. In formulating a feeding standard for cows considerable differ- ence of opinion exists as to the limits of variation, both in bulk and composition, that can be made without appreciable influence on production. The standard rations usually recommended for milch cows of average capacity require, for 1000 lbs. live weight: From 27 to 29 lbs. of total dry matter, From 13.4 to 16 lbs. of didgestible organic matter. From 2 to 2.5 lbs. of digestible protein, From 11 to 13 lbs. of digestible carbohydrates. From .4 to .5 lb. of digestible fats; with a nutritive ratio of from 1 :6 to 1 :5.7. For cows not approxi- mating 1000 lbs. in weight, for cows of inferior capacity or for very heavy milkers, the standard is modified. The standard 68 Report of Department of Animal Husbandry of the which has been longest used as a general guide gives for the average cow in milk, per 1000 lbs. live weight 24 lbs. total orgauic matter, 15.4 lbs. total digestible organic matter, 2.5 lbs. total digestible pi'otein, 12.5 lbs. total digestible carbohydrates; with a nutritive ratio of about 1:5.4 and a fuel value of about 30,000 Calories. These standards have been in large part drawn from averages. Slight individual modifications have therefore been occasionally desirable or without disadvantage. In practice, of course a close conformity may often be unprofitable because of relative market prices. The efficiency, however, of the modified standard ration is the point for first consideration. While the results of only a few weeks under a ration will not show its permanent sustaining power, they may perhaps suggest the stimulating effects. To these are often attributed results not fully explained by the knov/n utilization of nutrients in the product and the work of its construction. It is thought that the presence of a supply of protein considerably in excess of the indispensable amount which can be directly accounted for, tends to stimulate milk production. In feeding a dairy herd at this Station during several years the common standards were followed in a general way. From the daily records of this herd are drawn the data before referred to. They represent both winter and summer feeding at different times during seven years. Individual records were kept of the food consumed, the changes in live weight and the milk yield. Analyses were made of the foods and of the milk at regular inter vals so that the amounts of the dift'erent constituents in the food and in the milk have been calculated for each cow and averages made from these data. The records for short periods selected for use in these averages were taken at that stage of lactation when only very moderate changes in the milk flow might be expected. Of these limited-period, individual records, 981 have been used in making the averages considered here. When grouped according New York Agricultukal Expeiumknt Station. 69 to the same foods they form 111 groujis with an average of nine cows each. The food was changed often but no violent changes were made nor any rations fed that would appear in any respect radically deficient, as the primary object for keeping the herd would not have been furthered by the use of questionable rations. The change usually was only a substitution of one coarse fodder for another accompanied by a modification of the mixed grain. No unpalatable food was used. A moderate proportion of grain was always fed, varying from 5 to 9 lbs. per day, but generally about 7 lbs., the average for all the time being 6.63 lbs. per day. As a rule either silage, roots or green for{ige was fed with any hay or other dried fodder. Only 15 rations out of a tliou- sand were without some succulent food, the average moisture content of these being 12.2 per ct. The average percentage of moisture in all the rations was 61 per ct. Besides mixed grain, which was always fed, 336 rations con- sisted of silage and hay, 266 rations of green forage and hay, 76 rations of roots and hay; 72 rations contained two kinds of green forage and 47 rations one kind; 56 rations contained silage and green forage; 43 rations silage, hay and corn stover; 30 rations silage, forage and hay; 22 rations silage and corn stover; 15 rations hay alone; 10 rations silage, roots and hay, and 8 rations silage alone. The coarse foods principally used were clover hay, timothy hay, mixed hay, corn silage, alfalfa forage, oat-and-pea forage, corn forage and mangels. Others sometimes fed were oat-and-pea hay, orchard-grass hay, corn stover, barley-and-pea forage, sor- ghum forage, rye forage, rye-and-pea forage, timothy forage, clover silage, sugar beets and carrots. The grain foods most commonly used were wheat bran, corn meal, ground oats, wheat middlings, old process linseed meal, cottonseed meal, different gluten meals and gluten feed. Others occasionally used were new process linseed meal, brewers' grains, ground flaxseed, buck- wheat middlings and malt sprouts. In a grain mixture three kinds of ground feed were always used and generally more. 70 Report of Department of Animal Husbandry of the) Special analyses were made of all the foods. The coefiScionts of digestibility used in the calculations wore the averages from American determinations when several were to be found, but many of them were from European data. These calculations were made several years ago when factors from recent investiga- tions were not available. The coefficients used differ somewhat from the latest averages but not enough to have a recalculation affect the general results. In Table A are given those used for the several foods. Selecting records from a system of feeding conforming to other purposes did not permit an absolute uniformity in all relations essential to a direct and unqualified comparison of the different rations. The averaged groupings are therefore not by any means satisfactorily complete and full, although the volume of testimony in a degree compensates for some of the lack of uniformity. While a large number of records may cover one change in a ration, there may be comparatively few that afford data con- cerning some corresponding change, desired for comparison, so that in this respect also undesirable irregularity exists. Evi- dence only of the immediate, or perhaps stimulating, effect of the modified rations is offered. There is taken into consider- ation no intermediate period to permit full adjustment to the changed food, and the efficiency of the different rations for sus- tained milk production over longer periods is not shown. The record is taken for from two to four weeks before a change in a ration and for about the same time immediately following. Table A. — Average Coefficients of Digestibility. ProtPin. Whoali bran 78. Cornmcal ■ 76. Ground oats 78. Wheat middlings 82. O. P. linsoed meal 86. Cottonseed meal 89. Gluten meal 87. Gluten feed 86. N. P. linsood nioal S7. Brewers' grains 71. Fiber. N. free extract. Fats (ether extract). 33. 72. 76. 58. 87. 92. 26. 77. 84. 33. 88. s.-). 50. 80. 90. 33. 68. 95. 33. 91. 88. 34. 92. 82. 61. 86. 91. 46. 55. 8G. New York Agricultural Experiment Station, 71 N. free Fats (ether Protein. Fiber. extract, extract). Ground flaxseed 91. GO. 55. 86. Buckwheat middlings 78. 35. 75. 85. Malt sprouts S3. 41. 67. 90. Glover hay 58. 54. 64. 55. Timothy hay 49. 53. 63. 57. Mixed hay 42. 49. 57. 54. Corn silage 57. 69. 75. 84. Alfalfa forage 67. 53. 78. 64. Oat-and-pea forage ; . 78 . 57 . 65 . 71 . Corn forage 52. 52. 75. 77. Mangels 83. 71. 95. 50. Orchard-grass hny 59. 60. 55. 54. Oat-and-pea hay 81. 57. 66. 74. Corn stover 36. 64. 58. 70. Sorghum forage 47 . 59. 74 . 74 . Rye forage 74. 83. 73. 67. Clover silage 48. 60. 70. 67. Beets and carrots 84. 80. 95. 77. As a basis for comparison, the normal rate of shrinkage in millj; flow that would usually occur under an efficient average unchanged ration was assumed to be, for every period of one-half month, about 2 per ct. during the third month of lactation, about 2.5 per ct. during the fourth and fifth months, about 3 per ct. during the sixth and seventh months, about 3.5 per ct. during the eighth month and about 4 per ct. during the ninth and tenth mouths. In making this estimate, data from this Station were considered in connection with some from the Cornell Experi- ment Station, and some records published a number of years ago by Dr. Sturtevant. The amount fed was always about all that could apparently be used by the animal. As a rule there was a gain in live weight which indicates a usually sufficient amount of food. The cows were mostly of medium size and less than 1,000 lbs. in weight. The average weight per cow for all periods was 932 lbs. An idea of the prevailing composition of the rations and of their efficiency is given by the following data: Under 1 is the average from 90 individual records for one month with rations having a fuel value of considerably less than 30,000 Cal., and supplying less than 15.5 lbs. of total digestible organic matter. 9 5.6 967 4.5 25 33,937 17.0 2.3 1:7.2 25.9 3.4 4.0 72 Report of Depaktmext of Animal, Husbandry of the Under 2 is the average from 80 individual records for one month with rations having a fuel value considerably higher than 30,000 Cal., and supplying considerably more than 15.5 lbs. of total digestible organic matter. Table B. — Averages from Records for One Month. 1 Average age of cows, years 4 Average live weight per cow, pounds 894 Average month of lactation 6.2 Average gain in live weight for month, pounds 12 Fuel value of ration per KKX) lbs. live weiglit, Gal 28,855 Total digestible organic nutrients per 1000 lbs. live weight poiuids 14.6 Digestible protein per 100 lbs. live weight, pounds. .. . 2.1 Nutritive ratio 1 :6 . 6 Milk yield, average per day per cow, pounds 20.2 Total solids in milk, average per day per cow, pounds 2.8 Percentage of fat in milk 4.4 Digestible dry matter in food for one pound of milk solids produced, pounds 4.6 4.8 As the great mass of individual data is of no general interest, only the averages from different groupings are given. The cost of production is usually stated. While the more nitrogenous foods, especially grain foods, are usually higher priced, the lib- eral use of such foods as alfalfa forage, oat-and-pea forage, clover hay, etc., has somewhat modified this relation. The cost relation is an uncertain and fluctuating one, of course, but the prices assumed for calculation would fairly represent the aver- age cost of foods supplying rations of the stated composition. Hay was rated at -$10 per ton, and corn stover at |5, silage and roots were rated at |3, green fodders at |2. Corn meal, wheat middlings and brewers' grains were rated at $20 per ton, wheat bran, malt sprouts and gluten feed at .$18, ground oats, gluten meal and new process linseed meal at $25, old process linseed meal at |27, cottonseed meal at |30, and ground flaxseed, but little used, $60 per ton. The average food cost of milk for all the records considered was 73 cents per 100 lbs. Xeav York Agricultural Experimbnt Station. 73 TOTAL ORGANIC NUTRIENTS. Groupings of the different records according to the amount of the total digestible organic matter are first considered. The term " total nutrients " or " nutrients," whenever used without (lualification in the text or tables, refers to the total digestible organic matter of the feeding standards without reference to the mineral matter. The term " dry matter," however, includes the ash. CHANGE IN RATIONS WITHOUT CHANGE EST NTJTRIEINTS. The average from 126 individual records in which very little change was made in the amount of nutrients — no change on the average — shows just about the shrinkage in milk flow that would normally occur at tliis stage of lactation. The average month of lactation was 6.2 and the average age of the cows 4.8 years. There was a moderate gain in weight, the rate tending to increase after the change of ration. Table I. — A Change in Ration but Not in Amount of Total Nutrients. Average of 126 rec- ords from cows averaging 4.8 yis. old and 6.2 months in milk. . ■—1 cS cS T3 C a Before After Per 1000 Lbs. Livk Weight. Total digest- ible or- ganic matter. Lbs. 14.6 14.7 Fat !n milk. Per ct. 4.3 4.2 Digest- ible pro- tein. Lbs. 2.05 2.04 Fuel value. Cat 29,072 29,209 Nutri- tive ratio. 1:7.0 1:7.0 Average Per Dav Per Cow. Milk yield. Lbs. 20.2 19.9 Fat In milk. Lbs. .87 .83 Approxi- mate cost of food for one lb. or milk. Cts. .75 .75 Dry matter In food for one lb. of milk. Lbs. 1.0 1.0 Dry matter in food for one lb. of milk solids. Lbs. 7.5 7.7 Digest- ible dry matter In food for each lb. of milk solids. Lbs. 4.9 5.1 AN INCREASE IN THE AMOU'NT OF NUTRIENTS WITH BUT LITTLE CHANGE IN PROTEIN. The average of 73 records which cover periods when this change in the ration was made is shown in the following table. No shrinkage in milk flow followed, on the average, but instead a very slight increase. There was a little loss in live weight before the change and a good average rate of gain afterward. New York Agricultural Experiment Station. 81 Table IX.— The Amount of Total Nutrients Increased, "with Little Change of Protein. Per 1000 Lbs. Live Weight. Average Per Day Per Cow. Total digest- ible or- ganic matter. Digest- ible pro- tein. Fuel value. Nutri- tive ratio. Milk yield. Fat in milk. §D Before a ^ After 2 a ^ 00 Oi >j .^ '^ 1" f'S Before ^ After Lbs. 13.9 15.9 Lbs. 2.04 2.06 Cal. 27,985 32,575 1:6.4 1:7.7 Lbs. 22.2 22.3 Lbs. .91 .90 Average of 73 rec- ords from cows averaiiiug 4.7 yrs. oldand4.6mouths in milk. Fat In milk. Approxi- mate cost of food for one lb. of milk. Dry matter In food for one lb. of milk. Dry matter lu food for one lb. of milk solids. Digest- ible dry matter in food for each lb of milk solids. -- Per ct. 4.1 4.0 Cts. .70 .72 Lbs. .8 1.0 Lbs. 6.5 7.5 Lbs. 4.3 4.9 AN INCREASE IN AMOUNT OP NUTRIENTS AND ALSO OP THE PROTEIN. Tlie average from 132 records, each of which covers periods when this change in the ration was made, gives the data of the following table. Less than the usual diminution in milk flow followed the change. There was but a very moderate increase in live weight under either ration. 6 82 Report of Departiment of Animal Husbandry of thb Tabi,e X.— Thk Amount ok Total Nutrients Increasp:0 and also thk Protkin. ® a Before a After Avernrrc of 132 rec- o eoicls fioiM cons ii ve r :i ^ i nj; 4.6 .vr.->. old iiud 4.9 mouths iu milk. t- aj |2 etf — Before Q 03 After Per 1000 Lbs. Live Weight. Total digest- ible or- gan lo matter. Lbs. 14.5 16.0 Fat In milk. Per ct. 4.0 4.0 Dieest- ible pro- tein. Lbs. 2.03 2.44 P'liel value. Cat 28,«33 31,863 Nutri- tive ratio. 1:6.9 1:7.0 average Per Day Per Cow. Milk yield. Lbs. 22.4 22.1 Fat In milk. Lbs. .90 .89 Approxi- Dry mate co.it matter o( food In food for one for one lb. of lb. of milk. milk. Cts. Lbs. .70 .9 .71 1.0 Dry matter in food for one lb. of milk solids. Lijs. 7.1 7.8 Digest- ible dry matter in fooii for each lb of milk solids. Lbs. 4.5 5.1 AN INCREASE IN THE AMOUNT OF NUTRIENTS WITiI A RBDUCTON OF THE PROTEIN. Tlio .iverage of G3 records when this change in tlie ration occurred shows no shrinkage in the milk flow but a very slight increase. There was an average gain in weight of about one pound per day after the change, about twice as much as pre- ceded it. Nbw York Agkicultukal. Expe'Rimbint Station. 83 Table XI. —The Amount vv I'otal Nutkiknts Inorkased and the Protkin Reduced. Per 1000 Lbs. Live Weioht. AVERAGE Per Day Per Cow. Total digest- ible or- ganic matter. Digest- ible pro- tein. Fuel value. Nutri- tive ratio. Milk yield. Fat in milk. <» ^ Before eS <^ After 03 =2 o SI >. a, ~ H " c3 00 CO -I ^, ^-^ Before pti After Lbs. 14.7 16.5 Lbs. 2.68 2.22 Cal. 29,318 32,882 1:5.2 1:7.0 Lbs. 21.1 21.2 Lbs. .86 .88 Average of 63 rec- ords from cows a V e r a g i u <^ 5 yrs. old aud 5.9 months in milk. Fat in milk. Approxi- mate cost of food for one lb. of milk. Dry matter in food for one lb of milk. Dry matter in food for one lb. of milk solids. Dlge.st- Ibledry niattei' In food for each lb. of milk solids. Per ct. 4.1 4.2 Cts. .72 .74 Lbs. 1.0 1.1 Lbs. 7.8 8.4 Lbs. 5.0 5.5 AN INCREASE IN AMOUNT OF NUTRIENTS TO NOT MORE THAN 15.5 LBS. The average of 111 records, in each of which there was an increase in the amount of nutrients to not exceeding 15.5 lbs. per day per 1000 lbs. live weight, is shown in the following table. The average increase in nutrients was slight but the usual shrinkage in milk flow did not occur. In 61 instances there was a decided increase of protein in the ration, in 24 instances a decided reduction and in 26 instances little change. There was but slight average gain in live weight. With but little change in amount of protein. — The average of the 26 records in which little change in the amount of protein occurred shows a trifle less than the usual diminution of milk yield expected at this stage of lactation. The average increase in amount of nutrients was but little. A slight loss in live weight occurred. See A in Table XII. With an increase in the amount of protein. — Sixty-one records show a slightly greater average yield of milk following the mod- 84 Report of Department of Animal Husbandry of the O^ a ° cs o 5 So.. .'5 ' O" oj, a Ok-) CO i-- I- o CO CO cO ■B Q.O . £a Too §■23 ^05:3 00 com CO T-l e^^ 0000 ^3 • • OiOi C3 Ift CO CO 05 h-OO IMCS O T-l Q C^W t-C5 IN IN -t< CO 05 CO r-l IN 05(» CO ^' 0) O Q) 00 S TlKCO IN O CCCO QOOO CO 10 Tji CO coc» C5 ira I- IN CO CO IN IN ■-I 05 CO o csiiN cO(N CO -^ n o <^ ^ _ CJ t» ^ .9 ^ ej o wg CO aj -*^ '♦-' it t>i a o g 3 fcfj g CO CO S O U co^i:.s n Kkw York Acjiiicultuual Exfehumbnt Station. 85 erate increase of the total uutrients and more pronounced in- crease of protein. There was practically no change in live weight on the average. See B. Table XII. With reduction of the proteim,. — In C, Table XII, are the data from the 24 records when with a little increase of the total nutrients there was considerable reduction of the protein. But very slight shrinkage in the milk flow occurred. There was a moderate rate of gain in weight before the change and a more rapid one afterward. AN INCREASE OP TOTAL NUTRIENTS FROM LESS THAN 15.5 LBS. TO MORE THAN 15.5 LBS. In the following table are the average data from IIG records when this increase in the ration was made. The average milk 3'ield suffered none of the usual diminution. There was a slight average loss in live weight before the change of ration and a gain of over one pound per day per cow afterward. With hut little change in amount of protein. — The average of 47 records when with the above increase of nutrients there was but little change in protein is shown in A of Table XIII. Con- siderable increase in milk yield followed. There was consider- able loss in live weight before the change and an average gain of about two pounds per day afterward. With an increase of the protein. — The average data from 43 records ai'e given in B of the same table. In each of these there was an increase in the amount of protein averaging about 20 per ct. The shrinkage in milk flow slightly exceeded the normal rate. There was a loss in live weight before the change of ration and a good rate of grain afterward. With a reduction of the protein. — The average of 26 records in each of which the amount of protein was reduced is given in C of the table. While the amount of digestible protein p«M- day was redu(;ed about one-half pound, the average amount did not fall below 21 lbs. An increase in milk flow followed the change and but slight increase in the cost of production. There was a good rate of gain in live weight. 86 Report of Dki'autmkxt of Animal Husbandry of the •-si^s-^ <0 '- C3 — — — •=> 0« O « S O >-J a o p ^ ^ £■" o o ce o » ^ c o a i«5 oi 050 k0 05 do CO O CD l> OO CO O COCO OtH a,= c o 2 S o r o o a '" "^ COO I- t- I- L- 1-1 o O r-1 CO -^ ao Mi- . ; ^ "r: CO 00 1_Q t--; 2 1-1 T-l 1-1 iH iH ^^ i^' r-( ® SSi lO o CO c^' -I^M O 3 00 OO o ?* TfiTti ^■3 H t- t- coo rH I- COO s,^ 5 M CO CO lO M CO ^'^' M CO i 6 . « r; a >oeo CD CO «=!:;; MM e3 CM CO MM T-1M M M oiM - r i. '-< r, »0 iH CO O rilO no +j o a IS p 6oii « cj i -^ l- -rji t— ^O -ti t- >^ T-l rH 1-1 t-i r^ '^ r-l T-l a; Ol <3J o '-' t~i ^ (U tl u '-• t-l o S o S o i «-^ <*-! ?^'S ., '^ o O '^^ a =^ >^, «. a O «-l K'' P » C B\Q § ^ 2ii a C(M < 2 s|« (M(N IN C^ c5^ » h-l »Ob- t-»o oo t?s d d «o^ cpo 3 1=2 A5 iH rH T-l iH A^ S ■* •* O© — «5 -J 2 3 e r-l 05 ;5 0) S| ^ 1-1 -rt* rH OO ^"^ tj o 0) o (—1 o S ri tM § a ^ , » bCQO c; ?s « ^ ^^ i_. CO ' - ^^ ^ s (- ci o a .9 ?5 =«»■ •■t-l ? ki -t— ' ?5 'o O "^ a ® O, ^ *^ 15 g o o- '~ 2 o o . (/J a o s •< ^ >® "^ a 00 ? rH t>» 00 "'g Before fa After Lbs. 15.4 14.4 Lbs. 2.12 2.05 Cal. 30,628 28,530 1:6.9 1:6.7 Lbs. 21.3 20.6 Lbs. .93 .90 cVverajre of 82 rec- ords from cows a V e r aff i n g 4.7 yis. old and 5.7 months in milk. Fat In milk. Approxi- mate cost of food for one lb. of milk. Dry matter In food for one lb. of milk. Dry matter In food for one lb. of milk solids. Digest- ible dry matter 1 u food for each lb. of milk solids. Per ot. 4.4 4.4 Cts. .73 .74 Lbs. 1.0 1.0 Lbs. 7.7 7.3 Lbs. 5.0 4.9 A RBDU'CTION IN AMOUNT OF TOTAL NUTRIENTS WITH AN INCREASE OF PROTEIN. The following table shows the average from 72 records when this change in the ration occurred. The falling off in milk yield was considerably greater than is usual. The rate of gain in live weight was about three-fourths pound per day before the change and less than half as fast afterward. There was an increase in cost of milk production. 90 Rkport op Dbpahtmbnt of Animal Husbandry of the Table XVI. — Thk Amount of Nutrients Reduckd with an Increase of Pkotein. Average of 72 rec- ords from cows averaging 5.2 yrs. old and o.3 mouths in milk. to a o ,-1 =0 ^-3 Before Alter I- « -3 Before rt Eti After Pbb 1000 Lbs. Livb Weight. Total digest- ible or- ganic matter. Lbs. 1.16 14.6 Fattn milk. Per ct. 4.0 4.1 Dlpest- Ible pro- tein. Lbs. 2.10 2.54 Fuel value. Cal. 32,494 29,02.3 Nutri- tive ratio. 1:7.5 1:5.2 AVHRAGE Per Day Per Cow. Milk yield. Lbs 22.6 21.6 Fat In milk. 7.6s. .90 .88 Approxi- Dry mate eost matter of food in food for one for one lb of lb. ot milk. milk. Cts. Lbs. .70 1.0 .78 1.0 Dry matter In food for one lb. of milk eollds. illspsf- ible dr.y matter In food for ca.-h II). of milk solids. Lbs. 7.8 7-5 Lis. 4.9 A REDUCTION IN AMOUNT OF TOTAL NUTRIENTS AND OF PROTEIN. Ill Table XVII are the average data from 109 records, eacli of which covered periods when this reduction was made. The diminution in millc vield was about twice as much as woukl uor- mally occur under ample unchanged rations. The rate of gain in live weight, which was over one-half pound per day, became very slow after the change of ration. New York Agricultural Experiment Station. 91 Table XVII.— The Amount op Nutrients Reduced and Also the Protein. Average of 100 rec- ords fioni tows averaginj; 4.8 yrs. uld and r>.d mouths ill milk. to a o s> t-l o 3 0» 00 t-^ eooj r-Kft te^ (3 a. •^'^ o P3 la H H P ij O H o !z; O n EH Hi •a n Si Sa S^3 > o ~ (B a> 03« ,H;5a»a 3? 1^ CO 06 t-o t- (N CO CD CO r-i 05 CO CO ro C3 I- IM CI lO (N f— 1 r-t 00 C3 "* CO ai O a; OS 00 eoos CO co" CO o O CO tH (N CO 1-1 oo (Nci t-05 TjHCO O (U $2« cot- CJiH OS 05 OSN coo CO ^^ Ol- CO C^l 00 CO t-l (N O tH iO Ttl O m l^l^ TM r-t 00 t- lO CO -f CI en I- CI c^ lO 1-1 00 t- "* CO fcci ^CD^?'2M t> CS ci o o . t; to « 5 ft) t-. ■ to (.y'T3 • tn C^ U "3 " <^ t^ c3 o ca ^ =! I- a c ." (U 'CO ■; o a, a ^ fr, cr ^ 2 a ^ t ^ M O 2 o w ^ a tOOJ ^ 53" •* CO CO -"I* »-l (N ^ ca CO M r-i I- 05 CD -^ ^ CD -Ii O fli lire CO 03 t- co OS to 5i CO ■* IN CD IN (N ooo CD -i* (X> O Oi CO o CO l- CO?! r-l 00 iN r-i OCO o a} (U A ^ e0 5m «* * • o u Q. ^ . SM 15 ga « to ooo ooo c^ En ^ en s 1 ^ 0-* 00-* ■ • b-t- • • •• .J 3 10 to «DCO CO CO coco «9 s§5 i3 CO CD oos CO CD 0) O rt) CQ<1 «D b- ^ O fT. The average data from 91 records which covered periods when little change in the amount of protein was made, although there was some increase of the total organic nutrients in the ration, are found in Table XXI. The falling off in milk yield was less than the normal amount. The rate of gain in live weight was slow but somewhat faster before the change than after. 98 Report of Department of Animal IIusbaxdry of thb Tahlk XX[. — Little Chaxgk ix the Amount oi- Protkin. Avempje of 91 rec- ords from cows ca V e r ii y; i II f; 4.5 yrs. old and 4.9 munths iu milk. Before -g After o -*^ ~z ~ ■— ( '^ n Before u 5 Pi, After Fob 1000 Lbs. Live Weight. Total dgest- ible or- ganic matter. i6s. 14.6 15.5 Fat iu milk. Per of. 4.3 4.4 i>>le pro- tein. Lbs. 2.01 2.03 Fuel value. Cat. 28,813 31,611 Nutri- tive ratio. AvERAOK Per Day Per Cow. Milk yield. 1:6.8 1:7.5 Lbs. 21.1 20.7 Fat In milk. Lbs. .91 .91 Approxi- Dry mate cost matter Of food In food for one for one lb. of lb. of milk. milk. CIS. Lbs. .73 .9 .75 1.0 Dry matter in food for line II). of milk solids. Lbs. 6.9 7.5 Digest- ible dry jnatter in food for each lb. of milk, solids. Lbs. 4 6 5.0 AN INCREASE OP PROTEIN. The nvornffo of 273 records, eacli of wliicli covers a period when the amount of protein in the ration was increased, is sho^'h in Table XXII. There was, on the average, also a slight increase in fuel value. The shrinkage in milk flow was about half as great as would normally occur. The very moderate rate of gain in live weight was somewhat increased by the change. New York Agricultural Expeirimext Station. 99 Table XXII. — An Increase in Amount of Protein. Avernge of 273 rec- ords from cows avera^, iug 4.5 jrs. old and 5.2 months in milk. After £P Bofore .a o o c3 s Before Aftei- Per 1000 Lbs. Live Weight. Total (lltsest- Ible or- ganic matter. Lbs. 14.9 15.3 Digest- ible pro- tein. Fuel value. Lbs. Cal. 2.01 29,725 2.42 30,680 Nutri- tive ratio. 1:7.0 1:6.0 Average Per Day Per Cow. Milk yield. Lbs. 22.5 22.2 Fat In milk. Lbs. .92 .91 Fat In millc. Fer ct. 4.1 4.1 Approxi- Dry mate cost matter of food In food for one for cue lb. of lb. of mlllc. inillv. Cts. Lbs. .69 .9 .73 1.0 Dry matter in food for one 111. of milk solids. Lbs. 7.2 7.6 Digest- ible dry matter In food for each lb. of milk solids. Lbs. 4.7 4.9 A REDUCTION OF PROTEIN. The average from 297 records wliicli cover periods when the amoimt of protein in the ration was reduced give the data in the following table. There was, with this reduction, a slight lower- ing of the fuel value. The shrinkage in milk yield was at con- siderably more than the normal rate. The moderate increase in live weight was not affected by the change. 100 Report of Depautmknt of Animal Husbandry of the Table XXIII.— A Reduction in Amount of Protkin. Per 1000 Lbs. Live Weight. AVBKAOE Per Day Per Cow. Total digest- ible or- ganic matter. Digest- ible pro- tflii. Fuel value. Nutri- tive ratio. Milk yU'ld. Fat in milk. 1 Before "S After F-c 173 |si I- cS ^^ H "^^ Before ^ =» After Lbs. 15.8 15.1 Lba. 2.38 2.02 Cal. 31,456 30,197 1:6.4 1:7.3 Lbs. 21.5 20.5 Lbs. .90 .86 Average of 297 rec- ords from cows averaging 4.8 JT8. old and 6.1 months in milk. Fat In milk. Approxl- niule cost of food for one lb. of milk. Dry matter In food for one lb. of milk. Dry matter In food for one lb. of milk solids. Digesti- ble dry matter In food for each lb. of milk solids. Per ct. 4.2 4.*J Cts. .74 .76 Lbs. 1.1 1.1 Lbs. 8 7.8 Lbs. 5.3 5.2 LITTLE CHANGE OF PROTEIN AND IN THE AMOUNT OF NUTRIENTS. On the average for 49 records when there was little change in the amount of protein or of the total nutrients there followed about the normal reduction In the milk yield. The moderate rate of gain in weight was not much influenced. The following table contains the average data. New York Agricultural Experiment Station. 101 Table XXIV.— Little Change in Protein and Amount of Total Nutrients. ^ Pee 1000 Lbs. Live Weight. AVEEIOII PB« DAT Pbr Cow. Total digest- ible or- ganic matter. DlffciV lbl» pro- tein. value. NuttV tlre raUo. mik yield. Fat In milk. 1 Before » After o is flO 00 ooo -^Z before ^ «• After Lbs. 14.3 14.4 Lbs. 1.84 1.84 Cat, 28,096 28,121 1:7.3 1:7.4 Lbs. 19.1 18.6 Lbs. .86 .82 Average of 49 rec- ords from cows averaginji; 4.1 jrs. old and 4.9 mouths in milk. Fat In milk. Approxi- mate cost of food for one lb. of milk. Dry matter In food for one lb. of milk. Dry matter In food for one lb. of milk solids. Digest- ible dry matter in food for each lb. of milk solids. Per ct. 4.5 4.4 Ctt. .75 .78 LbB. 1.0 1.0 Lbs. 7.2 7.4 Lbs. 4.8 4.9 little CHANGE OP PROTEIN WITH AN INCREASE IN AMOUNT OF NUTRIENTS. Thirty-five records for periods when, with little change in the amount of protein, there was an increase of total nutrients show on the aA-^erage a slight increase in the milk flow — without change in the cost of production. There was a slight gain in weight before the change of ration and some losa in weight afterward. 102 R,EroRT OF Department of Animal HcsBANDRr of the Table XX7. — Little Change of Protein with an Increase of Total Nutrients. Average of 35 rec- ords from cows averaging 4.5 jrs. old aud 5.0 tuouths iu milk. ® tc a rt rS tl ; u C5 =H ' -*J rt P ■o o on ,^ l-H m T-i o P Before After Before ^ =« After Per 1000 Lbs^Live WEiairr. Average Per Day Per Cow. Total digest- ible or- ganic matter. U>B. 14.2 16.8 Fat In millt. Fer ct. 4.3 4.3 Digest- ible pro- tein. Lbs. 2,23 2,28 Fuel value. Cat 28,197 35,832 Nutri- tive ratio. 1:5.8 1:7.4 Milk yield. Lbs. 22.4 23.5 Fat in milk. Lbs. .96 .97 Approxi- Dry mate cost matter of food in food for one- for one lb. of lb. of milk. milk. eta. Lbs. .74 .9 .74 1.1 Dry matter in fooii for one lb. of milk solids. Lbs. 6.4 7.8 Digest- ible dry matter In focitl for each 11). of milk soilrls. Lbs. 4.2 5.1 AN INCREASE OF PROTEIN WITH LITTLE CHANGE IN AMOUNT OF NUTRIENTS. The average of 65 records shows, following this change in tlie ration, about the normal diminution in milk flow or slightly less. There was a slight average loss in live weight before and none after the change. New York Agricultural Experiment Station. 103 TAiiLi: XXVI.— An Increase of Pkotein with J.ittle Change OF Nutrients. Average at 65 rec- ords from cows averaj^iug 4.5 yrs. old aud 5.0 mouths in milk. Per 1000 Lbs. Live Weight. Average Per Day Per Cow. Total digest- Digest- ible Fuel Nutri- Milk Fat In ible or- ganic matter. pro- tein. value. tive ratio yield. milk. Lbs. Lbs. Cal. Lbs. Lbs. 0) Before 14.3 1.91 28,825 1:7.2 21.9 .92 to a rt After 14.5 2.28 29,167 1:6.0 21.4 .88 ,a o ® u a Approxi- Dry Dry Digest- ible dry matter i( food fof mate cost matter matter in Ti Zi Fat In or food In food food for milk. for one lb. of milk. for one lb. of milk. one lb. of milk solids. each lb. of milk solids. in >, .-t c3 2'-' Per cf. Cts. Lbs. Lbs. Lbs. -^-2 Before 4.2 .65 .9 6.8 4.4 S ^ PM After 4.1 .68 1. 7.3 4.7 AN INCREASE OP PROTEIN AND OF TOTAL NUTRIENTS. One hundred and thirty-four records, each of which covers a period when there was an increase of total nutrients with that of the protein, show, on the average, no shrinkage in the milk yield. Without change of ration, there is usually expected, dur- ing a similar period, at this stage of lactation, a diminution of about 2..5 per cent. The very slow rate of gain in live weight was moderately increased. 104 Report of Drpartmknt of Animai, IIusjjanduy of tiik Table XXVII.— An Increase ov Protein and op Total Nutkifnts. Average of 134 rec- ords from cows averaginp; 4.9 yrs. old and 5.0 montba in milk. Before After to a e3 V o 39 b >-. yith this same increase of protein, a reduction in the amount of total nutrients. In these few instances there was a rapid diminution of milk flow with a much greater cost of pro- duction. AX INCREASE OF THE PROTEIN WHE'N ABOVE 2 LBS. AND LESS THAN 2.25 LBS. There are 56 records which cover periods when this change in the ration was made. There followed a diminution of milk flow at a little less than the normal rate. The gain in live weight, fairly good before, was very little after the change. The data are found in Table XXX. With an increase of nutrients. — After an increase of protein, with an increase in the amount of nutrients, in 28 instances there followed ver^^ much less than the usual diminution in milk flow. There was a good rate of increase in live weight before but a moderate loss after the change. See B in Table XXX. With a reduction of- nutrients. — When, with the increase of pro- tein, there was a reduction in amount of nutrients in 28 in- stances, the following decrease in milk production was at a faster rate than normal. There was a good rate of gain in weight, but sloAver after the change. AN INCREASE OP PROTEIN WHEN BETWEEN 2.25 LBS. AND 2.5 LBS. The average data from 72 records which cover periods when this increase in protein content was made are found in Table XXXI. The milk How diminished but very slightly. The live weight increased on the average faster after the change than before. With an increase of total nutrients. — Thirty-two records cover periods when with the increase of protein there was an increase of total nutrients. The milk yield was very slightly increased. There was a loss in live weight before the change and a con- siderable gain afterward. New York Agricut^tural Experiment Station. 107 .it-' lO 00 CO Tt< . , O .MOT t'a3'^£ = :2 C8 o C! cfl EW O O Id O CO ■s o * a 5 o c » £w o OiO o o Oi o ftc n o V a ftC ^ o a « g S^. rH CO CO CO COCl CD t- 1^ ■a 3ii WT3 . CO 'll 05 CO CI cJ IN Ol l-T-l CI CO dH as cj >5 " — » a) 3 ~ fa S O ■^ M * a ii h5 fa CD CD l^ CD CO ira CI C4 coo" CJ fO O Ci CO iH i CI ►-1 rH CO 05 O o 1-1 4-5 pq<1 CD CO t- CO oo iO C) o l-O CI CI 1-1 lO OOi-l tH CI CO -t^ o a> I- C5 I- CD I- -H CI CD CI lO CO T-T Cl CO OCI c» CI r-Ici CI CO o a> CO feces o o . CD fee a ^ p ^ c o ^ ^ be o o 0) fcrj >.: o ^. ■'-• ^ ^* f-1 M3* ft* H OO t2[- OO ^§8 00 IN CI --1 O i-l CO fO (N O CO i-^ lO in lO -»1 ^' OO CO 00 OtH «§3 (M CO (31 Oi CO (M So ci (M* 05 O OiOi CD H Oi-I Oi I COO ^,^ rJJO O l£^ IC O CO n S3 S3' o a w ts. '-O o O CO Q O Si 0) ci h^. , r ? 'Z- a o <5 oi c B yj vj a; ^a-^3 5 tj a • tt =;^ C c^ g '-^j »:* t« o t S Moj Oj S- O • ^ =(-( .3 Tt< m 110 Rkport of Department of Amimal Husbandry of the With soms reduction of nutrients. — The average of 28 records which cover periods when there was a reduction of nutrients with the same increase of protein as above is shown in C of Table XXXI. There occurred a diminution in milk jield at about half the normal rate slower after the change than before. The gain in live weight was much AN INCREASE IN THE AMOUNT OF PROTEIN ABOVE 2.5 LBS. Only 18 records cover periods when this change in the ration occurred. With half of them there was an increase of total nutrients and with half a reduction. There was a more than usually rapid falling off in milk for both groups. On the aver- age there was a slight increase in amount of nutrients. The average shrinkage in milk was twice as much as it would normally be. The gain in live weight was much slower after the change of ration. Little change in the cost of production occurred. Table XXXII. — An Increase of Protein when Above 2.5 Lbs. Avirage of 18 rec- ords from cows a ve ra t; ing 5.3 yrs. old and 5.0 months in milk. Before After -a o 9 O a '"■ <-= ^' ri Ik'fore ^ g c ™ P4 After Per 1000 Lbs Lr?E Weight. Total digest- ible or- ganic matter Lhs. 15.4 16.0 Digest- ible pro- tein. Fuel value. 1,6.9. Cat 2.52 30,759 2.89 31,792 Nutri- tive ratio. 1:5.8 1:5-0 AVERAGE Per Day Per Cow. Milk yield. /,6s. 25.5 23.3 Fatin milk. Lbs. 1.05 .9fi Fat In milk. Per ct. 4.1 4.1 Approxi- Dry mate coat matter of food In food for one for one lb. of lb. of milk. milk. Cts. Lbs. .60 .8 .60 .9 Dry matter In food for one lb. of milk solids. Lbs. 6.6 7.4 Digest- ible dry matter in food for each lb. of milk solids. Lbs. 4 3 4.9 New York Agricultural Experiment Station. Ill A RISDUCTION OF PROTEIN WITH LITTLE CHANGE IN AMOUNT OF NUTRIENTS. One hundred records which cover periods when there was a reduction of the protein with little change in amount of nutri- ents give the average found in Table XXXIII. Following this change there was just about the normal decrease in milk yield. The moderate rate of increase in live weight was slightly increased. Table XXXIII.— A Reduction of Protein with Little Change op NUTKIENTS. Average of 100 ree- urds from cows averaffing 4.7 yrs. old a)i» .a c3 O Boforo After Before After Per 1000 Lbs. Live Weight. Total digest- ible or- ganic matter. Lbs. 14.6 16.1 Digest- ible pro- tein. Lbs. 2 56 2.16 Fuel value. Cal. 29,211 32,191 Nutri- tive ratio. 1:5.5 1:7.2 AVEHAOE Per Day Per Cow. MIIli yield. Lbs. 22.6 22.4 Fat In milk. Lbs. .93 .90 Fat in luillc. Per ct. 4.1 4.0 Approxi- Dry mate cost mattpr of food In food for one for one lb of lb. of millc. mllic. Cts. Lbs. .67 1.0 .70 1.1 Dry matter in food for one lb. of millc solids. Lbs. 7.4 8.0 Digest- ible dry matter in food for eacli lb. of milk solid*. Lbs. 4.7 6.2 A REiDUCTION OF PROTEIN AND OF THE TOTAL NUTRIEINTS. The average data from 139 records when both the protein and total nutrients were reduced follow in Table XXXV. The milk yield fell off at a rate about twice as fast as the normal one. There was a moderate increase in live weight before the change of ration and none afterward. New York Agricultural Experiment Station. 113 Table XXXV. — A Reduction of Protein and of Total Nutrients. Average of 139 rec- ords from cows averaging 4.8 yra. old and 5.8 mouths in milk. 4) ,a o «?. 00 '^ T-t r^ - 3 o c8 O Before After Before After Per 1000 Lbs. Live Weight. Total cliges^ ible or- ganic matter. Lbs. 16.7 14.8 Digest- ible pro- tein. Lbs. 2.41 2.09 Fuel value. Cal. 33,328 29,495 Nutri- tive ratio. 1:6.6 1:6.9 AVERAGE Per Day Per Covir. Milk yield. Lbs 22.5 21.0 Fat in niUk. IA)s. .92 .90 Approxi- Dry Dry mate cost matter matter In Fat In of fOO'l In food food for milk. for one for one one lb. lb. Of lb. of of milk milk. milk. BOlids. Per ct. Cts. Lbs. Lbs. 4.1 .72 1.0 7.9 4.3 .74 1.0 7.3 Dlgestl- Ible dry matter in food for each lb. of milk solids. Lbs. 5.4 4.9 A REDUCTION OF THE PROTEIN WHEN MORE THAN 2.5 LBS. One hundrt^d twenty-three records when there was this reduc- tion of protein give the average data found in Table XXXVI. The average yield of milk diminished at about half the normal rate. The rate of increase in live weight, at about one-half pound per day, was not affected. With little change in amount of nutrients. — When there was little change in the amount of nutrients in 31 instances the average result shown in A of Table XXXVI was obtained — which was practically the same as the total average above given. With an increase of total nutrients. — In 47 instances, when an increase of total nutrients occurred, there followed no reduction in the milk yield, but a slight average increase. The rate of gain in weight from about one-half pound per day became about one pound per day. (B, Table XXXVI.) With a reduction of total nutrients. — When there was a moder- ate reduction in the amount of total nutrients there followed, 8 114: Ui.ix)KT UK Dki'artment of Animal HusiiVNDiiv of iue 1 •; -Ti^ X 5 = = 5 O^ 2 = J o •^ O -^ • aa E^ «;-;£ = H yA /^ ^ • t, -J lO -* M CO • O >ft • Tii CO lOO lO lO -^ lO ':oio CO iH Mi- 6(5'* iXO ce l- I- CO GO CO ea o a ~ to -s°>- tS^^a i rHrH OO Oi-I T-l tH «ils^^ -^■'^* iH T-l tH iH • • C*^ o - T-H-H rH M lO CO BS t- s ^ *- -^ - -o 1-- i^ c:: :^ l-l- (XI CO n 1-^ • • »n N , ^ 5^ «■ oo OOi c o Ot-I > o |a o ^-^ CO CO Tj<^' CO Tji n «i ^ ' a i! . H -= -g COM COM O 1^ l-CS w >> e6 SS " <»=^ COCO CO 00 I- l- ^ ^ ^ • • * • • • • • 15 IM D t, ' E^ tcu 5 2 '^ o > „ T3 . 00 lO =S ^ do coo ■* in r-l ^ I- M 05 C5 Ph < m:^ M (M tH rH H L n H . . -^CO t>co M O CO 00 O « 1=2 a;;; LO l;-J o ■is CO 05 M lO M •- M lO H' 'S -S 2 . lO O CO CO rH -H 1- rH > 00^-3 o'rH Ci ri C5 CO m'S5 » S . CO CO CO CO M CO COC^ w 0} J> . - .O M o O 5 « CO t;; lO o ■^ VD CiTf* 1 i Ss^sg t-^l CD O l^M t-:^_ [ o Q (MM MM M'm' mm' t-H u ■s^^-^S' . t-00 fl- lOCO 1-; LC 05 O -TtH iH i-l 1-1 rH rH rH a> O a> 0) c S ^3 .■^ Jl? "H O ■+-! O •+-! OJ tt-i E-t m«ii VX< pq-^ V\< OT crc^i 05 .' i^ .^ m V *^ -> '11 J \ — ■ O it O o do.S 3 o > - ij iH I— IQ S? " :^ CO CO OJ JO "f y. M J5 Ttl y- ■/ '^ C^l r— P l-s > 1- ^ ri — ^ -^=1 O o o ^a «-, ? a 0) ^^w o t) '~ 2 o ~ o o O o . •-' o EC aj c: = ^ s It s "^ s £f g-t 3 CJ r;^ !-i S ito ;_! 5 tcco 2 P itrJ4 ^^S"a 0) -4^ .- O ? > ■^ .S .,::; ?i vl .So S O ,-, ^ ^ ^ CJ Ui '-. - "• < kL !i_r -H S < 23 o New York Aguicultukal Experiment Station. 115 on tho {ivorac;e for 45 iiistauees, about tlie normal shrinkage in milk yield. The rate of increase in live weight was noticeably reduced. A REDUCTION OF THE PROTEIN WHEN ABOVE 2.25 AND LESS THAN 2.5 LBS. The only records available which show this reduction of pro- tein show a corresponding reduction of the total nutrients; 54 records give, following this change in the ration, an average reduction in the milk flow about three times as great as should occur at the same stage of lactation. There was some loss in live weight but much less before the change of ration than after- ward. Table XXXVIT.— A Reduction of the Protein When BkTWEEN 2.25 AND 2.5 Ebs. Pep. 1000 Lbs. Live Weight. Average Per Day Per Cow. • Total digest- ible or- ganic matter. Digest- ible pro- tein. Fuel value. Nutri- tive ratio. Milk yield. Fat In milk. %^ Before ci "S After -a TO —3 ^^ O C5 '^ -^ Before ^ After Lbs. 17.3 14.9 Lbs. 2.38 2.05 Cal. 34,455 29,961 1:6.7 1:7.1 1,6s. 24.2 22.1 Lbs. .99 .93 Average of 54 rec- ords from cows averaging 4.9 yrs. old aod 5.4 mouths iu milk. Fat in milk. Approxi- mate cost of food for one lb. of milk. Dry matter in food for one lb. of milk. Dry matter in food for one ih. of milk solUls. Digest- ible dry matter in food for eacli lb. of milk solids. Per ct. 4.1 4.2 Cts. .63 .72 Lbs. 1.0 .9 Lbs. 7.5 6.7 Lbs. 5.0 4.6 A REDUCTION OP THE PROTEIN WHEN ABOVE 2 LBS., UNDER 2.25 LBS. The 65 records which cover periods when this change in the ration was made show an average decrease in milk flow consider- ably greater than it should normally be. There was a very good increase in live weight. 116 IIei'oet of JJia'AKiJMioNT OF Animal IIusijandkv of IHF ^b^ C.OM rB Qi ■a „ - - ,^ tf J) " -s •?, -5 "T — *-• 5 w ■■ O W ,Q TO X rt VI X ^2^-^ = 2 » a"" o 00 00 T-tiH CO !3 coco (MM Ol o 00 OJ CO OS a6 L-^ IM 1(0 I- I- • • .Si} OOO LO CO 00 ao OrH 05 00 §8 COM O5 0> rHlO 00 2 _ ^ w •■'J ^co 3 ^ ' .Ji CD «w Kew York Agricultural Experiment Station. 117 O 1^ q3 rt O S n * „ 9 ?i fi o Spa to s-3 3 --i gagSsa 5j £a «5fl i^ H O M CM w o H O P o M M M -^ 3" is « o 00 05 00 COM CON MO CO o\ t-00 T-l 1U> CO oi $»6d 2 tu "I » a ^ 2 ■1-a flJ -* -3 *-» 05 00 a> (-1 o <(-l CO I O C£ ^ ^ a o o .•" « W CO * ^■^ ^ P '-I 2 «0 -V QOQO iHM CO 10 00 CO CON 58 00 I- T-t T-l O r-< CO c?i CO 10 CO -f 05 Tt< CO 00 00 t- 05 O 05 tH 10 10 CO in 05 04 o o co^ cOt- >H O • • CD -H O >o C^' CO GO 10 T-l (N rl4 CO iH OS CO d (Z5 IQ CO CD tHt4 CO CO Y^ !/i m o '^ Oi V. Oi 1^ - i*. . 2 o *'- -*■ 118 Rkpurt of Departmknt of Ammal Husbandry of thk With hut little change in tuial nutrients. — The amount of total nutrients was but little changed in 43 instances. The milk flow diminished at a little more than the usual rate. The average increase in live weight was considerably greater after the change. With some reduction of nutrients. — The average for 22 instances, when considerable reduction in amount of nutrients occurred, shows a falling off in milk twice as great as is usual. The gain in live weight was much less after the change of ration. A REDUOTION OF THE PROTEIN WHEN LESS THAN 2 LBS., MORE THAN 1.6 LBS. The average data from '55 records when such a reduction occurred are found in Table XXXIX. The average shrinkage in milk was at about twice the normal rate. There was but little change in live weight. With hut little change in amount of nutrients. — After this change in the ration, when the amount of nutrients was but little affected, there was, on the average in 26 instances, a reduction of milk yield considerably greater than should normally occur. The data are in A of Table XXXIX. With some increase in amount of nutrients. — Only 11 records cover periods when with such a reduction of protein the nutri- ents were increased. In these cases the amount of total nutri- ents was small, and there was a continual loss of live weight. The milk flow diminished but little faster than usual. There was some increase in the cost of production. With same reduction in amount of nutriaits. — There are only IS of the^'ecords which show this change of ration. In these the shrinkage in milk flow was at about three times the normal rate. KEMAEKS. In general changes in the amount of protein within the ordin- ary limits i)roduced less effect than changes in the amount of total nutrients. The average results which followed the differ- New York Agricultural Expeiriment Station. 119 out inodilieations of the ration as related to the protein content are briefly summarized on p. Gl at the beginning of this bulletin. THE NUTRITIVE RATIO. The effects of changes in the nutritive ratio, as a rule, of course were directly in line with those evidenced by the groui)- ings of the rations on the basis of protein content as related to differing amounts of total organic nutrients. The unusual amount of fat in a few foods caused considerable difference in some of these relations of ratio to the actual amount of protein in the ration. it is unnecessary to give all the data from the averages made to show the general effect of changes in nutritive ratio. In con- sidering the records on this basis they were grouped with rela- tion to the ratio of 1:6, although the majority of the rations had a wider ratio, principally because the standards in general call for a ration with this nutritive ratio or one narrower. Moderate changes in the nutritive ratio within the ordinary limits had considerable less effect on the milk flow than did changes in the amount of total digestible organic matter. The general results accompanying different modifications of the ration which affected the nutritive ratio are stated on p. Go of the general summary. REPORT OF THE Department of Botany. F. C. Stewart, Botanist. n. J. Eustace, Student Assistant. Table of Contents. I. An ppi(3eniic of currant antliracnose. 11. Notes from the Botanical Department. AN EPIDEMIC OF CURRANT ANTIIRACNOSE.* F. C. STEWART AND H. J. EUSTACE. SUMMARY. During the past season the currant crop in the Hudson Valley has been seriously injured by anthracnose, a fungous disease causing the appearance of numerous small, dark brown spots on the leaves, which turn yellow and fall prematurely. Currant canes were quite generall}' defoliated early in the season and the exposure of the ripening fruit to the sun brought about sun- scald, resulting in a loss of nearly one-half the crop in some plantations. The disease attacks the leaves, petioles, fruit, fruit stems and canes. lu New York State it is present among currants almost every season, but there is no record of its destructive occurrence since 1889. Although it attacks also gooseberries and black currants it has not injured them in the same locality where red currants have been seriously damaged by it. It is readily dis- tinguished from the ordinary leaf spot by the size of the spots, which are much smaller. The weather conditions last spring seem to have been partic- ularly favorable to it; but judging from the past history of the disease it is not likely to become a constant pest. Fruit growers need not be alarmed. Probably, it will become epi- demic only occasionally. Although there are scarcely any experimental data at hand, it is likely that anthracnose may be prevented by spraying with Bordeaux mixture; and since currant worms make neces- sary at least one application of Bordeaux, and leaf spot (a *A rcpriut of Bulletin No. 199. 123 124 liKPORT OF THE DEPARTMENT OF JBoTANY OF TUE disease known to be preventable by spraying) is always more or, less prevalent, and it seems likely that the destructive disease known as cane blight may be checked, it is recommended that currants in the Hudson Valley be sprayed regularly every sea- son. INTRODUCTION. The region between Highland and Newburgh in the Hudson River Valley is the principal fruit-growing section of Eastern New York. Grapes, peaches, raspberries and currants are grown extensively. Currants are grown more extensively here than in any other part of the State. They constitute one of the leading fruit crops in this famous fruit-growing section. While visiting this locality June 13 and 14, 1901, we observed that the currant foliage was quite generally affected with a form of leaf blight or anthracnose caused by the fungus Gloeosporium ribis. The lower leaves were yellow and thickly covered with very small brown spots. Almost all the currant plantations were more or less affected and the presence of the disease could be detected at a considerable distance by the yellow color of the foliage. In some cases the leaves were already dropping quite freely. Fruit growers were alarmed. They were not accus- tomed to see the currant foliage behave in this way. Since there seemed liable to be an epidemic of this somewhat unusual disease we planned to watch its progress. During the remainder of the season we made frequent visits to the locality and kept close watch on the disease, particularly in a badly affected plantation on the farm of Mr. J. A. Hepworth near INIil- ton. This plantation consisted of about five acres in a peach orchard on high, well-drained, slaty soil. SYMPTOMS. The disease works from below, upward. The lower loaves become thickly covered with small dark-brown spots, turn j-ellow and fall. The disease appears in June and continues active throughout the season or until the bushes have been completely defoliated. In the present case it must have appeared rather New York Agricultural Expeiriment Station. 125 suddenly and become epidemic about June 8. When we made our first observations, June 13, it was already so abundant that fruit growers were cognizant of it. Ten days earlier we had spent two days visiting fruit plantations in this same locality and at that time we neither saw nor heard of any trouble with currants except cane blight which is always destructive there.^ Although we were seeking the diseases of raspberries rather than those of currants, it is likely that the currant anthracnose would have come to our attention had it been at all abundant at that time. In a letter dated June 10, Mr. A. B. Clarke, of Milton, states that it was very abundant in his plantation at that date. During June the affected plantations were readily recognized, even at a considerable distance, by the yellow color of the foli- age; but in July this was much less noticeable. By July 10 the few leaves still remaining on the bushes were scarcely at all yel- low although thickly covered with anthracnose spots. By June 26 the fruit was beginning to ripen and thereafter the affected plantations were to be recognized by their conspicuous red color. The falling of the leaves left the load of ripening fruit exposed to view. In addition to the leaves, the fungus attacked the leaf stalks or petioles, causing conspicuous black, slightly sunken spots. It also attacked the fruit stems, the berries and the new canes. The spots on the fruit stems were black and resembled those on the petioles. They were from one-fourth to one-half inch in length and extended half way or more around the stem. On the berries the spots were black and circular and bore some resem- blance to fly specks. While the berries were green the spots on them were fairly numerous and readily seen; but as the berries ripened the spots became less conspicuous. This may have been due to the fact that the small berries toward the tip of the cluster were the ones most severely attacked and as a result many of them dropped before ripening. The affected berries did not rot; and the presence of the spots on the fruit stems 'See Bill. 167 of this Station, p. 202. 126 Hepoict of the Dei'Aktmknt of Botany of the seemed to affect the berries but sliglitly. Very rarely did the berries wither from this cause. Peck's^ statement that the fungus does not attack the berries is certainly an error. Thinking it possible that the fungus attacks also the wood, we made a close examination of the canes in the badly affected Hepworth plantation and were immediately rewarded by the discovery of 3'ellowish pustules which upon microscopic examina- tion proved to be the acervuli or spore conceptacles of Gloeosporium rihis. This was on July 10. Most of the acervuli seemed immature, but some of them contained spores identical with those found on the leaves, thus leaving no doubt that Glaosporium rihis occurs on currant canes. At our next visit, July 23, it was found that the acervuli were mostly mature and contained an abundance of typical G. rihis spores. A quality of the affected canes was collected and preserved. They will prob- ably be distributed in Seymour and Earle's Econamic Fungi. So far as observed, the acervuli occur only on wood of the present season's growth. The color of the acervuli is pale yellow or light brown and differs but little from that of the cane. Conse- quently, they are inconspicuous. However, when they are num- erous, one acquainted with them may locate them with the unaided eye. The fungus seems to do very little harm to the cane, producing but a trifling discoloration of the bark and none, at all of the wood. AYe believe this to be the first account of the discovery of Glwosporinm rihis on currant canes. Considering the inconspicu- ousness of the acervuli, it is not strange that they have been overlooked. It is also possible that under ordinary circum- stances the fungus does not attack the canes. Whenever a plant disease becomes epidemic it is likely to behave somewhat differently from its usual manner. However, be this as it may, the discovery is an important one because it shows where the fungus probably passes the winter and that the canes are to be considered a source of infection in the spring. 'Peck, C. H. Rep. N. Y. State Mus. Nat. Hist, 43:52. IS'ew York Agiucultural Expeiriment Station. 127 now DISTINGUISHED FROM OTHER CURRANT LEAF DISEASES. Among fruit growers the currant disease under consideration is usually known as leaf blight or sometimes as leaf spot. Since there are at least two other common currant leaf diseases which go by the same name much confusion would be avoided if fruit growers would follow the custom of mycologists and call this disease anthracnose. Mycologists apply the name anthracnose to diseases caused by species of fungi belonging to Glccosporium, Gollototnclmm and a few other closely related genera. The currant disease which is properly called leaf spot is the one caused by the fungus Septoria rihis Desm. This produces on the leaves dead, brow^n (or gray) spots which are usually circular in outline and have a dimeter of about one-eighth inch (See Plate I, Fig. 8). As a rule, leaf spot is readily distinguished from anthracnose by the size of the spots, anthracnose spots being much smaller — often no larger than a pin head. However, the spots formed by Septoria ribis on both red and black cur- rants, may sometimes be angular and quite small, although always larger than those of Glccosporium rihis. A notable exam- ple of this came under our observation at Milton where a large plantation of black currants, liibes nigrum, was quite severely attacked by leaf spot as early as July 10, Since, at this date, Septoria rihis had showm itself only in traces on red currants in this locality, and the character of the spots was so much out of the ordinary, we were much surprised to find that the trouble was due to Septoria rihis. The spots were quite angular and scarcely more than one-third their usual size. The variety of currant is one said to have originated near Milton where it is known as the Mackey. The Septoria leaf spot is very common in New York and is usually the chief cause of the dropping of currant leaves in this State; but during the past season it was almost wholly absent from the locality where anthracnose was epidemic until about 12b liiiPOKT Off 1U.M1 DiirAlilMKWT Off JiuTAi^i: Vh' TUE EXPLANATION OF PLATE I. Fig. 1. A leaf of red currant affected with QntJiracnose, Glcieos- poriiim ribis. Natural size. Fig. 2. Spores of GloeosiJorium ribls. Magnification 825 diame- ters. Fig. 3. A leaf of red currant affected icith leaf spot, Septoi-iu ribis. Natural si::€. Ft.g. 4. A leaf of red currant shotving the wo7-Jc of the four-lined leaf-lug, Poecilocapsus lineatus. Natural size. Plate I. — Common Leaf Diseases of the Currant. Kew York Agricultural Expbrimext Station. 129 July 23, when it appeared in abundance and destroyed the few leaves left by anthracnose. Another form of so-called leaf spot which occurs on currant leaves in the Hudson Valley, sometimes in considerable abun- dance is that caused by the four-lined leaf-bug, Pcecilocapsus Uneatus} The spots caused by this insect are angular and trans- lucent or else black with a water-logged appearance. (See Plate I, Fig. 4). They are wholly different in appearance from anthracnose spots and, moreover, they occur on leaves at the tips of the canes; whereas, anthracnose appears first on the lower leaves and may attack leaves on any part of the plant. A third leaf disease of currants is one which may be called leaf blight. It is caused by the fungus Cercospora anr/ulata Wint. According to PammeP this fungus is common on cur- rants in Iowa. In New York State it seems to be rare. In 1S07 we received specimens of it from Highland, and in 1900 specimens were sent us from Long Island. During the past season we have sought for it in the Hudson Valley, but have not found even a trace of it. The spots formed by it are considerably larger than anthrac- nc« spots. O asionally we have met with a form of leaf spot caused by a sp* cies of Phijllosticta. The spots are larger even than those of Sfeptwia rihis so there need be no danger of confusing them with anthracnose spots. THE FUNGUS. Glocosporiiim rihis (Lib.) Mont. & Desm. The fungus which is the cause of currant anthracnose was named Gloeosporium rihis by Montague and Desmazieres^ in 1867. For some time previous it had been known as Lcptothi/rium rihis, which name is, therefore, a synonym. Cnjptosporium ribls Fckl. is also a synonym. As already stated, it attacks the leaves, petioles, fruit stems, 'See Bui. 167 of this Station, p. 291: also Cornell Agr. Exp. Sta. Bui. 58. * rammel, L. H. Iowa Exp. Sta. Bui. 13.07. "Montague & Desmazieres. Kickx' Flore crypt. Flandres 2:95. 9 130 Keport of the Depaktment of Botany of the fruit and canes. The spores are formed in pustules, technically known as acervuli, which originate underneath the epidermis of the leaf, chiefly on the upper surface. The epidermis becomes blackened and elevated so as to form a small pimple. At matur- ity this pimple is ruptured at the summit and the spores escape in a gelatinous mass which appears as a whitish or flesh colored speck at the center of the spot. The spores, which are one-celled and uncolored, are somewhat variable as to size and shape. Usually they are strongly curved and somewhat larger at one end. (See Plate 1, Fig. 2.) As we have found them, the spores measure 12 to 24, a in length by 5 to 9 ja in width, the most com- mon size being 19 by 7/^. In our exjierience there has never been any difiijculty to find the spores in abundance on the affected leaves. They are also fairly abundant on the new canes and on the petioles. On the canes they are much more easily found while the canes are fresh. Upon drying, the contrast of color distinguishing the acervuli largely disappears. From dried specimens of the canes the spores are most easily obtained by scraping the bark after a brief immersion in water. On the fruit stems and berries the spores are found less frequently. So far as known, Glccosporium ribis has but the one spore form above described. However, it is quite possible that there exists, also, an ascigerous form in which the fungus passes the winter. Fuckel^ has suggested such a relationship with Sphcoria circinata Fckl. [=Onomaniella circinata (Fckl.) Sacc] By means of artificial cultures Miss Stoneman^ has shown that two other species of Glceosporium, G. cingulatum Atk. and G. viperatum E. & E., have in their life cycle ascigerous forms referable to a pyrenomycetous genus for which she proposes the name Gnommiiopsis. Excellent figures of Glwosporium rihis are found in Briosi & Cavara's Funghi parassiti delle piante coUivate od utili, Fasc IX, Nr. 222. 'Fuckel. L, Symbola Mycologieie, p. 111. 'Stoueman. Bertha. A Comparative Study of the Development of some Anthracnoses. Botanical Gazette, 26:101-100. ;New York Agricultural Experiment Station. 131 Other species of Glceospm^ium attacking members of the genus Rihes, the genus to which the cultivated currants and gooseber- ries belong, are G. curvatum Oud. on leaves of R. nigrum, the black currant; G. iiibercularioides Sacc. on leaves of R. aureum, the Missouri currant j and G. riUcolum E. & E. on fruit of the English gooseberry. AMOUNT OF DAMAGE DONE. Although the fungus Glceospw'ium ribis is widely distributed over Europe, Asia, Australia and North America, and has long been known to mycologists, it seems to have attracted very little attention as a fungus of economic importance. While it is fre- quently mentioned in works on fungi, it is not often spoken of as doing any serious damage to currants. The first mention of its occurrence in this country seems to have been that made by Berkeley,^ in 1873, who reported it on leaves of black currant collected in Connecticut. In 1884 Peck^ found it on the leaves of the fetid currant, Ribes prostratum, in the Adirondacks. According to Dudley^° and also Peck^ there was a serious outbreak of the disease in New York State in the season of 1889. Prof. Dudley, at that time Cryptogamic Botanist of the Cornell Experiment Station, made the disease the subject of a two-page article which was published as a part of Bulletin 15 of that Station and also in the Annual Report of the same Station for 1889. Although so brief that Prof. Dudley himself called it a note, the article is, even to the present time, the most compre- hensive published account of currant anthracnose as it occurs in America. He reports^^ i\^q disease abundant on white currants at Ithaca and destructive to red currantsin the vicinity of Rochester. Peck^2 says: "A currant-leaf fungus, GlooospoHum ribis, has also ^ Berkeley, M. J. Grevillea, 2:83. "Peck, C. H. Rep. N. Y. State Mus. Nat, Hist, 38:98. '"Dudley, W. R. Cornell Agr. Exp. Sta. Bui. 15:196-198; same in Second Ann. Rep. Cornell Agr. Exp. Sta.. 1889, pp. 196-198. "Peck, C. H. Rep. N. Y. State Mus. Nat. Hist, 43:52. "Dudley. Loc. cit "Peck. Loc. cit 132 Report of the Department of Botany of the been excessively virulent. In some localities currant leaves have been so severely attacked by it that their vigor was destroyed and they fell to the ground long before the usual time. In my own garden the currant bushes were as destitute of foliage in August as thev usuallv are in November." Since 1889 it has been mentioned by PammeP* as occurring on red currants in Iowa and Halsted^^ has reported its occurrence on cultivated gooseberries in New Jersey; but we find nothing in the literature to indicate that ithasbeen at all destructive during the past eleven years. However, from our own observations we are inclined to believe that in New York, particularly in the Hud- son Valley, it occurs to some extent nearly every season and that, in some instances, it has been destructive. June 12, 1897, Mr. H. R. Leeder of New Paltz reported to the station that his currants were dropping their leaves badly. The specimen leaves accompanying his letter showed an abundance of Gloposporium ribis which was probably the cause of the leaves dropping. It is noteworthy that this outbreak, like the one of the present season, occurred before the middle of June. On July 7 of the same year Mr. F. A. Sirrine observed that in the vicinity of Highland cur- rants were dropping their leaves badly. Specimens of the fallen leaves were examined by one of the writers of this article and found to be infested with Cercospora ancjulata and Gloeosporium rihis. June 28, 1900, we observed a plantation of red currants on Long Island which was severely attacked by GlceospoHum rihis. Septoria ribis was also present in small amount. In this planta- tion the Gloeosporiiim had attacked the fruit stems to so great an extent as to attract the attention of the foreman in charge. Nevertheless, we saw no evidence of damage from this cause. None of the berries were dropping or shriveling. Dr. B. M. Duggar informs us that (ilccosporium ribis was abundant on cur- "rammel, L. H. Journal of Mycolofn/, 7:101. In a letter dated Novem- ber 5, 1901, Px-of. Pammel writes us that he has not observed the disease ill Iowa since 1891. ''Ilalsted, B. D. N. J. Agr. Coll. Exp. Sta. Report for ISDo, p. 331. New York Agricultural Experiment Station. 133 rants in tlie Hudson valley in the autumn of 1900. In a planta- tion at Rochester we found a few currant bushes quite severely attacked by G. rihis, August 30, 1900; but this was the only case of the disease observed in western New York last year. The season was an excessively dry one. During the past season currant anthracnose became epidemic in the Hudson valley about June 8. By June 13 many leaves were falling and it was already evident that the crop would be considerably injured. In some plantations one-half the foliage was gone by June 26 and by July 10 the bushes were completely defoliated except for small tufts of leaves at the tips of some of the canes. The fruit commenced to ripen about June 26 and by July 10 the harvest was in progress. About July 1 there was a week of excessive heat with a clear sky. As a result, currants throughout the Hudson Valley suffered severely from sunscald. Most of the leaves having fallen, the fruit was left exposed to the direct rays of the sun. However, it is likely that the injury was not all due to exposure to the sun. Some of it was prob- ably due to the inability of the defoliated canes to supply the ber- ries with water notwithstanding the fact that the soil was filled with water owing to frequent showers. The loss from sunscald and shriveling of the berries was enormous. Mr. Hepworth has. 18 acres of currants from which he sold, in 1900, 50,000 quarts of fruit. In 1901 the same plantation yielded only 26,000 quarts. This loss of nearly one-half the crop Mr. HepAvorth attributes to the effect of anthracnose and the accompanying sunscald. In the five-acre plantation mentioned in the introduction to this bulletin the loss was estimated to be about two-thirds of the crop. The fruit set as well in 1901 as in 1900 and there was no other disease besides anthracnose except cane blight, which was no more destructive in 1901 than in 1900. Therefore, had it not been for the anthracnose the crop of 1901 would probably have been as large as that of 1900. Moreover, the loss on the present season's fruit crop is not all. The dropping of the leaves so early in the season must seriously interfere with the proper ripening of the 134 E,EPOKT OF THE UePAKTMENT OF BoTANY OF THE wood and the formation of fruit buds for next year. How great will be the damage from this cause can not be determined until next season. As already stated, some plantations were almost completely defoliated by July 10. By July 22 many plantations were completely defoliated and many more had lost from one- half to two-thirds of their foliage. As a rough estimate we would say that in the region between Highland and Kewburgh probably two-thirds of all currant leaves (excepting black currants) had fallen by July 22. About this time Septoria rihis also appeared and assisted in completing the destruction. At what time the defoliation was completed we are unable to say, since we did not visit the region between July 22 and September 2. On the latter date very few green currant leaves were to be found ; and yet, normally, currants hold their leaves until heavy frost. On the station grounds, at Geneva, sprayed currants of many differ- ent varieties were in nearly full foliage as late as October 15. The disease was more destructive in old plantations than among young plants. Plants in the nursery row were attacked latest of all and consequently suffered least. It was a common observation among fruit gix>wers that the disease was more severe on high, dry soil than in lower situations where the soil was heavier and naturally moister. Our own observations con- firmed this. The disease was also somewhat less severe on plants which were partially shaded. It is a common practice in the Hudson Valley to plant currants between the rows in) peach orchards. Hence, it comes about that many bushes are in par- tial shade. The shaded plants were not attacked so early as were those fully exposed to the sun. Concerning the amount of damage done by currant anthrac- nose elsewhere than in the Hudson Valley, we have little infor- mation. At Geneva, some plantations lost a large part of their foliage because of anthracnose, and it was present in greater or less amount in almost all plantations; but the damage done by it does not appear to have been great. Prof. Craig informs us that the disease was common at Ithaca. jS'kav York Agricultural Experiment Station. 135 HOST PLANTS. While Gloeosporium ribls may attack several dififerent species of Rihes, it has a decided preference for R. ruhrum to which belong the red and white varieties of cultivated currants. It has been frequently reported on R. nigrum, the black currant, but according to our observations it is not at all destructive to black currants, to say the least. While watching the progress of the disease in the Hudson Valley we examined several plan- tations of black currants, but in no case found any damage done to them by anthracnose. In one case a row of black currants stood between two rows of red ones. The red currants were all severely attacked by anthracnose, but the foliage of the black currants was perfect and apparently free from the disease. The cultivated gooseberry, Ribes grossularice, is also said to be subject to anthracnose. In the region where anthracnose was epidemic on currants there are several commercial plantations of gooseberries none of which were affected by the disease to anv extent. ft/ It also appears that among the red currants some varieties are somewhat more susceptible than others. Our observations on this point are not as full as they should be and so we are unable to give a list of resistant varieties; but it is probable that this difference in susceptibility is sufficiently great to be turned to I>ractical account in case anthracnose should become an im- portant factor in currant culture. On July 23, when the disease w^as in full sway, we made some observations at Middle Hope where four varieties of red currant, Fay Prolific, Victoria, Prince Albert and Pres. Wilder, were growing in the same plantation under practically the same con- ditions. On Fay Prolific, anthracnose had caused about two- thirds of the foliage to drop and Victoria had lost about one- third of its foliage; while Prince Albert and Pres. Wilder were perfect in foliage and practically free from the disease. Goose- berries growing nearby were also unaffected. 136 liEl'UKT OF THE DEPAK•1ME^T OF BoTANY OF TUE THE OUTLOOK FOR THE FUTURE. The question has been asked, Will anthracnose be destructive next season? Also, Is it likely to appear regularly every season hereafter and become a menace to the currant industry? It is our opinion that currant growers need not be alarmed. Anthrac- nose is by no means a new disease of currants. It has existed in the currant plantations of New York for at least twelve 3'ears and probably longer. In 1889 it was destructive; but since that time there is no published record of any damage done by it in this State. Judging from the past history of the disease it seems unlikely that it will become troublesome except in an occasional season when all conditions are favorable to it.^** How- ever, we are not unmindful of the fact that diseases which spring suddenly into prominence as the currant anthracnose has done during the past season sometimes continue to be very destruc- tive. Striking examples of this are afforded by the cucumber downy mildew, Plasmopara cuhensis, and the asparagus rust, Puccinla asparagi. The former first appeared in this country in 1889 and has since become so destructive in the Eastern United States that the growing of late cucumbers must have been aban- doned had it not been discovered that the disease can be con- trolled by spraying.^^ The first epidemic of asparagus rust occurred in 1S96 in New Jersey, Long Island and Southern New England.is Prior to 189G it was practically unknown in America; but each season since 1896 it has been destructive and seems to be established as a permanent scourge of asparagus. '"Exactly what weather conditions are most favorable to the disease is not known. The two epidemics of recent years in tliis State have both occurred in wet seasons (ISSO and 1901) and naturally we infer that wet weather is favorable to the disease. However, Dr. Weiss states (Weiss, J. E. Die Blattfallkrankheit der Johannisbeerstraiicher. Praktische Blatter fiir Pfianzcnsclnitz, 3:3), that in soutlieru Bavaria the disease was epidemic in the dry seasons of 1898 and 1899, but scarcely any damage was done in the wet season of 1897. '■For the history of Plasmopara cuhensis see Bui. 119 of this Station, p. 1G4. "Ilalsted. B. D. N. i. Exp. Sta. Bui. 129. New York Agricultural Experiment Station. 137 Concerning the outlook for currants in 1902, it is safe to pre- dict that the crop in the 'Hudson Valley will be somewhat short- i-ned, owing to the premature falling of the leaves last summer; but the virulence of anthracnose will probably depend very largely upon the nature of the weather next spring The preva- lence of the disease in 1901 is certainly favorable to another epidemic in 1902, provided the weather conditions are favorable. The new wood and fallen leaves are everywhere covered with multitudes of the spores ready to start infection again next spring if they have a chance. In the Hudson Valley, the spring of 1901 was a very wet one as was also the spring of 1889 when the other epidemic oecurred; so it appears that the disease is favored by wet weather. TREATMENT. If it becomes necessary to fight currant anthracnose resort must be had to spraying, which seems to be the only promising line of treatment, except, perhaps, the planting of resistant varieties. Spraying with the copper compounds, particularly Bordeaux mixture, is effective against many fungous diseases of foliage and there is little doubt that currant anthracnose may be controlled in this way. However, there is but little experi- mental data bearing on this point. Prof. Pammel^^ at the Iowa Experiment Station, has conducted more experiments on the spraying of currants than any one else in this country and shown that ^S(!l)toHa ribis and Cercospora angidata may be controlled by spraying with Bordeaux mixture; but Gloeosparium rihis was not a factor in any of his experiments. Dr. Halsted^ made the fol- lowing experiment: " In a row of eight gooseberry bushes, two were selected for treatment. Beginning April 2.5, three appli- cations of Bordeaux were made previous to May 22. The bushes were again sprayed August 13. The foliage was somewhat injured by an anthracnose (Gloeospovium rihis Lib.), but there was 'Tamniel, L. H. Iowa Agr. Exp. Stn. Bui. 13:45-4G; Bui. 17:419-421; tJul. 20:710-718; Bui. 24:987-088; Bui, 30:289-291. ="H:)lsted, B. D. N. J. Agr. Coll. Exp. Sta. Rep. for 1895, p. 331. 13S Report of the Department of Botany of the no practical difference between the sprayed and unsprayed plants." As far as they go, the results of this experiment are unfavorable to the control of currant anthracnose by spraying. Currant growers in the Hudson Valley fully realize the impor- tance of protecting their plants against the ravages of currant worms^i which strip the bushes of their leaves in a surprisingly short time. Of late years they have abandoned the use of helle- bore, the standard remedy for currant worms, and substituted for it Bordeaux mixture containing Paris green, green arsenoid or some other arsenical poison. Promptly upon the first appear- ance of the worms the bushes are given a thoiM3ugh spraying with the poisoned Bordeaux mixture. If the work is well done, and rains not too frequent, a single application suthces for the season. Whereas, if hellebore is used it is usually necessary to make two or more applications, because there are generally two and sometimes three broods of worms during the season and the ^^Two distinct species of currant worms occur in the Hudson Valley, which not only differ in appearance but also in habits. The one generally known as the currant span-worm, called gooseberry span-worm in some sections (Diastictis ribearia), is single brooded; while the imported currant- worm or currant saw-fly {Netnattis ventricostis), has two broods each year. The larva of the first is a caterpillar. They appear early, sometimes before the currant leaves are even fairly expanded. They grow rapidly and feed voraciously. By the last of May or first of June they are full grown and stop feeding. At this time they are about one inch long, of a bright yellow color, marked with white lines on the sides together with numerous black spots and dots. They can also be distinguished from the imported currant-worm by their habit of looping the body when they travel. These worms leave the bushes about the first of June and go into the ground where they change to the chrysalis form. Early in July they issue as adult moths or millers and can be seen flying over the fields during July and pai't of August. In color the adult moth is pale yellow with dusky spots or bands on the wings. Seen at a distance it could easily be mistaken for the butterfly of the cabbage-worm flying over the currant fields. The eggs are deposited on the branches of the currants and do no-t hatch until the following spring. The imported currant-worm is the slug-like catei-pillar of a saw-fly. The flies appear about the time the span-wonn hatches from the egg. They pair first, then lay their eggs ui>on the underside of the currant leaves, usually along the larger veins. The eggs hatch a week or ten days after being deposited. Owing to the time required for laying and hatching the eggs, the worms do not appear until one or two weeks after the span-worm has commenced feeding. The larvoe of the saw-fly reach New York Agricultural Experiment Station. 139 hellebore applied for the first brood is washed off by rain before the appearance of the second brood. Bordeaux mixture, on th« contrary, is not readily removed by rain and enough of it still remains on the leaves to kill the second brood of worms. Besides requiring but a single application, the Bordeaux mixture has an additional advantage in that it protects the foliage, to a consid- erable extent, against leaf spot. The superiority of Bordeaux mixture22 is so evident that the use of hellebore has been almost entirely abandoned, except in cases where the application has been postponed until the fruit is so large that there is danger of spotting it if Bordeaux is used. The application of the poisoned Bordeaux is made upon the first appearance of worms; but last spring the worms appeared somewhat later than usual and so the Bordeaux was applied later. In fact, many persons accus- tomed to spray for worms did not do so the past season because there were so few worms that it seemed unnecessary. maturity in June, at which time they are about three-quarters of an inch long. They go to the ground and spin cocoons around themselves in which they change to chrysalides. During July they change again to adult flies; as a result a second brood of worms occurs after the crop of fruit is gathered. This worm can be distinguished from the span-worm by its color, which is usually green covered with black dots, with the extremities sometimes tinged with yellow; also by the fact that it does not loop the body when it travels, but does frequently curl itself up side- wise when feeding. In most sections of the country the last described species is usually the most common currant pest. When hellebore is recommended, this is the worm that is supposed to be doing the damage. The currant growers of the Hudson Valley have two distinct species of worms to combat and these worms appear at three distinct periods. This would require not only frequent applications of hellebore but also large quantities of it. Such treatment is expensive. The use of hellebore has also proven worthless as a remedy for the span-worm, as shown by the fact that in 1897 the fields in the vicinity of Highland, even where helle- bore was applied frequently, were completely stripped by this pest. These conditions have done much to induce growers to use some arsenical compound in Bordeaux mixture. — F, A. Sirrine. '^-It appears that poisoned Bordeaux mixture as a remedy for currant worms came into use in the Hudson Valley about 1898. It was recom- mended by Mr. F. A. Sirrine in a short article published in the Eastern New York Horticulnrist for October, 1897. Mr. J. A. Hepworth of Marl- borough and Messrs. W. D. Barns & Son of Middle Hope were among the first to use it. 140 K.EPOKT OF THE DEPARTMENT OF BoTANY OF THE Some persons thought they saw evidence that the single appli- cation of Bordeaux for worms had lessened the amount of dam- age from anthracnose. In the plantation of Mr. A. B. Clarke at Milton, we observed that in one portion anthracnose was con- siderably more severe than in an adjacent portion. Upon inquiry as to the cause we were informed that one portion had lL>een sprayed once with Bordeaux mixture while the other had not. In this case there appeared to be a marked benefit from spray- ing; but in general the Bordeaux applied for worms did not have much effect on the anthracnose. Probably the application was made too late. In the absence of experimental data we can only make sugges- tions as to treatment. Bordeaux mixture will pr\)bably control the disease, but the spraying must be commenced early. In view of the fact that the anthracnose fungus inhabits the canes, the first application should be made on the bare canes before the leaves appear.^^ Special attention should be given to the new wood because there is where the spores are most abundant. In fact no spores have yet been found on the old wood. However, the old wood should also be sprayed, because it is possible that some spores do occur on it, and also because of the possible effect on cane blight. How the fungus of cane blight gets into the canes is not known, but there is good reason for believing that thorough spraying of the canes will have a tendency to prevent "For the first treatment a strong solution of copperas (iron sulphate) may be used instead of the Bordeaux. Make a saturated solution (that is, add copperas to water until no more will dissolve) and apply while the buds are swelling but before they break. By some, this treatment is thought to be beneficial for grape anthracnose (See N. Y. Agr. Exp. Sta. Bui. 86:79; and Bui. 170:410). particularly when about one per ci'ut. of sulphuric acid is added to the copperas solution. But if tlie sulphuric acid is added the mixture can not be applied with a spraying machine, because it is so very corrosive. In that case it must be applied wiMi a swab or whisk broom. The fungus of grape anthracnose is closely related to tliat of currant anthracnose and there is some I'eason for believing that any treatment which is successful for the one AAOuld be successful for the other. Nevertheless we have i-ecommended Bordeaux mixture for the first treatment for the following reasons: (1) Bordeaux is likely to be equally effective; (2) The treatment is less complicated; (3) There is no danger of injury to the plants or to the sprayer. New York Agricultural Experiment Station. 141 its attacks. The second spraying should be made while the leaves are unfolding, and thereafter the treatment should be repeated at intervals of ten to fourteen days until there is danger of permanently spotting the fruit. Upon the appearance of worms add Paris green or green arsenoid to the mixture. In wet seasons one or two applications should be made after the fruit is gathered. ' Spraying in the early part of the season should be done with especial thoroughness and regularity in order, if possible, to keep the diseases completely under control until the time when the spraying must be discontinued on account of spotting the fruit. To restate the matter briefly: Spray thoroughly with Bor- deaux mixture, commencing before the leaves appear. Make the second treatment as the leaves are unfolding and thereafter at intervals of ten to fourteen days until the fruit is two'-thirds grown. In wet seasons make one or two applications after the fruit is gathered. When worms appear add Paris green or green arsenoid to the Bordeaux. It seems to us probable that currant growers in the Hudson Valley will find spraying, as suggested above, a profitable prac- tice. Anthracnose may not be epidemic except occasionally, but it probably does some damage nearly every season. Leaf spot is nearly always plentiful in the latter part of the sesiaon, and sometimes causes the leaves to fall before the fruit is ripe. Cane blight is always destructive, and one application must be made for the worms anyway. We believe that loss from all these troubles may be materially lessened by spraying. While the cur- rant bears premature defoliation remarkably well, preservation of the foliage must result in increased vigor of the plants and consequently, larger yields of fruit. NOTES FKOM THE BOTANICAL DEPARTMENT * p. C. STEWART AND H. J. E)USTACE. SUMMAKY. I. In a nursery cellar at Rochester 25,000 pear trees were seriously injured by thawing too suddenly. The sand covering the roots of the trees had become frozen, and in order to facili- tate the removal of the trees a fire was built in the cellar. A few days later it was found that the upper parts of all the trees had turned black. Although the trees were practically unin- jured for planting, it was impossible to dispose of them at wholesale, and they were almost a total loss to their owner. II. The shot-hole fungus so destructive to the foliage of cher- ries and plums has been discovered attacking the fruit-pedicels of cherries. This discovery is of considerable scientific interest, but it has little or no practical bearing on the control of the disease. III. The fungus of antirrhinum anthracnose which was sup- posed to be confined exclusively to the antirrhinum has recently been found on a common weed called yellow toad-flax. Since this weed may communicate the disease to the antirrhinum, the treatment of the disease on the latter is a more complicated mat- ter than has been supposed. IV. It has been observed that imperfectly fertilized peaches may attain considerable size and remain hanging on the trees until September. In such cases this trouble may be mistaken for the " little peach " disease by persons unfamiliar with the latter. However, in the " little peach " disease the pit is of normal size and provided with a well-developed kernel; while in *A reprint of Bulletin No. 200. 14S 2sEW York Agricultural Experiment Station. 143 cases of imperfect fertilization the pit is abnormally small and has no kernel, or at least only a partially developed one. This difference will enable anyone to distinguish readily between the two troubles. V. At Milton, N. Y., the tile drain to a vinegar cellar was clogged by a luxuriant growth of the fungus Leptomltws lacteus. The obstruction was easily and effectually removed by placing a small quantity of copper sulphate crystals in the upper end of the drain. VI. Drain pipes to refrigerators frequently become clogged with a slimy, gray growth of fungus which has its origin in the ice, but is not an accumulation of matter from the ice. It may be easily controlled by occasionally washing out the drain pipe and ice chamber with boiling water. I. TROUBLE WITH PEARS IN A NURSERY' CELLLAR. In March of the present year the Station received a letter from a Rochester nurseryman requesting that an expert be sent to his place to inquire into the cause of a serious trouble among the pear trees in his nursery cellar. One of the writers of this arti- cle was sent to investigate. It was found that 2.5,000 three-year old standard pear trees had been tied into bundles of ten to fif- teen trees each and placed in the nursery cellar in an upright position. The bundles of trees were set in rows and the roots covered with sand, after the usual custom in such cases. The bark on the trunks and branches of the trees was of normal color and apparently all right up to a height of about three and one- half feet, but above this point the bark was black, and many of the branches were evidently dead. This condition prevailed throughout the cellar in a strikingly uniform manner. All parts of the trees below three and one-half feet were healthy and all parts above that point blackened. This blackening of the branches was suggestive of the bacterial fire blight and the owner was fearful that it might be an outbreak of that disease. Observing that a fire had been built in the cellar, suspicion at once pointed in that direction, and after an inquiry into all the 1-ii liLFOKT OF THE Jjiil^AKTMKJST OF BoTAJSIY OK 1 UK details of tlie case it became plain that the trouble was due to the trees having been thawed out too suddenly. The trees were of many different varieties, and yet all were equally affected. Had it been due to lire blight or any other parasitic diseases, some varieties would have been injured more than others and some individuals more than others. On the disease hypothesis it is also impossible to account for the uni- formity of height at which the trees were affected. When the trees were placed in the cellar in the autumn they were all right,, and an examination of some trees in the same blocks which had remained over winter in the field showed that none of them had blackened branches. Also, some of the same lot of trees which had been stored in another cellar were free from the trouble. For many years it has been the practice of the owner of the trees to keep the temperature of the cellar as nearly as possible at 32° F., and whenever the temperature tends to fall below 32° an open wood fire is built on the floor of the cellar. In the present case, however, no fire was built during the winter. Hence, early in the winter the sand about the roots of the trees froze to a depth of i>erhaps three inches and remained frozen until February. On February 25, 1200 of the trees were dug out of 'the frozen sand and packed for shipment. No complaint was received concerning the condition of these trees, so it may be assumed that they were not affected with the branch black- ening either before or after removal from the cellar. At this time all trees in the cellar appeared to be all right. So much difficulty was experienced in removing the trees from the frozen sand that it was decided to build a little fire in the cellar and thaw the sand. The fire was built February 27 in one corner of the cellar where the 1200 trees had been removed two davs earlier. A few davs later the trees were observed to be in the unhealthy condition above described. Our own observa- tions were made March 15. The fire had not been suspected as being the cause of the trouble because it had long been the cus- tom to build fires in cold weather. The man who built the fire admitted that it had been made a little larger than usual in order New York Agricultural Experiment Station. 145 to thaw the sand as quickly as possible. However, it is unlikely that the fii-e was excessiA'ely hot, because, if it had been, some bundles of trees standing close to it would have been badly scorched, whereas only a few of the most exposed trees were slightly scorched on the exposed side of the trunk. Otherwise these trees were scarcely more injured than trees at the opposite end of the cellar. The heated air rose to the ceiling (which was about seven anu one-half feet above the floor and very tight), spread out over its entire surface and then accumulated in a layer of uniform thick- ness. This layer of warm air was warmest at the ceiling and became cooler the nearer it apx^roached the floor. The tips of the branches, being nearer the ceiling, were enveloped in air warmer than that surrounding the basal portions of the branches and the trunks. They were also smaller. Consequently the upper parts of the trees thawed out more quickly than the trunks. Now, it is a well-known fact that frozen plants which may be thawed with- out injury, if the thawing is done slowly, may be ruined if thawed suddenly. It appears that the pear trees were thawed too sud- denly, and that the line marking the boundary between the injured and uninjured portions marks the height above which thawing progressed too rapidly for safety. That the temperature of the air was a more important factor than the size of the branches is shown by the fact that one bundle of Bartletts, in which the trees were so short that they did not project above the danger line, was wholly uninjured. The majority of the trees were of such a height that their branches were blackened for a distance of six to eighteen inches. Only in a few instances did the injury extend quite to the trunk. With a few exceptions, the blackened branches might have been cut away without removing more of the tops than is customary in transplanting; and since it is unlikely that the branches were injured below the point of discoloration, the trees were practi- cally unhurt for planters' use. Nevertheless, the trees, which were worth about |2000, were almost a total loss to their owner. Twelve thousand of them were sold for |100. to a man who cut m 140 Repokt of the Department of Botany of the them back and planted them, losing less than two per cent, although they were not set until May 1. The owner states that had he been doing a retail business he could undoubtedly have disposed of a large proportion of the stock at a fair price, but it was impossible to sell it at wholesale. II. SHOT-HOLE FUNGUS ON CHERRY FRUIT PEDICELS. In New York State the shot-hole fungus, Cylindrosporium padi Karst., does more or less damage every season. It is destruc- tive to both plums and cherries in the nursery and in the orchard. During the past season it was unusually destructive. Among cherries, the variety English Morello is especially sus- ceptible to the disease. Trees of this variety were dropping their leaves quite freely as early as June 26 and in some cases the trees were nearly defoliated by August 1. On June 26, while examining some seriously affected English Morello trees at Milton, it was observed that many of the fruit- pedicels bore brown spots of considerable size. Upon micro- scopic examination it was found that the spots were caused by the shot-hole fungus, Cylindrosporium padi. On July 11 the same thing was observed at Highland. In this case there was a long row of English Morello trees, all heavily loaded with fruit. So many leaves had fallen that the trees looked bare. The fruit-pedicels were so generally attacked by the fungus that it was somewhat difficult to find one which was entirely free from the brown spots. The spots were from one- eighth to one-fourth inch in length and extended one-third to one-half the distance around the pedicel. In many cases they completely encircled the pedicel. Often the spots coalesced, and then a large portion, or even all, of the pedicel was brown. Even with the unaided eye one could detect a white speck or, more often, a white rift, at the center of each spot. With the aid of a hand lens it could be plainly seen that the white specks were gelatinous spore masses. The affected pedicels almost invariably showed an abundance of the spores. The same was i^Ew York Agricultural Experiment Station. 147 true at Milton two weeks earlier and also at Geneva, on July 13. There was no diflfieulty whatever in finding the spores. The presence of the spots on the pedicels caused the fruit to ripen unevenly. Many of the fruits were dwarfed and some of those most severely attacked withered. However, these injuries cannot, with justice, be attributed wholly to the spots on the pedicels. The premature falling of the leaves, also, had some- thing to do with it. We believe this to be the first record of the occurrence of G jlindrosporium padi on the fruit-pedicels of cherry. We do not say positively that such is the case, because we have not made an exhaustive examination of the literature; but it is at least safe to say that the fact is not generally known, because it is not mentioned in any of the many accounts examined by us. In connection with the appearance of CyUndrospormm on the fruit-pedicels we have observed a spotting of the green fruits which gave cherry growers in the vicinity of Geneva consider- able concern last spring. It was first brought to our attention by the Station Horticulturist, Mr. Beach, about June 15. The fruits, which were at that time about the size of peas, showed numerous small, brown, slightly sunken spots. As the fruits grew many of them became somewhat misshapen, seemingly as a consequence of the presence of the spots. The spots enlarged but little and there was no tendency to rot. In the vicinity of Geneva this trouble was exceedingly com- mon on English Morello and Montmorency Ordinaire, and fruit growers were fearful that the crop would be injured; but as the cherries began to swell and color in ripening the spots seemed to disappear, so there was little or no loss from it. The cause of this spotting is unknown to us. Because of its constant association with Cylindrospor'mm padi on English Morello at Geneva, Milton and Highland it was at first sus- pected that it might be due to that fungus. However, no evi- dence of the presence of any fungus could be found on the spots. Moreover, Montmorency Ordinaire, which was little affected by CyUndrosporium on the foliage, had nearly if not quite as much lis Report of the JJepaktment of Ijot/,nv of the of the fruit spot as had English Morello. These two facts, particularly the latter, are opposed to the theory that the spots were due to Cylindrosporium padi. III. ANTHRACNOSE OF YELLOW TOAD-FLAX. On June 26, 1901, while passing through a peach orchard infested with the common weed A'ariously known as Yellow Toad-flax, Butter-and-Eggs, and Ramsted, it was observed that some of the plants were dying. Upon making an examination of the affected plants it was found that the trouble was due to an anthracnose which was attacking the plants near the surface of the ground. For a distance of two to four inches above the surface of the ground the stems were pitted with elliptical sunken spots almost identically like those produced by Colletotri- chum antirrJiml on stems of the cultivated snapdragon, An- tirrhinum majus?- Since the Yellow Toad-flax, Linaria vulgaris Mill., belongs to the same family, Scrophulariaceie, as the cultivated snap- dragon, it is not strange that it should be attacked by the snap- dragon anthracnose. However, no case of the kind had ever been observed, although we had sought carefully for it. In fact, the disease was known only on the snapdragon, hence the fol- lowing statement in our Bulletin- 179: '* So far as known at present, this anthracnose attacks no other plant besides the Antirrhinum. Therefore, the florist whose grounds are free from the disease will have no trouble so long as he propagates only from his own stock or from seed. In such a case the source of danger is in diseased cuttings and plants from other estab- lishments." Upon the discovery of an anthracnose on the Yellow Toad- flax we immediately became interested to know if it was really the same as the snapdragon anthracnose. It is important to know this, because Yellow Toad-flax is a common weed of wide 'For an account of anthracnose on snapdragon, see Bui. 170 of this. Station 'Stewart. F. C. An anthracnose and a Stem Rot of the Oultivated £napdragon. N. Y. Agr. Exp. Sta. Bui. 179:109. New York Agricultural Experiment Station. 149 distribution, and if it serves as a host plant for the fungus of snapdragon anthracnose the problem of controlling the latter disease is a more complicated matter than has been supposed. Accordingly we made a thorough examination of the disease and the fungus causing it. The majority of the spots were black with the acervuli of a Colletotrichum. Setre and spores -vere abundant. The leaves on the diseased portion of the stem were nearly all dead and brown. Close examination revealed the presence of anthracnose spots on the dead leaves and there were also a few spots on the living leaves, but the leaf spots were in- conspicuous and not abundant. In all morphological characters the fungus agrees fully with Colletotrichum ontirrhini and there is little doubt but it is that fungus. However, positive proof depends on cross inoculations with pure cultures. These have not been made. It was found that many small plants had been killed outright by the disease, but that there were also many others which, although their stems were covered with the spots, were, never- theless, flowering and apparently thriving. While the disease evidently does some damage to the wood, it seems unlikely that it can be turned to any practical account as an aid in its eradica- tion. The original place of discovery was near Milton on a steep hillside in a rather dry situation where the plants were partially shaded by peach trees. Later it was found in similar situations on two other farms at Milton and also at Middle Hope. IV. IMPEEFECT FERTILIZATION AND THE LITTLE PEACH DISEASE. During the past few years peach growers in Michigan and in Western New York have been much concerned over the ap- pearance of a new and destructive disease known as the '" little peach " disease. It appears to have been first described by Taft^ in March, 1898. In October of the same jear a more extensive = Taft, L. R. Mich. Agr. Exp. Sta. Bui. 155:303-304. 150 Report of the Department of Botany of the account was published by Smith.'* The latter article has been widely quoted in the horticultural journals. Thus far no remedy for the disease has been found, and even the cause of it is still unknown. However, it is announced that Mr. M. B. Waite, an expert connected with the United States Department of Agri- culture, has the subject under investigation and it is confidently believed that we shall know considerably more about the disease in the near future. Since so much has been said about the disease and it is known to occur in various parts of New York State, particularly in Niagara County, our fruit growers are constantly on the look- out for it. During the past season a fruit grower of Penn Yan suspected that the " little peach " disease had made its appearance in his orchard. Upon investigation it proved to be simply a case of imperfect fertilization. Of course imperfect fertilization is com- mon among peaches, but this case had some unusual features making it worthy of record. Moreover, there are undoubtedly many fruit growers, like the one at Penn Yan, who have read of the " little peach " disease, but having never seen it are unable to distinguish it with certainty from the effects of imperfect fer- tilization. Hence, it seems desirable to give a detailed account of the Penn Yan case. The orchard was composed of 150 ten-year-old trees of the variety Globe. Occasional trees of several other varieties were intermingled. The owner stated that enough fruit had set to make a full crop. In fact, he expected to be obliged to thin it; but the great majority of the fruits failed to develop, although most of them remained hanging on the trees until ripening time. He estimated that the yield of marketable fruit was between one-eighth and one-sixth of a full crop, the money loss being about |oOO. Our observations were made September 25. At that time most of the marketable fruit had been gathered, but the majority of the small imperfect fruits were still on the 'Smith, Erwin F. Notes on the Michigan Disease Known as "Little Peach." The FennviUe (Mich.) Herald. Oct. 15, 1S9S. New York Agricultural Experiment Station. 151 trees. On the same tree and even on the same branch one could find fruits of all sizes from one-half inch in length up to normal fruits having a circumference of about eight inches. (Plates II-V). The majority of them were smaller than a normal peach pit. For the most part the little fruits were normal in color and free from rot. However, some of the smallest were somewhat shriveled. Nearly all of them below the size of a walnut could be cut, without much difficulty, directly through the pit, which was abnormally small and rather soft. Fruits of this size were usually without any kernel in the pit. Those which were one- half to two-thirds normal size often had pits with kernels which had partially developed and then decayed. Frequently the cavity was filled with gum. The little fruits were often mis- shapen. Many were double and some triple. It is not unusual to find unfertilized peach fruits in the spring, little woolly things which fall early in the season in what is called the " June drop." The unusual feature of the present case is the fact that the unfertilized fruits hung on the trees until ripening time and some of them made considerable growth. Had they fallen at the usual time they would not have attracted attention, but it w^ould simply have been said that the fruit did not set well. Why this particular orchard should behave in this way is not clear. So far as can be learned the orchard has received no unu- sual treatment which would account for such a condition. That it was partly due to some peculiarity of the variety is shown by the fact that trees of other varieties, viz., Old Mixon, Stevens Rareripe, Hill Chili, Smock, Stump and Elberta, which were intermingled with the Globe trees, all bore a full crop and with the exception of Elberta none of them showed any sign of the trouble. Elberta showed a little of it. Still it cannot be wholly a question of varieties, because last year the same trees bore a full crop of fine fruit; and the owner has never before noticed any of the trouble. Most of the trees were in a fair condition of general health. For the most part the leaves were dark green and there had been a fairly good growth of new wood. Last year there was a full 1.52 Report of tiik Dkpaktjient of Botany of the crop of fruil,but it was thinned so that the trees were not injured by overbearing. The soil is a sandy loam, well drained, and the air drainage is fairlj^ good. The soil has been cultivated every year and no other crop has been grown between the rows except when the trees were small. Last spring the orchard was not plowed until about June 1, and then the soil baked so hard that there w^as much difficulty in pulverizing it again. No manure was applied in the fall of 1900 and none in the spring of 1901. In the early life of the trees the owner thought they grew' too fast and so manure was withheld from them somewhat. The intermingling of the other varieties seemed to have no effect upon the Globe. Globe trees standing adjacent to trees of other varieties having a full crop of fruit were quite as much affected as trees standing at a considerable distance from other varieties. ]n the "little peach" disease the pit is of normal size and contains a well developed kernel, whereas in this case the pit is abnormally small and contains no kernel or at most only an abortive one. Herein lies the most striking difference between " little peach " and the effects of imperfect fertilization. Plates III and lYshownatural-size photogi-aphs of thirteen peaches, all from one tree. Plate IV also shows natural-size photographs of the pits from these thirteen fruits. Number one was a normal fruit, while the others were undersized as a consequence of imperfect fertilization. By comparing the photographs of the fruits with the photographs of their pits it will be seen that there is an intimate relation between the size of a fruit and the size of its pit. Also that the majority of the pits were far below the normal size. The latter is also shown in a striking manner by the w^eights of the pits as given in the accompanying table: Table Showixg Weights of Peach Pits. Pit No. 1 -weighed.... 6.9G grams. Pit Xo. 2 Aveiglietl. . . . G.24 grams. Pit No. 3 weighed.... 5.05 grams. Pit No. 4 weiglipd. . . . 2.41 grams. Pit No. 5 weighed.... 1.15 grams. Pit No. 6 weighed.... 1.24 grams. Pit No. 7 weighed. . . . 80 grams Pit No. 8 weighed. . . . 51) grams Pit No. weighed. .. . 40 grams Pit No. 10 weighed. . . . 20 grams Pit No. n weiglicd. . . . 20 grams Pit No. 12 weiglied. . . . 20 grams Pit No. V.i weighed. . . . 05 grams Plate II. — Full-sized Peach on Twig with Imperfectly Fertilized Small Peaches. I'LATE III. — 1, Perfect Peach; 2-5, Imperfectly Fertilized Peaches. Plate IV. — Imperfectly Fertilized Peaches: with Pits from Perfect and Imperfect Fertilization. CO H K o Q w Pi H O w New York Agricultural Experiment Station. 153 When a tree is affected with " little peach " all of the fruits on any given branch are affected and are fairly uniform in size; whereas, in the case under consideration, a normal fruit and small fruits of various sizes may be found on the same small branch. (Plate 11.) There are other important differences between " little peach " disease and the effects of imperfect fertilization; but the two above stated are sufficient to enable anyone to distinguish be- tween them. It is important for fruit growers to note these differences. Trees affected with " little peach " should be promptly removed. They do not recover and it is possible that they may be a source of infection to healthy trees. Imperfect fertilization, on the contrary, is certainly not infectious, and trees seriously affected one season may bear a full crop the fol- lowing season. Consequently, it would be unwise to destroy trees because of imperfect fertilization. Mr. G. Hiester,^ writing in the Country Gentleman for Novem- ber 24, 1S98, states that in 1806 his orchard of 3,000 trees bore a crop of imperfectly fertilized peaches. The following j-ear the same trees gave " an abundant crop of perfect peaches." Evi- dently Mr. Hiester had to do with a case similar to that observed by us at Penn Yan, but he makes the serious mistake of con- fusing it with the '' little peach " disease. Another case of imperfect fertilization was observed in a peach orchard near Geneva. On the east side of the orchard there were six rows of the variety Crosby and on the opposite side six other rows of the same variety. Between the two blocks of Crosby there were several rows of Brigdon and Red Cheek Melocoton. The Crosby was so much affected with im- perfect fertilization that the yield was only about one-sixth of a full crop; while the other two varieties were affected but little. According to the foreman in charge, the Crosbys were similarly, but not so much, affected in 1900. ■Hiestpr. Gabriel. Tbe Cause of Little Peaches. Country Gent., G3:92S. 24 X. ISOS. 154 liEPORT OF THE DePARTMKNT OF BoTANY oF THE V. TILE DRAIN CLOGGED BY FUNGUS. On June 13, 1901, while investigating an outbreak of currant autliracnose in the vicinity of Milton, we met Mr. H. H. Hallock, a vinegar manufacturer of that place. Mr. Hallock informed us that the tile drain to his vinegar cellar had become clogged some time during the previous May and upon investigation he had found that the cause of the trouble was a fungous growth re- sembling the " mother " of vinegar. He removed some of the tiles at intervals of about twenty-five feet and laboriously poked out the fungus until the drain was clear. In about three weeks it clogged again. Knowing the destructive effect of copper sul- phate on fungi in general it occurred to him to try to remove the fungus by putting some of the chemical into the upper end of the drain. Accordingly, this was done. About one-fourth pound of copper sulphate crystals was placed in the upper end of the drain on Saturday. The following Monday it was found that a large quantity of the fungus had been discharged from the outlet and the drain was again clear. However, in a few days it clogged for the third time, and the copper sulphate treat- ment was applied again with beneficial results. Fully one-half barrel of the fungus was discharged. This was about June 10. During the remainder of the season the fungus gave no further trouble. Our visit on June 13 was timely. A large quantity of the fungus lay in a pool of water at the mouth of the drain where it could be readily examined. It consisted of brownish, ropy, slip- pery masses of various sizes somewhat resembling the so called *•' mother " of vinegar. A small quantity was obtained for microscopic examination. It was found to consist almost exclusively of hyphje having a diameter of 8 to 11 ix. Some of the hyphaj were almost wholly destitute of contents, while others contained brownish granules which gave the brownish tinge in mass. The hyphae were sparingly branched in a dicho- tomous fashion. At regular intervals they were sharply con- stricted and at each constriction there was a single spherical New York Agricultural Experiment Station. 155 body, steel blue in color and having a diameter slightly less than that of the hypha. On account of the presence of these bodies it was not easy to determine whether there were septa ai the points of constriction, but it was finally decided that the hyphse were non-septate. No sign of fructification was present. After a vain endeavor to determine the fungus it was sub- mitted to Prof. Geo. F. Atkinson, who at once identified it as Leptomitiis lacteus Ag. With the name of the fungus known, its literature became accessible and it was learned that the fungus is one which lives in water contaminated with organic mat- ter. In the present case it was feeding upon the small quantity of cider drained from the floor of the vinegar cellar. Hum- phrey^ reports its occurrence at Bridgeport, Conn., in a stream below a tripe house; and Goeppert^ observed it growing in a small stream below a beet-molasses manufactory near Schweid- nitz, in Silesia. Humphrey^ states that in his studies " it ap- peared in fly cultures from waters from the outlets of drains containing decaying vegetable matter;" but so far as we can learn it has not been previously reported troublesome in drains except, perhaps, in a single instance. In the Country Gentleman (Vol. 61, p. 406) for May 21, 1896, there is a short article headed, "Fungus in Drain." In this article C. W. B[eak] of South Onondaga, N. Y., gives an account of the clogging of his barn- yard drain by " a thick scum — looks like the ' mother ' in vine- gar." By correspondence with Mr. Beak we have obtained ad- ditional details of the case and it appears probable that the cause of the trouble was Lcptomitus lacteus. The spherical bodies at the points of constriction in the hyphre are so constant and so characteristic that they should serve as a mark of identification.^ (Plate VI, Fig. 1). They "Humphrey, J. E. The Saprolegniaceai of the United States, with Notes on Other Species. Trans. Am. Phil. Soc, 17(III):13G. "Goeppert, H. E. Ueber Leptomitus lacteus in der Weistritz. Bcr. d. l^chlrs. GeseUsch. f. vaterl. Cultur, 1852, p. 54. (Reference taken from Humphrey.) *Loc. cit., p. 135. "lu all of the material examined by us the cellulin grains (cellulin- korner) were found almost invariably at the points of constriction. 156 Repokt of the Department of Botany of the prove to be the " Cellulinkdrner " of Prmgslieim.^^ According to Pringsheim^i they are not homogeneous in structure, but show stratification. At first we did not notice this, but upon closer inspection it was found to be true. Before our study of the fungus wns finished and before camera-lucida drawings had been made the fungus decayed and it was found impossible to obtain more of it. About October 7 the drain became clogged and Mr. Hallock, thinking that prob- ably the fungus was the cause, applied copper sulphate as be- fore. But this time the remedy did not work and upon investi- gation it was found that rats had removed the wire screen from the upper end of the drain, thereby permitting the ingress of sticks and rubbish. When the obstruction was finally removed a small quantity of light brown fungus came away with it. While to the unaided eye this fungus bore some resemblance to the fungus which had clogged the drain in June, the microscope revealed the fact that it was quite a different thing. It was a mixture, chiefly of two kinds of fungi: (1) A fungus with large hyphie bearing a striking resemblance to Rhizoctonia. They had a brownish tinge, usually branched at right angles, the branches somewhat constricted at the point of departure and with the first septum at a distance from the wall of the parent hypha. (Plate VI, Fig. 2.) However, the septa were not clearly defined and in many cases it was uncertain whether any real septa existed. The diameter of the hyphie varied from 12 Occasioually a constriction was without a cellulin grain and sometimes cellulin grains were found elsewhere than at the constrictions; but, as a rule, there was a single cellulin grain at each constriction. However, it appears that this condition of affairs is not to be expected in all cases, and may, perhaps, be the exception rather than the rule. Humphrey [Trans. A»i. Phil. Soc, 17(III):69], in speaking of cellulin grains, saj's: " In L. lacteus they often become lodged in constrictions of the hyphii?." He also cites Rothert's observation that they maj' disappear during the formation of sporangia. Pringsheim's figures {Bcr. d. deutsch. hot. GescUsch., 1, Taf. VII, Figs. 1-9) show the cellulin grains distributed seem- ingly without reference to the constrictions. '"Priugshoim, X. Ueber Cellulinlv("»rner. eine Modification der Cellulose in Kiiruerform. Ber. d. dcuisch. hot. Gcscllschaft, 1:288-308. Mit Taf el VII. "Loc. cit. ]N'ew York Agricultural Experiment Station. 157 to 24 11, the most common size being 15 ,a: (2) A fungus with unbranched, colorless, seemingly non-septate hyphre having a diameter of about 2 //. (Plate VI, Fig. 3.) Neither fungus showed any fructification and neither one was determined. Subsequently to our study of the fungus in June Mr. Hallock^- prepared for publication a brief article on the subject, which appeared in the Rural Xew Yorker for July 27, 1901. In addition to the circumstances which we have already related, he states that the tile drain was put into place in the autumn of 1900 to replace a stone drain, which, although it had not run as freely as it should, had, nevertheless, never become completely clogged during the several years in which it was in operation. The new tile drain was made of three-inch porous tiles and worked all right during the fall and winter, but clogged in the spring at a time when there was plenty of rain to keep the drain flushed out. In the fall, at cider making time, considerable pomace is run off through the drain, and had it clogged at that time it would have been less strange. In this connection it is interesting to note that Mr. Beak's barnyard drain at South Onondaga had been in place fifteen years before it became clogged. He removed the fungus by mechanical means. In his recent letter to us he states that he did not use the sulphuric acid recommended by the Country Gen- tleman; neither did he use any other chemical, and yet the drain has not clogged since the spring of 1896. Last spring he again saw indications of the presence of the fungus, but by turning a large quantity of water into the upper end of the drain he suc- ceeded in washing out the fungus and prevented clogging. Mr. Hallock's method of clearing his drain of fungus by the use of copper sulphate is so simple and so cheap that it is worthy of recommendation in all cases of this kind. Sulphuric acid, carbolic acid and other strong chemicals are also destruc- tive to fungi and may, perhaps, answer equally well. We think it likely that the clogging of drains by fungus may be more common than is generallv known. '=H[allock], H. H. Blue Vitriol Cleans a Drain. Rural New Yorker, 60:515. 158 Kepokt of the Depabtment of Butaj^y of the EXPLANATION OF PLATE VI. Fig. 1. Leptomitus lacteus froin tile drain; a, a cellulin grain; Fig. 2. Large lujphw from tile drain. Magnification 225 diam- eters; Fig. 3. SniaZl hyplia from tile drain. Magnification 1060 diam- eters; Figs. 4-8. Refrigerator fungus: 4, a living hypha; 5, four spoirs (?) ; 6, portion of a liypha with forming spore home later- all y; 7, portion of a hypha icith spore home termiiwlly ; 8, hyphce after four months in formalin. Magnifica- tion 650 diameters. Note. — Figs. 2-8 made with the aid of a earner a-lucida.. Fig. 1, diagrammatic. Plate VI. — Fungi From Tile Drain and Refrigerator. New York Agricdltukal Experiment Station. 159 YI. A FUNGUS IN REFRIGERATOKS. Last July our attention was called to a refrigerator which was not working properly. The provision compartment was flooded with water. Upon investigation it was found that the drain pipe was plugged throughout its entire length with a fungous growth. The conical cap over the lower end of the drain pipe was likewise filled with it, as was also the tube of a large funnel set to catch the water and conduct it through the floor. Being, at that time, interested in the tile drain fungus dis- cussed in the preceding article, we at once became interested in this somewhat analogous case and decided to make a study of it. The fungous growth was gray or dirty gray in color; but on account of admixture with dirt from the ice some of it was quite dark. It had a slimy, slippery feel and clung together in sheets or rope-like masses which were often several inches in extent. Microscopic examination showed the slimy, gray masses to be composed of small, uncolored fungous hyphae loosely woven together. The hyphae were branched and had a diameter of 3 ■j-Q 5/i. They contained numerous roundish granules of various sizes, and appeared to be non-septate. The most striking char- acter of the fungus was the presence of curved spore-like bodies resembling the spores of Fusarium except that they were non- septate. They measured 28 to 43 /^ in length by 4^//. in width. They were abundant and most of them were free, but occasion- ally they were found attached to the hyphae both laterally and terminally (Plate VI, Figs. 4-8). We have been unable to iden- tify the fungus. In the fresh condition we were unable to find any traces of septatlon, either in the hyphae or spores; but after the fungus had been preserved four months in a 4 per ct. solu- ion of formalin, some of the hyphae had the appearance of being septate (Plate VI, Fig. 8). However, the small size of the hyphas makes it diflScult to determine this point with certainty; there- fore, the identity of the fungus is very uncertain. If the hyphte are really non-septate (and we incline to this opinion) the fungus belongs to the Phycomyceteae, a group which contains many 160 Report of the Depakiment of Botany of the species of water-inhabiting fungi. On the other hand, if the hyphae are septate it belongs either to Fusarium or Fusisporium, and the species of these genera rarely live in water. It appears that this gray, slimy fungus is of common occur- rence and wide distribution in refrigerators. Upon inquiry among the members of the Station staff it was found that several of them are familiar with the fungus. Five of them furnished us with samples, all of which proved to be identical with the original sample. In each case the fungus with small, colorless hyphte and curved spores was found to predominate. Sometimes traces of other fungi, Oscillaria and bacteria were found but never in quantity. It is plain that the chief culprit is the fungus above described. Mr. Harding, the Station Bacteri- ologist, informs us that while he was an assistant in the bacte- riological laboratory of the University of Wisconsin a refrig- erator kept in the laboratory clogged at frequent intervals with a fungus probably the same as that found by us. Correspondence with some firms manufacturing refrigerators indicate that the trouble is a general one. The Wilke Manufac- turing Co., Anderson, Ind., write as follows: " Replying to yours of the loth, we have encountered, in a commercial way, the fun- gus growth to which you refer. We have always referred to it as "slime from the melted ice.' It is a peculiar deposit or growth, and will in time choke up the drain pipe. There seems to be little or no difference whether the ice is natural or artificial — from distilled water. In our Instruction Card, which accom- panies each refrigerator, we refer to this ' slime ' and request the users to remove drain pipe and scald it at least once a month during the summer season." The Bowen Manufacturing Co., Fond du Lac, Wis., write: " Our attention has at times been called to clogged drain pipes, which on being emptied proved to be filled with a substance having the appearance of jelly, with firmness enough to hold together in lengths of several inches. We had never looked upon this as a fungus growth, but rather as gelatinous matter coming from the ice, or condensed from the vapors which arise from the articles placed in the provision compartment." isEW York Agricultural Experiment Station. 161 In this connection we will call attention to a popular error concerning the origin of the '' slime." In the main, it is a growth and not a deposit or accumulation of matter from the melted ice. In all probability the trouble originates with the ice; that is, the ice contains spores or fragments of the fungus which, upon the melting of the ice, become lodged in the drain pipe and then commence to grow and multiply to an enormous extent. In all cases coming under our observation the principal part of the obstruction has been made up in this way; but if there is dirt or other foreign matter in the ice it lodges with the fungus and adds to its bulk. The nourishment of the fungus consists, chiefly, of waste material from food placed in the ice chamber. With many housewives it is a common practice to use the ice chamber for storing provisions whenever the provision compartment becomes crowded. As a consequence, milk, meat juices, parti- cles of butter, etc., find their way into the drain pipe to furnish nourishment for the fungus growing there. In one of the letters quoted above it is stated that it seems to make little difference whether the ice used is natural or manu- factured. This needs explanation. Ice made from distilled water cannot contain the germs of the fungus and if used in a new refrigerator there would probably be no trouble with slime in the drain pipe. But a change from natural ice to manufac^ tured ice will not result in the disappearance of the slime unless the precaution is taken to thoroughly disinfect the drain pipe and the ice chamber. Otherwise, the fungus contiues to grow as before, because the drain pipe is already " seeded " with the fun- gus before the manufactured ice comes into use. The presence of the fungus should not be regarded as evidence that the ice is dangerously impure. A mere trace of the fungus in the ice may bring about a luxuriant growth in the drain pipe. The simplest and most effective way of getting rid of the fun- gus is to occasionally wash out the drain pipe and ice chamber with boiling water. 11 REPORT OF THE Chemical Department. L. L. Van Slyke, Choinst. C. G. Jenter, Assistant Chemist. W, H. Andrews, Assistant Chonist. J. A. Le Clerc,! Assistant Cliemist. F. D. Flller, Assistant Chemist. E. B. Hart, Assistant Chemist. C. W. Mudge, Assistant Chemist. A. J. Patten, Assistant Chemist, Table of Contents. I. A study of enzymes in cheese. II. Conditions affecting weight loss by cheese in curing. 'Absent on leave after September 1, 1901. A STUDY OF ENZYMES IN CHEESE * L. L, VAX SLYKE, H. A. HARDING AND B. B. HART. SUMMARY. 7. Introduction.— Cheddar cheese contains enzymes coming from (1) bacteria, (2) milk glands of cows and (3) rennet. These enzymes, or chemical ferments, change insoluble cheese-casein into soluble nitrogen-compounds. The investigation has aimed to exclude bacterial action in cheese and limit the action to results produced by enzymes present in milk when made into cheese. //. Historical outline. — Early work done to show whether enzymes were active in cheese ripening gave negative results, owing to faulty methods of investigation, Babcock and Russell furnished the first positive evidence in 1897 in the discovery of the enzyme galactase in milk. They and others have also shown the power of rennet-enzymes to render cheese-casein soluble. ///. Methods of chemical analysis used. — Outlines are given to show methods used in determining total nitrogen, water-soluble nitrogen and nitrogen in forms such as albumoses, peptones, amides and ammonia; also method of determining chloroform in cheese. In milk, the term soluble nitrogen includes aJl nitro- gen-compounds except casein and albumin; in cheese, it includes all the nitrogen soluble in water under the conditions indicated. IV. Effect of chloroform, ether and formalin on the action' of enzymes. — (1) The effect of quantities of chloroform, varying from 2^ to 30 per ct. in milk, covering a period of 102 days, is shown to be apparently small in respect to restraining enzyme action. (2) It is shown that increase of fat in a mixture has little or no *A reprint of Bulletin No. 203, 165 1G6 Report of the Chemical Department or the effect upon the antiseptic value of chloroform. (3) Chloroforra is shown to be somewhat more effective in repressing bacterial activitv than ether or a mixture of chloroform and ether, with- out interfering with enzyme action. (4) Formalin appears to restrain enzyme activity much more than does chloroform. y. Connection hetween bacteria in the udder and enzymes in the milk. — Examination of milk drawn from different quarters of the udder with all necessary precautions shows that there is an apparent relation between the number of bacteria in the udder and the rapidity with which soluble nitrogen-compounds are formed in the milk. VI. Ripening process in normal cheese and in cheese made icith chloroform. — (1) Method of manufacture. Chloroform is intro- duced into milk at beginning of operation of cheese-making, using 4 or 5 per ct. of chloroform for the milk employed. The cheese thus made contains 12 to 15 per ct. of chloroform. It is then kept in an atmosphere of chloroform. All bacterial action can thus be prevented. (2) Chemical changes in cheese under chloroform compared with the normal cheese. In normal cheese, more soluble nitrogen is formed than in cheese made and cured with chloroform; the soluble nitrogen at the end of a year being about 37 per ct. in normal cheese and less than 23 per ct. in chloroform cheese. The amount of soluble nitrogen formed in chloroform cheese is regarded as representing work done by enzymes present in the milk when made into cheese. (3) Influence of acid upon enzyme action. Two-tenths of one per ct. of lactic acid added in the operation of making cheese with chloroform increased the amount of soluble nitrogen in ' a marked degree. (4) The use of salt retards the ripening of cheese to a degree quite marked. (5) Difference in character of chemical change in normal and in chloroform cheese, (a) In normal cheese the proportion of amides is large in comparison with albumoses and peptones. In chloroform cheese the reverse is true, (b) In chloroform cIkh'sc little or no ammonia is formed, while in normal cheese ammonia ai)pears early and increases steadily. isEw York Agricultural ExpMiment Station. 167 I. INTRODUCTION. Fermentation, or the breaking down of complex organic com- pounds into simpler ones, was first looked upon as a purely chemical process. Later it was studied from the standpoint of germ life and now we are coming to see that most of the work of fermentation is accomplished by the action of unorganized ferments commonl\i called enzymes. Enzymes are chemical sub- stances, without life, capable of causing deep-seated changes in certain substances, the enzymes themselyes undergoing little or no change. They are produced by the activity of plant or animal cells. As we shall see, Cheddar cheese as ordinarily manufactured, contains enzymes derived from three sources — (1) bacteria, (2) milk glands of cows and (3) rennet. During the past three years, in which we have constantly been working upon the problem of cheese ripening, it has been our hope to determine what proportion of the casein decomposition in normal cheese can be justly ascribed to the activity of enzymes. A direct determination under normal conditions is rendered impossible by the continued activity of germ life within the cheese mass. A separation of the activities of the various groups of enzymes is also rendered difficult under ordinary circumstances, owing to the intimate way in which the enzymes are mingled during the process of manufacture. Work on these lines is going forward under more satisfactory conditions than before and this report is to be considered only as a record of a portion of the data secured. The difficulties attending an investigation of this kind have been reduced to a minimum by the convenient location of all the departments and supplies concerned in the work. Our bac- teriological laboratory, cheese-curing rooms and dairy are in one building, while the chemical laboratory and the cattle barns are only a few rods distant. In making the cheese used in our investigation, we have had the valuable assistance of Mr. George A. Smith, Dairy Expert, 1G8 TtKI'ilUr OF TIIK (JllKinCAL DbPAK'niENT OF TIIK Mr. L. A. Rogers, Assistant Daetoviologist, lias donp much of the routine work connected with the bacteriological examina- tions. Mr. J. A. Le Clerc and Mr. A. J. Patten have rendered efficient assistance in some of the chemical work. 11. HISTOEICAL OUTLINE. The following outline of the work previously done in relation to cheese ripening covers only those features that relate to the special problems we have been studying. In 1887, Benecke,^ in discussing the r-jle of bacteria in cheese ripening, stated that, while they probably caused such changes, yet the ripening might really be due to the activity of some unor- ganized ferment. He pointed out that, if bactf^'ia are not essen- tial to cheese ripening, this fact could be made clear by the preparation of cheese under conditions which would exclude bacterial activity. Acting on this suggestion Adametz^ pre- pared a number of Hauskiise, a form of soft cheese, in the nor- mal way, except that he added various disinfectants to the milk or to the curd derived from it. When cheese made with the addition of kreolin or of thymole, were examined bacteriologically, they were pronounced sterile; but even when kept for double the normal length of time, they did not take on the appearance of ripened cheese. His experi- ments with salicylic acid, oxalic acid and with vapors of carbon disulphide and iodine were less satisfactory in repressing the microorganisms in the cheese, but these agents seemed to hold back the ripening in proportion as they inhibited the activity of germ life. (Cheese investigators quite generally accepted these results as settling the point raised by Beupcke, and during the succeeding ten years the work on cheese ripening was based upon the theory of germ action. The phenomena of ripening in cheese may be divided into two classes, (1) the chemical decomposition of casein and (2) the lor- •Benecke. Cent. f. Bak., I Abt.. I:n2l (ISST). »Adametz. Thiel's Laudwiitschaftliche Jabib., 18:227 (1889). Kew York Agricultural Experiment Station. 169 mation of cheese flavors. These may or may not arise from common causes. The casein begins to nndergo change at once, while the formation of flavors begins some time later, after which the two progress simultaneously. These two groups of phenomena cannot be measured by the same means or standards. Duclaux, Adametz, Weigmann and their disciples have directed their attention to the formation of flaA'ors and have quite generally relied upon the odor and physical appearance of their material in judging of the rate and character of the ripening. In those cases in which they have gone more fully into the solubility of the casein, they have usually determined this point by its ability to pass through a porcelain filter, a method wliich von Freudenrich & Jensen'^ have shown to be extremely liable to error in practice. They have rarely attempted to show that the species of bacteria which they look upon as the causal ones are present in cheese in any considerable quantities. They have, for the most part, confined themselves to showing that pure cultures of these species are able, by means of excreted enzymes, to digest the casein of milk and at the same time to form cheese-like odors. In some cases they have made cheese with the addition of pure cultures or of solutions of their enzymes and have stated that the resulting product was better flavored than cheese made in the usual way. In establishing this point, however, they were handicapped by the lack of accurate standards for measuring such relations. Recently Adametz and Winkler^ have placed a culture of one of these bacilli upon the market under the trade name of " Tyrogene," its use being expected to result in the production of a desirable Emmenthaler flavor in cheese. Some preliminary tests by von Freudenrich^ have failed to indicate that it will accomplish this desired end. When the study of the kinds of bacteria present in cheese was extended so as to include the numbers of each kind, it was found that the enzyme-forming bacteria previously mentioned »v. Freiidenreieh and Jensen. Lanrlvr. Jabrb. d. Scbweiz., 14:109 (1899). * Winkler. Molkerei Ztg.. 14:817 (1900). •v. Freudenreich. Ann. Agr. Suisse (1901), 170 E.KPORT OF THE ChEMICAL DePAKTMENT OF THE were present only in small numbers. Even when large numbers were added to milk before making it into cheese, these bacteria ceased to grow almost as soon as the curd was put into the press and rapidly disappeared in the cheese. It was found that from the time cheese was made until fully ripened there were present few besides lactic acid bacteria, so called because they curdle milk by production of acid without subsequent digestion of the casein. From these results von Freudenreich was led to question the connection between enzyme-forming bacteria and ripening process. He made numer- ous attempts to produce cheese with the addition of cultures of enzyme-forming bacteria, which uniformly resulted in a product of poorer flavor, according to his opinion. He then became the champion of the theory that lactic acid bacteria are the principal, if not the only, cause of cheese ripening.^ The chief objection to this theory is the fact that no one has yet been able to demonstrate the production of an enzyme on the part of lactic acid bacteria. Without such aid it is difficult to understand how a becterium is to attack an insoluble sub- stance such as the coagulated casein in cheese. Von Freudenreich' added chalk to milk cultures of these lactic acid bacteria for the purpose of preventing the accumulation of acid and of simulating in this respect the conditions found in cheese. He was thus able to demonstrate the ability of these organisms to increase materially the amount of soluble nitrogen. However, Chodat and Hofifman-Bang^ have pointed out that this is not equivalent to attacking the casein after it has been coagulated by rennet. They maintain that lactic acid becteria are unable to attack coagulated casein, even when sugar is not present. In a later publication Jensen,^ without bringing forward adequate experimental evidence, has suggested that lactic acid bacteria are able to elaborate an enzyme. •v. Freudonreich. Cent. f. Bak., II Abt., 1:384 (ISOH). 'v. Freudenreich. Cent. f. Bak.. II Abt.. .3;2.31 (1897). •Chodat and IIoffman-Bans. Ann. In^t. Basteur. l."):3G (1901). 'Jensen. Landw. Jahrb. d. Schweiz., 14:197 (1900). Nkw York Agricoltckal Expkkimknt Station. 171 In 1807 Babiock & Rnssclpo aiinoimccd that the milk of ail maninials coiitaius, in addition to the pi-oviously known sub- stances, (jalactusc, a tryptie-like ferment capable of producing digestion of casein, and tliev suggested that this substance might play a considerable role in cheese ripening. The correct- ness of their statements regarding the existence of this sub- stance has been substantiated by Storch,ii von Freudenreich^^ and •Jenseii.^'^ In 1901 Babcock, Russell and Vivian ^^ and Jensen ^^ almost simultaneously called attention to the ability of pepsin, con- tained in rennet solution, to render casein soluble, and they presented experimental evidence to establish this point. Thus, we see that cheese, as ordinarily manufactured, contains enzymes derived from three different sources, (1) bacteria, (2) milk glands of cows and (3) rennet. Enzymes, in acting upon casein, cause its decomposition and probably produce compounds that furnish some of the cheese flavors. While we appreciate as highly important, from a practical standpoint, the study of cheese flavors, we have devoted our time chiefly to a study of enzyme action upon cheese- casein. It seems that this donstitutcs so fundamental a prob- lem in cheese ripening that it should be first studied, and more- over its solution will doubtless go far toward solving the problem of flavors. III. METHODS OF CHEMICAL ANALYSIS USED. In a later bulletin the methods of chemical analvsis used in determining the amounts of nitrogen present in different forms in cheese will be presented and discussed in full detail. In this connection it seems suftieieut to present only a brief outline of such methods. "Babcoek and Russell Auu. Rc^'pt. Wis. E.xp. Sta. 14:101 (1897). "Storcli. 40 Rept. Copenhagen Exp. Sta. (Denmark). "v. Freiulenreioh. Cent. f. Bak., II Abt., 5:241 (1899). "Jensen. See footnote 9. "Babcock. Russell and Vivian. Ann. Rept. Wis. Exp. Sta. 17:102 (1900>, also Cent. f. Bak.. II Abt., G:817 (1900). '■Jensen. I.andw. Jaliib. d. Seliweiz., 14:197 (1900), also Cent. f. Bak., II Abt., 6:734 (1900). 172 KePORT of TUK CniiMIOAL DKPAKlIUh.NT OF 'JDE PREPARATIOX OF CHEESE EXTRACT. Twonty-fivo grams of cheese are mixed witli quartz sand and treated at 122° to 140° F. (50° to 60° C.) for a half hour with each of several successive portions of water, decanting and filtering each portion of extract until 500 cc. have been accuramu- lated. Portions of the solution thus prepared are used in making the various determinations. DETERMIXATION OP NITROGEN-COMPOTJN'DS IN CHEESE EXTRACTS. (a) Total icater-solnUe nitrogen is determined in an aliquot part of the water extract. (b) Precipitation hy ahim. — To 100 cc, of water extract, 2 cc. of saturated alum solution are added and digested at 104° to 108° F. (40° to 42° C.) until precipitation is complete. The precipitate is filtered, washed and then treated by Kjeldahl method to determine nitrogen. (c) Coagulation by neutralising and boiling. — The clear filtrate from (b) is exactly neutralized by dilute fixed alkali and heated on water bath until coagulation is complete. The precipitate is filtered, washed and its nitrogen determined by Kjeldahl method. (d) Alhimoses. — To the filtrate from (c) two or three drops of dilute (one to three) sulphuric acid are added, and then powdered zinc sulphate to saturation. The mixture is heated on water bath until precipitation is complete and the nitrogen is deter- mined in the precipitate washed with saturated solution of zinc sulphate. (e) Peptones. — To the filtrate of (d) two or three drops of ' strong hydrochloric acid are added and then bromine in succes- sive portions of a few drops at a time, accompanied by vigorous shaking until the liquid becomes super-saturated. The nitrogen in the washed precipitate is determined as before. (f) Amides. — (1) First Method. The nitrogen in filtrate from (e) is determined directly by Kjeldahl method and this, less the nitrogen present as ammonia, is the amide nitrogen. (2) Second Method. To 100 cc. of the original cheese extract Kkw York Agriccltdral Experiment Station. I'i3 there is added about one gram of common salt, together with an excess of ten per ct. tannic acid solution. The precipitate formed is filtered and washed and the nitrogen determined in an aliquot part of the filtrate. From this amount of nitrogen is deducted the amount of nitrogen found as ammonia in (g) and the remainder is the amount of amide nitrogen. (g) Ammonia. — To an aliquot portion of the filtrate obtained in (f), magnesium oxide is e.dded and the ammonia separated bj distillation. DETERMINATION OP NITROGEN-COMPOUNDS IN MILK. Casein.— To 20 gms. of milk, diluted with water to about 100 cc, are added 2 to 2J cc. of saturated alum solution. The determination is completed as under (b) in cheese extract, and the other determinations are made as described above in the cheese extract. DETERMINATION OF CHLOROFORM IN CHEESE AND MiLK. About 5 gms. of milk or cheese are placed in a pressure bottle with about 10 cc. of alcohol and 5 gms. caustic potash. The bottle and contents are then heated 30 minutes at 230° F. (110° C.) in an autoclave. The resulting chloride is determined volumetrically as in case of chlorine in sodium chloride. FORM OF STATING RESULTS. The figures given in the various tables represent percentages of the total nitrogen in milk and cheese. This form of state- ment is usually preferable, as figures representing the actual percentages in milk and cheese are often very small. Hence, considerable variations, expressed in percentages of nitrogen, often represent very small variations when expressed in actual amounts present in cheese and milk. The soluble nitrogen in milk, as used in this bulletin, includes all nitrogen compounds except casein and albumin. The water- soluble nitrogen in cheese includes all the nitrogen soluble under the conditions indicated in preparing the water extract of cheese. 174 Report of the Chemical Depautmejst of the lY. EFFECT OF CHLOROFOR^r, ETITER AND FOR^MA- LIN ON THE ACTION OF ENZYMES. In an investigation of tbis kind a prime necessity is a means of totally suppressing the action of genu life. It is equally important that the action of the agents employed shall not be so violent as to alter the enzymes Qr the casein. The worlv of Babcock and Russell has suggested two sub- stances suitable for this jjurpose, ether and chloroform. Of the two we haA'e used chloroform almost exclusively for several reasons: (1) As an auipsthetic it is more efficient; (2) its pro- portion in any mixtm^e can be quantitatively determined with approximate accuracy by chemical analysis; {?>) the amount required to prevent germ growth does not so largely increase the bulk of the mixture; (4) being less volatile, there is less loss in sampling materials under investigation; (5) it is not inflam- mable. In all our work with solutions it has been our aim to mix carefully by shaking at least once a day during the entire course of the experiment. Too much stress cannot be laid upon this point, since mixtures of milk with ether or chloroform tend to separate on standing and thereby produce conditions favoring the germination of spores in certain portions of the mixture. EFFECT OF VARYING PERCENTAGES OF CHLOROFORM ON ENZYME ACTIVITY. Since the relation of chloroform to the activity of these enzymes has not been investigated, except in a very general way by Babcock and Russell, the following study of its action on galactase and bacterial enzymes was made. Duplicate bottles of separator skim-milk containing only a trace of fat were ])re])ar(Hl containing 2.5, 5, 10, 20 and 30 per ct. of chloroform by volume. Th(^se bottles were kept at 00° F. (15.5° C.j, and examined both chemically and bucteriologically. JN'ew Yokk Agkioultukal Experiment Station. 175 Table I. — I^'FLUENCE of Varying Amounts of Chloroform upon the Activity of Enzymes. In 100 lbs. total nitrogen. -Amount of cluoroform. Age. Total soluble nitrogen. Nitrogen In albumoses and peptones. Nitrogen in amides. No. of germs pel Per ct. Days. fresh Lbs. 9.33 Lbs. 4.58 Lbs. 4.75 cc. 2.5 7 11.60 6.81 4.79 28 5 7 11. G3 6.94 4.69 46 10 7 11.72 7.07 4.65 26 20 7 11.63 6.79 4.84 25 30 7 12.28 7.37 4.91 18 2.5 21 16.80 9.94 6.86 42 5 21 16.16 8.47 7.69 36 10 21 16.39 8.99 7.40 25 20 21 16.37 8.34 8.03 26 30 21 13.09 7.07 6.62 30 2.5 49 21.79 14.75 7.03 ^ 49 21.65 14.63 7.02 o 10 49 21.40 14.93 6.47 20 49 20.11 14.36 5.75 30 49 22.99 16.17 6.82 2.5 112 33.15 18.81 14.. 34 9 5 112 33.62 18.59 15.03 10 10 112 30.51 16.. 38 14.13 11 20 112 33.78 16.39 17.39 6 30 112 33.06 19.15 13.91 5 2.5 192 41.98 19.18 22.80 13 5 192 39.. 37 14.83 24.52 14 10 192 35.36 15.13 20.23 7 20 192 35.65 16.37 19.28 6 30 192 35.78 17.30 18.48 During 112 days the amount of soluble nitrogen varied within very narrow limits in the different bottles. During the next 80 days the bottles containing 2.5 and 5 per ct. of chloroform showed a little more soluble nitrogen than the others, in which the amounts of soluble nitrogen were almost identical. These results could hardly be interpreted as indicating, even after the lapse of 192 days, any marked difference in the effect of definite quantities of chloroform upon the activity of enzymes. The germ content, as shown by bacteriological analysis, is in entire agreement with the results of chemical analysis. From th^se results we see that in the presence of 2.5 per ct. of chloroform the increase of soluble nitrogen is continuous and 17(3 Ukpout of the Chemical Depaktmei^t of thb considerable. However, these results do not enable us to know whether the chloroform exercised any restraining influence upon the activity of the enzyme. Any such repressing effect of chlo- roform upon enzyme action could be directly shown only by using as a means of comparison milk containing no chloroform, but under such conditions the action of bacteria would render the comparison worthless. A comparison of the changes produced in the bottles contain- ing the different percentages of chloroform shows a surprisingly small decrease of change in bottles having the larger propor- tions of chloroform. This tends to show that chloroform restrains enzyme action only slightly. The germ content, even in the bottles containing only 2.5 per ct. of chloroform, was so small that the observed changes were undoubtedly due to the enzymes present in the milk at the begin- ning of the experiment. EFFECT OF VARYING PERCEXTAGES OF FAT UPON THB ANTISEPTIC TALCB OF CHLOROFORM. In the case of ether, Babcock and Russell ^^ have shown that it has a strong tendency to combine with the fat present in such a way as not to exert its anaesthetic influence. For this reason rich cream could hardly be kept from decomposing through bacterial action when ether was used. To test this phase of the question with chloroform, two series of bottles were prepared. The first contained 10 per ct. and the second 20 per ct. of butter-fat and in each series duplicate bottles contained 2.5, 5, 10 and 20 per ct. of chloroform. In order that the transformations in each of the bottles in the two series should be directly comparable when expressed in per- centages of total nitrogen, it was necessary that for a given quantity of nitrogen in any bottle there should also be present a corresponding amount of enzyme. In order to maintain these relations, each bottle contained 900 cc. of a mixture made up of 540 cc. of whole milk, together with sufficient chloroform and melted butter-fat to give the "Babcock and Russell. See footnote 10. New Yokk Agricoltural EiPKKiMENT Station. 177 d€sired percentages by volume. Water was then added to bring the total up to 900 cc. The butter-fat used, after being heated above 185° F. (85° C.) for 10 minutes to kill the enzymes present, was filtered to remove the coagulated casein and was then decanted to free from water and salt. The chloroform assisted in emulsifying the fat and it was only in those bottles containing the smaller percentages of chloroform that difiSculty was experienced in getting satisfactory samples for chemical analysis. In order to minimize this difficulty, the bottles were warmed at 99° F. (37° C.) for a few hours before sampling in order to melt the fat. During the rest of the time they were kept at 60° F. (15.5° C). Table II. — Effect of Varying Amounts of Fat upox the Ajitiseptio Value of CHLOEoroEii. In 100 lbs. total nitrogen. Proportions of Nltn sen m Number »— -^ Total soluble aHiuinoses Nitrogen of germa Chloroform. Fat. Age. nitrogen. and peptones. In amides. per ec. Per ct. Per et. Days. if)s. Lbs. Lbs. Fresh: 17,124 20 10 14 19.31 10.52 8.79 • 20 20 14 22.35 10.07 12.28 10 10 14 21.92 12.86 9.06 10 20 14 19.47 9.38 10.09 • 5 10 14 24.45 15.09 9.36 . 5 20 14 23.58 12.23 11.. 35 2.5 10 14 28.14 16.21 11.93 . 2.5 20 14 28.82 15.65 13.17 20 10 56 35.47 21.. 56 13.91 165 20 20 56 34.85 20.04 14.78 1G3 10 10 56 .38.06 23.20 14.86 167 10 20 56 38.23 21.26 16.97 5 10 56 39.24 23.03 16.21 113 5 20 56 39.48- 22.36 17.12 82 2.5 10 56 38.26 24.44 13.82 209^ 20 10 112 34.60 18.28 16.38 298 20 20 112 36.74 20.24 16.50 10 10 112 37.08 18.62 18.46 2Gi» 10 20 112 41.78 21.40 20.38 142 5 10 112 40.82 19.35 21.48 196 5 20 112 43.05 23.95 19.10 107 2.5 10 112 86.48 18.53 17.95 274 2.5 20 112 38.81 19.09 19.72 34 12 178 Keport of the Chemical DEPAKiMEiST of the The preceding table does not show any marked influence due to Ibe presence of such varying amounts of fat. There are more bacterial spores present than are shown in the results given in Table I. This is probably due to the combined action of a number of factors: (1) The heating of the butter was not high enough to kill the spores introduced from that source; (2) the presence of many small globules of fat in the cultures makes counting difficult and tends to give too high figures; (3) this experiment was started in midsummer, when the air is better supplied with spores, than in midwinter, when the former investigation was begun. A comparison of the percentages of change shown in Tables I and II after corresponding intervals shows the transformation to have been more rapid in the case of Table II. This is easily accounted for by the fact that here whole milk was used and the proportion of enzyme to nitrogen was greater than in the former case where the skim-milk was poorer in enzyme on account of the amount lost in the separator slime and in the cream. COMPARISON OF EFFEiCT OF ETHER, CHLOROFORM AND A MIXTURE OF BOTH UPON EXZYME ACTION. Milk was obtained from two cows, care being taken to brush and moisten the flank and udder and to steam the pail, but by mistake the fore-milk was used in the case of one cow. The milk was taken directly to the laboratory and plates made, which later showed a germ content of 2719 per cc. The fat content of the milk was 4.5 to 5 per ct. Duplicate bottles were prepared in three series containing (1) 15 per ct. of ether, (2) 3 per ct. chloro- form and (3) a mixture containing 2.9 per ct. of ether and 2.1 per ct. chloroform. The bottles were kept at 99° F. (37° C). New Yokk Agricultural Experimrnt Station. 1T9 Table III. — Comparison of Effects of Ether. Chloroform, and Mix- ture OF Both Upon the Activity of Enzymes. Ancesthetic used. In 100 pounds total nitrogen. , . , , . ■ , Mi.xture. Toral Niiro- Nitro- , ■ , soluble gen la geu Nitro- Number Chloro- (;hloro- nuio- alhu- in pep- gen in of genus Ether. form. Ether, form. Age. gen. nioses. toues. amides. per cc. Fer ct. Per ct. Volumes. Days. Lbs. Lbs. Lbs. Lbs. 15 2 11.92 3. 'Jo 2.93 5.01 132 Fresh 2719 3 2 2 13.61 4.54 3.87 5.19 140 3 2 13.18 5.66 2.-38 5.14 93 15 5 16.75 4.81 . 5.64 6.30 19 3 2 5 14.90 4. 20 4.25 6.-39 116 3 5 13.23 3.70 3.60 5.93 28 15 8 20.73 6.38 5.65 8.70 3 2 8 17.67 5.18 4.71 7.78 3 8 17.31 4.26 4.62 8.43 15 14 25.64 9.90 4.72 11.02 100 3 2 14 25.92 8.98 3.98 12.96 5 3 14 24.82 9.16 3.61 12.04 6 15 21 32.49 14.34 4.90 13.25 121 3 2 21 29.71 11.29 4.-53 13.89 02 3 21 27.89 10.-55 4.81 12.50 6 15 28 35.81 14.89 8.42 12. .50 101 3 2 28 32.12 10.74 6.85 14.55 18 3 28 29.80 11.66 5.64 12.50 4 15 35 36.47 15.18 7.31 13.89 3 2 35 32.49 13.24 5.46 13.79 11 3 35 30.36 12.22 6.01 12.13 5 15 42 39.63 18.98 7.87 12.78 138 3 2 42 36.57 15.56 7.59 13.42 13 3 42 35.36 15.74 7.61 12.00 6 15 56 45.75 19.45 11.39 14.91 183 3 2 50 40.28 17.50 10.83 11.95 5 3 56 Lost. 15 84 55.01 22.22 15. .38 17.41 320 3 2 84 48.44 17.88 12.60 17.96 4 3 84 45.83 18.52 10.83 16.58 3 15 137 60.74 15.43 24.03 21.28 113 3 2 1.37 57.-33 14.73 24.08 18.52 2 3 137 Lost. 1 From these results it is seen that a slightly greater amount of soluble nitrogen was formed in the presence of 15 per ct. of ether 180 Report of the Che:micvl Department of the than under either of the other two conditions. From this it mi?;ht be inferred that 15 per ct. of ether was more favorable to enzyme action than 3 per ct. chloroform, but the results of the bacteriological analyses give some reason for believing that there had taken place a growth of bacteria. There had probably been a corresponding increase in the amount of bacterial enzyme. This is rendered more likely by the fact that the bacteria in this case were almost entirely of a single kind, which showed ability to grow in the presence of ether, formed spores quickly in almost every cell and elaborated enzyme with great freedom. This experience has made us slow to accept as trustworthy any results obtained with the use of ether, when the conditions are not constantly controlled by quantitative examination of the bacterial content. OOMPARISON OF EFFECTS OF CHLOROFORM AND FORMALIN UPON ACTIVITY OF ENZYMES. Jensen^'^ in a suggestive article on the enzymes of cheese ripen- ing has called attention to the use of 0.1 per ct. of formalin in studying their activity. Babcock and RusselP^ have stated that comparatively small amounts of this substance completely inhibit enzyme activity. The use of even the amounts recom- mended by Jensen is to be looked upon with suspicion until the influence of formalin upon enzyme action is more fully investigated. In order to facilitate comparisons at some future time, we give the results of parallel examinations of four samples of milk con- taining respectively 4 per ct. of chloroform and 0.1 per ct. of formalin by volume. Unfortunately the strength of formalin was not redetermined but it was the 40 per ct. article of commerce. The milk in this case was obtained from the four quarters of a single cow at one milking. The flank and udder were brushed and moistened. The hands of the milker were smeared with vaselin and the milk was caught in four-inch glass funnels lead- "Jeiisen. See No. 9. "Babcock aud Russell. Add. Kept. Wis. Exp. Sla. 15:77 (189S). New York Agricultural Experiment Station. 181 ing into glass bottles, all of which had been carefully steamed. The milk was taken at once to the laboratory and placed under the influence of chloroform and formalin at 99° F. (37° C). Table IV. — Compakison of Effects of Foemalin and Chloroform: xtpow Activity of E>zymes. MILK DRAWN OCT. 10. In 100 lbs. total nitrogpn. t Germ'cidp u":ccl. FoimiiliU Clilm oform. a. I per ct. 4 per ct. II II II Age. Total sriUiole nitrogen. Niiriigtri in albunio^es and peptones. Nitmj^enln amides. Number of germs per cc. Days. Lbs. Lbs. Lbs. Fresh 14 87 1 37.50 20.47 17.03 14 53.09 23.84 29.25 51 42 50.26 2S.19 22.07 42 64.22 24.57 39.65 38 77 59.18 35.87 23.31 1 77 72.57 33.63 38.94 5 152 69.94 22.12 47.82 * 152 67.62 43.58 24.04 20 II Fresh • • 6 II 14 21.28 12.81 8.47 IV 14 35.46 21.97 13.49 9 II 42 23.01 13.73 9.28 6 IV 42 36.56 21.36 15.20 4 III 77 24.32 16.67 7.05 IV 77 42.96 27.78 15.18 1 III 152 24.07 17.70 6.97 1 IV 152 43.59 31.40 12.19 2 Fresh ■ — • 232 V 14 37.42 20.98 10.44 VI 14 49.83 25.63 24.20 3 V 42 42. S9 24.96 17.93 3 VI 42 03.58 27.36 30.22 IS V * 77 53.50 31.95 21. .55 2 VI 77 08.50 31.12 37.38 1 V 152 48.70 27.96 20.74 VI 152 63.89 30.10 27.79 ♦Bacteria present In large niiniV)prs. NoTK.— I and II, front right quarter; III and IV, front left; V and VI, back right: VII and VIII, back left. 182 liAroKT OF THE ChKMICAL DiiPARTMENT OF THE Table IV.— Coyiihiued. In '00 lbs. total nitrogen. G»rnili •Ide used. / - Total soluble Nltroneii in albunioses and ' Number or germs per Formalin Chloroform Nltrogf^n in 0.1 per ct. 4 per cent. Age. iiurogrn. peptones. amides. cc. Days. Lbs. Lbs. Lbs. Fresh 14 14 138 3 VII 32.41 50.67 19.63 29.42 12.78 21.25 VIII 3 VII 42 42 40.12 57.23 27.47 29.72 12.65 27.51 2 VIII 1 VII 77 77 47.00 64.38 29.40 39.88 17.60 24.. 50 VIII VII 152 152 49.91 62.48 34.37 39.75 15.-54 23.73 VIII 3 From the above results we see that the number of bacteria in all the bottles remained very low. In all cases the decompo- sition has gone on more slowly in the presence of formalin than with chloroform, as is clearly shown by the following tabulated summary of results. Table IV A. — Average of Four Quarters. Total RoluV)le nitrogen. With chloroform. Per ct. 47. ,26 55. ,39 62 ,10 60.63 Age. With formalin. Days, Per ct. 14 32.15 42 39.07 77 46.00 152 45.62 In the article by Jensen previously referred to, he notes the same relation in the action of these two substances and he is inclined to hold the view that 0.1 per ct. formalin completely inhibits the action of galactase but allows bacterial enzymes to work. If this view is correct, we must consider that over 70 per ct. of the decomposition here produced in the presence of chloroform is caused by enzymes other than galactase. It seems hardly possible that sufficient bacterial enzyme could have been formed in the cases of No. Ill to account for the changes observed in the presence of formalin. The milk in this quarter of the udder was unusually free fiom bacteria, having ^EW York Ageicultckal Experiment Station. Ibo been caught under most favorable conditions and placed under the influence of formalin within a few minutes. CONNECTION BETWEEN BACTERIA IN THE UDDER AND ENZYMES IN THE MILK. Previous investigators^^ have noted that there is considerable difference in the rate of change caused b}^ enzymes in different samples of freshly drawn milk. These differences have been attributed to variations in the enzyme-forming activity of the milk glands, but we have been led to look for another explana- tion of these irregularities. The production of enyzmes on the part of certain classes of bacteria is well known, but the bacterial formation of enzymes in the udder, able to perform work in cheese ripening, is a possibility which has not been seriously considered. The work of Ward-^ has called attention to the fact that in many cases the interior of the udder is inhabited by certain microorganisms which find the conditions favorable to their continued uevelopment. In working -with certain Station cows we have found that in some cases large numbers of germs were present in the milk last drawn. This condition existed when- ever examinations were made during a period of some months. By comparing the germ content of the whole mess of milk, after rejecting the milk first drawn, with the germ content of the milk last drawn, or strippings, it is often found that the number present in the whole mess exceeds that in the strippings by an amount hardly larger than would be expected as a result of unavoidable contamination during milking. This is shown in the following table which gives the number of bacteria found per cubic centimeter in the whole mess and in the strippings from each (]uarter of a single cow at three successive milkings. In all cases the first few streams from each quarter were rejected. "Babcock and Russell. See No. 10. '"Ward. Bui. No. 178 Cornell Exp. Sta. (1900). 1S4 IIkport of the Chemical Dkpariment of tue "■ Taule V. — XuMBEB OF Bacteeia Per Cubic Centimeter in Whole-Milk AND Strippings. Front left Front right Baok left Back rlt;ht quarter. quarter. quarter. quaner Whole- Strip- Whole- Strip- Whole- Strip- Whole- Strlp- nillk. pings. milk. pings. milk. pings. milk. pings. June 11, p. m 53 2G 5G 140 173 401 710 June 12, a. m 646 244 210 429 442 403 G29 1870 June 12, p. m 88 22 30 305 90 105 789 975 These data strongly support the idea that the interior of the udder in such cases is seeded with these organisms, which are generally yellow cocci, capable of liquefying gelatin. Most striking are thase cases in which the interior of certain quarters of the udder is highly contaminated with certain organ- isms for long periods, while, at the same time, one or more quarters of the udder in the same animal may remain comi)ara- tively free from germ life. In the case of the cow used in collecting the data shown in the above table, examinations of the strippings were made extending over four months. Samples were collected by catching one of the last streams from each quarter in a sterile test tube, except in a few cases in which they were drawn with a sterile milking tube. The samples were taken at once to the laboratory and plates prepared containing 1 cc. and 0.5 cc. of the milk. The results are shown in the following table; Table V A. — Number of Bacteria Per Cubic Centimeter in the Strippings of Cow No. 8. Front Front Baok Back Date. left right left right quarter, quartt-r. quarter. quarter. May 26 22 296 372 June 11, p. m 26 56 173 716 June 12, a. m 244 429 493 1870 June 12, p. m 22 305 105 975 July 12 22 48 2106 488 July IS 36 55 280 86S July 27 211 10 3SS 628 Sept. 7 391 3684 631 656 Sppt. 20 132 450 356 9967 Oct. 10 6 87 138 232 This table shows that in general the strippings from the back right quarter had a germ content of .")00 to 800 per cubic centi- Nkw York Agkicultdkal Expkkiment Station. 185 meter; the back left quarter had slightly less; the front loft quarter had often less than 100 per cubic centimeter, and the front right quarter but little more. Making allowance for the work done by galactase, the milk from different quarters of the udder of the above mentioned cow should show different rates of chemical change proportional ta the number of germs present in the respective quarters of the udder, if these changes are to be associated with contamination within the udder. The results already given in Table IV, under chloroform, relate to this point. The quarters of the udder are there designated as follows: II, front right; IV, front left; VI, back right; VIII, back left. The second determination was made in the presence of 4 per ct. of chloroform. In order to obtain sufficient material for a large number of analyses, three sue* cessive messes of milk were collected and united. Care was taken to reject the fore-milk and keep out bacteria from other sources. The following table shows the results in this test up to 15 weeks: Table V b. — Soluble Nitrogen Formed in Milk from Different Quarters of Udder. milk drawn june 11 and 12. Soluble nitrogen in 100 lbs. total nitrogen. AoE OF Milk , When Analyzed. Front left quarter. Front right quarter. Back left quarter. Back right quarter. Days. Lbs. Lbs. Lbs. Lbs. 7 27.01 29.23 48.23 46 . CG 21 36.25 39.75 5G.61 56. (;6 35 36.65 40.60 59.37 57.54 49 lost 40.48 60.68 59.63 105 lost 55.31 77.29 71.63 The results given in Tables IV and Vb show in a general way that there is a relation between the numbers of bacteria present in the udder and the rapidity with which the milk pro- duced there undergoes self-digestion in the presence of chloro* form or formalin. It may be held that the presence of these bacteria has merely stimulated the production of an extra amount of galactase, but 180 Report of the Chemical Departmunt of ihe many of these bacteria are able to bring about the liquefaction of gelatin, a fact which suggests that they have played a part in enzyme formation within the udder. However, it is impossible to assign even an approximate value to the work performed by bacteria within the udder in the production of their enzymes, until we understand the conditions which relate to the normal formation of galactase. V. COMPARISON OF RIPENING PROCESS IN CHEESE MADE ^^•ITH CHLOROFORM AND IN NORMAL CHEESE. Previous attempts to study the part played by enzymes in cheese ripening have proceeded indirectly by a study of enzyme action in milk or have been carried out with cheese in a frag- mentarj' manner. In addition to the early work of Adametz, Babcock and Russell report that they have observed the changes that have taken place at the end of about a year in a cheese containing chloroform. They also added rennet to milk con- taining ether and determined the general changes taking place in the coagulum. Jensen" also reports the changes taking place in a cheese to which he had added trypsin and ether. How- ever, so far as we can learn, no cheese has, hitherto, been pre- pared under conditions essentially normal except for the presence of an anaesthetic, and been kept for a long period com- pletely under the influence of that anaesthetic, with S3^stematic chemical and bacteriological examinations at frequent intervals. METHOD OF MANUFACTURE AND SAMTLIXG. The preparation of a chloroform cheese presents no extreme difficulties. Chloroform added directly to the milk tends to settle to the bottom but the stirring which accompanies the manufacture serves to keep it distributed without any consider- able loss from evaporation. The addition of rennet at 84 to 88° F. (29 to 31° C), cutting and heating to 1)8 to 100° F. (37 to 38° (\). ]>rof('<*d in the usual way, except that both the curdling ="JcU!ieu. Tidakr. for Fijsik. oy Kcini, 2:U2-114 (1S9T). New York Agricultural Experiment Station. 187 of the milk and the expulsion of whey take place more slowly than in normal cheese. The expulsion of the whey is especially prolonged because of the absence of acid, and the moisture con- tent of the resulting cheese may be somewhat higher than in a flrst-class normal Cheddar. After the whey is drawn and the curd is fairly well drained, it is put to press with or without previous salting. In making more than a dozen of these cheeses at different times, we have added to the milk from 2 to 5 per ct. of chloro- form by volume, and we find that the percentage of chloroform by weight in the resulting cheese mass is about three times the figure given for the milk. The cheese is kept continuously under pressure 18 to 24 hours, and is then transferred to a room with a temperature varying only one or two degrees from 60° F. (15.5° C.) and placed under a bell jar in an atmosphere of chloroform. The moisture of cheese under bell jars remains fairly uniform. After testing a number of receivers we have settled upon bell- jars, or carefully soldered cans which are inverted over the cheese, and fit into a groove in a heavy wooden base. . The base is first boiled in paraffin to fill all the pores, and melted paraffin is used as a seal in fastening the cover into the grooves, thus reducing the loss of chloroform and moisture to insignificant amounts. At regular intervals the cover is moved and samples taken with a sterilized tryer for chemical and bacteriological analysis. The former includes a quantitative determination of the chloro- form present in the cheese. To replace the small amounts lost by leakage and evaporation, measured amounts of chloroform are added to a dish within the container at the time of each examination. DECOMPOSITION IN CHEESE UNDER CHLOROFORM COMPARED WITH that IN NORMAL CHEiESE. In order to get an idea of the changes brought about by the combined influence of all the enzymes present at the time a 188 litPoRT OF THE Chemical Department of the cheese is made, 3.5 lbs. of chloroform were added to 125 lbs. of night's and morning's milk having the degree of acidity suitable for Cheddar cheese-making. One-half ounce of Hansen's liquid rennet was added at 88° F. (31° C), and the cheese made as described above. One-half of the resulting curd, without salting, was pressed into form of a Young America cheese. On the third day it was foimd to contain 35 per ct. of water and 15 per ct. of chloroform. As a basis for comparison there is also given the analysis of a normal cheese ripened at the same temperature and having originally about the same percentage of moisture. However, since under normal conditions the moisture in a cheese rapidly decreases, while in the chloroform cheese this factor remains practically constant, there is also given the analysis of a Ghees'^ normal in every way except that it was coated with a layer of paraffin to lessen the loss of moisture. Table VI. — Comparison of- Normal Cheeses, Cured with axb without Paraffin Covering, with a Cheese Made and Cured with Chloroform. Total water-soluble nitrogen formed for 100 lbs. nitroRen In cheeRe. CoxDiTioxs OF CuRixo. 2 Weeks. 1 month. 2 months. 6 months. 12 mouths. 15 months. Cbeese No. 31A, cured under nor- mal conditiou.s. . 11.50 IS.r.O 25.10 33.70 37.30 3S.66 Cheese No. 31B, covered with paraffin 12.50 19.30 25.40 37.80 40.90 44.14 Cheese No. 30A, made and cured with chloroform 5.30 5.70 8 20 14.50 22.60 27.70 In Tables VI, VII, VIII and IX, the figures given for total water-soluble nitrogen represent the amount rendered soluble after the cheese was taken from the press. Samples of the green cheese fresh from the press were analyzed, and it was found that the amount of soluble nitrogen varied considerably in different cheeses. Therefore, for the sake of more accurate comparison, the amounts of water soluble nitrogen found in the ISLW YoKK Agkiclltukal Expkkiment Station. 189 green clieese have been deducted and so are not included in the figures presented in these tables. The data in Table VI show that at the end of one month the water-soluble nitrogen in the normal cheese was more than three times that contained in the chloroform cheese; gradually the difference decreased until at the end of 15 months the total decomposition in the case of the chloroform cheese amounted to 27.7 per ct. of the total nitrogen, while in a normal cheese of the same age the amount was 38.66 per ct. The enzymes present in this cheese were therefore able under favorable circumstances to accomplish about 72 per ct. as much decomposition of casein as occurred in a normal cheese. That they accomplish this fraction of the work under ordinary conditions does not neces- sarily follow. These results show merely that the peculiar con- ditions of manufacture in the presence of chloroform were not such as to prevent the enzymes from rendering cheese-casein soluble. INFLUENCE OF SMALL AMOTDsTS OF ACID ON ENZYME ACTION. In the ordinary process of manufacture there is a gradual formation of acid within the mass through the action of becteria. In the preceding experiment acid was necessarily absent. To remedy this, another cheese was made like the preceding, except that lactic acid was added. As before, 3.5 lbs. of chloroform were added to 125 lbs. of night's and morning's milk, sufficiently acid for cheese-making. This was curdled by one-quarter ounce of Hansen's liquid rennet added at 86° F. (30° C). After cutting the curd and applying heat, pure lactic acid was added in small quantities at a time until the whole amounted to nearly .2 per ct. of the milk used. One-half of the resulting curd, unsalted, was pressed into a Young America cheese which, fresh from the press, contained 32 per ct. of water and 15 per ct. of chloroform. 190 Repokt of the Chemical Department of the The results of the examinations are shown below: Table VII. — Comparison of Chloroform Cheeses Made with and WITHOUT Lactic Acid. Total water-soluble nitrogen formed for 100 lbs. nitrogen in cheese. Conditions of Making r- , AND Curing. l month. 2 months. 3 mouths. 6 months. 9 months. 12 months. Cheese No. 30A, made and cured with chloroform. 5.70 8.20 11.60 14.50 19.50 22.60 Cheese No. 32A, made and cured with chloroform and lactic acid. 5.70 9.40 14.00 20.60 23.20 31.65 It will be seen that in cheese 32A the amount of soluble nitro- gen is greater than in .30A after the first month and continues to become greater up to the end of 12 months, the age of 32A at its last anah'sis. This more rapid ripening in 32A took place in spite of the fact that only one-half as much rennet was used in 32A as in 30A. Acid appears to favor enzyme action. INFLUENCE OF SALT UPON ENZYME ACTION. In the two preceding experiments it has been noted that one-half the curd was pressed without salting and the results previously given represent the changes taking place in unsalted cheese. However in the manufacture of Cheddar cheese, salt is never omitted and, in order to make the comparison between the chloroform cheese and normal cheese complete, the aduition of salt is required. In each of the two experiments, one-half of the curd was salted just before putting to press, the first receiving 2 ounces and the second 2\ ounces. In each case the percentage of chlo- roform and water was essentially the same as in the correspond- ing unsalted portion. The results of analysis are shown in the following table. To facilitate comparison, the results from the unsalted portions are also repeated. New York Agricultural Experiment Station. 191 Table VIII. — Comparison of Cheeses Made and Cured with Chloro- form, Salted and Unsalted. Total water-soluble nitrogen formed for 100 lbs. nitrogen in cheese. Conditions ok Corino. 1 mo. 2 mos. 3 luos. 6 mos. 9 mos. 12 mos. 15 mos. (1) Without lactic acid: Clieese No. oOA, made and cured with chloro- form — not salted 5.70 8.20 11.60 14.50 19.50 22.60 27.70 Cheese No. SOB, same as 30A, but salted 2.25 3.20 5.50 7.80 11.00 17.20 24.00 (2) With lactic acid: Cheese No. 32A, made ■ and cured with chloro- form and lactic acid — not salted 5.70 9.40 14.00 20.60 OO OQ 31.65 Cheese No. 32B, same as 32 A, but salted 3.00 4.90 6.70 9.75 12.45 19.65 From the results here given it is seen that salt in the propor- tion usually present in cheese exerts a strong repressing influ- ence upon the activit^^ of the enzymes present. On comparing this effect of salt in the case of the cheese containing added acid with the cheese in which acid was omitted, it is seen that acid favored enzyme action here also as well as in unsalted cheese. The results of out work up to this time appear to show-, (1) that the use of chloroform excludes bacterial action in milk and cheese and limits the work of ripening to those enzymes con- tained in milk when made into cheese; (2) that the presence of salt noticeably decreases the effect of such enzymes; (3) that the presence of two-tenths of one per ct. of lactic acid increases the ripening action, at least of rennet enzymes; (4) that the percentage of cheese-casein made soluble by the enzymes under consideration in nine months (which may be regarded as the extreme limit of the commercial life of Cheddar cheese, kept under usual conditions) is about 12 per ct., or one-third the amount of soluble nitrogen found in normal cheese; and (5) that the amount of ripening caused by enzymes present in the milk when made into cheese is apparently more limited than was previously supposed. We may say that the limited part apparently taken by such enzymes in ripening cheese is a result we did not anticipate when 102 RkPORT of TirE CuKMIOAL DtiPAKTME.VT OF THIC uiidei'takiDg the work. We have additional experimental work under way for the purpose of testing these results more rigidly. DIFFERENCE IN CHARACTER OF CHEMICAL CHANGES IN NORMAL AND IN CHLOROFORM CHEESE. An examination of the detailed data secured with normal and with chloroform cheese shows clearly a marked difference in the character of the changes taking place in the soluble niti-ogen- compounds. This difference is seen if we study the amounts of albumo'ses and peptones in relation to amides, and also the relative amounts of ammonia found. The following tabulated comparison in case of cheese 34C and 34B, which were made with and without chloroform from differ- ent portions of the same milk, illustrate the points in question. Table IX. — Showing Difference in Character of Chemical Changes IN Normal and in Chloroform Cheese. Character of Cheese. Age. Months. Cheese 34C — normal 1 Choese 3-lB — chloroform 1 Normal cheese 1V> Chloroform cheese IV2 Normal cheese SY2 Chloi'oform cheese 3Y2 Normal cheese 5I/2 Chloroform cheese 0V2 Noi-mal cheese 7 Chloroform cheese 7 Normal cheese 9 Chloroform cheese 9 stated in a general way, these results show (1) that, in cheese made and cured with chloroform, the amount of albumoses and peptones is largely in excess of the amount of amides; (2) that the reverse is true in normal cheese; and (3) that ammonia appears in normal cheese much earlier and in larger quantities than in chloroform cheese. N. In albu- moses and pep- tones. N. In amides. R'itio (1) to (2). fN'. In anmiouli (1) (2) 2.95 5.42 1:1.80 .86 3.71 0.86 1:0.23 2.51 8.49 1:3.40 1.29 7.31 1.82 1:0.25 5.37 12.60 1:2.40 2.51 10.20 3.22 1:0.31 4.97 18.50 1:3.70 3. 38 12.40 4.73 1:0.39 3.0s 20.10 1:6.50 4.42 10.90 8.11 1:0.74 2.70 23., 50 1:8.70 4. 87 12.52 11. GO 1:0.93 Kew York Agricultural Experiment Station. 193 Making a detailed comparison, we note the following points: (1) In the normal cheese at the age of one month, the amount of amides was 1.8 lbs. for each pound of albumoses and pep- tones. This ratio increased until at nine months it was 8.7, nearlj' five times as great as at the end of one month. (2) In the chloroform cheese, the amount of amides was not quite one-fourth of the amount of albumoses and peptones at the age of one month. The relative amount slowly increased, until at the end of nine months the amount of amides was nearly equal to that of albumoses and peptones, (3) In chloroform cheese, no ammonia had appeared at the end of nine months; in the normal cheese, nearly one per ct. of the total nitrogen was present as ammonia at the end of one month and this amount steadily increased. - From these results it is seen that, in a normal cheese, the amides steadily increase, while the albumoses and peptones increase for some months and then decrease. In a chloroform cheese, the different classes of compounds under discussion all increase continuously from the beginning for many months. In normal cheese, traces of ammonia appear at an early stage of ripening, while, in chloroform cheese, the first traces usually appear only after the lapse of six months or more, and the in- crease is very slow, so that even after a year only minute amounts are present. FoT these data it appears that there is some agent at work in normal cheese which is not active in cheese made with chloro- form. Just what this additional factor is our present data do not explain, but our efforts are being directed to the task of identifying this agent. 13 CONDITIONS AFFECTING AVEIGIIT LOST BY CHEESE IN CURING* L. L. VAN 8LYKE. SUMMARY. I. The loss of weight by cheese in curing has not received systematic study in America under carefully controlled conditions. //. Equipment far investigation. — Six curing-rooms have been so built and equipped as to keep temperature and moisture under control. Each room is kept at a fixed temperature, and the different degrees represented in the work are the following: 55°, G0°, 65°, 70°, 75°, 80° F. The temperature varies only one m' two degrees from the desired point, and then only for brief periods. The moisture is kept mostly between 70 and 80 per ct. of saturation. The method is given for determining proportion of moisture in air, with necessary tables. ///. Conditions affecting loss of weight in cJieese-curing. — The weight lost by cheese in curing is due almost entirely to evapora- tion of moisture from cheese, except at temperatures above 70° F., when there may be some added loss due to leakage of fat. The rapidity and extent of loss per 100 pounds of cheese vary with the following conditions: (1) The percentage of moisture originally present in the cheese. The more moist the cheese, the greater and more rapid is the loss of weight. (2) The texture of cheese. The more open the texture, the greater the loss of moisture. (3) Temperature. Loss of weight increases with increase of temperature. ,(4) Size and shape of cheese. *A reprint of BuTletin No. 207. 104 New York Agricultural Experiment Station. 195 Loss of weight increases, when the size of cheese decreases. Increase of height or diameter of cheese decreases loss of weight. (5) Proportion of water-vapor in air. The greater the moisture in the air of the curing-room, the smaller is the loss of weight of cheese. lY. Some practical applications. — (1) Value of water in cheese to dairymen. Water, put in cheese in right proportions and kept there, is money to the dairyman, increasing amount of cheese to be sold. (2) Moisture in cheese in relation to quality. Exces- sive loss of moisture in curing seriously injures commercial quality of cheese. (3) What percentage of moisture should cheese have? When consumed, cheese should have not less than 33 per ct. moisture. If cured at low temperatures, larger amounts can be held to advantage of quality. (4) Value of water in cheese to consumers. Cheese with fairly laiige amount of moisture, cured at proper temperature, is more palatable to most consumers. Less rind is thrown away. (5) Variation of loss of moisture with size of cheese. As small cheese loses moisture more rapidly than larger cheese, greater pains must be taken with small cheese to prevent excessive loss of moisture. (6) Loiss of moisture and loss of fat. To avoid loss of fat by leakage and excessive loss of moisture, cheese should not be kept above 70° F. for any length of time. y. Prevention of loss of moisture in curing cheese. — Three systems have been proposed: (1) Immediate sale and removal of newly- made cheese. In this case the buyer assumes responsibility of curing in cold-storage and secures all the benefits. (2) Central curing-rooms, located so as to care for product of several factor- ies and equipped with complete facilities for controlling temper- ature and moisture. This system has greater promise than any other. (3) Special curing-rooms in each factory are desir- able when central curing-rooms cannot be had. Details are given, taken mainly from Bulletin No. 70 of the Wisconsin Agri- cultural Experiment Station, describing construction of curing- rooms and various kinds of sub-earth ducts. 196 KEPOiiT OF THE Chemical Department of the I. INTRODUCTION. It is well known among cbeesemakers that cheese begins to lose weight immediately from the time it is taken from the j)ress and placed upon the shelves of the curing-room; this loss con- tinues indefinitely. While there has been some study in Europe relating to the conditions and extent of loss of weight in cheese- curing, the results thus obtained are not generally applicable to the conditions prevailing in this country. Some study of this question has been made in America, but it has been rather desultory in character, lacking in systematic plan and thorough- ness, and under circumstances not permitting careful control of conditions. II. EQUIPMENT FOR INVESTIGATION. CURING-ROOMS. For the past three years at this Station we have been making a systematic study of the various conditions that affect loss of weight in cheese during the progress of curing. The special equipment in the way of cheese-curing rooms has given us unusual opportunities to carry on such study under well con- trolled conditions. We have a block of six distinct curing-rooms, separated from the outer walls of the building by a passage four feet wide. The rooms are farther insulated by double walls and air spaces on every side of each room. These rooms are 9 by 10 feet and about 8 feet high, and the wall space on three sides is provided with shelves 12 inches apart. OONTROL OF TEMPERATURE. The temperature and moisture in each room can be controlled independently of the other rooms. It is possible to obtain a range of temperature varying from 40 to 90 degrees Fahrenheit in every room. Each room is provided with a hot-air flue from below and a cold-air flue above, leading from the chamber in the attic, which contains ammonia expansion coils and brine tanks. These two flues, one for cold and one for hot air, are closed by New York Agkicultural Experiment Station. 197 dampers, and these dampers are operated by means of com- pressed air tubes controlled by metallic thermostats. There is also a ventilating flue in the ceiling of each room. The thermostat is fixed so as to register some definite temperature in each room. For example, in one room the thermostat is set at 70° F. When the temperature falls one degree below 70° F., the thermostat is affected in such a manner that a valve is turned and this causes compressed air to close the cold-air damper in the ceiling and to open the hot-air damper in the floor, thus restoring the temperature to 70° F. On the other hand, when the temperature rises to 71° F., the cold-air flue in the ceiling is opened and the hot-air flue is closed, when the temperature soon begins to drop. Thus we have an alternate admission and exclusion of hot air and cold air, causing the temperature to rise a little above or fall slightly below the given point at which it is desired to hold the temperature of the room. So delicate is the operation of this system that merely breathing upon the thermostat will open the cold-air flue, while fanning the thermostat will open the hot-air damper. We are able, therefore, by this system to hold temperature within a very limited range. Under most favorable conditions, the limit of variation is only two degrees. Even with a much wider temporary variation the temperature of the interior of a cheese would not be affected to the extent of more than a small fraction of a degree, as we have shown by placing a thermometer inside a cheese and keeping it there under observation for several weeks. CONTROL OP MOISTURE), It is more difficult to control moisture than temperature, so as to hold it within narrow limits. The most practicable and efficient method we have found adapted to our conditions is to make use of yard-wide pieces of coarse felt, having a strong capillary power. One end of the felt dips in a trough of water situated near the top of the room and the lower end drops into a trough placed on the floor. The water is sucked in by the felt at the upper end and gradually distributes itself throughout 198 Kepokt of thb Chemical Department of the the whole piece, the excess of water dripping into the lower trough. It is necessary occasionally to boil these pieces of felt cloth in water slightly acidulated with some acid, like acetic or hydrochloric, in order to remove mineral salts tha.t accumulate and interfere with capillary action. The use of distilled or rain water would obviate this difficulty. Thus far we have kept the moisture as nearly as possible at 75 i^er ct. of saturation, though variations of 5 per ct. below and 10 per ct. above may occur at times. The natural tendency is toward a higher percentage of relative moisture at lower temperatures. METHOD OF DETERMINING MOISTURE IN AIR. The relative amount of moisture in air can be determined by means of an instrument known as a hygrometer, of which there are several forms. One form indicates the percentage of mois- ture directly by means of a needle or hand; this is the most convenient kind of hygrometer, and is probably sufficiently accurate for ordinary purposes. A more accurate instrument consists of two sensitive, standard thermometers. The bulb of one is exposed to the air directly, like any thermometer, and is known as the dry-bulb or dry thermometer, while the other has its bulb wrapped in a piece of muslin to hold water, and is known as the wet-bulb or wet thermometer. The wet ther- mometer should be fixed in a frame that enables one to whirl it easily. The form of hygrometer used by us is made by Julian P. Friez, Baltimore, Md. The dry thermometer indicates the temperature of the air in the room. The wet thermometer, properly used, indicates a lower temperature than does the dry thermometer, because the water in the muslin bound about the bulb evaporates and the evaporation is accompanied by a lower- ing of the temperature immediately around the bulb. The less moisture there is in the air, the more rapidly does evaporation take place and the lower is the temperature indicated by the wet thermometer. The greater the moisture in the air, the less rapid is the evaporation and the smaller the difference between New York Agkiclltukal Expekimkjst Station. 11)9 the temperature indicated by the dry and wet thermometers. When the two thermometers indicate the same temperature, then there is no evaporation taking place from the bulb of the wet thermometer, because the air is saturated with moisture, that is, holds as water-vapor all it can at that temperature. If the moisture is increased beyond this point or if the tempera- ture is lowered, some of the water-vapor will be condensed into visible drops of water. In order to use a hygrometer for the purpose of ascertaining the proportion of moisture in air, we note first the temperature indicated by the dry thermometer. Then we dip in water the bulb (wrapped in muslin) of the wet thermometer, whirl it vigorously for one or two minutes, and then quickly read the temperature. The whirling is for the purpose of quickly causing evaporation. It is well to repeat the whirling two or three times, noting the temperature of the wet thermometer after each whirling. The different temperature readings should agree if the whirling operation is equally thorough each time. Pains must be taken to keep the muslin about the bulb moist during the different whirlings. After getting the temperatures of the two thermometers, we subtract the number indicating the temperature of the wet thermometer from the number showing the temperature of the dry thermometer. Then we turn to pre- pared tables of figures and find the column of figures, at the top of which is the difference obtained by the foregoing subtraction. If the exact figure is not there we take the one nearest it. We then follow down this column until the figure is found opposite the number in the left hand column which is the same as the temperature indicated by the dry thermometer. The number thus found indicates the relative amount of moisture in the air, or percentage of saturation; that is, how much moisture the air actually holds compared with what it could hold at that temperature if saturated. The preceding statements can be better understood by use of a specific illustration. Suppose we find by actual trial that the 200 Report of the Chemical Department of the readings of the two thermometers of our hygrometer are as follows: Dry thermomptor TO^F. Wet thormomotpr 65° F. Difference 5°F. We turn to the tables given at the end of this bulletin (taken from Weather Bulletin No. 127, U. S. Dept. Agr.) and look in the upper horizontal row for the number 5. HaAing found this, we follow down the column until we come opposite the number indicated by the dry thermometer (70) in the vertical column at the extreme left. This brings us to the figure 77, which indicates the relative amount of moisture in the air; in other words, the air contains 77 i)cr ct. as much moisture as it can hold at 70° F. III. CONDITIONS AFFECTING LOSS OF WEIGHT IN CHEESE-CURING. The loss of weight in cheese during the process of curing under proper conditions may be regarded for practical purposes as being due entirely to the evaporation of water from the cheese. Of course, the mechanical loss of fat by exudation from cheese kept at high temperatures must be considered, but with proper control of "temperature such loss will not take place. The small amount of loss due to the formation and escape of carbon dioxide and other gases from cheese can be neglected for the purpose we now have in view. The rapidity and extent of loss of moisture in cheese during the process of curing vary with several conditions, chief of which are the following: (1) The percentage of moisture originally present in the cheese. (2) The texture of the cheese. (3) The temperature of the curing-room. (4) The size and shape of the cheese. (5) The proportion of water-vapor present in the air of the curing-room. JSIew York AGKicciyruEAL Experiment fcjiA'iioN. 20 1 LOSS OF MOISTURE AS INFLUEiNCED BY THE PERCENTAGE OF WATER PRESENT IN GREEN CHEESE. In presenting the results of our study under this division of our subject, we will first make use of some extreme cases, in which the percentage of water in the cheese varied from 55 to 35. In the following table we give the percentage of water originally present in the cheese fresh from press and the amount of water lost per 100 pounds of cheese for each of four weeks, the conditions of temperature and moisture of air being the same for the different cheeses. Table I. — Loss of Moisture in Cheeses Containing Different Per- centages OF Water. Water lost by 100 lbs. of green cheese. Water in 100 Lbs. Green Cheese. In 1 week. In 2 weeks. In 3 weeks. In 4 weeks. Lbs. Lbs. Lbs. Lbs. Lbs. 55 9.0 11.2 12.3 16.8 50 5.5 9.2 11.0 12.9 45 4.5 6.3 8.0 9.5 35 3.3 4.2 4.9 5.7 An examination of these figures suggests the following statements: (1) There is a marked general tendency for very moist cheese to lose water more rapidly than cheese having less moisture, other conditions being uniform. Thus, the cheese containing 55 per ct. of moisture lost nearly three times as much moisture by evaporation each week as did the cheese containing 35 per ct. of water, and nearly twice as much as the cheese containing 45 per ct. of moisture. (2) At the end of four weeks, the cheese containing 55 per ct. of moisture had lost about one-third of its water; the one with 50 per ct. had lost one-fourth; the one containing 45 per ct., one- fifth; and the one with 35 per ct., one-sixth. It is thus seen that the more moist the cheese the greater is the proportion of its water lost by evaporation; and, hence, the moisture in the different cheeses tends to become more nearly alike. However, they would not all reach the same condition of moisture-content, except under very unusual conditions. 41.7 88.7 37.6 35.4 5.3 4.6 4.5 4.2 202 Report of tiik Chemical Departmp:nt of the The results presented in such extreme cases fire full of inter- est, but do not have practical application to conditions com- monly present in cheese making. There is, however, a practical question in this connection to be considered later. We will consider one more illustration, in which the varia- tions of moisture in the green cheeses are within narrow limits and essentially similar to cases occurring in factory work. The data in the following figures represent averages obtained with four different lots of cheese. The cheeses weighed about 30 lbs. each. Water in 100 lbs. green 1 cheese, lbs. [ Water lost by 100 lbs. cheese ) in 6 weeks, lbs. ) These data show that the loss of moisture increases as the amount of water in the green cheese increases, even though the amount of moisture in the green cheese varies within compar- atively narrow limits. Variation in other conditions may, of course, interfere with this general tendency. LOSS OP MOISTURE AS INPLUEiNCED BY TEXTURE OP CHEESE. Cheese filled with holes will occupy more volume than the same weight of cheese free from holes. Hence, cheese with such faulty texture has a larger surface exposed for evaporation relative to its weight and will lose more moisture. Then, in addition, the presence of numerous holes in cheese greatly facili- tates the escape of moisture from the interior of the cheese to the surface. This is a partial explanation of the fact that cheese high in moisture loses water more rapidly than cheese containing less moisture. It is well known that cheese contain- ing high percentages of water usually develops holes abun- dantly, especially when cured at or above ordinary temperatures. LOSS OF MOISTURE AS INFLUENCED BY TEMPERATURE'. In our study of the influence of temperature upon loss of moisture we used six different temperatures, viz.: 55°, G0°, 65°, New York Aguiccltukal Experiment Station. 203 70°, 75°, 80° F. In one case a temperature of about 32° F. was employed. The degree of moisture was kept as nearly uniform as possible in the different curing-rooms. In this connection we will present the results secured with cheeses 15 inches in diameter, and weighing, fresh from press, about Go lbs., the usual standard size of the most common tjjye of American Cheddar cheese. Work with the cheeses at 75° and 80° F. was discontinued after 16 weeks. The number of cheeses of this size available for our work has not been suffi- cient to cover the ground as fully as is desirable. Table II. — Loss of Moisture at Different Temperatures. TKMr. Water lost by 100 lbs. of green cheese In OF , CUIilNG- KOOM. 1 week. 2 weeks. 3 weeks. 4 weeks. 8 weeks. 12 weeks. 16 weeks. 20 weeks. 24 weeks. 28 weeks. Deg. F. Lbs. Lbs. Lbs. Lbs. Lbs. Lbs. Lbs. Lbs. Lbs. Lbs. 55 1.6 2.G 3.2 3.7 5.2 6.1 6.8 7.5 8.1 8.6 GO 1.7 2.8 3.4 3.9 5.5 6.5 7.5 8.5 9.3 9.9 Go 1.9 3.0 3.G 4.1 5.8 7.0 8.2 9.2 10.1 10.5 70 2.0 3.1 3.7 4.3 6.0 7.8 9.0 10.1 11.1 12.0 75 2.2 3.3 4.0 4.7 7.2 9.7 11.4 SO 2.4 3.7 4.5 5.2 8.3 11.6 15.5 Attention is called to the following points: (1) At 55° F. the total loss of moisture is less than it is at the higher temperatures. This is true at the end of the first week and continues so through all the weeks following. (2) The loss of weight increases in a marked degree with increase of temperature. During the first four weeks the loss of weight increased about three ounces for each increase of five degrees of temperature between the limits of 55° and 70° F. From 70° to 75° F., the increase was six and one-half ounces, and from 75° to 80° F., the increase was eight ounces for each 5° F. additional. As between 55° and 80° F. the loss increased on an average one ounce per 100 lbs. of cheese for each addi- tional degree Fahr. At the end of two months, comparing 55° and 80° F. the loss increased two ounces per 100 lbs. of cheese for each degree; and at the end of three months, three and one- half ounces. 204 E.EP(ET OF THE ChEMICAL DEPARTMENT OF THE (3) The average weekly loss of weight or rate of loss increases with increase of temperature. This statement can be more clearly understood by means of the subjoined table, which has been prepared from the data given in Table 11. Table III. — Average Weekly Loss at Different Temperatures. Tkmp. Average loss per week. Water lost by lOJ lbs. of green cheese. Lbs. to- OF , ' 1 tal loss Curing- Ist 2d 3cl 4th 2il 3d 4th 5th 6th for six Room. week. week. week. week, mouth, month, mouth, month, mouth, months. Deg. F. Ozs. Ozs. Ozs. Ozs. Ozs. Ozs. Ozs. Ozs. Ozs. Lbs. 55 25.6 16.0 9.6 8.0 6.0 3.6 2.8 2.8 2.4 8.1 60 27.2 17.6 9.6 S.O 6.4 4.0 4.0 4.0 3.2 9.3 65 30.4 17.6 9.6 8.0 6.8 4.8 4.8 4,0 3.6 10.1 70 32.0 17.6 9.6 9.6 6.8 4.8 4.8 4.4 4.0 11.1 75 35.2 17.6 10.2 10.2 10.0 10.0 0.8 80 38.4 20.8 12.8 10.2 12.4 13.2 15.6 An examination of this table shows the smallest weekly loss at 55° F. in every case and a clear tendency for the loss to increase with increase of temperature. (4) It is noticeable that the loss is greater the first week than during any other week. At 55° and G0° F. the loss the first week is equal to the combined losses of the second and third weeks. At the higher temperatures the loss during the first week is nearly equal to the combined losses of the second, third and fourth weeks. (5) The weekly loss decreases continuously as the cheese grows older. This is true at all temperatures. (6) The comparatively rapid loss of moisture during the early stage of curing is entirely consistent with the fact previously shown, that loss of moisture increases with the moisture content of the cheese. The cheese contains its maximum of moisture when new. In addition, the bandage holds considerable water which quickly evaporates. Then, again, the outer surface of the cheese, in drying, begins to harden, the pores of the cheese-cloth filling to some extent with dried matter, and this condition tends constantly more and more to diminish evaporation, provided cracking is prevented. (7) An examination of Table III shows that the cheese at 80° F., after the fourth week had an increased weekly loss of New Yokk Agkiodltukal Experiment Station. 205 weight, while at the lower temperatures the weekly loss fell gradually. This extra loss was due to leakage of fat from the cheese, which was very noticeable on the surface of the cheese and on the shelf. The cheese at 75° F. also lost some fat ,by leakage, as the figures in the table indicate for the second and third months. (8) To illustrate the influence of temperature below 55° F. upon loss of moisture in cheese curing, we give some results secured with cheeses weighing 30 pounds, 13 inches in diameter. "The last weighing was taken when the cheeses were five weeks old. Temperature, degrees F. 32 55 GO 70 Weight, loss by 100 lbs. of cheese in five weeks, lbs., 3.0 4.G 4.6 4.9 The reduction in temperature below 55° F. is seen to be attended with decreased loss of moisture in a marked degree. LOSS OF MOISTXJRD AS INFDUENCEID BY SIZE AND SHAPE! OF CHEESE. We will first present data secured with cheeses having the same diameter but varying in height. ,These cheeses were 7 inches in diameter, being of the type commonly known as " Young Americas." They were made from one vat of milk and subjected to uniform conditions. They were all kept at a uni- form temperature of 65° F. Table IV.— VYeig£ [T Lost ■ BY Cheese OF Varying Height AND Uniform DiAMETEB. TTpTOHT Weight of green cheese. Wa ter lost by 100 lbs. of green cheese In OF Cheese. 1 week. 2 weeks. 3 weeks 4 weeks. 8 weeks. 12 weeks. 16 weeks. 20 weeks. 24 weeks. Inches. Lbs. Lbs. Lbs. Lbs. Lbs. Lbs. Lbs. Lbs. Lbs. Lbs. 3 4.6 3.4 5.3 6.4 7.0 10.7 12.9 13.9 15.9 17.0 4 6.1 3.3 5.1 6.1 6.7 9.7 11.5 13.0 14.0 15. G 5 7.9 2.8 4.2 5.5 6.3 8.3 9.8 11.2 12.6 13.4 G 9.3 2.5 3.9 5.2 6.0 7.8 9.4 10.6 11. G 12.8 7 11.0 2.3 3.4 4.7 5.6 7.4 8.9 10.5 11.2 12.4 ,The data in this table suggests the following statements: (1) The loss of weight was greatest in the cheese whose height was least. The loss decreased with increase of height. Taking 206 Repokt of the Chemical Department of the the total loss for the first four weeks, it is seen that an increase of one inch in height reduced the loss of moisture about 5^ ounces per 100 pounds of cheese. (2) The general tendency in all the cheeses was a decrease in the weekly rate of loss of weight as the cheese grew older. The weekly rate of loss was greater in the smaller cheeses for the first two months, after which the rate was fairly uniform in all the cheeses. We will now consider data furnished by cheeses having approximately the same height but varying in diameter. The' results represent, in case of the smaller cheeses, averages cover- ing from ten to twenty-five separate lots of cheese. Table V. — Weight Lost by Cheese of Varying Diameter and Uniform Height. Diame- ter of cheese. Weight of greeu Clieese. Tem- pera- ture or /^ Water lost by 100 lbs. of cheese. eurins rooms. 1 week. 2 wcel£s. 4 weeks. 8 weeks. weeks. 16 weeks. wet ks. 24 weeks. Inches. Lbs. Deg. F. Lbs. Lbs. Lbs. Lbs. Lbs. Lbs. Lbs. Lbs. 15 7 65 9 SO so 2.4 3.6 3.7 5.2 5.2 7.3 8.3 10.9 11.6 12.7 15.5 14.5 16.3 17.1 15 7 65 9 75 75 2.2 3.1 3.3 4.8 4.7 6.6 7.2 9.2 9.7 11.1 11.4 12.7 14.1 l.-).l 15 65 70 2.0 3.1 4.3 6.0 7.8 9.0 10.1 11.1 11 23 70 3.0 4.2 6.1 7.7 9.2 10.6 11.6 12.4 7 9 70 2.9 4.5 6.2 8.9 10.9 12.2 13.9 14.6 15 65 65 1.9 3.0 4.1 5.8 7.0 8.2 9.2 10.1 13 21 65 2.0 3.4 5.1 6.2 7.7 8.7 9.3 10.2 11 22 65 2.6 3.7 5.3 6.9 8.1 9.5 10.4 11.3 7 9 65 2.5 3.9 5.6 7.9 9.5 10.9 12.1 13.1 15 65 60 1.7 2.8 3.9 5.5 6.5 7.5 8.5 9.3 13 31 60 1.7 2.7 4.3 6.1 7.3 8.4 9.5 11 22 60 1.9 3.6 4.5 6.3 7.5 8.7 9.6 10.5 7 9 60 2.4 3.7 5.5 7.7 9.3 10.6 11.9 12.8 15 65 55 1.6 2.0 3.7 5.2 6.1 6.8 7.5 8.1 13 29 55 1.5 2.7 4.2 5.7 7.2 7.9 8.9 9.4 11 20 55 2.1 3.6 4.6 6.4 7.4 8.8 9.4 10.1 7 9 55 2.2 3.6 5.1 7.2 8.8 9.8 11.0 12.0 A study of the preceding table brings out the following points: (1) In general, at all temiieratures, the loss of weight in cheese New Yoke Agkicdltckal Experiment Station. 207 increases when the diameter of the cheese decreases. Taking the cheeses having diameters of 15 and 7 inches respectively, at the age of four weeks, we see that at a temperature of 80° F. the smaller cheese has lost 2.1 pounds more per hundred pounds of cheese than has the larger cheese. (2) The difference in loss of weight between cheeses of differ- ent diameters is greatest at 80° F. and gradually decreases with decrease of temjjerature. Illustrating this point with the 15 and 7 inch cheeses at the age of four weeks, we have the small cheese losing more than the large cheese by the following amounts per hundred cheese: at 80° F., 2.1 lbs.; at 75° F., 1.9 lbs.; at 70° F., 1.1) lbs.; at G5° F., 1.5 lbs.; at G0° F., 1.6 lbs.; at 55° F., 1.4 lbs. (3) At 65° F. we find that an increase of two inches in diam- eter reduces the loss of weight about one-half pound per hundred Ijounds of cheese, when the cheeses are four weeks old. ^Vhen the cheeses are 16 weeks old, the decrease in loss of weight is one pound for an increase of two inches in diameter. We have two additional illustrations to present, in which cheeses were made from the same milk and cured under the same conditions. We present these data in the following ta.ble: Taule VI. — Weight Lost by Cheeses of Different Diameters. Diame- Weight of green cheese. Tern pera- ture of curing- room. Water lost by 100 lbs. of cheese in ter of cheeses. 2 weeks. 3 weeks. 5 weeks. 7 weeks. 8 weeks. 12 weeks. 16 weeks. 20 weeks. Jnc/ies. Lbs. J3eg. F. Lbs. Lbs. Lbs. Lbs. Lbs. Lbs. Lbs. Lbs. 13 36 55 2.9 3.G 4.5 4.9 5.4 6.4 7.2 8.1 7 10 55 3.9 4.9 6.3 6.9 7.2 8.7 9.9 11.5 13 29 60 2.8 3.7 4.7 5.7 6.0 7.3 8.2 9.3 7 9 60 3.8 4.8 6.7 7.9 8.2 9.7 11.2 12.5 LOSS OF MOISTURE AS IXFLUEiXCED BY rROPORTION OP WATEIR- VAPOR PRESENT IN AIR OF GURING-ROOM. The relative amount of moisture in air, or, more properly, the degree of saturation, exercises a marked influence upon loss of water in cheese-ripening. While we have not carried systematic investigation far in this line, we can j)n s nt data that will 208 Rkpokt of the Ouemical Depaktment of the clearly illustrate tlie influence of this factor. Cheeses, which were made from the same milk, were placed in the curing-room at 60° F. One cheese was kept on the shelf in the ordinary manner, the air of the room containing from 75 to 80 per ct. of all the moisture it could hold at G0° F. The other cheese was placed under a bell-jar and kept in an atmosphere completely saturated with moisture. The results secured by this treatment are presented in the following table. The amount of moisture in the fresh cheese was not determined and we start, therefore, with the moisture in the cheese at two weeks. Table YII. — Loss of Moisture in Cheese Kept in Air Completely and Partially Saturated with Moisture. In air partially saturated. In air coinpl etely saturated with moisture. Age of Cheese. Moisture In cheese. Water lost by 100 lbs. of cheese. Moisture In cheese. Water gained by 100 lbs. of cheese. Per ct. Lhs. Pei- ct. Lhs. 2 weeks .... 1 mouth .... 35.99 35.23 35.93 35.87 0.76 2 moutlis . . . 3i.S6 1.13 36.01 0.08 6 months . . . 31.87 4.12 37.04 0.11 12 months . . . 2i;.30 9.G9 37.63 1.70 15 months . . . 21.85 11.14 37.85 1.92 Attention is called to the following points in connection with this table: (1) In case of the cheese kept in air partially saturated with moisture, there is a loss of moisture from the first, which at the end of 15 months has reached the total of 11.14 lbs. per hundred pounds of cheese. (2) In the cheese kept in a moisture-saturated atmovsphere, there was practically no loss of moisture in the cheese, but at the end of 2 months the moisture in the cheese had actually increased and continued to increase steadily, until, at the end of 15 months, there had been an actual gain of 1.92 lbs. of moisture per 100 lbs. of cheese. (3) The two cheeses containing the same amount of moisture at the beginning were found to differ, at the end of 15 months, 13 per ct. in moisture, as the result of being kept in air contain- ing different degrees of niui^lure. New Yokk Agricdltukal Expeeiment Station. 209 IV. SOME PRACTICAL APPLICATIONS. We have been considering those conditions that are most prominent in influencing the loss of moisture in cheese and have called attention to the results secured by us. We come now to consider these results in their practical application to the inter- ests of the factory owner, his patrons and the consumers of cheese. In this connection we will discuss the following topics: (1) Value of water in cheese to dairymen. (2) Moisture in cheese in relation to commercial quality. (3) What percentage of moisture should cheese have? (4) Value of water in cheese to consumer. (5) Variation of loss of moisture with different kinds of cheese. (G) Loss of moisture and loss of fat. VALUE OF WATER IN CHEESE TO DAIRYMEN. To the cheese-maker and producer of milk, water in cheese is money, when put there in the right way and in proper pro- portions. It is essential, in the process of manufacture, to incor- porate water in cheese in quantities best suited to the require- ment of the market for which the cheese is intended, and then it is equally essential that the water be kept there with the least jiossible loss. From the dairyman's standpoint, it is desirable to sell as much water in cheese as will suit the consumer. In preventing excessive loss of moisture there is more water to sell at cheese prices. From inquiries made among cheese-makers we find quite a variation in respect to the loss of moisture experienced by them in curing cheese. One of the most complete records, covering an entire season, furnished by a cheese-maker and factory owner who has better than average conditions for curing rooms, makes the average loss of weight during thirty days amount to about five pounds per hundred pounds of cheese. Others report an average loss for the first thirty days as high as ten pounds per hundred pounds of cheese. The average loss lies somewhere between these two extremes and would probably not be far from seven pounds per hundred pounds of cheese. U 210 RliPOKT OF THE CllKMICAL DEPARTMENT OF TUE An examination of Table II shows that the loss of moisture can be reduced to four pounds per hundred pounds of cheese. Using this figure as a basis for calculation, we find that, for every hundred pounds of cheese, from one to six i)ounds, with an average of three pounds of water could be saved to sell at cheese prices. This would mean an increase of 8 to 48 cents, with an average of 30 cents, received for every hundred pounds of cheese. This would mean an average saving of three hundred dollars a season for a factory with a total season's outj)ut of one hundred thousand pounds of cheese. One cheese-maker reports that he calculated one season's loss from shrinkage and found it over six hundred dollars. While such losses may not be regarded as large in comparison with the total receipts, they constitute a noticeable percentage when viewed as a decrease of profits, and are well worth saving. MOISTURE IN CHEESE IN RELATION TO COMMERCIAL QUALITY. We have just called attention to increased receipts coming from cheese, as a result of preventing excessive loss of moisture. Such saving of moisture not only increases the amount of cheese to be sold but also increases the value of the cheese from the standpoint of commercial quality. In Bulletin No. 184, of this Station, Mr. Geo. A. Smith, Dairy Expert, has presented the results of work showing the influence of temperature upon the commercial quality of cheese. No attempt is there made to analyze the results and point out the immediate causes afifecting quality, and attention is, therefore, called to the subject here. The relations existing between moisture and flavor are known only in a very general way. But we know something of the general relation between moisture and texture. Excessive moisture produces undesirable softness, from a commercial standpoint, and at ordinary temperatures favors the formation of holes, a serious fault in the texture of Cheddar cheese. On the other hand, deficient moisture favors the production of a crumbly, dry, mealy texture, which is an undesirable condition. Texture of cheese. (Peifoct texture is 2D.) Ml l3ture lost by 100 lbs. of cheese. Lbs. 24.6 8.5 24.4 9.0 23.6 9.2 22.0 10.2 21.4 10.7 20.6 13.1 New Yokk Agkicultdbal ExpEKiMEiqT Station. 211 High temperatures cause excessive loss of moisture and result in the production of crumbly texture. This condition injures the commercial quality of cheese and results in lower prices for such cheese. The following figures represent averages taken from data given on page 202, Bulletin 184, showing the general relation between texture and loss of moisture. Table VIII. — Effect of Temperature op Curing on Texture and Mois- ture OF Ciieese. Temi'Euature of Curing-Room. 55 degrees F. 60 degrees F. 65 degrees F. 70 degrees F. 75 degrees F. 80 degrees F, WHAT PERCElNTAGE OF MOISTURE SHOULD CHEE'SB HAVE? Much of the cheese made in New York State contains, in the fresh state, from 36 to 37.5 per ct. of water. The home-trade cheese, much of which is made in the fall, contains 38 to 40 per ct. of water. For the average consumer, it is safe to say, the amount of moisture in cheese should be not less than between 33 and 35 per ct. at the time of consumption. Taking everything into consideration, it is reasonable to expect better results in reference to quality by holding a moderate amount of moisture in the green cheese and so curing as to lose only a small amount of water, than by holding an excessive amount of moisture in the green cheese and so curing as to lose a larger amount of moisture. Some cheese-makers expect that they must lose ten pounds of weight per hundred pounds of cheese in curing, and they attempt to meet this loss by retain- ing 40 per ct. or more of moisture in the cheese. Such a prac- tice can not lead to good results from any point of view. A fact that should not be lost sight of in this connection is this: Cheese cured at such low temperatures as are favorable to diminishing the loss of moisture can carry larger amounts of moisture from the start without impairing the quality. 212 Report of the Chemical Depaktment of the value of water in cheese to consumers. In the first place, cheese that has not lost too much of its moisture is more pleasing to the taste of the average consumer. In the next place, the more completely a cheese dries out, the harder and thicker is the rind and the greater the loss to the consumer. Most people have become accustomed to such a waste, but much of it is unnecessary. In a carefully cured cheese, the rind is comparatively moist and only a very thin I)ortion need be lost, and even this can be used in cooking. VARIATION OF LOSS OF MOISTURE WITH DIFFERENT KINDS OP CHEESE. It has been pointed out that cheeses of small size lose more moisture per hundred pounds than do cheeses of larger size. In making small cheeses like " Young Americas " the proportion of loss is much greater, and hence the demand is still more imperative that these shall be cured under conditions where the loss of moisture shall be greatly reduced. This applies also to such sizes as " Flats " and " Twins." It is not surprising that the manufacture of small cheeses of the Cheddar type has been discouraged. Even at the higher prices that they bring, the extra loss of moisture and additional cost of manufacture are not satisfactorily covered. In the manufacture of small fancy kinds of soft cheese, these statements do not apply, because an essential part of the equipment consists of curing-cellars of fairly low temperature and high moisture content. LOSS OF MOISTURE AND LOSS OF FAT. High temperatures, which favor increased loss of moisture, also favor loss of fat by exudation from the surface of the cheese. When cheese is kept at a constant temperature even of 70° F,, there is evidence of some, though small, loss. At 75° F. the loss becomes considerable and increasingly large with increase of temperature above 75° F. New York Agricultural Experiment Station. 213 V. PKEVENTION OF LOSS OF MOISTURE IN CURING CHEESE. From the data previously presented, it has been seen that loss of weight in cheese curing is due to lack of control of tempera- ture and moisture in the curing-room. Three methods or sys- tems have been proposed for the purpose of controlling these conditions or obviating the need of controlling them: (1) Immediate sale and removal of green cheese. (2) Central curing-rooms for the use of several factories. (3) Special curing-room in each factory. IMMEDIATE SALE AND REMOVAL OF GRETON CHEESE. It was formerly a common custom to keep cheese at the factory for thirty days or more before selling it. For some time there has been a tendency to dispose of cheese at more frequent intervals, sales and shipments being made, in some cases, of cheese a week old. There appears to be an increasing desire to place cheese in the hands of buyers just as soon as they were willing to take it. Many buyers who have cold-storage facilities prefer to remove the cheese from the factory before it has had a chance to deteriorate under the adverse conditions of curing commonly present in factory curing-rooms. The sys- tem of removing cheese by buyers from the factory when less than a week old has the advantage for the cheese-maker of relieving him from all responsibility in re.lation to the curing process. There is, however, under such a plan the disadvantage of turning over to the bu3^er all the advantage that comes from increase of value as a result of good curing. With proper curing facilities, the cheese could be retained by the factory and held until it had increased very materially in value as a result of curing under good conditions. When cheese is sold green, or nearly so, the opportunity for increased profits, due to proper curing, is wholly lost. CENTRAL CURING-ROOMS. Four or five years ago Drs. Babcock and Russell made the sug- gestion that buildings, centrally located with reference to 214 KtPORT OF THE ChEMICAL DEPARTMENT OF THE several cheese factories, be erected especially for curing pur- poses and designed to take care of the product of the several factories. Such a system has several advantages: (1) Enough money could be easily secured to build and equip a central curing-house that would be complete in its details and thor- oughly efficient for controlling temperature and moisture. In fact, ideal conditions could be assured. No single factory could afford to provide itself an equally effective curing-room, or would be likely to do so. The cost for one central cheese-curing building, distributed among several factories, would be no more than would the cost of providing an inefficient curing-room in each individual factory. (2) Cheese stored in a central curing- house could receive more skillful and efficient attention than it could in curing-rooms located in each factory. (3) The cheese could be examined more economically by buyers, being collected in large quantities in a central curing-house. The buyer would be saved the necessity of visiting each factory separately. (4) The maximum saving could be effected in decreasing loss of moisture and in improving quality of cheese. Moreover, the cheese from any one factory or any number of factories would be more uniform in character when cured than under present conditions, or even with curing-rooms in individual factories. (5) Cheese kept under ideal conditions during the curing process can be held subject to market conditions without risk of injury in respect to quality. Under the conditions commonly prevail- ing, cheese has to be sold to avoid the injury in quality that might result from longer holding at the factory. This is especially applicable in hot weather, a time when prices are likely to be lowest. Cheese kept in proper curing-rooms can be held for higher prices and will constantly improve in quality for quite a long period of time. SPECIAL CURING-ROOM IN BACH FACTORY. When it is impossible to cooperate with other factories in the construction and use of a central cheese-curing building, then it is desirable that one shall make a cheese-curing room in the New Yoke Agricultdkal Experiment Station. 215 factory, even though the results secured may not be perfect. Some attempt to control temperature and moisture in curing- cheese will give better results than are possible in the absence of any system, a condition too general at present in the cheese factories of New York. The subject of a special curing-room in each cheese factory has been very fully discussed in Bulletin No. 70 of the Wisr-ousin Agricultural Experiment Station, and several factories in that State have made such curing-rooms. The system has also been studied and applied in Canada by Prof. James Robertson, Com- missioner of Agriculture and Dairying for the Dominion. The following statements are, for the most part, condensed from Prof. F. H. King's Wisconsin Bulletin No. 70. The cuts are from the same source. Curing-rooms may be constructed above ground or under ground, and may be of wood or masonry or a combination. Con- sidering moderate cost, convenience, and efficiency, a curing- room built of wood entirely above ground is the most practical for the average factory. " (1) Location. — A curing-room above ground should be placed on the north side of a building in order to be protected as much as possible from the direct rays of the sun. It is advantageous also if the room can be shut off on the other three sides by hall- ways, stairways, other rooms or building screens. (2) W indoles in a curing-room should be as few and as small as consistent with the amount of light necessary. They should be mado double, as nearly air-tight as possible, and preferably in one section, fitted closely and permanently in place. If neces- sary to exclude direct sunshine, blinds or awnings should be placed outside. \ ..--.- (3) The door of a curing-room should be built to resemble that of a refrigerator. , ^ - (4) Walls should be built like those of cold storage and ice- houses. The studding outside should be covered with matched sheathing and drop siding, with a layer of three-ply acid and water-proof paper between. The paper recommended by Prof. 216 Kepoet of the Chemical Deparimknt of the King is manufactured by the Standard Paint Co., New York and Chicago. On the inside a layer of matched sheathing is nailed to the studding, then strips of inch furring two inches wide, to which are nailed two thicknesses of matched sheathing with paper between. The outer air space between the studding is filled with sawdust or similar material, and the spaces left by the furring are closed air-tight at the ceiling and floor. (See Plate VII.) (5) Gelling and floor should also consist of two thicknesses of matched lumber with paper between, and joints made at corners should be very tight. In constructing curing-rooms two things should be kept in mind: First, that the walls should be as nearly air-tight as possi- ble in order to keep out the warmer air outside, and, second, that the walls should be poor conductors of heat. It is advan- tageous to cover the inside walls with two coats of shellac. (6) Yentilating flue in ceiling. — It is desirable to provide a tight ventilating flue in the ceiling of the curing-room, extending above the roof. Its diameter may be six or eight inches. It should be provided with a damper. (See Plate VIII, H, I.) (7) Methods of Gontrolling temperature and moisture in cheese curing-rooms placed above ground. — After constructing a proper curing-room, it is essential to provide arrangements for control- ling temperature and moisture. The construction of a curing- room is only a partial means toward this end. The following methods have been found effective in keeping the temperature during summer between 58° and 70° F. and at the same time modifying the moisture content of the air favorably: (a) Ven- tilation by air forced through horizontal sub-earth ducts or deep vertical sub-earth ducts and wells, (b) Ventilating over ice. (c) Evaporation of water. Plate VIII illustrates the construction of a horizontal sub- earth duct, which should be 12 feet or more below the surface of the ground and 100 feet or more in length. It is recom- mended that the sub-earth duct consist of three rows of 10-inch drain tile laid side by side at the bottom of the trench, or the Plate VII. — Showing the construction of wood curing-room: 1, 1, 1, sill; 2, '^, 'J., a two-by-ten spilled to ends of joist; 3, .3, 3, a two-by-four spilied down, after first layer of floor is laid, to toe-nail studs to; 4, 4, 4, a two-by-four spiked to upper ends of studding of first story. A, A, A, A, three-ply acid and water- proof paper. The drawing in the center shows space between studding filled with sawdust and another dead-air space to be used when the best ducts cannot be provided. (From Wis. Agr. Exp. Sta. Bui. 70.) ■CXX)000' coxooa C Plate VIII. — Section of cheese-curing room and liorizontal multiple sub-oarth duct. A, inlet to curing room; B, end of sub-earth duct in briclsed enti-ance to factory; C, cross-section of the multiple ducts; D, E, bricked entrance under funnel at outer end of sub-earth duct; F, funnel with mouth 36 inches across; G, vane to hold funnel to the wind; H, ventilating flue with damper. (Prom Wis. Agr. Exp. Sta. Bui. 70.) Plate IX. — Showing how funnel and vane may be mounted. A, funnel; B, shaft of funnel; C, C, C, 1-inch gas pipe; D, D, 1^-inch gas pipe; E, cap for support of 1-inch gas pipe; F, G, H and M M and N N are stays of band iron bolted together and to the sides of the shaft to support the axis of the funnel; J, weather collar to turn rain out ot shaft; K, L, band-iron to stiffen vane and attach it to funnel. (From Wis. Agr. Exp. Sta. Bui. 70.) Plate X. — Showing vertical sub-earth duct. A, brick chamber 25 to 30 feet below surface and 40 inches inside diameter; B, tile or conductor pipe of galvanized iron; C, main shaft of funnel; D, brick chamber at upper end of duct. The circle and section represent a cast-iron plate to cover brick chamber A, and can be had of King & Walker, Madison, Wis. (From Wis. Agr. Exp. Sta. Bui. 70.) Plate XI. — Showing vertical section of factory and sub-earth duct in well. A, A, funnel taking air into well; B, B, duct leading air from wall to curing room, C; D, ventilator. (Prom Wis. Agr. Exp. Sta. Bui. 70.) > ^KJa-eAftTH over Plate XII. — Showing metbocl of cooling air with cold water. A, curing room; B, duct leading into curing room; C, E, galvanized iron drums, air and water tight; F, thirteen or more 5-inch flues of galvanized iron, 10 ft. long, soldered water tight to drums to cool air; D, main air duct from funnel; G, water pipes from pump; H, overflow pipe; 1, damper in main shaft; J, 4-inch pipe leading from blower to use when there is no wind; K, smokestack of boiler; L, venti- lator from curing room to smokestack; N, boiler. (From Wis. Agr. li}xp. Sta. Bui. 70.) New York Agkicdltukal Experiment Station. 217 trench may be dug narrower and one or two feet deeper and the tile lolaced one above the other. The shaft for carrying the funnel must be made tight; it may be 12 inches square, if made of plank, or 12 inches in ^diameter, if made of galvanized iron. The height should be sufficient to enable the funnel to catch the wind readily. The construction and mounting of the funnel are illustrated in Plate IX. The extreme diameter of the funnel should be about 3G inches. The inlet from the sub-earth duct into the curing-room must be provided with some arrangement of valves that will permit the air to be shut off wholly or partly. Too rapid entrance of air in warm weather will not permit enough cooling during pas- sage through the duct. In case of dry winds, too rapid entrance would reduce the moisture too much. In Plate X there is illustrated a deep vertical sub-earth duct. Such a duct has the advantage of requiring less piping and also less wind will suffice to produce a current of air. The vertical duct should have a depth of not less than twenty-five or thirty feet, provided water is far enough from the surface. Thirteen lines of 6-inch drain or 5-inch galvanized iron conductor pipe may be used and placed as in the cut. The duct should be located near the north end of the curing-room or directly beneath it. A hanging platform can be used in placing the pipes or tubes in position and the earth packed carefully around the pipes. An excavation of proper size, made as for an ordinary well, will answer the purpose. After the duct has been placed in position the earth that has been removed can be used for filling around the duct. In Plate XI there is represented a duct connected with a well of water. In the particular instance illustrated, the well is 64 feet deep; the intake pipe is 10 inches in diameter, rising just barely above the roof of the factory, entering the well, as shown at A, two feet below the surface of the ground and then descend- ing inside the well a distance of 8 feet. Another 10-inch gal- vanized iron pipe starts 40 feet below the surface of the ground and rises to within 5 feet of it, when it turns and passes hori- 218 Report of tue Curmical Dj-partmi£nt of the zontally until it comos under the curing-room which it enters directly, as shown at B B I> C. The top of the well is tightly closed. ' In Plate XII the cut illustrates the cooling of air in a curing- room by forcing the air through cold water. When the ground water is within 12 or 15 feet of the surface, then a cistern 5 or 6 feet in diameter, shaped like a well, may be built, plastering with cement as in the case of ordinary cisterns. In this cistern can be placed an air duct made of galvanized iron as given in Plate XII. The duct should be water-tight. By connecting the cistern with the well, fresh water may be added from time to time as may be found necessary to keep water sufficiently cool to be effective. In Canada, considerable work has been done in using ice in curing rooms to control temperature. Where ice can be obtained conveniently and cheaply, this method may be advant- ageously utilized. One or more ice boxes are placed in the curing-room, so built that air can circulate about the ice and into the curing-room. Also compartments, filled with ice, may be made adjoining the curing-room on the side or above, pro- vided with openings into the curing-room which will allow a flow of air over the ice and into the curing-room. Where special means are needed to secure moisture, this can be effectively done by means of yard-wide strips of any cloth material that has good capillary power. The pieces of cloth are hung about the room and kept more or less saturated with water. Experience will tell how much evaporating surface is needed to provide the degree of moisture needed. New Yokk Agricdltdral Experiment Station. 219 Table Showing percentage of Saturation op Moisture in Am at Vakious Tempkhatures According to Hygrometer. Difference between the dry and wet thermomet ers in deg rees Fahrenheit. Dry ther- 1 1 mometer. Degrees Fahrenheit. 0.5 1.0 1.5 3.0 3.5 3.0 3.5 4.0 4.5|5.0|5.5|6.0|6.5|7.0|7.5|8.0|8.5|9.0 9.5 10.0 10.5]ll.0|ll.5|l3.0 1 ' Percentages of moisture In air. 40 96 93 83 84 80 76 73 68 64 60 56 53 49 45 41 33 34 30 26 22 19 16 13 8 41 96 93 88 84 80 76 73 69 65 61 57 54 50 46 43 39 36 33 29 34 21 18 11 10 4-4 96 93 88 81 81 77 73 69 65 63 58 55 51 48 44 40 37 34 30 27 23 20 16 13 43 96 9i 88 85 81 77 74 70 66 63 59 56 52 49 46 42 38 35 33 29 25 23 19 15 44 96 93 88 85 81 78 74 70 67 63 60 57 53 50 47 43 40 37 33 30 27 24 21 17 45 96 93 89 85 82 78 75 71 67 64 61 58 51 51 48 44 41 38 35 32 29 25 22 19 46 96 93 89 85 82 79 75 72 68 65 61 58 55 53 49 46 42 39 36 33 30 27 23 21 47 96 93 89 86 83 79 76 72 69 66 63 59 36 53 50 47 44 40 38 34 31 28 25 83 48 96 93 89 86 83 79 76 73 69 66 63 60 56 53 51 48 45 42 39 36 83 30 37 24 49 97 9i 90 86 83 80 76 73 70 67 63 6u 57 54 53 49 46 43 40 37 34 31 29 26 50 97 93 90 87 83 80 77 74 70 67 61 61 58 55 53 50 47 44 41 38 36 33 .30 27 51 97 93 90 87 84 81 77 74 71 63 65 62 59 56 53 50 48 45 42 39 37 34 31 28 53 97 94 90 87 84 81 78 75 72 69 66 63 60 57 54 51 48 46 43 40 3S 35 33 30 53 97 94 91 87 84 81 78 75 72 69 66 63 61 58 55 53 49 47 44 43 39 36 34 31 54 97 94 91 88 85 83 79 76 73 70 67 64 61 59 56 53 50 48 45 43 40 38 35 33 55 97 94 91 88 85 82 79 76 73 70 68 65 63 59 57 54 51 49 46 43 41 39 86 34 56 97 94 91 88 85 82 80 77 74 71 68 65 63 60 57 55 53 50 47 44 42 40 37 35 57 97 94 91 88 86 83 80 77 74 71 69 66 64 61 58 55 53 50 48 45 43 40 38 36 58 97 94 91 89 86 83 80 78 75 72 69 67 64 61 59 56 53 51 49 46 44 42 39 37 59 97 94 93 89 86 83 81 78 75 72 70 67 65 62 60 57 54 52 49 47 45 43 40 38 60 97 94 92 86 86 84 81 78 75 73 70 68 65 63 60 58 55 53 50 48 46 44 41 39 61 97 94 93 89 87 8C> 81 78 76 73 71 68 66 63 61 58 56 54 51 49 47 41 43 40 63 97 95 93 89 87 ■34 82 79 76 74 71 69 66 64 61 59 57 54 52 50 47 45 43 41 63 97 95 92 89 87 84 83 79 77 74 73 69 67 64 63 60 57 55 53 51 48 46 44 43 64 97 95 93 90 87 85 83 79 77 74 72 70 67 65 63 60 58 56 53 51 49 47 45 43 65 97 95 92 90 87 85 83 80 77 75 72 70 63 65 63 61 59 56 54 52 50 48 46 44 66 97 95 93 90 87 85 82 80 78 75 73 71 68 66 63 61 59 57 55 53 51 49 47 45 67 98 95 93 90 88 85 83 80 78 76 73 71 69 66 64 63 60 58 55 53 51 49 47 45 68 98 95 93 90 88 86 83 81 78 76 74 71 69 67 65 63 60 58 56 54 53 50 48 46 69 9S 95 93 90 88 86 83 81 78 76 74 73 70 67 65 63 61 59 57 55 53 51 49 47 70 98 95 93 90 88 86 83 81 79 77 74 73 70 68 66 64 63 60 57 55 53 52 50 48 71 93 95 93 91 88 86 84 81 79 77 75 73 70 68 66 64 63 60 58 56 54 52 50 48 73 98 95 93 91 88 86 84 83 79 77 75 73 71 69 67 65 63 61 59 57 55 53 51 49 73 98 95 93 91 8S 86 84 8i 80 78 75 73 71 69 67 65 63 61 59 57 55 53 52 50 74 98 95 93 91 88 86 84 83 80 78 76 74 72 70 68 66 64 62 60 58 56 54 5? 50 75 98 95 93 91 89 87 84 82 80 78 76 74 73 70 68 66 64 62 60 58 56 55 53 51 76 98 95 93 91 8't 87 85 82 80 78 76 74 73 70 68 66 64 63 61 59 57 55 53 52 77 98 95 93 91 89 87 85 83 80 78 76 74 73 71 69 67 65 63 61 59 57 56 54 52 78 98 96 93 91 89 87 85 83 81 79 77 75 73 71 69 67 65 63 62 60 58 56 51 53 79 98 96 94 91 89 87 85 83 81 79 7 ( 75 73 71 70 68 66 64 63 60 58 57 55 53 80 98 96 94 92 89 87 85 83 81 79 77 75 73 72 7C 68 66 64 63 61 59 57 55 54 REPORT OTiT Crop Production. W. n. Jordan, Director. G. W. Churchill, Agriculturist F. A. SiRRiNE, In charge of Second Judicial Depa/rtment Experiments, Table of Contents. I. Influence of manure upon sugar beets, II. Commercial fertilizers for onions. INFLUENCE OF MANUKE UPON SUGAR BEETS* W. H. JORDAN AND G. W. CHURCHILL,. SUMMARY. (1) Tlipse experiments were undertaken to test the accuracy of the statement that sugar beets are of an inferior quality when grown on land to which stable manure is applied in the spring. (2) The experiments have been conducted during four consecu- tive years, mostly on the Station farm. Comparisons have been made of the quality of beets not manured, those grown with commercial fertilizer, mostly 1,000 lbs per acre, and those grown on land receiving in the spring, before planting the beets, from 40,000 lbs. to 80,000 lbs. stable manure per acre. Beets from at least six varieties of seed were grown during the four years. (.3) The results are almost unanimous in one direction. The beets have been of high quality with all three methods of treat- ment, averaging somewhat better with the farm manure than with no manure or with commercial fertilizers. INTRODUCTION. The value of a given lot of sugar beets for sugar-maldng pur- poses depends chiefly upon two factors, viz.: the percentage of saccharose in the beets and the percentage and character of the solub\e compounds accomijanying the sugar. In general a beet is valuable in proportion to its content of crystallizable sugar, but if this is attended by too large an amount of certain soluble non-sugars, the effect is to prevent the crystallization of some of the saccharose which under better conditions would be secured in the manufactured product. -A reprint of Bulletiu No. 205. 223 224 Repout on Crop Pkoduction of the The relation of crystallizable sugar to the total solids in solution in the juice of beets is known as the coefficient of purity. It is generally taught that the percentage of sugar in beets and also the coefficient of purity are materially influenced by the kind and amount of fertilizing material which is used in growing the crop. Growers are especially cautioned against planting beets on laud freshly fertilized with stable manure and against heavy nitrogenous manuring with chemicals. It is stated that past experience has shown that beets raised where a generous application of farm manure is made in the spriug are inferior for manufacturing purposes, and it is suggested that while a large application of nitrate of soda may not cause a diminution of the sugar content it may so lower the coefficient of purity as to lessen materially the proportion of available sugar. In 1898 experiments conducted by this Station in growing beets with the use of farm manures and with commercial fertil- izers in varying quantities gave results in apparent conflict with prevailing views.^ These results have led to the continuation, during the past three years, of experiments of a similar char- acter, the outcome of which is presented in this bulletin. THE EXPERIMENTS. GENERAL PLAN AND CONDITIONS. The experiments were planned with reference to comparing the composition of beets grown with commercial fertilizers and those grown with stable manure applied in the spring just before planting. Check plats have also been used in order to ascertain how the beets would grow without the application of any manure whatever. All the experiments have been conducted on the Station farm, excepting in the year 1898, when one set was carried out on the farm of F. E. Dawley, Fayetteville. Excepting the year 1899, when texture conditions were ' N. Y. Agrl. Expt. Sta. Bui. No. 155. New Yokk Agkicdltural Experiment Station. 225 unfavorable on some of the plats, owing to a lack of rain at the time of germination of the seed, the beets were successfully grown, both as to quantity and as to the type and healthfulness of the plants. The temperature and rainfall as shown by records kept at the Station are of interest in this connection. Table I. — Temperatuee at Geneva. 1893. 1899. Month. Max. MIn. Av. Max. Mln. At. Deg. Deg. Deg. Deg. Deg. Deg. April 52.7 33.6 43.2 56.8 36.3 46.6 May 65.5 47.4 57 68.7 46.4 57.6 June 7S.2 57.2 67.7 82.3 56.6 69.5 July 86.5 61.8 74.2 82.9 59.5 71.2 August 80.6 61.4 71 83.9 59.2 71.6 September 77.1 54.7 65.9 71.1 50 60.8 October 61.4 42.8 52.1 63.8 42.9 53.4 1900. 1901. Slontb. Max. Mln. Av. Max. Min. Av. Deg. Deg. Deg. Deg. Deg. Deg. April 52.7 34.2 43.5 56 37.1 46.5 May 68.5 44.9 56.7 67 46.8 56.9 June 81.4 55.4 68.4 80.5 57.4 68.9 July 83.2 62 72.6 88.2 65.1 76.6 August! 85 63.1 74.1 80.9 61.1 71 September 78.9 53.3 66.1 75 53 64 October 68.4 47.4 57.9 62.1 47 51.4 Table II. — Precipitation at Geneva. Month. 18'J>!. 1899. 19C0. 1901. Itk In. In. In. March 1.54 1.22 .02 2.19 April 2.03 1.12 .95 4.43 May 1.90 1.69 ].71 3.80 June 2.39 1.71 1.45 2.07 July 1.32 4.15 6. .-.3 3.97 August 3.60 1.05 1.75 5.62 September 1.86 2.23 .91 2.46 October 3.83 2.69 3.65 1..35 Totals eight months 18.47 15.86 16.97 25.89 Total for May, June, July and August 9.21 8.60 11.44 15. 4Q 15 226 Repokt on Crop Fkoddction of the Attention is called to the great variations in tlie rainfall during the four months most important in the life of the plants. In 1899 there was a deficiency in available water but not in 1901. The soil of the Station farm is a strong clay loam, well adapted to general farming and capable of producing large crops when well handled. The plats used in the beet experiments could not be regarded as especially deficient in fertility, and would not respond to the application of manures as would poorer land. The methods used in the cultivation of the crop were those approved by past practice and no detailed statement concerning them is necessary in this connection. THE MANURE. The stable manure used was a mixture of that coming from the cow and horse stables of the Station, sufficiently composted and mixed to render it fine and of uniform composition. The manure actually applied to the beet soil was analyzed only one year, 1901. Manure from the same general source and used in another experiment was analyzed in 1897, 1898 and 1899. From the data thus secured it is possible to know approximately the amounts of plant food supplied to the crops from this source. Table III. — Composition of Manure Made in the Station Stables. Phosphoric Year. Water. Nitrogen. acid. Potash. Per ct. Per ct. Per ct. Per ct. ' 1897 73.9 .389 .3G0 .342 Used on corn. ISOS 76.1 .303 .241 .593 Used on corn. 1899 74.3 .529 .576 .851 Used on wheat. 1901 78.3 .445 .382 .738 Used on beets. In 1901 the manure was applied at the rate of 80,000 lbs. per acre, in all other years at the rate of 40,000 lbs. The commercial fertilizer mixture was essentially the same tln-oughout. Its ingredients and approximate composition are given in Table IV. New York Agricultural Experiment Station. 22- Table IV. — Fertilizer Mixturei Used on Sugar Beets. Coutaining approximately — Material. Acid phosphate Sulphate of potash. Dried blood Nitrate of soda In one ton. Lbs. 900 300 400 400 Nitro- gen. Phos- phoric acid. Pot- ash. Lbs. Lbs. 126 . Lbs. 150 40 8 , 60 Total 2000 100 134 150 In 1000 lbs. of the mixture • ••••' ••••••' 50 67 75 In percentages , >•••• •••••• 5 6.7 7.5 EXPEROIEMTS OF 1898.^ These were carried on both on the Station farm and on the farm of F. E. Dawley, Fayetteville, N. Y. The plats were 1-12 acre at the Station and at Mr. Dawley's 1-16 acre. In the tables which follow the yields are given for one acre. Table V. — Commercial Fertilizers on Sugar Beets, 1898. Amount of fertilizer used. Yield of trimmed and ■washed beets per acre. Sugar In beets. • Coefficient of purity of juice. Average weight of beets analyzed. . Place of experiment. Lbs. Lbs. Per et. Ozs. 20,425 15.2 85.2 161/2 Station. 500 21,375 15.6 85.7 161/2 Station. 500 27,140 14.5 86.0 15 Station. 1000 26,928 14.4 83.6 20 Station. 1000 26,250 14,7 85.4 16 Station. 1500 23,822 14.3 84.5 17 Station. 1500 27,920 14.9 85.8 1514 Station. 2O0O 22,073 15.0 85.6 I614 Station. 2000 27,875 17.0 87.1 131/2 Station. 18,585 15.4 81.6 121/2 Fayetteville. 17,740 17.2 85.0 9 Fayetteville, 50O 23,373 15.2 77.1 141/^ Fayetteville, 500 24,075 14.3 79.8 161^ Fayetteville, 1000 24,220 14.5 78.3 131/2 Fayetteville. 1000 24.220 15.9 81.3 10 Fayetteville, 150O 26,890 15.3 80.1 I81/2 Fayetteville. 1500 26,330 15.2 79.7 131/2 Fayetteville, 'Reported in Bulletin No. 155. ^The percentages of sugar in the beetfe as given in the bulletins are the results of actual determinations on the basis of a weighed quantity of beet and are not calculated from the sugar in the .juice. 'J'/S Keport on Crop Production of the Table VI. — Summary Showing Effect of Fertilizees on Yield of Sugab Beets in 1S98. Number of experl- meuta. Yield per acre. used per acre. Lowest. Highest. Average. Increased average. Lbs. Lbs. Lbs. Lbs. Lbs. 500 3 4 17,740 21,375 20, 425 27.140 19.294 23,990 4,696 IWO 4 24,220 26,928 25.405 6,111 1500 4 2.3.822 27.920 26. 240 6.946 2000 2 22,073 27,875 24,974 5,680 Table VII. — Summary Showing Effect of Fertilizers Upon Percent- age OF Sugar in Beets in 1898. Amount of sugar In beets. Fertilizer used per acre. Number of experiments. Lbs. 3 500 4 1000 4 1500 4 2000 2 Lowest. Per ct. 15.2 14.3 14.4 14.3 15.0 Highest. Per ct. 17.2 15.6 15.9 15.3 17.0 Average. Per ct. 15.9 14.9 14.9 14.9 16.0 Table VIII. — Summary Showing Effect of Fertilizers Upon Coeffi- cient OF Purity of Sugar Beets in 1898. Coefficient of purity. Fertilizer used Number of per acre. experiments. Lbs. 3 500 4 1000 4 1500 4 2000 2 Lowest. 81.6 77.1 78.3 79.7 85.8 Highest. 85.2 86.0 85.4 85.8 87.1 Average. 83, ,9 82. ,1 82. ,1 82. .5 86.3 xSEW YoKK AgEICDLTDRAL EXPERIMENT STATION. 229 Table IX. — Results ob- Applying Stable Manuee in Geowing SuGxYr Beets, 1S98. Amount of stable manure applied per acre. 20 tons. 20 tons. 20 tons. 20 tons. 20 tons. 20 tons. 20 tons. 20 tons. 20 tons. 20 tons. 20 tons. 20 tons. 20 tons. 20 tons. 20 tons. Yield of trimmed and washed beets. Lbp. 20.425 2.5. 3G0 2i;>.::;40 28,690 27,100 28,354 28,630 29,656 29,533 31,944 16,050 18,022 23,514 25,025 24,780 25,485 27,034 26,750 Sugar in beets. Per et. 15.2 18.5 17.2 16.4 15.7 16.2 17.2 17. S 17.9 17.7 14.4 15.5 18.2 15.7 13.1 14.3 15.2 17.9 Coefficient of purity of juice. 85.2 85.2 86.2 86.7 85.2 85.7 87.4 86.4 87.7 87.8 77.8 82.0 81.3 78.8 78.0 79.0 80.3 87.5 Average weight of beets analyzed. Ozs. 161/2 12 13 15 11 121/2 13 11 14 12 131/2 16 * 81/2 111^ 141/^ 111/2 151/2 121/2 Distance between beets in row. In. 8 6 8 10 6 8 10 6 8 10 8 8 8 8 8 8 8 8 Place of experiment. Station. Station. Station. Station. Station. Station. Station. Station. Station. Station. Fayetteville. Fayetteville. Fayetteville. Fayetteville. Fayetteville. Fayetteville. Fayetteville. Fayetteville. Table X. — Summary Showing Effect of Stable Manure on Yield op Amount of stable manure used per acre. 20 tons. Sugar Beets, 1898. Number of experiments. -Lowest. Lbs. 3 15 16,050 23,514 Yield per acre. Highest. Lbs. 20,425 31,944 Average. Lbs. 18.730 27,450 Increased average. Lbs. 8,720 Table XI. — Summary Showing Effect of Stable Manure on Percent- age OF Sugar in Sugar Beets, 1898. Amount of sugar in beets. Number of , Amount of stable mmure used per acre. 20 tons. Number of experiments. Lowest. Highest. Average. Per ct. Per ct. Per ct. 3 14.4 15.5 15.1 15 13.1 18.5 10.6 Table XII. — Summary Showing Effect of Stable Manure Upon Co- efficient OF Purity of Sugar Beets. 1898. Coefficient of purity. Amount of stable manure used per acre. Lowest. Highest. Average. 77.8 85.2 82.6 20 tons 78.0 87.8 84.2 2ao Report on Crop Production of the EXPERIMENT OF lSf)9. Tliis was similar in plan to those of 1&08 and was conducted on the Station farm. As before stated, the crop was a failure on part of the plats, owing to a failure of the seed to germinate. Only on the check plat and on the plats to which farm manure was applied was there a uniform stand of plants. Because of lack of rain at critical times the crop was reduced below the normal. The seed was sown June 1 and the crop was harvested November 22. The plats were 1-36 acre in area. To avoid error, 20 beets from each plat were analyzed. Table XIII.- -SUGAK Beets Grown with Stable Mantjee, 1899. Plat No. Stable manure; per acre applied Yield trimmed beets per acre. Sugar in beets. Coefficient of purity. Average weigbt of beets. Distance apart of beets. Lbs. Per ct. Per ct. Oza. In. 11 15,840 14.8 84.2 10 • • 1? 20 tons 25.200 23.400 21,960 14.9 15.7 15.1 8:3.9 85.9 85.7 8 8.7 10.4 6 18 20 tons 8 14 20 tons 10 15 20 tons 22. 320 22.320 15.4 15.5 83.3 84. 6 8.7 9.5 6 16 20 tons 8 17 20 tons 21,020 16 86 11.2 10 18 20 tons 23.940 15.8 86.9 7.7 6 19 20 tons 21,800 16.3 88.1 10.3 8 20 20 tons 22,080 15.8 90.2 11.7 10 Table XIV. — SujiiiARY' Showing Effect of Stable Manure on the Yield and Composition of Sugar Beets, 1899. Uauuring per acre. stable manure, 20 tons Stable manure, 20 tons Stable manure, 20 tons General average for manure plats. Yield beets per acre. Sugar in beets. Coef. of purity. Distance apart of beets. Lbs. Per ct. Per ct. In. 15,840 14.8 84.2 • • 23,820 15.4 84.7 6 22,507 15.8 86.2 8 21,887 15.6 87.3 10 15.6 86.1 • • EXPERIMENT OF 1900. In the experiment of this year three varieties of beets were grown, the seed of which was supplied by the U. S. Department of Agriculture. Each variety was planted in triplicate plats with commercial fertilizers and the same with stable manure. New Yoek Agricultdeal Experiment Station. 231 The time of planting was June 4 and of harvesting Nov. 23. The dimensions of the plats were 165x6J feet. Twenty beets from oach plat were analyzed. The results appear in Tables XV and XVI. Table XV. — Results feom Manuring Sugae Beets, 1900. No. Plat. 1 2 3 4 5 6 7 8 9 10 11 12 Variety of beets. Plats with Commercial Fertilizer, 1000 Lbs. per Acre: Vilmorins Improved White. . White Queen of the North . . Austrian Special Kleinwan- zlebener Vihnorins Improved White.. White Queen of the North.. Austrian Special Kleinwan- zlebener Plats with Farm Manure, 40,000 Lbs. per Acre: Vilmorins Improved White. . White Queen of the North.. Austrian Special Kleinwan- zlebener Vilmorins Improved White. . White Queen of the North.. Austrian Special Kleinwan- zlebener Yield triiniued beets per acre. Lbs. Sugar iu beets. Per ct. Suijar iu juice. Per ct. ;7,171 14.2 15 Coef- ficient of purity. 42,568 15.9 S.3.6 39,029 IG.l 84. S 38,822 14.6 81. G 40,051 13.4 15.8 83 35,035 14.9 16. 6 84 84. 4 41. .395 16.1 83.7 38,077 17.6 85.2 40,. 550 15.7 84.3 43,908 15.5 16.8 86.2 38,784 15.4 17.5 83.9 44,851 14.7 IG.l 85 Average weight of beets analyzed. Ozs. 19 20 18 18 18 ITi^ 21 19 19 18 19 20 Table XVI. — Summary Showing Effect Composition of Sugar Beets OF Manure Upon the Yield and (BY VARIETI£S), 1900. Yield per Varieties and manures. aere. Lbs. Commercial Fertilizer, 1000 Lbs. per Acre: Vilmorins Improved White 41.310 White Queen of the North 37, 632 Austrian Special Kleinwanzlebener 37,996 Stable Manure, 40,000 Lbs. per Acre: Vilmorins Improved White 42,681 White Queen of the North 39, 430 Austrian Special Kleinwanzlebener 42,700 General Summary': Beets with commercial fertilizers 38,979 Beets with stable manure 41 , G04 Sugar lu beet. Sugar iu juice. Coeffi- cient of purity. Per ct. Per ct. 13.4 15.8 83.3 14.9 16.3 84.4 14.2 14.8 83 15.5 16.4 85 15.4 17.5 84.5 14.7 15.9 84.6 14.2 15.6 83.6 15.2 16.6 84.7 232 Report on Ckop Pkoductio:"! of the EXl'EIilMEXT OF 1001. The seed of the two varieties of beets came from the U. S. Department of Agriculture. Each variety was planted in dupli- cate on both commercial fertilizer and farm manure plats Table XVII.— Results of Manuring Sugar Beets, 1901. Coef. Yield ficients of of Av. -,, beets Sugar Sugar purity weinht *['*"• per ill iti of beets No. Variety and treatment. acre. beets, juice, juice, au'lyz'd Lbs. Per ct. Per ct. Ozs. Dep't No. 6359, Meyers Friedb- rickswerter: 1 80,000 lbs. stable manure per acre. 40,710 1.3.2 14.9 78.1 13.7 2 1000 lbs. commercial fertilizer 36,660 12.3 16.5 83.7 14.2 3 No manm-e 25,226 13.1 17.1 82. 9 13. & Dep't No. 5772, Dippes German: 4 80.000 lbs. stable manure per acre. 33,570 13.4 18.6 80 13.1 5 1000 lbs. commercial fertilizer 28,190 15.6 20.7 87.7 12 Table XVIII. — Summary Showing Effect of Manures upon Yield and Composition of Sugar Beets, 1901. Yield Sugar Sugar Coef- of beers in In flcifnr of per acre. beets. juice. purity. Z,bs. Per ct. Per ct. No manure 35,226 13.1 17.1 82.9 1000 lbs. commercial fertilizer 32,425 13.9 18.6 85.7 80,000 lbs. stable manure 37,140 13.3 16.7 79 In this experiment the commercial fertilizer used was similar to that applied in former years, both in kind and quantity, but the stable manure was increased from 40,000 to 80,000 lbs. per acre. This was an excessive application of an animal manure, twice as much as what most farmers would consider a liberal quantity. Fifty beets from each plat were analyzed. Tables XVII and XVIII show results. DISCUSSION OF KESULTS. These experiments, which have been carried through four years, have included the growing of beets from high grade seed from various sources, at least six different varieties (names) New York Agricultural Experiment Station. 233 boing present. Tlie main question at issue in this work lias been the effect of commercial fertilizers and stable manure upon the manufacturing value of the beets, with especial reference to the possibility of depressing the quality of beets by growing them on land to which stable manure has been freshly applied. A determination of the percentages of sugar and of the coefficients of purity has been the means of judging of the quality of the beets grown. No determination has been made of the character of the non-sugars present in the juice. If beets may be standardized as to quality by the proportion of sugar in them, together with the coefficient of purity, then the conclu- sions to be drawn from the data herewith presented are plainly indicated. Attention is directed to the figures of the preceding tables but more especially to the general summary in Table XTX. Table XIX. — General SuinfARY of Results Showixg the Influence of Manuke Upon the Quality op Sugar Beets, 1898-1901. No manure. Commercial fertilizer. Stable manure. Sugar Sugar Coef- Sugar Sugar Coef- Sugar Sugar Coef- ja in fleipntof iii in ficlent of in in flcient ot Tear grown. beets, juice, purity, beets, juice, purity. beets. juice, purity. Per ct. Per ct. Per ct. Per ct. Per ct. Per ct. 1898, station. 15.2 85.2 15 85.4 17.2 86.5 1898, Dawley 15.6 81.6 15 79.4 15.9 80.8 1899 14.8 16.2 84.2 15.6 17.9 86.1 1900 14.2 15.6 8.3.6 15.2 16.6 84.7 1901 13.1 17.1 82.9 13.9 18.6 85.7 13.3 16.7 79 The data here presented are strikingly opposed to what is regarded as the orthodox method of manuring sugar beet land. It so happens that, with the exception of the crop of 1001, not only does the stable manure fail to depress the quality of the beets, but the crops grown where it was applied in the spring show a higher percentage of sugar than where commercial ferti- lizer was used or where no manure was applied. In 1901 the percentage of sugar was but little lower but the coefficient of purity appeared to drop. In this case the stable manure was used in an excessive quantity. 234 Kepokt on Chop Pruduciion of the Table XX. — Influence of Manures on the Relation of the Roots and Tops of Sugar Beets, 1900, 1901. Treatment. Commercial fertilizer plats: Plat 1 (2, 1901) Plat 2 (5, 1901) Plat 3 Plat 4 Plat 5 Plat 6 Average Stable manure plats: Plat 7 (1, 1901) Plat 8 (4, 1901) Plat 9 Plat 10 Plat 11 Plat 12 Average No fertilizer (3, 1901) . . . Single beet, commercial fertilizers Single beet, stable manure Experiment 1900. Experiment 1901. Weight Weight Per ct. No. of beets beets weight beets wiih without of ■w'hed. tops, tops, Ounces. Ounces. roofs. Per ct. Weight Weight No. of beets of beets beets with without weighed, tops. tops. Ounces. Ou7ices. 40 20 20 20 20 20 40 40 40 20 20 20 761 409 3G9 307 3ol 3G5 837 778 763 354 318 435 572 309 284 267 272 2Q1 263 253 322 75.1 5 9 7 5 4 76 72 77 81 6.5 676 80.7 619 79.6 626 82 74.3 79.5 to 78.3 i)0 55 952 839 55 920 55 881 778 698 Per ct. weight of roots. Per ct. 81.7 a3.2 82. 4 765 82.0 735 83.4 19.3 15.3 • ■ • » 16.3 13.4 16.4 1.36 83 55 951 786 82.6 Some indication of the relative effect of the two kinds of manures may be gained from knowing the relation by weight of roots and tops. In 1900 and 1901 carefully selected beets from all the plats were weighed before and after trimming. The results of these weighings are given in Table XX. It does not appear from the above data that stable manure induced an excessive growth of leaves as compared with com- mercial manures. It is fair to inquire if, in the nature of things, there are good reasons for expecting results different from those detailed here. Why should plant food derived from stable manure cause growth unlike that sustained by chemical manures? While we cannot enter into all the secrets of the life of the plant, such a New York Agricultural Expekiment Station. 235 specific difference does not appear to be rational. It does not appear in llie experiments here recorded that the yield of beets was greatly larger with 40,000 to 80,000 lbs. of stable manure per acre than with 1000 lbs. of commercial fertilizer. It seems extremely probable that if a fresh application of stable manure ever causes the ranker vegetative growth, this result must be attributed either to the larger quantities of fertilizing ingredi- ents w^hich are usually supplied in comparison with chemical manures, or to the modifying influences upon the soil of its organic matter as affecting texture and water holding power. Granting that the excessive amounts of nitrogen and other ele- ments of plant food contained in the usual application of stable manure are sometimes the cause of too rank growth, then the use of less manure will modify this effect which is undesirable with some plants. In case the better texture and greater water holding power that the manure induces, and which are so desir- able conditions to secure for most soils, are the explanation of the great vigor of the plants, it would not seem wise to with- hold the manure, but to control the character of the growth by regulating the quantity of manure and number of plants on a given space and by other means. The evidence which this bulletin presents shows clearly that under the conditions involved, stable manure was freely applied to the beet land in the spring just before planting the seed without injury to the quality of the beets. If this proves to be true in general, no time limitations are to be placed on the use of such manure, and sugar beet produc- tion will not demand of the farmer an annual expenditure of cash for commercial fertilizers because they are a necessity in this line of farming. COMMERCIAL FERTILIZERS FOR ONIONS. * W. H. JORDAN AND F. A. SIRRINE. SUMMARY. (1) Experiments in the use of different quantities of a com- plete fertilizer in growing onions were conducted at Florida, Orange county, N. Y., for four years on the same field and for one year on a field of another farm. (2) The quantities of fertilizer used were 0, 500 lbs., 1000 lbs., 1500 lbs. and 2000 lbs. per acre. (3) On the Purdy field (four years), when only 500 lbs. of fer- tilizer was used, the manure cost of the increase of crop was 16.6 cts. per barrel; with 1000 lbs., 79.3 cts.j with 1500 lbs., 80.4 cts., and with 2000 lbs., 227.8 cts. (4) The profit from using the fertilizer came mostly from the first 500 lbs, api>lied, averaging |35.84 per acre. With onions at 11.25 per barrel the profit was slightly larger (about |3 per acre) with both the 1000 lbs. and 1500 lbs. of fertilizer per acre; but 2000 lbs. was used at a loss. (5) On the Mars field one experiment was conducted which showed no increase of yield from applying commercial fertilizer even in the larger quantities, (6) The results of these experiments show clearly that the crops were limited more by other conditions than by the extent of the plant-food supply. With the best conditions of season and water supply the smallest amount of fertilizer supported the maximum crop. *A reprint of Bulletin No. 206. New York Agricultural Experiment Station. 287 (7) Considering the varjing market price of onions from one year to another and the various vicissitudes to which the crop is subjected, the use of the larger quantities of fertilizer (above 500 lbs.) was attended by danger of financial loss. GENERAL CONDITIONS. Experiments and investigations were begun by the Station in the Second Judicial Department of New York in the year 1894. One of the conditions of practice prevalent in that portion of the State, especially with the market gardeners and potato and onion growers, was the excessive use of commercial fertilizers. The application of one ton or more per acre of a high grade, complete fertilizer was frequently observed. Reasoning from general facts, it did not seem clear that such a large expenditure for commercial plant-food was justified from the standpoint of profit. In order to determine the correctness of this view, field experiments with fertilizers on potatoes were begun on Long Island in 1895, which were continued until 1900, during the last four years of which time observations were made on four farms located at different points in potato growing dis- tricts. The general outcome of these experiments was to show that, so far as profit from the potato crops was concerned, the use of 1000 lbs. of fertilizer per acre was more profitable than the use of 500 lbs., 1500 lbs. or 2000 lbs. In 1898 similar observations were begun at Florida, Orange County, on the use of commercial fertilizers in growing onions. These have been continued each year since, the experiment of 1901 being regarded as concluding the series. THE EXPERIMENTS. PLAN. In these experiments, conducted for four years on one farm (Purdy field) and for one year on another (Mars field), approx- imately one acre of land was utilized in each locality. This acre 23S E.EPOKT ON CkOP PbODUCTION OF THB was divided into ten plats, wliich wore treated in accordance with the diagram shown below. On tlie field where the experi- ment was continued for four years, each plat received the treatment as indicated each year of the entire time, with the excejjtion noted under " Fertilizers used." Arrangements of Plats in Onion Fertilizer Experi- ments. No fertilizer. noo lbs. fertilizer per acre. FERTILIZEES USED. The fertilizer was applied annu- ally. For three years it was com- pounded in accordance with the formula for some time so popular with Long Island farmers, viz.: four per ct. nitrogen, eight per ct. phosphoric acid and ten per ct. potash. In 1901 the potash was changed to five per ct. Crimson clover was sown on the Purdy field in August of 1900, which grew to a height of from four to six inches and was plowed under the very last of November. W tilizer was the only means employed land, other than the usual cultivation, applied to plats 6 to 10 of the Purdy lOOO lbs. fertilizer per acre, i l.-iOO lbs. 4. fertilizer per acre. 2000 lbs. 5. fertilizer per acre. 0. No fertilizer. 500 lbs. 7. fertilizer per acre. 1000 lbs. 8. fertilizer per acre. 1500 lbs. 9. fertilizer per acre. 2000 lbs. 10. fertilizer per acre. ith this exception the fer- of adding fertility to the In 1901 no fertilizer was field. LOCATION AND CONDITIONS OF THE EXrERIMENTS. The location of the experiments was at Florida, Orange county, N. Y., a region where onion growing is an important industry. The soil is the kind so highly regarded by onion growers, being black, peaty and friable, with a water table about two feet below the surface, except in the time of a severe drought. Such soil appears to allow the continuous production of the same crop Avithout the apjiearance of the unfaA'orable conditions which N EW YoKK Agricdltural Experiment Station. 230 follow with most soils where a rotation of crops is not practiced. During- the course of the experiment insect and fungus troubles and excess of water caused more or less damage, the instances of which will be mentioned in the proper connections. As stated, two fields were used, the Purdy field for four years and the Mars field for one year. In 1897 the former field pro- duced a crop of onions, receiving a small application of commer- cial fertilizer. Previous to 1897 the crops had been grass, corn and potatoes. The Mars field had been generously manured in previous years. ' NOTES. In the conduct of these experiments approved methods of cul- ture were followed at the hands of experienced onion growers. The fertilizer was sown broadcast before the drilling of the seed. The planting generally occurred late in April and the harvesting of the crop during the last half of August. Unfavorable conditions prevailed to some extent every year of the experiments. In 1898 Plat 10 of the Purdy field was flooded for a short time soon after the young plants made their appearance. Again in 1900 Plats 7, 8, 9 and 10 were partially flooded on two occa- sions, but this occurred late in August, not long before the crop was gathered, and as the onions which had rotted were weighed, the figures given show the approximate yield. The crops suffered more or less every year from smut, mildew and the maggot, but the plats appear not to have been injured to a sufficiently unlike extent to seriously impair the accuracy of the work in measuring the yield. In 1900 and 1901 a mixture of sulphur and lime in the propor- tion of 2 to 1 by weight was sown with the seed at the rate of 150 lbs. per acre. This was sown as a preventive of smut. Several tables follow showing the plat yield, the acreage yield calculated both in pounds and bushels and the outcome of the experiments considered from a financial point of view. 240 Report on Ckop Pkodcction of the Table I. — Yield of Onions on Purdy Field foe Four Years, 1898-1901, BY Plats. Yield per plats.* Quantitv of fertilizer , • ^ Plat No. per acre. 1K!18. 1899. 1900. 1901. Averasie. Lhs. Lbs. Lbs. Lbn. Lbs. 1 None 494 823 17041^ 605 906 2 500 lbs 681^2 104.51/0 2704 1257 1422 3 1000 lbs 7021/2 1148 28131^ 1425 1522 4 1500 lbs 8311/2 1416 2736 1423 1601 5 2000 lbs 821 13S21/2 20981/2 1499 1600 . 6 None 602 858 2118 923^ 1125 7 500 lbs 693 1108 2605 1117» 13!J6 8 1000 lbs 721 12591/2 2G3G 1218^ 1458 9 1500 lbs 835 13411/2 25881/2 1398- 1541 10 2000 lbs 8141^ 1298 2540 1371' 1599» 1 Size of Plats 1 to 9, .0794 acre ; Plat 10, .0843 acre. * Plats 6 to 10 received no fertilizer in 1901. 3 Calculated to yield for 0794 acre. Table II. — ^Acre Yield of Puedy Field Showing Increase From Feb TILIZERS. Average yield per acre for four years. Quantity of fertilizer Increase over Plat No. per acre. Pounds. Barrels.' no fertilizer. £bl8. 1 None 11,415 76.1* 2 500 lbs 17,917 119.4 34.1 3 1000 lbs 19.177 127.8 42.5 4 1500 lbs 20.173 134.5 49.2 5 2000 lbs 20,160 134.4 49.1 6 None 14.175 94.5* 7 500 lbs 17, .590 117.3 32. 8 1000 lbs 18,371 122.5 37.2 9 1500 lbs 19,417 129.4 44.1 10 2000 lbs 20,147 134.3 49. 1 Barrel. 150 Ib.s. S Average Plats I and 6, 85.3 bbls., taken as yield with no fertilizer. Increase for each ariilition of .Wii lbs. feitilizer. £hls. 34.1 8.4 0.7 S2. 5.2 6.9 4.9 Table III. — Average Yearly Profits Per Acre from Use of Fer- tilizers on Purdy Field. Quantity of fertilizer Cost fertilizer Plat No. used per acre. per acre. 2 500 lbs $6.25 3 1000 lbs 12.50 4 15JDITI0N AND CLASSIFICATION. The experiments were conducted in a number of different orchards, all of them in western New York. In each case .the condition of the trees was carefully noted. The crude petroleum used was purchased of the Standard Oil Company. It was dark green in color and had a specific gravity of 44°. An emulsifying pump of the type that emulsifies the oil and water at the nozzles was used in all of the experiments. The spray was very fine and the emulsion thus made was excellent, the oil being so thoroughly broken up that it did not wholly separate from the water in the 1,000 cc. graduate in which the tests were made for over two days. To avoid error in the amount of oil applied the pump was frequently tested. These tests showed only very slight varia- tions in the percentages. Much pains was also taken to make the applications thorough and uniform. The spray was directed upon the tree until it New York Agricultural Experiment Station. 251 began to drip slightly. By so doing all of the emnlsion that would adhere was applied to each tree. It is apparent that if less were applied it would be very difficult to tell whether the work had been thorough or not. On the other hand prolonged spraying and consequent over drenching would be merely a waste of material and would have practically no effect upon the quantity of petroleum that finally clings to the tree because the excess of emulsion does not adhere to it. Also if the machine emulsifies the oil and water as thoroughly as in our own experi- ence there will be no danger from free oil as the excess of emul- sion will drain off before separation can take place. Hence a tree sprayed to a moderate excess with an emulsion containing say 25 per ct. of petroleum would finally retain no more petro- leum than if sprayed merely to the dripping point. If, there- fore, the spray is directed upon each tree until its full capacity to hold the emulsion is reached, that is until it begins to drip, an amount of oil in proportion to the percentage indicated is practically maintained. However, it is to be noted that in the case of unusually rough or cracked bark or open wounds that would hold the emulsion the accumulation of oil may cause injury. It is also possible that, in case of over drenching, the oil that would soak into the ground might cause injury to the roots. character of crude petroleum. Crude petroleum is an oily, inflammable liquid varying in color from very dark brown to greenish tints. By refining it yields a number of valuable products including paraffin, lubricating and illuminating oils and a series of highly volatile oils. It is the heavier oils that make it especially valuable as an insecti- cide. Crude petroleum varies in appearance and composition according to the locality from which it was taken. The eastern oils are said to vary greatly from the western and most foreign oils, the former having a paralfin and the latter an asphalt base. The true indication of the safety of petroleum as an insecticide evidently depends upon its specific gravity; as it has been found that petroleum having a specific gravity of 43° or 252 Kp:port of the Department of Entomology of the above (Baum^ oil scale at a temijerature of GU° F.) is less likely to injure the trees tliau i)eti'oleum of a lower specific gravity, although oils of a lower specific gravity have been successfully used in some instances, notably in Canada. REASON FOR EMULSIFYING THE PETROLEUM. A very thin film of petroleum covering the entire tree is all that is required to kill the scale. Theoretically, by using a very fine nozzle, the undiluted petroleum might be applied in a thin film but in practice it has been found very difficult if not iui})os- sible to make the treatment thorough without applying a dangerous and wasteful excess. For this reason emulsifying the petroleum with water is desirable as the tree can then be thoroughly wet without applying an excess of oil. SERIES I. EXPERIMENTS TO DETERMINE THE EFFECTS OF CRUDE PETROLEUM UPON SOUND TREES. ORCHARD i: PLUMS, PEARS AND CHERRIES. This orchard consists of 152 plum, 13 pear and 13 cherry trees. The i3lums consist of Monarch, Reine Claude and Quackenboss varieties; the pears Bartlett and the cherries Montmorency. The Reine Claudes, which include about one-third of the plums, are old trees that have been weakened by disease and decay but usually bear a small crop of fruit. The Monarch and Quackenboss trees are sound, especially the latter which are unusually vigorous. The orchard has been kept continually under high cultivation. Summaries of treatments and results in Orchard I are given in Tables I, II and III. The checks, which are not given in the tables, consisted of a large number of trees of the same varieties in adjoining rows. In every case they showed no indications of injury during the winter. 'Weather duviiKj tests in Tabic I. — Winter treatment. Trees sprayed Dec. 22 to 24. Average temperature of the three days 39°, cloud}'. Weather during the week following usually cloudy with average temperature of 21)°. New York Agricultukal Experiment Station. 253 Table I. — Winter Spraying in Orchard I. Trees. Slniigth of Number Kind. treated. Age and description. petro- leum. Ver ct. Plum; Reiue Claude. 7 Old, weakened by dis- 2.J ^ ease. Results. 2.J Two slightly injured, remainder unii> jured. 25 No injury. 25 No injury. 40 Four seriously In- jured, remainder slightly. 40 All slightly injured. 40 All slightly injured. 60 GO- GO 100 100 Five dead and three seriously injured. All slightly injured. All slightly injured.. All dead. Monarch 4 Full bearing., sound.. Quaclieuboss . G Full bearing, sound and unusually vig- orous. Reine Claude. 8 Old, weakened by dis- ease. Monarch 5 Full bearing, sound.. Quackenboss . G P^ull bearing, sound and unusually vig- orous. Reine Claude. 8 Old, weakened by dis- ease. Monarch 6 Full bearing, sound.. Quackenboss . G Full bearing, sound and unusually vig- orous. Reine Claude. 6 Old, weakened by dis- ease. Monarch ..... 7 Full bearing, sound.. 100 Three dead, remain- der seriously in- jured. Quackenboss . 6 Full bearing, sound 100 All seriously injured. and unusually vig- orous. Pear: Bartlett 3 One year planted, vigr orous. Bartlett 3 One year planted, vig- orous. Bartlett 1 One year planted, vig- orous. Bartlett 1 One year planted, vig- orous. Cherry: Montmorency. 2 One year planted, vig- 25 No injury. orous. Montmorency. 2 One year planted, vig- 40 No injury. orous. Montmorency. 3 One year planted, vig- GO No injury. orous. Montmorency. 2 One year planted, vig- KJO No injury. orous. 25 No injury. 40 No injury. GO No injury. 100 No injury. Weather during tests in Table II. — Spring treatment. Trees sprayed April 18. Temperature 52°, cloudv. Weather during the week following usually cloudy with average temperature of 48°. I 254 Report of the Departme>;t of Entomology of thh Table II. — Spring Spraying in Orchard L Trees. Strength f • s of Niinibor petro- Kind. tieaied. Age aud (Jeacription. leum. Besults. Per ct. Plum: Keiiie Claude. 6 Old but usually bear 25 Two dead, three seri- a small crop. ously injured. Mouarch 5 Full bearing, sound.. 25 Uninjured. Quackenboss. . 3 Full bearing, unusu- 25 Uninjured. ally vigorous. Reine Claude. 4 Old but usually bear 40 Three dead, one seri- a small crop. ously injured. Monarch ..... 5 Full bearing, sound.. 40 Two sli^iitly injured, remainder unin- jured. Quackenboss.. 3 Full bearing, unusu- 40 Uninjured. ally vigorous. Reine Claude. 4 Old but usually bear 60 All dead. a small crop. Monarch 6 Full bearing, sound.. 60 Two dead, remninder seriously injured. Quackenboss. , 3 Full bearing, unusu- 60 Slightly injured. ally vigorous. Pear: Bartlett 1 One year planted, vig- 25 Uninjured. orous. Bartlett 1 One year iilanted, vig- 1 60 Uninjured. orous. Cherry: Montmorency. 2 One year planted, vig- 40 Uninjured. orous. Montmorency. 1 One year planted, vig- 60 Uninjured. orous. Time of tests in Table III. — Winter and spring treatment. Trees sprayed Dec. 24 and April 18. Table III. — Winter and Spring Spraying in Orchard I. Trees. Strength , > of XniL.ber petro- Kind. treated. Description. leum. Besults. Per ct Plum: Quackenboss. . 3 Full bearing, unusu- 25 Uninjured. ally vigorous. Reine Claude. 1 Old, but usually bear 40 Dead. a small crop. Monarch 1 Full bearing, vigor- 40 Dead. ous. Monarch 3 Full bearing, vigor- 60 Dead. ous. Reine Claude. 2 Old, but usually beai-s 100 Dead. small crop. Monarch 1 Full bearing, vigor- 100 Dead. ous. Pear: Bartlett 1 One year planted, vig- 40 Seriously injured. orous. Bartlett 1 One year planted, vig- GO orous. Dead. Cherry: Montmorency. 3 One year planted.... 40 All seriously injured. New York Agricultural Experiment Station. 2dS SUMMARY for ORCHARD I. In the winter treatment (Table I) none of the healthy plum trees showed evidence of injury by the 25 per ct. emulsion but all were injured and some killed by the 40 per ct. and above. In all cases the old trees (Reine Claude) were more sensitive to the treatment than the others, some of them being injured by the weakest emulsion, half of those treated with tlie 4 per ct. seriously injured and all killed or seriously injured by the stronger emulsion and the undiluted petroleum. The younger trees, Monarch and Quackenboss, especially the latter, stood the treatment better, being uninjured by the 25 per ct. and only slightly by the 40 per ct. emulsion. The pears and cherries were uninjured. The spring treatment (Table II) shows even more serious injury to the Reine Claudes than the winter treatment but the other varieties consisting of younger trees were less affected being uninjured by the 40 per ct. and only slightly by the 60 per ct. emulsion. Both the pears and cherries were uninjured by the highest percentage of petroleum (60 per ct.) used in this series. As was to be expected the winter and spring treatments com- bined (Table III) caused more serious injury than either of the single applications. The Quackenboss was the only variety treated with 25 per ct. emulsion and the trees were not in- jured. Reine Claudes and Monarchs were treated with the higher percentages with the result that every tree was killed. From these results it appears that the spring treatment was slightly less injurious than the winter treatment while the two combined proved fatal except with the 25 per ct. emulsion. The 40 per ct. and stronger emulsions caused so much injury to the plum trees as to indicate that they are dangerous. The only exception was the Quackenboss trees which were uninjurc d by the 40 per ct. although unable to withstand the higher percentages. The lack of injury to the pear and cherry trees even by the undiluted petroleum indicates strongly that these trees are much less susceptible to crude petroleum than the plums. 256 Report of the Department of Entomology of the SERIES II. SPRAYING EXPERIMENTS TO DETERMINE THE EFFECTS OF CRUDE PETROLEUM UPON HIBERNATING SCALES. The experiments of this series were conducted in Orchards II, III and IV. orchard II : PEARS. The experiments of Series II were begun in this orchard which consists of 101 pear trees, nearly all of which are standard Rartletts. The trees were planted about six years ago and except for the first two years have received but little care. During the past four years the orchard has been in sod. It is quite probable that some of the trees were infested when planted but the scale was not noticed until the spring of 1898. ^Yhen the experiments were begun in 1900 most of the trees were extensively infested, many being encrusted on the trunk and larger limbs. Before the experiments were begun they were carefully trimmed and the orchard was divided into two sections. The first, containing 54 trees, was treated with crude petroleum, and the second, containing all of the remaining trees, except a few reserved for checks, with hydrocyanic acid gas as noted in a subsequent section of this bulletin. Summaries of the spraying experiments and results in this orchard are given in Tables IV, V, VI and VII. The check trees are not recorded in the tables. In every case the scales had multiplied rapidly. The live ones were abundant upon both the new and old growth. New York Agricultural. Experiment Station. 257 Weather during tests in Table IV. — Fall treatment. Trees sprayed Oct. 23. Temperature 64°, cloudj'. Weather during tlie week following usually bright, with average temperature of 59°. Table IV. — Fall Spraying in Orchard II, Trees. Strength of Number petio- Kiud. treated. Degree of infestaiion. leiim. Results. Per et. Pear: Bartlett 2 Both extensively' in- 25 Scales not affected, fested. trees uninjured, r 1 Extensively in- "1 I fested. I Bartlett 2<( ^ 40 Scales dead, trees un- j 1 Moderately in- I injured. I fested. J - " Estrnsively infested" as used in this and other tables me.ins that the trees were encrusted on parts of the trunk and some of the larger limbs. Modetately and slightly infested mean to a less degree. Time of tests in Table V. — Winter treatment. Trees sprayed Dec. 24. Table V. — Winter Spraying in Orchard II. Trees. Strength of Number potro- Kind. treated. Degree of infestation. leum. Results. Per et. Pear: Bartlett 4 1 Extensively, 1 mild- 25 Scales not affected. ly and 2 slightly in- Trees uninjured, fested. Bartlett 4 2 Extensively and 2 40 Scales dead.' Tree.s slightly infested. uninjured. Bartlett 4 3 Extensively infested 60 Scales dead.' Trees i (2 nearly dead), 1 uninjured. moderately infested. Bartlett 4 3 Extensively infested 100 Scales dead. The two (2 nearly dead), 1 trees most seriously moderately infested. infested were killed the others slightly injured. 3 On some of the trees spraypd with the high percentages of crude petroleum an occasional live scRle was found, but as they were always upon some small twig tint might easily have escaped thorough treatrni'tit tbey were not considered as affecting the results. 17 258 Keport of the DepartxMent of Entomology of the Time of tests in Tahlc VI. — Spring treatment. Trees sprayed April 18. Table VI. — Spring Spraying in Orchard II. Trees. Strength , • . of Nil 'II Iter petro- Kirid. Heated. Degree of infestation. leuin. Kesiilta. ]'er ct. Pear: Bartlett 4 2 Extensively and 2 2o Scales not afferted. mildly infested. Trees uninjured. Bartlett 4 2 Extensively and 2 40 Scales dead. Trees mildly infested. uninjured. Bartlett 4 3 Extensively and 1 GO Scales dead. Trees mildly infested. uninjured. Bartlett 4 All mildly infested. . . 100 Scales dead. Trees uninjured. Weather during tests in Table VII. — Winter and spring treat- ment. Trees sprayed Dee. 8 and April 18. Dec. 8 temperature 29°, cloudy. Weather during the week following alternating cloudy and fair with average temperature of 22°. Tap.le VII. — Winter and Spring Spraying in Orchard II. Trees. Strength • ^ ot Nniriber petro. ICiutl. tioaled. Degree of infestation. leum. Kesults. Per et. Pear: Bartlett 3 1 Extensively and 2 25 Scales show slight mildly infested. effect of treatment. Trees uninjured. Bartlett 4 2 Moderately and 2 40 Scales dead. Trees slightly infested. uninjured. Bartlett 4 2 Moderately and 2 60 Scales dead, 1 tree slightly Infested. dead, 1 nearly dead and 2 seriously in- jured. Bartlett 4 1 Extensively and 3 100 Scales dead, 3 trees moderately infested. dead, 1 nearly dead. summary for orchard II. The experiments in this orchard indicate that the 25 per ct. emulsion cannot be depended upon to kill the dormant scales, while the 40 per ct. emulsion gives satisfactory results. The power of the pear tree to resist the injurious effects of crude petroleum is also indicated. There was no apparent injury to any of the trees sprayed once, although many were much weak- ened by the scale, except in one case where the trees were nearly New York Agricultural Experiment Station. 259 dead at the time time of spraying. These trees were killed by the undiluted petroleum (Table V.), The trees sprayed twice (Table VII), with 60 per ct. and undiluted petroleum were killed or seriously injured in every case. But those receiving the weaker emulsions, 25 and 40 per ct., were uninjured in- dicating that pear trees may be spraj^ed twice, once during the winter and once during the early spring with a petroleum emul- sion strong enough (40 per ct.) to kill the scale without being injured. ORCHARD hi: APPLES. This orchard consists of thirty-two Baldwin apple trees in full bearing. All were infested but none sufficiently to be seriously weakened. They have been well cared for and, except for the scale, are in good condition. The experiments were undertaken principally to ascertain whether large trees moderately infested with the scale could be satisfactorily treated with crude petroleum. The trees were too large to make thorough spraying practicable without severe pruning. They were therefore cut back severely in October and finally sprayed, with the results shown in Tables VIII and IX. WmtJier during tests in Table VIII. — Winter treatment. Trees sprayed Dec. 20 to 22. Average temperature of the three days 33°. Table VIII. — Winter Spraying in Orchard III. Trees. Strength > of Degree of infestation. leum. Resnlts. Per ct, ^-. ^ .. 1 Extensively, 3 mod- 25 Scales not affected, erately and 4 slight- Ti-ees uninjured, ly infested. 1 Extensively, 2 mod- 40 Scales dead, except erately and 5 slight- on some of the ly infested. small branches where many live ones were found. 5 trees dead, re- mainder seriously Injui'ed. Kind. Number treated Apple: Baldwin . . • • • o Baldwin .. • • > o 200 Report of the Depautment of Entomology of tub Weather chiving tests in Table IX. — Spring treatment. Trees sprayed April 19. Temperature 34°, cloudy with slight rain. Weather durinj;; the week following cloudy with frequent show- ers. Average temperature 48°. Tahle IX. — Spring Spraying in Orchard III. Ee^^nlts. Scales not affoct'efl. Trees nn injured. Scales dead, except on some of the small branches where many live ones were found. Trees un- injured. Trees St I en at li of ' Number "" perro- Kind. tleattd. Degree of infestation. ii mil. rer et. Apple: Baldwin . . ... 8 2 Extensively and G 25 i moderately infestt ■d. Baldwin .. ... 8 1 Extensively and 3 40 i moderately infested. SUMMARY FOR ORCHARD III. The results in this orchard show only partial success for the treatment. As with the other experiments the 25 j^er ct. emulsion had no noticeable effect on the insect. The lack of thorough work with the 40 per ct. emulsion appeared to be due to the difficulty of reaching every limb and twig on large trees. This seems evident because nearly all of the scales v.-ere dead, the live ones being found only on a few small branches that might easily have escaped thorough treatment. The serious injury to the eight trees sprayed during the winter with the 40 per ct. mixture was unexpected. As apples are not considered especially sensitive to treatment with crude petro- leum and similar insecticides and as the other apple trees in- cluded in the experiments were not seriously injured by similar treatment it seems probable that some other factor besides the petroleum must have had an important inlluence. The apple trees that were uninjured by the winter treatment of 40 per ct. emulsion were not trimmed just before being sprayed as was the case with the injured trees, and as the pruning was unusually severe it may have weakened the trees sufficiently to cause them to succumb to the treatment. New York Agricultxjkal Experiment Station. 2G1 ORCHARD IV : PEACH, PEAR AND APPLE TREES. This small orcbard consists of ten peach, pear and apple trees just coming into bearing. The orchard has evidently received fairl}^ good care, and until two or three years ago the trees were thrifty. Recently most of them have shown signs of weakness, probably due in part to the San Jos6 scale. The treatment and results are summarized in Table X. Time of tests in Table X. — Winter treatment. Trees sprayed Dec. 20 to 24. Table X. — Winter Spraying in Orchard IV. Trees. Strength , • , of Number petro- Kiiid. treated. Degree of iufeatation. leum. Resalts. Per ct. Peach: Var. unknown. 2 Slightly infested 25 Scales not affocfed. Trees slightly in- jured. Yar. unknown. 1 Slightly infested 40 Scales dead. Tree seriously injured. Pear: Var. unknown, 2 Slightly infested 40 Scales dead. Tree uninjured. Apple: Var. unknown. 6 Extensively infested.. 40 Scales dead, 2 trees seriously injured, remainder unin- jured. SUMMARY FOR ORCHARD IV. In these experiments also, the 25 per ct. emulsion did not Idll the scales while the 40 per ct. was effectual. The peach trees, although no more seriously infested than the pears, were slightly injured by the 25 per ct., and seriously injured by the 40 per ct. emulsion. Two of the apple trees were injured by the 40 per ct. emulsion but not seriously. ORCHARD V: PLUM TREES. This orchard consists of twenty plum trees which have recently come into full bearing. All of them are extensively infested and somewhat weakened by the San Jos^ scale. As shown in the following table half were sprayed in the spring 262 Rki'Ort of the Departmext of Entomology of the with a resin wash and the remainder with a whitewash known as government whitewash. Both have been suggested as being effectual against the San Jos6 scale. They were made after the following formulae: Resin Wash. Resin ICA pounds Soap 21 pounds Fish oil 1% quarts Water 21 gallons Boil the resin, soap and fish oil in about one-fourth of the water until dissolved. While boiling, gradually add remainder of the water. Care should be taken not to add the cold water too fast as it has a tendency to precipitate the resin. Government tcliitcicash. Slaked lime ^4 bushel Salt % bushel Rice 6 pounds Glue 2 pounds Water 10 gallons Boil the rice with enough of the water to make a moderately thin paste. Dissolve the glue in a small amount of hot water and boil the salt and lime in what is left of the ten gallons of water until a thin whitewash is formed. Then add the rice and glue solutions to the whitewash and boil for at least half an hour. Weather during tests in Table XI. — Spring treatment. Trees sprayed April 12. Temperature 47°, fair. Weather during the week following usually fair with average temperature of 51°. Table XL — Treatments in Orchard V. Trees. Nnn)bet Dpgrt e of Kind. treated. infestation. Mixture used. Results. Pr.tiM: Var. unknown. 10 8 extensively R es i n-1 i m e Scales not af- and 7 moder- mixture. fected. Trees ately infested. uninjured. Var. unknown. 10 4 extensively Government Scales not af- and G moder- whitewash. fected. Trees ately infested. uninjured. ^Ew York Agricultural ExPERniEXT Station. 203 SUMMARY FOR ORCHARD V. Although twenty infested trees were used in these experi- ments and unusual pains taken to make the applications thor- ough there were no beneficial effects apparent in either case. The scales were breeding as rapidly during the summer on the treated trees as on the checks. Although there was a week of dry weather immediately following the applications, the un- favorable results may have been due in large part to an unusu- ally wet spring. The heavy rains washed both compounds almost entirely off before the summer was over. Further experi- ments with these washes seem desirable as they have not yet been sufficiently tested to prove or disprove their value. GENERAL SUMMARY. The experiments with crude petroleum include 321 fruit trees consisting of apples, cherries, pears, peaches and plums. The results were fairly uniform. In the experiments of Series I no injury was caused by the 25 per ct. emulsion, but in every case the 40 and higher percentages caused serious injury to the plum trees while the pear and cherry trees were practically un- harmed. The younger and more vigorous plum trees were injured less than the old and weaker ones. The experiments included in Series II show serious injury to peach trees by the 25 per ct. emulsion and equally serious injury to plum and apple trees by the 40 per ct. emulsion. In all cases of injury it is to be noted that the most serious injury was caused by the fall applications and by two applications- one in the fall and one in the spring. These results do not agree with those of Smith and Corbett previously referred to but agree in the main with those of Felt who, as previously stated, found that the undiluted petroleum caused serious injury to the treated trees. The experiments to ascertain the percentage of petroleum in the petroleum and water emulsion required to kill the hibernat- ing scales also gave uniform results. The 25 per ct. emulsion failed to affect the scales materially while the 40 per ct. killed 264 Report of the Departmrxt of EntomoivOGy of the) them in eA'erj instance. The faihire of the 25 per ct. to liill the scales does not agree with the results of Felt and Corbett who report success with a 20 per ct. emulsion. The reason for this is not readily apparent. It is to be noted, however, that although an examination of the treated trees made in the spring may indicate that the treatment has been successful, definite and final results cannot be obtained without several examina- tions during the following season. This is especially true in the latitude of New York State where a large percentage of the scales die during the winter so that during the spring but few live ones can be found. But later in the season after breeding begins the real condition can be much more easily determined. Taken as a whole the spraying experiments reported in this bulletin indicate the following: 1. Vigorous trees are probably less liable to injury by crude petroleum than weak ones. 2. Peach and i^lum trees are more sensitive to crude petro- leum than apples, cherries, or pears. 3. There is less danger of injury if trees are sprayed in early spring than during the fall or winter. 4. The 25 per ct. emulsion of crude petroleum and water can- not be depended upon to kill the hibernating scales in the lati- tude of Western New York, while the 40 per ct. has proven efficient. 5. Much pains should be taken to avoid over-drenching the trees. Onl^^ enough of the emulsion should be applied to wet the bark evenly and thoroughly. II. FUMIGATION EXPEROIENTS WITH HYDROCYANIC ACID GAS. introductory. Fumigation with hydrocyanic acid gas is now recognized as one of the best known methods of combating scale insects. The gas was first brought into prominence as an insecticide in 1886 New Youk Agricultikal Experiment Station. 265 by Mr. D. W. Coquillett who made extensive experiments in California. Altliongb it was being extensively used in Cali- fornia it received but little attention in the east until in 1897 when Prof. W. G. Johnson took up the problem of successfully combating the San Jos^ scale in Maryland. After extensive experiments he decided that two-tenths gram of 98 per ct. potassium cyanide per cubic foot was sufficient for outdoor fumi- gation of deciduous trees when in the dormant state, that dor- mant nursery trees should be fumigated with .25 gram of cyanide and buds, grafts and scions with not more than .16 gram. In connection with these experiments Johnson developed better methods of handling and applying the gas than had been l>reviously in use and called attention to its wide range of use- fulness until now it is employed in green houses, graneries, mills and other buildings subject to infestation by insects. In this State the gas is used extensively for fumigating dor- mant nursery trees. Until the past two years, however, but few attempts have been made to use it in the orchard and in this capacity it may still be considered in the experimental stage so far as this State is concerned. The experiments reported in this bulletin were begun during the fall of 1900 and continued during the following winter and spring. The principal objects of the experiments were to deter- mine the effects, if any, of the gas upon healthy buds and the strength of the gas required to kill the hibernating scales. CONDITIONS. The buds were fumigated in small box fumigators made espec- ially for the purpose. For the orchard trees fumigators of the type described in Bulletin 181 were used. All of the fumigators were carefully tested and found to be gas tight. The gas was generated in the manner described in Bulletin 194, page 382. The amount of cyanide used varied from .18 gram per cubic foot of air space to .3 gram as shown in the tables. 266 Rkpokt of the 1)i:i'autm1':int of Entomology of thb character op hydrocyanic acid gas. Hydrocyanic acid gas may be generated by bringing cyanide of potassium in contact with sulphuric acid. It is colorless, has a faint odor of almonds, and when inhaled, unless largely diluted with air, is very dangerous. Much care should therefore be used in handling it. CLASSIFICATION. The experiments were divided into two series. Series I included the experiments with uninfested buds and Series II the Hxperiments with the hibernating scales. SERIES I. EEFEOT OF THE GAS UPOIN BUDS. The following experiments with buds of a variety of fruits were undertaken to ascertain if possible whether bud sticks could be safely fumigated with the gas strong enough to kill the scale. The conditions were not entirely satisfactory as the treatment was somewhat delayed and the treated buds were not set in until the first week of August. This was out of season for most of the varieties. Also the treated buds were not set in until after the checks, which were budded at the proper time, and were placed about four inches above them where they were too high to be protected by the earth thrown against the trees during fall cultivation. In addition to this they were neces- sarily placed on the furrow sides of the trees thus endangering them to injury during cultivation. These unfavorable condi- tions must be in part, and probably in large part, the cause for the failure of the treated buds to set or grow, on the average, equally as well as the checks. New Yokk Agricultural Experiment Station. 2G7 < o 'A < o o Pi R M O H O < P P fQ 1-1 o p ft o Cu 3s o o u ^ (0 K e . e ■ CO .2 g a 5 Q "O 'O tj "3 ^ o aj o o o c •- 5^ fi O O o O rt X, 'f^ ot- ic o oD in o QO O 10 Cl I- CO o t- I- t- (M ^ CO rH C<> rH ci .P" = c: 2 o C5 CO 1-1 00 CI 10 CO d ai oj <,^muiTf2 -^ <1 ■* l^ l- CO CO CO rJ d CO d CO CO 10 c cc I- 00 a, o ira CJ CO TtH i!0 CO CI iH CC CI 1-1 CI CI IQ IC CO LC Oj 05 tH CO CI CJ CO CO iH CO C C< ji '^ I) s c: •- P ci CO O) c: ^ o _ q o a ';::'C S 9 f* Q K i— I C CS t-l o CO ^ (^ :-> ^4 tj tj p p p p p p 000000 .a ^ fl a ^ fl o tn u o .a coco CO CO CO CO CO ■PH tH r-, T-l iH T-( tH <6 d> <6 <6 d' <6 d t^ i- o 00 CJ CO CJ CO I- I- o CO I- o CJ o Eh "C '3 ij' '3 sJ o o c c .a O O c C c3 CO CO CO 05 >0 CO 1(0 CO o 05 CO C5 CO' CO CO CJ CO o I- CI CJ 04 T-l T-l ^ IfO -fi O T-l CJ CJ a CO CJ 02 CJ CJ O CS p o *•* -4-) rfi oj .— 5i p '^ p c ai o 4J a .Q O rt ai tc cc ■— ^ ^ ® O) (— < ^ O) w C3 a cu OJ a a c: C3 KW in CO CD TH C r-/ CO O c: IX CO ir: I- CJ o CJ CI CI CI CI I- CI r- C1 t 1- CICJ I- c p. c; .- - - 0. ce„ ^ a T' a " 'c '-' ^ t- 5 c J:; c: 1^ t^ ui tj t-l p p p p p cS c c to. ja ^ j3 ^ -a «M ^J.';?.';^;?^;?^ J3 M R ?- CI Cl CI CI CI CI CI CI CI CI X 00000 2C8 KErouT OF the Department of Entomology of the o o - :3 — , ii 'A H a ^ /'. -3 3 -a d % o If o "A 6£ a fl Oi o 0) 0) o w 1^ M w w -t1 C5 ^ o ■'1' o CO (M o Ol CO 'SI o ^ ZJ d cj ^ 0) o 0) be c; a ci 0* > K* c3 c3 rt a ?5 0) X u a 7= 1^ s -^ =5 'Z> 'i> <:i' <:> z> C^ CI CO o I- 05 O) CI CO CI o a OJ o •c -a ;; -3 . • O O o = •- O O o ^ rt CO' C-l CO -H lo >o CO O i-t '3' o CO Ci o; I- C5 CO CO O 05 O CO CI CI Tfl CI CI T-l -ti I— CO C5 >0 05 CI CI -fH C) CO iH a o C5 C5 CO CI CO • 05 • 05 ■ OJ m VI 72 t; ^ id ^ ^ -M 'ctt^ ,x ^Station. 2GU d a 0) 0) r— < • r~; . ■ ri .• OJ "3 .^ S o.=! o o r: o o O K X « t^C^H W H •^ -* CO C5 00 Tt* ^ 1- c o d I- r^ CO c: -^ I- -^ o GO GO T— 1 ,H :o ?.-. 35 lO o C5 S (N Ci !M <-< rH CO C^ rH 1- :c -tH 1- in iH CO 00 ci tM :o c-i s\i CO CO 03 ■M • . !» • CO 1 • . •/J -a a. IS o 1 5^ =^ P&5 is r-« r- 0) ) ) O +j O o c > 01 ii s OQ C p ^ o^-iD ;: 2 t> -t^ > ^ *? '.^<< \^ i^ < o Li Tt< -+ lO O GO CO -t< '^ - « n o O GO r1 "S^ O o t2i 6 o o o a a o H O M H H 3 - 2; a Q 2 a a ^ O) (D a) O +J -l-j -4-5 3 fl a a) 01 O) o o y X M M C-1 c« -o o C« 3i O o -t I- 00 CO tn o o ^ m xh o o ■-- oj aj 3 o o 0) M M rt a a ^ ci d IC :3 o t- ^PSJ;^ CJ (M ( TjH lO UO CO(N CO o o a 01 a. 3 G o m iH tj ^ 3 3 3 o o o ja x5 J T-\ T-\ T->. 3 -t-> a ^ 0) _o _aj ^~t r— t "o o 0) o o k! o H'^H I- i M CO 1-1 I O CO Ci C5 CO Ol CO r-( CI OJ CO c) CI OJ CO » id^ ^ o CJ o o> o 4) J3^ d o o O m M OT a a Oj o o 0) aa a ci ci a co mmm 05 CO O lO CZ) 05 l- CD OD CO CI T-< r-( CO i5i c5 -3 a o a (-1 O) --J !>. t*. 3 "S 3 3 3 O O O ^ J ^ v^-I ^1 vffi ;?: Cl CI CI CI C) C^l odd I- CO o I- CI o -p3 d 4) a CJ .3 O t»< (rf « -f< lO o I- C5 O ^ ^ t- ?| CI CI GO tr.' CI o I- CO CI 00 03 0) .a o W M o o ■p O) OJ o" 01 M CO . C3 Cj •2 O) (1) CO CI lo CO 03 CO Oi W (X) CO GO CO CI T-H CI CO I 3 Q 1 CO a Q a m >M a TI tj tj t-i 3 3 3 o o o AAA CI CI CI CI CI CI S S <6 CD CO CO CJ CO CD "3 o -ti -l-> 4J a a a 0) O) O) O) 4) 0) CJ CJ CJ y, ^ >< 1-00 O) o o CO O CD 00 i-H CI 01 O CO CO +^ il ^ __ o o "h 4; 0> 4) CO CO ■j:; aj 0) d a a Z 71' M OD Tfi CO ICCC o I- CDCO KO CO tH c^ th cj Tt< oOOi CD 00 CO tH CI Oi 01 ; A 1-hS °^ 000 AAA i-hr-i^i-t^ 000 00 CO CO t- Ci 05 CD r -^ I- CI l-H CI I- 00 I- CO 10 I- 05 10 00 I- d a a =1 a .0 a a ■a a a o a d o p. a c a a t-. -a ■B 01 a CO 272 RBroRT of the Dei-autmk.nt of Entomology of tuk a a a dj Qj ^ o o f3 3^ a . -=^ 00 ■ ■CO O a> o o M oo ■^ ^ J^ _ o o 7^ o 1) = — -^ - o y C' 02 M rt H a l-M O O I- o o t- 00 o CI i-l tH . o g ESS ~ * o o o " 'o 2 .G .s ;a i '=i'^.<^ CO c CO fc: CL 0. cy o Coy i^ XXX :3 KKS oj o t- <' J a, ■ fc'i' c: o i-I <3J 4) n c/j ci Ci O) O ■a . CO 'Jj CI CO -' m CO i-H (N I- 6 ^ . (S ■13 3 00 O'* C0C4M CI CO e 1 ii; Tf< I- CD a « 9» o o 4 00 O '^ C > 1 CO CI CI a c3 o , &< •*•* r— o ^ w « a C 3 ^ O ^ a _y V. S ^ rH c. ^ C OQ •^* o H u i;>« <5 K* r- 1 *i c New York. Agricultural Experiment Station. 273 Four hundred and eiglity-seven cherry buds were treated of which 77.8 per ct. set. The total number of checks was 497, of which 91.7 per ct. set. While the difference in these percent- ages is much greater than in the case of the apples it is hardly sufficient to indicate important injury by the gas. As pre- viously stated the somewhat unfavorable conditions under which the buds were grown, together with the fact that they were put in about two weeks late, might easily account for this difference. A more definite indication that the gas had little if any injurious effect is shown by the comparatively high percentages of buds set that were fumigated with .22 gram of cyanide. The strongest gas and longest exposure had practically no more effect than the weakest gas and shortest exposure. The growth also was in all cases equal or nearly equal to that of the checks. 18 274 Kepout of the Department of Entomology of the O 'J o c a I I— I 1^ ^ a xi .^rf Ef -o -d 0) -d o o o y o k> O O ',^ o o OOSO 4J O (M OGO OQ " ; .M ^t o ci CI CO § • e « 00 O C5 05 .a ►-* o -« 3 w >Q 1- CO O C^ ^ C^l CO 6 ^ Vj '^ iH ^GO IN .a CO lO (N CO 6 » 3 •{ a ^ 1 IS o M CO .id ^ 05 t» O O o o 3 3 02 03 01 <1) a 3 3 s cj rt ,^ ,^ Oi M O O O (M O lO (» I- .a ^ Tj^ t- O CO .5 iH i-l (M r-( o >5 CO ■* inco W CO (N iH ^ fQ r '^ 3. o ■ 5 0)'^ a ;:: ;= ;^ ''OS ^ 1.1 E-i ti = s :3 3 o o o o .3 .:= .3 .a M X e .-( rH I-l tH a « g . . • . o 05 00 IN IN Good. Excellen Exeellen Good. 05 IN CO CO CO 05 c; 05 1-1 T-i -t CO IN CO IN CO CO ■*! CO iro IN CO M CO o o CO o: id ^ .a ^ to CO o o a a -w *-> oi a; '3 rs ^^ I § CO CO 4) O) 1^ 1^ ^^ ►^ a a ci rt CO I- O ■*' CO IC 05 coco O) OO (N IN -O tH IN O H o -t-i ^ id — H o '^ a d O) o to ' «> i k. t~t ^ a CJ 4) ^^-, CI CI 05 CO CO CO Cl tH »n 05 IN IN CO CI 05 CC CO CI ■ CO 05 aS •r-i iw a ci <5M "3^ a a .a .a 3 a .a,a ;??;^ ;?;;?; MCI CI M CI IN CI C^l ©o 00 o a OJ 0) o o cj o O X X o th 11 eo CO t- Tf C5 CI CO CI CO 00 00 co 05 o Eh IN Cl 05 CI 05 00 C/D 00 CI CO CI -tH — ' CO CI CI CI ?5? C5Tt< I CO o .a (0 03 O => 05 oj 2 .Id •3-5 9 O cj OJ .a M O) O a ci aj o 3 3 00 00 ja x3 .a .a eg a e u a .a 3 u .a o e ."2 '5 (9 o s o M es U) o .0 tic a CI CI CI CI CI IN CI CI 00 o© New York Agricultural Experiment Station. 275 "O t3 -73 nd o o o O O O o O oooo a a • a r: • o y o o c M M o OOQCOO t- T-l O O -* iM O 00 oboo lO CD CO T+i ifl © o M !M (N 00 o CO CO 1-i CO CO C4 IN CO ift 00 gi lo o (N iJJ IN -i< m o oj a* t) o TO to 0) O) aa c3 ci COM +-> -^ O Ol to ■* O CO O 05 lO o I- CO 05 l- CiCicO'i* iH IN (N i-l tw ;ij Ui ;j S 3 3 3 0000 S J ^ J >JI vCI vff( --Jl (M II CO 10 lA O IN CO CO W ■* CD CD I- t-05 Tt< O O (N (N IN C^ O TO TO CO t« ^ ^ ^ ^ 0000 01 Qj 0) O) ^ ^ ^ ^ 0000 TO TO TO < i^ X X >^ CO 0? CO CI o ift t- CO ic CO CO lo -<< X X CO I- I- CO CO I- 00 I- S^ ci I': CO CO 1- -ti vi Ci a X X -^o X lo ■r-' ri -H O 1^ T-^ CI 71 T— ( ^H T— 1 t- ^H CI M T^ Tl " X O >0 IQ -t< (M CI C4 M CI 1-1 tH CI CO -t CI -^ CI -t •+ 1-t CI 1— 1 r-l CI iH CI r-l T~* a a a a a a "oJ " "o "o "cI "3 c :^ :^ :j o o X X X ;^ X S^ ICi C5 CI CI C5 IC X 5 CI -S) X O GO O X l- -. I- O '^ -+ :- "-D Ci O X -t X iH T-i CI T-l T^ CI CI I- X fc xti C) m 'fi m in ^ ^ +^^ ^^ ^ '■h o o o y o 2; n S ^ S - O g, i» o o y O! aj 03 M CO 03 ei , c: cj cj ci >i o T^ a^ CJ c^ gj a 3 a g 3 a cj .i* a ci c3 cj V^Av.r/^'Jirri ui <~, m vi m m .— -H -^ -^ -M ^ i) ^ o o o y a^ '^ w ^ o g, o o o o to Cj m ci , cS c5 M CO 0? (i> T^ i> aj cj a^ a rt a a a a c: ^^ c: cs ci rt xfiA'fimrjixn m (:, m m Tfi m *^ -fc-. /.^ ►> >> '^ o _ o o o o 0/ "^zj o) a.* u a> ^ y ^ ^ ^ ^ o g, o o o o ui 'ii in xn 'ji m ci , c3 ci c; cS o ii o a rt s a - a rt ,0 05 Tt< CI CI .-a >> . =5 ^ - o i; U « a a a a a 3 o o o o o o a ^ ;a j^ ^ X X XX XX r-l r-l r-t rH rH r-l dddddd o o d ,5 03 ^ 1- g j- S^ CO r-< ci c a a a a a a c c c o o o ^ ^ ^ ^ ^ J CI CI CI CI CI c« Cj 03 c; , cs 3 c: 3 c: I' cj 10 ca CO CO ^ CO CO KO CO M^ CO Ci ^ CO LO rH C5 CO i-O 1- CO I- L- C5 CO 10 X I- 1.0 I- X CO 1- 10 X rH I- CO X I- I- I- X >0 -tH KO rH I- C5 c: t- CO rH CI X -f CO X 01- 1- CO I- X o t- CO CO I rH C5 C^ o ^ a c T- a BS« a a a coo CI CI CI CI CI CI JSJeW iOUK AgKICLLTUKAL Exi'EKlMENT JSXATlON. 277 a a a a> cj o) "oj "al "a; o o o y, y y, EqHH C5 CC C X l~ X l- C5 0) ;= — ^ o o o a; a: a; ci CJ rt C C' i- Oi a; aj CO CO -f I- lO X w o 'J2 -j: 'Ji 02 >v« -i-J /> /> ^^ /-^ CJ O O CJ CJ o "^ c- o o o 02 aj O! t» CC CO c5 . c3 Cw C3 Cw a; o a> 3 K S a 2 9 « 5" « « rt <^ o I- ■ c^ O X iH -ti O CI X X Tf< CO CO t- CO to m o X ^ ^ cs ^ s a "o ^ CJ C^ O CJ VI 'iJ r/1 Hi m Hi c; . c! c3 c3 c: a; T' o c u qj a S a a a a rt ^ K c: rt cj CD CO Co o CI iH l-^ r a a. 2i a -3 ^oi a -^^^ o t/! CD -H O O CD CO T-l CI rH T-l CI O I 1^ CD' O X CI CI C4 CI CI C^l i-l '^ O — ' •-1 a jj CO I- M rt o s o oj a c H t; ^.^ t-. . ct ^ ^ Ct «-i C3 Cj rX HHpqHM*)^ T-l Cl C5 CO lO C5 Vi ^ ^' Cj j_, ^ j5 -5 5 a "3 -S i "^ ai a a^ T^ a T" r^ >^ ■^ ;:: a^ p :^ ^ S H W H W ^ lO CO X CO CI Cl CD CO T-l CI CO CI T-l T-l CO o a S 3 O P. P 3 O a t*. CI CI 01 CI Ti n d do o o o o o o X5 ^ ^ ;a ^ ^ o c o o o o CO CO CO CO CO CO iS S S S o c c c o o ;a ,a ^ ^ ^ ^ o o o o o o CO CO CO CO -.'1 CO <6 ^