116 HOLSTEIN -FRIESI AN CATTLE. 



3d. The change of the time of calving of many of the cows, during the 

 continuance of the experiment, from spring to fall introduced an element of 

 uncertainty and unfairness entirely destructive of fair competition. 



4th. The method of comparing the different individuals and breeds by cost 

 of fat production per pound instead of by net profit for a unit of time", fails 

 to show the most profitable cows and breeds. 



5th. The difference in age of the different heifers at the time of commenc- 

 ing their first and subsequent periods of lactation introduced another element 

 of unfairness. 



6th. The failure in the tabulation of results to take account of several 

 important items of income and of cost was grossly unfair to certain individuals, 

 and, alone, had the cows been properly selected, might easily have been suffi- 

 cient to totally change results. 



Some of these criticisms, it will be observed, refer to the fundamental 

 methods of conducting the experiment, while others less serious merely refer 

 to the methods of reporting the results. To understand the matter fully, 

 reference should be had to the tables of the report already mentioned. 



A few words of further explanation may perhaps be permitted. Where the 

 character of a breed of cows is sought to be tested by the capacity of from one 

 to four individuals of that breed, it is clear that if the test is to be of any value, 

 those individuals must be fair representatives of their respective breeds. The 

 number used is too small for the elimination of individual variations by the 

 process of averaging. The question therefore meets us at the outset what is 

 "a fair representative " of a breed ? Is it one of the best ? Is it an average of 

 the entire number of individuals of the breed ? Is it one selected by lot ? If 

 none of these, what is it? It should be noted that the selection and rearing of 

 heifer calves for the test in no respect meets the difficulty. It merely throws 

 the question of selection back one generation and introduces an additional 

 uncertainty the question of heredity. Without discussing the matter, I will 

 merely indicate my own impression, that as these tests are conducted for the 

 purpose of assisting practical dairymen in their selection of herds of practical 

 dairy cows, much the same method should be employed in selecting the herd 

 for test as would be employed by an intelligent dairyman in building up his 

 private herd a somewhat extensive process of selection and weeding. (Com- 

 pare work of the Cornell Station, Bulletin No. 52.) Were this process applied 

 to the Geneva herd, it is apparent that, notwithstanding Dr. Collier's boast, 

 " one of the finest herds ever brought together" (see page 17 Eleventh An. 

 Rep. ) , many members of that herd would no longer remain to disgrace the breeds 

 they are supposed to represent. To one who has any knowledge of the Holstein - 

 Friesian cow, the claim that Essel 2d, Tolsma Artis or Beauty Pledge represent 

 that breed is ridiculous. Two of these cows, aged at the date of second calving 

 three years and two months and four years and two months respectively, pro- 

 duced during the first month of their second period of lactation an average of 

 less than 1 Ib. of butter fat per day. The weight of the cows was 1,195 and 1,045 

 Ibs. respectively. The average per cent of fat for one during the first three 

 months of lactation was 3.31, of the other 2.36. Beauty Pledge (age not given 

 and record not included in tables for some reason which I have not seen stated) 

 gave in her first seven months, first period, a daily average of 20.8 Ibs. of milk 

 testing 3.14. Her best month showed a daily average of nearly 0.71 of a Ib. of 

 butter fat. Let us compare the records of these cows with those of others of 

 the same breed at other stations. In 1892 the three mature Holstein cows of 

 the Vermont Station herd gave milk averaging 3.63 per cent fat and 1.11 Ib. 

 butter fat per day throughout the entire milking period ; the total amounts of 

 butter fat for the three cows for the year being 381, 323 and 364 Ibs. respect- 

 ively, an average butter capacity of over 400 Ibs. The Minnesota Station herd, 

 the record of which for 1893, bulletin 35, is before me, contained one Holstein- 

 Friesian and one grade of that breed. The pure-bred cow weighed 1,127 Ibs., 

 was milked 331 days, gave 10,087 Ibs. of milk and 374 Ibs. of butter fat; costing 

 11.6 cents per Ib. The grade gave 408 Ibs. butter fat costing 10.4 cents per Ib. 

 the best record of any cow in the herd. As this last cow was only one-half blood 

 Holstein, her dam's breeding being unknown but surmised to be a grade Jer- 

 sey, her record is perhaps of little pertinence here. The best Jersey cow in the 

 herd gave 354 Ibs. of butter fat, costing 10.6 cents per Ib. She weighed 877 Ibs. 

 and was milked 346 days. No credit was given these cows for skim milk or 

 manure. 



