1920] DAIRY FARMING DAIRYING. 677 



herd, which is located In western Virginia, has been under a uniform system of 

 management and apparently selection on a basis of yield has not been exten- 

 sively practiced. Kecords of 1,741 lactations were used in the study. 



The data showing the change in production with age were fitted by the 

 method of least squares to an equation of the type previously used by Pearl and 

 his collaborators for similar data, the equation in this case being 



1/=33S7.912— 99.S8;U— 0.4.87a.-'+289G.219 log.oO? 

 where y is the 8-nionths milk yield in pounds and the age in years equals 

 1.25-|-0.5a'. According to this foruuila the age of maxinuim yield is 7 years 2.4 

 months. Since similar equations have been used successfully to describe growth 

 changes, it is suggested that the change in milk yield with age is due to growth 

 of the udder, 



Pearson's generalized probability curves were fitted to the frequency distribu- 

 tion of eacli age. Eight of the nine curves belonged to tyiies I or II, and sinte 

 they were not markedly unsymnietrical it was found that the distribution could 

 also be satisfactorily graduated by the normal curve — a result which justifies 

 the procedure of Rietz (E. S. R., 22, p. 278) in estimating the " tail frequencies " 

 of advanced registry data (i. e., the records eliminated through the requirement 

 of mininmm production for entry) by means of the normal curve. 



The last sections of the paper deal with the mutual correlations of the differ- 

 ent lactation records of a cow and also the correlation between particular lacta- 

 tion records and the total of the first five lactations. The general results have 

 been noted from the author's abstract in Maine Station Bulletin 283 (E. S. R., 

 43. p. 174). 



The variation of milk secretion with age in Jersey cattle, J. W. Gowen 

 {Maine Sta. Bui. 2S6 {1920), pp. ^5-60).— This is an abstract of the preceding 

 publication with the exception of the material in the last sections. 



Rules for testing dairy cows for advanced registration, J. B. Fitch and 

 F. W. Atkeson {Kansas Sta. Circ. 82 {1920), pp. 12, fig. 1). — Includes breed 

 regulations and rules for supervisors. 



Sunflower silage {Michigan Sta. Quart. Bui., 2 {1920), No. J,, pp. 163, 16^).— 

 Sunflowers chopped and put in a silo with corn silage above and below were fed 

 the following winter to dairy cows. During a week of sunflower silage feeding, 

 the milk produced by 24 cows was 11.65 per cent lower than 'the yield in the 

 previous week when corn silage was fed. There was some recovery in the third 

 weelc when both kinds of silage were fed in equal amounts, but a return almost 

 to the sunflower silage level in the fourth week when corn silage alone was fed. 

 " In this feeding trial sunflower silage compared very unfavorably with corn 

 silage. However, the drop in milk flow when the cows were changed from the 

 mixed corn and sunflower silage to clear corn silage would indicate that the 

 mixture may be preferable to straight corn silage, and work along this line will 

 be contiinictl during the coming season," 



Wild sunflowers for silage, J. B. Fitch {Breeder's Gaz., 78 {1920), No. 7, 

 p. 254). — Wild sunflowers in full bloom were cut at the Kansas Experiment 

 Station and " put into a silo between layers of corn silage." The material was 

 fed to dairy cows in place of corn silage, but they ate it sparingly, declined in 

 milk, and lost in weight. Part, at least, of the poor success is attributed 

 to the late cutting. 



Silage crops [at the Scottsbluff Substation] {Nebraska Sta. Rpt. 1919, p. 

 28). — It Is stated that a comparison between corn and sunflowers as silage 

 crops is under way. " The silage made from sunflowers is less desirable than 

 that made from corn. Dairy cows eat this silage fairly well, however, and their 

 milk flow seems to hold up nearly as well as on corn silage." 



