202 EXPEKIMENT STATION RECORD, 



variation is not likely to be found either in maternal impressions, the 

 direct action of environment, use-inheritance, or telegon3\ Instead 

 of simph' stating- that variation is due to the constant recurrence of 

 slight inequalities of the nutrition of the germ cells, he believes that 

 we may assert with some confidence that differences in age, vigor, and 

 health of the parents and differences in the ripeness of the germ cells 

 are potent causes of variation. He holds that intercrossing, though a 

 direct cause of retrogressive variation, is only an indirect cause of 

 progressive variation, while in-and-in breeding at the right moment is 

 a cause of progressive variation. He discusses at some length the 

 "swamping" effects of intercrossing, showing that "progress in a 

 single direction is probabl}" often due to new varieties SAvamping old — 

 it may be long established — varieties, and that several varieties may 

 be sufficienth^ exclusive to flourish side by side in the same area and 

 eventuallv (partly owing to their aloofness, i. e., to differential mating) 

 give rise to several species." 



In conclusion Dr. Ewart states his belief that the time has come 

 when a well-equipped institute for biological and other experiments 

 bearing upon these questions should be provided. 



The need of such experiments must be patent to all who are familiar 

 with the theories and traditions of stock breeding. Considering the 

 very great progress which has been made in recent years in agricultural 

 science, it is rather remarkable that our knowledge of the laws upon 

 which the breeding of animals for different purposes rests has remained 

 practically at a standstill. While attention has been given to nearl}^ 

 every other line of animal husbandrj^, this interesting and important 

 field has been passed over, probal^ly on account of the difficulties and 

 the time-consuming nature of such experiments. These, however, 

 should not present themselves as barriers to institutions whose per- 

 manence is assured and' which can safely la}^ plans for a long term of 

 years. Stock breeding suggests itself as a most inviting field for 

 experiment station work, and one in which very important, if not 

 epoch making, results are quite within the range of possibility. 



In this day of specializing, the field would seem to be sufliciently 

 inviting to attract men who should prepare themselves for it by the 

 very Ijest means possible, and be content to make the subject a life 

 study. What has long been needed is a master mind to work over and 

 digest the theories and traditions on which our so-called principles of 

 breeding rest. The}' rest too largely at present on experience, rather 

 than on systematic experiments, which affords instances in support of 

 almost an}' theory. The information needs to be classified and graded, 

 and the wheat separated from the chaff' that has been passed down to 

 us through successive manuals on the subject. It would seem that 

 some few of the experiment stations in this country might be in posi- 

 tion to enter upon some phase of this work, under the advice of such 



