No. 639] INHERIT AXCK OF BLUE IN FOVLTRY 



297 



lualiiius and ilu- itroiXT ideiitilicatioii of the cuus laid 

 each iViiiale In <.arli matin- Not only wa> the a^^iManl 

 in cliar^'o of the 1 ivipncstiim- selected because of liis liahit- 

 ual accnracy in details, Imi ilie e^i^j^s from each individual 

 lien were ke])t tou.M lier, x pai'ate from the egj^s of otlier 

 individuals, and eai'i'TuilN- compared one with anotli^r 

 before bein^ put into the incubator. \\\\ otV lype di' otT 

 colored eggs were discarded, so fai- as these exiicrinicni - 

 are concerned. In sjVite of tliese precautions it is ton 

 much to hope tliat sonu' eri-(u-s have not crept in, thouu'h 

 it is believed they ;irr Vel'V few. 



Owing to the I'act that the original stock was of rela- 

 tively unknown composition, it was necessary to make 

 such matings as would not only throw light on the 

 behavior of the factors under observation, but would also 

 be likely to bring to liirlit unsuspected factors whose 

 action might interefere with the action of the i;-enes 

 being studied. This neces.-itaird int rudnein-' lot females 

 in the matings where the males were nneeiinin. and of 

 mating many of the females with test males a second 

 season instead of repealinu' the matin- already made. 

 In both cases the result was to i-ednee -i-eatly the numbers 

 of offspring from some of the crucial tyi)es of matings, 

 and con^derable numbers of "te-t" (.ff^piin- ^^ere 

 hatche.l .•iiid de>ei'ibed, for the reporting of whieh here 

 there is no particular ol)ject. 



The count. hviie- cliick. Were mad, ' , ^n^^n at 

 hatchin- time ; ■ ■ -I'. . d, each 



