252 



Research in Animal Breeding. 



[JXJNE, 



RESEARCH IN ANIMAL BREEDING. 



III. 



E. C. PUNNETT, F.R.S. 



Professor of Genetics, University of Cambridge. 



In the previous articles of this series, -puhlished in the 

 April and May issues of the Journal, Prof. Punnett dealt 

 with the coat colours in cattle, and the crossing of polled 

 with horned cattle as illustrations of simple Mendelia-n 

 inheritance. 



The factorial hypothesis of heredity is, if substantiated, of 

 fimdamentai importance to the breeder, for it at once raises his 

 operations from an empirical to a scientific plane. It brings 

 certainty where before was only conjecture. Consequently, when 

 animal breeding experiments were started on the University 

 Farm at Cambridge in 1910, it was felt that among the first things 

 to do was to choose one or two cases of apparent blending inherit- 

 ance, and to study them critically in order to ascertain whether 

 they could be interpreted on the factorial hypothesis. The choice 

 of material was limited to small animals, for reasions of economy. 

 This, hovrever, Vv^as no drawback, for small animals can be bred 

 in reasonably large numbers ; and we can hardly doubt that 

 what we learn from them is applicable to bulkier and more costly 

 stock. Our work has, therefore, been entirely with poultry and 

 rabbits. 



One of the most extensive series of experiments undertaken 

 with poultry was designed to investigate the inheritance of 

 weight.' For this purpose two standard breeds were chosen, differ- 

 ing markedly in size, but not so much so as to prevent natural 

 crossing. For the larger breed we selected the Gold Pencilled 

 Hamburgh, and for the smaller one the Silver Sebright Bantam 

 (Plate 1, Fig. 1). As will appear later, the reason for choosing 

 these particular breeds was to make use of the same material for 

 the elucidation of more than one problem. From the point 

 of view of size the two breeds differed sufficiently, for the average 

 weights of cocks and hens respectively were for the Hamburgh 

 about 1,400 and 1.100 grammes, while for the Sebright they 

 were about 850 and 650 grammes, Eoughly the Sebrights were 

 about 3/5ths of the v>^eight of the Hamburghs. 



The first cross birds were intermediate in size, though approxi- 

 mating to the larger breed, the cockerels averaging about 1,200 

 grammes, and the pullets about 950 grammes. From several 



