102 Experimental Evolution, Variation, and Heredity 



to be permanent. Upon crossing the selected races with each other an intermedi- 

 ate condition is obtained which in the following generation shows increased vari- 

 ability about the same intermediate condition. Crossing the plus selected race 

 with wild rats renders it less plus when extracted in the second filial generation. 

 Crossing the minus selected races renders it less minus. These seemingly contra- 

 dictory results and other results described in this paper favor a hypothesis that the 

 immediate effect of selection is registered in the germ cell not along with the 

 chief determiner of the hooded pattern but independently of it. But ultimately, 

 and as an indirect result of selection the accumulated effects of the selection be- 

 come in some way incorporated in the same single unit when a simple Mendelian 

 behavior is observed. On this view mass-selection is an effective means of guiding 

 evolution not only by the cumulation of small variations but also by inducing "mu- 

 tation.'' These conclusions are based on the study of more than 25,000 rats bred 

 in the course of seven years. The observations are summarized in some fifty cor- 

 relation tables in which parents and offspring are compared as to the extent of 

 their pigmentation. 



No. 196. (Paper No. 22, Station for Experimental Evolution.) 

 MACDOWELL, E. C., and W. E. CASTLE. Size Inheritance in Rabbits. Octavo, 55 

 pages, 9 figures. Published 1914. Price $0.50. 



In the historical part of this paper a review is made of all accessible previous 

 observations on size inheritance in animals and plants with a view to ascertain 

 whether evidence of Mendelian segregation as regards size characters can be de- 

 tected. In the experimental part is described an extensive series of experiments 

 with rabbits in which crosses were made between small size and large size rab- 

 bits. The immediate offspring were in nearly all cases found to be of intermediate 

 size, as were also offspring of a second filial generation (F ), but this second 

 generation was in most cases found to be more variable than the preceding one, 

 which fact the author regards as evidence of the presence of size-determiners 

 which mendelize in crosses but without the occurrence of dominance. In an ap- 

 pendix, it is shown that a high degree of correlation exists within the body be- 

 tween one skeletal measurement and another, and the fact is pointed out that if 

 size-determiners exist they are general in character and affect all parts of the 

 skeleton simultaneously. 



No. 205. (Paper No. 23, Station for Experimental Evolution.) 

 DETLEFSEN, T. A. (with prefatory note by W. E. CASTLE). Genetic Studies on a 

 Cavy Species Cross. Octavo, 134 pages, 10 plates, 2 text figures. 

 Published 1914. Price $2.00. 



Cavia rufescens, a distinct species of wild Brazilian guinea-pig, was crossed with 

 C. porcellus, the tame domestic guinea-pig. The female hybrids were fertile, but 

 the male hybrids were sterile. The female hybrids were crossed back to males of 

 both parent species, the cross with guinea-pig males being more successful. By 

 eight successive back crosses to guinea-pig males, more and more dilute wild- 

 blooded generations were obtained. Over 1,800 animals of the various hybrid gen- 

 erations were thus produced. Part I, dealing with color and coat characters, shows 

 that they are transmitted in accordance with Mendel's law. Any combination of 

 characters possible in the guinea-pig is possible in the hybrids. Although acting as 

 units in transmission several characters were modified in the cross. Part II dis- 

 cusses growth, and morphological characters in the parent species and hybrids. The 

 guinea-pig is apparently dominant in size but segregation was not apparent. The 

 M-shaped nasal-frontal suture of the wild was dominant to the truncate suture of 

 the tame. The later hybrids showed a great range of variation. Morphological 

 anomalies occurred in the hybrids. Part III deals with fertility of the parent 

 species and hybrids. The peculiar sterility in the male hybrids seems to be due to 

 multiple factors (possibly about 8) the elimination of which gives recessive fertile 

 males. The sex ratio in the early hybrid generations showed a marked excess of 

 females. The wild females have a smaller litter-average. The hybrid females are 

 intermediate but back-crosses to guinea-pig males gradually raised this average. 



