GENERAL CONCLUSION. 119 



indicates that Cavia cutleri possesses the same agouti factor as tame 

 agouti guinea-pigs. Light agouti of Cavia cutleri and dark agouti of 

 Cavia rufescens are thus variations in a character in two wild species 

 which differ in heredity by a clear-cut Mendelian factor. 1 



7. There is a continuous series of variations between smooth fur 

 and very rough or resetted fur in guinea-pigs. The primary effects in 

 this series are due to two independent pairs of allelomorphs. One 

 factor, discovered by Castle (1905), is essential to any roughness of the 

 common type, and is completely dominant over its allelomorph found in 

 wild cavies and smooth guinea-pigs; the other, an incomplete recessive 

 to its allelomorph in the wild cavies and some tame guinea-pigs, is 

 necessary for the higher grades of roughness. Second-order effects 

 seem to be due to the residual heredity of different stocks, and probably 

 to non-hereditary irregularities hi development. There is a roughness 

 of a different type from the usual which is inherited independently. 



GENERAL CONCLUSION. 



Most of the successful earlier attempts at Mendelian analysis of 

 heredity naturally dealt with variations which were obviously dis- 

 continuous. But in nature such variations are much less common than 

 apparently continuous series of variations. It was thus a common 

 reproach against the Mendelian analysis that it dealt only with excep- 

 tional conditions. The work of Nilsson-Ehle, East, and others has 

 shown how quantitative variation may be brought under a Mendelian 

 explanation. MacDowell (1914) presents data on size inheritance 

 from this standpoint and discusses the literature up to that time. 

 Recently two very interesting papers have been published (Dexter, 1914, 

 Hoge, 1915) which analyze the heredity of certain very variable char- 

 acters in Drosophila by means of linkage relations. 



Several of the studies in this paper deal with inheritance in continu- 

 ous series of variations. The only general statement which can be 

 made about the results is that there is no general rule for such cases. 

 Intermediates between varieties which mendelize regularly have been 

 found to follow very definite modes of inheritance, which, however, are 

 very different in different cases and could not possibly be predicted 

 a priori. On the other hand, each mode of inheritance is exactly 

 paralleled by cases among the most diverse groups of animals and plants. 

 It may be interesting to summarize the modes of inheritance of inter- 

 mediates which have been found. 



An intermediate condition is sometimes found to be due to an inter- 

 mediate variation of the essential hereditary factor involved, i. e., to 

 an allelomorph. Thus yellows are intermediate between red and albino 



1 It should be pointed out, however, that the original stock of Cavia rufescens used in these experi- 

 ments included individuals of the light-agouti character as well as those classed as dark agouti. 

 It seems quite likely that dark agouti arose as a recessive mutation in C. rufescens. W. E. C. 



