206 PHYSIOLOGICAL GENETICS 



conclusion that this is the case (Boveri, Wilson, Conklin, Morgan, 

 Goldschmidt, etc.). Though it would be difficult to contradict 

 such an assumption, the question ought to be answered whether 

 or not facts are known supporting such a self-evident assumption. 

 Actually, there are not many such facts. This is to be expected, 

 as only rare cases are imaginable in which the action of mutant 

 genes upon Buch early processes could be analyzed. But those 

 which are known justify such a generalization. There is the 

 work of Toyama (1909) and Tanaka (1924) on characters of 

 silkworm eggs. These are different colors of yolk, already exist- 

 ing in the unfertilized egg, and colors of the serosa, a membrane 

 formed by the blastoderm cells. These early embryonic 

 characters are determined by mutant genes. (For an inter- 

 pretation see the pigmentation hormone on page 184.) A 

 more direct relation to the pattern problem is provided by the 

 facts regarding right- and left-hand coiling in molluscs. It has 

 been knoVn for a long time that this may be a modification in 

 some cases and a hereditary trait in others. The genetic analysis 

 of such cases by Boycott et al. (1930), Sturtevant (1923), and 

 Crampton (1924) showed a Mendelian behavior w r ith the addition 

 that Fi behaved like the mother, F 2 like Fi in the classic case, and 

 F 3 like F 2 . This means, as first shown by Toyama, that a 

 character of the egg is involved that has already been deter- 

 mined before fertilization, so that the influence of the sperm 

 will be felt only in the eggs of the next generation. The embryo- 

 logical work of Kofoid (1895) and Conklin (1903) had actually 

 shown that the direction of coiling is already determined when 

 segmentation begins; i.e., it is laid down in a promorphological 

 structure of the egg. Here, then, certainly, one of the primary 

 processes of pattern formation is controlled by mutant genes. 

 Another example of this type is furnished by Tanaka's work, 

 already mentioned. Among the characters of early stages that 

 wen 1 studied there was one type of pigmentation of the blastoderm 

 which was inherited in the way just described (wrongly so-called 

 maternal inheritance), which proves determination before 

 fertilization. Blastoderm formation is a form of pattern forma- 

 tion of a special type in the insect egg, and the pigment thus 

 serves as a marker of the separation (stratification) of one 

 embryonic area controlled by mutant genes. A similar case has 

 also been reported by Sexton and Pantin (1927) for larval lipo- 

 chrome colors in Gammarus. 



