256 PHYSIOLOGICAL GENETICS 



today, we have repeatedly passed over points tluit would have 

 required more consideration. We propose to discuss these now in 

 the form of additions. 



The decisive point, the analysis of which is still more or less 

 in the theoretical stage, is the formation of those primary patterns 

 which have been described as the distribution of the field centers 

 or points of outlet. This problem is, of course, identical with 

 the problem of embryonic stratification, as mentioned before. 

 We have often used the comparison with the production of 

 Liesegang rings in colloidal solutions after introduction of some 

 chemical, acting as an evocator. Gebhardt (1912) was the first 

 to draw attention to the great similarity of some patterns on 

 butterfly wings with Liesegang rings, and he believed that pig- 

 ment deposition in the scales was actually such a process. 

 Kuester (1913) applied this same principle to numerous rhythmi- 

 cal structures in plants. In its original form, the idea had to be 

 discarded as applied to the butterfly wing, when it became known 

 that the pattern was determined independently of pigment 

 formation (Goldschmidt, 1920d), though it was still possible to 

 make use of the same idea in explaining the arrangement of 

 different determining stuffs (growth substances) in the form of a 

 rhythmic pattern or stratification. As we have seen, this is 

 probably not the case for the primary patterns of the butterfly 

 wing, though certain secondary patterns within the fields or 

 spreading over the whole wing may be explained thus. (Simi- 

 larly with many rhythmic developmental patterns, like segmenta- 

 tion.) A few additional facts regarding this phenomenon may 

 therefore be mentioned. In detail, it may be produced by differ- 

 ent processes. A substance present in the system may spread 

 by diffusion and be arranged in a rhythmic way; or the phases 

 of the rhythm may be produced only as a reaction in loco; or a 

 precipitate may be formed in a rhythmic procession. We may 

 assume that one or the other of these phenomena occurs when 

 rhythmical patterns are involved, and we may accept this as a 

 general model for developmental, simultaneous stratification 

 in cases where no other related process of diffusion is known. 

 Thus far, no actual genetic or experimental fact is known to 

 show a Liesegang phenomenon at work, though some pat- 

 terns apparently require it. Quite recently, however, Becker 

 (1937) showed that a special pattern formed upon the tergites 



