252 PHYSIOLOGICAL GENETICS 



scales, only those scales can be impregnated with it which are 

 in the proper stage. This means that the substance in question 

 i- deposited only in definite areas. (See the facts of scale 

 development, page 241.) This one model serves for the descrip- 

 tion of many other types of patterning in similar terms, using 

 velocities, thresholds, time of onset or end of action, and similar 

 processes of simple quantitative behavior as elements. This has 

 been worked out in Goldschmidt's book (1927c). 



But the decisive point remains: This or any other explanation 

 requires that a primary stratification occurs at a definite time 

 leading to areas of different constitution. Only then can begin 

 the automatic action of a system, as the one described. 1 These 

 primary areas in our case were the areas of different speeds of 

 differentiation (caused primarily, so it seems, by different rates 

 of cell division, see page 244). Goldschmidt assumed that this 

 process occurs as a consequence of the production of a substance 

 at a definite time, which acts upon the substratum, i.e., the young 

 wing, like the evocator substance in the formation of Liesegang 

 rings within a colloidal medium. This means that the genes 

 responsible for this primary pattern or stratification act just as all 

 genes do, producing a definite substance at a definite time; but 

 that "the conditions of the system," i.e., the physicochemical 

 structure of the Anlage and the laws of chemical equilibrium, 

 cause a reaction with this evocator substance which leads to an 

 automatic distribution of the products of reaction in a definite 

 arrangement or pattern, after the model of the formation of 

 Liesegang rings (see discussion, page 205). On the assumption 

 that the whole wing acts as a unit, there was no possibility 

 of making more specific assumptions regarding this phenomenon 

 of stratification as applied to the pattern of the butterfly wing. 

 The work of Henke and Kuehn (1929-1936) has led one step 

 further. They showed that in the formation of the primary 

 pattern, the following processes take part. 



1. An already present general subdivision of the wing area into 

 different fields, which are, however, not yet determined in their 

 later extent. In the terms that we used, this would mean that 

 there is not one single process of evocation for the whole wing 

 but a number of centers from which this process starts. (These 



1 This is, of course, the same situation that was discussed repeatedly in 

 regard to embryonic pattern. 



