210 PHYSIOLOGICAL GENETICS 



It is only a short step from a growth pattern to a pattern of an 

 area type, as the work Oil certain patterns in insects has shown. 

 Long ago. GoldschmicH (19206, d) pointed out that the insect 

 wing and especially the wing of Lepidoptera might serve as a 

 model because rather complicated patterns of form, structure, 

 and color here arc found in a more or less bidimensional arrange- 

 ment, and because this pattern is accessible to genetical as well 

 as embryologies.! experimentation. The first steps in this 

 direction were published in 1920c/; more material was added later; 

 and a comprehensive theoretical treatment was given in 1927c. 

 Recently this problem has been attacked mainly by Henke, 

 Kuehn, and their students, and the body of facts now available 

 furnishes considerable information. 



These two main types of pattern formation have some addi- 

 tional special aspects. There are patterns of primary importance 

 for the progress of development, e.g., the pattern that is laid down 

 in the formation of an insect wing with the proper arrangement 

 of veins, etc. But there are also secondary patterns which are 

 superimposed upon the primary patterns late in development, 

 e.g., melanism, spread of dark pigment, upon an otherwise 

 completely patterned wing. There is, moreover, the problem 

 of symmetrical arrangement of the patterns on both sides of the 

 body, a problem that might furnish decisive information upon 

 the pattern problem in general. The following chapters will 

 deal with these and other aspects of the pattern problem as 

 related to gene action. 



1. Pattern of Growth and Form. If we refrain from a detailed 

 discussion of differential or heterogonic growth (which is found 

 in the book of Huxley, 1932), we may state in a few words the 

 points that are essential from the standpoint of physiological 

 genetics. Very different and specific forms and shapes in two or 

 three dimensions may be produced if the process of growth in 

 different directions occurs according to a definite rule, containing 

 variables the values of which produce a definite result. Huxley 

 found an exponential formula that accounts for all cases of 

 differential growth and contains two decisive constants. The 

 value of these constants determines the numerical system that 

 the differential growth will follow in each case. Whatever the 

 specific merits of this formula may be (see discussion by Hersh 

 and Feldman, 1936), it furnishes, if considered in a general way, 



