180 PHYSIOLOGICAL GENETICS 



however, speak of the Bame processes usually as being of a cyto- 

 plasmic type. This is not correct, since all the facts indicate that 

 the same typo of processes may occur either later in development 

 or earlier and even in the undeveloped egg cell (see page 206). 

 This misleading description becomes dangerous when facts are 

 analyzed to show the part played by the maternal cytoplasm in 

 early development as compared with the influence of the genes con- 

 tributed by the sperm. If it is found that certain developmental 

 processes occur under complete control of the egg, this does not 

 mean that these processes are not under control of genes. It 

 means actually that certain genes present in the egg nucleus con- 

 trol the pattern formation in the egg and in the following stages. 

 In a hybrid egg, therefore, such a pattern ought to be of a hybrid 

 type, if the parents had different patterns. But this experiment 

 has thus far not been performed. As a type of analysis of the first 

 generation, we mention Hamburger's recent work on species 

 hybrids in amphibia (1936). He crossed Triturus crislatus, T. 

 taeniatus, and T. pahnatus and followed in detail the development 

 of the hybrids. In this case, the parental forms are different only 

 in the later embryonic stages, the tail-bud and early larval stages. 

 Up to a somewhat later stage, the limb bud stage, the embryos 

 are absolutely maternal in reciprocal crosses, and only later the 

 influence of the father becomes visible in influencing growth and 

 pigmentation. This shows, then, that a certain amount of 

 pattern formation before as well as after fertilization occurs under 

 the exclusive influence of the egg-nucleus genes. In terms of 

 intracellular gene action, it would mean that there is no difference 

 between the diffusion through the cell of substances produced 

 by action of the genes (with consequent arrangement as in a 

 pattern) and the diffusion of these substances from cell to cell. 

 Embryonic effects of genes and other effects of the usual type 

 are hardly different in principle (see also page 205). 



We may point out finally that diffusion of gene-controlled 

 products from cell to cell may become visible in quite a number of 

 cases in which these products have only a general action. Baltzer 

 and de Roche (1936) discussed this problem in connection with 

 their experiments on growth of merogonic tissue upbn a different 

 host (see page 268). Here the otherwise impaired vitality of the 

 merogonic cells is reconstituted, and this might be interpreted 

 on a genie basis as a check to lethality by diffusion products 



