R. C. VON BORSTEL 189 



laid the groundwork for experimental study of insects; their 

 methods and interpretations have been used on examples from 

 a variety of insect orders. Differentiation in insects has been con- 

 sidered by Bodenstein ( 1955 ) in a short but comprehensive re- 

 view, and no attempt wdll be made here to reevaluate the tre- 

 mendous amount of work that has been done in the field of insect 

 morphogenesis. Suffice it to say that in general the Diptera are 

 considered as being completely determined embryologically and 

 Hymenoptera partially so. Of insects that have been studied ge- 

 netically, Drosophila embryos have been fairly well analyzed by 

 the methods available (Geigy, 1932; Howland and Child, 1935; 

 Rowland and Sonnenblick, 1936; Howland, 1941); Habrobracon 

 embryos have not. The available experimental data on Habro- 

 bracon embryos indicate that development is largely predeter- 

 mined. The few eggs that have been tied in half ha\'e not hatched, 

 and ultraviolet-irradiated eggs ( with the nucleus shielded ) raised 

 to imago have had distorted tergites which were not the result of 

 gene mutations. Bodenstein thoroughly considers the characteris- 

 tics of insect eggs with determinative development. These have 

 the various structures of the future embryo localized in their pe- 

 ripheral cytoplasm. He points out that experiments on determinate 

 eggs show quite conclusively that peripheral cytoplasm is a dif- 

 ferentiated continuum in which localized differentials exert spe- 

 cific influences on totipotent cleavage nuclei, leading them 

 toward special assignments. "One has to assume that the cyto- 

 plasmic regions possess theii^ specific qualities only when in 

 normal topographic relationship to the cortical cytoplasmic layers 

 as a whole. Their influence must be regarded as of a general di- 

 rective nature in that they set up differentials in cleavage nuclei, 

 thereby creating a different pattern within the framework of the 

 blastoderm, which forms the basis of the ensuing de\'elopmental 

 events." That is to say, if any organ-forming region of cytoplasm 

 on the egg surface is injured, no amount of totipotent nuclei com- 

 ing into that region can repair or redirect the cytoplasm. The 

 information for development is locked in the peripheral cyto- 

 plasm, and invading nuclei are guided in their destiny by this 

 cytoplasm. The work of Brauer and Taylor ( 1936 ) is of especial 



