150 EMBRYOLOGY OF INSECTS AND MYRIAPODS 



normal embryos. Duplications that result from splits induced by 

 cautery also show that the prospective potency of the egg parts is greater 

 than the prospective significance at the time the various anlagen are set 

 apart; each duplicated part is larger than one-half the size of the same 

 organ in a normal embryo, so that the sum of the two duplicated parts 

 is greater than the size of a single normal organ. This power of duplica- 

 tion is possessed by all organs, external and internal. It occurs never 

 in the longitudinal axis, seldom in the dorsoventral axis, but commonly 

 in the transverse axis. The interpretation given is that determination 

 occurs in these three axes in this same order. In Tachycines Krause 

 (1934) reports a correlation between the type of mechanical injury to 

 the germ band and the effect. Frontal fissures give duplications in the 

 dorsoventral axis; median and oblique fissures of various extents and 

 positions give symmetrical or asymmetrical, transverse anterior or pos- 

 terior duplications or parallel twins (division of differentiation center?); 

 and fine splits give duplications of single organs. Effects produced by 

 operating at different stages show that determination of the main axes is 

 followed by determination of segments in the longitudinal axis, then of 

 the lateral halves A\dthin these segments, and finally of individual organs. 

 Oka (1934) reports preliminary experiments showing only that the egg 

 of the cricket is of the regulative type. 



Incompletely determinate eggs are intermediate between the two 

 extreme types. They are capable of considerable regulation following 

 experimental interference in early stages, but the power is lost much 

 sooner than in indeterminate eggs. Dwarf formation has been observed, 

 but not duplication of parts. The absence of duplications indicates that 

 strict determination of the presumptive anlagen occurs early. 



In the Hj^menoptera Schnetter (19346) reports that constricting the 

 honeybee egg at the stage of the "uniform blastoderm" (12-hour embryo) 

 can result in a normal dwarf embryo. Later blastoderm stages (24-hour 

 embryo) are less labile and more inclined to mosaic formation with a 

 shift in the regions of different prospective significance (Fig. 60). In the 

 ant Camponotus, Reith (19316, 19326) shows by cauterization and con- 

 striction that the egg possesses considerable regulative power until the 

 completion of the visible zonation bj^ differential thickening of the 

 cortical layer (presumably under influence of the activation center). 

 After that it is determinate though not so highly so as dipterous eggs. 

 By centrifuging he shows that stratification of the egg parts prior to the 

 cortical zonation blocks development but that later centrifuging does 

 not prevent formation of a fairly normal germ band. These two sets 

 of data show that within this order there is a difference in the time of 

 determination, that of the ant occurring noticeably earlier than that of 

 the hone3^bee. DuBois (1938) has found that Sialis lutaria, a member 



