DEVELOPMENT OP PHRYGANIDS. 581 



proof that the invaginated ectoderm cells give rise either to 

 the long or the short commissures. 



Before the formation of the brain by the union of the 

 thickened procephalic lobes, a deep fold is formed on the lateral 

 surface of each lobe, which results in the formation of a thin 

 blade of ectoderm, which imperfectly divides the brain into an 

 anterior and a posterior portion. This fold is seen in surface 

 view in PI. XXXVIa, figs. 11 and 12, and a section of the same 

 in PI. XXXVIc, fig. 41. Although not indicated in fig. 41, 

 since the stage is too late, there is an actual infolding of the 

 ectoderm produced, which, according to Tichomerof (" Ueber 

 die Entwick. des Seidenwiirms,^^ ' Zool. Anzeiger/ ii, Jahr. No. 

 20 (Preliminary Notice) ), produces a part of the chitinous 

 internal support for the brain, while, on the other hand, 

 Hatschek claims that it forms a part of the ganglionic cells of 

 the brain. As an objection to Tichomerof s view, it can be 

 urged that this fold is fundamentally diff'erent from the thickened 

 rod-like ectodermic ingrowths which undoubtedly form a part 

 of the internal skull (PI. XXXVIb, fig. 16, a, and fig. 15, d). 

 These unfoldiugs, however, are exactly like those between the 

 post-oral ganglia, and hence lead to the conclusion that the brain 

 is formed of two pairs of coalesced ganglia, from the posterior 

 pair of which the antennaj are developed, the anterior pair 

 producing no appendages. The principal objection to this 

 view is that instead of having a single transverse fold as in the 

 post-oral segments, we have in reality two separate, lateral 

 iufoldings, one on each side. The lateral position, however, is 

 easily explained, since the procephalic lobes, originally strictly 

 ventral in position, have been bent in a dorsal direction, giving 

 them a lateral position as related to the germinal band. The 

 fact, however, that there are two folds is a real difficulty, and 

 can only be explained by assuming that the fold which origi- 

 nally extended along the whole ventral face of the brain has 

 disappeared in the middle line, thus leaving two separate and 

 lateral, although originally continuous, infoldings. We have, 

 however, no facts to warrant such a supposition. If the brain 

 in reality is made by the fusion of two pairs of ganglia, then 



VOL. XXIV. NEW SER, r r 



