DEVELOPMENT OF PHRYGANIDS. 577 



protoplasm (PL XXXVIb, figs. 6 and 7), and still others which 

 have completely separated themselves from the mesoderm and lie 

 in the yolk close to the ventral plate ; in general appearance all 

 the cells just described are very similar to each other. Although 

 I am not ready to believe without further evidence that yolk 

 cells arise by proliferation from these segmented portions of 

 mesoderm, yet it is necessary to recognise the fact that the 

 appearances just described could be employed as arguments in 

 favour of such a view, A great many cells similar to those repre- 

 sented in PI. XXX VIb, fig. 6, were observed in various stages 

 of separation from the mesoderm. In the upper part of fig. 6, 

 PL XXXVIb, is represented a cell which is connected with the 

 mesoderm by a very long filament of protoplasm. Such a condi- 

 tion would not be realised in cells approaching the mesoderm, 

 but very likely would be if the cells were separating from the 

 mesoderm ; for in the latter case it is very likely that a filament 

 would continue to hold the elements a long time in union. 

 It may be urged, on the other hand, that the absence of any 

 indication of a spindle metamorphosis is, to a certain extent, an 

 objection to the supposition of a proliferation on the part of 

 the mesoderm cells ; this objection, however, appears less 

 important when it is remembered that no spindle metamor- 

 phosis could be observed in other parts of the embryo where 

 there was undoubtedly rapid cell division. 



Stage 5. — The fifth period extends from the segmentation 

 of the mesoderm to the revolution of the embryo in the egg. 

 It is the longest and most complicated period in the develop- 

 ment of the insect; at its close nearly all the organs of the 

 body are differentiated. 



No important change occurs in the general form of the 

 germinal baud up to the time of revolution. The procephalic 

 lobes increase in thickness and at the same time the yolk with- 

 draws from between them, allowing their inner faces to come in 

 contact and unite with each other to form the brain. The 

 germinal band continues to increase in length until its extremi- 

 ties almost touch each other, and even at that late period it is 

 very easy to distinguish in the living egg the points where 



