﻿142 



William S. Marshall and Paul H. Dernehl, 



(Fig. 36) the appearance of a number of dark granules either sep- 

 arated, or mucli oftener appearing partially fused, and then taking 

 the form of a plate, not straight, but of a wavy appearance. At 

 either side of this traces of one or more vacuoles were seen giving 

 the plate the appearance of having been enclosed in a vacuole, or 

 vacuoles, which had nearly disappeared. Despite the fact that the 

 boundaries between the blastoderm cells have been described as 

 cntting in between the nuclei from the outside, and as appearing in 

 a similar way in Polistes, we have here to do with the formation 

 of boundaries, which, separating the nuclei from each other, give 

 rise to at least some of the blastoderm cells. In what order the 

 figures we give occur we could not say, it appearing to us probable 

 that the vacuoles are first to appear, and they are followed by the 

 granules and the cell boundaries. We call attention to the fact that 

 the boundaries are in the Keimhautblastem most of them formed 

 between resting nuclei, Caenoy (12) has figured the formation of 

 cell boundaries between resting nuclei in a Hymenopteron, Bombus, 

 noting rows of granules, very similar to what we find by Polistes, 

 but without the appearance of the vacuoles which we have described. 



The undulating outline of the egg has already been noted and 

 explained by the withdrawal of the Keimhautblastem from the egg 

 membranes at certain points. This appearance does not persist for 

 any length of time, the outer margin becoming again eveü and a 

 layer of nearly equal thickness is formed over the surface of the 

 egg. Cell boundaries are formed between the nuclei in this layer, 

 and the egg becomes covered by a Single layer of cells, the blasto- 

 derm. These cells are at first very much alike, differing in a slight 

 Variation in size and shape. In reality such a stage is probably 

 never present in Polistes, or, if present, has an exceedingly short 

 duration. The reason for this is found in the fact that the cleavage 

 nuclei reach the anterior surface before the posterior, and also that 

 the blastoderm formation does not go on similarly over the entire 

 surface of the egg. 



Most of the earlier students of insect embryology, in fact until 

 the last few years, described the division of the blastoderm cells 

 as occurring amitotically ; this has, however, been changed, mitosis 

 having been observed in the blastoderm cells of a number of diffe- 

 rent insects. In Polistes a mitotic division is the only one found, 

 different stages of which are easily seen and have been observed 

 from the different parts of the egg's surface. We have figured a 



