580 Wm. S. Marshall, 



the number of its nuclei due to its enlargement and the formation 

 of many new cells. Mitotic figures now become fairly abundant and 

 are found distributed throughout all parts of tbe disk. 



The disk continues to invaginate, increasing in size as it sinks 

 below the surface; the cytoplasm shows a marked striation and, in 

 many specimens, vacuoles are present. The nuclei erowd together in 

 the basal, inner, part of the disk and become so numerous that many 

 of them are pushed out towards the other, outer, surface giving the 

 appearance of two or three irregulär layers. That chitin is being se- 

 creted is easily seen when, as often happens in the sections, the cuticula 

 is pulled away from the disk; in such sections clear, newly secreted 

 chitin can be seen connecting the disk with the adjacent surface of 

 the cuticula (Figs. 8 and 13). 



Coincident with the removal of the disk from the surface comes 

 an increase in its size so that its diameter is soon greater than of the 

 open Space it has left in sinking away from the cuticula. This open- 

 ing, from a surface view (Fig. 9), is seen to be surrounded by a darker 

 ring due to the fact that where the hypodermis turns under to connect 

 with the disk (Fig. 9^) one must, in surface view, look through two 

 or three layers of cells. A comparison of these two figures, 9 and 9Ä, 

 will show this relationship — these two figures are the wing rudiments 

 from the mesothoracic segment of a larva 10 mm long — the wing 

 rudiment from one side was mounted entire (Fig. 9), and that from 

 the other side was sectioned (Fig. 9^). This invagination leaves the 

 disk below the surrounding hypodermis, the space thus left is the 

 peripodial cavity and the opening into it, now nearly as large as the 

 cavity itself, the peripodial pore. Surrounding the peripodial pore 

 is the turned over margin of the hypodermis, the lip. As the disk 

 enlarges the circular margin of the lip comes closer together thus de- 

 creasing the diameter of the peripodial pore and the irregulär growth 

 of the disk gives to it a more elongated shape (Figs. 10 and 11). At 

 the very beginning of invagination the outer surface of the disk was 

 slightly concave and the inner surface somewhat convex or flattened; 

 as the disk sinks below the surface this changes and we soon find the 

 outer surface convex and the inner surface either convex or flattened 

 (Fig. 10^). The nearly circular outline of the disk soon changes to 

 a more oval shape, when this happens the longitudinal axis of the 

 disk does not lie parallel to that of the body of the larva but the posterior 

 end~of the oval wing rudiment is tilted upward and its longitudinal 

 axis becomes oblique (Text figure I). As far as we have been able 



