582 Wra. S. Marshall, 



ends was an exceptional occurence. This surface view was taken from 

 one side of tlie mesothorax of a larva 10 mm long, the rudiment of 

 the other side was sectioned transversely and is represented in figure 11^. 

 The entire evaginated part is nearly circular, the peripodial pore is 

 elongated and not centrally situated but considerably nearer the dorsal 

 than the ventral surface. The darker outer ring that one sees in the 

 surface view is due to looking through a greater thickness of cells re- 

 presented by line a — a in figure WA than at any other part. The 

 relative distance between the boundary of the peripodial pore and 

 the outer margin of the disk, both dorsally and ventrally, is best seen 

 by a comparison in figure WA, of the lines h — h, with c — c, ventral. 

 The anterior and the posterior ends of the peripodial pore are also 

 unequally distant from the outer margin of the disk. 



From the time when the disk is well formed (Fig. 5), and espe- 

 cially at a little later stage, the nuclei nearly all lie nearer its inner 

 than its outer surface; a similar condition is present in the disks during 

 invagination. In the different stages mitotic figures are quite abun- 

 dant; they are seldoni found in the region occupied by the other 

 nuclei, but well out towards the outer surface of the disk (Fig. 13). 

 It is probable that the crowding of the nuclei into the inner portion 

 of the disk does not leave room here for the mitotic figures and that 

 dividing nuclei wander out into the less crowded outer region 



Regarding the differences, if any, between the wing rudiments 

 in the two wing-bearino; segments nothino- but slight variations were 

 found and these were not constant in the majority of the larvae 

 examined. Specimens of larvae both at the beginning and at the 

 completion of invagination were taken, the disks measured, and those 

 of the mesothorax compared with the measurements of those of the 

 metathorax. Very little difference in size was found in any of these 

 specimens and what Variation in this respect could be noted was not 

 constant; from this it was inferred that in neither the one nor the 

 other Segment was the disk the larger. Ten larvae of another group 

 in which the invagination had begun were taken and the wing rudi- 

 ments sectioned; these sections were drawn with a camera lucida 

 and compared with each other. From this series it was found that 

 no constant difference between the rudiment of the meso- and meta- 

 thoracic Segments exi&ted. 



Jn no specimen does the disk sink far below the surface and the 

 depth of the peripodial cavity is seldom found to be as great as the 

 thickness of the disk. Evas-ination of the disk now commences the 



I 



'1 



