INTKWNAL ANATOMY OF SQUILLA. 
419 
with the end-sacs of most antennal and maxillary glands, 
since they are figured by Grobben ( 10 )^ Claus ( 6 ), 
Vejdovsky (22), and many other authors. Finally, it will be 
noticed that these end-sac cells contain none of the vacuoles 
so characteristic of the cells composing the fully formed end- 
sac. 
As regards the squamous epithelial investment of the gland 
at this stage of development it will be remembered that in 
the adult Squilla the whole of the gland, excepting the lower 
part of the bladder, is enclosed in a double layer of 
squamous epithelium, the two layers enclosing a blood-space. 
On the other hand, in the early stage of development of the 
gland depicted in Text-fig. 1 no epithelial investment what- 
ever can be observed. Text- fig. 3 shows what may perhaps 
be regarded as an interniediate condition. As will be seen, the 
gland at this stage lies in close contact with the side of the 
body, and is only separated from the hypodermis by a. 
(apparently) single layer of squamous epithelium {8q. Ej:).), 
which is also to be found on the inner side of the gland, and 
penetrates between the closely apposed walls of the end-sac 
and the gland proper. This single layer of squamous epithe- 
lium also penetrates into the invaginations of the wall, the 
presence of which distinguishes the gland of the 14 mm. larva 
from that of the 12 mm. In this 14 mm. larva, tlierefore, we 
can observe the commencement of the formation of those 
involutions of the wall of the gland proper (two are shown in 
Text-fig. 3 — I.M. and I.P.), which ultimately so invade the 
cavity of this region as to reduce it to the winding system of 
narrow cracks already described as characteristic of the fully 
formed organ. Since the single layer of investing squamous 
epithelium is drawn into each of these involutions of the wall, 
it must therefore represent the inner layer of the two layers 
characteristic of the adult condition. The outer layer of 
squamous cells of the adult gland is possibly produced later by 
the splitting (delamination) of the inner (I admit that the 
only evidence I have for this suggestion is the occasional 
appearance, in sections, of a double condition to be found 
