318 Annals Entomological Society of America  [Vol. VII, 
I have described the appearance of the epithelial cells in one 
beetle larva, Dendroctonus. Another larva which I have 
sectioned, that of a Carabid, is entirely different. The cells” 
are arranged in lobe-like groups with a nidus (n) at the base 
between each two groups, from each side of which the cells 
arise and gradually grow and migrate until they become full- 
sized, when the contents is discharged and others replace them. 
The larger cells are extremely vacuolate, and irregularly granu- 
lar basally and distally. The nuclei are fairly large and deeply 
staining. 
The Diptera have been studied by various investigators. 
Van Gehuchten, in his complete work on Ptychoptera, a Tipulid, 
was one of the first to point out the method of digestive secre- 
tion in insects. Haseman’s recent paper on Psychoda describes 
the conditions occurring in another group, with habits not un- 
like the Tipulids. It is apparent that in larve such as these 
which live practically submerged in their food, the merocrine 
type of secretion prevails, and the arrangement of the cells 
secreting in this manner is manifestly entirely different from 
that representing the holocrine type. This latter, which we 
have seen in the Carabid larva, and elsewhere, demands cells 
capable of storing up the digestive fluids until such time as 
food may be taken into the canal, for predatory insects neces- 
sarily get their food irregularly. 
The silkworm is distinctly a continuous feeder. Hence we 
should expect to find no nidi or nuclei present. At first glance 
it seems otherwise (Fig. 7) but a closer scrutiny reveals the fact 
that these apparent groups of nuclei are quite different. In the 
first place there are fewer nuclei in a group than is usual, and 
then they are strictly not groups of mere nuclei, but groups of 
small cells. When we realize, too, that this particular insect 
was just upon the point of molting, we conclude, as I have pre- 
viously shown, that these are the new cells which form to re- 
place the old ones sloughed off at the molting period. 
But let us examine a just-hatched larva, which has taken no 
food except the portions of the egg-shell devoured in hatching. 
The cells are exceedingly regular, and none of the small basal 
cells to be observed. Most of the cells are deeply staining, 
granular, with an elongate, central, granular nucleus, and distal- 
ly containing a few small vacuoles. Frequently another type oc- 
curs, lighter, more homogeneous, with basal, rounded nuclei. 
