
No. 427.] HAIRS OF LEPIDOPTEROUS LARVÆ. 569 
these cells from the true multipolar nerve cells of Crustacea, 
regarding them as many-branched mesenchymatous cells. In 
lepidopterous larvae Holmgren states that there are no mesen- 
chymatous cells like those in Crustacea which one would con- 
fuse with nerve cells. 
In 1897 and 1898 Duboscq criticises the work of many pre- 
vious workers upon multipolar subhypodermal nerve cells and 
states as his opinion that the bipolar nerve cell is the only true 
nerve cell in the hypodermis of arthropods, and that there is 
no subepithelial plexus. Although much of the criticism of a 
subepithelial plexus by Duboscq is very just, it seems to me 
that simply calling all such structures so described connective 


Methylen blue. 
tissue is hardly justified when we look over the works of 
numerous observers. It may in part be true, as Holmgren 
points out, that much which has been described, especially in 
Crustacea, may come under the head of mesenchyma, but do 
we not have multipolar nerve cells as well? 
The occurrence of evidently multipolar nerve cells together 
with bipolar nerve cells and fibers I found to be very constant 
in lepidopterous larvae (Fig. 6). The nucleus of these multi- 
polar cells was a very well marked, clear area occupying a large 
part of the center of the cell; in the center of this clear 
area was a darker staining portion, possibly the nucleolus. 
These cells with their many fine branches always take a paler 
blue stain than the other nervous structures in the skin. The 
very fine branches of these multipolar cells run out long 
