456 F. G. HEATHCOTE. 
tion of their long axis; and, in fact, before they do begin to 
take part in the formation of the mesodermic keel, they undergo 
an alteration, which I shall describe. When first the endo- 
derm cells just mentioned begin to come together in the middle 
line near the ectoderm their appearance is somewhat peculiar ; 
their nucleus is small, round, and deeply stained ; their form is 
stellate and their outline very distinct. 
Processes pass from them to the ectoderm cells. This is 
shown in fig. 19, which is a transverse section through an ovum 
on the fourth day, taken in a plane such as to cut through the 
first beginning of the keel. When a fair number of these cells 
are assembled in the middle ventral line a change takes place 
in the cells of the ectoderm just outside them. The latter 
become more rounded, while their nuclei, instead of being long 
and oval, become round. In fact they undergo an alteration 
which causes them to resemble the cells which I have described 
as assembling immediately below them. This alteration is 
shown in figs. 19 and 20, which are transverse sections through 
the first beginning of the keel. 
The ectoderm cells in the middle line, after altering their 
shape as I have described, increase by division, and take a con- 
siderable share in the formation of the keel. The cells in the 
middle line, both ectoderm and endoderm, continue to increase, 
and are joined by more cells from the hypoderm, and eventually 
on the fifth day we find a keel in the middle ventral line, 
something like that described by Balfour in his paper on the 
development of Agelena labyrinthica (16). Both ecto- 
derm and endoderm have taken part in the formation of the 
keel. 
When the keel is fully formed the cells of which it is com- 
posed are large, somewhat irregular in shape, and have a large 
nucleus. They are all directly connected together, though, 
owing to their being closely packed together, it is difficult to 
see anything of their connections, except where one cell has 
been somewhat separated from the others. The keel is of con- 
siderable thickness, being about six or more cells deep in its 
thickest part. 
