1893.] Embryology. 747 
it from the thicker disk of cells, the inner or true indusium, lying upon 
the yolk. The indusium is then a double body lying anterior to the 
embryo and free. So it is not carried away when the embryo sinks 
into the yolk but remains and grows in all directions as a double layer 
between the serosa and the yolk. This double sac externally meets 
itself on the dorsal mid line of the egg and fuses with itself, inner 
indusium with inner, outer with outer. Thus the embryo becomes cov- 
ered in by four cellular membranes; of these the inner indusium also 
secretes a granular substance and a cuticle. 
As to the origin of these structures the author inclines to the 
mechanical conception regarding the amnion, since the insect may be 
regarded as sinking down from the surface in its formation much as in 
the case of some “invaginal disks,” heads of cysticercus, the nemertean 
in the pilidium, etc. The indusium is, however, astructure found only 
in the Locustide, but may be homologized with a solid mass of cells, 
the so-called micropyle of certain Poduride. This micropyle in turn 
is comparable to the “dorsal organ” of crustaceans which possibly 
goes back to some sucking disk of some remote annelid-like ancestor. 
The author’s discoveries regarding the formation of the nervous sys- 
tem form one of the most interesting chapters of the present contribu- 
tion and add to the evidence for the derivation of insects from annelid 
ancestors. The first recognized start of the nervous system is when the 
blastopore is still open and many large clear ectoderm cells are distin- 
guishable in groups amongst the superficial common ectodermal cells. 
These large nerve-formative cells soon sink beneath the surface and 
are arranged in four long rows on each side of the blastopore. When that 
closes a median row is formed between the above four right, and four 
left rows. From these lateral and median rows of large cells the whole 
central nervous system is formed, at least all its true nervous sub- 
stance. These cell rows do not, however, become directly converted 
into ganglion cells, but disappear, it is thought, after having budded 
off a mass of daughter cells. Each cell in one of the rows buds off 
successive cells that form vertical strings of cells; these strings form 
the thickness of the nerve cord and become the ganglion cells with 
their nerve processes. The continuous ganglionic mass so formed 
becomes divided into sixteen successive ganglia which are reduced by 
fusion to ten. 
The brain is directly continuous with the ventral cord and, at least 
in the middle and posterior of the three regions into which it may be - 
divided, differs from the ventral ganglia chiefly in the lack of the mid- 
dle row of cells. 
