MUNSON: SPERMATOGENESIS OF PAPILIO. 
55 
matter seems to have been given no attention. As to the origin of the 
cyst cells, which was so much discussed by the earlier writers, Meyer 
(’ 49 ), Bessels (’ 67 ), and others, without finding any proof or giving any 
satisfactory explanation, the later writers dismiss it with mere con¬ 
jectures. My studies on the grandmother stem cell, to be related 
presently, give, I believe, a complete explanation of the problem. 
The number of cells in a gonocyst increases by division till there are 
between one hundred and one hundred and fifty cells. The cysts are 
approximately spherical. They are surrounded by a thin membrane, 
that was thought by earlier writers to be the mother cell which by 
endogenous cell multiplication had given rise to the contained sperma¬ 
togones. The cyst wall.or membrane is composed of a few very thin 
broad cells with distinct nuclei. Surrounding the nucleus there is 
usually a thick zone of cytoplasm, the peripheral parts of each cell 
being so thin as to be scarcely visible in section. There is no evidence 
of mitosis in these nuclei after the cyst is formed. Their number is 
probably determined very early, a fact that is sustained by the small 
number of these cells that become so prominent in the developed sper- 
matocyst (pi. 12, fig. 14-23; pi. 13, figs. 43, 44, 48, 49, 50, 51, 54). 
A comparison of the different forms represented in these figures shows 
that one of these gonocyst cells becomes the head nurse-cell of the 
spermatocyst, while the rest become enveloping cells of the bundle of 
mature spermatozoa. The origin of these cyst cells is explained later 
in connection with the grandmother stem cell, and the origin of the 
primary spermatogone. It will there be seen that they divide by the 
usual mitosis in that early stage. 
The gonocysts increase in size as the cells multiply, and as the cells 
in this period of multiplication lie closely packed, there being no 
internal cavity, there must be some growth of the cells after each divi¬ 
sion, a conclusion which is sustained by a comparison of the relative 
size of the spermatogones in the younger and older cysts. I do not 
consider it strictly true, therefore, that all growth is confined to the 
“period of growth” just preceding maturation. This is illustrated 
in the following diagram (text-fig. A). 
Degenerating gonocysts .— Both in the early stage of the larva and 
in the mature butterfly, some follicles occasionally contain degenerating 
gonocysts. These can be recognized by the fact that the contained 
cells vary greatly in size, some being several times larger than others 
(pi. 13, fig. 33). They can also be known by the nuclei which are 
