
No. 426.] THE CILIATED SPONGE LARVA. 455 
conclusion with respect to the origin of the larva. Ijima’s 
observations are as follows : 
In Euplectella marshalli undifferentiated cells, styled ** archze- 
ocytes," are found in abundance scattered through the sponge 
body. Such cells occur “solitarily or in irregular groups of 
two, three, and so on, up to tens and in certain positions even 
to hundreds or thousands." The growth of the archzocyte 
groups ** may take place not only by cell division of their cells, 
but also by fusion of originally separate groups." The smaller 
groups are flat, the larger ones form solid and compact masses, 
the shape of which is at first rather irregular. '* With continued 
increase in the number of the cells and consequently in the 
size of the mass, the latter assumes a roundish, oval or broadly 
lobose shape, measuring up to 100 m or more across." Such 
masses are common in all large individuals. Ijima believes, 
on the strength of his own and earlier observations, that 
such congeries of archzeocytes are of general occurrence in 
the Hexactinellida. 
In certain archzeocyte congeries the cells undergo a histologi- 
cal change, the cell body developing spherules and increasing in 
size. The transformed cells are known as **thesocytes." The 
spherules are not firm solids but are of a soft, perhaps even 
fluid, nature. Ijima concurs in the opinion of F. E. Schulze 
that the formation of spherules is a metabolic process analogous 
to the formation of glycogen in liver cells, and that the spherule 
substance is to be looked on as reserve nutriment. The sub- 
‘stance of the spherules is probably of an albuminous nature. 
Ijima's description of the spherules strongly suggests that the 
bodies are comparable with the well-known G/anzkérper in the 
` rhizopod Pelomyxa (see Amer. Nat., Vol. XXXIV, No. 403). 
All the archzocyte congeries are not transformed into theso- 
cyte masses. Of those which are not so transformed, some 
possibly represent an early stage in spermatogenesis, but Ijima 
believes that “a good part of the primitive archzocyte congeries 
are directly and actively concerned in the formation of certain 
reproductive bodies, asexual or sexual but other than sperma- 
tozoa.” The further history of such reproductive masses in 
Euplectella marshalli was, however, not followed. But Ijima 
