OF THE SOFT PARTS IN EUPLECTELLA ASPERGILLUM. 669 
distinguish three different layers of tissue corresponding to the layers of the 
germ. 
The whole free external surface of Huplectella, as well as the inner wall and 
the hollow spaces and canals which conduct the water from the external pores 
to the ciliated chambers, are covered by an epithelium composed of a single 
layer of thin flat cells, which, from the developmental observations which I 
have recently made upon Plakina,* I regard as proceeding in this, as in other 
sponges, from the ectoderm of the larva, and I shall therefore simply call it 
ectoderm. 
A like simple layer of epithelium which lines the ciliated chambers them- 
selves, as well as the whole exhalent canal system extending from the mouths 
of the ciliated chambers to the oscular openings, I call endoderm for the same 
reason. Finally, I consider as mesoderm the whole mass of connective tissue 
between these two layers, which represents the stroma of the skeleton and of 
the genital products. 
The Ectoderm.—In most sponges we can only succeed under specially 
favourable circumstances in detecting in the living animals the mits of the 
ectodermal cells, though it is easy to observe them by the use of nitrate of 
silver. I never succeeded in showing the limits of the cells in spirit specimens, 
and consequently I could not expect to see them in this case, since they can be 
brought into view by no known process. Of course, it by no means follows 
that the layer of ectoderm cells is wanting. On the other hand, I find their 
presence indicated, if not by the outlines of the cells, by the characteristic cell- 
nuclei, which, like the neuclei of the ectoderm cells of other sponges, are dis- 
tinguished by their circular form, by their fine refractive nuclear corpuscles, 
and here especially by their minuteness compared with the oval, paler, and 
larger cells of the underlying connective tissue (Pl. XVII. fig. 5). 
Besides, the situation of these small round nuclei, in the uppermost layer, 
directly washed by the water, their partial projection when seen in profile, and 
their tolerably regular distribution, are in support of the view of the existence 
of such a layer of ectoderm cells. . 
The Endoderm.—That part of the epithelial layer termed endoderm, which 
lines the exhalent vessels from the openings of the ciliated chambers to the 
oscular openings of the cribriform operculum, here, as in all other sponges, 
closely resembles the flat ectodermal epithelium, whilst that portion of the 
endoderm which lines the ciliated chambers is of an entirely different 
character. 
Although even the excellent state of preservation of the materials at my 
* Zeit. fiir Wiss. Zool., Band xxxiv. 
VOL. XXIX. PART II. ux 
