EPITHELIOID MEMBRANES IX MOXAXOXID SPONGES 551 
Granules sufficiently large to be recognised individually and which 
are abundantly present in mesenchyme cells, are not found in the 
epidermis. The reticular character of the sheet is very distinct 
in places where the staining is both deep and clean. The meshes 
appear to be actual spaces. The}' look clear and empty and are 
bounded b}" the stained reticular lines. The thin pore membranes 
closing in pore canals are especially favorable for such observa- 
tions. The epidermal sheet is certainly of surprising delicac}' 
and thinness. Tt may often be traced over the large spicules which 
lie horizontally and form the superficial meshes of the skeletal 
network. In such places it rests upon the white background of 
the spicule and the reticular character comes out distinctly. 
There are places where mesenchyme cells of the dermal membrane 
also lie on top of the superficial spicules. But it is where the 
epidermis alone crosses the spicule that the opportunity for study 
is so especially good. 
Results with other methods 
Sublimate was given a good trial as a fixative. The epidermis 
cracks a great deal. The fragments are sometimes fit for study. 
They exhibit the reticular character of the sheet and an absence 
of cell boundaries, as noted above. 
Picro-sulphuric which is a good fixative for the epidermis in 
Stylotella does not give good results on Reniera. The epidermis 
cracks into pieces. When treated with this fluid the membrane 
seems to have no stiffness. Thus it often drops down into the 
pore canals and breaks away from the part left on the surface. 
The fragments are sometimes fit for study. They are frequently 
polynucleate but exhibit no cell boundaries. The reticular char- 
acter of the sheet could not be observed on this material. 
Several trials of the silver nitrate method were made on favor- 
able days. The silver entirely failed to show the presence of 
cell boundaries in the epidermis. Where the stain is deep, the 
outlines of mesenchyme cells and processes sometimes appear. 
The silver was used according to the method already described 
for Stylotella. Oscular tubes that had been so stained were 
