EPITHELIOID MEMBRANES IN MONAXONID SPONGES 555 
and at the free margin of a partial pore membrane. In fig. 17 
three pores are represented in which such activity is going on. 
In pore h the pseudopodia extend out from the incomplete pore 
membrane, in the other two cases from the margin of the pore 
itself. The fine pseudopodia were thrown out, moved about 
quickly, often fused more or less with one another, sometimes com- 
bining to form a network (pore c). They then were partially or 
completely drawn in, but reappeared after a short interval. This 
remarkable phenomenon was observed in the case of the three 
pores shown during one-half hour, at the end of which period the 
pores were still wide open and the pseudopodia! activity going on. 
At several other times I have noticed the form.ation of one or two 
flagellum-like pseudopods at the margin of open pores. Such 
pseudopods would quickly appear, move or wave from side to 
side, and be drawn in. There was nothing to indicate that this 
pseudopodial activity at the margin of pores was a pathological 
phenomenon. It is possible that it occurs commonly during life 
and that the pseudopodia are temporary, sensory processes which 
explore, so to speak, the region of the open aperture. The facts 
afford a further illustration of the widespread occurrence of filose 
phenomena,'' to the importance of which as an expression of the 
fundamental nature of protoplasm. Professor and Mrs. E. A. 
Andrews have repeatedly called attention (see especially, Andrews 
G. F., '97). 
Summary account of pore closure in Reniera 
The pore canals may undoubtedly contract, viz., while still open 
become smaller (comp. figs. 15 and 16). The entire thickness of 
the dermal membrane shares in this process. Actual closure 
is however brought about by an extension of the epidermal layer 
across the pore. This extension of the epidermis may at once 
constitute a simple and continuous pore membrane (fig. 10) 
similar to that present in Stylo tella. Or the epidermis may first 
extend across the pore in the shape of one or more strands of pro- 
toplasm which shift about in amoeboid fashion (figs. 12 and 16) 
and are then withdrawn into the general layer before the continu- 
