34° The Field Naturalist 's Quarterly November 



which they style respectively the ectoderm and endoderm ( = outer skin 

 and inner skin). 



It will perhaps be convenient here to make a little digression. We 

 mentioned in Chapter II. that all animals originate from a single cell, 

 the ovum, "which, after fertilisation, subdivides— becoming first a mass 

 of cells, and then, by a more or less complicated series of changes, a 

 separate individual." Now, however far apart in the scale the parents 

 may be, the ova in all cases begin to develop in identically the same 

 way. The single cell divides into two cells ; these divide, making four ; 

 the four divide, making eight ; the eight, sixteen ; sixteen, thirty-two ; 

 and so on. Very soon, however, the cells cease to be a mere bunch, 

 and arrange themselves into two layers — ectoderm and endoderm. 

 Eventually the different parts of the embryo (such, for instance, as the 

 alimentary canal and the spinal column in the case of a vertebrate) are 

 formed by the infolding of these two layers. 



The reader will perceive from this that if we could demonstrate the 

 existence of an ectoderm and endoderm in a sponge, we should be able 

 without hesitation to place it amongst the Metazoa, because all these in 

 their early stages possess these two germinal layers. 



We are confronted, however, by a serious difficulty — the nature of 

 the cells in sponges. The amoeboid cells are similar in all respects to 

 Amoeba described in Chapter 1 1., having a nucleus and contractile vesicle, 

 and changing their position and shape by means of pseudopodia. 



The collar-cells are highly characteristic, and merit careful study. 

 The drawing (Fig. 18) will give an idea of their shape and appearance. 

 Each consists of a mass of protoplasm, containing 

 nucleus and contractile vesicle. The base is em- 

 bedded in the jelly-like mass which connects all parts 

 of the sponge together. The opposite end carries a 

 long flagellum and a thin transparent cup-shaped 

 collar. This appendage is not constant, it is merely 

 a part of the protoplasm of the cell " spun out," and 

 can be entirely withdrawn at will. In fact, it is never 

 quiescent, for there is a continual slow interchange 

 going on between the particles of the collar and the 

 body of the cell. Particles of food brought by the 

 currents caused by the flagellum adhere to the surface 

 of the collar, sink into its substance, and are carried 

 Firr 1% —Collar- t ^ lus mt0 ^ e ce ^ body. ^ n f act > tne process is exactly 

 cell of sponge. ^ e same as m Actinophrys (Chapter III.), except 

 that the radiating pseudopodia are here modified into 

 a cup. I will even venture to say that functionally the pseudopodia 

 and collar are identical. 



In Fig. 19 is shown Monosiga, a flagellate Infusorian, with a retractile 

 collar, which in every detail closely resembles the sponge collar-cell. 



Some authorities state that the two kinds of cell in the sponge are 

 really only one — that they are interchangeable. A collar-cell will with- 

 draw the collar and flagellum, become amoeboid, and wander off in the 



