324 DR. G. P. BIDDER : NOTES ON THE 



Minchin (1900, p. 49) suggests that the pore-cells may primarily be 

 nutritively phagocytic; in function, and therefore have come to surround 

 pre-existing afferent pores. The hypothesis does not seem a priori per- 

 suasive ; and it does not fit well with the facts, since we have no evidence 

 of pre-existing pores, nor, in these sponges, of any prosopyles which do not 

 perforate porocytes *. 



[Postscript. 

 Note F. — Oogenesis in Sycon. 



Dendy's derivation of the oocytes from collar-cells (1914) in Grantia 

 compressa is not in harmony with my own views ; but on re-examination of 

 my preparations I cannot claim that they disprove it. 



His 'tigs. 2, 3, 12, 13, 14 (as also 62, 66, and 68) are from a sponge which 

 had been a week in the aquarium circulation. It appears possible, therefore, 

 that he observed the early stage of the pathological changes in the flagellate 

 chambers, of which I have recorded the condition after a month in the 

 aquarium (1895, p. 27). It is remarkable, however, that in my pathological 

 specimen, though the highly gelatinous mesogloea contains many cells of the 

 size and appearance of collar-cells, there are scarcely any recognisable 

 gonocytes or cells which resemble gonocytes in size or character. 



I agree with Dendy that the ova are found free in the cavity of the 

 chambers, strange as this appears, and that such is the interpretation- of the 

 " larger nucleate cells, possibly Protozoa, partly enveloped by the distended 

 collars, sometimes more than one cell converging on them," which I described 

 in 1895 (p. 31), and which are well indicated in his figs. 11 and 74 (in the 

 latter case the ovum is called a " primary spermatogonium "). I found 

 in sections of S. raphanus unmistakable ova in the lumen of the flagellate 

 chamber and one just passing through the wall, and I sketched at Plymouth 

 (April 1920) such an ovum alive in the cavity of a flagellate chamber of 

 G. compressa, seated with pseudopodia stretching over the free ends of half- 

 a-dozen collar-cells, in outline ludicrously like an octopus. Between the 

 vesicular nucleus and the free ovoid surface was a comet-shaped streak 

 or arc of minute green fragments, suggesting t na t it had used its freedom to 

 feed on and digest an alga, thus supporting Dendy's suggestion on pp. 323, 

 335 (1914;. 



Note G-. — Migration of the Nucleus in Collar-cells. 



In fig. 1 G the nucleus is visible in two cells, and is seen to have moved 

 downwards to allow of the escape of the fseces through the aperture in the 



* Cotte (1903, p. 434) considers that there is a general power of cell-perforation in the 

 pinacocytes of Reniera, and quotes Topsent for similar perforation? of the contractile cells of 

 this species and of Cliona. 



