242 MR J. ARTHUR THOMSON ON THE 



sectioned in the usual fashion. I found it useful for general survey to make 

 several large sections through the sponge. 



The ectoderm of Suberites exhibits no marked peculiarities, but consists of 

 a fine layer of small polygonal and apparently unequal cells, the contours of 

 which were readily demonstrable by the silver nitrate or gold chloride method. 

 With a lens the skin can be seen to be covered with fine pores, while numerous 

 larger apertures are irregularly distributed over the surface. I was, however, 

 unable to discover any oscular opening or openings. The larger apertures 

 seemed, on closer examination, to be widened canals occupied by the abundant 

 commensal polypes. Between the pores the points of the monact spicules 

 projected slightly above the surface. 



Not a little of the difficulty attending the investigation of Suberites is clue 

 to the very abundant occurrence of the large free siliceous needles. On a section 

 through the whole sponge a radiate disposition can be recognised. They 

 extend in large crowded brush-like bundles from the centre outwards, though 

 considerable irregularity of arrangement is also observable. The brush-like 

 bundles are best seen towards the surface between adjacent canals. The 

 needles exhibit a simple unaxial form, running to a point at one end, and 

 knobbed like a pin at the other (Plate XVI. fig. 4). I also observed a diact 

 form, with double median nodes and with both extremities pointed, a modifica- 

 tion readily derivable by fusion or doubling. 



The Ciliated Chambers (fig. 4). — The disposition of the ciliated chambers, which 

 is of course the principal point, is very difficult to determine owing to the number 

 of spicules, the compactness of the sponge, the minuteness of the chambers 

 themselves, and the adjacent relations of afferent and efferent canal systems. 

 In sections of fortunate thickness and staining, the small chambers can be seen 

 throughout the whole sponge, but more abundantly towards the periphery. 

 They seem to have a somewhat more than hemispherical form, and exhibit on 

 cross section as many as sixteen cells round the margin. These chambers are 

 in direct communication with the finer branches of the ordinary canal system, 

 which exhibits what is usually termed the fourth degree of complexity. An 

 inner flattened nucleated epithelium could be detected as the lining of some of 

 the canals. The afferent and efferent canals lie side by side, their parietal pores 

 are adjacent, and beyond the distribution of the chambers and the direction of 

 the increase in the diameter of the branch canals, I could detect no difference 

 between them, and no special oscular or efferent regions. 



The Connective Tissue. — The mesodermic connective tissue exhibits great 

 variety of composition in different regions. The cells vary greatly in shape, 

 from round and regular to polygonal and multipolar, or to long drawn out 

 spindle-like forms. Fine connecting threads between adjacent cells could be 

 readily recognised. In some cases, besides nucleus and nucleolus, the intra- 



