244 MR J. ARTHUR THOMSON ON THE 



structure, and thus doing something towards increasing our knowledge of the 

 almost unattached Monaxonia. 



Part II. 



Appended Note on the Capsules found on the surface of Spongelia pallescens. 



(Plate XVII.) 



In the course of some studies on sponges, pursued in the laboratory of the 

 Zoological Institute at Berlin, Professor F. E. Schulze directed my attention 

 to certain peculiar club-shaped knobs which were formed on the surface of a 

 Spongelia, and entrusted them to me for examination. The death of the 

 Spongelia prevented me from tracing their further history, and I can therefore 

 at present only note their interesting structure, in the hope that others who 

 may haye observed these or similar structures may be able to explain their 

 import. 



The knobs were of the size of a small pin's head, and were raised above the 

 level of the already contracted sponge by stalks formed from projecting peaks 

 of the horny skeletal framework. The shape of the knobs was oval or pear- 

 shaped, and their contour was always perfectly defined. 



Treatment with silver nitrate readily revealed the interesting fact that the 

 knobs were surrounded by a well-defined ectoderm composed of cells of varying 

 shape and size (fig. 11). 



The contents of a teased-out knob seemed to consist of generally round 

 cells (figs. 3, 6) of very varied size, and with distinct nuclei, while sections of 

 stained knobs exhibited what appeared as a compact and intricate meshwork 

 of fine filaments, the meshes of which were occupied by cells of varied size (figs. 

 4, 7). Towards the margins, and especially towards the base of the knob, the 

 apparent network was looser, and there especially it could be seen that the 

 structure was that of incipient tissue with undifferentiated cells, or of connective 

 tissue in which the matrix was apparently wrinkled, or in some way modified 

 round the cells, following their contours and producing the appearance of a 

 very intricate filamentous network. The cells round the boundary, directly 

 below the epithelium, were often very regular. Below these and at the base of 

 knob, certain larger round cells (fig. 6) occurred, though by no means confined 

 to these portions. Between these and the irregular cells of the meshwork 

 intermediate forms were obvious. In some knobs the cells were almost all 

 rounded (fig. 10), as if not compressed to the same extent. In others, as was 

 especially well seen at the base, where the cells were less abundant, they 

 exhibited the greatest variability of form (fig. 8). 



As these knobs present perfect definiteness of structure, and are only in 

 formal contact with the sponge, it seems possible that they may thus secure the 



