122 Prof. P. M. Duncan on the 



cells which is situated upon the muscular layer may deceive 

 a young microscopist into crediting that there are cross light 

 and dark marks in the muscular fibres ; but careful illumina- 

 tion and the use of a lens capable of good definition expose 

 the error. 



Diadema setomm has the jirimary tubercles with crenulated 

 bosses and ]-)erforated mamelons, and the part of the spine 

 between the lower end or edge of the cotyloid cavity and the 

 milled ring is very long. I find that the muscular investment of 

 the joint greatly resembles that of Temnoplewus, that it arises 

 from the outer edge of the scrobicule of the primary tubercle, 

 and is inserted at the lower edge of the milled ring, and 

 possibly here and there upon the outside of the spine below 

 the milled ring and the capsule of the joint. This capsule 

 covers the ends of the spine and the top of the boss and covers 

 the crenulation and extends as a flat layer beyond it. 



No muscular fibres penetrate this capsule, and none arise 

 from the crenulation. It is interesting to note that in the 

 same specimen of Diadema some of the bosses are perfectly 

 crenulated, others are half crenulated, and not a few may be 

 destitute of the structure. This is not uncommon in other 

 forms. 



It would therefore appear that crenulation and })erforation 

 of the tubercles is not of sufficient physiological importance to 

 distinguish genera, although it may be useful in grouping- 

 species. 



The Branchice. (PI. XI. figs. 17-21.) 



The branchige of Temnopleurus and Salmacis are small and 

 narrow, although moderately long. They are situated upon 

 the peristomial membrane, close to the small so-called " in- 

 cisions " or rather grooves, one of which is placed close to 

 each ambulacro-interradial suture at the peristomial margin. 

 The branchia3 protrude and reacii up over the edge of the 

 grooves and may be 2 millim. in vertical measurement. In 

 Teinnopleurus (fig. 17) the narrow branchias are in finger- 

 shaped processes, from six to twelve or more in number ; 

 they arise from a hollow stem and rarely bifurcate, but some 

 are in tufts which come from the same source. All are tumid, 

 rounded at the free end, and hollow. Decalcified specimens 

 show that there is an outer thick layer of columnar epitlie- 

 lium (fig. 18), the cells containing granules which are most 

 numerous at tiieir bases (fig. 19) ; but it appears that there 

 may be groups of cells so crowded with pigment corpuscules 

 that a definite pattern is presented on the outside of the 

 finger-shaped body, consisting of longitudinal, irregular, 





