Tin: HYDE0SPLEE8 AND si'ii; vt l.i.s. '.'7 



exisl in P. acutumi though in the specimen shown on PI. XIV. fig. 11 the) are 

 artificially enlarged bj the removal oftwo or three side plates. Thej are possiblj also 

 present in /'. Archiaci (PI. XIV. fig. 6), but we have not been able to make them 



out distinctly. Tin-) are verj evident, however, in OryptOScMsma Sckulzi, and vary 

 somewhat in their characters according to the relative sizes of the ambulacra and of 

 the deltoid plates respectively. In the individual shown on PI. XIII. fig. 20 each of 

 the ten sets of hydrospires has its own openingatthe proximal end of tin; radial sinus 

 a little below the level of the median ridge on the deltoid plate. But in the specimen 

 figured by Eoemer x , as in those represented on PI. V. figs. 23 & 24, the ambulacra 

 are not sunk so deep in the radial sinuses, and the two openings at the Mile of each 

 deltoid ridge thus become confluent into one, which may be quite small as seen in 

 Roemer's figure and in our fig. 24 on PI. V., or extend outward for some little distant e 

 above the deltoid as in fig. 23. The latter condition appears to be a somewhat 

 abnormal one, and to arise from the smallness or absence of the proximal side plates 

 which do not meet one another above the deltoid ridge as is usually the case. There 

 is thus the same difference between two individuals of Cryptoschisma as between the 

 two examples of Pentremites elongatus shown on PL I. figs. 4 & 5, or between the J', 

 sulcatus and P. Godoni represented on figs. 10 & 11 of the same plate. The spiracles 

 of P. Burlingtonensis are but little more distinctly double than those of either of our 

 specimens represented on PI. I. figs. 5 & 11. Put the species is placed by Ilambach 2 

 in a different division of the genus Pentremites from I', tttlctttux entirely on this 

 account. The facts mentioned above, however, will serve to show how little reliance 

 can be placed on Hambach's system of classification. His explanation of it will be 

 best considered when we come to study the spiracles of Pentremites. 



In the British species of Orophocrinus, as explained above, the hydrospires come to 

 be concentrated towards the bottom of the radial sinus, and are more or less com- 

 pletely covered up by the under lancet-plate. This is sometimes narrow and linear, 

 as in 0. pentangularis (PI. XV. fig. 10), and scarcely conceals auy of the hydrospires, 

 while the ambulacra lying above it are also narrow and do not come in contact with 

 the radials, so that the remainder of the radial sinus appears as " the spiracles," long 

 clefts on either side of the ambulacrum (PI. XV. figs. 5, 8). But in 0. verus the 

 under lancet-plate is relatively larger and wider, and comes in contact with the sides 

 of the sinus some little way from its distal end, so that the length of the spiracular 

 cleft is considerably reduced, but to a somewhat variable extent, as appears from 

 figs. 1-3 on PL XV. It is widest at the proximal end (PL XII. fig. 9), and gradu- 

 ally passes backwards into a mere slit until it finally disappears altogether. 



The rare American species 0. gracilis seems to be in much the same condition as 



1 Arclriv f. Xaturgcsch. 1851, Jahrg. xvii. Ed. i. Taf. iv. fig. lb b. 



2 Trans. St. Louis Acad. Bci. 1884, vol. iv. no. 4 6, p. 544. 



