THE HTDE0SPIBE8 AND SPIRACLES. 89 



P. Verneuili (PL XL figs. l-(i; PI. XIV. figs. 8, "■>) there arc deep radial Binu6es the 



Steep walls of which arc crowded with closely set hydrospire-slits, but few of which 

 arc concealed by the narrow ambulacra (PI. XI. fig. 4; PI. XIV. fig. 9). On the 

 other hand, in P. caryqphyllatum, I'. Arckiaci, and /'. acutum (PI. XIV. figs. 1-7, 

 11, 12) the sinuses are less deep and more nearly filled by the ambulacra. These 

 are broad and petaloid, and conceal a considerable number of the hydrospire-slits 

 which arc excavated partly in the deltoid plates, but mostly in the radial limbs 

 (PI. XIV. figs. 3, 6, 11). This is well shown in the transverse vertical sections of 

 the calyx in P. Arckiaci and P. caryophyllatum which are represented in PI. XVII. 

 figs. 15, 1G. The enlargement of each lamellar tube at its inner end so as to 

 form a kind of sac is very well seen in these sections. In these species, therefore, 

 while some of the hydrospires — those nearest the top of the sinus — open directly to 

 the exterior, others can only do so partially, and others — those nearest the ambu- 

 lacrum — are completely concealed. As regards the second group it may be noted 

 that while the partially exposed hydrospires of P. caryophyllatum have their distal 

 ends open 1 (PI. XIV. figs. 3, 4), as is also the case in P. Arckiaci (fig. G), it is the 

 proximal ends which seem to remain uncovered in P. acutum (fig. 11). This is due 

 to the differences in the form of the ambulacra, as will be apparent to any one who 

 examines the figures on PI. XIV. 



The side plates of P. Arckiaci rather rest upon the lancet-plate than against it 

 (PI. XIV. figs. 6, 7) ; but in P. acutum and P. caryophyllatum they lie against its 

 sides with much the same relation to it as the side plates have to the lancet-plate of 

 Pentremites (PI. I. figs. 5, 6 ; PI. XIV. figs. 2, 3. 11), and they thus bridge over the 

 deepest part of the radial sinus so as to convert it into a tunnel. The proximal end 

 of this tunnel opens externally by a small pore close to the peristome, which 

 represents the spiracle of Pentremites. In the only example of Phcenoschisma 

 acutum which shows the ventral side this spiracle is artificially enlarged by the 

 removal of some of the side plates, just as in the specimens of Pentremites elongatus 

 shown in PI. I. figs. 5, 6. 



This condition is seen very much more plainly in the remarkable genus Crypto- 

 schisma. The hydrospire-slits are excavated in the sides of the radial sinus and 

 cross the lines of the radio-deltoid sutures just as in Codaster (PL V. fig. 24 ; 

 PI. XIII. fig. 8) ; but the ambulacra are so wide that they are completely concealed, 

 owing to the great size of the side plates which rest against and not upon the 

 lancet-plate (PI. V. fig. 23; PL XIII. fig. 20). The side plates are quite small, 

 however, at the proximal ends of the ambulacra and fail to meet the deltoids, so 

 that they leave spiracular openings. These lead into the hydrospire-canal which 

 runs beneath the side plates, and into which the lamellar tubes open by their slit-like 

 upper ends. The condition of the spiracles varies according to the relative sizes of 



1 Sec antea, p. 77. 



