CR1N0ID FROM THE SOUTHERN SEA. 
921 
which it exhibits are (1) the appearance of a closed ring of basals on the exterior of 
the calyx ; (2) the separation of the radials by inter-radial plates; (3) the presence of 
a jointed arm-like appendage on the inter-radial of the anal side. 
The persistence of the oral and basal plates of the larva, together with the small 
size of the specimen, might be thought to indicate that Thaumatocrinus is merely a 
type in which the resorption of the orals and the metamorphosis of the basals into 
a rosette take place unusually late. But as I have already pointed out, the condition 
of the centrodorsal, and of the cirri which it bears, is evidence that the specimen must 
have been detached from its stem for some little time, while the orals show no signs 
of any approaching resorption. The existence of the interradials also decidedly 
indicates that the basals on which they rest form a permanently closed ring on the 
exterior of the calyx. This is the case in but a very few Co mat idee, either recent 
or fossil. 
Schluter mentions a Cretaceous species in which it occurs ;* while there are 
several forms, both of Cretaceous and of Jurassic age, in which the basal ring is 
incomplete, and the radials partly rest on the centrodorsal. But the only recent type 
in which the basals remain visible on the exterior of the calyx is the curious genus 
Atelecrinus ;t and here they are very small in proportion to the radials. In all 
other recent Comatulce the basals disappear from the exterior of the calyx towards 
the end of Pentacrinoid life, and become transformed into the rosette. Some species 
remain much longer in the Pentacrinoid stage than others; so that of two calices of 
equal size, the one may be still attached to a stem, the top joint of which bears but a 
few rudimentary cirri, and have large basals; while the other has a centrodorsal 
bearing 15 or 20 cirri, and concealing both the basals, and a part of the first radials. 
Antedon rosacea and A . dentata ( A . sarsi, Dub. and Kor.) are two excellent instances 
of this difference, the latter attaining a length of 40 millims. in the Pentacrinoid 
state. A similar condition is presented by a Pentacrinoid which was dredged by the 
“Porcupine,” and is probably to be referred to Ant. eschrichti, or to Ant. quadrata.% It 
is stouter and altogether more robust than any larva of Ant. dentata which I have 
seen ; and though its.radials are as large as those of the free Thaumatocrinus, yet its 
basals are actually higher than those of the single specimen of the latter type ; while 
the centrodorsal on which they rest has merely a few imperfect cirrus-stumps, and is 
scarcely larger than the stem-joints below it. On the other hand, another larva 
from near Ascension (S. 344 ; 420 fathoms) has equally large radials resting directly 
on the centrodorsal, which bears about eight well-developed cirri; but the basals 
have already disappeared from the exterior of the calyx. An earlier stage in the 
development of this same larva is shown in Plate 71, fig. 6, for comparison with 
Thaumatocrinus. Although the calyx and arms are well developed, the basals ('h) 
* Zeitschr. d. Deutsch. Geol. Gesellsch., Jalirg. 1878, p. 66. 
t Bull. Mus. Comp. Zool., vol. ix., No. 4, 1881, p. 16, pi. i., figs. 1-7. 
t This is the Ant. celtica of Marenzeller, and of Duncan and Sladen ; non Barrett. 
