REPORT ON THE CRINOIDEA. 169 



height of the cup. The second radials and axillaries are well developed, as are also the 

 arms, which are unfortunately broken at about the tenth joint or earlier. But even 

 under these circumstances the head has a length of 4 mm. A slightly bifid plate, having 

 a somewhat worn appearance, stands up in one of the interradii of the disk. It may be 

 one of the orals, or, as I am more inclined to think, the anal plate. For I cannot make 

 out anything corresponding to it in the other interradii, which are, however, but 

 imperfectly visible. A striking feature of this very robust larva, and one in which it 

 resembles Antedon tenella rather than Antedon rosacea, is the large development of the 

 arms before the appearance of the cirri. The radials and brachials are larger than those 

 of a recently detached individual of Antedon rosacea. This is also the case in the Penta- 

 crinoid of Antedoneschrichti and in that of Antedon midtispina, from near Ascension 

 (PI. XIV. fig. 7), which has a very robust appearance, like the larva now under con- 

 sideration. The latter can hardly be a younger stage of the Pentacrinoid of Antedon 

 eschrichti than that figured by Levinsen. 1 Their difference in relative age is not great, 

 while they are very unlike in many respects. The " Porcupine " larva has high basals 

 and relatively wide first radials, with short, wide, and well-formed axillaries (PI. XIV. 

 fig. 3) ; while in the Pentacrinoid of Antedon eschrichti the basals are low, the radials 

 relatively high, and the axillaries rhombic, about as wide as long. It would appear for 

 the same reason that this larva cannot belong to Antedon quadrata, which is most 

 closely allied to, if not identical with, Antedon eschrichti, while it is clearly not that of 

 Antedon tenella, and the only other Comatula known to occur in the cold area is Antedon 

 hystrix. 



The brachial ambulacra of this larva are protected by relatively large plates, not 

 unlike those which occur in some varieties of Antedon phalangium, but the armature of 

 the ambulacra in the mature Antedon hystrix consists of quite simple rods of limestone. 

 This difference may perhaps be explained by the fact that an absorption of the perisomatic 

 skeleton of the Pentacrinoid seems to take place in some forms of Antedon rosacea, as 

 noticed by Dr. Carpenter. 2 



3. Antedon tenella, Retzius, sp. (PL XIV. fig. 4 ; PI. XXXI. figs. 1-4). 



€ 



Specific formula — A. y. 



1783. Asterias tenella, Retzius, K. Svensk. Vetensk. Akad. Handl, 1783, t. iv. p. 241. 



1788. Asterias tenella, Linn., Systema Naturae, ed. xiii., cura, J. F. Gmelin, Lipsiaj, 1788, t. i. 



pars vi. p. 3166, 

 1805. Asterias tenella, Retzius, Dissertatio, sistens Species Cognitas Asteriarum, Lundae, 1805, 



p. 33. 

 1825. Alectro dentata, Say, Journ. Acad. Nat. Sci. Philad., 1825, vol. v. p. 153. 

 1835. Comatula mediterranea (?), Sars, Beskriv. og Jagtagels, Bergen, 1835, p. 40, pi. 8, fig. 19 a-j. 



1 Loc. cit, tab. xxxv. fig. 8. 2 Phil. Trans., 1866, p. 741. 



(ZOOL. CHALL. EXP. — PART LX. 1887.) OOO 22 



