9 2 



ECHINOIDEA. II. 



of Spatangoids. In Acropsis rostrata the number is even not larger than 7, the two genital plates of 

 each side being, generally, united into one (Fig. 15); in the smallest of the specimens in hand, 7'5 mm in 

 length, the two left genital plates are, however, separate. Duncan (Op. cit. p. 272) suggests that in 

 the beautiful drawing given in the Challenger-Report, pi. XXXIII. a, fig. 10, there is a possibility of 

 the existence of a fifth imperforate basal plate ; this figure, however, is too little detailed or exact 

 for founding a so remarkable conclusion upon, and no such plate exists in this species. The madre- 

 poric plate occupies only the middle of the right (composed) genital plate; it is somewhat elevated. 

 Wyv. Thomson and Agassiz have found 4 genital openings; two of the specimens before me have 

 only two genital pores, a third specimen has three pores, none of them has four pores. The genital 

 papillae are well developed as in the type specimen. The genital pores have not yet been formed in a 

 specimen of 15 length, in a specimen of 17""" length they have appeared; it may thus be concluded 

 that they appear at a size of ca. i6 mm length. The labrum reaches to the middle of the 2. ambula- 

 cral plate, as is also seen in the figures in the Chall.-Ech. ; in the smallest specimen (7'5 mm ) it reaches 

 only to the end of the ist adjoining ambulacral plate on each side. In one of the specimens (that 

 figured as denuded) the anal area is almost quite naked, in the other specimens it is covered by plates 

 as described and figured by Agassiz. -- It is to be emphasized, that in the general form of the test 

 the small specimens agree with the large ones one proof more that the differences between the large 

 form figured in Chall.-Ech. PI. XXXIII. i 5 and the short, typical form are not (Compatible with 

 differences due to age. 



The number of tube-feet in the odd anterior ambulacrum increases with age. The specimen 

 of 7'5 inm has only two large tube-feet (one pair), the largest specimen has 12 (6 on each side). The 

 size of the sucking disk is comparatively the same in both small and large specimens not distinctly 

 an embryonic feature. No large tube-feet are developed near the periproct. According to Agassiz 

 (Chall.-Ech. p. 193) there are only ten large tube-feet round the actinostome. I find all the 15 tube- 

 feet of the inner plates well developed and of the usual form; in the largest specimen those of the 

 second ambulacral plates likewise begin to develop into the ilsual form. No spicules are found in the 

 actinal tube-feet; in the large frontal tube-feet the spicules are very numerous, almost smooth, elongate 

 rods, not arranged in longitudinal series, but forming a close mail round the foot (PI. XV. Fig. 5). The 

 extremely elongate rosette plates consist of a small flat, pointed inner part and a very long outer 

 part, the edges of which are bent inwards on the lower side so as to form an almost closed, narrow 

 tube at the inner end; towards the outer end the edges become less and less incurved, the point of 

 the plate being quite flat. The inner and outer parts of the plate are separated by a distinct widening, 

 somewhat thickened and with a bow on the lower side, evidently serving as a support of muscles 

 (PI. XV. Figs. 19, 20). To be sure I have been unable to see these muscles with certainty, but as the term- 

 inal disk in the preserved specimens is often folded in different ways, it seems almost beyond doubt 

 that such muscles really occur. 



The spines along the odd anterior ambulacrum are long and straight, not widened in the point; 

 those on the anterior part of the test, inside and outside the fasciole, as well as round the peristome, 

 are short, spear-shaped, a little curved in the point; those on the posterior end are simple, of medium 

 length. De Meijere (Siboga-Ech. p. 195) mentions as a character of A. rostrata, distinguishing this 



