SOME WEST INDIAN ECHINOIDS. 9 



ostracum. I might venture to suggest that the smooth radioles found by Agassiz in 

 specimens with spinous radioles were really such young radioles. The third speci- 

 men at hand, on the contrary, has all the radioles of the smooth form; the same 

 is the case in the specimen figured by Clark. (The Cidaridte, Plates 8-9.) There 

 are, however, reasons for suggesting that this form of radiole does not really belong 

 to the species hartletti, the specimens bearing them being probably hybrids (see 

 p. 10). The radioles are banded with brown, the ground color being whitish; 

 the coarse spinelets are white, including those on the brown ])ands, which makes 

 them especially prominent. In sections the radioles are seen to have the ostracum 

 covered with fine, unbranched "hairs." (Plate 16, fig. 2.) In the smooth radioles 

 these hairs appear to be somewhat less numerous. The actinal radioles (Plate 

 15, figs. 8, 13-14) are very little s|iecializeil, faintly serrate along the striae of the 

 aboral side; the striae are very little developed on tlie adorai side and only toward 

 the point. The secondary spines are flat anil somewhat pointed; those around the 

 radioles are distinctly longer and broader than the primary ambulacral spines (not 

 nearly of the same size as mentioned in the description in the BlaJce Echini). The 

 inner ambulacral spines are very small, not (|uite 1 mm. long; they scarcely reach 

 the base of the primary aml)ulacral spines. These latter are about 3 mm., those 

 surrounding the radioles about 4-5 mm. in length. The spines on the anal plates 

 are rather large and are so bent as to cover the anal opening. Those on the geni- 

 tal and ocular plates are smaller, especially the latter; the spines forming the outer 

 series on the ocular plates, mentioned under the description of the tuberculation, 

 are bent outward so as to cover the ocular pore. "Ampullae" are well developed 

 on the abactinal spines, and also on those around the radioles. 



The spicules do not afford any specific features; they are of the form typical 

 of citlarids, and are arranged in the usual way so as to leave a naked space for the 

 nerve. 



The pedicellariff of this species have received considerable attention. The 

 large globiferous pedicellariffi are of the peculiar form described and figured in the 

 Ingolf Echinoidea" with a well-developed end tooth and the opening reduced to a 

 small pore. The stalk generally is provided with a limb of |)r()jecting rods, but this 

 is not always distinct (see Plate 12a, figs. 12-13 of A. Agassiz and Clark; Hawaiian 

 Echini, Cidaridae). 



The small globiferous |)edicellariae have, like the large form, a well-developed 

 end tooth, but tiie opening is larger and triangular (Plate 17, fig. 6). As is often 

 the case in Cidarids in which both large and small globiferous pedicellariae have 

 (or lack) an end tooth, intermediates occur of which it cannot be said with cer- 

 tainty whether they belong to the one or the other form, both forms varying con- 

 siderably in size and also in the size of the opening. The tridentate pedicellariaj 

 (Plate 17, fig. 1) are simple, narrow, the valves joining in the outer half of their 

 length; there are numerous cross beams in nearly the whole length of the blade. 

 They reach a size of 2 mm. in length of head. 



In the Hawaiian Echini, Cidarida^, Agassiz and Clark have figured a series 

 of large globiferous pedicellaria) from a single specimen of T?: hartletti (Plate 12fl, 

 figs. 6-11), showing a much larger degree of variation than described above, the 



nPt. 1, p. 16, pi. 10, figs. 22-23, 30-31. 



