ECHINOIDEA. II. 163 



can scarcely be any doubt that these curious eight-valved pedicellaria: are an abnormal case. If it 

 should prove a constant feature, it would certainly be a sufficient character for distinguishing this 

 form as at least a separate variety. 



From the Paris Museum 1 have received a specimen of Br. lyrifera-* from the Talisman , 

 1550 M. It has confluent petals like Br. atlantica, and the shape of the test is as in that species (PI. III. 

 Pig. 1. PI. IV. Figs. 5, 19), only the labrum is somewhat more prominent. There are only three pairs of 

 pores within the subanal fasciole. The pedicellarise, unfortunately, are very sparingly represented, only 

 one form of tridentate, rostrate and triphyllous pedicellarise being found. The tridentate differ some- 

 what from those of atlantica (PI. XIX. Figs. 13, 31). Also the rostrate pedicellarise (PI. XIX. Figs. 8, 16, 

 23) show some minor differences, especially in the basal part having often irregularly serrate edges. 

 I do not venture to state that this specimen belongs to Br. atlantica; but in any case it is not 

 Br. lyrifera. 



From Puerto Cabello we have in our Museum some specimens of a Brissopsis which prove to 

 belong to another, very distinct, new species; I shall describe it here under the name of Brissopsis 

 elongata u. sp. I have seen in the U. S. National Museum a specimen of this species from the Alba- 

 tross St. 2145 (25 fms., Caribbean Sea), referred to Br. lyrifcra, and further I have examined there 

 the specimens from Porto Rico mentioned as Brissopsis lyrifera by Clark (The Echinoderms of Porto 

 Rico. Bull. U. S. Fish. Coram. XX. Part II. 1900. p. 254) and find them likewise to belong to this 

 species. Probably also the specimens from the Sea between Jamaica and San Domingo mentioned by 

 Agassiz ( Blake >-Ech. p. 69) as « representing the extreme elongated form of Br. lyrifera will turn 

 out to be Br. elongata. In any case it is certain that this species also has been recorded as Br. lyrifera. 



The shape of the test (PI. III. Figs. 4, 14, 15, 19. PL IV. Figs. 1,4, 13) is upon the whole like that 

 of Br. atlantica. viz. the narrow form, only the posterior end is more vertical than is generally the 

 case in that species. The labrum is very little prominent, its anterior edge almost straight; its poster- 

 ior prolongation ends off the middle of the second ambulacral plate and it is much widened off the 

 border between the first and second ambulacral plate (PI. III. Fig. 19). The spines of the actinal plastron 

 accordingly do not reach so near to the mouth as in the other species. The first of the ambulacral 

 plates reaching within the subanal fasciole is the 7th. 1 This is a highly interesting case, showing that 

 the number of ventral plates is not everywhere limited to five, as maintained by Loven (On Pour- 

 talesia. p. 33); the same case is found in Toxobrissus pacificus. (I know of one more case, viz. in a new 

 species, which I am, however, not entitled to describe). There are 4 pairs of pores within the fasciole 

 (PI. IV. Fig. 18); sometimes the posterior one is indistinct and the tubefoot simple, not penicillate as 

 the others. In one case I have found only 3 pores on one side, while on the other side all 4 pores 

 were present. The spines of the subanal plastron are rather long and form two prominent tufts, 

 separated by a median belt of small spines. The posterior petals are almost as long as the anterior 

 ones; they are parallel in almost their whole length and very close together, separated only by some 

 very narrow interambulacral plates without tubercles (and spines), the latter beginning only on the 



■ In the specimen figured in PI. IV. Fig. 1 it is in the left ambulacrum exceptionally the 6th, which reaches the 

 fasciole. On the right side it is the 7th. 



21* 



