ECHINOIDEA. II. 103 



From the description given by Agassiz it is thns inapossible to find how to distinguisli H.gibbosiis 

 from rxpcrgitus. A comparison of the figures seems to give a somewhat better result, the petals and the odd 

 ambulacrum showiiig some difference: In the specimen of gihhosus figured by Agassiz in PL XX. 5 and 

 9 (ca. 30""" in length) the posterior petals are only a little shorter than the anterior ones, and the nnmber 

 of pores in both petals is almost the same. In the largest specimen of fxpergifns (37'"'") the posterior 

 petals are only half as long as the anterior ones and the nuraber of pores in the posterior petals is 

 likewise only abont half that in the anterior; further in rxpcrgitus the inner ca. 7 pairs of pores in the 

 median (anterior) row of the anterior petals are small, in gibbosus, according to Fig. g, the>- are all large 

 and conjngated. The nmnber of piates in the odd ambulacrum within the fasciole is in gibbosus (ac- 

 cording to Fig. 9) ca. 18, in rxpcrgifus 29. — These differences look very good. If, however, \ve com- 

 pare the specimen of gibbosus of 20'"" before me with equal-sized cxpergitus. these differences become 

 verj- slight. In both I find the anterior petals twice as long as tlie posterior and with the double 

 nnmber of pairs of pores. In gibbosus I find the 4 inner pairs in the median row of tlie anterior petals 

 small (in cxpergitus about 7). In the odd anterior ambulacrum I find in gibbosus 14—15 piates within 

 the fasciole, in rxpcrgitus 17 — 18. And in the specimens from the Challenger» in the British Museum 

 the posterior petals are only about half as long as the anterior ones, and the inner 5—7 pores of the 

 inner series of the anterior petals are small, not conjugate. No specimen in the British Museum 

 corresponds to the Fig. 9. PI. XX of the <,Challenger»-Echini. These differences thus become so slight 

 that the)- seem rather inappropriate for distinguishing two species thereby. But other distiuguishing 

 characters do not seem to be found in the structure of the test. The fasciole is alike in shape, like- 

 wise the spines. To be sure the labrum, according to Agassiz' Fig. 6 would seem to give some differ- 

 ence: Its posterior end reaches on the right side the middle of plate 3 in the adjoining ambulacrum, 

 011 the left side to the middle of plate 2. As, however, this figure gives in any case a quite wrong 

 representation of the piates in the left posterior ambulacrum (I), it probably cannot be relied upon for 

 the right side either, the more so as in the specimens in the British Museum the labrum reaches only 

 to the middle of the second ambulacral piates of the adjoining series. The specimen from the Siboga» 

 likewise agrees exacth- with equal-sized cxpergitus in this respect. — The number of buccal piates 

 and the form of the peristome is the same in both of them. The tube-feet and spicules are alike. — 

 The globiferous pedicellariæ (not seen in the <;Siboga »-specimen) present a small difference (PI. XV. 

 Fig. 46) : the blade is more elongate, with four teeth around the terminal opening, and the Ijasal part 

 is narrower than in cxpergitus. The rostrate pedicellariæ do not present any reliable differences, 

 whereas the large tridentate pedicellariæ (PI. XV. Fig. 42) differ from those of cxpergitus in having 

 the edge in the outer part, where the valves join, regnlarly serrate — but in view of only one speci- 

 men of this kind having been found in cxpergitus, it does not seem reasonable to la\- any stress upon 

 this feature. Ophicephalous pedicellariæ were not met with in any of the specimens of gibbosus e.\am- 

 ined. There seems then not to be a single reliable difference of any reasonable importance by which 

 to distingui.sh gibbosus from expergitiis (— also in the structure of the globiferous pedicellariæ there 

 is some variation in rxpcrgitus, as pointed out above, p. 100, so that they present no reliable difference 

 either — ). If specimens of both < species » were put together, I think it wonld be impossible to separate 

 them rightly again. Accordingl>- I must regard //. gibbosus as a synonym only of //. cxpergitus; but 



