Herbert L. Hawlcins — On the Holectypoida. 3 



In Biscoidea cylindrica (PI. I, Fig. 3) the pits are far less deep, 

 even than in Pfjgaster. Indeed, it was not until I was drawing an 

 ambulacrum of this species under a high power of magnification 

 with specially arranged ligliting, and examining each granule 

 separately, that I discovered them. Unfortunately none of the 

 material I have yet seen is sufficiently weathered around the 

 invaginated peristome to give a clear view of the adorally situated 

 plates. But, although the lower plates of tlie triads are demiplates, 

 and of very restricted area, the four next outside the border of the 

 invagination show the pits. They are crescentic, and similar, save in 

 point of size, to those of Pygaster. 



"Whatever may be the significance of these pits, and however much 

 the extent of their distribution may be enlarged by further researcli, 

 I feel confident that they have essentially the positions indicated 

 above, and that nothing like them occurs on the other plates of the 

 triads. The nearest approach to them seems to be found in the sunken 

 tubercles met with on the adapical surface of the Cornbrash form of 

 Holectypus depresses, but those are on the interambulacral plates 

 between the central tubercles of the main series (see Hawkins, Geol. 

 Mag., 1911, p. 444, Fig. 1, H). The restriction of the pits in 

 (}uestion to the ambulacra at once suggests a comparison with the 

 pockets for sphseridia, and this is the only suggestion as to their 

 meaning that I can offer. Their arrangement (in reduced numbers) 

 in C. serialis is strongly reminiscent of the distribution of the 

 sphgeridiain Echinon'eus (Westergren, 1911), and the presence of four 

 of them in one area, would seem to compare with the four spha^ridia 

 figured by Loven [Etudes) for that genus. But very great difficulties 

 are encountered if the comparison between the pits of the Holectypoida 

 and sphseridia be carried further. In Pygaster and Biscoidea they 

 extend to a considerable distance from the peristome, and seem not 

 to occur in its immediate neighbourhood. It is true that in many 

 advanced Spatangoids sphseridia, or at least their pits, occur at an 

 even greater proportionate distance from the peristome, but they 

 occur on every plate, there being no intervening plates as in the 

 Holectypoids. 



A comparison is possible between these pits and the curious sunken 

 granules (bearing ' glassy tubercles ') in Echinon'eus, and the 

 probably similar pits that occur thickly scattered over the test of 

 Conulus. However, the restricted distribution of these ambulacral 

 pits seems to point to some more special and localized function than 

 could be ascribed to glassy tubercles. 



As the matter stands at present, these depressions would seem to 

 be either the pits for sphaeridia (in which case their distribution is 

 unique) or some special structure not yet recognized in recent forms. 



II. The ScrPKABRiNCHiAL Depressions. 

 In all the Holectypoida, with the partial exception of Conulus, 

 the branchial slits make conspicuous indentations on the margin of 

 the peristome, the deepest ]>art of the incision being situated just 

 in the interambulacrum. Passing from the peristome, near the 

 adradial suture, but separated from it by a row of fairly prominent 



