28 UINTACKINUS: ITS STRUCTURE AND RELATIONS. 



There are other variations which point still farther to an instability of 

 the base. These are well shown by a series of figures on Plate II. Figs. 

 1 and 2 show maximum and minimum sizes of the centrale in Form M. In 

 a few cases the centrale is somewhat elongate, and here its angles are more 

 pointed than usual (Fig. 3). Then follows a series of specimens having 

 respectively a centrale, and one, two, three, and four other plates (Figs. 

 4, 5, 6, 7). In studying the collection I found myself noting specimens 

 like Fig. 4 as monocyclic, but having a " double centrale," and those like 

 Figs. 5, 6, and 7 as dicyclic, with 2, 3, or 4 IBB. I had no special reason 

 for this at first, but preserved the distinction on observing that in Fig. 

 4 both plates meet interbasal sutures by more than one angle. For the 

 same reason it would probably be consistent to say that in Fig. 5 there is 

 a double centrale and one infrabasal. It is important to note that in Fig. 

 6 the angles of the centrale next to the infrabasals are interradial, while 

 those which abut directly upon the basals are radial. In Fig. 8 all five 

 IBB are present, but unequally developed. In Fig. 9 the IBB are barely 

 visible as minute points, which truncate the corners of the centrale, leaving 

 it nearly round. In Fig. 10 the dicyclic base is fully developed. There 

 are also to be found cases with only four basals, and with 6 basals (Plate II., 

 Figs. 14 and 15). 



All these irregular cases are rare and exceptional, and constitute mere 

 individual variations. 



Form D has the essential basal structure of Marsiipites, viz. : basals, 

 infrabasals, and a centrale. As to it, to this extent, the essential dissimi- 

 larity between the two genera pointed out by Mr. Bather ( Oj^. cit. p. 996) 

 does not exist, the calyx from the radials down being cotnposed of the same 

 elements. 



The occurrence of this new Form D throws some light upon the ho- 

 mology of the centrale, — so far at least that it reduces the number of 

 hypotheses that may be entertained regarding it. One of the suppositions 

 heretofore suggested is disposed of: it cannot be (1) the fused infrabasals, 

 because it is found here co-existent with them, a distinct element, situated 

 within their circlet. There are thus left the other possibilities, viz. : (2) 

 that it is the representative of the proximal or distal stem ossicle ; or (3) 

 '■ a simple supplementary plate developed to fill up the gap left by the 

 disappearance of the stem." An objection to the second hypothesis is 

 found in the fact thnt the centrale is located within the circlet of infra- 



