LECAN0CRIN1DAE 1 49 



basalis (PL VI, fig. la) similar ramules are borne directly upon the main branches produced 

 by the first bifurcation of the ray. There are thus in the latter ten main rami, giving off 

 ramules toward the inside of the dichotom ; whereas in Calpiocrinus there is the same kind 

 of subordinate branching, but it occurs from the second main division of the ray, so that 

 there are twenty such ramule-bearing trunks. In other words the heterotomy in Hoinalo- 

 crinus begins on the axillary secundibrach ; while in Calpiocrinus that bifurcation is substan- 

 tially equal and the heterotomy begins only on the axillary tertibrach. In each case the out- 

 side branch of the respective dichotom remains strong, but diminishes in steps by reason of 

 the successive ramule branching until the final division is equal (PI. VI, fig. 10; VII, fig. 46) ; 

 while the inner branches, or ramules, taper gradually to the distal end and become simple, 

 except in some intermediate cases where the lower ramule is enlarged, tending to become an 

 arm, with secondary ramules which are relatively inconspicuous and can rarely be seen at all 

 from the dorsal side (PL VII, figs. 1, 2). In the various figures of H. parabasalis on Plate VI 

 it will be seen that the lower ramule is relatively the largest, and the series of specimens shows 

 some variation in the size of it. If we had a sufficiently complete series we should doubtless 

 find various stages of its growth, from a relatively small side branch (something like a pin- 

 nule) to forms like Plate VII, figures 1, 2, where it is a strong arm, rising to the full height 

 of the ray, and itself having subordinate ramules. If this lower ramule should increase still 

 further until it equaled the size of the outer branch, with its ramules proportionally large, 

 then we should have the perfect arm condition of Calpiocrinus (PL VIII, figs. 50, 6a). Or, 

 conversely, the twenty main rami of Calpiocrinus may gradually take on the condition of 

 Homalocrinus through a tendency of the inner ramus to abort, thereby losing its own branches 

 and diminishing to the size of a mere ramule. The evolution might go either way, according 

 as it represented progression or retrogression, and in a type otherwise strongly specialized as 

 this is such modifications are to be expected. 



In my paper of 1906 I noted a further complication in the definition of the two genera 

 arising out of the apparent inconstant occurrence of the radianal in Homalocrinus, which 

 I found to be quite unequally developed in the English specimens, being plainly seen and of 

 good size in some and not visible at all in others. I said, however, that " the difference in 

 this respect may be partly due to the unequal development of the infrabasals, which may 

 sometimes conceal the radianal." This proves to be the correct solution of the difficulty. The 

 radianal in this form is differently located from that in any of the preceding genera, being 

 within the ring of basals and touching the infrabasals, as is the case in Sagenocrinus, and in 

 the Inadunate genus Thenar ocrinus; but it is more primitive because directly under the right 

 posterior ray. It is thus of small size — a low, transverse plate, lying below the lower margin 

 of the radials, much- below the angular points of the basals whose ring it shares, and not 

 higher in any event than their lateral margins. 



Homalocrinus, as has been stated, shares with Calpiocrinus the unusual character of the 

 base by which the infrabasals overgrow and conceal the basals and other succeeding plates ; 

 and we have already seen that the extent to which this goes in Calpiocrinus is variable, some- 

 times leaving the points of the basals exposed, and sometimes covering not only basals but 

 still higher plates. If there were in Calpiocrinus a small radianal lying between the two 

 basals, it is evident that it would be invisible in all cases where the basals themselves were 

 hidden by overgrowth of the infrabasal ring. This is precisely what happens in Homalo- 

 crinus. Out of the 17 specimens of H. parabasalis used in this investigation, eight have the 

 radianal visible, eight fail to show it, and in one the right posterior ray is not exposed. Tak- 

 ing figures 3a, b on Plate VI as typical of those in which the radianal is visible, it will be 

 observed that the basal plates are well exposed almost to their lateral margins, and the 

 angular lower faces.of the radials are in plain view; the radianal, being at the level of the 

 lateral margins of the basals, is accordingly exposed, just as the basals are. In the type, 



