150 SMITHSONIAN INSTITUTION 



figures la-d, the radianal is just barely visible, and the basals to an equally small extent. A 

 step farther in infrabasal growth and both radianal and basals (except the posterior) would 

 be shut out (PI. VI, figs. 13a, b). 



The sketch at figure 12 on the same plate explains how this might happen in all stages of 

 growth ; the dotted line a .... a represents the line of infrabasal growth in cases where both 

 basals and radianal are left exposed to view ; if this line moved upward to b .... b the radianal 

 would be entirely covered while the triangular points of the basals remained in sight, as in 

 figure 9. And in general it may be stated that where the basals are well exposed nearly or 

 quite to their margins, so that the radials are about clear of the encircling infrabasal ring, 

 we can usually see the radianal; but where the basals appear separated, and covered with 

 the exception of their projecting angles, we cannot. This relation is not absolutely regular, 

 but the facts are sufficient to indicate that it is the rule, and that the radianal is therefore 

 probably present in all the species. 



This is corroborated by the evidence of the Swedish specimens. A second example of 

 H. parabasalis has been found since Angelin's time and is now in the Riks Museum — a very 

 perfect specimen much like the type except that the infrabasals are enlarged far enough to 

 overlap half of the radials, so that no part of the basals is visible (PL VI, figs. za, b). Here, 

 of course, the radianal has also disappeared, just as would have happened in the type species 

 (fig. id) if the infrabasals had grown a little higher up. In still another Gotland species, 

 which I have described as H. liljevalli, we have two specimens, in one of which the basals 

 and radianal are visible (PI. VI, figs. 14a, b), and in the other both are completely buried 

 (figs. 13a, b). There is an analogous case in Ichthyocrinus gotlandicus, where the radianal 

 is sometimes invisible simply because covered by the column. 



This proof seems to be sufficient to establish the radianal as a constant character of 

 Homalocrinus; but the question pertinently follows, why may it not equally exist in Calpio- 

 crinus, which would be simply in the condition of those Homalocrini in which the radianal 

 cannot be seen? For answer we must recur to the specimens on Plate VIII already men- 

 tioned, figures 4a, b, and especially to that of figures ya-c which we know has the fully de- 

 veloped Calpiocrinus arm structure ; these show that in two of the most characteristic species 

 there is no radianal. 



This, however, does not end the difficulties we encounter in endeavoring to adjust the 

 long confused relations between these two genera. I have, as I think, satisfactorily accounted 

 for the old and new species of Homalocrinus; but our troubles begin again when we under- 

 take to deal with the described species of Calpiocrinus. Leaving out the incorrect statement 

 about the basals, and all characters which are of family rank or are mere individual details, 

 the characters assigned by Angelin to the two genera may be stated as follows : 



fi. iBr 1, large, heptagonal ; 2 smaller, in series. 

 Homalocrinus: \ 2. Anals 3, subequal. 



[3. Arms repeatedly dichotomous. 



[1. iBr few, unequal. 

 Calpiocrinus: < 2. Anals variable, in longitudinal series. 



[3. Arms short, wide, unequally dichotomous. 



It is evident that the first two specifications of both are wholly indecisive, as they might 

 concur in the same specimen. This leaves only the third to be considered. Supposing this 

 specification in Homalocrinus to mean equally dichotomous, we know from the type specimen 

 of his only species that this is contrary to the fact, and therefore the third in Calpiocrinus 

 would apply to it also. There would thus be nothing in the diagnosis by which to distinguish 

 the two genera ; and upon these facts alone Homalocrinus, being the first described, would 

 stand. We have therefore to look to the specimens themselves, and not to the diagnoses of 

 Angelin, for the separation of the two genera. Under Calpiocrinus he described four species. 

 One of them, C. humilus, may be left out of consideration as not possibly belonging to either 



