4-20 SMITHSONIAN INSTITUTION 



The first is the extreme and typical form on which the genus was founded, and under 

 which it was discussed by Meek and Worthen, viz., the form with profuse interbrachials 

 and a tendency to extend the free rays to resemble a starfish ; the second is a form with 

 equally prominent heterotomy, but having a very slight development of interbrachials and 

 no tendency to horizontal extension, and also having much greater length of the rami. 

 These represent the two extremes also found in Taxocrimis, from which they may well be 

 derived by the addition of complete heterotomy in the arm branching — the exsculptus group 

 from species like Taxocrimis intermedins, and the ramulosus group from the typical Taxo- 

 crimis like T . macrodactylus. 



From an evolutionary standpoint sub-divisions of the two genera upon these lines 

 might perhaps be an advisable procedure. The difficulty would be to find a dividing line 

 in Taxocrimis, where the change from few to numerous interbrachials is so gradual. A still 

 stronger objection is that the course of development of interbrachials is not the same in the 

 two genera ; for whereas in Taxocrimis it is by way of a progressive increase, maintained 

 to the last, this is reversed in Onychocrimts, in which the later phases are those of retro- 

 gression — the most profuse growth of interbrachials being in the species from the lower 

 half of the Lower Carboniferous. But at all events this distinction will make two rather 

 conspicuous sections under Onychocrimts. The first group begins so far as we know in the 

 Burlington and ends in the Keokuk, while the other begins in the Keokuk and ends with 

 the family in the Kaskaskia ; and there is a somewhat intermediate form in the St. Louis. 



I have already alluded to the fact that the alternate branching of subordinate ramules 

 is suggestive of pinnulation, the absence of which is such a remarkable characteristic of 

 the Paleozoic Flexibilia. Not only so, but there is in the very last American species, on the 

 eve of the extinction of the entire group, a more direct tendency toward pinnulation, by the 

 frequent reduction of the intervals between the ramules to two brachials ; a very slight 

 further modification, reducing the intervening brachials to one or uniting them by syzygy, 

 would produce a condition of arms nowise essentially different from that of a ten-armed 

 comatulid. This condition is approached in a new species from Scotland, O. wrighti, found 

 since the foregoing was written, which in the upper arm regions has some ramules alternating 

 on every successive brachial. 



The stem of Onychocrimts reached the maximum length for the Flexibilia, being long- 

 est in 0. ramulosus, in which it attains a length of three feet. In exsculptus, where we also 

 have it complete, it is considerably shorter. 



Representing, as it does, a high specialization of the Taxocrinidae, the genus was not 

 very long lived, being limited to the Lower Carboniferous; and it had its chief development 

 in America. The Irish species, of which the known material is too meagre for critical study, 

 is from strata probably corresponding to the later American rocks of that period. In the 

 length of its free rays it is the most extreme form, but in the other characteristics it would 

 go with the ramulosus group. We have fragmentary evidence of the existence of the genus 

 in the Knobstone beds, but no trace of it has ever been found in the lower beds at Burling- 

 ton. Two well distinguished species of the exsculptus type occur in the Upper Burlington, 

 but the culmination of the genus was in the Keokuk. There three splendid and beautifully 

 distinct species flourished in abundance, one apparently a lineal descendant of the Burlington 

 species in the lower of the two Keokuk horizons ; and the other two, types of the two groups, 

 side by side in the same colony of the other and later Keokuk horizon. The St. Louis and 

 earlier Kaskaskia formations have two intermediate species, and the later Kaskaskia, which 

 marks the extinction of the genus, has a very distinct successor of the ramulosus type. The 

 last is now represented by a similar species from the closely equivalent Hurlet limestone of 

 Scotland. 



