TAXOCRINIDAE 419 



Forbesiocrinus, and which now represents one of the most important groups of the present 

 genus. Hall did not recognize Onychocrinns, and put asteriaeformis under Forbesiocrinus. 

 Meek and Worthen, on the other hand, referred not only this but some species of Taxocrinus 

 to it, on account of having an anal tube which they then considered a special character of 

 Onychocrinus. 



If characters were to be given the same value here as in some other cases, it would be 

 possible to subdivide this genus into several. It stands as a very distinct type, sharply differ- 

 entiated from all other Taxocrinidae by its heterotomous arms. Included within this type 

 are species characterized by three different kinds of heterotomy, and by very constant differ- 

 ences in the number of primibrachs and in the development of interbrachials, some or all 

 of which under other circumstances would be held to be generic characters. Here, however, 

 we have a highly specialized type limited in geological range to the Lower Carboniferous, 

 and exhibiting many and rapid changes on approaching the extinction of the family. Under 

 such conditions it may happen that characters which have been important elsewhere in the 

 same family become subordinate or wholly worthless. 



Other types have run the same course. There are cases where a species, just at the 

 close of the life of a genus, has been known to exhibit a jumble of characters which at an 

 earlier period in its history had been constant and reliable for several well-defined species. 

 An example of this is seen in the Camerate genus Dolatocrinus, in which during the Onon- 

 daga epoch the number of arms was a good specific character, while in a rampant develop- 

 ment in the Hamilton, on the eve of extinction, these same characters appear worthless, — 

 irregular variations within the limits of species otherwise perfectly well-defined. Therefore 

 it seems to me the most logical course to treat the present form as a somewhat comprehen- 

 sive and vigorous generic entity, embracing a number of unusually well-marked species. 



In 1902 I proposed to separate 0. asteriaeformis from the genus because it differed 

 from all the other American species in having the ramules unilateral, branching only to the 

 outside of the dichotom instead of alternately from each side, and accordingly published 

 the name Oligocrinus for that purpose. The considerations above mentioned have con- 

 vinced me that such a course is undesirable under the circumstances. The Irish species 

 Taxocrinus polydactylus of M'Coy, which is of this type, has the ramules branching only 

 toward the inside of the dichotom, and there would be equal reason for creating a new genus 

 for it. This would be a greater refinement than is justified with two such isolated species, 

 and I think it preferable to abandon the name already proposed. A better reason could be 

 given for making two genera out of those forms that would remain under Onychocrinus, 

 which fall into some rather well-marked sections containing assemblages of species. Between 

 two such species as 0. exsculptus and 0. ramulosus there is a difference in habitus, as well as 

 in strong morphological characters, far greater than between the former and O. asteriae- 

 formis, on account of which under some circumstances we should without hesitation pro- 

 pose a new genus. 



But in view of the stronger distinctive character by which all the species of this type 

 depart from the rest of the family, and by which they are so conspicuously united, I believe 

 it the better practice to regard the differences in heterotomy as minor details, wholly subor- 

 dinate to the more general plan. Adopting this course, the species form a compact, satis- 

 factory genus, well limited geologically, one in which the leading forms are sharply differen- 

 tiated from one another. Unilateral heterotomy is not uncommon in other groups ; but 

 bilateral, alternating heterotomy is a new feature pointing toward pinnulation, and the most 

 characteristic species of the genus are those possessing it. Treating the former as in this 

 case a minor variation, and separating 0. polydactylus as a very extreme form imperfectly 

 known, the species will form two groups, which from the two best known species repre- 

 senting them may be called the exsculptus group and the ramulosus group. 



