156 THE CRINOIDEA CAMERATA OF NORTH AMETIICA. 



easy to discover and define the points by which certain assemblages of these 

 individuals agree with each other, and difier from others, with sufficient con- 

 stancy to be called species. With but a single specimen in hand, and this 

 imperfectly preserved, as is often the case, it is impossible to decide whether 

 we have a good species or a mere variation, whether it is the young or the 

 adult; and whether or not the distinguishing character represents a mere 

 abnormal condition of some established species. In describing new species, 

 therefore, the utmost caution is required, especially since the number of 

 described species in America alone has increased to almost fifteen hundred. 

 That a specimen comes from virgin soil, distant from any other known 

 locality of the same horizon, does not make it a different species. The 

 geographical range of species is much wider than formerly supposed, and 

 careful comparison with authentic specimens of allied forms must always be 

 made before a form can be recognized as a valid species. Among the char- 

 acters to be considered as most important for distinguishing species, we 

 recognize the form and proportions of the calyx ; the relative proportions of 

 the plates and their ornamentation ; the number of arms and arm openings, 

 the direction of the latter, and their distribution around the calyx, whether 

 continuous or separated by the supplementary plates ; the form and position 

 of the orals, whether flat or tumid, symmetrically or asymmetrically arranged; 

 the presence of '' radial dome plates " or regular covering pieces, and other 

 characters of a more specialized nature ; not forgetting that some of them 

 depend on the more or less adult condition of the specimen, and its preserva- 

 tion, and that in certain groups some of them are wholly worthless. 



That a given character may be good in one group, and without any 

 value for classification in another, is a fact so fully recognized at the present 

 day that there is no need of citing instances to prove it. Every working 

 naturalist has encountered striking illustrations of its truth. 



To facilitate the identification of species among genera containing 

 a large number of forms, we have arranged our descriptions so as to place 

 species which are most closely related next to each other, thereby en- 

 abling the student to make satisfactory comparison with allied forms. In 

 Platycrinus, which contains an unusually large number, we have arranged 

 the species into subordinate groups. 



It is not our intention to go into details upon the classification of the 

 Inadunata, but a short review of them will be necessary for this work. 



We have stated that we divide the Inadunata into Larviformia and 



