XII BULLETIN 82, UNITED STATES NATIONAL MUSEUM VOLUME 1 



former primary division of the Zenometrinae from the rest by the columnar arrange- 

 ment of the cirrus sockets has been rendered unreliable by the inclusion of species such 

 as those of Eumorphometra, with more or less irregularly arranged cirri. 



Also introduced are some tables to illustrate the range of variation of species 

 where abundant material is known. Such species are in the minority in most groups 

 of comatulids, many of those included in this part being known only from single speci- 

 mens and these are often more or less incomplete. 



The loss of the outer parts of the arms and of some or all of the cirri makes an 

 estimate of the original size of many specimens very difficult. Until now the arm 

 length and to a lesser extent the length of the longest cirri have been used as criteria 

 of size but a crinoid received in the museum with complete arms is exceptional and often 

 at least half of each arm has been lost. Gislen has given occasional measurements from 

 the centrodorsal to the first syzygy on incomplete specimens and in 1955 the length 

 from the first brachial to the second syzygy, while John has given the lengths of vary- 

 ing numbers of brachials remaining attached. Mr. Austin Clark himself in a few 

 instances has used the length from the apex of the centrodorsal to the first syzygy, 

 but this proportion is more liable to error owing to the variable dorsoventral flexion 

 of the arms relative to the centrodorsal and also the often considerable variation in 

 shape of the centrodorsal itself, even within a single species (notably Leptometra celtica). 

 None of these authors has given comparable measurements in more complete specimens 

 of the same or related species. It seems to me that a criterion of size that can be 

 brought into general use is needed so that incomplete specimens can be compared with 

 others and differences in proportions taken into account. Since no consistency exists 

 between the measurements of other authors I decided to adopt two proportions, namely 

 the arm width at the first syzygy and the length from the proximal edge of the division 

 series (or distal edge of the radial or periphery of the centrodorsal if this overlies the 

 radials in the midradial line) to the second syzygy, the positions of the proximal syzy- 

 gies being sufficiently constant within the different groups dealt with here. Since 

 both these sutures are tangential in alignment, measurement is simplified. The dis- 

 tance to the second syzygy rather than to the first was chosen since the greater length 

 is less subject to error, though in both allowance should be made for dorsoventral 

 flexure of the arms where this occurs. An alternative would have been the distance 

 between the first two syzygies but this again is a shorter distance. The measurements 

 of width were made with a micrometer eyepiece and a magnification of about 12, while, 

 except in the smallest specimens, length was measured using a hand lens since accuracy 

 to more than half a millimeter is not essential, the measurement often varying within 

 that limit (or even more in large specimens) on different arms of the same specimen. 



These two measurements also have the merit of providing an estimate of the 

 stoutness of the arms, which appears to be characteristic of some species, notably in 

 the subfamily Heliometrinae. 



In the course of this work I have had occasion to examine many of the type 

 specimens of crinoids in the British Museum as well as a number from the U.S. National 

 Museum, for which I am indebted to the staff of the Division of Marine Invertebrates; 

 for the study of others, from the Siboga collections in the Amsterdam Museum, my 

 thanks are to Dr. Engel. I am also indebted to Mr. P.-A. Andersson of the Stockholm 

 Museum and to Dr. F. Jensenius Madsen of the Copenhagen Museum, who lent me 

 type specimens of Isometra vivipara. In all these specimens I have taken these two 



