96 Clark — On the Type Specimen of Alecto Purpurea. 



appreciate, and in some cases his original diagnoses have never 

 been revised, so that the identification of certain of his forms is 

 now more or less a matter of guesswork. Carpenter placed 

 some of Miiller's species in the synonomy of earlier species 

 described either by him or l)y Lamarck, without any notes in 

 regard to the type specimens. While in most cases this course 

 was no doubt justified, increasing knowledge in regard to the 

 differential specific characters of comatulids has raised certain 

 questions as to the propriety of his action in so doing in one or 

 two instances, and it has thus become imperative that Miiller's 

 types be reexamined, and described and figured along the lines 

 followed in the systematic work of the present day. Through 

 the kindness of the Berlin Museum I have been placed in so 

 fortunate a position as to be able to do this Avith the types in 

 their possession, and I take this opportunity of acknowledging 

 my deep indebtedness therefor to that institution, and in par- 

 ticular to Drs. W. W'eltner and R, Hartmeyer. 



Professor Miiller's systematic work on the comatulids, con- 

 sidering his limited amount of material, was exceptionally good. 

 Most of his descriptions even at the present day leave little to 

 be desired, being far more detailed and accurate than very 

 many of those subsequently drawn up by others. But he very 

 rarely gave any comparative notes; each of his descriptions he 

 regarded as a unit which needed no comparison with any other 

 similar unit. This has resulted in one or two instances in the 

 suppression of a species which, had a comparison with other 

 allied species been given, would have been recognized as valid. 



Alecto purpurea belongs to the Comasterida^ falling in the 

 subfamily Comactiniina^ and in the genus Comatula as now 

 understood. It is a rather small species, and is most nearly 

 related to C. fxrtinata from which, however, it is quite distinct, 

 being separal)l(' at once by the curious segregation of its cirri, 

 which are from five to ten in number and occur singly oi- in 

 pairs in the intei-radial angles of the centrodorsal, those of C. 

 pectinnta occurring in an, irregular row all around the margin. 

 I have examined a large number of specimens of C. purpurea, 

 all from Queensland, and compared them with some dozens of 

 C. pectinata from all parts of its range, and have never had any 

 difficulty in difT(>rentiating them. The type sju'cinien may be 

 described as follows: 



