There is no doubt that the spicules form by far the most important structures for dividing 

 the Muriccida- into generic groups, and it is impossible to properly identify species, or to place 

 them in their genera, without a microscopic examination of carefully cleaned spicules. In order 

 to test this point, the writer, when beginning work on the muriceids of the Siboga Expedition, 

 attempted to divide the great mass of new and unfamiliar material into genera by the use of 

 external characters, such as manner of branching, shape and disposition of calyces, and a super- 

 ficial examination with a dissecting lens of the spicules in situ. The result was a lamentable failure, 

 except in the case of some particularl)- well marked genera such as Acanthogorgia or Acis. In 

 many cases one could have done ju.st about as well by assorting the material when blind-folded. 



It thus becomes evident that superficial characters are comparativelv useless in defining 

 the genera of the Muriccida-, and this leads inevitably to the conclusion that it is exceedingly 

 hazardous to attempt to identify the old and long know species by the figures and descriptions 

 of the earlier writers; for these men used exclusively those characters which were superficial 

 and thus of little value for our present purpose. It is onlj- when some subsequent writer has 

 had access to the type specimen and given it a careful technical description that these old 

 species can be .safely identified. It thus appears impossible to identify with any certainty many 

 of the long known species, for many of the types are lost or inaccessible. 



The writer has, in such cases, preferred to carefully describe a species as new whenever 

 he is not reasonably sure that it should be referred to an already de.scribed form, rather than 

 to call it by an old name merely on account of some superficial resemblance. In such cases 

 more favored naturalists, who may have access to the types, can detect the error by comparing 

 the detailed descriptions and figures in the present work with the type. He can then, if necessary, 

 relegate the supposed new species to the ranks of synononi)-. In this way less harm is done 

 than by erroneously ascribing a really new form to an old species, particularly when it involves 

 an incorrect idea regarding distribution. 



In the present work the spicules are regarded as by far the most available structures 

 for generic diagnoses. It is only occasionally that the shape or the size of the calyces is used, 

 and it is but rarely that the mode of branching assumes generic importance in the Muriccidcc. 



In studying the spicules for systematic purposes it soon becomes evident that the matter 

 is far more complex than would appear from a study of the literature of the subject. In some 

 cases spicules of a dozen different types can be found on a single .slide from a single species, 

 while in others the spicules that are regarded as typical of a certain genus may be found in 

 several genera. Moreover it soon becomes plain that the same spicule may assume the form 

 of several types during the course of its individual development. 



But the investigator who perseveres in his work finds that, in spite of these and other 

 discouraging conditions, there are a number of forms of spicules that can readily be recognized 

 as typical of certain genera. They may be associated with many other forms on a given slide; 

 but nevertheless they are the dominant note of the composition, and can readily be 

 detected by the experienced observer. By a study of these dominant types we are enabled to 

 very much simplify the classification of the Muriccida-, and divide that troublesome family into 

 a fairly natural group of genera. 



