CARADOCIAN CYSTIDEA FROM GIRVAN. 365 



statement as to the mouth and anus, which applies equally to all my Amphoridea, and 

 indeed to all Cystidea. The further statement that the ambulacra are developed in 

 two radii cannot, of course, be maintained for those forms in which the subvective 

 cxrooves have left no traces on the skeleton ; and for those in which there are traces, 

 the statement is not universally correct, e.g. Dendrocystis. Moreover, the elements 

 that bear the subvective groove are not uniserial in Dendrocystis ; nor are there four 

 basals in that genus. The supposed genital appendages occur only in Rhipidocystis. 



In short, there is very little that can be said about the genera assembled in the 

 Carpoidea, except that most of them show some trace or other either of a compression, 

 which is certainly not " dorso- ventral " in any morphological sense, or of a bilateral 

 symmetry. Not only are these characters well recognised as generally due to mode of 

 life, but in this particular assemblage they affect now one set of organs now another. 

 Little then remains to knit the genera together but those negative characters which 

 have prevented either Dr Jaekel or myself from assigning these and other genera to 

 the better-defined Orders of Cystidea. 



§ 18. Some of the differences between the included genera have been taken by 

 Dr Jaekel to warrant the establishment of two Orders : " A, Heterostelea, with pro- 

 nouncedly biserial columnals, strongly compressed theca, apparently always four basals, 

 and ambulacral organs as a rule devoid of skeleton." This Order comprises the Anoma- 

 locystidae and Dendrocystidae with their immediate allies. " B, Eustelea. Stem 

 simply built of ring-shaped columnals, without appendages. Theca spheroidal or com- 

 pressed. Base, so far as known, trimerous. Ambulacra borne by uniserial elements, 

 either pushed over the theca [epithecal] in two simple or forked rays, or raised as free 

 arms [exothecal], which, as well as the ambulacral grooves, give off branches on their 

 left side. Mouth subcentral. Anus near it on the right." This Order comprises the 

 Malocystidae and Comarocystidae, which Families are turned into Suborders and 

 further divided. 



§ 19. It is clear that the Heterostelea coincide more closely with Dr Jaekel's own 

 diagnosis of the Carpoidea than do the Eustelea. In the latter, for instance, the 

 columnals are not biserial, the theca is often not compressed, and it always shows 

 distinct traces of the subvective skeleton. Dr Jaekel admits an unbridged gulf 

 between the two Orders, and when he speaks of their relationship it is not clear whether 

 he regards the Eustelea as derived from the Heterostelea (though to this he seems to 

 incline), or whether he supposes both Orders to have descended from a common stock. 

 His inclusion of the Eustelea in the Carpoidea was admittedly provisional (vorlaufig), 

 and I am unable at present to accept the suggestion. Dr Jaekel says that my refer- 

 ence of the Malocystidae, and presumably of the Comarocystidae also, to the Cystidea 

 Rhombifera is erroneous because they have no pore-folds. It is possible that these 

 forms do not fall within Dr Jaekel's Dichoporita as diagnosed by him ; but there is 

 nothing in my diagnosis of the Rhombifera with which they are discordant. Since this 

 is not a fitting occasion on which to discuss our differing bases of Cystid classification, 



