no. 1646. REVISION OF BEYRICHIID E—VLRICH AND B ISSLER. 3 2 1 



Considerable instability of character in these inequivalved Bey- 

 richiidas was recognized in the preceding publication; also the fact 

 that many of them exhibited more or less obvious resemblances to 

 Ordovician and Silurian types. These facts were explained as result- 

 ing from reversion and arrested development incident to the degen- 

 eration and extinction of the family in Carboniferous time. 



In now recognizing four genera instead of a single broad genus, 

 it is not intended to convey the impression that the writers have 

 materially changed their opinions respecting the close alliance of 

 the several groups. Neither are the groups of species thought to be 

 more natural genetic associations than they were believed to be in 

 1906. It is only in deference to the matter of convenience in classifi- 

 cation, and to insure greater clearness and brevity in definition, that 

 the change from one to four is made. Considering them as one genus, 

 it seemed impossible to draw up a reasonably brief diagnosis that 

 would not cover a variety of really very distinct pre-Carboniferous 

 types. Though it can not be denied that the whole group is bound 

 together by intimate alliances, it is yet a fact that but a single feature 

 of generic or family rank pertains to all its members, namely, slight 

 inequality of the valves. But this feature, of course, is not peculiar 

 to this group, since it occurs in many otherwise very different Ostra- 

 coda. By dividing the group into four genera and using the feature 

 common to them all as characterizing a subfamily of Beyrichiida?, 

 the resulting classification is at least convenient* and probably as 

 natural as it can be made so long as established characters are 'con- 

 sulted as not less important than genesis in the forming of zoological 

 classifications. 



In the present arrangement of the Carboniferous Kloedeniellimv, 

 ill of the British species described by Jones and Kirkby are ac- 

 counted for save Beyrichia tuberculospinosa, B. multiloba, and B. 

 varicosa. The published figures of these three species a indicate pecu- 

 larities, the value of which it would be unwise to decide without first 

 verifying them by study of good specimens. Until that is done, it is 

 idvisable to leave them in the still large residuum of undetermined 

 ind doubtful species of Beyrichia. In the meantime it may be said 

 hat B. multiloba and B. tmrieosa would have been placed with 

 Wma were it not that in both cases, as figured bv Jones and 

 virkby, the left instead of the right valve is the smaller. In this 

 especl the two species agree with Klcedenella. The figures of B. 

 uberculospina are indecisive on this point, and the most that may 

 •e said of this species at the present time is to suggest that it may 

 e an aberrant H oiling and as probably related to //. longispina. 



".Tones and Kirkby, Ann. and Mag. Nat. Hist. (5), XVIII. 1S86. pi. vm. 

 Proc. N. M. vol, xxxv— OS 21 



