KUNGL. SV. VET. AKADEMIENS HANDLINGAR. BAND 59. N:0 3. 21 



In the absence of the earliest stages of development, which should facilitate an 

 empiric solution of the question, vve are obliged to make comparisons with other 

 forms of Lamellibranchs and to draw conclusions as to the greatest probability. There 

 are two circumstances against alternative 1. In most of the Lamellibranchs examined 

 by Bernard, where a small tooth comes as the first in the left valve and the sub- 

 sequent one is larger, this first tooth is 2 a and the second 2 b; few cases are known 

 (e. g. Lutetia), where 4 a is proved to exist, and these cases are essentially different 

 from the ones described here. On the other hand, the foremost tooth of the right 

 valve is remote from the outer shell margin and, consequently, can scarcely be 3 a 

 but is either 1 or 3 b. These facts, which thus militate against alternative 1, also 

 support alternative 3. 



Against the second possibility an additional circumstance may be alleged: no 

 similar case of reduction of both anterior right teeth is known, for either 1 or 3 a 

 or a trace of division of 2 is discernible in hinges with three tooth elements in each 

 valve. 



In accepting the third alternative as the most natural possibility, \ve state 

 the following differences between young stages of »normal» and »inverse» Chamas. 



In the »normal» Chamas tooth 2 is undivided and 1 rudimentary, but 3 a 

 persistent; in the »inverse» forms 2 is deeply split into one anterior smaller and a 

 larger posterior tooth in analogy with, for example, Petricola (ef. Bernard 1895); 

 further 1 is highly developed and 3 a is entirely reduced. 



As growth continues, the primary dentition changes, and finally the adult 

 shells assume a bewildering resemblance as mirror images of each other, so that the 

 fixed valves on one hand and the free ones on the other exhibit a striking similarity 

 reciprocally. This phenomenon is of course a false or pseudo-convergence. 



As a result of the facts concerning the development of the hinges in Chamidae 

 we find, on the one hand, that the whole homologization of Munier-Chalmas and 

 his theory of the symmetric valves must be abandoned as being founded on a false 

 assumption of a similarity that has proved to be nothing but a superficial pseudo- 

 convergence. 



On the other hand the difference between the both types of hinges in the 

 recent Chamidae is so fundamental that there are sufficient reasons for keeping them 

 apart as separate genera, the »normal» forms retaining the old generic name, and 

 the »inverse» forms constituting a new genus Pseudochama which is more closely 

 allied to Ecliinochama, which it embraces, than to Chama sensu stricto. 



There remains only a single fact that seems to disagree with the results de- 

 duced above, and, on the contrary, seems to speak in favour of the theory of 

 Munier-Chalmas. Reeve (1847) in his monograph on Chama has described a spe- 

 cies Ch. pulchella, which lie denotes as indifferently »normal» or »inverse» by these 

 words: it may be observed that the umbones are turned in one example from left 

 to right, and in the other from right to left». In fig. 10 a lie illustrates the first 

 case (a »normal» specimen) and fig. 10 b represents a larger, »inverse», shell. 



Through the kind permission of Mr. S. F. Harmer of the Natural History 



