Dr. Jw E. Gray on some Families of Bivalve Shells. 399>\ 



therUj or at least have them so sHghtly developed as uot to be 

 distinguished ; thus, Anapa in Paphiadte agrees with Paphia in 

 all the characters except in being destitute of any appearance of 

 a siphoual inHection, but the species of Paphia and the other 

 allied genera differ greatly in the depth of the inflection. 



It may be remarked, that though some genera of families 

 which generally have a siphonal inflection are sometimes without 

 it, I have not yet observed any genus which appeared to be- 

 long to a family usually without any retractile siphons or siphonal 

 inflection, provided with one, and the absence or presence is 

 generally a good family character, but one which must not be 

 solely depended on. 



On the other hand, the absence of a (contractile) siphon cannot 

 be considered an unfailing character of a family ; for though the 

 genus Nucula has, like the other genera of Arcade, no siphons, 

 yet in Leda and Yoldia, both so nearly allied to Nucula, that some 

 authors have refused to separate them, the one has short and 

 the other long separate siphons. The genus Lithopliagus differs 

 from other Mijtilidce in having united siphons varying in length, 

 but in L. patagonicus (D'Orb. Voy. Amer. Merid. t. 82. f. 24) 

 they are half as long as the shell. 



The structure of the siphons appears to be of considerable im- 

 portance, that is to say, if they are furnished with retractor 

 muscles which have a fan-like end which forms the siphonal inflec- 

 tion on the inner surface of the shell, or are only contractile ; but 

 all siphons with or without these muscles are more or less con- 

 tractile, and, as I have shown in the preceding observations, the 

 size of the retractor muscles appears to decrease until it becomes 

 absolutely wanting in the same family. 



2. Some authors regard the entire or partial union and separa- 

 tion of the siphons as of considerable importance, but it is only 

 necessary to look at the figures of the animals of the different 

 genera and indeed species of the same genus of Veneridce, Sole- 

 nidcB and Mactrada, to discover that this is a character of 

 secondary importance, as being exceedingly liable to vary in 

 nearly allied species ; nevertheless it is often useful to distinguish 

 those species from one another, and also some groups of genera. 



3. The adhesion or separateness of the leaves of the mantle is 

 another character which has been considered by some as infal- 

 lible, yet the extent to which the leaves are united generally 

 varies in the species of the genera which have this character, 

 as the species of tlie genus Solen for example ; but sometimes 

 two genera of the same family have the mantle leaves free and 

 adherent ; this is the case with Mactra and Lutraria of Mac- 

 tradce. Leach indeed separates the latter from the family and 

 places it with Mya on that account, but no one who has studied 



