OF SOUTHEEN INDIA. 299 



The shells of the TJnionidm are extremely variable as to form and thickness ; 

 all stages from ovately rounded into ovate and linearly elongated, or rectangular 

 and trapezoidal shapes, are to be met with. The solidity of shells varies equally 

 as the size, and often depends upon the character of the locality where the animals 

 live. The hinge has either an anterior and one posterior tooth in each valve, or 

 only the anterior are present, or both are absent; the ligament is moderately 

 strong and supported by fulcra ; muscular impressions two. 



As regards geographical distribution the Unionid^ are world-wide ; they are 

 almost everywhere present where permanent fresh-water pools or rivers occur 

 though they are to be met with in greatest numbers and variety in the American 

 especially North American, rivers. There the recent species form quite a contrast 

 to the insignificantly small number of tertiary forms as yet on record. But also 

 in other countries fossil Tlniones are a great rarity, for true fresh-water deposits do 

 not appear to have been very extensive in any of the former geological forma- 

 tions. Eastern Europe was during the latter part of the miocene period {'' Sar- 

 matische Stuffe'') inhabited by a comparatively large number of Tlniones, some of 

 which greatly resemble North American types (see Homes' Eoss. MoUusken des 

 Wiener Beckons, vol. ii). Erom the German miocsene and oligocgene deposits only 

 very few are known, and India, during the Sivalik deposits, was also peopled 

 only with a small number of species. Erom the eocene deposits of the Paris 

 basin Deshayes describes only five species. Of cretaceous forms I may say there 

 are only two with sufficient certainty known from Europe ; two are doubtful, but ten 

 supposed new forms have lately been added from North America (New Jersey). 



The first typical forms of JJnio appear in the so-called Wealden deposits of 

 North Germany and England ; the forms partially represent true TJnio, partially 

 Margaritana, Erom the carboniferous beds I add two genera, but it yet remains 

 to be proved that their classification in this place is correct. 



The generic divisions which I adopt here are mostly those recorded by H. and 

 A. Adams. These genera are unquestionably as good as any others which have 

 been proposed in other families, but I do not enter into the sub-generic divisions, 

 there being no occasion to discuss those details here. I do not see any benefit 

 in arguing the great unity of the genus JJnio, for Deshayes seems to have 

 exhausted the subject, though, as the result of all his arguments, he at last finds it 

 necessary to describe two of the Paris species as Anodonta and three as JJnio I 

 Deshayes places also Cardinia, Ag., in the present family, but I do not think 

 that there can be much difi'erence in the opinions of other conchologists regard- 

 ing the classification of this genus, for it agrees in all essential characters with 

 the AstartiDjE, where it has been recorded. 



a. Suh-famili/— VNIONINJE. 

 1. Carhonicola, M'Coy, 1855, (Brit. Pal. Ptocks and Eoss., p. 5M). Shell 

 elongated, solid, with moderately tumescent, not eroded, beaks, somewhat impressed 

 in front of them ; ligament strong external ; surface concentrically striated ; 



4 D 



