144 Records of the Indian Mvscvw. [YoL, XV, 



study of the shell characters alone, placed the genera in the groups 

 established by him primarily on the structure of the marsupium. This 

 tentative classification has necessarily proved to be wrong in many cases. 

 It was shown in the first paper of this series, that the genus Physunio 

 should be placed in the sub-group Mesogenae of the group Exobranchiae 

 and not in the Endobranchiae as was done by Simpson. As a result of 

 my investigations on three of the common'^r Indian genera, it has been 

 found that these also have been assigned wrong places, Lamellidens 

 and Parreyssia were placed in the sub-family Hyrinae, which are, accord- 

 ing to Simpson, primitive forms, and carry the glochidia in the inner 

 pair of gills (Endobranchiae) Preston in the official " Fauna of British 

 India "^ has followed Simpson, loc. cit., adding nothing new so far as the 

 anatomy was concerned, and even neglected the various important 

 contributions on the anatomy of some of the Indian Unionidae published 

 since Simpson's revision was issued. These observations of mine are, in 

 part, a confirmation of what was found by Ortmann- as a result of his 

 study of the animals of Parreyssia wynegungaensis (Lea) and LameUidens 

 consohrinus, Lea. Preston on page 180 of the volume cited above 

 considers the latter to be only a subspecies of Law,ellidens marginalis 

 (Lamarck). Ortm_ann found that the position assigned by Simpson 

 to the genera Parreyssia and Latnellidens was quite wrong. Both these 

 genera were, from the structure of the marsupium, found to be exo- 

 branchous, and in his revision he placed them in the sub-family Unioninae 

 of the family Unionidae. This sub-family he characterised as having the 

 " Marsupium formed by all four gills or by the outer gills only ; edge of 

 marsupium always sharp and not distending ; water-tubes not divided 

 in the gravid female." Unfortunately the material which Ortmann had 

 at his disposal was very small and consisted of sterile or unripe females 

 from Bombay. 



In the case of the third genus Nodidaria, Conrad, it was found that in 

 the two Indian species investigated the animal was quite dift'erent from 

 that of N. aequitaria (Morelet), and N. japanensis (Lea), and that the 

 Indian forms of this genus unlike the Japanese and the African endo- 

 branchous forms should be placed in the Exobranchiae. This subject 

 is treated in detail further on in the account of the genus, which has for 

 this reason been separated from the genus Nodidaria and called Indonaia. 



Lamellidens. 



Two varieties of the common Indian species L. marginalis (Lamarck) 

 were studied. The first is a variety very little different from the typical 

 form and may for the purpose of the present paper be considered as such. 

 The nomenclature of the varieties and subspecies of L. marginalis is in 

 a very confused condition. Gravid specimens of this form were collected 

 in a tank at Bora near Serampur in the Hughly district, Bengal, on the 

 1st of April, 1918. The outer pair of gills, which alone formed the 

 marsupium, were found to contain large numbers of unripe glochidia. 

 Specimens were kept in large bowls in muddy water, but no further 



^ Mollusca, Gastropoda and Pelecypoda, p. 134 (1915). 



2 Ann. Carnegie Mus. VIII, pp. 222-3G5, pis. jvViii-xx (1911-12). 



