MOOEE : ON THE FAMILY MELANIIDiE. 233 



Before, however, going further, it is advisable to discuss somewhat 

 in detail the anatomical characters of some of the more typical 

 members of the Melaniidae. 



In their admirable work, " Die Siisswasser-Molluscen von Celebes," 

 the cousins Sarasin have divided the Melaniidse into three ' unter- 

 griippen,' namely, Palceo-, Neo-, and Tylo-Melania respectively, and 

 these groups are distinguished mainly by the characters of their 

 opercula and radulae. 



In the first series the radula has the conspicuous littorinoid 

 character which is apparent in Melania episcopalis and the genus 

 Pachychilus, while the operculum is multispiral. 



In the Neo-Melanian group the radula has the peculiar features 

 associated with forms more nearly related to Melania amarula ; as, 

 for example, Melania punctata (L.), and Melania tuberculata (MiilL), 

 while the operculum is littorinoid. 



Before receiving a copy of the Sarasins' work, I had been led, 

 from a study of the soft parts and radulae, to form conclusions which 

 were somewhat similar to the above, and had made use of the terms 

 Littorino- and Ceritho-Melania. 



The third type described by the Sarasins appears to be peculiar to 

 Celebes, not being represented among the African or the New World 

 forms. 



In Melania episcopalis we find that the radula-sac is long, that 

 the salivary glands have the true littorinoid character, while the 

 nerves, except in the probably unimportant feature of being more 

 normally dyaloneurous on the right, are littorinoid too. There is, 

 however, a remarkable development of the reproductive apparatus, 

 in the form of a subcutaneous dorsal pouch, opening beneath the eye, 

 and connected with the genital aperture by. a groove corresponding, 

 probably, to that present in the Opisthobranchs. So far as at present 

 known, this condition is only met with elsewhere in the Tsenioglossa 

 among some closely allied Philippine species of Melania, and in the 

 genus Tanganyicia, a form which, however, in many ways belongs 

 to a totally distinct type. I am therefore led to the conclusion that 

 the grooves and pouches are probably to be regarded as extremely 

 primitive characters, and to be looked upon as the last remains among 

 existing Prosobranchia of the grooves and introversible penes of the 

 Opisthobranchia. 



This opinion is strengthened by the fact that in the female Littorina 

 and some other forms, such as Strombus, part of this accessory 

 reproductive apparatus, the groove, still remains, although in these 

 cases it appears to be quite without function. 



The existence of this curious apparatus in M. episcopalis, therefore, 

 does not necessarily at all dissociate it from the rest of the Littorino- 

 Melanias in which the pouch is absent. 



I have come, therefore, to the conclusion that it is in the highest 

 degree probable the Littorino-Melanias (or Palaeo-Melanias of the 

 Sarasins) have arisen, independently from the Ceritho- or Neo-Melanias, 

 as fresh-water derivatives of an old littorinoid group ; but it is clear 

 that, if this be so, such conchological similarities as these two groups 



