530 J. Starlde Gardner — Eocene Aporrhdidce. 



modifications they have iindergone with the lapse of time may be as 

 apparent as possible. Each has its place in the long chain of descent 

 of which the whole Eocene period forms so to speak but a few links.. 

 Eightly interpreted, each should form a landmark, assisting us to fix 

 with some show of reason, not only the relative ages and mutual 

 relationship of the subdivisions of the Eocene, but even to form a 

 kind of idea of the time that had elapsed between their several 

 periods of deposition. Bearing these considerations in mind, fossil 

 shells, it will be seen, demand far more minute and critical examina- 

 tion than those that are living. This aspect of the case again forces 

 upon us the question, one becoming more and more pressing and 

 urgent, as to whether the binomial system can be made much longer 

 to suffice. Here is a group of forms, sej)arated by characters that 

 are in themselves not entitled to specific rank, and not of the same 

 value as existing species. We attach therefore a false importance to 

 them, in calling them species, while if we call them varieties or 

 sub-species, we depart from the ordinary practice, and at once intro- 

 duce a triple nomenclature. Personally I should like to see them 

 written with three names, as Ap. Soioerhii hantonensis, etc., a method 

 that would save much definition. 



In order to avoid tedious repetition, and to condense the descrip- 

 tions, I propose to describe one type sufficiently minutely, and to 

 jDoint out in what way the others depart from it. In a group of 

 shells so very closely allied to each other it does not seem likely that 

 this can lead to inconvenience. 



The first record of their occurrence in the English Eocenes is in 

 Sowerby's Mineral Conchology, vol. iv. p. 69, plate 349, figs. 1-4. 

 They are described as Rostellaria Parlcinsoni, and stated to occur 

 always above the Chalk. In the sixth vol. 1827, p. 112, pi. 558, 

 fig. 3, another illustration is given under the same name, but in the 

 systematic index, p. 248, published in 1835, the name is changed to 

 It. Sowerhii, Mantell having in the interval, 1829, pointed out that the 

 original B. Farhwsoni, Geol. Trans. 2nd ser. vol. 3, p. 203, is a wholly 

 different and a Cretaceous fossil. It was named in compliment to 

 the son, J, de C. Sowerby, who continued the Mineral Conchology, 

 and not, as supposed by Nyst, to the father, who was the author of it. 

 Nothing has been done with the Eocene species in England since, 

 except to transfer them to Aporrhais. 



At least three of the species to be described were figured in 

 Sowerby under the one name, and it is therefore immaterial to which 

 it becomes definitely attached. 



Apokehais Sowekbii, Mantell, sp. 1829. London Clay. 

 Plate XVII. Figs. 5 and 6. 



The characters are based upon 96 specimens in the Edwards' and 

 other collections in the British Museum. 



The maximum dimensions are, length 24 mm., width across wing 

 20 mm. : the smallest adult shell measured 13 by 11 mm., the wing 

 being thickened in this case to 2 mm. It is never composed of less 

 than 5 or more than 8 whorls, the average being 6. The whorls are 



