OSTREA. 351 



but in some of the formei' it is small/ while in some of the latter it attains a large 

 size (fig. 123), and such specimens are indistinguishable from 0. mncroptera. It 

 seems, therefore, that no line can be drawn between 0. macro2)ter(t and the Upper 

 Cretaceous forms, especially since it is found that the size of the wing is directly 

 related to that of the attached surface of the left valve which was almost certainly 

 determined by external conditions. When the attached surface is large and 

 mainly posterior to the umbo, then the macropfera type with a large wing arises. 



The examples of this species (here known as 0. diluviana, L.), found in the 

 Upper Cretaceous deposits, have received numerous names, of which the principal 

 are 0. pcctlnata, Lamarck, 0. carinata, Lamarck, 0. coluhrina, Lamarck, 0. froiis, 

 Parkinson, 0. t^errafa, Brongniart, 0. pn'onota, Goldfuss, 0. Millrtidvu, d'Orbigny, 

 0. BlrorJcana, d'Orbigny, 0. Zeilleri (Bayle). In the LTpper Cretaceous of England 

 this species is not known to occur above the zone of Holaster suhjlohnsus and has 

 been usually named O.frons or 0. mrlnnta, but on the continent of Europe and in 

 other parts of the world it is represented in the higher parts of the Chalk. 

 D'Orbigny and Coquand, believing in the principle that specimens found at different 

 horizons belong to different species, have regarded as distinct forms the following 

 amongst others : 0. Milletiana from the Gault, 0. carinata from the Cenomanian, 0. 

 pedviafa (O.fronx) from the Lower Senonian, 0. scrrata from the Upper Senonian. 

 But later authors have found the o'reatest difficult v in distiuQ-uishino- these 

 " species." An examination of a large number of specimens, even when collected 

 from one locality and one horizon only, shows that the shell is -extraordinarily 

 variable, and that every gradation is found between the different types ; apart from 

 variation in the number and size of the ribs, the curvature of the shell, etc., there 

 are differences which are obviousl}^ duo to the size, shape and position of the 

 attached surface. Similar modifications occur at all horizons, and it seems 

 impossible to separate as distinct species the forms found at different levels. 

 Sometimes at one locality or in one kind of deposit a particular form of the shell 

 may be more common than others, and may, if only a small number of specimens 

 are available, give the impression of l)eing a distinct species, but whenever a large 

 collection is made other varieties are found which make it impossible to regard as 

 a species the form which at first sight appeared to be distinct. The difficulty of 

 separating these supposed species is shown by the fact that although d'Orljigny 

 states that 0. froiis is characteristic of the Senonian, yet he includes in his synonymy 

 the specimens figured by Sowerby from the Upper Greensand and Chalk Marl. 

 Similarly, in quoting foreign examples of a " species," authors have sometimes 

 unconsciously cited them from horizons in which they believed the species did 

 not occur. 



' See, for example, a specimen from the Lower Greeusaud of Faringdon (figs. 106, 107), in 

 whieh the left valve was attached to a long narrow object, and tlie posterior wing is not developed. 



