1889.] On some Variations of Cardium edule. 



209 



chocolate colour. The shells are very long in proportion to their 

 breadth. (See tables.) 



Eamleh Lakes No. 2 and No. 3. — By the construction of the railway 

 from Alexandria to Cairo another portion of Mareotis has been cut 

 off by an embankment, and the lake thus formed was again divided 

 into two by the second embankment lately made to connect the Cairo 

 Railway with the Ramleh line. In this way two lakes have been 

 formed — an eastern (No. 2) and a western (No. 3). Both these lakes 

 are fresh owing to irrigation- waters. In No. 3 there are no shells of 

 Cardium at all, but in No. 2 I found quantities of living specimens. 

 These fresh- water cockles were in texture like the shells found in 

 Ramleh Lake No. 1, but the colour and other features were different. 

 The colour of the outside of the shells is almost uniformly yellowish- 

 white, but on the inside the region of the posterior 3 — 6 ribs is 

 chocolate colour. The rest of the inside of the shell has the same 

 bright white colour which characterises those of Ramleh Lake No. 1. 

 The proportion of length to breadth in these shells is very great. 

 Another character of these fresh- water shells is the frequent occur- 

 rence of specimens with the free ventral margins of the valves bent 

 inwards. 



Sub-fossil Shells. — At Mandara and elsewhere I found considerable 

 deposits of very large, thick shells, like those found occasionally at 

 Jaksi Klich in the Aral Sea district. Probably those shells were 

 deposited at the time when Abu Kir and Mareotis formed one or more 

 large lagoons in communication with the open sea. 



Recapitulation. 



The most important feature of these observations lies in the fact 

 that the shells of each sample, ivhether it be from a separate lake or 

 only from a particular level, have special characters, and are more 

 like to each other than to the shells of one of the other lakes or of 

 another level. The next feature of importance is the fact that in the 

 four independent cases, Shumish Kul, Jaksi Klich, Jaman Klich, and 

 the Egyptian lagoon Abu Kir, the shells which have lived under 

 similar conditions, i.e., in very salt water, resemble each other, having 

 the characters of thinness, light colour, small beaks, ribbing on the 

 inside of the shell, and great relative length. Similarly the shells 

 from the two isolated and independent fresh-water lakes at Ramleh 

 also present similar characters, viz., thickness, similar texture, and 

 shape. It may be remarked that the resemblance between the cockle- 

 shells from an Asiatic lagoon and those from Abu Kir becomes still 

 more striking when it is remembered that their immediate ancestry 

 is very different. For the Asiatic shells had been living for many 

 generations in the brackish waters of the Aral Sea, and had already 



Q 2 



