306 
MR. W. BATESOX ON SOME VARIATIONS OF CARDIUM EDULE 
thick shells, pale in colour, having from 18 to 22 ribs, the region behind the ilth to 
14th ribs being purplish in colour. 
No paired shells were found at this level. In 30 shells, all between 21 mm. and 
16 mm. in length, the average ratio of length to breadth is 1 : 0799 ; that is to say, 
that the average breadth of a shell 20 mm. long would be 15'98 mm. 
The Second Terrace .—This is a Hat about 50 paces across. Upon it ai’e two well 
marked ridges of shells, the lowest of which is about 10 feet belov/ the level of the 
first terrace. These ridges were obviously formed by the casting-up of shells on the 
beach during gales, as may be seen on the shore of the Aral Sea in many places 
(Meregen Sai, &c.). They contain no paired shells with ligaments, such as are found 
lower down in places where the bottom of the lake has been exposed and not after¬ 
wards disturbed. 
Shells on this terrace were found of the maximum length of 26 mm. They have 
from 18 to 21 ribs, the region behind the 11th to 16th being purplish in colour. In 
20 shells taken from the lower of the two ridges of shells on tliis terrace the average 
ratio of length to breadth is 1 : 0770 in shells between 26 mm. and 20 mm. in length, 
and 1 : 0782 in shells between 21 mm. and 16 mm. in length ; that is to say, that 
among shells similar in size to those of the first terrace the average breadth of a shell 
20 mm. long would be 15'64 mm. The shells do not differ materially in consistency 
from those of the first terrace [vide Table of Comparative Weights) ; they are, 
however, slightly more highly coloured. 
The Tldrd Terrace consists of a strip of small sand-hills about 180 yards wide. 
The division between it and the region which I have called the fourth terrace is not 
sharply defined, but is indicated by a ring of old tamarisks. Such rings of tamarisks 
occur round many of the salt lakes of this steppe, and always show that the water 
stood at a definite level below them for a sufficiently long period to influence the 
vegetation. Some of the lakes in the Turgai district v^^ere surrounded by several 
concentric rings of tamarisks, showing several distinct periods in the progressive 
drying up of the water. This ring of tamarisks stands at a level about 20 feet below 
that of the ridge of shells which marked the lower limit of the second terrace. 
Amongst the bases of these sand-hills are many Cockles in situ, with their ligaments 
preserved, indicating that this part of the shore remains as it was when it formed part 
of the bottom of the lake. The shells on this terrace differ from those of the second 
terrace, being thinner, and showing that appearance of grooving on the inside of the 
shell which was referred to above (4). In shells of this terrace the grooving is not 
much marked in the case of ribs anterior to about the 11th. The number of ribs and 
distribution of colour are as they were in the last terrace. 
In 30 shells between 22 mm. and 18 mm. lono' the average ratio of leno-th to 
breadth was 1 : 0751 ; that is to say, that the average breadth of a shell 20 mm. 
long would be 15’02 mm. 
'The Fourth 'Terrace is like the last, in that it is a stretch of shelving sand about 
