Manchester Memoirs, Vol. Iv. (191 1), No. 8. 3 



maximum, corresponding to the two maxima of the first 

 curve. 



The non-Keuper curve, however, still contained a 

 second sub-maximum at 74. Investigation showed that 

 this was largely due to forms from the New River and 

 Thames at Twickenham, both districts in the London 

 Clay. The latter specimens did not give such high ratios 

 as those from the New River, possibly owing to the greater 

 amount of chalk in the Thames. 



A marked discrepancy in the Keuper marl curve was 

 modified by the removal of forms from Congleton, where 

 the river in which they were found, although actually 

 passing through Keuper marl, yet flows for the greater 

 part of its course through Carboniferous beds. 



Forms from Evesham resembled those from the 

 Keuper marl in outline, but not in breadth and thickness 

 ratio. At Evesham the Avon is running through Lias, but 

 the greater part of its course is through Keuper marl. 

 Their removal smoothed out the curve. 



The curves for tiunidus gave a similar result with 

 maxima at 74 and •66-65. In the A. cygnea curve a 

 similar result was hinted at, but the number of specimens 

 from Keuper marl districts was too few to make a curve as 

 definite as in Unios. 



2. Another difference of form is visible amongst the 

 thinner shelled non-Keuper specimens. The shells from the 

 Marple, and Burnley and Accrington {Figs. B & C oi Plate) 

 canals are remarkable for their extreme length, and many of 

 the Marple specimens for a forward throw of the umbo, 

 giving a form very like the intertidal Modiolas. This altera- 

 tion in shape occurs during the growth of the shell as is 

 shown by the growth lines. Anteriorly the growth lines are 

 close together, posteriorly they are far apart. This points to 



