Re te tts ack A LAMELLIBRANCHIATA. 483 
outline and consequently of the size of the specimen. The rostral region of the 
right valve was greatly reduced and flattened, that of the left not so much reduced 
and made more strongly convex at a. In the lower half of the shell the result of 
distortion was reversed in the two valves. In the restoration, assuming the two 
valves to have been equal, we draw a curve through the point a that is intermediate 
in convexity between the flattened curve of the rostral half of the right valve and 
the sharpened one of the left. An equivalent curve is then drawn for the right 
valve and then continued to and beyond the point >. We now have the original 
outline of one of the valves as it would appear in an end view. The outline of the 
other valve being equally curved, only in an opposite direction, is then easily 
finished. 
Having satisfied ourselves as to the original shape of the shell, it is first desirable 
to determine whether the valves are equal, as in Modiolopsis and Whitella, or unequal, 
as in Pterinea and Aristerella. Next we note the relation of the various parts of the 
outline to each other, the relative width of the two ends and other features bearing 
upon the determination of the contour. Now the position, altitude and degree of 
fulness of the beaks and umbones is taken into account. The former may be terminal 
(i. e., situated at the anterior extremity of the hinge line and projecting as far forward 
as the margin beneath them), ‘as in Ambonychia; or they may be nearly central in 
position, as in some species of Ctenodonta. Then the umbones may be strongly 
gibbous and the beaks curve over the hinge line (Cuneamya, Whitella), or they may 
be less full and comparatively erect (Clionychia), or depressed, or scarcely distinguish- 
able (Cycloconcha and Clidophorus). Decided deviations in the position and altitude 
of the beaks are generally of generic value, but lesser modifications are likely to 
prove of merely specific importance. The character of the surface markings will 
probably have been taken into account at once. 
In the next order, and here we usually credit them with generic and greater 
value, the student should observe the presence or absence of a byssal opening, of the 
lunule and escutcheon, and the character of the area. He should note also whether 
the edges of the valves fit tightly or gape at one or both ends or ventrally. His next 
step is to observe the position, distinctness and relative size of the various muscular 
impressions, the adductors particularly. Nor is he to forget to trace out the pallial 
line. Next he may find internal sockets, plates or ridges, that supported internal 
ligaments, or to which muscles were attached. Finally, he will observe the method 
of hingement. The hinge may be edentulous, in which case an external ligament 
(perhaps internal also) may usually be assumed if not found (see fig. 35, IV). In 
Modiolopsis there may be a slight thickening or rudimentary cardinal tooth in each 
