STATION IN FRESH-WATER MUSSELS. 275 
average, are more swollen than old ones. This does not, however, 
introduce a serious error, since from most localities young and old 
shells are at hand. On the other hand, this serves to explain, in 
part, the irregularities observed. Very old shells are often un- 
usually produced or elongated at the posterior end, and this tends to 
lower the percentage of the diameter. Just such shells have been 
found frequently in the Ohio near Pittsburgh, and these have, con- 
sequently, unduly depressed the average. Also the young shells, 
or, for that matter, shells of about the same size, when compared 
among themselves, show the decrease of the obesity in the upstream 
direction. The same phenomenon is found in other forms to be 
discussed later. 
Fusconaia subrotunda also enters several tributaries of the 
Ohio. First of all, we have it in Elk River in West Virginia 
(tributary to Kanawha). The Kanawha empties into the Ohio in 
a region where typical subrotunda is found. I have no material 
from this river. But in Elk River is a form, which I have called 
F. subrotunda leucogona, and which I have characterized as a 
“rather small and somewhat flattened subrotunda,” also differing 
by (mostly) pale soft parts and eggs. I doubt now the propriety 
of introducing a separate name for this form, for it might very 
well be included under both, subrotunda and kirtlandiana, these also 
having sometimes pale soft parts. The following table gives the 
measurements of the Elk River specimen. 
Lac. No. Max. Min. Av. 
Shelitonbetncttiseahthuaa see sacra 12 52 45 49.5 kirtl. (subrot.) 
(Gassawavae cocston ei rseoeh oes 12 53 38 48 kirtl. (subrot.) 
Simiiake): Wine soloigencis Ge ice ceca 12 49 43 46 kirtl. 
The decrease of the average diameter is also here clearly shown, 
the maximum being found in the two lower stations (with specimens 
which would fall under subrotunda). The minimum has been ob- 
served in the middle station, but only in one individual, and another 
with the diameter of 41 per cent., while all the rest do not fall 
under 45 per cent. This is one of the irregularities which should 
be expected. 
Another river which contains this species is Beaver River in 
