100 THE AMERICAN MIDLAND NATURALIST 
Tennessee basin may furnish some interesting data for the recon- 
struction of ancient geographical features for the central Missis- 
sippi Valley. 
FamiLy II. Unionidae Swainson (restricted). 
‘Diaphragm complete, formed only by the gills; the outer 
lamina of the outer gills connected with the mantle at its posterior 
end. Anterior end of inner gills separated from palpi by a more 
or less wide gap. Margins of the mantle held together by the gill- 
diaphragm, but not united, thus separating the anal from the 
branchial opening, and the anal is generally closed above by the 
union of the margins of the mantle, (it rarely remains open), 
and when closed, it always leaves a supra-anal opening (which 
is very rarely obliterated). Gills always with water tubes formed 
by interlamellar connections developed as continuous septa, 
running parallel to the gill-filaments. Marsupium formed by all 
four gills, or by the outer gills alone, or by parts of the outer gills. 
Glochidia of various shapes, suboval, or subtriangular, or celt- 
shaped, with or without hooks on the ventral margin.’’—(Ortmann 
1912b.) 
Simpson’s terse diagnosis of this family is:—‘‘ Hinge with schizo- 
dont teeth; embryo a glochidium.”’ 
The family, Unionidae, naturally falls, into three divisions - 
on the basis of physiological and morphological characters; how- 
ever, this family may fall into two sub-divisions on the sole basis 
of reproductive functions. The Unioninae and Anodontinae 
would form the first and the Lampsilinae the second group from 
the fact that the discharge of the glochidia takes place in the 
former through the primitive and natural way of passage from the 
oyisacs through natural openings into the suprabranchial canals 
and then on out through the anal opening and in the latter the dis- 
charge is effected in a more direct and seemingly unnatural manner; 
that is, in the passage from openings forced through the ventral 
edges of the ovisacs, and thence out through the branchial open- 
ing. Yet the two sub-families, Unioninae and Anadontinae, have 
morphological differences in marsupial characters and in struct- 
ures of the glochidial masses that are correlated with physiological 
differentiation in breeding habits. On the same grounds, Lamp- 
stlinae is set aside as well a defined group; however the latter, 
although the modern group, is related to the primitive one Uniont- 
