stone form abundant in certain localities, but not generally distributed. Thus in the State of New York, according to Professor Hall "it abounds at Trenton Falls and at Sugar River in Lewis County," but "is scarcely known as occuring in the Champlaim valley."* In Canada it is found at Lachine and at several other places, but there are hundreds of good exposures of the rock in the Province that have been carefully examined, where not a single specimen has been seen, although in all the localities S. alternata is more or less common. S. camerata occurs at one spot in the vicinity of Ottawa, but I have never met with it elsewhere. S. tenuistriata may be collected in the hard black limestone around the base of the mountain of Montreal, particularly in the neighbourhood of the McTavish monument and also at Ottawa and two or three other places, but does not occur at all in the majority of the localities of the Trenton limestone. These three species, therefore, must have been capable of existing in certain places only, on the bottom of the ocean during the period of the accumulation of this rock, while S. alternata, flourished everywhere. Whether or not, therefore, they are to be regarded as distinct species, this much at least seems probable, that they were by some peculiarity in their habits or in their organization, unfitted for so wide a range through the seas as that enjoyed by S. alternata.

The question, whether or not these supposed species are really distinct, cannot be answered until naturalists shall have discovered some general law of life by an appeal to which they may in all cases determine what is a species as distinguished from a mere variety. It is scarcely necessary to state that such a law if it do exist at all may remain unknown to man for ages, and in the meantime nearly all determinations of species from varieties where the forms are very closely related may be regarded as not positively established. The physical geologist is more interested in the results of investigations which show that certain forms are confined to particular geological horizons than in those, whose sole object is to determine the exact zoological relations of such forms. If it be true, for instance, that that particular form of the genus called Strophomena incrassata is confined to the limestones lying next under the Trenton in the fossiliferous series, it makes no difference to the geologist, whether it

* Palæontology of New York, vol. 1, page 107.

be in reality a distinct species of itself or only a variety of some other species. Its value to him as a guide, while tracing out the geographical distribution of these Rocks, is not at all affected by the zoological question. In demonstrating the physical structure of the country, he can reason upon varieties with as much safety as upon true species, provided that such varieties are confined to and consequently characteristic of particular portions of the geological edifice. It is therefore, of great importance, not only to ascertain to what particular level each variety is confined, but also to determine whether or not such species as range through several formations exhibit any and what change in form on passing from one group of rocks to another. Should it be hereafter, proved that the supposed species above quoted, constitute on purely zoological grounds, but one extensive and variable species still it would be convenient for geological purposes to have a separate name for each variety that can be shewn to be characteristic of a particular geological horizon.

Before entering upon the description of the new species, I shall give a general account of such characters as are common to all the forms of the group typified by S. alternata.

STROPHOMENA ALTERNATA, (Conrad.)

In all the forms of this important type, the convexity of the ventral valve has a peculiar contour which may be seen not only in the Lower Silurian but also in the Upper Silurian and even in such Devonian species as S. Pattersoni, S. inequiradiata, S. demissa, S. concava and others. That part of the valve which is usually called the visceral disc occupies all the central region of the shell and terminates in a point at the beak. Just in front of the beak it forms a more or less well defined low rounded umbo on each side of which there is a flattened or sub-concave depression extending obliquely outwards to the margin just in front of the cardinal angles. These latter are usually reflected or a little curved upwards from the plane of the lateral margins. The visceral disc is somewhat flattened, gently convex or only slightly elevated throughout the greater part of its extent. In the upper half of the shell it is bounded by the depressions that have been mentioned as existing between the umbo and cardinal. angles, but in front and at the sides it terminates where the shell.

begins to be bent down to form the deflected margin which runs all round the edge and becomes obsolete on approaching the cardinal angles. This margin varies in width from one-twelfth to two-thirds the whole length of the shell and therefore the disc in some of the varieties occupies nearly the whole superficies of the valve, bnt in others, less than half. In the very young shells. in most of the specimens that I have seen, there is no deflected margin and occasionally adult individuals may be found, which on a side view give the outline of an uniform flattened arch from beak to front. In by far the greater number of the specimens. however the deflected margin is well defined. The contour of the front of the visceral disc varies according to the form of the deflected margin and is thus either broadly rounded or more or less pointed. In S. depressa, which also belongs to this group, the front of the disc, and its sides also, are often nearly straight.

The dorsal valve is flat or only gently concave beneath the visceral disc of the ventral valve, but all round, its curvature conforms to that of the deflected margin.

In the true S. alternata the areas of the ventral and dorsal valves are inclined towards each o her at an angle varying from 75° to 80°, but this angle never amounts to 90°. It will be observed that in some of the new species hereinafter described it is greater than 90°.

The surface in most of the species exhibits two sets of radiating striæ, the larger of which are about one-twelfth of a line wide in large specimens, and the smaller half that size, from one to ten of the smaller between each two of the larger, the more common numbers being from three to five. Sometimes also the shell is marked with a series of concentric wrinkles.

Of the above characters, those which are corfined to the upper half of the shell such as the form of the beak, the umbo, the concave depressions or hollows on each side of the umbo, and the reflected cardinal extremities, are common to all the species and in order to avoid repetition will not be particularly dwelt upon in the following descriptions. The radiating striæ are also very constant in the aspect they present. The only parts which appear to afford permanent variations of much value are the front of the visceral disc the deflected margin and the hinge line. The proportional length and breadth of the shell seems also to be of much importance especially if accompanied by a variation in two or three of the other characters.

Fig. 1.-Strophomena nitens. a is a section shewing the curvature and obtuse angles formed by the inclination of the areas.

Description.-Transversely semi-oval, sides somewhat straight for one third or a little more of the length from the cardinal angles, and slightly converging towards each other; front angles broadly rounded; front margin gently convex or nearly straight for about one third the width in the middle portion. Widh on hinge-line from nine to twelve lines. Length from six to eight lines.

The beak, umbo, depressions on each side of the umbo and the cardinal angles of the ventral valve the same as in S. alternata. The deflected margin forms an angle of between 1000 and 1100 with the general plane of the visceral dise, and occupies on the median line (in all the specimens I have seen) from one third to nearly one half the whole length of the shell.

The dorsal valve is quite flat, or even a little concave, just in front of the beak, but elsewhere curved to correspond with the ventral valve.

The area of the ventral valve lies nearly in the plane of the lateral margins, and the area of the dorsal valve forms with it an angle of about 95o. The height of the area of the ventral valve at the foramen is three fourths of a line in a specimen nine lines wide, and of the dorsal valve about one third of a line. Foramen of ventral valve partly closed by a V-shaped deltidium, the lower open part of which is closed by the strongly projecting deltidium of the dorsal valve.

The width of the foramen is about equal to its height.

The surface is the same as in S. alternata, and, when a little worn, presents a smooth shining silken lustre.

When compared with S. incrassata, S. alternata, S. deltoidea, S. camerata, or S. tenuistriata, it will be seen that this species is shorter in proportion to the width than any of them, and also that the inclination of the areas towards each other differs in forming an obtuse instead of an acute angle.

Locality and Formation.-This species occurs at Charleton Point, Anticosti, in the upper part of the Hudson River group. Collector.-J. Richardson.

STROPHOMENA CERES. N. s.

Description.-Semi-oval, sides rather straight and a little converging for one third their length; front angles and margins broadly rounded. Width on hinge-line twelve to fifteen lines; length ten to twelve lines.

The ventral valve varies greatly in the amount of its convexity. In some specimens it is depressed convex, and these have almost precisely the aspect of the more flattened forms of S. alternata. Others are strongly convex, nearly hemispherical, uniformly arched from beak to front, no deflected margin distinct from the visceral disc, the latter occupying the whole of the shell except a small triangular space at the hinge-angles. Between these two extremes there are individuals which present all the intermediate degrees of convexity, and some in which the deflected margin. can be detected with a width equal to half the whole length of the shell.

The surface is the same as that of S. alternata.

The area of the ventral valve is one line high in a specimen fourteen lines wide, and lies very nearly in the plane of the lateral margin. The foramen is as wide as high, and closed by a strongly convex deltidium, the lower margin of which is concave to admit the equally convex deltidium of the dorsal valve, whose area is almost half a line wide and forms an obtuse angle of between 90° and 1000 with that of the ventral valve. The beak of the ventral valve exhibits in some specimens a small round perforation.

This species differs from S. nitens in being in general a little longer proportionally, larger, and more uniformly convex, with scarcely a distinct deflected margin. In S. nitens the length is in general only two thirds of the width, but in this species it is always over five sixths.

The angle formed by the inclination of the areas being obtuse instead of acute furnishes the only character as far as I can ascertain by which it can be separated from S. alternata.

Locality and Formation.-Charleton Point, Hudson River group, and also at East Point in the Middle Silurian, Anticosti. Collector.-J. Richardson.