rather large, the lateral line is so high near the back, that it is not covered by it anteriorly. The general color is of a reddish brown, mottled with red above and passing gradually into a uniform bright brick-red color prevailing upon the lower part of the body, and sprinkled with irregular light dots.

2. Pomotis inscriptus, Agass.-Small species, the outline of which is more elongated than in P. sanguinolentus. The gill covers are marked as in that species with three or four lines of a metallic steel blue color; opercular appendage long, directed more obliquely upwards than in any other species here described, black, with a light border which is a continuation of two of the lines of the cheeks, the one running below the eye, the other terminating behind the eye. Each scale of the back and sides is marked in its centre with a short narrow black line, hence the sides are regularly striped with dark interrupted lines as numerous as the rows of scales. Spinous rays all comparatively long and slender; the passage from the anterior to the posterior part of the dorsal gradual. All the fins except the pectorals are tinged with black at the extremity. General color dark olive above, lighter beneath.

3. Pomotis notatus, Agass.-Called Pond Perch at Huntsville. Body more elongated than in P. vulgaris; its upper and lower curve nearly equal. Opercular appendage very short, not extending beyond the base of the pectorals; its hinder margin is orange-colored, with a black spot in front, from which a faint dusky band extends to the eye. The spinous rays of the dorsal and anal are more slender than in P. vulgaris, and the articulated rays are crossed by fewer dotted or broken dark lines. The pectoral fins are long, extending beyond the base of the anal, as in Ichth. macrochirus, Raf. The color is of a uniform light olive; the sides, gill cover and belly being silvery; scales not dotted with black as in many similar species.

4. Pomotis incisor, Val.-Also called Pond Perch at Huntsville. This species resembles very closely the preceding, and is considered the same by the fishermen; but its profile is more arched and slants more abruptly; the black opercular appendage is not encircled with a brighter margin. Sides of the head not banded, but of a uniform color throughout. Dorsal and anal not banded, but darker colored than in the preceding species. There is moreover a dark black spot near the base of the hind rays of the dorsal in P. incisor which is wanting in P. notatus. General color of the body the same in the two species.

5. Pomotis obscurus, Agass.-Also called Pond Perch at Huntsville. Resembles P. incisor in the outline of the body, except that the profile is still more precipitate and the body somewhat more elongated as well as much stouter, especially in the region

of the head and across the pectorals. The opercular appendage is longer and broader, but also without a light posterior margin. The posterior soft rays of the dorsal are marked with a black spot as in the preceding species, but all the spinous rays of that fin are shorter and stouter. It is a dark colored fish throughout the lower as well as the upper side of the body, almost uniformly brown, the belly only being somewhat lighter in hue. The face and lower jaw are of a leaden color. The fins are all darker than in P. incisor, especially the ventrals.

6. Pomotis bombifrons, Agass.—Body higher than in P. obsenrus and profile even more arched. Forehead prominent especially over the eyes. Head quite broad and short. Opercular appendage black, and small; a light narrow band runs along its lower margin. No black spot upon the hind part of the dorsal. The last spinous rays of this fin are shorter than in P. obscurus, thus making the passage to the soft rays more abrupt and marked, the soft portion of the fin being almost as prominent as in Ambloplites and Calliurus when compared with the spinous rays. Body light brown, fins lighter colored; scales of the belly and sides dotted with golden orange. The face and under jaw have not the leaden color of P. obscurus. Considering the peculiar form of the vertical fins and of the forehead, it may become necessary to separate this species from the other Pomotis. Indeed, I know already several species which agree in these respects with one another and must at all events form a distinct group in the genus.

7. Pomotis pallidus, Agass.-This species resembles P. incisor in the outline of the body, the nature and coloration of the scales, and in the size and form of the fius, but it differs greatly from it by its large mouth, the free extremity of the upper jaws reaching the vertical line of the middle of the eye, by the presence of teeth upon the palate, and by the ventral fins being placed immediately under the pectorals. The black opercular appendage which is very short, has a narrow orange border behind. There is a black spot at the base of the posterior rays of the dorsal. Both dorsal and anal are marked by one or two dark stripes; the caudal is crossed by several dotted vertical lines. There are eight or nine dusky bars across the sides, between the head and tail. This species bears the same relation to Pomotis, that Pomoxis bears to the true Centrarchus, in the size of the mouth, and the form of the body, and I have no doubt it will some day become the type of a distinct genus.

ETHEOSTOMOIDS, Agass.-There are comparatively few natural families in the animal kingdom so limited in their geographical distribution as to be entirely circumscribed within the boundaries of a single continent, and these few belong mostly to the type of Vertebrata. Though among fishes we should least ex

pect such local groups, considering the greater uniformity of the conditions of existence prevailing in the medium they inhabit, when compared with the main land, yet there are several families of this class, the geographical range of which is quite limited. I need only mention the Goniodonts of South America, the Labyrinthici of the Indian Ocean and the Sunda Islands, the Lepidostei of North America, &c. Another natural family thus located within narrow limits is that of Etheostomoids, which I have for the first time characterized in my work "Lake Superior,” p. 298. This family is founded upon the genus Etheostoma of Rafinesque, to which are added the genera Pilcoma and Boleosoma of Dekay (of which the genus Percina of Haldeman is a synonym) and my genus Pacilichthys. The three first of these genera were referred by their authors to the family of Percoids; but the absence of an air-bladder and of pseudobranchia, and the incomplete suborbital arch precludes such an association. Indeed these fishes are more closely allied to the true Cottoids and in particular to the genus Gasterostens than to the Percoids, though the want of connection between the single suboperculum and the preoperculum forbids also a more intimate alliance with that family. The form of the ventrals of the Etheostomoids reminds us somewhat of those Gobioids in which the two ventrals are distinct. Since the publication of the work above mentioned, I have become acquainted with three new genera of this family, for which I would propose the names of Hyostoma, Catonotus, and Hadropterus.

The more extensive knowledge I have acquired of this family by these recent accessions enables me to give more precision to the characters assigned at first to its genera; as follows:

1. ETHEOSTOMA, Rafin.-Head elongated pointed; mouth terminal, widely open, not protractile, broad; jaws of equal length. Opercular apparatus and cheeks bare. First dorsal distinctly separated from the second. Anal and second dorsal smaller than the

*The genus Preilichthys was first mentioned under the name of Pacilosoma. Being however at the time of its publication far away from Cambridge, and unable to consult my library or any other, I did not perceive that that name was already preoccupied; I would therefore change it now to Pocilichthys. Several new species of this genus have been discovered since. One described by Dr. Kirtland as P. erythrogaster from the vicinity of Cleveland, Ohio. Ann. of Sci., Jan., 1854, p. 4. Another collected by Mr. Geo. Stolley in the Osage River, Mo., remarkable for its brilliant colors, the body being light brown, with dark black lines upon the sides of the back and with broader transverse bands alternately black and orange red, especially bright upon the sides of the tail; dorsals banded with black, white and red. I call this species P. spectabilis. Another found by Dr. L. Watson in small creeks near Quincy, Illinois, similar in color to the preceding but without black stripes along the back, also less compressed. I call this species P. versicolor. Specimens of this species were also received from Osage River. A fourth species from the Osage River, Mo also discovered by Mr. Geo. Stolley, is of a greenish color mottled with black, the second dorsal, the caudal, the anal, the ventrals and the pectorals being dotted all over with minute dark specks. I call this species P. punctulatus.

first dorsal, but equal to one another. Caudal lunate. Type of the genus: Ethblennioides, Raf.

2. CATONOTUS, Agass.-Head elongated, obtuse; mouth terminal, widely open, not protractile, lower jaw longer than the upper. Opercular apparatus, cheeks and neck destitue of scales. First dorsal much lower than the second, with clubshaped rays when full grown; membrane of this fin extending to the base of the second dorsal. Anal smaller than the second dorsal. Caudal rounded. Only one species known: C. lineolatus, Agass., discovered by Dr. L. Watson in small creeks near Quincy, Ill. The whole body olive green with close narrow interrupted black longitudinal lines; transverse lines of the same color across the caudal.

3. PILEOMA, Dekay.-Head conical, pointed, truncated at the end, in form of a hog's snout; mouth moderate, in form of an oblique arc of a circle, opening below the end of the snout, very slightly protractile. Lower jaw shorter than the upper. Operculum and cheeks scaly. Membrane of the first dorsal not reaching the base of the second. Anal smaller than the second dorsal. Caudal truncate or slightly lunate. Type of the genus: P. caprodes. (Etheostoma caprodes, Rafin.)

4. HADROPTERUS, Agass.-Head conical, obtusely pointed, rounded at the end; mouth moderate, terminal, not protractile, jaws nearly equal. Operculum and cheeks scaly. Membrane of the first dorsal extending to the base of the second. Anal and second dorsal large and equal. Caudal truncate or slightly lunate. Only one species known: H. nigrofasciatus, Agass. From the neighborhood of Mobile, Alabama. Discovered by Albert Stein, Esq. Brown above, lighter below, with transverse black bands, wider in the middle than nearer to the back or the belly.

5. HYOSTOMA, Agass.-Head short, blunt, rounded, with swollen cheeks. Mouth comparatively small below the snout, slightly protractile. Lower jaw shorter than the upper, which may be concealed in a deep furrow below the snout. Opercular apparatus and cheeks scaly. First dorsal long, but not reaching the base of the second. Anal smaller than the second dorsal. Caudal slightly lunate. Only one species known: H. Newmanii, Agass. Discovered by Dr. Newman in the vicinity of Huntsville, Alabama, where it is called "Salmon." This fish is uniformly brown with irregular transverse black blotches. A red stripe along the base of the first dorsal.

6. PECILICHTHYS, Agass.-Head short and strong, tapering into a rounded snout. Mouth terminal, proportionally broad, not protractile, though the maxillary bone be moveable. Opercular apparatus scaly, cheeks bare. First dorsal distinctly separated from the second. Anal smaller than the second dorsal. Caudal truncate or slightly rounded. The species of this genus are among SECOND SERIES, Vol. XVII, No. 50.-March, 1854.

89

the most brilliant freshwater fishes in the world. Type of the genus: Etheostoma variatum, Kirtl. Several new species are mentioned in the note above.

7. BOLEOSOMA, Dekay.-Head short, rounded; mouth below the end of the snout, small, horizontal, slightly protractile. Opercular apparatus and cheeks scaly; neck scaleless. Membrane of the first dorsal reaching the base of the second, though the two fius are distinctly separated. Second dorsal much larger than the anal. Candal rounded. Type of the genus: Boleosoma tessellatum, Dekay. For references to other species, see “Lake Superior," page 299.

All the representatives of this family are confined, as far as we know, to the fresh waters of North America; not a single species having thus far been noticed either in Europe or Asia. To this circumstance we must no doubt ascribe the total neglect of the genus Etheostoma of Rafinesque by European ichthyologists.

The genus Hyostoma is the only type of this family I am acquainted with from the southern bend of the Tennessee River. It is true, Dr. Storer has described two species of Etheostoma from the vicinity of Florence, Alabama, but they do not seem to occur farther east; at least I have found nothing to remind me of his species in the collection forwarded by Dr. Newinan.

It is a fact worthy of notice that not a single species of Gasterosteus has as yet been discovered in the Mississippi River or its tributaries, or in any of the rivers emptying into the Gulf of Mexico. I have also searched in vain for them in the southern Atlantic states, though they are common in the northern states and in the waters emptying into the St. Lawrence.

SCLENOIDS, Cuv.-In the old world no representative of this family is known to inhabit the fresh waters, whilst in North America a remarkable species has been found in Lake Champlain, Lake Erie, Lake Ontario and the Ohio River, which truly belongs to this family and has generally been referred to the genus Corvina, under the name of Corvina Oscula. It should however be remarked that this species is but remotely allied to the genus Corvina and must in reality be considered as the type of a distinct genus, which has already been characterized, thirty-four years ago by that indefatigable naturalist, Rafinesque, under the name of Amblodon. Nobody has however thus far taken the trouble to examine the value of this genus, nor even to state on what ground it has been rejected by those who have incidentally noticed it as a synonym of Corvina. The truth is that Rafinesque was right in considering this Corvina Osenla as a distinct genus, the characters of which he has well defined, as may be seen by comparing his description in the Ichthyologia Ohiensis with that below. Moreover I have lately ascertained that there are several