As the subject stands at present, then, we have no right to infer from the presence of an Orbitolite however abundant, that the stratum in which it occurs belongs to one period more than another between the commencement of the cretaceous epoch and our own times.*

In my former paper, I endeavored to point out the cause of the obscurity in which the true age of the Orbitolite limestone of Alabama had been involved, it having been considered sometimes as an upper cretaceous group, and at others as intermediate between the cretaceous and the eocene formations. The accompanying section from Claiborne bluff to Bettis's Hill, near Macon in Alabama, may serve to explain the relations which I found to exist between the white limestone group of the south, comprising the successive formations 1, 2, 3, and the overlying group 4, which is perhaps of equal thickness but which from the absence of calcareous matter rarely yields organic remains, and those consisting only of silicified casts of shells and corals. This upper formation (4) is composed of variously colored red, pink and white sands, and of yellow ochre-colored sands, white quartzose gravel and sand with beds of chert and flint, blood-red and pink clays and clays of white kaolin or porcelain earth all horizontally stratified. I could find no fossils in those in Alabama, and only conjectured that they are of eocene date from the analogy of Georgia, where a deposit of the like aspect and nature and occupying a similar position contains eocene shells and corals. I formerly explained in 1841-42, the relative position of the upper clay and sands with flint of the burrstone formation of Georgia to the underlying white limestone and marl of the state of South Carolina, in a diagram published in the Journal of the Geol. Soc., vol. i, p. 438, where the newer group is represented as resting on the eocene limestone of Jacksonborough, near the Savannah river. It appeared in that case as in Alabama, that the older calcareous strata of limestone and marl, had undergone great denudation and had acquired a very uneven surface, having been shaped into hills and valleys, before the incumbent clays and sands were thrown down.

At the bluff on the Alabama River at Claiborne, where so rich a harvest of fossils has been obtained, especially in some of the beds of No. 2, we see at one spot called "The old landing," that nearly the whole precipice in its lower one hundred and sixty feet, exposes to view the calcareous beds 1 and 2, covered with about twenty feet of red clay and sand. Whereas at the distance of less than a mile from this spot, the upper formation No. 4, occupies more than one hundred feet of the face of the same cliff from its summit, while at the base the lower members of the cal

"The Plagiostoma dumosum of Morton, is decidedly a Spondylus."

careous series crop out from beneath the horizontal and incumbent beds of sand and clay. This twofold composition of the mass of strata in the bluff at Claiborne, is expressed at A, in the annexed wood-cut, (fig. 1,) and I verified a similar mode of juxtaposition

Clai Alabama borne. River.

Fig. 1.

Clarke County.

Bettis's Hill.

2

1. Eocene sand, marl, &c., with numerous fossils.-2. White or rotten limestone; Zeuglodon, nautilus, &c., eocene.-3. Orbitolite limestone, eocene.4. Overlying sand, clay, &c., eocene.

of the two series of beds in several places in the interior of Clarke County, where the limestone often ends abruptly and is succeeded sometimes in the same ridge or hill by the newer beds, (No. 4.) the latter having evidently filled up the inequalities of a previously denuded deposit, after which the whole was again denuded.

I have suppressed several details and repetitions of the same phenomena in the country represented in the above diagram, (fig. 1,) and have been obliged to give a considerable inclination to the strata, because in the distance of twelve or more miles between Claiborne and Bettis's Hill, although the dip is not perceptible to the eye, the same beds are at the latter place more than twice as high above the Alabama river as at Claiborne. The mass No. 1, about one hundred feet thick at Claiborne, which constitutes the lowest visible member of the eocene series in this region, comprises marly beds with Astrea sellaformis seen at the base of the cliff at Claiborne, and an argillaceous stratum with impressions of leaves and sandy beds, with marine shells among which are found Cardita alta and Cardita planicosta, Cardita parva, Crassatella prætexta, Cytherea æquorea, Oliva Alabamensis, Pleurotoma (several species), Solarium canalicaletum, Crepidula lyrata, Endopachys alatum, Lonsdale, and two hundred other species. No. 2, about fifty feet thick, is the white or rotten limestone, which is sometimes soft and argillaceous, but in parts very compact and calcareous, and contains Flabellum cuneiforme, Lonsdale, Scutella Lyelli, Conrad, Lunulites, and several shells, some peculiar, others common to the formation below. Mr. Conrad has already described this section at Claiborne, and I hope soon to give a fuller notice of it with the observations which I made there in 1846. Of the limestone, No. 2, only the lower portion is seen here, for it

is cut off at the top of the bluff by the newer series of beds No. 4, but in many parts of Clarke County, as near Bettis's Hill and near Clarksville, the same No. 2 is found more largely developed. It is characterized among other organic remains by a large Nautilus allied to N. ziczac and by the gigantic zeuglodon of Owen. Near the juncture of the mass with the incumbent Orbitolite limestone, we find Spondylus dumosus, (Plagiostoma dumosum, Morton,) Pecten Poulsoni, Pecten perplanus, Ostrea cretacea, in abundance. No. 3 is a pure limestone, sometimes hard and full of Orbitolites Mantelli. At Bettis's Hill the formation is about sixty feet thick, and the upper beds are composed of a cream-colored soft stone, which hardens on exposure to the air, is not divided by lines of stratification, and is for the most part made up of Orbitolites of various sizes, with occasionally a Lunulite and other small corals, with specimens of Pecten Poulsoni. The origin of this limestone like that of our white chalk, the softer varieties of which it much resembles, is I believe due to the decomposition of corals, and like our chalk downs, the surface of the country where it prevails is sometimes marked by the absence of wood, by which all the other deposits in this part of Alabama are continuously covered. The spots where few or no trees appear are called "bald prairies," but in some places, and at Bettis's Hill among others, the Orbitolite rock produces what is termed a "cedar knoll;" the red cedar, Juniperus virginiana having exclusive possession of the ground. I was much struck with the resemblance of such calcareous tracks, covered with the trees above mentioned, to certain chalk regions in the south of England, where the only wood which grows on the white rock consists of yew trees accompanied here and there by shrubs of juniper.

At St. Stephens, on the left bank of the Tombeckbee river, in Alabama, a similar limestone with Orbitolites forms a perpendicular bluff. The water of the river at the time of my visit was too high to enable me to collect fossils from the older beds at the base of the cliff, but I was afterwards furnished with them through the kindness of Prof. Brumby of Tuscaloosa. They are imbedded in a ferruginous ochreous-colored sand, and consist in part of shells common to Claiborne bluff, such as Terebra costata, Conrad, Cardita parva, Dentalium thalloides, Flabellum cuneiforme, Lonsdale, Scutella Lyelli, Con., and several more.

I shall now conclude by adverting briefly to the result of a comparison which I made of the fossils contained in the eocene strata of Vicksburg, on the left bank of the Mississippi river, the position of which is indicated at 4a in the wood-cut on the next page, with those of other eocene beds forty-five miles further inland or eastward at Jackson, in the same state. In the former of these

at 4a, the Orbitolites Mantelli abounds, together with Pecten Poulsoni, Dentalium thalloides, Sigaretus arctatus, Con., Terebra

costata, Con., and a few others common to Claiborne; but the great bulk of the associate fossils do not agree specifically with those of Claiborne bluff. I found these distinct species at Vicksburg to be referable to the genera Voluta, Conus, Terebra, Fusus, Murex, Cassis, Pleurotoma, Oliva, Solarium, Natica, Turritella, Corbula, Panopæa, Crassatella, Lucina, Venus, Cardium, Arca, Pinna, Pecten, and Ostrea, with several corals, the whole having a decidedly tertiary and eocene aspect. The genus Pleurotoma, for example, which is represented by several species, is one of the forms most characteristic of tertiary as distinguished from secondary formations.

1. Mud of alluvial plain of Mississippi.-2. Superficial drift.-3. Freshwater loam with land shells.-4. Eocene strata. – 5. Cretaceous strata -Length of section fifty miles.

At Jackson, which, as before stated, is more than forty miles to the eastward, eocene beds, older than those of Vicksburg, crop out near to the area occupied by cretaceous deposits, as at 4b, wood-cut, (No. 2.) Here on the Pearl river I found no specimens of Orbitolites Mantelli, although some are said to have been met with in the vicinity. But I observed that a larger proportion among the fossils, were specimens common to Claiborne bluff, than at Vicksburg. Among these may be mentioned Cardita planicostata, Cardita rotunda, Cytherea æquorea, Natica, like one which I obtained at Claiborne, Flabellum cuneiforme, Lonsdale, and Endopachys alatum, Lonsd., (Turbinolia Maclurii of Lea.) These I found in strata of yellow loam, sand, and marl, on the Pearl river and in the banks and bed of one of its tributaries. The other shells collected by me at the same place, several of them I believe identical with Claiborne species, belong to the genera Voluta, Oliva, Terebra, Rostellaria, Murex, Plurotoma, Umbrella, Natica, Turritella, Crepidula, Dentalium, Corbula, Mactra, Lucina, Cytherea, Cardium, Cardita, Pectunculus, Nucula, Pinna, Pecten, and Ostrea. With these are corals, teeth of fish, &c. I was shown the remains of a Zeuglodon procured from the neighborhood, at a place five miles south of Jackson, on the right bank of the Pearl river, but as I did not visit the locality I cannot point out the precise place in the eocene series which it occupies. Some of the accompanying corals, however, were the same specifically as those occurring with the shells above mentioned at Jackson, and one of my informants stated that this Zeuglodon bed was immediately under "the rotten limestone." London, June 18, 1847.

ART. XVI.-Notice of some recent Additions to our Knowledge of the Magnetism of the United States and its Vicinity; by ELIAS LOOMIS, Professor of Mathematics and Natural Philosophy in the University of the City of New York.

In the forty-third volume of this Journal I have given a discussion of all the observations of magnetic dip in the United States with which I was then acquainted; and in the fortyseventh volume I have given the comparison of these observations with Gauss's theory. Quite recently we have received a most important addition to our stock of observations, from Major Graham, Prof. Locke, and Capt. Lefroy, published in the Transactions of the American Philosophical Society and the Royal Society of London. In order to make the American Journal a complete depository of our knowledge upon this subject, it is proposed to present an abstract of these new observations accompanied by a few remarks.

The observations of Major Graham were made in the years 1840, 41, 43, '44, and '45. Those of 1840 were made with a dipping needle constructed by Troughton and Simms, having a vertical circle eight inches in diameter, graduated to fifteen minutes. The azimuth circle reads to single minutes by the aid of a vernier. The observations subsequent to 1840 were made with an instrument by Gambey, having a vertical circle ten inches in diameter, and graduated to ten minutes. In conducting the observations, the usual precautions were observed of reversing the poles of the needles, and reading with face alternately east and west. The following is a summary of Major Graham's ob