have been found in it at some places; (2) it is the first calcareous bed of a series dominantly calcareous and succeeds noncalcareous sands; (3) at Willow Branch and one or two other places there is evidence strongly suggestive of unconformity between this bed and the underlying Gosport sand.

The other four divisions are the same as those described by Hopkins1o and need no further comment here. Sections illustrating them are given by Hopkins and Cooke in the papers cited.

The Jackson formation contains a large marine fauna. From Jackson, Mississippi, I have listed 200 species of mollusks and Vaughan has identified 12 species of corals; of these, about 49 are survivals from the Claiborne and about 15 are supposed to have lived also in Vicksburg time. Canu and Bassler list 67 species of Bryozoa from Jackson, of which 15 are known from the Claiborne group and 9 from the Vicksburg. The commonest and most significant vertebrate is Basilosaurus cetoides.

OCALA LIMESTONE

East of Tombigbee River a rather abrupt change is noticeable in the stratigraphy of the deposits of Jackson age. The beds become progressively more calcareous, lose their individuality, and assume more and more the lithologic and faunal aspects of the Ocala limestone of Florida. In some places a dual division of these beds may be distinguished, but it is not everywhere possible to draw a sharp line of demarkation between the upper and the lower members. The lower part consists chiefly of very argillaceous and somewhat glauconitic limestone, and on Sepulga River the approximate position of the yellow sand at Cocoa (division 3 of the generalized section) is occupied by calcareous sandstone. The upper part, corresponding to the "Zeuglodon bed" and the Yazoo clay, consists of soft, cream-colored, amorphous limestone which closely resembles the "chimney rock" of the overlying Marianna limestone.

10 Op. cit., 296.

11 CANU, FERDINAND, and BASSLER, R. S., Manuscript list of Eocene and Oligocene Cheilostome Bryozoa.

As the upper Eocene limestone of southeastern Alabama is continuous with the Ocala limestone of Florida and southwestern Georgia and does not differ materially from it in lithology or in fossils, the name Ocala limestone is extended to all of the deposits of Jackson age in that part of the state, but future more detailed field work may show the propriety of restricting the name Ocala to the upper part of the formation. Just where the boundary between the Ocala limestone and the Jackson formation should be drawn is a matter of expediency, for the transition area, although narrow, is without definite natural limits. Either the Tombigbee River or the 88th meridian might conveniently be selected.

VICKSBURG GROUP

In Mississippi the Vicksburg group falls naturally into three divisions, the upper, middle, and lower Vicksburg, which differ from one another in both lithology and fossils. The first of these, which corresponds to the "Higher Vicksburgian" of Meyer1 and to the "Upper Vicksburgian" of Casey, 13 is herein named Byram calcareous marl; for the second, which is approximately equivalent to the "Middle and Lower Vicksburgian" of Meyer and to the "Lower Vicksburgian" of Casey, the name Marianna limestone, already in use in Florida, is available; the third includes two facies, a shallow-water or nonmarine facies in western Mississippi, which will be called the Forest Hill sand, and a marine facies in eastern Mississippi and western Alabama known as the Red Bluff clay. In the middle division, or Marianna, two subdivisions are recognized, herein named Mint Spring calcareous marl member and Glendon14 limestone member. East of Clarke County, Alabama, the middle and lower Vicksburg are similar lithologically and are both included in the Marianna limestone.

12 MEYER, OTTO, Amer. Journ. Sci., 2d. ser., 30: 71. 1885.

13 CASEY, T. L., Philadelphia Acad. Nat. Sci. Proc. 53: 515. 1901.

14 The name Glendon limestone has been adopted, with my consent, by O. B. Hopkins (U. S. Geol. Survey Bull. 661-H. 1917) who had access to my notes and manuscripts.

FOREST HILL SAND

The name Forest Hill sand (from Forest Hill, 5 miles southwest of Jackson, Mississippi) replaces the "Madison sands" of Lowe,15 which name is preoccupied. The Forest Hill sand appears to rest conformably upon the Yazoo clay member of the Jackson formation. Although the character of the sediments indicates a change from marine to very shallow water or palustrine conditions at the close of Jackson time, it is probable that the change was gradual and that deposition was nearly continuous. The Forest Hill is overlain conformably by the Mint Spring marl member of the Marianna limestone. The relations of the Forest Hill to the Red Bluff clay are not definitely known, but it is believed that the two were formed contemporaneously, the latter having been deposited under more strictly marine conditions than the Forest Hill sand.

In the type area, the Forest Hill sand consists chiefly of crossbedded or laminated, more or less ferruginous, silicious sand and some clay.16 West of this area, the formation becomes more argillaceous and contains lenses of lignite and lignitic clay.

In Warren and southern Yazoo counties, the Forest Hill sand is estimated to be about 60 or 70 feet thick, and at Forest Hill it is between 50 and 60 feet thick.

Petrified wood, leaves, and other plant remains are common in the Forest Hill sand, but recognizable forms are not abundant. No animal remains have been found in the formation. The Forest Hill sand crops out along the bluff from Vicksburg northward to within a few miles of Satartia. Exposures are numerous in eastern Hinds County and in Rankin County as far east as Rankin. Outliers of the Vicksburg group in Madison County afford good exposures of the Forest Hill sand. Southeast of Rankin the country has not been explored in sufficient detail to determine the extent of the formation in that

15 LOWE, E. N., Op. cit., 82.

16 A section at Forest Hill School has been published by O. B. Hopkins (U. S. Geol. Survey Bull. 641-D: 100. 1916). I consider the lower 7 beds of his section as typical Forest Hill sand and refer the upper 6 beds to the Marianna limestone.

direction, but lignitic clays that are tentatively referred to the Forest Hill have been observed at a number of places in Smith County.

RED BLUFF CLAY

Fossiliferous deposits on Chickasawhay River at Red Bluff, 11⁄2 miles below Shubuta, were first noted by Harper17 in 1857. Three years later they were called the Red Bluff group by Hilgard, who correctly announced that their stratigraphic position lies between the Jackson and typical Vicksburg strata and that their fossils are more closely related to those of the Vicksburg than to those of the Jackson.18

Wherever the contact of the Red Bluff clay with the underlying Jackson has been observed, the two appear to be conformable. The upper limits of the formation are less well known, but there seems to be no break between it and the Marianna limestone. As the formation has not been traced west of Wayne County, its relations to the Forest Hill sand are conjectural, but it is believed that the two were approximately contemporaneous in origin and that the Red Bluff clay represents the marine equivalent of the exceedingly shallow water deposits of the Forest Hill sand in the Mississippi Embayment. The formation extends eastward into Alabama but rapidly thins, becomes calcareous, and merges laterally into the Marianna limestone.

The Red Bluff consists chiefly of stiff blue or greenish gypseous clay, but contains also discontinuous ledges of indurated marl or sandstone and a thin bed of shell marl. On Buccatunna Creek the formation is 70 feet thick.

The Red Bluff fauna includes more than 128 mollusks, 6 corals, and a considerable number of Bryozoa. Of the 134 species listed from Mississippi, about 60 appear to be restricted to the Red Bluff beds; about 55 are present in the Mint Spring marl or have varieties there; and about 49 species or varieties are known in the Byram marl, of these species 10 have not yet

17 HARPER, L., Preliminary Report on the Geology and Agriculture of the State of Mississippi, p. 142. 1857.

18 HILGARD, E. W., Report on the Geology and Agriculture of the State of Mississippi, p. 136. 1860.

been found in the Mint Spring marl. Twelve mollusks, 3 of which range through the Vicksburg group, are listed also from the Jackson formation, but some of these are characterless species of supposedly very long range.

MARIANNA LIMESTONE

The name Marianna limestone was given by Matson and Clapp19 in 1909 to the soft, porous, light-gray to white limestones at Marianna and other places in western Florida "which are characterized by an abundance of Orbitoides mantelli and other Foraminifera associated with many other fossils, prominent among which are Pecten poulsoni and P. perplanus."20 The last named species has since been found to be restricted to underlying Eocene strata" and was referred to the Marianna limestone by mistake.

The Marianna limestone was included in the Vicksburg group by Matson and Clapp, by whom it was regarded as the stratigraphic equivalent of the upper part of the bluff at Vicksburg (Byram marl). It was later found to lie conformably upon the Ocala limestone,22 which had been thought to be the highest formation of the Vicksburg group.

The typical Marianna limestone is very homogeneous, white or cream-colored, and when first quarried is so soft that it is easily sawed into building blocks which harden on exposure. Because of its extensive use for building chimneys, it is popularly known as "chimney rock." This facies of the Marianna limestone extends with remarkable uniformity from Marianna, Florida, nearly to Pearl River, Mississippi. It is characterized nearly everywhere by a great profusion of Bryozoa and an abundance of Lepidocyclina mantelli, Pecten poulsoni, and 19 MATSON, G. C., and CLAPP, F. G., A preliminary report on the geology of Florida: Second Ann. Rept. Florida Geol. Survey, p. 51. 1909.

20 Idem, 52.

21 Not having seen the type of Pecten perplanus, I am accepting as correct the species so named in the collection of the U. S. National Museum and described by Dr. Dall (Tertiary Fauna of Florida, p. 732). Hopkins has figured a specimen

on plate 27, in U. S. Geol. Survey Bull. 661-H.

22 COOKE, C. W., The age of the Ocala limestone. U. S. Geol. Survey Prof. Paper 95: 109. 1915.