The species of vertebrates found along Peace River in the vicinity of Arcadia and their geological age are discussed on pages 380–381.
6. _Palmetto, Manatee County._—From Mr. J. C. Hennessy, of Palmetto, the U. S. National Museum has received a part of a lower left hindermost molar of _Elephas imperator_, found by him on January 10, 1917, on the north shore of Manatee River, within the corporate limits of Palmetto. The specimen presents seven ridge-plates and part of an eighth. Portions of the tooth are missing from both ends. The distance across five plates is 106 mm. The width across the worn face is 100 mm., the height of the hindermost plate present 150 mm. The enamel is strongly plicated. The tooth certainly belongs to _Elephas imperator_. The whole length of the tooth in its complete state was about 360 mm. Had it belonged to _E. columbi_, with 24 plates, the length would have been about 480 mm. (19 inches).
1. _Bogue Chitto, Dallas County._—In the U. S. National Museum is a lower left molar which belongs to this species. It was collected by Lawrence Johnson, of the U. S. Geological Survey. It is worn down to the base in front and some plates have thus disappeared. Parts of seven plates and the hinder talon remain. The width of the grinding-face is 90 mm. At the third plate from the rear the height of the crown is 97 mm. The hinder border of the tooth is obtusely keeled and there are no indications that there was another tooth behind it. It seems necessary, therefore, to regard it as the hindermost molar. The large hinder root was developed, but hollow to contain the pulp. The anterior root is entirely missing. The plates of the crown turn backward strongly. Of these plates there are on the inner face of the tooth hardly four in a 100–mm. line; on the outer face, only four. The enamel is rather strongly folded and of moderate thickness.
With this tooth there came from the same place a molar of _Equus leidyi_ and some fragments of teeth of _Mammut americanum_. The writer believes that these species show the presence, along Bogue Chitto, of Pleistocene deposits of about Aftonian age.
2. “_Near Gulf of Mexico._”—J. C. Warren, in the second edition of his work, “The _Mastodon giganteus_ of North America,” 1855, page 162, plate XXVIII, figure A, described and figured a part of a large upper molar, probably the hindermost, of an elephant which, as the writer believes, belongs to _Elephas imperator_. Warren stated merely that this tooth had been found in Alabama, near the Gulf of Mexico. He regarded the tooth as belonging to _Elephas primigenius_ and representing a form with extremely thick plates. Falconer (Palæont. Mem., vol. I, p. 227) described the tooth with somewhat more accuracy than did Warren, although he had only a cast of the tooth. He stated that the specimen presented the middle portion of an enormous last upper molar of the right side. This tooth had lost part of the front by wear and the rear by fracture. There were preserved eight complete ridges and a half of another in front. Falconer said that it bore a close resemblance to the Bollaert tooth found at San Filipe, in Texas, a tooth described in The Geologist, of London, in 1861, 1862, volumes IV and V. He gave the length of the fragment, measured at the base, as 7 inches; the length of the eight hinder ridges, at the base, 6.6 inches; the width of the crown at the third ridge, 4.6 inches; the greatest width behind, 4.9 inches; the height of the last ridge, 8 inches. The average thickness of the plates, including the cement, was 0.8 inch. Warren’s figure shows that the enamel is well crimped. Falconer referred the tooth, with some doubt, to _Elephas columbi_, but he was not well acquainted with _E. imperator_. The present writer believes that the tooth belongs to the last species named. It is now in the American Museum of Natural History, New York. The width of the grinding-surface is 110 mm. There are 5 plates in a 100–mm. line. The plates are not curved. The enamel is thick and festooned.