AEECALES. 



177 



hvrfrnl BronQ;niart, the species to whicli 

 Seward and Arber refer all the nipa fruits from 

 the lower, middle, and upper Eocene of Bel- 

 gium. Though it is improbable that a single 

 species ranges throughout the I<]ocenc, and' 

 though I also regard it as improbable that the " 

 American and FAU'opean forms arc specifically 

 identical, in spite of their contemporaneity and 

 their distribution by ocean currents, no reliable 

 characters separate the American from the 

 European form. 



The occurrence of Nipadites in the early 

 Eocene of the Mississi])]ii embayment is of the 

 greatest interest, as it throws so much light on 

 the contemporaneous physical conditions. In 

 the existing flora the genus Nipa is monotypic 

 and stands in an isolated position among the 

 palms, formerly being placed with the family 

 Pandanacea=. The existing nipa palm is a 

 stemJess form, some of whose large puuiate 

 leaves attain 25 feet in length. It inhabits the 

 tidal waters of the Indian Ocean, ranging from 

 India through the Malay Archipelago to the 

 Philippines and vying with the mangroves for 

 possession of the tidal flats. It produces clus- 

 ters of large fruits which are distributed l)y 

 ocean currents. During the Eocene the closely 

 allied if not identical genus Nipadites is repre- 

 sented by the characteristic fruits in southern 

 England, Belgium, France, northern Italy, 

 southern Russia, and northern Egypt. Until 

 the present discovery in Mississippi neither 

 Nipa or Nipadites had been found in the living 

 or fossil floras of the Western Hemisphere. 



A characteristic view of the habit and the 

 habitat of the modern Nipa is shown on Plate 

 VII, A. 



The path of migration by which these palms 

 were introduced into the Eocene Gulf of Mexico 

 is worth considering. Smce their remains are 

 so Avidespread and common in the early Ter- 

 tiary deposits of the Mediterranean region, and 

 since they occur there earlier than in America, 

 it seems probable that they represent an mtro- 

 duced element in the Wilcox flora. Their 

 fruits may have reached this hemisphere by 

 floating across the Atlantic, which woukl not l)e 

 possible if the Atlantic Ocean currents of tlie 

 Eocene were at all similar to those of the ])rcs- 

 eut time. On the otlier hand, the more pro])a- 

 ])le hypothesis is that their range may ha\'(^ 

 covered Oceanica during the late Cretaceous 

 and early Eoc'cne, and they may have been 

 50243°— IG 12 



carried ])y ocean currents across the sub- 

 merged lauds of Central America and into the 

 Mississippi Gulf. 



Occurrence. — Grenada formation, Grenada, 

 Grenada County, Miss, (collected Ijy Lowe and 

 Berry ) . 



CnlJeefion.—V. S. National ^luseiim. 



Genus SABALITES Saporta. 



Sabalites tiRAVANUs Lcsciuercux. 

 Plates XII, (igures l-:i, and XIV, figure 1. 



S(il)iil (Irin/anii. Losquercux, Am. Philo.s. Soc. Trans., 

 vol. 13, p. 412, pi. 14, figs. 4-(), ISO!). (Not Lesque- 

 reux, 1871, 1874, 1876, 187S, or Knowlton, 1900.) 



Flahcllaria eocenica. Lesquereu.x'. Tlu' Tertiar}' flora, 

 p. Ill, pi. 13, figs. 1-3, 187S. 



Description. — Lesquereux's description, pub- 

 lished in 1869, is as follows: 



S. fronde petiolate, rachide in piano posteriore subplaiia, 

 e basi dilatata ovata, linear! cuspidata 6-8 policari; 

 foliis flabelliforniibus, radiis numerosis, elongati;;, sensim 

 dilatatis, nervulis distantibus. grar-ilimis. 



This somewhat protean species was de- 

 scribed by Lesquereux from the soft white 

 ("Eolignitic") clay of Lafayette County, Miss. 

 None of Lesquereux's specimens of it can at 

 present be found in the Hilgard collection, Init 

 it seems very probable that the type came from 

 the raih'oad cut just north of Oxford. Not only 

 is this the sole locality in the county known 

 to the writer where these white clays are fossilif- 

 erous but there are several specimens of this 

 species from this locality in the collections of 

 the Univei-sity of Mississippi, which have been 

 collected at different times by different indi- 

 viduals, and some of these fragments may pos- 

 sibly represent Lesquereux's type material. 

 There is also a specimen from this outcrop in 

 the collections of the LTnited States National 

 Museum, collected some L"} or 20 years ago by 

 Mr. T. O. Mabiy. 



Within the next 10 years after liis original 

 characterization of this species Lesc[uereux 

 idcntitied it from a large number of western lo- 

 calities in Colorado, Wyoming, and on Van- 

 couver Island. Nearly all this material is now 

 in the National Museum. In my judgment all 

 these determinations are open to very grave 

 doubt, not only on account of the inadequacy 

 of the material but also liecause of tlie a priori 

 improbability of a single species raugmg from 

 the Ci-etace(ms Montana group to a horizon 

 well a])ove the ])ase of the Eocene at such widely 



