176 PROCEEDINGS OF THE NATIONAL MUSEUM. vol.59. 



Order THYMELEALES. 

 Family LAURACEAE. 



Genus GOEPPERTIA Nees. 



GOEPPERTIA TERTIARIA, new species. 



Plate 25, fig. 1. 



Description. — Leaves of medium size, elliptical oval in general out- 

 line, widest below the middle, with an acute tip and a rounded base. 

 Margins entire, full and evenly rounded. Texture coriaceous. 

 Length, about 12 cm. Maximum width, about 5.75 cm. Petiole 

 missing. Primaries 3, supra basilar, all prominent on the lower 

 surface, midrib stoutest; the laterals diverge from the midrib about 

 5 mm. above its base at acute angles of about 25°, curving upward. 

 Secondaries comprise a few camptodrome pairs in the tip of the leaf, 

 several broadly curved ascending ones from the outer side of the 

 lateral primaries and an opposite pair from near the base of the 

 midrib. The tertiaries are thin and percurrent or inosculate midway 

 between primaries or secondaries and primaries. Aerolation forms 

 a fine polygonal lauraceous mesh. 



This leaf is of a type that has uniformly been referred to the genus 

 Cinnamomwn except in a few instances in Engelhardt's work upon 

 South American fossil plants. I know of no certain characters for 

 distinguishing certain forms of Goeppertia or Cryptocarya from 

 Cinnmnomum and am therefore inconsistent in not adhering to cus- 

 tom. It is a problem which paleobotanists will be obliged to face 

 sooner or later in connection with a great many fossil species which 

 have been referred to Cinnmnomum. I am influenced by the large 

 number of species of the almost exclusively tropical American tribe 

 Cryptocaryeae which have leaves of the Cinnamomum type; in fact, 

 the modern Cinnamomum campJiora Nees has leaves very much like 

 the present fossil species. It does not have the characters of Strychnos 

 nor of the many American Melastomataceae, but is very similar to 

 various species of Goeppertia, a genus with numerous species of tropical 

 America, to which region it is confined, and which is sometimes, as 

 by Pax in Engler and Prantl, made a subgenus of Aydcndron Nees. 

 With the exception of a species from the Chattian of Bohemia, the 

 only fossil forms that have been heretofore recognized comprise one 

 from Colombia and two from southern Chile, and all probably of 

 lower Miocene age. 



Holotype.— Cat. No. 35467, U.S.N.M. 



