56 Later Extinct Floras of North Atm rica, 



Planera Ungeri, Ettings. (Tert. flor. der (Estr. Monarch, p. 14, 

 Taf. ii. figs. 5-18). Ulm us Z Ikovcefolia. Ung. (Chloris Protogcea 

 Taf. xxiv. figs. 7-12, etc.), but it is apparently considerably 

 smaller, narrower, and more coarsely toothed. 



Formation and Locality. Miocene strata. Fort Union, 

 Dacotah. (Dr. Harden.) 



Catalpa crassifblia (n. sp.) 



Leaves large, fleshy, ovate, heart-shaped at base, pointed above, 

 sometimes unsymmetrical ; margins entire; nervation strongly de- 

 veloped ; midrib straight or flexuous; lateral nerves about 7 pairs; 

 lower pair strongest, not reaching the middle of the leaf, giving off 

 eacli about 4 branches on the lower side, of which the lower ones 

 spring from the base of the laterals and are much branched; upper 

 laterals branched at their summits, branches uniting to form a fes- 

 toon somewhat remote from the margin ; tertiary nervation 

 invisible. 



In its general aspect this leaf bears a marked resemblance to 

 those of the common Catalpa which grows spontaneously in 

 Kentucky and Tennessee, and is generally cultivated through- 

 out the Northern States. The leaves of the Catalpa are, how- 

 ever, broader, and the basilar pair of lateral nerves are stronger, 

 reaching to and sometimes above the middle of the margin. 

 The number of lateral nerves is also less, but they branch at 

 their summits, and form a marginal festoon very much in the 

 same way as in the fossil. 



Tin- haves of some species of FicU8^ of Gonolobu8, and Aris- 



tolochia exhibit considerable resemblance to this, bul scarcely 

 a- much as do those of Catalpa. In Gonolobus and Ari&tolo- 

 chia the nervation is lighter, more open, radiate and waved, 



while in the broad-leaved species of FtCUS {F. tilioBfolia, F. 



popidina, etc.), the basilar nerves are much more developed, 

 and the marginal festoou nearer the margin. In F. Morloti, 

 on the contrary, the basilar nerves are loss strong. Catalpa, 

 therefore, seems to offer the greatest resemblance to the fossil, 



and I place it provisionally in that genus. This is 80 COnspiCU- 



