DESCUIPTION OF THE SPECIES. 293 



widest, iuul then gnidually narrowed to the petiole, subacute at the tip, 

 where the margins for some distance down are generally straight; midrih 

 comparatively very thick, and apparently somewhat succulent; petiole 

 somewhat succulent and very thick ; the leaves probably pinnately ar- 

 ranged ; primary nerves very slender in proportion to the size of the mid- 

 rib, departing at an angle of about 45°, arching around to connect with 

 those next above, and sending oti' branches mostly on the outer side, form- 

 ing by their union several series of arches between the midrib and the mar- 

 gin; between the main primaries subordinate ones si)ring from the midril), 

 sending off branches to unite with tlie main primaries; prinuirv net-work 

 of nerves formed of the principal branches of the primaries meeting nearly 

 at a right angle, composed of large, lax, obtuse-angled meshes, four to six 

 sided; ultimate net-work composed of nerves of uniforudy equal strength, 

 meshes nearly of the same size, obtuse-angled, and generally four-sided. 



Localities: Road-side near Potomac Hun, very rare; Fredericksburg; 

 common. 



The petiole of this plant is, in proportion to the primary nervation, 

 even more massive than that of F. crasstnervc. The primary nerves are 

 nt)tably slender, and their branches meet more nearly at a right angle, 

 while the series of arches running parallel with the mai'gin are more num- 

 erous than in F. crassinrm'. Plate CXLI, P^ig. 2, indicates that some ot 

 the leaves must have attained dimensions rivaling those of F. crassincrve. 

 This shows well also the straight margins towards the tips of the leaves, 

 which are caused here to have a triangular outline. On the other hand, 

 the small leaf, PI. CXLIX, Fig. 3, gives a degree of smallness never found 

 in F. cruss'iHcrve. The leaves in Fig. o of the same plate occur in such a 

 position as to suggest they were attached in pairs or piimately. The upper 

 leaf, on its left-hand basal margin, has the lamina torn loose and pressed 

 over on the midrib, while the similarly situated portion of the lower leaf is 

 crushed back and doul)led over u])on itself. This figure shows a great 

 thickness of petiole, but it was evidently somewhat succulent, as it is 

 wrinkled and slightl}' puckered. Plate CXLV, Fig. 4, gives a restoration 

 made out from a large number of specimens which give all parts of the leaf, 

 so that it may be relied upon as giving a pretty accurate idea of it. 



