NO. 187:;. CHARACTERS OF GIGANT0PTERI8— WHITE. 499 



in many large leaves in which the tooth is prolonged into an elongated 

 lobe. The thick basal portion of the secondary nerve of such an 

 enlarged lobe, broken away in exposing the young leaves, is seen on 

 the lower left of the leaf fragment shown on the right in the figure. 

 The downward contmuation of the lamina, as shown in the smallest 

 leaf, and its lobation may have to do, I beheve, with the fructifica- 

 tion of the plant, as will presently be explained 



None of the specimens yet gathered present conclusive evidence as 

 to wliether the fronds of Gigantopteris, seen essentially in miniature 

 m plate 46, figure 3, were borne on an aerial axis or stem, an elongat- 

 ing rhizome, or in tufts. The general form of the frond and the ar- 

 rangement of the leaves in the specimen suggest, however, that they 

 may have been developed along a prostrate stem or rhizome. 



Seeds. — At every locality where Gigantopteris has been found in 

 America there are associated with it numerous flat cordiform, alate 

 seeds at first glance suggestmg Cardiocarpon. On closer examination 

 the seed, which is small and obovate, is found to lie in the slightly 

 concave face of a very broadly round-obovate, slightly cuneate and 

 asymmetrical bract, as shown in plate 49, figure 6. The latter has 

 in every case a narrow keel, or rib, along one of its bcnxlers, which 

 runs down to its point of attachment, as illustrated by plate 49, 

 figure 4. From the convex side the seed appears somewhat thickly 

 covered by the wing or bract substance so that it is somewhat dimly 

 outlined (plate 49, figures 3, 4, and 6) ; but on the concave surface it is 

 fairly clearly defined, apparently lying beneath a thin envelope, which, 

 in a few cases, as in the specimen shown in figure 1, is scaled off, 

 revealing the seed itself, which appears to have a thin, hard test like 

 that of Cardiocarpon. At the broader end of the seed, which is 

 slightly apiculate, a depression or small zonulate configuration, usually 

 accompanied by a slight gathering or puckering of the "wing," may 

 often be noted. This, which might casually be mistaken for a chalaza, 

 appears to lack the hardness, fiber, and all other evidence of attach- 

 ment to an axis. It lias the appearance of a collapsed pollen chamber, 

 the puckered wing being susceptible of interpretation as a micro pylar 

 environment. Further, it is to be noted that a forldng vascular sys- 

 tem radiates from the base toward the broad upper end of the wing 

 where the nerves terminate at the margin. The origin of this vascular 

 system is clearly seen in the specimen shown in figure 2, which is still 

 attached to its axis In this example the strands emerging from the 

 stem or pedicel curve into the base of the wing, or bract, some of the 

 forking bundles being clearly seen to traverse the wing. The seed 

 seems originally to have been placed in a somewhat twisted position 

 on the axis, the marginal rib being on the lower (proximal) side. 



The nervation of the bract- wmg of the seed is not clear throughout ; 

 but in most of the specimens, among the hundreds found in the coUec- 



