500 PROCEEDINGS OF THE NATIONAL MUSEUM. vol.41. 



tions, it may bo more or less indistinctly seen branching radiately in 

 passing to the distal margin, while in a few of the clearer examples 

 it seems to unatomose near the distal border so as to form a mesh like 

 the areolation of the Gigantopteris frond. Wliile this areolation seems 

 to exist, its occurrence is not, however, unquestionable. The small 

 roundish body showing through the envelope in the upper part of the 

 seed may be interpreted as the impression of the megaspore. 



The seeds described above are provisionally referred to Gigantop- 

 teris americana on account of their generally abundant and most inti- 

 mate association with the leaves of that plant wherever it has been 

 found; on account of their asymmetry and solitary lateral attachment, 

 as though on the ventral faces of lateral lobes or modified pinnules, 

 along a longitudinal axis; and on account of the similarity of texture 

 and the apparent though not indubitable similarity in the nervation 

 of the bract-wing to the lamina and nervation of Gigantopteris. The 

 characters of the asymmetrical and unilaterally ribbed lamina or 

 wing-bract, as described above, strongly suggest a modified and re- 

 duced pinnule or secondary pinna of a fern-like frond. Such, accord- 

 ing to my interpretation, it probably is. Mention has already been 

 made of the deeper dissection of the shortening lobes at the base of 

 some of the large fronds as well as in the young one shown in plate 46, 

 figure 3. As to whether, supposing the seeds to belong to Gigantop- 

 teris, they were borne on reduced lobes at the bases of the fronds or 

 on special seed-bearing pinnse cannot at present be ascertained, though 

 it is perhaps more likely that they were distichously and" obliquely 

 placed on a short, special rachis, possibly situated in the axil of one 

 of the fronds. 



Anticipating that all doubt as to the connection between seed and 

 leaf will later be fully removed, I do not give any other name to this 

 new generic type of seed. 



Supposed polleniferous scales. — In addition to the seeds which ac- 

 company the remains of Gigantopter^is in the plant beds near Fulda, 

 there are found several examples of a peculiar strobilus. This consists, 

 as shown in figure 4, plate 48, of a rather thick, short axis bearing 

 oppositely two distichous rows of closely placed broadly reniform or 

 broadly ovate-reniform bracts, each about 1 centimeter in length. 

 The bracts, wliich stand nearly at a right angle to the plane of the axis, 

 partially clasp the latter in an oblique direction and are concavo- 

 convex, the hollow side being downward, as shown in the figures 

 (3 and 4, pi. 48). A border zone of the bracts, many of which are 

 found detached, as shown in plate 49, figure 7, is smooth and bent 

 slightly downward to form a curtain. The inner portion of the bract 

 is thickened, more fibrous, and provided on the lower side with great 

 numbers of small oval pendant sacs, probably for containing pollen, 

 though they possibly are sporangia. The examples seen in plate 49, 



