184 :mi:soz()T(' floras of fxited states. 



\(). .")()(). (iSl represents a luuch larfjci' ti'unk, Imt is much less perfect 

 than tlie one last described, and none of its missing parts could he found 

 among the fragments. It consists of considerably more than half of the 

 original trunk, which was symmetrically subconical, rounded at the 

 summit, anil longitudinally flattened, but to a much less degree than the 

 other. The principal loss is at tlie base by a rather even transverse 

 fracture, but a large piece is also broken away from one of the thinner 

 sides, the fracture deepening toward the summit and carrying away the 

 apex altogether. It is now 24 cm. liigh and IS 1)V 25 cm. in diameter at 

 the basal fracture, and has a girth of (H) cm. It weighs 11.34 kg. This, 

 therefore, represents a larger trunk than any that were originally referred 

 to this species. It also differs in some other respects from the types. 

 The outer coating of ramentum is also wliolly removed and the surface 

 thus exposed presents a somewhat different appearance from that of the 

 other specimens. The l)ases of the petioles rise above the walls so as to 

 give the surface a rough, warty aspect. The rock sul)stance is harder 

 and heavier than that of the other trimks. These differences do not, 

 however, seem to be specific, and I prefer to retain the specimen in this 

 species. 



Xo. 100.227 is a segment from near the base of a still larger trunk. 

 The basal fracture is nearly horizontal, Ijut the upper one is somewhat 

 oblique, so that while the thicker edge is 12 cm. high the thinner edge is 

 only 6 cm. This trunk was also somewhat elliptical, the diameters at 

 the base being 19 cm. and 27 cm., respectively, giving a girth of 74 cm. 

 The segment weighs 8.62 kg. The external surface of this specimen 

 closely resembles that of No. 500.681 and it represents the same general 

 type. These two specimens may ultimately require to be placed in a 

 distinct species. 



PL XLVI is a side view, also showing part of the Ijase, of the trunk 

 formed by joining Xos. 500.88, 500.513, and 500.525. 



A larger number of specimens had to be referred to this species than 

 to any other, but all except those treated above are mere fragments, 

 most of them quite small. Many of these fragments closely resemble 

 the interior of the type specimens Xos. 500.7, 500.8, 500.20, and 500.67, 

 and there can be no doubt that some of them are parts of the same trunk 

 or trunks represented by those specimens, but in only one case has this 

 been proved l)y finding the complementary parts. This case is that of 



