SUPPLEMENT— CRASSULACE;E. 697 



the sterile flowers. Rudimentary stamens none. Disk none. Ovary 3-CPlIed, 

 with apparently about 6 erect or ascending ovules. Immature fruit ovoid 

 (somewhat fleshy?), gibbous, densely echinate witli weak prickles, at length 

 1-celled? Seeds 3-4 or G, large, flat. — A slender trailing or climbing herb, 

 with pedately dissected leaves, and simple tendrils. Flowers (while) very 

 small; the sterile in filiform often somewhat compound racemes; the fertile 

 ones solitary in the same axils, on short peduncles. 



D. disseda. 



Texas, Drummond ! — Plant nearly glabrous. Stem slender. Leaves 

 ternately divided ; the divisions attenuated and linear at the base, or petiolu- 

 late, the petiolules slightly margined; tiic terminal division 3-parIod; the 

 lobes irregularly toothed or sinuate ; the middle one oblong, conspicuously 

 mucronate; the lateral ones shorter, and often 2-3-lobed : lateral divisions 

 2-parted; the segments deeply U-S-lobed, and sinuate-toothed. Sterile ra- 

 cemes as long as tlie leaves; the flowers (scarcely 3 lines in diameter) on 

 short pedicels. The colunm consists of a very short flat filament (of 2 

 united), bearing a peltate Hat disk, which probably is composeil of 2 united 

 anthers, opening by a continuous even line all around the margin, with no 

 interruption to mark the points of connexion; neither is the antlicr in any 

 degree tortuous, but the disk, after the pollen is slicd, is slightly folded up- 

 wards: the anther is manifestly 2-celled : within the margin, both on the 

 upper and lower side of the disk, is a circle of minute radiating cilia", borne 

 apparently by the margin of a disk which is closely api)licd to the faces of 

 the anllicr. Peduncleof the fertile flowers scarcely as long as the half- 

 grown fruit. Ovary 3-celled ; the cells probably disappearing during the 

 growth of the fruit : style and stigmas not observed: ovules erect from near 

 the base of the cell; the young fruit tiiickly clothed with long, weak, and 

 soft smootli i)rickles. Seeds apparently large and flat, erect from near the 

 base of the fruit. — This plant, which we have in Drummond's Texan col- 

 lection, had escaped our notice until after our account of this family was 

 printed. Our specimens are unfortunately somewhat incomplete. The 

 genus belongs to the same division with Cyclanthera of Schrader (founded on 

 a Mexican plant), with which it accords in the remarkable structure of the 

 anthers, and in some other respects; it is also allied to Elateriuni. This 

 last genus appears to need revision, and some of the described species may 

 be found to agree either with Discanthera or Cyclanthera. The fruit of 

 Elaterium pubescens is 3-, or perhaps by suppression 2-celled, and the 

 seeds erect as in the present plant. Does E.? hastatum, H. B. S^' K., which 

 is said to have minute campanulate-rotate flowers and a 6-seeded fruit, 

 belong to the same genus with the plant here described ? 



Elaterium trifoliatum (Linn, manliss.) is founded on a description of a plant 

 of Clayton's, which we did not find in his herbarium, and are unable to identify 

 it by the characters given, unless it should bo Sicyos angulalus. Clayton does 

 not describe the leaves as trifoliolate, but as 3-Iobcd. 



Order CRASSULACE.^. 



1. TILL^A, p. 557. 



1. T. minima. — We have it also from Douglas's Californian collection. 



88 



