424 TAMABIX GALLICA. 



their colour entirely to their base, alternate with the ten 

 "■lands which form the disk, confirm this opinion. In further 

 confirmation we find that the floral whorls in the young 

 Tamarix are all pentamerous, except the innermost, or 

 pistil, which is composed of three phyllidia* or ovarian leaves. 

 On examining these (see J^g. 2.) we perceive that the 

 interior is opposed to the axis of the vegetable, and what is 

 altogether anomalous, to one of the stamens. On the con- 

 trary, in the pentamerous Crassulacea, the two interior phyl- 

 lidia alternate with the axis. In Tamarix, therefore, it is 

 probable that whilst two of the phyllidia have disappeared, 

 the remaining three have filled up the vacant space, and thus 

 displaced themselves. If however we insert the two that are 

 wanting, one on the side of the axis, and the other in the 

 space opposed to it, the normal position of the whole will be 

 restored, and we shall have a pentamerous flower on the 

 same plan as those of Crassulacece, the fourth whorl alter- 

 nating with the stamens, and its two interior members with 

 the axis of the plant, each opposed, as in Crassulacece, to 

 a staminode, though in this latter case the staminodes are 

 united together in a continuous cupule. This cupule exists 

 in the whole order, and the description of the genus Myri- 

 caria, by Professor Ehrenberg, glandula scutellaris germen 

 suffitlciens nulla, is in this respect faulty ; the only difference 

 that exists is, that in Myricaria, the enlarged bases of the 

 filaments are united above the disk in a tube very distin- 

 guishable from the disk itself by its different colour. Hence 

 M. A. de St Hilaire very justly remarks that the lower part 

 of the staminal tube in 31yricaria is of a glandulous consis- 

 tence. Nor does the difference in the disks of T. Africatia, 

 Poir,f and T. Anglica, hereafter to be described, in which 



* This word from the Greek (fuwihov, a leaflet, I employ in Latin to 

 express the term ovarian leaves, created by M. A. de St Hilaire, and 

 which represent in the ovarium what are afterwards the valves in the fruit. 

 See Phijtogr. Can. Sect. i. p. 202. 



t From having examined an imperfect specimen in which the filament 

 was shrivelled, I advanced most erroneously {Phytogr. Can. Sect. i. p. 



