[ 411 ] XXXIV. Notes on the Structure and Affinities of Batidea3, Callitrichacese, Vochysiaceae, and Cassytlieae. By B. Clarke, Esq., F.L.S. ReadJune 16th, 1859. 1. On the Structure and Affinities of Batidese and Callitricliaceae. i HE principal points in the structure of the genus Batis, P. Br., have become well known, but its affinities have not been decided, while those which have been suggested for it are proved to be, for the most part, if not entirely, impracticable ; further analysis therefore seems required before its affinities can be brought to light ; and the follo\ving additional particulars, as regards its structiu'c, and especially the relative position of the oatiIcs to the axis of the ovary, will, I believe, make it evident that it must be a near ally of the Verbenacece. The male flowers are arranged in 4-rowed spikes, giving them a habit like that of the Verbenacece with 4-rowed spikes of flowers ; but in one species the inflorescence becomes compound, which gives it in some degree the habit of a Globularia, as of G. orientalis. The calyx is reduced to one sepal, posterior ; but, since ta the Acanthace(S also the calyx is sometimes very nearly obsolete, tliis negative character is of Little value. The corolla is attached by a very slender basis to the calyx, and consists of four petals with long claws and oval or almost circular laminae, and of four stamens alternate with them, two of which are j)laced anterior and posterior, and two right and left of the axis *, — the two lateral stamens having the rudiment of an ovary slightly adhering to them, which is sometimes bifid, the segments being lateral. The female flowers are in shorter spikes, which are 4-rowed, as in the male. The floral envelopes are deficient ; but each ovary has at its base a quickly deciduous bractea (Tab. LXVIII. fig. 27), so that in the early stage the spike is completely enveloped in scales, and is also terminated with three or four pairs of minute flowerless scales. The stigmas are two, right and left of the axis, and the cells of the ovary four, so placed that two of them stand right, and two left of the axis, so as to correspond in theu* position with the two stigmas, each pair having no bundle of vessels between them, but being separated by very distinct bundles from the opposite pair (fig. 27 a). Each cell contains an erect anatropal ovule, the raphe and foramen of which are very distinct. The raphe is lateral in relation to the inner angle of each cell ; but in each jjair of cells the raphes are always turned towards each other (fig. 27 a), as in the Verbenacece and LabiatcB, proving beyond doubt that the structure of the ovary is the same, consisting of only two carpels. According to this view of its structure, Batis differs from the Verbenacece only in being diclinous and polypetalous, imless it is in the position of the stamens ; for in the female * This, at least, is their position in B. marifima, but whether it is always so may be a question. 3 H 2
BioStor
