AMARANTACEJl. 61 



tapering atthe base, dilated at the truncate top, the teeth of which are minute, blunt or ^' 

 States to'peraT'"™''"*' ^*^-~^''^- '^"'^ ^^'^"^='' ^'«^'»-' [seashores from the United J 



5. BATIS, L. 



Flowers dioecious, amentaceous, naked. $ : Stamens 4, alternating with as many mem- 

 branaceous scales, enclosed in a bifid, delicate involucre, and inserted upon the base of an 

 orbicnlar bract. ? : Flowers semi-immersed in the fleshy axis, exinvolucrate, supported by 

 a single bract. Ovan, 4-celled, crowned by » bilobed stigma: ovule in each cell single 

 basilai-, anatropous. Syncarpmm baccate: endocarp coriaceous; testa membranaceous: 

 embryo arcuate, exalbuminous. — Shrubby halophytes; ]sa.vea opposite, fleshy ; aments com- 

 pact, oblong-cyhndncal, axillary and terminal. 



The combination of this genus with Chenopodea was proposed by Kunth at a time when 

 very little of its structure was known, but now, siuce Torrey has lately given a complete 

 andysis of it, the difSculty is apparently increased. The chief points of anomaly in this 

 Order may however be lessened by the following considerations : 



1. Torrey adopts in Satis a complete flower, interpreting the involucre as a calyx, and 

 the scales as a tetrapetalous corolla : thus the stamens would alternate with the petals. But 

 this anomaly disappears if we compare the scales, which in our specimens are smaller than 

 in Torrey's figure and exceedingly delicate, with the staminodia in other Ohenofodece ; and 

 the bifid involucre with the analogous organ in the female flower of Atriplex. 



3. In the female amentum of Batis, the structure of which I know only from Ton-ey's 

 analysis, the flowers are united to the middle and with the basilar bracts of the system; each 

 flower, supported by a bract analogous to that of the male one, is reduced to a single pistil* 

 which may be considered as a 4-ceIled oVary, terminated with a bilobed stigma. According 

 to this view, each ovary-cell would contain a single basilar ovule, and this is a kind of pla- 

 centation, of which analogies in allied plants are completely wanting, while the structure of 

 such a cell might be compared with that of the entire ovary of Chenopodets. Now the pistil 

 in this family is a reduced one, the two styles showing its compound nature, and that of 

 Batis therefore may be considered as showing its fall development. The strength of this 

 argument is increased by the similar habit of SaKcomiete, their flowers immersed in the axis 

 (though not adherent), the fleshy parts, the coriaceous endocarp, and the membranaceous 

 testa. 



3. The exalbuminons embryo of other Chenopodete is usually combined with cochleate 

 cotyledons, while in Batis the embryo is only slightly arcuate ; but the fleshy, large cotyle- 

 dons, with a short conical and inferior radicle, are adapted to the asymmetrical form of the 

 seed, thus indicating an eccentrical development; and such a structure might be compared 

 with the redaction of the common annular embryo in Caryophyllea to a straight lateral one 

 in THanthus. 



/ 7. B. maritima, L. Leaves oblong-linear or linear, flat above, convex beneath. — Jacq. y 

 ' Amer. Pict. t. 246 : the fruit-bearing plant (copied in Desc. Fl. 7. t. 496) ; Torr. in Smith- ^ 

 son. Contribut. 6. t. 11. — A diflnse shrub, with the young branches upright, 3-4' high ; 

 leaves about 1" long, exceeding the aments ; aments paniculate. — Hab. Jamaica I, March, 

 common in the salt-marshes of the south side of the island ; Turk Islands !, Hjalm. ; Carib- 

 bean Islands; [Florida, Venezuela], 



XXIII. AMAEANTACEJ]. 



Flowers apetalous (2)-3-bracteolate. Stamens hypogynous, opposite to the calyx, which 

 is usually scarious. Ovary unilocular : placentation basilar. Embryo excentrical : perisperm 

 central, mealy. — Leaves exstipulate, usually quite entire. 



Some species are used in Colonial medicine, being slightly emollient, resolvent drugs : thus 

 in Jaihaica Iresine celosioides (Juba's-bush) is used as a stomachic by the Negroes (Prf.). 

 • In the arrangement of this Order I recur chiefly to R. Brown's generic characters. The 

 characters taken from the staminodia (or, rather, lateral teeth of the filaments), if employed, 

 as Endlicher and Moquin-Tandon did, for the discrimination of the genera, destroy those 

 natm'al assemblages of species, which from their habit, and especially their inflorescence, are 

 sufficiently obvious. 



