Rosacee.] THE HIMALAYAN MOUNTAINS. 201 
- Parochetus owvalidifolia ; foliolis obovatis retusis integerrimis.—P. major. Wall. Cat. Herb. Ind. n. 
5525.—P. communis. 1. c. n. 5972.—T ab. 35. f. 1,—(a.) standard ;—(6.) alz ;—(c.) carina of corolla ;— 
(e.) quadrifid calyx, upper lip bidentate, and pistil with single stamen ;—(d.) nine united stamens ;— 
(f.) half of legume, with seven seeds, 
This genus has been. referred by Mr. Don to the Phaseolee, but Mr. Bentham is of opinion that it 
has greater affinity to the T'rifoliee, with which it certainly agrees in habit, while the parts of fructi- 
fication present no discrepancies. The species was at first referred to P. communis, as on referring to 
the E. I. Herbarium, no difference could be perceived between it and P. major. n. 5525, which appeared 
to be only a variety of P. communis. n. 5972. The latter name, as the more general and applicable, 
was therefore adopted ; but Mr. Don has since shown me the original specimens, of which the leaflets are 
clearly dentato-serrate, of his P. communis; and as P. major is described with crenate leaves, this species, 
the most nearly allied to it and which I have never seen except with the most completely entire leaves, 
must be considered distinct, unless specimens, showing the regular gradations, can be found. 
Smithia ciliata; caule erecto, calycis labiis reticulato-venosis dentato-ciliatis.—Tab. 35. f. 2.— 
(a.) calyx and bractez, with young legume ;—(b.) ripe fruit. 
This plant brought me from near Kedarkanta, is easily distinguished from §. sensitiva, found in the 
Deyra Doon, by its habit, the form of the leaves, and its calyx. I am indebted to my friend, W. Saun- 
ders, Esq. for the drawing. 
Uraria lagopus ; DC. Prod. 2. p. 324. Wall. Ic. ined. 577—U. lagopodioides. Wall. Cat. Herb. Ind. 
n. 5676.—An Doodia alopecuroides? Roxb. FI. Ind. 3. p.368.—Uraria lagopoides. 'T. 33. f. 1. 
Though this species varies a good deal in the form of its leaves, there is no doubt of its being the same 
plant as n. 5676 of Dr.Wallich’s Catalogue, there named: U. lagopodioides, and in consequence so named 
by me in the figure T. 33. f.1. though in my own Catalogue it had been referred to U. lagopus, DC. ; 
so in the E. I. Herbarium, the specimens are marked U. lagopus, as in Dr.W.’s drawing referred to. 
This cannot be Doodia lagopodioides, Roxb., as that is described as being “ prostrate and rooting at the 
joints ;° but may very probably be D. alopecwroides, from which it differs little in character, and is found 
in the same parts of the country, that having been sent by General Hardwicke from the northern parts 
of India, and this found by myself in the Kheree Pass, Deyra Doon, and Suen Range. 
Cassia lanceolata.—Forskal Catal. Arab. p. cxi. n. 270. Descrip. Plant. p. 85. 
In conformity with the opinion of many botanists, this species has been considered as distinct from 
C. acutifolia of Delile, though the two are united by M. De Candolle, and are certainly very nearly 
allied to one another. The latter, called C. lanceolata, Lam., by Nectoux, is considered by M. Delile 
to be the same as his C.acutifolia, and may be distinguished by its suffruticose habit, shorter, ovate- 
lanceolate acute leaves, long-linear stipules, and by the shorter, less membranous legumes. ‘The Cassia 
lanceolata, Forsk., or that grown from the seed of Swna-mukee, though an annual, may with care be made 
to live throughout the year, and thus assume a suffruticose habit. The leaves are truly lanceolate, but 
differ in length in the lower and upper parts of the stem, but the sessile glands on which so much stress 
has been Jaid, do not appear to be a constant character, as I have very seldom seen them. The figure 
will give a correct idea of the form of the thin and membranous legume. Tab. 37.—(a.) A flower; 
—(b.) the same, with the calyx and corolla removed ;—(d. ¢.) sepals and petals separated ;—(f.g.) fertile 
stamens ;—(e.) abortive ditto ;—(A.) pistil;—(é. &.) legume ;—(/. m. n.) seed and embryo. 
63. ROSACEZ. 
The Rosacea, like the Leguminosae, have been divided by some botanists into several 
orders, which are by others considered only tribes of that under consideration. As the 
multiplication of families renders advisable the retaining together as many as possible in 
one group, whether this be called a class or order, I have preferred following M. De 
Candolle, in considering them together, instead of separating the Chrysobalanea, Amyg- 
dalee, and Pomacee, from the Sanguisorbee, Spireacee, Dryadee, and Rosee ; particu- 
larly as they correspond in many points of structure, geographical distribution, and, 
2D with 
