5(3 THE .TOUK>'AL OF BOTANY 



No diagnosis is attached to Wallich's name {List, no. 7227). 

 Decaisne's description (in Jacquemont's Voyage dans VInde, iv. 61) 

 is tolerably full, and most of the dried specimens which I have had an 

 opportunity of examining agree fairly satisfactorily with it. The 

 majority of the plants raised from Maire's seed may be referred to 

 the same form, which may be taken as the type — plant glabrous, 

 rosettes lax, leaves Hat, alternate, spathulate, acuminate, stem 4-10 

 inches. The only differences of any moment between my series of 

 the t^^pical plant and Decaisne's description are that he describes the 

 petals as lanceolate, subattenuate, and twice as long as the sepals, 

 and his figure shows a campanulate flower with the tips of the petals 

 tapering and erect ; in my series the petals were oblong or oblong- 

 lanceolate with recurved tips, and the flowers resembled in shape those 

 of the lil3^-of -the- valley. C. B. Clarke's description (Fl. Brit. Ind. ii. 

 413) is veiy short ; he describes the petals as " dull rose, scarcely 

 twice the sepals." 



The tips of the petals are erect in bud, and also after flowering, 

 and tend to assume that position in drWng ; this no doubt accounts 

 for the absence in all the descriptions of reference to their character- 

 istic reflexed habit. A peculiar thickening on the upper part of the 

 face of the fleshy j^etals is likewise undescribed, doubtless because in 

 dried specimens it is obscure. This is a marked feature of the type 

 as represented by Maire's plants. In longitudinal section the S-shaped 

 petal is seen to increase in thickness from the tip to half way down, 

 when it contracts abruptly to about one-third of its maximum thic'k- 

 ness, and continues so to the base, the scale occupying the hollow 

 thus formed. In front view the thickened portion shows a bluntly 

 bilobed lower edge. 



Two varieties have been described — var. Forresfi E. Hamet (in 

 Notes R. Bot. Grard. Edinb. v. 115; type in Herb. Edinb.), a tall 

 green plant with very broad ovate-suborbicular acute leaves, of which 

 all but the uppermost are opposite; and var. yunnanense R. Hamet 

 (in Journ. de Bot. x. 284 — Crassula yimnanensis Franchet) a densely 

 hair}^ form with mucronate leaves. A number of Maire's seedlings 

 are referable to this latter variety, of which I am able to amplify the 

 description ; it is a noteworthy form, almost worthy of specific rank. 

 Franchet separated it from Crassnia indica Decne mainly on account 

 of its general pilosity and its mucronate leaves. In my plants the 

 size of stem, leaf, and inflorescence was much less than in my plants 

 of the type (Franchet says "Port et dimensions de C. indica"'). 

 The leaves were very thick, being so convex on the under side that the 

 breadth was only from once to twice the thickness, not thi-ee to four 

 times the thickness as in the type. Flowers rather larger than in 

 type, calyx narrower, petals more erect at base, so that the cor(.)lla is 

 narrower, and less reflexed at apex, making the whole petal much 

 sti-aighter ; the peculiar thickening of the upper half of the petal, 

 which is so marked a feature of the type, is quite absent. The whole 

 plant, as stated by Franchet, is shortly i)ilose, even to the backs of 

 the petals. 



Another form deserving of varietal rank ap])eared in some numbers 

 among the plants grown from Maire's seed. This was very unifoim 



