4 FLOKA INDICA. [Ranunculocea. 



less, but solid, corky, and not eeed-bearing. The leaves are tlie same as in N. lauri- 

 foHa. 



2. CLEMATIS, L. 



Sepala 4-8. Petala nulla vel calyce breviora. Achenia sessilia, 

 caudata vel rostrata. — Frutices scandentea vel erecti. 



This very large genus, which is common in all temperate climates, is represented 

 in the mountains of India by numerous fine species. These ai'e all scandent, and 

 vary much in appearance at different stages of their growth, so that their sepa- 

 ration in the herbarium is not always easy, the specimens which are preserved there 

 very imperfectly illustrating the character of the species and the amount of variation 

 in the different parts of the plant. The shape and degree of division of the leaves, 

 the size and form of the flowers, the inflorescence, and the bracts, have all been em- 

 ployed to yield specific characters. The mode in which the leaves are divided, 

 whether decompound, pinnate, or ternate, appears pretty constant in each species, 

 but the shape and degree of division of the leaflets is extremely variable. AVith 

 regard to the inflorescence, the species with single-flowered peduncles seem never 

 to pass into the paniculate group ; but among the latter every amount of variation 

 occurs, the terminal panicles in the larger species being often a mass of flowers with 

 few and very small leaves, while those on the lower part of the stem are long-pedun- 

 cled, much more leafy, and altogether quite different in aspect. The bracts too are 

 extremely variable, and specific characters founded upon them will, we believe, be 

 found invariably to break down when a large series of specimens are examined. 

 Good characters are afforded by the position of the sepals in the expanded fiower, 

 but the size and pubescence of the flowers are very uncertain. The buds in all 

 the species are at first globose, and afterwards become oval or oblong ; still, within 

 certain limits, their shape at the period of expansion of the flower appears tolerably 

 constant. The shape and degree of hairiness of the stamens form an important 

 character, which, however, unfortunately fails in those species which are most difli- 

 cult of discrimination. 



The sections into which this genus has been divided by De CandoUe are on the 

 whole natural. The majority of the Indian species belong to the subgenus Flam- 

 mula, which is everywhere the most extensive. Cheiropsis is in habit nudistinguish- 

 able from Atragene ; and the amount of development of the petals or barren sta- 

 mens in Atragene being very variable, it is not improbable that on a general revision 

 of the genus these two gi-oups will be united. C. montmia has so exactly the habit 

 of Cheiropsis (in which indeed De Candolle has placed it, notwithstanding its having 

 no involucral bracts), that it appears desirable to rest the distinction of that section, 

 mainly on the one-flowered pedicels, in preference to the involucre, by doing which 

 we include C. montana. Behceanthera of Edgeworth has also so much the habit 

 of Cheiropsis, that we prefer its union to that group to its retention as a distinct 

 subgenus, especially as a transition is established between its very remai'kable sta- 

 mens and the ordinary form by C acutangula. 



The genus Clematis is universally diffused throughout temperate conntries, if we 

 except extratropical South America, in which its species are very rare. In Australia 

 it is as abundant as in the Northern Hemisphere. There are no arctic or alpine 

 species, and within the tropics they are for the most part natives of mountainous 

 countries. Madagascar possesses a number of very remarkable forms. 



In India the species of Clematis are most numerous in the temperate zone of the 

 Himalaya, only two occurring in the tropical regions. Of these, one (C. Gouriana) 

 is found in the open plains, but only exceptionally, as it is much more plentiful 

 in subtropical hills. The other (C. Cadmia) is a rare and local plant. The S|)ccies 

 of liie Eastern Himalaya and Malayan Peninsula are analogous to those of China. 

 In the Western Himalaya and the mountains of Hindostau and the Dekhan, African 



