44 floba indica. [Ranunculacece. 



The genus Aquilegia is limited to a few species, all natives of the north temperate 

 zone. In Europe, Northern Asia, and North America, they are common in moun- 

 tain woods and pastures, rising into the alpine region. In India the genus occurs 

 only in the Western Himalaya and in the mountains of Tibet, to which at least five 

 species have been considered peculiar. We have, however, found that all the Euro- 

 pean species distinguished by Linnaeus and subsequent authors occur in the Himalaya. 

 It has therefore been necessary to submit all these to a critical examination, the re- 

 sult of which has very unexpectedly been that all the European and many of the Sibe- 

 rian forms generally recognized belong to one very variable species. We do not 

 include A. parviflora, Ledeb., which, judging from the figure and description, and 

 from a siugle specimen, is very distinct ; nor A. Canadensis, L., to which A. Sibi- 

 rica, Lam., A. atropurpurea, DC, A. Davurica, Patr., A.formosa, Fisch., and per- 

 haps A. lactijlora, Kar. et Kir., ought probably to be referred as synonyms. This 

 species is universally recognized as distinct by American botanists, and appears 

 readily distinguished by the exserted stamens, the shape of the petals, and the small, 

 straight, inflated, and suddenly contracted spurs. A. carulea, Torrey, or A. lepto- 

 ceras, Nutt. non Fischer, is probably a large-flowered form of the same species ; and 

 even A. hybrida, Sims (Bot. Mag. t. 1221), which has hooked spurs, retains the 

 other characteristics of A. Canadensis. None of these varieties have been found in 



the Himalaya. 



We feel that it is difficult to explain briefly, and at the same time clearly, the 

 grounds on which we have come to the conclusion that all the synonyms quoted 

 below must be considered states of one very variable species. Our Indian specimens 

 are numerous, and exhibit many different forms, which it was not difficult to throw 

 into tolerably well marked groups by their general appearance, with the exception 

 of a few intermediate specimens. On comparing them with the general herbarium, 

 it was at once apparent that these groups corresponded pretty closely to the com- 

 monly recognized species of authors, so that our course appeared easy. As soon, 

 however, as we attempted to frame diagnoses which should be applicable not only 

 to the Indian plants, but to those of Europe with which we had identified them, we 

 found that the great amount of variation to which this genus is subject interposed 



insuperable difficulties. 



Authors have availed themselves of four classes of characters to distinguish from one 

 another the species of Aquilegia. 1. The shape of the floral organs. 2. I he nature 

 and degree of pubescence. 3. The height of the stem, the number of its leaves and 

 the amount of ramification. 4. The degree of division of the leaves, and the stalked 

 or sessile leaflets. Linmeus described only two European species, A. alpina, witn 

 straight spurs, and A. vulgaris, with hooked spurs; but subsequent authors consider 

 the pubescence a prominent character, as may be seen by the names viscosa, gl**f*~ 

 losa, pubijlora. Both De Candolle and Treviranus, however, have long ago admitted 

 the inefficacy of this character, and stated their belief that A. viscosa cannot be dis- 

 tinguished from A. vulgaris; and though systematists in general have not followed 

 their example, that is only because the wish to make species prevails over the au- 

 thority of scientific inquirers,- we cannot say over their example, since both tne 



have kept them sepanile. 



The shape of the 'leaves, though not noticed by Linmrus has been relied upon 



by De Candolle and others, for the separation of A. alpina from A. vulgaris. Ifte 

 result has been that specimens which would otherwise be referred, from the shape tM 

 size of the flowers, to A. alpina, have been separated from it by more recent authors 

 Tinder various names, because the leaves were less deeply cut. Species h«e even 

 been distinguished bv the leaflets being sessile or stalked W e do not repeat the re- 

 marks which we have already so frequently had occasion to make regarding the great 

 degree of variation to which the foliage of Ranuncutacea, and indeed of all cut-Ieavea 

 „ c i- e i n ,i,Jia,i+i/\Tt t\f ct^ppittiptis or a caie- 



