EAST NEPAL AND THE SIKKIM HIMALAYA MOUNTAINS. 71 



alpine locality, and the young plants of the same species from 

 wooded regions lo^ye^ down, than between some widely distinct 

 species. I have hereby been led into much error in my illus- 

 trated work on Sikkim Rhododendrons, which I shall endeavour 

 to remedy in the present essay. 



Hybridization in a Native State, — It has been insisted that 

 many of the so-called species of this genus are naturally-produced 

 hybrids. This is by no means clear to me, though I am far 

 from denying its possibility, and I am aware of the many obvious 

 facilities for such a process : it shares the plausibility of all hy- 

 potheses against which negative proof only can be brought to 

 bear. A gardener's opinion is in such cases of value, as he can 

 best appreciate the power of an agent he employs himself with 

 great effect, and to which the attention of botanists is being 

 drawn through the pages of this Journal especially. Con- 

 sidering, however, the consequences which have attended the 

 process of hybridizing Rhododendrons in our gardens, it appears 

 to me that were such an operation actively pursued by nature, 

 or tardily throughout a series of years, in the Rhododen- 

 dron region of the Himalaya, the species would be in the con- 

 dition of Salix, Ruhiis, Rosa, Mentha, and many other familiar 

 examples of hopelessly entangled assemblages of species. Where 

 the Rhododendron "scrub" prevails, it is impossible for any 

 but very strong growing plants to establish a footing ; a mule 

 would have little chance of a flourishing existence anti numerous 

 offspring'in so dense a mass ; but the prevalence of land and 

 snow-slips often lays bare broad tracts of land, \\hich are for 

 years but scantily clothed with vegetation. From the constant 

 local changes of surface that hence ensue, there is no want of 

 opportunity and space for hybrids to establish themselves. Be- 

 lieving it probable that this genus has survived slow but gi'eat 

 changes in the physical geography and climate of the countrj'^, 

 especially with relation to the snow -level, I cannot but conclude 

 that varieties, perhaps permanent ones, may have been induced, 

 which are now regarded as species ; and this has prompted me 

 to unite some plants from very different elevations which, though 

 varying much, present no well-marked specific characters. Such 

 are R. eltsagnoides, R. salignum, and R. obovatum. All these 

 I include under Wallich's R. lepidotum, to which a range in ele- 

 vation of upwards of 7000 feet is consequently given, but I can- 

 not trace the influence of hybridization in itself or any allied 

 species. It is accompanied by twenty congeners in its ascent from 

 7000 to 15.000 feet, but except that its flowers varj' from 

 yellow to a dirty purple, at various heights, neither it nor they 

 present any characters that can be attributed to hybridization. 

 ^.fulgens is the most suspicious- looking plant in this respect, 



