Xot.1- BEES AND POISONOUS HOXEY. 



pa, shrubby Conqtosita, and very many plants absent 

 at similar elevations on the wet outer Dorjiling ranges. 



In the contracted parts of the valley, the mountains often 

 dip to the river-bed, in precipices of gneiss., under the ledges 

 of which wild bees build pendulous nests., looking like huge 

 bats suspended by their wings : they are two or three feet 

 long, and as broad at the top, whence they taper down- 

 wards: the honey is much sought for, except in spring, when 

 it is said to be poisoned by Rhododendron flowers, just as 

 that, eaten by the soldiers in the retreat of the 

 Thousand, was by the flowers of the R. jxmficum. 



Above these gorges are enormous accumulations of rocks, 

 especially at the confluence of lateral valleys, where they 

 rest upon little flats, like the river-terraees of Mywa, but 

 wholly formed of angular shingle, flanked with beds of 

 river-formed gravel : some of these boulders were thirty or 

 forty yards across, and split as if they had fallen from a 

 height ; the path passing between the fragments." At first 

 I imagined that they had been precipitated from the 

 mountains around ; and I referred the shingle to land- 

 shoots, which during the rains descend several thousand feet 

 in devastating avalanches, damming up the rivers, and 

 destroying houses, cattle, and cultivation ; but though I 

 still refer the materials of many such terraces to this cause, 

 I consider those at the mouths of valleys to be due to 

 ancient glacial action, especially when laden with such 

 enormous blocks as are probably ice-transported. 



A change in the population accompanies that in the 

 natural features of the country. Tibetans replacing the 



* The split fragments I was wholly unable to account for, till my attention 



?cted by Mr. Darwin to the observations of Charpentier and Agassiz, who 



refer s imilar ones met with in the Alps, to rocks which have fallen through 



crevasses in glaciers. — See '• Darwin on Glaciers and Transported Boulders in 



Xorth Wales." London, " PhiL Mag/" xxi. p. ISO. 



