ORNITHOLOGICAL OBSERVATIONS 267 



Should the nest adhere to the side of a wall, then the 

 bird supplies it with a dome of mud, but if it is 

 situated in the angle between the wall and the roof, 

 no dome is required since the bird utilizes the roof 

 for a dome. The latter position is therefore more 

 economical for the bird and is thus more frequently 

 chosen. An underlying support for the nest was also 

 adopted, for occasionally the martin used to squeeze 

 its nest in between the rafters and the roof and support 

 its nest on the underlying wooden rafter. This latter 

 change of instinct was less fixed, and I have no doubt 

 was more recently acquired, as I have noticed that 

 some nests, though constructed in a suitable place for 

 deriving support from the rafter, yet were separated 

 by a space of about one inch from the rafter and were 

 completely closed in by mud below. Every instinct, 

 like every structure, is plastic. Given sufficient time, 

 it will mould itself to new ends. 



A very characteristic feature of the bird life of the 

 Himalaya is the congregation of insectivorous birds 

 that traverse the forest in a compact troop. Often 

 the woods seem quite deserted and not a sound is to 

 be heard. Suddenly a hustling throng of little birds 

 appears. All is then activity and bustle. The trees 

 swarm w r ith feathered life. Creepers, different species 

 of tits and warblers arrive and flit in company from 

 tree to tree ; chats, flycatchers, an occasional nuthatch, 

 perhaps a family t of shrike-tits or laughing-thrushes, a 

 fire-crest or a troop of minivets also join in the common 

 throng. A few seed-eating birds like the rose-finches 

 or the cinnamon sparrows sometimes appear, but the 

 main body is composed of insectivorous species. Some 

 hang upon the smaller branches, others search the 



