62 PASSERES. — HIRUNDINID^. 



dently old ones, for the Bomhax had not yet 

 perfected its cotton, and hence I infer that these 

 birds continue from year to year to occupy the 

 same nests, until they are thrown off by the growth 

 of the tree. The entrance to the nests, which were 

 subglobular, was near the bottom. 



Near the middle of May, my servant Sam, being 

 engaged at Culloden, in Westmoreland parish, cut- 

 ting the fronds of the palmetto (Chamcerops) for 

 thatching, found these little birds nestling in abun- 

 dance, and procured for me many nests of the 

 present season. Their recent construction, and 

 perhaps the diversity of their situation — for instead 

 of the hollow of a spathe, these were attached to 

 the plaited surface of the fronds, — gave them a 

 different appearance from the former specimens. 

 Many of these 1 have now in my possession. They 

 have a singularly hairy appearance, being composed 

 almost exclusively of the flax-like cotton of the 

 Bombax, and when separated, are not unlike a 

 doll's wig. They are in the form of those watch - 

 fobs, which are hung at beds' heads, the backs 

 being firmly glued by saliva to the under surface 

 of the fronds, the impressions of the plaits of which 

 are conspicuous on the nests when separated. The 

 thickness is slight in the upper part, but in the 

 lower it is much increased, the depth of the cup 

 descending very little below the opening. The 

 cotton is cemented firmly together as in the case of 

 the others, but externally it is allowed to hang 

 in filamentous locks, having a woolly, but not al- 

 together a ragged appearance. A few feathers are 



