28 EAMBLES OF A NATUEALIST. [Oh. II. 



this would hardly account for the fact that, considerably- 

 above high- water mark, many lay half-embedded in the dried 

 mud and thick confervoid growth which had long lain above 

 high-water mark, and bore the signs of having been well 

 baked and cracked by many a noonday sun. The banks of 

 the lagoon had evidently been under water comparatively 

 recently, and much higher up than the tide now reached. 



But although some classes of animals were poorly repre- 

 sented upon Pratas Island, there were plenty of birds, and 

 of several species, both sea and land birds. A buzzard I 

 noticed several times ; but it was too wary to allow me to 

 come within gunshot, although it offered a tantalising mark 

 just out of range. I observed a very handsome shrike, with 

 an ash-coloured head and black moustache. The blue- 

 jackets reported that they had seen a canary ,• and I after- 

 wards saw myself a yellowish bird resembling the English 

 siskin, which was probably the bird they had noticed. 

 Another bird (Petrocinclus manihensis), about the size of a 

 blackbird, was of a glossy metallic blue above and fawn- 

 coloured beneath. Its stomach contained the elytra of 

 beetles. A fifth species presented aU the appearance of a 

 veritable blackbird, but I could not get near enough to ex- 

 amine it closely. A species of swallow, probably Hirundo 

 gutturalis, with glossy bluish back, chestnut throat, and 

 with a speckled fawn-colour imderneath, was flying about ia 

 considerable numbers ; and on the banks of the shallow in- 

 let I saw a bright-coloured kingfisher, very similar in appear- 

 ance and size to our own species. There were also some 

 small birds which crossed our path from time to time, with 

 the jerking flight and the chirrup of the hard-billed perchers. 

 Large flocks of Tringas (sandpipers), of at least two species, 

 were visible on the sandy flats of the inlet which were left 



