308 BIRDS or AUSTRALIA. 



the low wooded islands it loves to frequent ; altliongh gene- 

 rally a wary bird even when little disturbed by man, yet on 

 one occasion on Heron Island I knocked down several with a 

 stick. The nest is usually placed on a tree, but on those 

 islands where there are none, such as Raine's Islet and else- 

 where, it breeds among the recesses of the rocks ; where the 

 trees are tall, as on Oomaga or Keat's Island, the nests are 

 placed near the summit ; on Dugong Island they were placed 

 on the root of a tree, on a low stump, or halfway up a low 

 bushy tree ; they are shallow in form, eighteen inches in diame- 

 ter, and constructed of small sticks and lined with twigs ; the 

 eggs are two in number, and of a pale bluish white, one inch 

 and seven-eighths long by one inch and a quarter broad." 

 Strange says, " I procured specimens about ten miles north of 

 Sydney Heads ; it appears to be strictly confined to the rocky 

 cliffs and ledges of rocks, where it takes great delight in 

 allowing the spray to beat over it. It is very shy and 

 wary, and never stops long in one place." In his notes 

 from Port Essington, Gilbert states that " it is abundant 

 on all the small islands and rocks immediately adjacent 

 to the mainland. It is gregarious in its habits the whole 

 year round, for I remarked that it was congregated in as 

 large numbers before* as after the breeding-season, which is 

 the month of August. The nest is built of sticks on the 

 ground, and is perfectly round and from twelve to eighteen 

 inches in height, with a considerable depression for the recep- 

 tion of the eggs ; they are always placed in thickets or under- 

 wood, and as near the outer edge of the rock as possible. 

 On one small rock I found at least fifty of these nests, some 

 of which were so close as nearly to touch each other. The 

 eggs were sometimes two, and at others three in number." 



The sexes are so similar that dissection must be resorted to 

 to distinguish the one from the other. 



It will be observed that these statements are contradictory 

 in some particulars, which may perhaps be accounted foi" by 



