754 Birds of Celebes: Charadriidae. 



America, collaris of S. America, and ruficapilla of Australia, New Guinea and 

 New Zealand. It would be a matter for no great surprise if three of these 

 closely allied forms were to be found in Celebes besides Aegialitis peroni; namely 

 A. cantiaim and dealbata as winter visitors, and A. ritficapilla as a straggler from 

 Australia or New Guinea. A. cantiana, which is already known from Borneo 

 and some of the Philippines, may be recognised by its black (or leaden") legs 

 and larger size (wing 104 — 114 mm: Seebohm); A. dealbata has light yellowish 

 brown or flesh-coloured legs like pero7ii, but is fully as large as cantiana and, 

 as in that bu'd, the black collar (or rather the remains of it) is found only on 

 the sides of the neck, not on the hind-neck; A. ruficapilla is of about the same 

 size as A. peroni, but, as Seebohm remarks, it has neither the black nor the 

 white nuchal collar; the tarsi are light grey, feet blackish brown. An example 

 in the Dresden Museum (Nr. 11440) with the wing 1 14 mm, labelled "Philipiiines", 

 seems to be dealbata. 



Compared with A. curonica and its closest allies, the cantiana-gioup differs 

 in having a relatively longer tarsus, shorter tail (shorter by about '/«) and much 

 longer and stouter bill; the three outer tail-feathers are pure white, the shafts 

 of all the remiges are white (except at the tip), and the head of the adult is 

 rufous behind the black bar of the fore crown. 



A recent writer on migi-ation asserts that northern birds do not increase 

 their breeding range towards the tropics. Swinhoe (d J) remarks of A. dealbata 

 of China, Hainan and Formosa, that no one can doubt the fact of its being 

 derived from A. cantiana. In the same way it might be urged that peroni and 

 the other forms of the south were sprung from that species, in other words, 

 that A. cantiana has spread its breeding range southward, where it has under- 

 gone local modifications. On the other hand it may be supposed that one of 

 the tropical races gave rise to the northern cantiana. The former view seems 

 the more plausible one, because, as has been shown, certain allied migratory 

 species stay the summer in their winter quarters, and cases are known of their 

 breeding there'). As to the modifications of A. peroni, Seebohm remarks that 

 "its small size is probably connected with the fact that it lives upon islands and 

 not upon a continent", and believes that, being of a somewhat arctic genus, it 

 has decreased in size in the uncongenial climate of the troi^ics. If peroni is 

 really sprung from cantiana, it has of course decreased in size (supposing cantiana 

 has not since grown larger), but more proof of a decrease in size of northern 

 forms in the tropics is desirable, for as a rule a more luxuriant growth pertains 

 to the tropics, and it does not appear that islands are likely to harbour smaller 

 forms than continents (except, perhaps, among the Raptores) ; the small Sangi 

 Islands at all events produce larger races than the large island of Celebes. 



') Charadrius fulvus, Aegialitis geoffroyi, Aeg. mongola, Strepsilas interpres and Limosa novaezealandiae. 



