In view of the above observations by tliis careful observer, we have gone over the series of 

 Sand-Martins in the British Museum, and we find that there is considerable variation in the 

 depth of colour and in the amount of furcation in the tail. These characters are not even 

 constant in the American series, and the latter can be matched in every respect by examples 

 from the Old World. We cannot, therefore, uphold the separation of the Palaearctic and 

 Nearctic Sand-Martins, even as subspecies. 



Hab. Breeding in the northern parts of the Old and New Worlds, wintering in more southern localities 

 in the Indian, Ethiopian, and Neotropical Regions. 



The Sancl-Martin, or Bank-Martin as it is sometimes called, from its habit of nesting 

 in sandbanks, is a common species in nearly every part of Europe, becoming rarer in the 

 northern parts of the continent. It has been found throughout the whole extent of the 

 British Islands, and is a regular summer visitor all over Ireland, arriving at the end of 

 March or early in April, and leaving in the early part of September. As with the 

 House-Martin, later instances of its stay are frequently recorded, and jMr. Howard 

 Saunders, in his ' Manual,' refers to an exceptional record of the bird's occurrence in 

 county Limerick as late as the 30th of November in 1859. 



Mr. Eobert Gray writes:— "The Sand-Martin is widely distributed throughout 

 Scotland, extending to all the remoter islands. It is a regular summer visitant to Lewis, 

 Harris, and North List, breeding in sandbanks on the western sides of these islands. 

 It also inhabits South List and Barra, but I have not seen it on Benbecula, which 

 probably does not furnish suitable banks for a bird of its mining habits." 



In the Orkney Islands, according to Messrs. Buckley and Harvie-Browu, it was 

 formerly a common summer resident, and bred in large numbers, but is now much less 

 numerous, so that it can only be considered an irregular visitant, being more plentiful 

 in some years than in others. It has been recorded by Mr. H. A. Evans as breeding 

 near Lerwick in the Shetland Islands. 



So far as is at present known, the Sand-Martin has never been noticed either in 

 the Tvevoe Islands or in Iceland. Mr. Dresser, in his ' Birds of Europe,' gives a full 

 account of its distribution in Northern Europe, and publishes the following notes from 

 Prof. Ilobcrt Collett, of Christiania, and other observers, Avritiug as follows :— " ' It breeds 

 from Southern Norway up to the Prussian frontier, north of tlic Polar Circle, and along 

 the west coast it is one of the commonest species. It breeds up in the birch-region in the 

 fells both on the Dovre and in Ptouderne.' In a further note Prol'. Collett tells me that 

 it aiTives in Norway about the middle of May, and leaves again early in September. In 

 the Gudbrandsdal and Osterdal he found them breeding in the roofs of liouses, which 

 are there frequently made of turf, into which the Martins had burrowed, though many 

 other more suitable localities for the purpose of nidification were close at linnd. Pastoi- 

 Sommerfelt says that it is 'not common on the A'aranger iioid, l)ut breeds in several 

 localities, as, for instance, at Seida, on the Tana river at Nuorgan, at Polniak, and at 

 Oxevandct ; ' and he tliinks that of latter years it has increased in numbers. Nilsson says 

 that it is numerous throughout Sweden up into the Polar Circle, arriving rather later 



