The spring tide of migration appears to set in along the entire coast-line of the Mediterranean, and in a direction almost due north. I do

not remember to have seen any record of the occurrence of the Martin on the west coast of Africa, although there seems to be no reason why it should not accompany the Swallow there in winter.

Both species will rear two broods in a season; and this fact, doubtless, will account for the prolonged stay in autumn of the later fledged birds, which are not sufficiently strong on the wing to join the main body of emigrants at the usual time of their departure.

THE

range than either of the foregoing species, being found in the New as well as in the Old World. In British North America M. Bourgeau obtained both birds and eggs on the Saskatchewan plains. Dr. Coues met with it in Arizona, and Professor Baird has recorded it from California. He says: "It furnishes almost a solitary instance amongst land birds of the same species inhabiting both continents permanently, and not as an accidental or occasional

visitor in either." Mr. H. E. Dresser found it

common in Southern Texas, and Mr. O. Salvin obtained several specimens in Guatemala.

It

has even been met with in the Bermudas, 600 miles from Cape Hatteras, the nearest point of the North American coast.2

In Europe the Sand Martin generally makes its appearance in the spring somewhat earlier than any of the other Swallows, and departs sooner. From different stations on the Mediterranean large flocks have been observed at the period of the vernal migration winging their way northward, returning even in greater numbers in the autumn. Mr. O. Salvin saw this species between Tunis and Kef during the third week in March. Canon Tristram, who found it abundant in Palestine in the sandy banks of the Jordan, has suggested that it is double-brooded, since he found it nesting in Egypt in February. The same observer met with it in November on its autumn migration through the Sahara. When

1 "Birds of North America," p. 313.

Jones's "Naturalist in Bermudas," p. 34.

passing down the Red Sea, early in November, Mr. Swinhoe saw numerous Sand Martins, which followed the ship for some days, and on arriving at his destination found these birds very common about the marshes at Takoo and before Tientsin in North China. Dr. Leith Adams says' that Sand Martins build in numbers along the banks of the Indus, and that in consequence in some places the banks are quite riddled with their holes. Hence it will be seen that this delicate little bird enjoys a more extensive range than any other species of the family.

Before leaving this country in autumn, they assemble in vast flocks, and go through a variety of evolutions on the wing, as if practising for a long flight, alighting from time to time upon the ground, or on willows or reeds by the river-side, to rest. Swallows and Martins do the same, but never congregate-so far as I have observed -in such large numbers.2

1 "Wanderings of a Naturalist in India," p. 49.

2 See "The Birds of Middlesex," p. 126.

The Purple Martin (H. purpurea) of America is recorded to have been procured once at Kingstown near Dublin; and Yarrell included it in his "History of British Birds," relying on a statement that two specimens had been shot. at Kingsbury Reservoir, in Middlesex, in September, 1842. It has since been ascertained, however, that he was misinformed on the subject. A specimen of this bird, said to have been shot near Macclesfield, was sold at Stevens's, with other birds from the Macclesfield Museum, on the 14th June, 1861, and realized twentyeight shillings. With these exceptions, so far as I am aware, no other instance of its occurrence in Europe has been published.