the year. Clustering in numbers against these gloomy cliffs, its "pendent bed" may be observed in many places throughout the range; among others, at the Gobbins, where some hundreds annually breed. About the Giant's Causeway, their admirably buoyant flight has particularly attracted me, as they skimmed the varied surface of the shelving banks and rugged rocks, keeping with an easy grace the same distance from the ground, notwithstanding the extreme inequality of surface. The house-martin "is the most numerous of the genus in Rathlin, where it is found in all parts of the island, as well inland as along the cliffs which overhang the sea." It builds in "the range of white cliffs running along the north-western side of Church bay."* I observed considerable numbers of their nests in May, 1842, built in a similar manner, against the high and white limestone precipices of the Little Deer Park, Glenarm (county of Antrim). So many as twenty of them were in some places in juxtaposition. They almost overhang one side of the new line of road lately formed at the base of the cliffs, previously washed by the sea.†

About the sea-girt rocks of the peninsula of the Horn, in Donegal; those near to Ardmore in the county of Waterford; at the island of Lambay off the Dublin coast; and other similar localities, I have remarked the presence of the martin. It is said by Capt. Cook, to breed in vast numbers among the rocks of the Alps and Pyrenees, often far from the habitation of man.

Martins occasionally build under the arches of bridges. At Toome bridge, where the river Bann leaves Lough Neagh, I saw many of their nests in 1834, and was told that for a long period it had been a favourite haunt. In 1846, the following notes were made during a stay of three days at Toome:--August the 1st.

* Dr. J. D. Marshall, in the same memoir, mentions, that one of these birds which he shot, "had its mouth completely filled with insects, among which were a large dragon-fly, and one of the Tipula [T. oleracea]." White of Selborne has remarked, that swifts and sand-martins feed on Libellulæ.

† In rocks of a similar kind, though in a very different scene, the chalk-cliffs, which rise above the river Derwent, near Cromford, Derbyshire,-I observed many nests of the martin, in June, 1835. They were built as far in as possible beneath the overhanging rocks, in the same manner as under a projecting roof.

# Sketches in Spain, vol. ii. p. 275.

When we arrived at the bridge at twenty minutes past 8 o'clock in the evening, great numbers of house-martins were most noisy before retiring for the night, which they had done at half-past eight. Next morning, their call-notes were equally loud; not less than three hundred birds being on wing together. It is an excellent feeding station. My companion (A. H. Haliday, Esq.,) found quite a harvest of insects (Phryganea, &c.) on the railing of the bridge. This is a fine structure of nine arches, having its walls surmounted by a stone coping with an iron railing. The martins' nests are all in an unusual site, (though it would not be so for the Hirundo rustica,) beneath the arches of the bridge. This is not owing to any want of their usual building places, the eaves of houses; those in the village of Toome, distant only a few hundred paces from the bridge being perfectly suitable for the purpose, and similar to what are chosen in other parts of the country :--but not a nest is to be seen there. Hundreds of birds, too, covered the slated roof of the inn during the forenoons. We went under the arches of the bridge on our way to and from Church Island, and saw innumerable nests, looking most interesting, from the variety of form assumed in consequence of the nature of the locality; many were of a rude Florence-flask-like shape. The pellets of clay on the outside were remarkably large, appearing at first sight like gravel of pretty uniform size. The apertures to the nests on the southern side of the arches looked south, as those on the opposite side, did north, doubtless on account of the greater facility thus afforded during the progress of building, and afterwards for the egress and ingress of the birds. At ten minutes past 8 o'clock this evening, an amazing increase to the number of these martins appeared above the bridge: there could not have been less than five hundred together. At twenty minutes past eight, the whole body made a sweep to the arches of the bridge and a number retired to their nests, but several downward sweeps from on high, like those of starlings to their roosting place, were necessary, before they all got settled for the night. Their numbers reminded me of a swarm of bees hurrying to their hive. Until about roosting time they would seem to range to

some distance to feed. I saw many, early this evening, about little bays of the lake, and believed them to belong to the Toome bridge colony. After the martins had retired to roost, a few swifts continued flying above the bridge, in the neighbourhood of which only, they were before observed. On the following morning, a sand-martin appeared at the bridge with its congeners. Not a swallow (H. rustica) was seen there during the three days.

Houses are however, the best known building places of this species.

White, in the sixteenth letter of his Natural History of Selborne, says :"It has been observed that martins usually build to a north-east or north-west aspect, that the heat of the sun may not crack and destroy their nests; but instances are also remembered where they bred for many years in vast abundance in a hot, stifled innyard, against a wall facing to the south." On this subject the following note was made on the 15th of July, 1832-I this day observed twelve or thirteen nests of the Hirundo urbica built against a two-story house at Wolf-hill. These were all on the north-west side or front, excepting one, which was at the north-east corner. The other two sides of this house have in part a southerly exposure (S.W. and S.E.), and being fenced in, are consequently more private. A road passes those preferred by the martin::-on every side the facilities for its building operations are the same. In front of a thatched cottage not more than eight feet high, which is not only at the side of the highway, but constantly resorted to as a public-house, I remarked several nests of the martin. In the rear of this cottage, which is fenced off from the road, and its walls (from the building being on the side of a hill) considerably higher than in front, none of the nests appear. Some years ago a few pair built annually in front of the dwelling-house at Wolf-hill, but not more than a single pair occupied either gable. Nests were also displayed in considerable numbers in front of two lofty four-story houses in Belfast. Judging from the situations selected by the martin for its nests on these five houses (the three first mentioned being only a few hundred yards apart), it would seem that the bird is more influenced by the front of a house than by aspect; as the first faces the north-west, the second and third the southeast, and the fourth and fifth the south. In innumerable other instances, I have remarked that where facilities for building are similar on all sides of the house, the front was thus preferred by the martin, although the nests were opposite every point of the compass. This is particularly apparent in houses situated in streets which intersect each other at right angles. The aspect of the cliffs before mentioned, as being tenanted by the martin, is as different as that of the houses. One reason for the fronts of houses being thus preferred (in the instances mentioned, the low cottage and the four-story house are equally so,) is probably, on account of the more open space in front allowing of a freer range of flight to and from the nest. The following was noted as a striking instance to the same effect. It refers to Hever Castle, in Kent, a square building, fronting the south, well known historically in connection with Henry the Eighth, and Anne Boleyn :-When there in October, 1847, I remarked

great numbers of house-martins nests about the tops of the fine old windows of all the stories at three sides of the castle. They were equally common on each of these sides, and picturesquely clustered on the top of each other with the apertures above, below, and in all positions. The three sides on which they were, are the north, south, and west, not one appearing on the east, although it presents every facility that the three others do, in windows, &c. This side only is retired, and forms a boundary to a garden, the gate of which is kept locked. There are doors in the three sides of the castle resorted to by the birds, but no entrance on that they do not occupy. In front, the bustle of an occupied farm-house is superadded. The colour of the nests harmonises well with the lichen-stained sandstone building.

It has been remarked of this species, that the nest "is closed all around, except a small orifice, usually on the most sheltered side," &c. My observation on the side of the nest chosen for entrance in the north of Ireland, does not accord with this, as in nests closely adjoining, the apertures are on every side. Of nine nests in front of a house before alluded to, the entrances appeared on the north, south, and west sides, the wall against which they were placed occupying the eastern. On this house,— as is not unfrequently the case,-several of these architects had, like certain other bipeds when erecting their habitations, taken advantage of their neighbours' gables, and it may be presumed, for a similar reason,-that of being saved trouble and expense of labour: All the nests of the martin that I have particularly noticed here, had, with a single exception, the entrance close to the top. In this instance, although the nest was built against the wall of the house, beneath a projecting roof, the aperture was placed rather above the centre, in the same manner as that in the nest of the wren (Troglodytes Europæus). The entrances to other nests on the same house, were as usual.

Although the nest of the poor martin is often, in the north of the island, torn away from the houses of persons who imagine themselves to be possessed of good taste, yet there are others, who, disliking the harsh contrast between its clay-built shed and the snow-white walls of their mansions, are unwilling to disturb the summer wanderer, and for the sake of uniformity, have had its domicile whitewashed at the same time with their own. I first noticed this in the town of Antrim, where, on two houses, several nests thus appeared: their architects flying in and out, evinced their contentment with the change. In Hillsborough, I afterwards observed, that the same practice had been adopted, and without annoyance to the birds. When visiting the old church at Helmsley, Yorkshire, in Oct., 1844, I remarked that the nests of the martin built at the top of the windows so perfectly harmonised in colour with the venerable lichen-stained structure as almost to escape notice.

The statement of several continental authors, that house-martins, on finding sparrows in possession of their nests, had been known to rise en masse, and fill up the entrance when the intruders were within, would seem, from the silence of some of the latest British writers of authority on the subject, not to be credited by them. The compiler of the Architecture of Birds sets it down as a “fanciful legend;" but I have unquestionable testimony, that a case precisely similar to those related by the authors alluded to, occurred in 1832, in the next farm to our own, near Belfast.

When the house-martins returned in that year to a thatched cottage, belonging to Mr. John Clements, where they had annually built for a long period—and which then displayed fourteen of their nests—a pair found that sparrows had taken possession of their domicile. On perceiving this, they kept up such "a chattering about the nest" as to attract the attention of the owner of the house. After its continuance for some time, apparently until they were convinced that the sparrow was determined to retain possession, they flew away, and did not return for a considerable time, when they re-appeared with about twenty of their kindred. They now immediately commenced "claying up the entrance to the nest," which was done in the course of the day; next morning, the construction of a new nest was commenced against the side of the old one, and in it they reared their brood undisturbed. After some time, the proprietor of the cottage, who had never heard of any similar case, pulled down both nests, and in that occupied by the sparrow found its "rotten corpse," together with several eggs. A particular note of the entire proceeding, as related by Mr. Clements, was made by my brother soon after the occurrence; but to make assurance doubly sure" before publishing the account in 1842, I inquired of Mr. Clements whether he remembered such a circumstance, and he repeated it just as narrated nine years before. Some other persons too of our mutual acquaintance, were witness to the chief parts of the proceeding, and saw the sparrow, together with the eggs, in the sealed-up nest.*

66

* Three recent occurrences of a similar nature are recorded by Mr. Weir (Macgillivray, British Birds, vol. iii. p. 591), and two others are alluded to under the head of "Swallow" by the Bishop of Norwich, in his Familiar History of Birds, vol. ii. p. 55, 3rd edition.