Sword-sheaths.—Up to the present time four sheaths (Fig. 123, Nos. 1 to 4) have been recovered, but only one is entire, the rest being more or less in a fragmentary condition. They are all made of thin bronze riveted together at the margins, and over this there is a bead which, towards the lower third, develops into an elegant ornamentation very similar to that which we have already seen on the sword-sheaths of La Tène (See Fig. 87.) The perfect sheath (No. 1) is devoid of ornamentation, except that formed by the marginal bead; but the other three (of which only one side of each remains) are decorated with highly artistic designs formed by incised lines, the details of which will be readily seen from the illustrations. No. 1, which still retains the blade of an iron sword firmly adhered to it, is 17½ inches long. No. 2 is defective at the tip, besides having lost its surrounding bead; it measures, in its present condition, 19 inches in length. The illustration here is a reproduction of a rubbing reduced half size and shows only the upper half. The design which comes out in white lines is in reality incised lines and corresponds to the dark lines in the other two. No. 3, the largest of the group, measures 22 inches in length. Both ends are here represented. No. 4 is the smallest, being only 16½ inches in length. It is supposed that the incised lines, which are sharply defined and deeply cut, contained a black enamel, but no traces of it now remain. The circular cavities in the surrounding bead at the tips were also intended for the reception of enamel, probably of some brilliant colour. These designs, which belong to the so-called "late Celtic" style of ornamentation, when so treated must have had a striking effect on the bright bronze surface.

In Nos. 1 and 4 it will be observed that there is a transverse raised band, the purpose of which was, no doubt, to strengthen the sheath. Such bands were a feature in the La Tène sheaths, which in some instances were repeated several times at intervals on the body of the sheath. In Canon Grainger's collection I saw one or two broad rings compressed in the middle, which I took to be the cross-bands of other sheaths.

Swords.—The swords which belonged to these sheaths were all of iron, a fact which probably accounts for their being in a more dilapidated condition. Of four recovered up to this date only one is in a good and perfect condition (Fig. 124, No. 1). Its total length is 19½ inches, of which the handle takes up five inches, measuring from the extremity of the tang to the nearest part of the curved ridge which separates it from the blade. The blade has a sharply defined ridge and tapers gracefully to a point. The other swords are all fragmentary. One, as already mentioned, is still in its sheath; another, portion of a blade 14 inches long, has a sharp central ridge and otherwise corresponds exactly with the entire weapon. Of the fourth there remains merely the handle (No. 2), the blade having apparently disappeared by oxidation. So far it also closely corresponds with the entire weapon. Both handles have bronze mountings, which, though differing in some details, are so peculiar in their plan and method of execution, that we have no hesitation in recognising them as belonging to one and the same group—a group which, so far as my knowledge goes, finds its parallel only in a few examples from La Tène. (See Fig. 87, Nos. 7 and 8.) As to the material of which the grip was made there is no evidence. The bronze circlets and flanges on the tangs, if in their original position, involve the necessity of having the handles, whether of bone or horn, divided in several pieces. That these bronze sheaths and iron swords were counterparts of each other there can be no reasonable doubt. Their points of agreement, besides general dimensions and style of manufacture, are too remarkable to be accidental. Thus the peculiar curve at the opening of the sheaths fits that of the band of separation between the handle and blade of the swords. Also the prominent midribs in the two sword-blades have corresponding prominences in two of the sheaths (Nos. 3 and 4).

Lances.—Though there is only one spear-head in the Lisnacroghera collection, there is ample evidence to show that it was not an isolated example of this kind of weapon. Certain objects, like the brass knob of a door (Nos. 28, 29, and 30), amounting in all to seven or eight, are now known to have adorned the butt end of the wooden handles of so many spears. One of these handles, 8 feet in length, though now greatly shrunk and contorted, is still preserved by Canon Grainger and conclusively proved their use as well as that of some other objects which previously seemed equally mysterious. At the other end of the spear-shaft, close to where it entered the socket, was a ferrule of the same style of art as the sword-sheaths (Nos. 23, 24, and 25). Several of these ferrules have been found which, like the ornaments on the butt end, differ considerably in regard to size. They are all ornamented with elegant designs in incised lines which also, it is supposed, contained enamel. One is here represented with a piece of the handle and a bronze rivet (No. 23), probably that which fixed the spear more firmly in its shaft; but of the spear itself nothing remains. The only spear-head that has been found is of iron. It is a magnificent blade 16½ inches in length, with a slim socket containing two rivets (No. 3).

A curious object here represented as No. 27 consists of an oval ring, "richly decorated with bands, in which are remains of white and red enamelled designs in a chevron or wavy pattern." On one edge there are two forms like that of a bird. Mr. Wakeman considers it to be the terminal ornament or pommel to a sword or dagger. (B. 411, p. 391.)

Various Bronze Mountings.—Among the other objects which from their character and style of art belonged to the same group as we have just described are the following:—(a) Two ornaments of thin bronze in repoussé. One is a disc (No. 22), slightly impaired at the margin, bearing in the centre a triquetrum of symmetrical spirals, and surrounded by a slightly raised border. The other (No. 20) is in a more fragmentary condition, but sufficient remains to show the design to be a swastika or croix gammée, worked after the same fashion as the former and probably serving as the central ornament of a disc. (b) Three cup-shaped ornaments each having a small aperture in the centre. One (No. 18) is surrounded by a broad margin like a wide-awake hat, which is copiously ornamented with incised lines. On the supposition that this was intended to adorn the face of a wooden shield the cup must have been embedded in the wood, as it is clear that its concave side is the front, there being no ornamentation on the other. Another of the same kind is represented by No. 19, and differs from the former only by having a smaller and less decorated border. In a third specimen the ornamented border entirely disappears, and its place is taken by a marginal bead. (d) A variety of annular and penannular rings, some hollow (Nos. 6 and 7) and others solid (No. 17). Some, as Nos. 8 and 9, are ornamented with designs (one of which still retains an enamel of a vermilion tint) and were probably attached to pins and used as ring-brooches. A curious object like a doubly coiled bangle with the coils adherent and ornamented with zig-zag lines is supposed to have been the rim of an oval brooch (No. 15). (e) A plain but elegantly shaped drinking-cup of thin bronze (No. 10) has neither handle nor any marks of rivets. (f) Among minor things are a couple of bronze pins, one with a round top and the other with a small ring; also a rude finger ring (No. 16), an ornamental stud (No. 21), and a rivet with one side deeply serrated like a cog-wheel (No. 26).

In addition to the above list of objects, all of which may be considered as part of one special and indeed unique group which by some chance found their way into the Lisnacroghera crannog, there are others similar to the usual class of crannog relics; and among these I have noted the following:—

Iron.—Two large iron cauldrons in a fragmentary state; a large curved knife, like that of a hedge-cutter, still retaining its wooden handle; an axe (No. 4), also containing a portion of the handle; an adze (No. 5); a reaping hook; portion of a gunlock, together with various bits of an undefined character.

Beads.—Several coloured and variegated beads (blue, red, and white) (Nos. 11, 12, and 14). One bead is of stone, another large one is of jet, and another of amber (No. 13).

Stone.—One small stone axe of a dark colour and wedge-shaped, some elongated four-sided hones, hammer-stones, etc.; a few flint flakes, and rude arrow-points.

Pottery.—A few fragments of coarse earthenware, indicating large wide-mouthed vessels.

Miscellaneous.—A portion of "bog-butter" bearing the impression of a coarse cloth. Several pieces of wood with round and square-cut holes.

LOUGH MOURNE.

In consequence of the partial drainage of Lough Mourne during the summer of 1882, while its basin was being converted into a reservoir for the supply of water to the town of Belfast, two artificial islands became exposed which were at once recognised to be the remains of submerged lake-dwellings. Shortly after exposure I visited them in company with Mr. Robinson, the assistant engineer to the Belfast waterworks, and subsequently recorded my observations in the Proceedings of the Society of Antiquaries of Scotland (B. 439), of which the following is a short abstract.

The first, which was easily accessible on foot, was very much disturbed by the repeated "howkings" of visitors. Its form, as determined by the area containing the stumps of piles, was irregularly circular, but nowhere exceeding 60 feet in diameter. Included within this area were four small separate elevations, composed of a few stones, clay, and ashes interspersed with burnt twigs and bits of charcoal. None of these elevations had as large a diameter as 10 feet, and it only required one turn over of the spade to reach the undisturbed bed of the lake. Digging on these so-called islets, and in the spaces around them, I found stumps of upright piles and fragments of burnt faggots, the latter forming a thin layer over the true lake-sediment. The piles were closely set, about one or one and a half foot apart, and uniformly occupied the whole area of the dwelling without any appearance of circular stockades; nor was there any semblance of a fascine-structure indicating an artificially constructed island. From the margin or outskirts of the area occupied by these piles a double row of stumps extended towards the shore, which, on being traced, were found to have stopped half way to the ordinary water-level mark. The conclusion which I came to was that this lacustrine dwelling was a true palafitte, over whose wooden platform two or three hearths were constructed of incombustible materials, as stone and clay. Probably these hearths would have been surrounded by some kind of huts. During the conflagration which had destroyed the entire wooden structures (of which final catastrophe there appeared little doubt from the amount of burnt faggots and beams that lay scattered over the lake bed all over the area containing piles), the materials of these hearths would ultimately drop down to the bottom of the lake, still, however, retaining their relative position, and so present the appearance of low mounds over the bed of the lake with the stumps of their supporting piles penetrating them.

Among the relics recorded from this lake-dwelling are "two small stone crucibles" (B. 320), an iron hatchet, part of a canoe, a hammer-stone, a rubbing-stone, two small urns (?), part of a large crucible, charred bones, teeth, shells of hazel-nuts. (B. 439, p. 324.) Flint flakes, scrapers, and arrow-points have been found in the vicinity along the shore, but on the crannog itself only a quantity of fine chips was discovered. Some of these relics I saw in the possession of a young man of the name of Macdonald, who lived in the neighbourhood and took an active part in searching for relics. The iron hatchet turned out to be of exceptional interest. It is a socketed celt, with a loop at one side for fixing the handle, part of which was still remaining in the socket. (Fig. 125, No. 1.) Bronze celts of this description are, of course, very common, but specimens made of iron are extremely rare, especially in the British Isles, and only exceptionally to be seen in the museums of Europe. This is the only one hitherto recorded from a British lake-dwelling, and, moreover, it is, as regards size and form, quite unique. It measures 6¾ inches long and 3¼ inches broad immediately above the cutting edge. The longest diameter of the socket is 2½ inches, and this breadth is continued for about two-thirds of its length, except where the loophole causes it to bulge outwards.

One of the crucibles, which I believe is now presented to the Belfast Museum, is rather peculiar in having a short projecting handle (No. 2).

It would thus appear that the inhabitants of this lake-dwelling practised the art of metallurgy, which would entail at least one fire-proof furnace, and thus partly account for the existence of so many hearths on such a small area.

The second dwelling was about 150 yards from the shore, in deep water, and lying over a great depth of quaking mud. It was a small example of the ordinary stockaded island, but owing to the softness of the mud and oozing of water no satisfactory investigation could be undertaken. Mr. Robinson calculates the extent of its submergence at 15 feet. Its entire surface was occupied by a heap of stones which gave it the appearance of a hillside cairn. Digging underneath these stones we came upon a thick bed of heather and brackens interspersed with beams and brushwood. Near the margin were to be seen a few piles and beams rudely mortised, from which one or two long beams radiated towards the centre like the spokes of a wheel.

In the first described lake-dwelling no submergence could have taken place, as the lake bottom was composed of compact sand in which the piles had a firm hold.

In different parts of the lake two single-tree canoes were found embedded in the mud, both of which are now preserved in the Belfast Museum. One is a plain trough of a rectangular shape with slightly sloping sides, measuring 12½ feet long, 2½ wide and 9 inches deep (No. 3). Its flat base is perforated with six circular holes ¾ inch in diameter. The other is pointed at both ends, and altogether much more artistic in its structure. It was originally about 13 feet long and 3½ feet wide, but the fore-part is considerably damaged. The stern portion is here figured (No. 4). It had two seats formed of neat planks of oak, seven inches broad, and supported on projecting ledges on each side cut out of the solid, as shown in the section No. 6. The seats (No. 5) were kept in position by two wooden pins at each end which penetrated into the solid supports. For the oars there were also two arrangements on each side consisting of perforated projections left out of the solid as shown on the margin of the portion here sketched. But the most remarkable feature of this canoe consists of four prominences with abrupt edges (also left in the solid) for the feet of the rowers, as seen in the illustration. The distances from the centre of the seats to their corresponding foot marks were 33 and 34 inches respectively.

LIST OF IRISH CRANNOGS

ALPHABETICALLY ARRANGED,
WITH NOTES AND REFERENCES.

Aconnick Lough, co. Cavan. B. 13, Ap. p. 43.
Acrussel Lough, co. Fermanagh. B. 444, p. 187.
Allen Lough, co. Leitrim. B. 9, p. 45.
[75]Aghakilconnel Lough, co. Leitrim. B. 13, Ap. p. 43.
Aghnamullen ("Glebe Island"), co. Monaghan. B. 93b, p. 229.
[76]Annagh Lough, two crannogs, between King and Queen's County. B. 149, p. 154.
Annagh, parish of Kilbarron, co. Tipperary. B. 444, p. 212.
Ardakillen, four crannogs. B. 13, p. 208, and Ap. p. 48; B. 18, p. 230.
[77]Ardmore Bay, submarine crannog, co. Waterford. B. 329, p. 61, and B. 330, p. 154.
Arrow Lough, co. Sligo, several stone islands near the Abbey of Ballindoon. B. 444, p. 245.
Aughlish, about five miles from Enniskillen, co. Fermanagh. B. 217, pp. 323.
Ballaghmore, co. Fermanagh. B. 217, p. 322.
Ballinafad, co. Galway. B. 214, p. 12.
[78]Ballinahinch, co. Galway. B. 214, p. 12.
Ballinderry, near Moate, co. Westmeath. B. 391, p. 196.
Ballinlough, four crannogs, co. Galway. B. 70a.
Ballydoolough, five miles from Enniskillen, co. Fermanagh. B. 191, pp. 360 and 569, and B. 217, p. 314.
Ballygawley Lake, co. Sligo. B. 444, p. 246.
[79]Ballyhoe Lough, two crannogs, near Carrickmacross, co. Monaghan. B. 13, p. 417, and B. 135, p. 8.
Ballykinler, co. Down. B. 29, p. 86.
Ballylough Baile-an-Locha. "Annals of the Four Masters," B. 27, p. 193, and B. 438, p. 168.
Ballywoolen, co. Down. B. 29, p. 86.
Bohermeen, co. Meath. B. 444, p. 82.
Bola Lough, co. Galway, lake stone-dwelling. B. 214, p. 11.
[80]Breagho, co. Fermanagh. B. 217, p. 322.
Camlough, co. Armagh. B. 444, p. 178.
[81]Cargaghoge, Barony of Farney, co. Monaghan. B. 162, p. 269, and vol. v., 4th S., p. 330.
Castleforbes, co. Longford. B. 24, p. 150.
Castlefore Lough, two crannogs, co. Leitrim. B. 13, Ap. p. 43; B. 438, p. 409.
Clogherny, co. Tyrone. B. 119, 2nd ed., p. 649.
[82]Cloncorick Castle L., co. Leitrim. B. 440, p. 408.
Cloneygonnell L. (Tonymore), three crannogs, co. Cavan. B. 60.
Cloonbo L., two crannogs, co. Leitrim. B. 13, Ap. p. 43.
Cloonboniagh L., co. Leitrim. B. 13, Ap. p. 43, and B. 438, p. 408.
Cloonfinnen L., co. Leitrim. Ibid., and B. 438, p. 408.
Cloonfinlough, two crannogs, co. Roscommon. B. 13, p. 208, and Ap. p. 40.
Cloonfree, two crannogs. Ibid., p. 219, and Ap. p. 48.
Cloonturk L., two crannogs, co. Leitrim. Ibid., Ap. p. 43.
[83]Cloughwater Bog, near Ballymena, co. Antrim. B. 148, p. 21.
Coal Bog (Kilnamaddo), near Boho, co. Fermanagh. Jour. Arch. Association, xxxvi. p. 271; B. 345, p. 66.
Coolcranoge, co. Limerick. B. 444, p. 28.
Corcreevy (Loch-Laoghaire) co. Tyrone. "Annals of the Four Masters," B. 13, p. 215.
Corrib L., a few lake stone-dwellings, co. Galway. B. 214, p. 11.
[84]Cornagall L., co. Cavan. B. 191a, p. 461.
[85]Cornaseer, co. Cavan. B. 438, p. 148.
Craigywarren, co. Antrim. B. 444, p. 110.
Crannagh MacKnavin, co. Leitrim. B. 18, p. 233.
Crannagh Lough, co. Antrim. B. 24, p. 157.
Crannog-na-n-Duini, co. Donegal. B. 18, p. 233.
Crannog-boy, parish of Inishkeel, co. Donegal. B. 444, p. 28.
Crannog Mac Samhradhain, co. Cavan. "Annals of the Four Masters.," B. 9, p. 45.
Creenagh L., two crannogs, co. Leitrim. B. 438, p. 408.
Cullina, near Maryborough, Queen's County. B. 444, p. 210.
[86]Currygrane L., two crannogs, co. Longford. B. 443, p. 410.
Derreen L., co. Roscommon. B. 13, Ap. p. 62.
Derreskit L., co. Cavan. B. 13, Ap. p. 43.
Drumaleague L., two crannogs, co. Leitrim. B. 18; B. 13, Ap. p. 43.
Drumdarragh or Trillick, co. Fermanagh. B. 441; B. 217, p. 324.
[87]Drumgay, three crannogs and one stone-island, co. Fermanagh. B. 189, p. 232, and B. 217, p. 314.
[88]Drumkeery L., near Bailieborough, co. Cavan, B. 57, p. 483.
[89]Drumkelin, parish of Inver, co. Donegal. B. 3, p. 361.
[90]Drumlane, eight miles from Cavan, two crannogs, co. Cavan. B. 438, p. 149.
Drumskimly, three crannogs, co. Fermanagh. B. 189, p. 583, and B. 217, p. 320.
[91]Drumsloe, co. Fermanagh. B. 217, p. 321.
Effernan, co. Clare. B. 346, p. 336.
Eyes Lough, six crannogs, co. Fermanagh. B. 189, p. 553 and B. 217, p. 317.
Faughan L., co. Down. B. 24 and 25; B. 18, p. 158.
Fort L., co. Donegal. B. 444, p. 181.
Funshinagh L., co. Leitrim. B. 13, Ap. p. 43.
Galbally, co. Tyrone, B. 217, p. 197.
Glencar L., five crannogs, co. Sligo and Leitrim. B. 444, p. 243.
Gortalough, co. Fermanagh. B. 346, p. 336.
Grantstown, co. Queen. B. 93, p. 228.
Guile L., co. Antrim. B. 148, p. 20.
Gur L., co. Limerick. Evans' "Ancient Bronze Implements, etc.," p. 436; B. 18, p. 223.
Hackett Lough (L. Cimbe), co. Galway. "Annals of Lough Cé," 1067; B. 119, 2nd ed., p. 654; B. 18, p. 230.
Hilbert L., Goromna Island, a lake stone-dwelling, co. Galway. B. 214.
Inishrush (Green Lough), co. Derry. B. 25, p. 212.
Joristown, in the river Deal, co. Westmeath. B. 13, Ap. p. 55, and B. 444, p. 205.
Kilglass L. B. 13, Ap., p. 48.
Killynure, near Enniskillen, co. Fermanagh. B. 217, p. 323.
Kilmore L., two crannogs, co. Monaghan. B. 13, Ap. p. 43.
Kilknock L., co. Antrim. B. 24, p. 153.
Knockany (Lough Cend), co. Limerick. B. 444, p. 156.
Lagore, or Dunshaughlin, co. Meath. B. 175, p. 462; B. 14, p. 35; B. 18; B. 4, p. 425; B. 10, p. 101.
Lane L., co. Roscommon. Cal. State Papers Ireland, vol. 156, p. 374.
Lankhill, near Enniskillen, co. Fermanagh. B. 441, p. 372.
Leesborough L., co. Monaghan. B. 13, p. 43.
Lenaghan, co. Fermanagh. B. 444, p. 188.
Lisanisk, co. Monaghan. "The island Ever McCooley's house"; B. 18, p. 231; B. 9, p. 46, and B. 8, p. 94.
Lisnacrogliera, near Broughshane, co. Antrim. B. 411.
Lochanacrannog, co. Sligo. B. 444, p. 246.
[92]Loughran's Island ("Innis-an-Lochan"), in the river Bann, near Coleraine, co. Antrim. B. 13, p. 417; B. 27, p. 192.
Loughannaderriga, Achille Island, co. Mayo. B. 444, p. 230.
Loughinsholin, co. Derry. B. 18, p. 233; B. 25, p. 157.
Loughavarra, co. Antrim. Ulster J. Arch., vol. vii. p. 192; "Annals of the Four Masters," 1544.
Loughavilly, co. Fermanagh. B. 217, p. 321; B. 346, p. 332.
[93]Lochlea, three crannogs, co. Roscommon. B. 13, Ap. pp. 48 and 61; B. 18, p. 29.
Lough-na-Glack, co. Monaghan. B. 9, p. 46.
Lough Cam, lake stone-dwelling, co. Galway. B. 214, p. 12.
Loughmagarry, co. Antrim. B. 24, p. 156.
[94]Loughtamend ("Louglitoman"), co. Antrim. Ibid., p. 155.
Loughtown, co. Leitrim. B. 13, Ap. 43.
Loughrea, four crannogs, co. Galway, B. 58, p. 412.
Lough Oughter, three or more crannogs, co. Cavan. B. 438, p. 151.
Lynch Lough ("Loch-Leith-innsi"), co. Antrim. B. 27, p. 193.
Mac Hugh L., two crannogs, co. Leitrim. B. 13, Ap. p. 43, and B. 440, p. 408.
Macnean L., three crannogs, co. Fermanagh. B. 217, p. 323.
Mac Nevin (Mac Cnaimhain) crannog, co. Galway. B. 70, p. 176.
Manorhamilton, co. Leitrim. B. 25, p. 346, and B. 18, p. 552.
Marlacoo, co. Armagh. R. H. A. A., vol. vi., 4th S., p. 432.
Mask L., Hag's Castle, lake stone-dwelling. B. 214, p. 11.
Melvin L. (Melge), between co. Fermanagh and Leitrim. "Annals of the Four Masters"; B. 18, p. 231, and B. 13, p. 215.
Moinenoe, co. Fermanagh. B. 217, p. 322.
Monaincha, co. Tipperary. B. 444, p. 212.
Monalty, half a mile from Carrickmacross, co. Monaghan. B. 8, p. 94, and B. 9, p. 46.
Monea, co. Fermanagh. B. 217, p. 318.
Mongavlin, co. Donegal. "Annals of Lough Cé"; B. 444, p. 151.
Monnachin L., co. Monaghan. Ibid., p. 151.
Mourne L., two crannogs, co. Antrim. B. 390, pp. 194 and 371, and B. 439, p. 321.
Moynagh L., co. Meath. Notes by Col. Wood-Martin.
Muickeanagh L. (Lough Leisi), co. Roscommon. B. 18, p. 230.
Mucknoe L., co. Monaghan. B. 444, p. 151.
Muintir Eolais, co. Leitrim. "Annals of the Four Masters"; B. 18, p. 231; B. 9, p. 45.
Nahinch L., Tipperary and King's County. B. 70b; Wakefield's "Ireland," vol. i. p. 94.
Naneevin L., co. Galway. B. 118, p. 31.
Ooney L. ("Loch-n-Uithne"), co. Monaghan. "Annals of Lough Cé"; B. 444, p. 156.
Owel L., co. Westmeath. Pro. R. I. A., vol. ix. p. 210.
Pad or Boat L., near Lough Eyes, co. Fermanagh. B. 217, p. 318.
Ravel L., "Derryhollow," "Aghaloughan," near Randalstown, co. Antrim. B. 24 and 25; B. 148, p. 22; B. 215, pp. 74 and 194; Kilk. A. S., 2nd S., vols. iii. p. 88, and iv. p. 36.
[95]Rahan's L., near Carrickmacross, co. Monaghan. K. A. S., vol. iv., 2nd S., p. 379.
Ramor L., co. Cavan. B. 438, p. 152.
[96]Rinn L., three or four crannogs, co. Leitrim. B. 24, p. 147; B. 440, p. 408.
Ross L., near Crossmaglen, co. Armagh. J. R. H. A. As., vol. vi., 4th S., p. 432.
Roughan L., near Dungannon, co. Tyrone. B. 438, p. 152.
Rouskey L., co. Monaghan. B. 444, p. 151.
Scur L., two crannogs, co. Leitrim. "Annals of Lough Cé," 1345, 1390, and 1580; B. 13, Ap. p. 43. B. 18, p. 223.
[97]St. John's Lough, four crannogs, co. Leitrim. B. 13, Ap., pp. 43 and 59.
Talogh L., at Feenagh, several crannogs, co. Leitrim. B. 440, p. 408.
The Miracles, co. Fermanagh. B. 217, p. 319, and B. 346, p. 331.
Toome Bar, co. Antrim. B. 92, p. 227.
[98]Tully L., three crannogs, co. Cavan. B. 438, p. 150.
Tullyline, co. Cavan. B. 13, p. 215.
Veagh L., co. Donegal. "Annals," B. 18, p. 231.
Yoan L., co. Fermanagh. B. 217, p. 324.

II.—SCOTTISH LAKE-DWELLINGS OR CRANNOGS.

It was not till after the discoveries on the Continent had attracted universal attention that archæologists began to look for similar remains in Britain. It was then found that early historic references to island forts, and some incidental notices of the exposure of buried islands artificially constructed of wood and stone, and other remains of lacustrine abodes, during the drainage of lochs and marshes in the last and early part of this century, had been entirely overlooked. The merit of correctly interpreting these remains in Scotland, and bringing them systematically before antiquaries, belongs to the late Joseph Robertson, Esq., F.S.A., Scotland, who read a paper on the subject to the Society of Antiquaries of Scotland on the 14th of December, 1857, entitled, "Notices of the Isle of the Loch of Banchory, the Isle of Loch Canmor, and other Scottish examples of the artificial or stockaded islands, called crannogs in Ireland, and Keltischen Pfahlbauten in Switzerland."

Mr. Robertson's paper, though not published, at once attracted attention, and stimulated so much further inquiry on the part of the members, that, at the very next meeting of the Society, another contribution on the subject was read by Mr. John Mackinlay, F.S.A., Scotland (B. 21), from which it appeared that as early as 1812 this gentleman had observed some remains (now surmised to be a crannog) in Dhu Loch, in the island of Bute, which were described in a letter dated the 13th February, 1813. This communication found its way to George Chalmers, Esq., author of "Caledonia," regarding which, writing on the 26th of April, 1813, he says:—"It goes directly to illustrate some of the obscurest antiquities of Scotland. I mean the wooden castles, which belong to the Scottish period when stone and lime were not much used in building. I will make proper use of this discovery of Mr. Mackinlay." In 1863, Dr. John Grigor, of Nairn, described "two ancient lake-dwellings or crannoges in the Loch of the Clans, Nairnshire." (B. 55.) The remains, however, were too imperfect to be of value in illustrating their structure, and the only relics found were a portion of a small stone cup or lamp, two whetstones, an iron axe-head, and some charcoal and bits of bone.

A more important discovery, made about the same time, was a group of artificial islands in Loch Dowalton, Wigtownshire, which were first described by his Grace the Duke of Northumberland (then Lord Lovaine) in a paper read at the Newcastle-upon-Tyne meeting of the British Association in 1863. (B. 56.) About two years later Mr. John Stuart, Secretary to the Society of Antiquaries of Scotland, visited Dowalton, and, owing to a greater drainage of the loch having been made in the interval, was enabled to re-examine the Dowalton islands under more favourable circumstances. The result of his labours was an elaborate paper to the Society, in which he gave a detailed account of the structure and relics of these crannogs, and also took the opportunity of incorporating into his article all the facts he could glean, so as to afford a basis for comparing the Scottish examples with those in other countries. (B. 94.)

Since the publication of Dr. Stuart's paper in 1866, little progress was made in the investigation of Scottish crannogs, though traces of them were occasionally noticed in various parts of Scotland, till the discovery and examination of the Lochlee crannog, Ayrshire, in 1878-9. The work done at Lochlee was important, not only because of the varied collection of relics secured, but also on account of the interest it had excited in archæological research, the fruit of which has already been reaped by the discovery of no less than five other lake-dwellings in the south-west of Scotland, all of which have now been carefully investigated. Full details of these investigations are given in the Collections of the Ayrshire and Galloway Archæological Association, as well as in my recent work on the "Scottish Lake-Dwellings." (B. 331, 344, 373, and 426.)

While such general indications of lake-dwellings can hardly be said to limit their geographical distribution to any given area in Scotland, it is a singular fact that, so far as the discovery of actual remains illustrative of the civilisation and social condition of their occupiers is concerned, we are almost entirely dependent on the investigations made at Dowalton, Lochlee, Lochspouts, Buston, Airrieoulland, Barhapple, and Friar's Carse, all of which are situated within the counties of Ayr, Dumfries, and Wigtown. In instituting a comparison between the relics of these respective groups their resemblance is so wonderfully alike that we have no difficulty in dispensing with the necessity of discussing the merits of each group separately; so that whatever inferences can be legitimately derived from a critical examination of any one group may be safely applied to the whole.

As a preliminary to this inquiry the following details of the investigation of lake-dwellings in Scotland will, I trust, be sufficient to give general readers a tolerably correct notion of the social conditions and environments of the people whose history, solely from an archæological standpoint, it is our object here to pourtray.

DOWALTON.

The loch of Dowalton was of an irregular form, about 1½ mile long, and about three-quarters of a mile in greatest breadth, and without any marked outfall for drainage. Sir William Maxwell effected this by making a cut, 25 feet deep, through the wall of whinstone and slate which closed it in at its south-eastern boundary. When the waters were allowed to run off in the summer of 1863 no less than five artificially constructed islands became visible. One of these had a cairn of stones on it which always remained above water, and was known as the "Miller's Cairn," from the fact of its having been used, like the Nilometer, to measure the quantity of water in the lake, and thereby to regulate its supply to neighbouring mills. "On approaching the cairn," says Dr. Stuart, "the numerous rows of piles which surrounded it first attracted notice. These piles were formed of young oak-trees. Lying on the north-east were mortised frames of beams of oak, like hurdles, and, below these, round trees laid horizontally. In some cases the vertical piles were mortised into horizontal bars. Below them were layers of hazel and birch branches, and under these were masses of ferns, the whole mixed with large boulders, and penetrated by piles. Above all was a surface of stone and soil, which was several feet under water till the recent drainage took place. The hurdle frames were neatly mortised together, and were secured by pegs in the mortise holes. On one side of the island a round space of a few feet in size appeared, on which was a layer of white clay, browned and calcined as from the action of fire, and around it were bones of animals and ashes of wood.... Lines of piles, apparently to support a causeway, led from it to the shore." (B. 94.)

The other islands were constructed in a similar manner, and of like materials. The largest and farthest from the shore measured twenty-three yards in diameter, and its surface was three feet lower than that of the others. Several canoes and bronze dishes were found in the mud in the vicinity of these islands, and in making excavations on them many relics and broken bones were collected, of which the following is a list as far as known up to the present time:—

RELICS FROM THE CRANNOGS IN LOCH DOWALTON.

Metal.—"Pot or patella of yellowish-coloured bronze (Fig. 126), with a handle springing from the upper edge, 7 inches in length, on which is stamped the letters P. CIPIPOLIBI. At the farther extremity is a circular opening. The bottom is ornamented with five projecting rings, and measures in diameter 6 inches; it is 8 inches in diameter across the mouth; the inside appears to be coated with tin, and has a series of incised lines at various distances. The vessel is ornamented on the outside opposite to the handle by a human face in relief, surrounded by a movable ring, which could be used in lifting the pot." (B. 94, p. 109.)

A bronze basin, measuring 10 inches in diameter and 4 inches in depth, shows several patches or mendings (Fig. 127). It is formed of several separate pieces of sheet-metal riveted together, and appears to have had an iron handle.

Two bronze dishes, hammered out of the solid. One measures 12 inches in diameter and 4 inches in depth. The other has the same diameter as the former, but is 1 inch less in depth, and has a turned-over rim 1 inch in breadth (Fig. 128).

A bronze ring, having attached to it a portion of the vessel of which it had been a handle (Fig. 129).

A penannular brooch and a circular ornament, with trumpet-shaped spaces, probably intended for enamel[99] (Fig. 130). Also a small ring, a fragment of bronze, and iron slag.

Three iron hammers or axe-hammers (Fig. 131).

Glass.—Two beads of earthenware of a ribbed pattern, showing traces of a green glaze; one of vitreous paste of a whitish colour, with red spots; half of another bead of white glass, streaked with blue; and a large bead, 1¼ inch in diameter, of a somewhat remarkable character. The latter has in the central perforation a tube of bronze, and the edge of both sides of the perforation is ornamented by three minute bands of twisted yellow glass, while the body is of blue glass, of a ribbed pattern (Fig. 132).

Several portions of armlets of glass. Half of one is of white glass, and streaked with blue. Others are of a yellowish or whitish colour (Fig. 132). A small portion of blue glass.

Leather.—Portion of a leather shoe, 7 inches in length, nearly covered with ornamental stamped patterns (Fig. 133).

Stone.—A thin flat stone, of a rectangular shape and highly polished, is supposed to have been used as a mirror; five querns, a flake of yellow flint, and several whetstones.

Pottery.—A small fragment of Samian ware and an earthen crucible (Fig. 134).

Miscellaneous.—A small shale ring, unfinished, a bead of amber, and a copper coin.

LOCHLEE.

The site of the Lochlee crannog was a small lake, which formerly occupied about nineteen acres of what is now, and has been for many years, arable land, on the farm of Lochlee, near Tarbolton. Before it was artificially drained, some fifty years ago, no one appears to have surmised that a small island, which became visible in the summer-time, and formed a safe habitation for gulls and other sea-birds during their breeding season, had been formerly the residence of man; nor does it appear to have attracted the attention of the poet Burns, who lived on this farm for four years as ploughman to his father, then tenant of the place. The crannog was near the outlet, of the lake, and the nearest land, its southern bank, was about seventy-five yards distant. When the first drainage of the lake was carried out, the wrought woodwork exposed in the drains running through the island, and especially the discovery of two canoes buried in the moss, attracted some attention. It was not, however, till 1878, in consequence of some discoveries during the re-drainage of the locality, that this most important crannog was investigated.