surface, on the farm of Luggtonrigge, near Giffin Castle, Ayrshire. The shields were regularly disposed in a circle, and one of them, which passed into the possession of Dr. Ferris, was subsequently presented by him to the Society of Antiquaries of London. It has a semiglobular umbo, surrounded by twenty-nine concentric rows of small studs, with intervening ribs, and measures 26 inches in diameter.1 Like all the primitive British bucklers, it will be seen that it was designed to be held in the hand, the raised umbo in the centre being hollow to receive and protect the hand where it grasped the cross-bar, seen on the under side in the annexed engraving. This central umbo is invariably surrounded with a series of rings in relief, with studs between; and the two pins seen on the inner side have perhaps secured a strap for suspending it to the neck of the wearer when not in use. Two remarkably fine bronze shields of this description, twenty-four inches in diameter, and with twenty-four concentric circles,-exhibited to the Society of Antiquaries of Scotland by Mr. George Wauchope of Niddry, in 1837, and since added to the collection,- were found near Yetholm, about eight miles from Kelso, at a depth of four feet, by a labourer engaged in digging a drain. Such is not the form of shield introduced on the gold coins of Tasciovanus, Cunobelin, and others of the native rulers contemporary with the first intercourse with Rome. On one of the coins of Tasciovanus a horseman wears a long double-pointed shield, and others, though round, are large, dished, and of very different construction from those described here, and supposed to pertain to an earlier period. Sir Robert

1 Catalogue of Antiquities, etc., Soc. Antiq. Lond. 1847, by Albert Way, Esq., p. 16. Mr. Way adds in a note, "The description of the shield found in Ayrshire, as given in the minutes, corresponds with the buckler now in the Society's possession in every particular, with the exception of the diameter, which is stated to have been about 15 inches, possibly an error of transcript."

Sibbald describes among Scottish antiquities obtained on the sites of ancient camps, "pieces of harness of brass : some for the arms and some for the legs. Shields also are found; some oblong and oval, and some orbicular. Some of these are of brass, and some of wood full of brass nails."1 It is probable that many of the shields of the same period were made chiefly of wood and leather, with the central umbo of bronze. In the later AngloSaxon grave the iron umbo and other metal portions of the perished shield are of common occurrence; and in the circular Highland target, still to be met with among collected relics of the clans, we find a curious imitation of the earlier model. Though the Roman fashion of wearing the shield on the arm has been followed by the Scottish mountaineer, rendering the hollow umbo no longer of use, yet it appears to the last in the boss of his target furnishing another striking illustration of the unreasoning tenacity with which the Celtic race clings to ancient customs, and perpetuates, amid all the progressive civilisation with which it is surrounded, customs and traditions inherited from remote pagan centuries.

Among the specimens of defensive armour preserved in the Museum of the Scottish Antiquaries, are two pieces of thin copper, decorated with indented ornaments, which were presented to the Society by Sir George Mackenzie of Coull, Bart., in 1828. They are described by the donor as pieces of copper, supposed to be plate armour, or the covering of a shield, found in a cairn, under an oak-tree at Craigdarroch, Ross-shire. Various other portions were found along with these, and their appearance seems to justify the supposition of the donor. In the autumn of 1849 a remarkable discovery of bronze arms and other antiquities was made in the Isle of Skye. They included swords, spear-heads, celts, and a bronze

1 Portes, Coloniæ, etc. App. pp. 17, 18.

pin with a hollow cup-shaped head similar to one figured in the Archæological Journal: a relic of one of the Irish Crannoges, or island strengths. A gold armilla and other ornaments of the same precious metal are also said to have been obtained along with these ancient remains, and beside them lay the fragments of an oaken chest in which the whole appeared to have been deposited. The most of those valuable relics were secured by Lord Macdonald, but one curious and probably unique implement fell into private hands, and has since been deposited in the Museum of the Scottish Antiquaries. In general appearance it resembles a bent spear-head; but it has a

raised central ridge on the inside, while it is nearly plain and smooth on the outer side. Its hollow socket is perforated with holes for securing it to a handle by means of a pin. The most probable use for which it has been designed would seem to be for scraping out the interior of canoes and other large vessels made from the trunk of the oak. But we necessarily reason from very imperfect data when we ascribe a specific purpose to the implements of a period the arts and habits of which must have differed so essentially from our own.

Another class of bronze implements, includes what are generally described as sickles, or reaping and pruning hooks. One of these, which was found at a depth of six

1 Archæol. Jour. vol. iii. p. 48.

feet in a bog in the neighbourhood of Ballygawley, county of Tyrone, now preserved in the British Museum, is figured in the Archæological Journal. Another, engraved in General Vallancey's Collectanea, is described as "a small securis, called by the Irish a searr, to cut herbs, acorns, misletoe, etc.," and a fine series, varying in form and decorative details enriches the collection of the Royal Irish Academy. Among older writers on antiquities such relics were invariably described as the pruning-hooks with which the Druid priests were wont to cut the sacred misletoe. About the year 1790, an instrument of this class was discovered at Ledberg, in the county of Sutherland, by some labourers cutting

3

peats, and was pronounced by the Earl of Bristol, then Bishop of Derry, to whom it was presented, to be a Druidical pruning-hook, similar to several found in England. The example here engraved (Fig. 70) was dredged up in the river Tay, and is now preserved in the Perth Museum. Perhaps among the same relics of primitive agricultural skill ought also to be reckoned a curious weapon or implement of bronze, occasionally found in Scotland, two examples of which are shown in Fig. 71. One of them is from the original in the Museum of the Scottish Antiquaries. This was found among the

1 Archeol. Jour. vol. ii. p. 186.

2 No. 13, Plate x. Fig. 4.

3 Catalogue of the Museum of the Royal Irish Academy, vol. i. Sinclair's Statist. Acc. vol. xvi. p. 206.

VOL. I.

2 C

remains of many large oak-trees, on the farm of Rottenmoss or Moss-side, in the vicinity of Crossraguel Abbey, Ayrshire, and is not inaptly described by its donor as nearly resembling one of the common forms of the Malay Creess. It measures fourteen inches in length. The other and more finished implement of the same kind is in the collection formed by the distinguished Scottish antiquary, Sir John Clerk, at Penicuick House. It is furnished with a hollow shaft or socket for the handle. The same interesting and valuable collection includes other specimens of this primitive implement, constructed like that in the Museum of the Scottish Antiquaries, with only a metal spike for insertion into the haft.

FIG. 71. Bronze Reapers.

Some examples of this relic of old agricultural skill are of extremely small dimensions, measuring only from six to eight inches in the length of the blade, and should perhaps more correctly be described as pruning-hooks or knives. But in this, as in so many other attempts to assign a use to obsolete implements, the most probable suggestions of their original purpose are at best but guesses at the truth.

Such may suffice in illustration of some of the most characteristic weapons and implements of the Bronze Period. In the detailed enumeration of specific examples the course of inductive reasoning has been necessarily interrupted; but in this and subsequent accumulations of needful details, the argument is in reality presented