CHAPTER XXI.

CHAPTER XXI.METAL, MOULDS, AND THE METHOD OF MANUFACTURE.Havingnow passed in review the various forms of weapons, tools, ornaments, and vessels belonging to the Bronze Period of this country, it will be well to consider the nature of the metal of which they are formed, and the various processes by which they were produced and finished ready for use. Some of these processes, as for instance the hammering out of the cutting-edges of tools and weapons, and the production of ornamental designs by means of the hammer and punch, have already been mentioned, and need be but cursorily noticed. The main process, indeed, of which this chapter will treat is that of casting.Bronze, as already stated, is an alloy of copper and tin, and therefore distinct from brass, which is an alloy of copper and zinc. Many varieties of bronze—or, as it is now more commonly called, gun-metal—are in use at the present day; and one remarkable feature in bronze is that the admixture with copper of the much softer metal tin, in varying proportions, produces an alloy in most if not all cases harder than the original copper; and when the tin is much in excess, as in the metal used for the specula of telescopes, so much harder that,à priori, such a result of the mixture of two soft metals would have been thought impossible. The following table compiled from a paper inDesign and Work, reprinted in Martineau and Smith’sHardware Trade Journal,[1602]gives some of the alloys now in most common use and the purposes to which they are applied:—Tin.Copper.Per cent.of Copper.11108=90·76{A common metal for cannon and machine brasses, used also for bronze statues.1199=90·00}Gun-metal proper, used for cannon.1196=89·721184=88·44{Used for bearings of machinery, frequently called gun-metal.1172=86·75Rather harder.1160=84·50Harder, not malleable.1144=80·00Used for cymbals and Chinese gongs.1148=81·35Very hard, used for culinary vessels.1136}or{76·69}Bell-metal.123675·601124=68·57}Yellowish, very hard, sonorous.114=26·66Very white, sometimes used for specula with some other slight admixture.Lord Rosse, however, in casting specula, preferred using copper and tin in their atomic proportions, or 68·21 per cent. of copper and 31·79 of tin.The addition of tin, while increasing the hardness of copper, also renders it more fusible. In small proportions it but little affects the colour of the copper,[1603]and it is difficult to recognise its presence from the physical characters of the copper, except from that of increased hardness. What appear, therefore, to be copper instruments may, and indeed often do, contain an appreciable admixture of tin, which, however, can only be recognised by analysis.Besides the superiority of one alloy over another, it appears probable that the method of treatment of the metal may somewhat affect its properties. M. Tresca[1604]found that a gun-metal cast by Messieurs Laveissière, consisting of—Copper89·47Tin9·78Zinc0·66Lead0·09was superior in all respects to either the common gun-metal A or the phosphor-bronze B cast at Bourges, the constituents of which were as follows:—ABCopper89·8790·60Tin9·458.82Zinc0·310·27Lead0·370·31——–——–100·00100·00The results of both ancient and modern experience as to the proportions in which copper and tin should be mixed, in order to produce a tough and hard though not brittle metal, appear to be nearly the same; and nine parts of copper to one part of tin may be regarded as the constituents of the most serviceable bronze or gun-metal.In the following table I have given the results of some of the more recent analyses of bronze antiquities found in the United Kingdom, and have omitted the early analyses of Dr. Pearson[1605]in 1796 as being only approximative. I have arranged them so far as practicable in accordance with the different forms of the objects analyzed; and one feature which is thus brought out tends strongly to confirm the conclusion which has been arrived at from other premises, that certain forms of bronze weapons and other instruments and utensils are of later date than others.It will be seen, for instance, that in the flat and flanged celts, the palstaves, and even spear-heads, lead, if present at all, exists in but very minute quantity; whereas in the socketed celts and swords, which are probably later forms, and especially in those from Ireland, this metal occurs in several cases in considerable proportions.This prevalence of lead is very remarkable in some of the small socketed celts found in very large numbers in Brittany, which from their diminutive size have been regarded as “votive” rather than as destined for actual use. In some of these Professor Pelligot[1606]found as much as 28·50 and even 32·50 per cent. of lead, with only 1½ per cent. or a small trace of tin. In others, with a large per-centage of tin, there was from 8 to 16 per cent. of lead. Some of the bronze ornaments of the Early Iron Period also contain a considerable proportion of this metal, which, in the early Romanas[1607]and its parts, is found to the extent of from 20 to 30 per cent. Although some such proportion as 9 to 1 appears to have been aimed at, there is great variation in the proportions of the principal ingredients even in cutting tools of the same general character, the tin being sometimes upwards of 18 per cent. and sometimes less than 5 per cent. of the whole.This variation was no doubt partly due to occasional scarcity of tin; but, as Dr. W. K. Sullivan has pointed out,[1608]there are twoother causes for it: first, the separation of the constituent metals in the fused mass, and the accumulation of the tin in the lower portion of the castings; and, second, the throwing off of the tin by oxidation when the alloys were re-melted. M. Dusaussoy[1609]found that an alloy containing 90·4 per cent. of copper and 9·6 per cent. of tin lost so much of the latter metal by six fusions that it ultimately consisted of 95 per cent. of copper and only 5 per cent. of tin.With regard to the early sources of the copper and tin used in this country, and in general through Western Europe, it will not be in my power to add much to what has already been published on this subject.It seems probable that gold, which commonly occurs native and brilliant, was the first metal that attracted the attention of mankind. The next metal to be discovered would, in all probability, be copper, which also occurs native, and has many points of resemblance with gold.The use of this metal, as I have observed in the Introductory Chapter, no doubt originated in some part of the world where, as on the shore of Lake Superior, it occurs in a pure metallic state. When once it was discovered that copper was fusible by heat, the production of the metal from some of the more metallic-looking ores, such as copper pyrites, would follow; and in due time, either from association with the metal, or from their colour and weight, some of the other ores, both sulphuretted and non-sulphuretted, would become known.[1610]When once the production of copper in this manner was effected, it is probable that the ores of other metals, such as tin, would also become known, and that tin ores would eitherbe treated conjointly with the ores of copper, as suggested by Dr. Wibel, so as at once to produce bronze; or added to crude copper, as suggested by Professor Sullivan; or again, be smelted by themselves so as to produce metallic tin. At what date it was generally known that “brass is molten out of the stone”[1611]is, however, a question difficult to answer.Native copper and many of its ores occur in Hungary, Norway, Sweden, Saxony, and Cornwall; but copper pyrites is far more generally distributed, and is found in most countries of the world. So far, therefore, as the existence of this metal is concerned, there was no necessity for the Britons in Cæsar’s time to make use of imported bronze, especially as tin was found in abundance in Cornwall, and long before Cæsar’s time was exported in considerable quantities to the Continent. And yet his account may to some extent be true, as a socketed celt of what is almost undoubtedly Breton manufacture has been found near Weymouth,[1612]and several instruments of recognised French types have been found in our southern counties. Bronze vessels also may have been imported.Copper and its ores are abundant in Ireland, especially copper pyrites and gray copper.Although tin was formerly found in abundance in some parts of Spain, and also in less quantity in Brittany,[1613]there can be but little doubt that the Cassiterides, with which either directly or indirectly the Phœnicians traded for tin,[1614]are rightly identified with Britain. But, with due deference to Professor Nilsson and other antiquaries, I must confess that the traces of Phœnician influence in this country are to my mind at present imperceptible; and it may well be that their system of commerce or barter was such as intentionally left the barbarian tribes with whom they traded in much the same stage of civilisation as that in which they found them, always assuming that they dealt directly with Britain and not through the intervention of Gaulish merchants.The argument, however, that the Phœnician bronze would have been lead-bronze, because the Phœnicians derived their civilisation and arts from Egypt, and had continual intercourse with that country, where lead-bronze was early known, appears to me wanting in cogency. For though the Egyptians may have usedlead-bronzes for statues and ornaments, the Egyptian dagger[1615]analyzed by Vauquelin gave copper 85, tin 14, and iron 1 per cent., and showed no trace of lead. Of one point we may be fairly certain, that the discovery of bronze did not originate in the British Isles, but that the knowledge of that useful metal was communicated from abroad, and probably from the neighbouring country, France. When and in what manner that and the other countries of Western and Central Europe derived their knowledge of bronze it is not my intention here to discuss. I will only say that the tendency of the evidence at present gathered is to place the original source of bronze, like that of the Aryan family, in an Asiatic rather than an European centre.The presence in greater or less proportions of other metals than copper and tin in bronze antiquities may eventually lead to the recognition of the sources from which in each country the principal supplies of metal were obtained. Professor Sullivan, in the book already cited, arrives at the following among other conclusions from the chemical facts at his command:—1. The northern nations in ancient times used only true bronzes—those formed of copper and tin—of greater or lesser purity according to the kind of ores used.2. Many of these bronzes contain small quantities of lead, zinc, nickel, cobalt, iron, and silver, derived from the copper from which the bronze was made.3. Though some bronzes may have been produced directly by melting a mixture of copper and tin ores, the usual mode of making them was by treating fused crude copper with tin-stone.[1616]In later times bronze was made by mixing the two metals together.4. The copper of the ancient bronzes seems to have been smelted in many different localities.Some analyses of bronze antiquities found in other countries are given in the works indicated below,[1617]in addition to those mentioned on page 418.ANALYSES OF BRONZE ANTIQUITIES.* In this case oxygen to the extent of 3·83 was present. The bronze had become so friable as to be easily pulverised in a mortar. Mr. J. Arthur Phillips writes about it as follows:—“When a freshly-broken fragment of it is examined under a low magnifying power, it is seen to consist of a metallic net-work enclosing distinct and perfectly formed crystals of cuprite, surrounded by a greyish white substance which is chiefly bioxide of tin. In this alloy the nickel, silver, and iron are evidently accidental impurities, but the lead is no doubt an intentional ingredient.” The specific gravity after pulverization is about 7·26 only.** Specific gravity 8·59.I have here given most of the trustworthy analyses already published, and have only added two new analyses kindly made for me by Mr. J. A. Phillips, F.G.S., of a socketed celt from Yorkshire and of a small dagger from Newton, near Cambridge.Those who wish for detailed information as to the composition of the bronze antiquities found in other countries are referred to De Fellenberg’s essays and to Von Bibra’s comprehensive work.[1618]The copper which was used by the bronze-founders of old times appears to have been smelted from the ore and run into a shallow concave mould open at top, in which the metal assumed the form of a circular cake, convex below and flat above; but before becoming sufficiently cold to be quite set into tough metal, these cakes seem as a rule to have been disturbed and broken up into numerous pieces, better adapted for re-melting than the whole cakes would have been. This method of breaking up the solid cakes while hot saved also an infinity of labour; as to cut such masses into small pieces when cold would, even with modern appliances, be a difficult task; and with only bronze and stone tools at command would have been nearly impossible. Many of the cakes are, however, interspersed with cavities formed in the metal, and in some cases there seems reason to think that this may have been produced intentionally, so as to render the breaking of the cakes even when cold more readily practicable.Many of the blocks of metal cast in rough moulds, and known by Italian antiquaries asæs signatum, have a similar broken appearance at the ends. Professor Chierici[1619]has suggested that the moulds in which they were cast were of considerable length, and that from time to time clay and sand were thrown in so as to break the continuity of the metal, which indeed was poured in at intervals, after the insertion of the sand or clay, to form the break in the mould.Some pieces of metal which have been regarded as ingots, and which not improbably are really such, have the form of a double-ended axe with a very small shaft hole. They have been discovered with several of the bronze-founders’ hoards in France. Dr. V. Gross, of Neuveville, has a fine example of this kind found at Locras, in the Lac de Bienne.[1620]It is about 16½ inches long and 4¾ inches wide at the ends, the hole through the centre beingabout ¼ inch in diameter, and the weight of the ingot, which is of pure copper, is about 6½ lbs.Rough lumps of metal have frequently been found with deposits of bronze implements in Britain, these latter being sometimes in a worn-out or broken condition, and apparently brought together as old metal for re-casting. In other deposits the instruments seem new and ready for use, or again they are in an unfinished condition. All the circumstances of these discoveries, however, go to prove that they are in fact the stock-in-trade of the ancient bronze-founders. The jets or waste pieces from the castings, of which I shall subsequently have to speak, are often found mixed with the rude lumps. These lumps have usually the appearance of pure copper, and in many cases have proved to be so on analysis.Some copper cakes appear, however, to belong to Roman times. They differ in shape from those already described, in being of nearly even thickness, but with the edge inclined as if they had been cast in a small frying-pan. They are from 10 to 13 inches in diameter and about 2 inches thick; and on more than one found in Anglesea[1621]there are inscriptions in Roman characters. They weigh from 30 to 50 lbs.Turning now to the instances of lumps of rough metal being found with bronze weapons and tools, the following may be cited, though other instances are given in the tables at page 462:—Lanant, Cornwall,[1622]heavy lumps of fine copper, found with broken socketed celts, &c.Kenidjack Cliff, Cornwall,[1623]with palstaves and socketed celts.St. Hilary, Cornwall,[1624]lumps weighing 14 or 15 lbs. each, said to have been found with spear-heads.Near Worthing, Sussex, several lumps of metal, with palstaves and socketed celts.Beachey Head,[1625]three lumps of raw copper, apparently very pure, with palstaves, socketed celts, &c.Wick Park, Stogursey, Somerset,[1626]with palstaves, socketed celts, broken swords, spears, &c.Kingston Hill, Surrey,[1627]with socketed celts, fragments of swords, and spear-head.Beddington, Surrey,[1628]with mould, socketed celts, gouge, spear-heads, &c.Wickham Park, Croydon, Surrey,[1629]with palstave, gouge, hammer, &c.Danesbury, near Welwyn, Herts,[1630]lumps of metal with damaged socketed celts.Cumberlow, Herts,[1631]with palstaves, socketed celts, fragments of swords, &c.Westwick Row, Hemel Hempsted,[1632]several lumps, with socketed celts.Romford, Essex,[1633]lumps of metal in waste pieces and imperfect castings, untrimmed socketed celts, &c.Fifield, Essex,[1634]upwards of 50 lbs. of metal, with socketed celts.High Roding, Essex,[1635]with socketed celts, &c.Kensington,[1636]with socketed celt, gouge, &c.Sittingbourne, Kent,[1637]with socketed celts, gouges, &c.Meldreth, Cambs,[1638]with socketed celts, chisel, ring of caldron, &c.Carlton Rode, Norfolk,[1639]lumps of metal, with socketed celts, gouges, &c.Helsdon Hall, Norwich,[1640]pieces of copper, socketed celts, &c.Earsley Common, York,[1641]several lumps of metal, with nearly a hundred socketed celts.Martlesham, Suffolk,[1642]a large quantity of metal, including some lumps weighing 5 or 6 lbs., with socketed celts, gouge, &c.West Halton, Lincolnshire,[1643]with socketed celts and broken sword.Roseberry Topping, Yorkshire,[1644]with socketed celts, gouges, hammer, &c.In the Heathery Burn Cave, Durham, and in the Guilsfield find, there was in each case at least one lump of metal.Besides the cakes of copper, bars of that metal appear to have been hammered into an oblong form, and then cut into lengths of from 4 to 5 inches, weighing each about ¼ lb., and in that state to have served as the raw material for the bronze-founders. Thirteen of these short bars were found at Therfield, near Royston, Herts,[1645]and Dr. Percy found on analysis that they contained about 98½ per cent. of copper with a small alloy of tin or antimony, probably the latter. Some fifteen or sixteen “pieces of long triangular brass” are described as having been found with about the same number of celts at Hinton, near Christchurch, Hants.[1646]These bars “seemed to be pieces of the metal out of which the celts were cast.”In Scotland some “lumps of brass” were found with the swords, spears, &c., in Duddingston Loch.[1647]Probably other lumps of metal have been found in that country, but they seem to be scarcer in Scotland and Ireland than in England.Although, as already observed, Spain may have been the principal Western source of tin in early times, and possibly Malacca[1648]in the East, the trade with Britain for that metal musthave commenced at a very remote epoch. We might expect, therefore, that fragments of tin would be frequently found in the old bronze-founders’ hoards. But though lumps of copper have so often been discovered in them, tin is at present conspicuous by its absence. The only instance to which I am able to refer is the discovery at Achtertyre,[1649]Morayshire, of four “broken bits of tin,” in company with socketed celts, spear-heads, and bracelets. These pieces seem to be fragments of a single bar which was about 6 inches in length, of oval section, and somewhat curved, and in weight about 3 ounces. Though spoken of as tin, the metal is in fact a soft solder composed, according to Dr. Stevenson Macadam, of—Tin78·66Lead21·34——–100·00This, he points out, is a more fusible alloy than the ordinary plumbers’ solder, which consists of 1 of tin to 2 of lead, and fuses at 441 degrees Fahr., as it contains nearly 4 of tin to 1 of lead, and would fuse at 365 degrees. Whether this bar was intended for use as solder, or represents a base tin exported to Scotland from the tin-producing districts, is an interesting question. Professor Daniel Wilson[1650]has called attention to the fact that in all the bronze instruments found in Scotland which have been submitted to analysis lead is uniformly present, though in varying proportions. Soldering[1651]is considered to have been entirely unknown in the Bronze Age, and even during the earlier times of the Iron Age; but the art of burning bronze on to bronze was certainly known, and instances of its having been practised are given in preceding pages.Some fragments of pure metallic tin have from time to time been found on the Continent. A small hammered bar found at the Lake-dwelling of Estavayer,[1652]and analyzed by M. de Fellenberg, was free from lead, zinc, iron, and copper.Besides being found in Cornwall, tin occurs in France,[1653]Saxony, Silesia, Bohemia, Sweden, Spain, and Portugal. It also occurs in Etruria,[1654]and is said to be found in Chorassan.[1655]This metal is said by Dionysius[1656]to have been struck into coins at Syracuse, but none such are at present known. Among the Ancient Britons,[1657]however, tin coins cast for the most part in wooden moulds were in circulation, not in the tin-producing districts, but in Kent and the neighbouring parts of England. Their date is probably within a century of our era, either before or after Christ.Fig. 514.—Falmouth.1/12A large ingot of tin, in shape like the letterH, was dredged up in Falmouth harbour.[1658]It is 2 feet 11 inches long and about 11 inches wide, and 3 inches thick, and, though a small piece has been cut off at one end, it still weighs 158 lbs. It is shown in Fig. 514. The late Sir Henry James, F.R.S.,[1659]has pointed out that the form in which the ingot is cast adapts it for being laid in the keel of a boat, and for being slung on a horse’s side, two of them thus forming a proper load for a pack-horse. He has also suggested that this was the form of ingot in which the tin produced in Cornwall was transported to Gaul, and thence carried overland, as described by Diodorus Siculus, to the mouths of the Rhone. Curiously enough this author speaks of the blocks being in the form of astragali, with which this ingot fairly coincides. Other ingots[1660]of tin of different form have also been found in Cornwall, but there appears to me hardly sufficient evidence to determine their approximate date, and I therefore content myself with mentioning them. A lump cast in a basin-shaped mould, with two holes in the flat face converging so as to form a V-shaped receptacle for a cord, is in the Blackmore Museum at Salisbury.What appear to be ingots of copper rather than votive or mortuary tablets have been found in Sardinia,[1661]and in their form present a close analogy with this ingot of tin, though they are of muchsmaller dimensions. Both the sides and ends curve inwards, the notch at the ends of some being semicircular. They are counter-marked with a kind of doubleT.As to the method of melting the metal but little is known. It seems probable, however, that the crucibles employed must have been vessels of burnt clay provided with handles for moving them; while for pouring out the metal small ladles of earthenware may have been used. At Robenhausen,[1662]on Lake Pfäffikon, Switzerland, small crucibles of a ladle-like form have been found, in some cases with lumps of bronze still in them. Crucibles without handles have been discovered at Unter-Uhldingen,[1663]in the Ueberlinger See.The methods of casting were various. Objects were cast—1. In a single mould formed of loam, sand, stone, or metal, the upper surface of the casting exhibiting the flat surface of the molten metal, which was left open to the air. In the case of loam or sand castings a pattern or model would be used, which might be an object already in use, or made of the desired form in wood or other soft substance.2. In double moulds of similar materials. The castings produced in this manner when in unfinished condition show the joints of the moulds. When sand was employed a frame or flask of some kind must have been used to retain the material in place when the upper half of the mould was lifted off the pattern. The loam moulds were probably burnt hard before being used. In many cases cores for producing hollows in the casting were employed in conjunction with these moulds.3. In what may be termed solid moulds. For this process the model was made of wax, wood, or some combustible material which was encased in a mass of loam, possibly mixed with cow-dung or vegetable matter, which on exposure to heat left the loam or clay in a porous condition. This exposure to fire also burnt out the wax or wood model and left a cavity for the reception of the metal, which was probably poured in while the mould was still hot.Sir John Lubbock[1664]regards this as the commonest mode of casting during the Bronze Age, but so far as this country is concernedit appears to me to have been very seldom, if ever, in use. Except in highly complicated castings, such as ring within ring, no advantage would be gained by adopting the process, as the same result could usually be obtained by the use of a mould in two halves, while the pattern would then be preserved. In comparing a number of objects together, though, like the six hundred and eighty-eight specimens of celts in the Dublin Museum, no two may appear to have been cast in the same mould, it does not follow that this was actually the case, for allowance must be made for hammering, polishing, and ornamenting, which were subsequent processes, and also for wear at the edge. Even in castings from the same metal mould there will be considerable variations, from differences in the amount of coating used to prevent the hot metal from adhering to mould, and the length stopped off by the core. But of this I shall shortly speak.The moulds formed of burnt clay have but rarely lasted to our times, though some have been found on the continent of Europe.One for a perforated axe found among the remains of Lake-dwellings near Laibach, in Carniola, is in the museum of that town. Others will subsequently be mentioned.The single moulds found within the United Kingdom are all of stone, and are adapted for the production of flat celts, rings, knives, and small chisels. In some cases it is hard to say whether a mould was intended to be used alone or in conjunction with another of the same kind, so as in fact to be only the half of a mould.The single mould, which I have engraved as Fig. 515, was found near Ballymena, Co. Antrim, and, as will be seen, is for a flat celt of the ordinary form. The material is a micaceous sandstone, which a recent possessor of the mould has thought so well adapted for use as a whetstone, that the mould is in places scored with the marks where apparently a cobbler’s awl has been sharpened. A celt cast in such a mould would be flatter on one face than the other, and be blunt at the ends, though much thinner there than in the middle. Before being used it would be submitted to a hammering process, which would render the two faces nearly symmetrical, and at the same time condense the metal and render it harder and fitter for cutting purposes, especially at the edge which was drawn out. In an Irish specimen in my collection there is in one face a deep conical depression, apparently caused by the contraction of the metal in cooling. It was probably necessary to add a little molten metal to the casting while coolingin order to avoid such defects. The sides as well as the faces of these plain celts have usually been wrought with the hammer, and it seems probable that some even of the flanged celts were originally plain castings in an open mould.Fig. 515.—Ballymena. ½Moulds of the same kind have been found, though rarely, in England. In a field near Cambo,[1665]near Wallington, Northumberland,was found a block of sandstone, having on one face two moulds for flat celts of different sizes, and on the other face another such mould, and also one for a flat ring. It is now in the British Museum.Stone blocks with moulds cut in them have been found in Scotland.One with a mould for a large celt in the centre, and near it in one corner of the slab a mould for a very small celt, was found in a cairn near Kintore, Aberdeenshire.[1666]Another large block, forming the end of a cist, near Kilmartin, Argyleshire,[1667]has nine depressions in it in the form of flat celts, which may have been used as moulds. They are barely an eighth of an inch in depth, and on this account have been thought to be pictorial representations rather than moulds. With a metal so imperfectly fluid as melted bronze, castings could be made thicker than the depth of the moulds, and it is by no means impossible that this stone and another forming part of the same cist may have been intended for the production of castings. The second slab of stone may have served for casting pins.The stone moulds from Trochrig, near Girvan, Ayrshire,[1668]and Alford, Aberdeenshire,[1669]with depressions of various forms upon them, not improbably belong to a later period than that of which I am treating.A mould for casting rings, 2½ inches in diameter, found at Kilmailie, Invernessshire, is in the Museum at Edinburgh.One for two flat celts on the one face, and for a larger celt and perhaps a knife on the other, is in the Antiquarian Museum at Edinburgh.[1670]These moulds are more abundant in Ireland.One in the Belfast Museum,[1671]polyhedral in shape, has moulds upon four of its faces for flat celts of different sizes. In the Bateman Collection is a slab of schistose stone (7 inches by 6 inches) with three such moulds upon it. It was found near Carrickfergus, Co. Antrim.[1672]On a slab in the Museum of the Royal Irish Academy[1673]there are moulds for two flat celts, and also for one with a stop-ridge and a loop. It would appear as if the founder must have possessed a second half of this latter mould.Two moulds formed of stone, and apparently intended for flat or slightly flanged celts, have been found at Bodio in the Lago di Varese.[1674]Moulds for palstaves and socketed celts have been found both of stone and of bronze, but it will be well to reserve the latter until all the forms of moulds made of stone have been considered. Such celt moulds have always been made in halves.

METAL, MOULDS, AND THE METHOD OF MANUFACTURE.

Havingnow passed in review the various forms of weapons, tools, ornaments, and vessels belonging to the Bronze Period of this country, it will be well to consider the nature of the metal of which they are formed, and the various processes by which they were produced and finished ready for use. Some of these processes, as for instance the hammering out of the cutting-edges of tools and weapons, and the production of ornamental designs by means of the hammer and punch, have already been mentioned, and need be but cursorily noticed. The main process, indeed, of which this chapter will treat is that of casting.

Bronze, as already stated, is an alloy of copper and tin, and therefore distinct from brass, which is an alloy of copper and zinc. Many varieties of bronze—or, as it is now more commonly called, gun-metal—are in use at the present day; and one remarkable feature in bronze is that the admixture with copper of the much softer metal tin, in varying proportions, produces an alloy in most if not all cases harder than the original copper; and when the tin is much in excess, as in the metal used for the specula of telescopes, so much harder that,à priori, such a result of the mixture of two soft metals would have been thought impossible. The following table compiled from a paper inDesign and Work, reprinted in Martineau and Smith’sHardware Trade Journal,[1602]gives some of the alloys now in most common use and the purposes to which they are applied:—

Lord Rosse, however, in casting specula, preferred using copper and tin in their atomic proportions, or 68·21 per cent. of copper and 31·79 of tin.

The addition of tin, while increasing the hardness of copper, also renders it more fusible. In small proportions it but little affects the colour of the copper,[1603]and it is difficult to recognise its presence from the physical characters of the copper, except from that of increased hardness. What appear, therefore, to be copper instruments may, and indeed often do, contain an appreciable admixture of tin, which, however, can only be recognised by analysis.

Besides the superiority of one alloy over another, it appears probable that the method of treatment of the metal may somewhat affect its properties. M. Tresca[1604]found that a gun-metal cast by Messieurs Laveissière, consisting of—

was superior in all respects to either the common gun-metal A or the phosphor-bronze B cast at Bourges, the constituents of which were as follows:—

The results of both ancient and modern experience as to the proportions in which copper and tin should be mixed, in order to produce a tough and hard though not brittle metal, appear to be nearly the same; and nine parts of copper to one part of tin may be regarded as the constituents of the most serviceable bronze or gun-metal.

In the following table I have given the results of some of the more recent analyses of bronze antiquities found in the United Kingdom, and have omitted the early analyses of Dr. Pearson[1605]in 1796 as being only approximative. I have arranged them so far as practicable in accordance with the different forms of the objects analyzed; and one feature which is thus brought out tends strongly to confirm the conclusion which has been arrived at from other premises, that certain forms of bronze weapons and other instruments and utensils are of later date than others.

It will be seen, for instance, that in the flat and flanged celts, the palstaves, and even spear-heads, lead, if present at all, exists in but very minute quantity; whereas in the socketed celts and swords, which are probably later forms, and especially in those from Ireland, this metal occurs in several cases in considerable proportions.

This prevalence of lead is very remarkable in some of the small socketed celts found in very large numbers in Brittany, which from their diminutive size have been regarded as “votive” rather than as destined for actual use. In some of these Professor Pelligot[1606]found as much as 28·50 and even 32·50 per cent. of lead, with only 1½ per cent. or a small trace of tin. In others, with a large per-centage of tin, there was from 8 to 16 per cent. of lead. Some of the bronze ornaments of the Early Iron Period also contain a considerable proportion of this metal, which, in the early Romanas[1607]and its parts, is found to the extent of from 20 to 30 per cent. Although some such proportion as 9 to 1 appears to have been aimed at, there is great variation in the proportions of the principal ingredients even in cutting tools of the same general character, the tin being sometimes upwards of 18 per cent. and sometimes less than 5 per cent. of the whole.

This variation was no doubt partly due to occasional scarcity of tin; but, as Dr. W. K. Sullivan has pointed out,[1608]there are twoother causes for it: first, the separation of the constituent metals in the fused mass, and the accumulation of the tin in the lower portion of the castings; and, second, the throwing off of the tin by oxidation when the alloys were re-melted. M. Dusaussoy[1609]found that an alloy containing 90·4 per cent. of copper and 9·6 per cent. of tin lost so much of the latter metal by six fusions that it ultimately consisted of 95 per cent. of copper and only 5 per cent. of tin.

With regard to the early sources of the copper and tin used in this country, and in general through Western Europe, it will not be in my power to add much to what has already been published on this subject.

It seems probable that gold, which commonly occurs native and brilliant, was the first metal that attracted the attention of mankind. The next metal to be discovered would, in all probability, be copper, which also occurs native, and has many points of resemblance with gold.

The use of this metal, as I have observed in the Introductory Chapter, no doubt originated in some part of the world where, as on the shore of Lake Superior, it occurs in a pure metallic state. When once it was discovered that copper was fusible by heat, the production of the metal from some of the more metallic-looking ores, such as copper pyrites, would follow; and in due time, either from association with the metal, or from their colour and weight, some of the other ores, both sulphuretted and non-sulphuretted, would become known.[1610]

When once the production of copper in this manner was effected, it is probable that the ores of other metals, such as tin, would also become known, and that tin ores would eitherbe treated conjointly with the ores of copper, as suggested by Dr. Wibel, so as at once to produce bronze; or added to crude copper, as suggested by Professor Sullivan; or again, be smelted by themselves so as to produce metallic tin. At what date it was generally known that “brass is molten out of the stone”[1611]is, however, a question difficult to answer.

Native copper and many of its ores occur in Hungary, Norway, Sweden, Saxony, and Cornwall; but copper pyrites is far more generally distributed, and is found in most countries of the world. So far, therefore, as the existence of this metal is concerned, there was no necessity for the Britons in Cæsar’s time to make use of imported bronze, especially as tin was found in abundance in Cornwall, and long before Cæsar’s time was exported in considerable quantities to the Continent. And yet his account may to some extent be true, as a socketed celt of what is almost undoubtedly Breton manufacture has been found near Weymouth,[1612]and several instruments of recognised French types have been found in our southern counties. Bronze vessels also may have been imported.

Copper and its ores are abundant in Ireland, especially copper pyrites and gray copper.

Although tin was formerly found in abundance in some parts of Spain, and also in less quantity in Brittany,[1613]there can be but little doubt that the Cassiterides, with which either directly or indirectly the Phœnicians traded for tin,[1614]are rightly identified with Britain. But, with due deference to Professor Nilsson and other antiquaries, I must confess that the traces of Phœnician influence in this country are to my mind at present imperceptible; and it may well be that their system of commerce or barter was such as intentionally left the barbarian tribes with whom they traded in much the same stage of civilisation as that in which they found them, always assuming that they dealt directly with Britain and not through the intervention of Gaulish merchants.

The argument, however, that the Phœnician bronze would have been lead-bronze, because the Phœnicians derived their civilisation and arts from Egypt, and had continual intercourse with that country, where lead-bronze was early known, appears to me wanting in cogency. For though the Egyptians may have usedlead-bronzes for statues and ornaments, the Egyptian dagger[1615]analyzed by Vauquelin gave copper 85, tin 14, and iron 1 per cent., and showed no trace of lead. Of one point we may be fairly certain, that the discovery of bronze did not originate in the British Isles, but that the knowledge of that useful metal was communicated from abroad, and probably from the neighbouring country, France. When and in what manner that and the other countries of Western and Central Europe derived their knowledge of bronze it is not my intention here to discuss. I will only say that the tendency of the evidence at present gathered is to place the original source of bronze, like that of the Aryan family, in an Asiatic rather than an European centre.

The presence in greater or less proportions of other metals than copper and tin in bronze antiquities may eventually lead to the recognition of the sources from which in each country the principal supplies of metal were obtained. Professor Sullivan, in the book already cited, arrives at the following among other conclusions from the chemical facts at his command:—

1. The northern nations in ancient times used only true bronzes—those formed of copper and tin—of greater or lesser purity according to the kind of ores used.

2. Many of these bronzes contain small quantities of lead, zinc, nickel, cobalt, iron, and silver, derived from the copper from which the bronze was made.

3. Though some bronzes may have been produced directly by melting a mixture of copper and tin ores, the usual mode of making them was by treating fused crude copper with tin-stone.[1616]In later times bronze was made by mixing the two metals together.

4. The copper of the ancient bronzes seems to have been smelted in many different localities.

Some analyses of bronze antiquities found in other countries are given in the works indicated below,[1617]in addition to those mentioned on page 418.

ANALYSES OF BRONZE ANTIQUITIES.

* In this case oxygen to the extent of 3·83 was present. The bronze had become so friable as to be easily pulverised in a mortar. Mr. J. Arthur Phillips writes about it as follows:—“When a freshly-broken fragment of it is examined under a low magnifying power, it is seen to consist of a metallic net-work enclosing distinct and perfectly formed crystals of cuprite, surrounded by a greyish white substance which is chiefly bioxide of tin. In this alloy the nickel, silver, and iron are evidently accidental impurities, but the lead is no doubt an intentional ingredient.” The specific gravity after pulverization is about 7·26 only.

** Specific gravity 8·59.

I have here given most of the trustworthy analyses already published, and have only added two new analyses kindly made for me by Mr. J. A. Phillips, F.G.S., of a socketed celt from Yorkshire and of a small dagger from Newton, near Cambridge.

Those who wish for detailed information as to the composition of the bronze antiquities found in other countries are referred to De Fellenberg’s essays and to Von Bibra’s comprehensive work.[1618]

The copper which was used by the bronze-founders of old times appears to have been smelted from the ore and run into a shallow concave mould open at top, in which the metal assumed the form of a circular cake, convex below and flat above; but before becoming sufficiently cold to be quite set into tough metal, these cakes seem as a rule to have been disturbed and broken up into numerous pieces, better adapted for re-melting than the whole cakes would have been. This method of breaking up the solid cakes while hot saved also an infinity of labour; as to cut such masses into small pieces when cold would, even with modern appliances, be a difficult task; and with only bronze and stone tools at command would have been nearly impossible. Many of the cakes are, however, interspersed with cavities formed in the metal, and in some cases there seems reason to think that this may have been produced intentionally, so as to render the breaking of the cakes even when cold more readily practicable.

Many of the blocks of metal cast in rough moulds, and known by Italian antiquaries asæs signatum, have a similar broken appearance at the ends. Professor Chierici[1619]has suggested that the moulds in which they were cast were of considerable length, and that from time to time clay and sand were thrown in so as to break the continuity of the metal, which indeed was poured in at intervals, after the insertion of the sand or clay, to form the break in the mould.

Some pieces of metal which have been regarded as ingots, and which not improbably are really such, have the form of a double-ended axe with a very small shaft hole. They have been discovered with several of the bronze-founders’ hoards in France. Dr. V. Gross, of Neuveville, has a fine example of this kind found at Locras, in the Lac de Bienne.[1620]It is about 16½ inches long and 4¾ inches wide at the ends, the hole through the centre beingabout ¼ inch in diameter, and the weight of the ingot, which is of pure copper, is about 6½ lbs.

Rough lumps of metal have frequently been found with deposits of bronze implements in Britain, these latter being sometimes in a worn-out or broken condition, and apparently brought together as old metal for re-casting. In other deposits the instruments seem new and ready for use, or again they are in an unfinished condition. All the circumstances of these discoveries, however, go to prove that they are in fact the stock-in-trade of the ancient bronze-founders. The jets or waste pieces from the castings, of which I shall subsequently have to speak, are often found mixed with the rude lumps. These lumps have usually the appearance of pure copper, and in many cases have proved to be so on analysis.

Some copper cakes appear, however, to belong to Roman times. They differ in shape from those already described, in being of nearly even thickness, but with the edge inclined as if they had been cast in a small frying-pan. They are from 10 to 13 inches in diameter and about 2 inches thick; and on more than one found in Anglesea[1621]there are inscriptions in Roman characters. They weigh from 30 to 50 lbs.

Turning now to the instances of lumps of rough metal being found with bronze weapons and tools, the following may be cited, though other instances are given in the tables at page 462:—Lanant, Cornwall,[1622]heavy lumps of fine copper, found with broken socketed celts, &c.Kenidjack Cliff, Cornwall,[1623]with palstaves and socketed celts.St. Hilary, Cornwall,[1624]lumps weighing 14 or 15 lbs. each, said to have been found with spear-heads.Near Worthing, Sussex, several lumps of metal, with palstaves and socketed celts.Beachey Head,[1625]three lumps of raw copper, apparently very pure, with palstaves, socketed celts, &c.Wick Park, Stogursey, Somerset,[1626]with palstaves, socketed celts, broken swords, spears, &c.Kingston Hill, Surrey,[1627]with socketed celts, fragments of swords, and spear-head.Beddington, Surrey,[1628]with mould, socketed celts, gouge, spear-heads, &c.Wickham Park, Croydon, Surrey,[1629]with palstave, gouge, hammer, &c.Danesbury, near Welwyn, Herts,[1630]lumps of metal with damaged socketed celts.Cumberlow, Herts,[1631]with palstaves, socketed celts, fragments of swords, &c.Westwick Row, Hemel Hempsted,[1632]several lumps, with socketed celts.Romford, Essex,[1633]lumps of metal in waste pieces and imperfect castings, untrimmed socketed celts, &c.Fifield, Essex,[1634]upwards of 50 lbs. of metal, with socketed celts.High Roding, Essex,[1635]with socketed celts, &c.Kensington,[1636]with socketed celt, gouge, &c.Sittingbourne, Kent,[1637]with socketed celts, gouges, &c.Meldreth, Cambs,[1638]with socketed celts, chisel, ring of caldron, &c.Carlton Rode, Norfolk,[1639]lumps of metal, with socketed celts, gouges, &c.Helsdon Hall, Norwich,[1640]pieces of copper, socketed celts, &c.Earsley Common, York,[1641]several lumps of metal, with nearly a hundred socketed celts.Martlesham, Suffolk,[1642]a large quantity of metal, including some lumps weighing 5 or 6 lbs., with socketed celts, gouge, &c.West Halton, Lincolnshire,[1643]with socketed celts and broken sword.Roseberry Topping, Yorkshire,[1644]with socketed celts, gouges, hammer, &c.In the Heathery Burn Cave, Durham, and in the Guilsfield find, there was in each case at least one lump of metal.Besides the cakes of copper, bars of that metal appear to have been hammered into an oblong form, and then cut into lengths of from 4 to 5 inches, weighing each about ¼ lb., and in that state to have served as the raw material for the bronze-founders. Thirteen of these short bars were found at Therfield, near Royston, Herts,[1645]and Dr. Percy found on analysis that they contained about 98½ per cent. of copper with a small alloy of tin or antimony, probably the latter. Some fifteen or sixteen “pieces of long triangular brass” are described as having been found with about the same number of celts at Hinton, near Christchurch, Hants.[1646]These bars “seemed to be pieces of the metal out of which the celts were cast.”In Scotland some “lumps of brass” were found with the swords, spears, &c., in Duddingston Loch.[1647]Probably other lumps of metal have been found in that country, but they seem to be scarcer in Scotland and Ireland than in England.

Turning now to the instances of lumps of rough metal being found with bronze weapons and tools, the following may be cited, though other instances are given in the tables at page 462:—

Lanant, Cornwall,[1622]heavy lumps of fine copper, found with broken socketed celts, &c.

Kenidjack Cliff, Cornwall,[1623]with palstaves and socketed celts.

St. Hilary, Cornwall,[1624]lumps weighing 14 or 15 lbs. each, said to have been found with spear-heads.

Near Worthing, Sussex, several lumps of metal, with palstaves and socketed celts.

Beachey Head,[1625]three lumps of raw copper, apparently very pure, with palstaves, socketed celts, &c.

Wick Park, Stogursey, Somerset,[1626]with palstaves, socketed celts, broken swords, spears, &c.

Kingston Hill, Surrey,[1627]with socketed celts, fragments of swords, and spear-head.

Beddington, Surrey,[1628]with mould, socketed celts, gouge, spear-heads, &c.

Wickham Park, Croydon, Surrey,[1629]with palstave, gouge, hammer, &c.

Danesbury, near Welwyn, Herts,[1630]lumps of metal with damaged socketed celts.

Cumberlow, Herts,[1631]with palstaves, socketed celts, fragments of swords, &c.

Westwick Row, Hemel Hempsted,[1632]several lumps, with socketed celts.

Romford, Essex,[1633]lumps of metal in waste pieces and imperfect castings, untrimmed socketed celts, &c.

Fifield, Essex,[1634]upwards of 50 lbs. of metal, with socketed celts.

High Roding, Essex,[1635]with socketed celts, &c.

Kensington,[1636]with socketed celt, gouge, &c.

Sittingbourne, Kent,[1637]with socketed celts, gouges, &c.

Meldreth, Cambs,[1638]with socketed celts, chisel, ring of caldron, &c.

Carlton Rode, Norfolk,[1639]lumps of metal, with socketed celts, gouges, &c.

Helsdon Hall, Norwich,[1640]pieces of copper, socketed celts, &c.

Earsley Common, York,[1641]several lumps of metal, with nearly a hundred socketed celts.

Martlesham, Suffolk,[1642]a large quantity of metal, including some lumps weighing 5 or 6 lbs., with socketed celts, gouge, &c.

West Halton, Lincolnshire,[1643]with socketed celts and broken sword.

Roseberry Topping, Yorkshire,[1644]with socketed celts, gouges, hammer, &c.

In the Heathery Burn Cave, Durham, and in the Guilsfield find, there was in each case at least one lump of metal.

Besides the cakes of copper, bars of that metal appear to have been hammered into an oblong form, and then cut into lengths of from 4 to 5 inches, weighing each about ¼ lb., and in that state to have served as the raw material for the bronze-founders. Thirteen of these short bars were found at Therfield, near Royston, Herts,[1645]and Dr. Percy found on analysis that they contained about 98½ per cent. of copper with a small alloy of tin or antimony, probably the latter. Some fifteen or sixteen “pieces of long triangular brass” are described as having been found with about the same number of celts at Hinton, near Christchurch, Hants.[1646]These bars “seemed to be pieces of the metal out of which the celts were cast.”

In Scotland some “lumps of brass” were found with the swords, spears, &c., in Duddingston Loch.[1647]Probably other lumps of metal have been found in that country, but they seem to be scarcer in Scotland and Ireland than in England.

Although, as already observed, Spain may have been the principal Western source of tin in early times, and possibly Malacca[1648]in the East, the trade with Britain for that metal musthave commenced at a very remote epoch. We might expect, therefore, that fragments of tin would be frequently found in the old bronze-founders’ hoards. But though lumps of copper have so often been discovered in them, tin is at present conspicuous by its absence. The only instance to which I am able to refer is the discovery at Achtertyre,[1649]Morayshire, of four “broken bits of tin,” in company with socketed celts, spear-heads, and bracelets. These pieces seem to be fragments of a single bar which was about 6 inches in length, of oval section, and somewhat curved, and in weight about 3 ounces. Though spoken of as tin, the metal is in fact a soft solder composed, according to Dr. Stevenson Macadam, of—

This, he points out, is a more fusible alloy than the ordinary plumbers’ solder, which consists of 1 of tin to 2 of lead, and fuses at 441 degrees Fahr., as it contains nearly 4 of tin to 1 of lead, and would fuse at 365 degrees. Whether this bar was intended for use as solder, or represents a base tin exported to Scotland from the tin-producing districts, is an interesting question. Professor Daniel Wilson[1650]has called attention to the fact that in all the bronze instruments found in Scotland which have been submitted to analysis lead is uniformly present, though in varying proportions. Soldering[1651]is considered to have been entirely unknown in the Bronze Age, and even during the earlier times of the Iron Age; but the art of burning bronze on to bronze was certainly known, and instances of its having been practised are given in preceding pages.

Some fragments of pure metallic tin have from time to time been found on the Continent. A small hammered bar found at the Lake-dwelling of Estavayer,[1652]and analyzed by M. de Fellenberg, was free from lead, zinc, iron, and copper.

Besides being found in Cornwall, tin occurs in France,[1653]Saxony, Silesia, Bohemia, Sweden, Spain, and Portugal. It also occurs in Etruria,[1654]and is said to be found in Chorassan.[1655]

This metal is said by Dionysius[1656]to have been struck into coins at Syracuse, but none such are at present known. Among the Ancient Britons,[1657]however, tin coins cast for the most part in wooden moulds were in circulation, not in the tin-producing districts, but in Kent and the neighbouring parts of England. Their date is probably within a century of our era, either before or after Christ.

Fig. 514.—Falmouth.1/12

A large ingot of tin, in shape like the letterH, was dredged up in Falmouth harbour.[1658]It is 2 feet 11 inches long and about 11 inches wide, and 3 inches thick, and, though a small piece has been cut off at one end, it still weighs 158 lbs. It is shown in Fig. 514. The late Sir Henry James, F.R.S.,[1659]has pointed out that the form in which the ingot is cast adapts it for being laid in the keel of a boat, and for being slung on a horse’s side, two of them thus forming a proper load for a pack-horse. He has also suggested that this was the form of ingot in which the tin produced in Cornwall was transported to Gaul, and thence carried overland, as described by Diodorus Siculus, to the mouths of the Rhone. Curiously enough this author speaks of the blocks being in the form of astragali, with which this ingot fairly coincides. Other ingots[1660]of tin of different form have also been found in Cornwall, but there appears to me hardly sufficient evidence to determine their approximate date, and I therefore content myself with mentioning them. A lump cast in a basin-shaped mould, with two holes in the flat face converging so as to form a V-shaped receptacle for a cord, is in the Blackmore Museum at Salisbury.

What appear to be ingots of copper rather than votive or mortuary tablets have been found in Sardinia,[1661]and in their form present a close analogy with this ingot of tin, though they are of muchsmaller dimensions. Both the sides and ends curve inwards, the notch at the ends of some being semicircular. They are counter-marked with a kind of doubleT.

As to the method of melting the metal but little is known. It seems probable, however, that the crucibles employed must have been vessels of burnt clay provided with handles for moving them; while for pouring out the metal small ladles of earthenware may have been used. At Robenhausen,[1662]on Lake Pfäffikon, Switzerland, small crucibles of a ladle-like form have been found, in some cases with lumps of bronze still in them. Crucibles without handles have been discovered at Unter-Uhldingen,[1663]in the Ueberlinger See.

The methods of casting were various. Objects were cast—

1. In a single mould formed of loam, sand, stone, or metal, the upper surface of the casting exhibiting the flat surface of the molten metal, which was left open to the air. In the case of loam or sand castings a pattern or model would be used, which might be an object already in use, or made of the desired form in wood or other soft substance.

2. In double moulds of similar materials. The castings produced in this manner when in unfinished condition show the joints of the moulds. When sand was employed a frame or flask of some kind must have been used to retain the material in place when the upper half of the mould was lifted off the pattern. The loam moulds were probably burnt hard before being used. In many cases cores for producing hollows in the casting were employed in conjunction with these moulds.

3. In what may be termed solid moulds. For this process the model was made of wax, wood, or some combustible material which was encased in a mass of loam, possibly mixed with cow-dung or vegetable matter, which on exposure to heat left the loam or clay in a porous condition. This exposure to fire also burnt out the wax or wood model and left a cavity for the reception of the metal, which was probably poured in while the mould was still hot.

Sir John Lubbock[1664]regards this as the commonest mode of casting during the Bronze Age, but so far as this country is concernedit appears to me to have been very seldom, if ever, in use. Except in highly complicated castings, such as ring within ring, no advantage would be gained by adopting the process, as the same result could usually be obtained by the use of a mould in two halves, while the pattern would then be preserved. In comparing a number of objects together, though, like the six hundred and eighty-eight specimens of celts in the Dublin Museum, no two may appear to have been cast in the same mould, it does not follow that this was actually the case, for allowance must be made for hammering, polishing, and ornamenting, which were subsequent processes, and also for wear at the edge. Even in castings from the same metal mould there will be considerable variations, from differences in the amount of coating used to prevent the hot metal from adhering to mould, and the length stopped off by the core. But of this I shall shortly speak.

The moulds formed of burnt clay have but rarely lasted to our times, though some have been found on the continent of Europe.

One for a perforated axe found among the remains of Lake-dwellings near Laibach, in Carniola, is in the museum of that town. Others will subsequently be mentioned.

The single moulds found within the United Kingdom are all of stone, and are adapted for the production of flat celts, rings, knives, and small chisels. In some cases it is hard to say whether a mould was intended to be used alone or in conjunction with another of the same kind, so as in fact to be only the half of a mould.

The single mould, which I have engraved as Fig. 515, was found near Ballymena, Co. Antrim, and, as will be seen, is for a flat celt of the ordinary form. The material is a micaceous sandstone, which a recent possessor of the mould has thought so well adapted for use as a whetstone, that the mould is in places scored with the marks where apparently a cobbler’s awl has been sharpened. A celt cast in such a mould would be flatter on one face than the other, and be blunt at the ends, though much thinner there than in the middle. Before being used it would be submitted to a hammering process, which would render the two faces nearly symmetrical, and at the same time condense the metal and render it harder and fitter for cutting purposes, especially at the edge which was drawn out. In an Irish specimen in my collection there is in one face a deep conical depression, apparently caused by the contraction of the metal in cooling. It was probably necessary to add a little molten metal to the casting while coolingin order to avoid such defects. The sides as well as the faces of these plain celts have usually been wrought with the hammer, and it seems probable that some even of the flanged celts were originally plain castings in an open mould.

Fig. 515.—Ballymena. ½

Moulds of the same kind have been found, though rarely, in England. In a field near Cambo,[1665]near Wallington, Northumberland,was found a block of sandstone, having on one face two moulds for flat celts of different sizes, and on the other face another such mould, and also one for a flat ring. It is now in the British Museum.

Stone blocks with moulds cut in them have been found in Scotland.

One with a mould for a large celt in the centre, and near it in one corner of the slab a mould for a very small celt, was found in a cairn near Kintore, Aberdeenshire.[1666]Another large block, forming the end of a cist, near Kilmartin, Argyleshire,[1667]has nine depressions in it in the form of flat celts, which may have been used as moulds. They are barely an eighth of an inch in depth, and on this account have been thought to be pictorial representations rather than moulds. With a metal so imperfectly fluid as melted bronze, castings could be made thicker than the depth of the moulds, and it is by no means impossible that this stone and another forming part of the same cist may have been intended for the production of castings. The second slab of stone may have served for casting pins.The stone moulds from Trochrig, near Girvan, Ayrshire,[1668]and Alford, Aberdeenshire,[1669]with depressions of various forms upon them, not improbably belong to a later period than that of which I am treating.A mould for casting rings, 2½ inches in diameter, found at Kilmailie, Invernessshire, is in the Museum at Edinburgh.One for two flat celts on the one face, and for a larger celt and perhaps a knife on the other, is in the Antiquarian Museum at Edinburgh.[1670]

One with a mould for a large celt in the centre, and near it in one corner of the slab a mould for a very small celt, was found in a cairn near Kintore, Aberdeenshire.[1666]

Another large block, forming the end of a cist, near Kilmartin, Argyleshire,[1667]has nine depressions in it in the form of flat celts, which may have been used as moulds. They are barely an eighth of an inch in depth, and on this account have been thought to be pictorial representations rather than moulds. With a metal so imperfectly fluid as melted bronze, castings could be made thicker than the depth of the moulds, and it is by no means impossible that this stone and another forming part of the same cist may have been intended for the production of castings. The second slab of stone may have served for casting pins.

The stone moulds from Trochrig, near Girvan, Ayrshire,[1668]and Alford, Aberdeenshire,[1669]with depressions of various forms upon them, not improbably belong to a later period than that of which I am treating.

A mould for casting rings, 2½ inches in diameter, found at Kilmailie, Invernessshire, is in the Museum at Edinburgh.

One for two flat celts on the one face, and for a larger celt and perhaps a knife on the other, is in the Antiquarian Museum at Edinburgh.[1670]

These moulds are more abundant in Ireland.

One in the Belfast Museum,[1671]polyhedral in shape, has moulds upon four of its faces for flat celts of different sizes. In the Bateman Collection is a slab of schistose stone (7 inches by 6 inches) with three such moulds upon it. It was found near Carrickfergus, Co. Antrim.[1672]On a slab in the Museum of the Royal Irish Academy[1673]there are moulds for two flat celts, and also for one with a stop-ridge and a loop. It would appear as if the founder must have possessed a second half of this latter mould.Two moulds formed of stone, and apparently intended for flat or slightly flanged celts, have been found at Bodio in the Lago di Varese.[1674]

On a slab in the Museum of the Royal Irish Academy[1673]there are moulds for two flat celts, and also for one with a stop-ridge and a loop. It would appear as if the founder must have possessed a second half of this latter mould.

Two moulds formed of stone, and apparently intended for flat or slightly flanged celts, have been found at Bodio in the Lago di Varese.[1674]

Moulds for palstaves and socketed celts have been found both of stone and of bronze, but it will be well to reserve the latter until all the forms of moulds made of stone have been considered. Such celt moulds have always been made in halves.

Back to Index Next