Chapter VIII: France (2)

About a mile from this example, Mr. Lukis mentions a still larger one. It measures 81 feet in length by 6 feet in width, is divided into two compartments like the one just described, and has also a holed entrance. He also measured two in Finistère, one 76 feet, the other 66 feet, in length, and both 6 feet wide. Both, however, had been rifled long ago, and are now mere ruins. More, no doubt, would be found if looked for. Indeed, these straight-lined "allées couvertes," or "Grottes des fées," without cells, as the French call them, as before mentioned, are the most characteristic, if not the most common, form of French rude-stone monuments. The only other place where they are equally common is Drenthe, and it may be that this side hole at Kerlescant is an approach to the side entrance so usual in that province.

At Plouharnel, about a mile and a half westward from Mont St.-Michel, a double dolmen was opened a good many years ago. In it were found some beautiful gold ornaments, others in bronze, and some celts or stone axes in jade[420]--all these, like those of Mont St.-Michel, belonging evidently to what antiquaries call the latest period of the Polished Stone age; but until it is determined what that age is, it does not help us much to a date.

To the north of Kerlescant, at about the distance of half a mile, is another long barrow, called Moustoir or Moustoir-Carnac, which was opened in 1865, also by M. René Galles. It was found to contain four separate interments, dispersed along its length, which exceeds 280 feet, the height varying from 15 to 20 feet. The western chamber is a regular dolmen, of the class called "Grottes des fées," and is apparently the oldest of the group. The centre one (_b_) is a very irregular chamber, the plan of which it is difficult to make out; the third (_c_) is a dolmen, irregular in plan, but roofed with three large stones; but the fourth (_d_) is a circular chamber, the walls of which are formed of tolerably large stones, the roof being built up into the form of a horizontal dome (woodcut No. 144), by stones projecting and overlapping, instead of the simpler ceiling of single blocks as on all the earlier monuments. This, as well as the walls, being built with small stones, I take to be a certain indication of a more modern age. A considerable number of flint implements were found in the western chamber, with some beads and a partially pierced cylinder in serpentine, but no coins, nor any object of an age which can be positively dated. Here, however, these troublesome Roman tiles make their appearance as at Crubelz. "Ici, comme à Mané er H'roëk, nous trouvons les traces caractéristiques du conquerant (les Romains): des tuiles à rebord ont croulé, au pied de notre butte funéraire, et plusieurs même se sont glissées à travers les couches supérieures des pierres, qui forment une partie de la masse."[421]

If these monuments are really prehistoric, it is to me incomprehensible that these traces of the Romans should be so generally prevalent in their structure. If it is objected that these are not found in the chambers of the tombs themselves, the answer seems only too evident that hardly one of them is virgin: all, or nearly all, have been entered before the time of recent explorers, and all their more valuable contents removed. Celts and beads and stone implements were not likely to attract the attention of early pilferers, and these they left; but except in the instance of the sepulchre at Plouharnel, metal is very rarely found in any. But the presence of Roman pottery, or other evidence of that people, in the long barrows in Gloucestershire, at Kennet, and at Carnac, are too frequent to be accidental. In so far as proving that the monument is not prehistoric, the presence of a single fragment of Roman pottery is as conclusive as a hoard of coins would be, provided it is found so placed that it could not have been inserted there after the mound was complete; and this I fancy is the case in all the instances mentioned above.

LOCMARIAKER.

It is rather to be regretted that no good survey exists of this cemetery. Not that much depends on the juxtaposition of the monuments, but that, as the French are continually changing their names, and most of them have two, it is not always easy to feel sure which monument is being spoken of at any particular time. Those on the mainland are situated in a zone about a mile in length, running north and south, between Mané Lud, the most northern, and Mané er H'roëk, the most southern. The first-named is a long barrow, 260 feet by about 165, but not, as in England, of one age or containing only one, but, like Moustoir-Carnac, several sepulchres, which may either be of the same age or erected at different though hardly distant periods, and joined together by being buried under one great mound. Of the three which Mané Lud contains, the most interesting is the partially covered dolmen at the west end. It consists of a chamber of somewhat irregular form, but measuring 12 feet by 10 feet, and covered by one enormous block of stone, measuring 29 feet by 15 feet, and with a passage leading to it, making the whole length from the entrance to the central block of the chamber 20 feet. According to Mr. Ferguson,[422] five of the blocks of this dolmen are sculptured; according to M. René Galles,[423] nine are so ornamented. The stone, however, is so rough and the place so dark that it is difficult at times to distinguish them and always so to draw them. The principal objects represented seem to be intended for boats and hatchets, but there are other figures which cannot be so classed, and, though it may be rash to call them writing, they may mean numbers or cyphers of some sort. Their great interest is, however, their similarity to the engravings on Irish monuments. If any one will, for instance, compare this woodcut (No. 145) and woodcut No. 68 from New Grange, he can hardly fail to see a likeness which cannot well be accidental; and in like manner the curvilinear forms of woodcut No. 146, in a manner hardly to be mistaken, resemble those from Clover Hill (woodcut No. 77).

]

Close by Mané Lud, but a little nearer to Locmariaker, stands what may be considered as the most interesting, if not the finest, free-standing dolmen in France. Its roof consists of two stones: one of these measures 18 feet by 9 feet,[425] and more than 3 feet in thickness. The second stone is very much smaller, and seems to form a sort of porch to it. The great stone rests, like that of most free-standing dolmens, on three points, their architects having early learned how difficult it was to make sure of their resting on more; so that unless they wanted a wall to keep out the stuff out of which the tumulus was to be composed, they generally poised them on three points like that at Castle Wellan (woodcut No. 7).

The great interest in this dolmen, however, lies in its sculptures. The stone which closes the east end is shaped into the form of two sides of an equilateral spherical triangle and covered with sculptures, which this time are neither characters nor representations of living things, but purely decorative. At one time I thought the form of a cross could be traced on the stone. The central stem and the upper arm are shown clearly enough in the drawing by Mr. Ferguson; but all the drawings show a lower cross-arm--though I confess I did not see it--which quite destroys this idea. On the roof a well-sculptured plumed[426] hatchet can be traced very distinctly, as shown in the woodcut copied from Mr. Ferguson. He fancies he can also trace the form of a plough in the sculptures of the roof, but this seems doubtful.

It is to this dolmen that the great fallen obelisk belongs. If it was one stone, it measured 64 feet in length and 13 feet across its greatest diameter; but I confess I cannot, from the mode in which it has fallen, rid myself of the idea that it was in reality two obelisks, and not one. Whether this was the case or not, it is a remarkable work of art for a rude people, for it certainly has been shaped with care, and with the same amount of labour might have been made square or round or any other shape that might have been desired. This, however, is one of the peculiarities of the style. No one will dispute that this obelisk and the stones of the Dol ar Marchant are hewn; but instead of adopting the geometrical forms, of which we are so fond, they preferred those that reminded them of their old rude monuments, and which to their eyes were more beautiful than the straight lines of the Romans. I do not feel quite sure that artistically they were not right.

If we compare this dolmen with that at Krukenho (woodcut No. 126), the difference between them appears very striking. The Del ar Marchant is a regular tripod dolmen, carefully built of shaped stones and engraved. The other is a magnificent cist, walled with rude stones, and such as would form a chamber in a tumulus if buried in one, though whether this particular example was ever intended to be so treated or not is by no means clear. Be this as it may, there are two modes of accounting for the difference between two monuments so nearly alike in dimensions and situated so near to one another. The first would be to assume that the Krukenho example is the oldest, it being the rudest and approaching more nearly to the primitive form of the monuments: the second would be to assume that the one was the memorial of some warrior, erected in haste on the battle-field where he fell, by his companions in arms; and that the other was a royal sepulchre, prepared at leisure either by the king himself or by those who succeeded him in times of peace, and consequently who had leisure for such works. We must know more of these monuments before a satisfactory choice can be made between these two hypotheses. At present I rather incline to the belief that the circumstances under which they were erected may have more to do with their differences than their relative ages.

To return to Locmariaker. Close to the town there is, or was, a long allée couverte.[427] It is 70 feet long, and divided towards its inner end into a square chamber, to which a long-slightly curved gallery led, composed of fourteen stones on each side. Five of these are covered with ornaments, and characters engraved on them. One might be considered as representing the leaf of a fern, or possibly a palm; the rest are ovals, circles, and similar ornaments, which may or may not have more meaning than those at New Grange or other monuments in the locality.

On the other side of the village is the tumulus already mentioned as Mané er H'roëk, where the twelve Roman coins were found, and inside it an immense collection of polished celts, but all broken, and one slab, which apparently originally closed the door, and is covered with sculptured hatchets, similar in character to that on the roof of the Dol ar Marchant, but not so carefully drawn nor so well engraved.

Besides these there are several--probably as many as a dozen--monuments of the same class, within what may fairly be considered the limits of this cemetery; but of these the most interesting, as well as the most perfect, is that on the island of Gavr Innis, about 2 miles eastward from Locmariaker.

]

The plan of the chamber of this monument will be understood from the annexed plan.[429] The gallery of entrance measures 44 feet from where the lining stones begin to the chamber, which is quadrangular in form, and measures 9 feet by 8 feet. All the six stones forming the three sides of the chamber, and most of those which line the entrance on either hand, are most elaborately sculptured with patterns, the character of which will be understood from the annexed woodcuts. The pattern, it will be observed, is not so flowing or graceful as those found at New Grange or Dowth, and nowhere, I believe, can it be said to imitate vegetable forms; and in the woodcut on the left-hand stone are some seventeen or eighteen figures, which are generally supposed to represent celts, and probably do so; but if they do, from their position they must mean something more, either numbers or names, but, whatever it may be, its meaning has not yet been guessed. On other stones there are waving lines, which are very generally assumed to represent serpents, and, I believe, correctly so; but as that is somewhat doubtful, it is as well to refrain from citing them. Besides these, the general pattern is circles within circles, and flowing lines nearly equidistant, but, except on one stone, never of spirals, and then less graceful than the Irish. The sculpture, however, on some of the stones at Lough Crew, and that in the centre especially of woodcut No. 75, is absolutely identical with the patterns found here; and altogether there is more similarity between these sculptures and those at Lough Crew than between almost any other monuments of the class that I know of.

In the chamber on the left-hand side is a stone (woodcut No. 153), with three holes in it, which have given rise to an unlimited amount of speculation. Generally it is assumed that it was here that the Druids tied up the human victims whom they were about to sacrifice. But, without going back to the question as to whether there ever were any Druids in the Morbihan, would any priest choose a small dungeon 8 feet square and absolutely dark for the performance of one of their greatest and most solemn rites? So far as we know anything of human sacrifices, they were always performed in the open day and in the presence of multitudes. Assuming for the moment, however, that these holes were intended for some such purpose, two would have sufficed, and these of a form much simpler and more easily cut. As will be seen from the woodcut, not only are the three holes joined, but a ledge or trough is sunk below them which might hold oil or holy water, and must, it appears to me, have been intended for some such purpose.

The existence of these holes seems to set at rest another question of some interest. Generally it has been assumed that the tattooing on the stones of the chambers, &c., may have been done with stone implements. This cannot be denied, though it seems improbable; but the undercutting of the passages between these holes and the formation of the trough could only be effected by a tool which would bear a blow on its head, and a heavy one too, or, in other words, by some well-tempered metal tool.

At Tumiac, opposite Gavr Innis, existed a very large tumulus, which was opened in 1853 by Messrs. Fouquet and L. Galles. It was found to contain a small chamber, partly formed of large slabs, partly of small stones. Some of the former had rude carvings upon them, but without any meaning that can now be made out.

The whole has the appearance of being considerably more modern than Gavr Innis.

* * * * *

Besides these, in the neighbourhood of Carnac and Locmariaker, there are at least three other groups of stones in France which deserve much more attention than has hitherto been bestowed upon them. The first is in the peninsula of Crozon, forming the southern side of the roadstead of Brest. It consists, among others, of three alignments of stones. The principal one is at a place called Kerdouadec, and consists of a single line of stones 1600 feet in length, arranged on a slightly curved plan, and terminating in a curious "Swastica"-like cross. The second, at Carmaret, is a single line, 900 feet long, and with two branches at right angles to it, near its centre. The third, at Leuré, is likewise a single line with a slight elbow in the centre, from which starts a short branch at right angles.[430]

I am not able to offer a conjecture what these alignments represent, nor why or when they were placed here. Whether an inspection on the spot might suggest some clue is not clear, but they are so unlike anything found anywhere else, either in France or any other country, that they must for the present, I fear, remain a mystery.

The second group, known as the Gré de Cojou, is situated about halfway between Rennes and Redon. The remains here consist of a short double alignment some 500 feet long, several tumuli--one at least surmounted by a circle of stones--several stone enclosures, and frequent dolmens. They have been imperfectly described by M. Ramé,[431] and planned, but not published, by Sir Henry Dryden. Until these are given to the world more in detail than has hitherto been done, it is impossible to say whether they represent a battle-field or a cemetery. From their position--a bleak, barren heath, far from any centre of population--I would guess the former; but I have not visited the place myself, and the information at my command is too meagre to enable me to speak with any confidence regarding them.

The third group is in the department of the Lot, near Preissac, in the parish of Junies, and extends over half a mile (800 metres) in length. Unfortunately we have nothing but verbal descriptions of it, and from these it is impossible to realise its form, or predicate its destination.[432] We are, indeed, in a state of great ignorance with regard to all these megalithic remains in the south of France, but as they seem as important and as numerous as those in the north, it is to be hoped some one will devote an autumn to their illustration. There are probably several other groups as important as those at Junies, but they are quite unknown to us at present. These groups must therefore be put aside for the present, and any argument regarding age or use of this class of monuments must be based wholly on what we know of those of the Morbihan.

* * * * *

So far as I know, no reasonable tradition attaches to any of the monuments in the Locmariaker cemetery which would enable us to fix their dates with anything like certainty, nor are there any local circumstances, except the Roman coins and tiles above alluded to, which aid us in our researches. We are thus left to such general inferences as the case admits of, and to a comparison with other similar monuments whose dates are nearer and better ascertained. No one, however, who is familiar with the two great cemeteries of Meath will probably hesitate in admitting that the two groups cannot be far separated in date. Of course, it is impossible in a general work like the present to put the evidence forward in anything like a complete state. In order to do this in a satisfactory manner would require a large volume to itself, and the illustrations both of the French and Irish examples should be drawn by the same person. Even the few illustrations that have been given are probably sufficient to show a similarity so great that it can hardly be accidental, and I may be allowed to add, from personal familiarity with both groups of monuments, that it seems impossible to escape the conviction that they are monuments of the same class, probably of the same or a closely allied race, and of about the same age. This last must always be the most uncertain premiss of the three, as we can scarcely hope ever to know the relative state of civilization of the two countries at a given time; and consequently, even if we could prove that two ornaments in the two countries were identical in form, this would not prove that there might not be a difference of fifty or a hundred years between them. Even at a later age, in the thirteenth century, for instance, the same form and the same style in France and England did not prevent a difference of fifty years existing between any two examples. In the fourteenth the two were abreast, and in the fifteenth century they again diverged, so that, although the architecture of both was still Gothic, a comparison of style for this purpose became almost impossible.

In like manner, though the central ornament in the middle stone at Lough Crew (woodcut No. 75) is almost identical with some of the ornaments at Gavr Innis (woodcut No. 152), it by no means necessarily follows that the two are exactly of the same age. So, too, the foliage at New Grange (woodcut No. 67) and that in the allée--now, I fear, destroyed--at Locmariaker are evidently of one style, but still admit of a certain latitude of date. On the whole, judging from style alone, I should feel inclined to range Gavr Innis rather with the cemetery at Lough Crew than with that on the Boyne; as well from its ornaments as because I fancy that those monuments which are roofed with flat stones only are earlier than those which make some attempt at construction. But, on the other hand, I believe that Mané er H'roëk and Mané Lud may more probably range with New Grange and Dowth; and as I look upon it as quite certain that the monuments on the Boyne were all erected in the first four centuries after the birth of Christ, it seems impossible that the age of those at Locmariaker can be very distant from that date.

To many it will no doubt seem improbable that these monuments should have been erected during the occupation of the country by the Romans. If, however, they would take the trouble of studying what is now going on in India, their incredulity would, I fancy, soon disappear. The natives there at the present day are in many parts of the country building temples which it requires a practised eye to distinguish from those erected before any European settled in the land; and they follow their own customs, and worship their own gods, utterly irrespective of, and uninfluenced by, the strangers who have held the chief sway in the country for more than a hundred years. It must also be borne in mind that the Romans never really settled in Brittany. The country was poor then as now, and it led to nowhere. So long as the Bretons remained quiet, the Romans seem to have left them to themselves, and certainly have left no traces of any establishment of importance in their country--nothing that would lead us to suspect such intimate relations with the natives as would induce them to change their faith or fashions and copy the institutions of the foreigners.

On the other hand, it seems not only possible, but probable, that intercourse with the Romans may first have inspired the inhabitants of Brittany with a desire to attain greater durability and more magnificence, by the employment of stone, instead of earth or wood, for their monuments. This they might do, without its creating in their minds the smallest desire to copy either Roman forms or Roman institutions. On the contrary, we have every reason to believe that in these remote districts the Romans would be hated as conquerors, and that their religion and their customs would be held in abhorrence as strange and unsuited to the land they occupied.

Be this as it may, a comparison with the Irish examples reduces the questions at issue with regard to dates within very narrow limits. Either these monuments were erected immediately before or during the time of the Roman occupation or immediately after their departure, but prior to the conversion of the natives to Christianity. We are not yet in a position to decide positively between these two hypotheses, but the presence of Roman coins and Roman tiles in some of the mounds and the whole aspect of the argument seem to me to incline the balance in favour of their belonging to Roman times. Some may be anterior to the Christian era, but I am very much mistaken if it be not eventually admitted that the greater number of them are subsequent to that epoch.

Even, however, if the age of the monuments of the cemetery of Locmariaker could be ascertained, it would by no means necessarily carry with it that of the stone rows at Carnac. They belong to a different category altogether, and may be of a different age.

No one now, I presume, after what has been said above, especially with regard to the Scandinavian examples, will think it necessary that I should go over the ground to prove that they are not temples. Every argument that could be adduced to prove that Avebury or Stonehenge are not temples tells with tenfold force here. A temple extending over six or seven miles of country is more improbable than one covering only 28 acres. This one, too, is open everywhere, and has no enclosure or "temenos" of any sort, and there being an uneven number of equally spaced rows of stones in the principal monument is sufficient to show it was not intended and could not be used for processions. In fact I hardly know of any proposition that appears to me so manifestly absurd as that these stone rows were temples, and I feel sure that no one who thinks twice of the matter will venture again to affirm it.

It seems equally clear that they were not erected for any civic or civil purpose. No meetings could be held, and no administrative functions could be carried on in or around them. Nor are they sepulchral in any ordinary sense of the term. In the first place because, though men were buried in tumuli or under dolmens, or had single head-stones, nowhere were men buried in rows like this, extending over miles of heath and barren country. But the great fact is that the French savants have dug repeatedly about these stones and found no trace of burials. The most conclusive experiment of the sort was made by a road surveyor some six or seven years ago. Wishing to raise the road from Auray to Carnac, he dug out the sand and gravel on the east side of the road, over a considerable area, to a depth of from three to four feet; but being of a conservative turn of mind, he left the eleven rows of stones each standing on a little pillar of sand. It was then easy to trace the undisturbed strata of differently coloured earth round and almost under the stones, and to feel perfectly certain that it had never been disturbed by any inhumation. It, no doubt, is true that the long barrow at Kerlescant, the dolmen at Kermario, and the enclosure at Maenec, may have been, indeed most probably were, all of them, burying-places, but they can no more be considered the monument than the drums and fifes can be considered the regiment. They are only the adjuncts; the great rows must be considered as essentially the monuments.

If, therefore, they are neither temples, nor town-halls, nor even sepulchres, we are driven back on the only remaining group of motives which, so far as I know, ever induced mankind to expend time and labour on the erection of perfectly unutilitarian erections. They must be trophies--the memorials of some great battle or battles that at some time or other were fought out on this plain. The fact of the head of each division being a tomb is in favour of this hypothesis; but if it is considered as the principal part, it is like drawing a jackdaw with a peacock's tail--an absurdity into which these men of the olden time would hardly fall.

It is more difficult to answer the questions, Are Carnac and Erdeven parts of one great design, or two separate monuments? Is Carnac the march, St.-Barbe the position before the battle, Erdeven the scene of the final struggle for the heights that gave the victory, and the tombs scattered over the plain between these alignments the graves of those who fell in that fight? Such appears to me the only feasible explanation of what we here find; but the great question still remains, What fight?

There is, probably, no single instance in which the negative argument derived from the silence of the classical authors applies with such force as to this. If these stones existed when Cæsar waged war against the Veneti in this quarter, he must have seen them, and as it may be presumed that the monument was then more complete than it is now, he could hardly have failed to be struck with it, and, if so, to have mentioned it in his 'Commentaries.' Even, however, if he neglected them, the officers of his army must have seen these stones. They must have been talked about in Rome, and some gossip like Pliny, when writing about stones, must have heard of this wonderful group, and have alluded to it in some way. The silence, however, is absolute. No mediæval rhapsodist even attempts to give them a pre-Roman origin. Such traditions as that of St. Cornely, or Cornelius the Centurion, though absurd enough, point, as such traditions generally do, to the transition time between paganism and Christianity, when, apparently, all mediæval chroniclers seem to have believed that all these rude-stone monuments were erected. Till, therefore, some stronger argument than has yet been adduced, or some new analogy be suggested, the pre-Roman theory must be set aside; and if this is so, we are tolerably safe in assuming that no battle of sufficient importance was fought which these stones could be erected to commemorate during the time when the Romans held supreme sway in the country.

If this is so, our choice of an event to be represented by these great stone rows is limited to the period which elapsed between the overthrow of the Roman power by Maximus, A.D. 383, and the time when the people of the country were completely converted to Christianity--which happened in the early part of the sixth century.[433] But if the history of England is confused and uncertain during that century and a half, that of Brittany is even more so, and has not yet been elucidated by the French authorities to the same extent as ours has been.

No one, I believe, doubts that Maximus, coming with an army from Britain, landed somewhere in Brittany, where he fought a great battle with the forces of Gratian, whom he defeated, and that afterwards, in a second battle near Lyons, he expelled the legitimate government of the Romans from Gaul.[434] I also see no reason for doubting that he was accompanied by a British prince Conan Meriadec, who afterwards settled in the country with thousands of his emigrant countrymen, over whom he was enabled to establish his chieftainship on the ruins of the Roman power.

If this is so, the battle which destroyed the Roman power, and gave rise to the native dynasty, would be worthy of such a monument as that at Carnac; but so far as local traditions go, the place where Maximus and his British allies landed was near St. Malo, and the battle was fought at a place called Alleth, near St. Servan.[435] If this is so, it was too far off to have any connection with the Carnac stones. Two other wars seem to have been carried on by Conan, one in 410 against a people who are merely called barbarians,[436] a second against the Romans under Exuperantius in 416;[437] but we have no local particulars which would enable us to connect these wars with our stones. A war of liberation against Rome would be worthy of a national monument, and it may be that this is such a one, but I know of nothing to connect the two together, though local enquiries on the spot might remove this difficulty.

On the whole, however, I am more inclined to look among the events of the next reign for a key to the riddle. Grallon was engaged in two wars at least: one against the Roman consul Liberius in 439,[438] in which he succeeded in frustrating the attempts of that people to recover their lost power; the other against the "Norman pirates;"[439] and it is to this, as connecting the stone monuments with a Northern people, that I should be inclined to ascribe the erection of the Carnac alignments. From Grallon being the reputed founder of Landevenec, it might seem more probable that the alignments at Crozon marked the position of this battle, and I am not prepared to dispute that it may be so. The question is not of importance; if either group marked a battle-field of this period, the other certainly did so also, and I would prefer to refrain from offering any opinion as to what particular battle these stones commemorate. That must be determined by some local antiquary with much more intimate knowledge of the history and traditions of the province than I possess. All I wish to show here is that there was a period of a century and a half between the departure of the Romans and the time when the Bretons were so completely converted to Christianity as to abandon their old habits and customs, and that during that period there were wars with the Romans and the Northern barbarians of sufficient importance to justify the erection of any monuments within the competence of the people. If this is so, and we are limited to this period, enough is established in so far as the argument of this work is concerned, and the rest may fairly be left to be discussed and determined by the local antiquaries. All that it is necessary to contend for here is, that the alignments at Carnac are neither temples, nor tombs, nor town-halls, and that they were not erected before the time of the Romans. If these negative propositions are answered, there will not, probably, be much difficulty in admitting that they must be trophies, and that the battle or campaign which they commemorate was fought between the years 380 and 550 A.D.--in fact in the Arthurian age, to which we have ascribed most of those in this country.

The monuments in the cemetery at Locmariaker are probably older, but some of them extend down to the time when Carnac "closed the line in glory."

NUMBER OF DOLMENS IN THIRTY-ONE DEPARTMENTS OF FRANCE ACCORDING TO M. BERTRAND, 1864.[440]

Dolmens. Terminations in _ac_.
Lot 500 71
Finistère 500 3
Morbihan 250 26
Ardèche 155 16
Aveyron 125 35
Dordogne 100 75
Vienne (Haute et Basse) 82 41
Côtes du Nord 56 8
Maine-et-Loire 53 --
Eure-et-Loir 40 --
Gard 32 16
Aube 28 1
Indre-et-Loire 28 --
Charente 26 50
Creuse 26 6
Charente-Inférieure 24 21
Lozère 19 16
Corrèze 17 42
Vendée 17 --
Loire-Inférieure 16 11
Sarthe 15 --
Ille-et-Vilaine 15 18
Deux-Sèvres 15 --
Orne 14 --
Indre 13 3
Manche 13 --
Pyrénées-Orientales 12 2
Puy-de-Dôme 10 3
Oise 9 --
Cantal 8 37
Tarn-et-Garonne 7 16

[Footnote 379: 'Revue archéologique,' August, 1864, 148 _et seq._]

[Footnote 380: Livy, v. chap. 34.]

[Footnote 381: Walcknaer, 'Géographie des Gaules.' The earlier
chapters and Map V.]

[Footnote 382: 'Revue archéologique,' new series, vii. 228.]

[Footnote 383: _Ibid._]

[Footnote 384: 'De Bello Gall.' i. 1.]

[Footnote 385: Strabo, vi. 176, 189.]

[Footnote 386: 'Archæological Journal,' 1870, cviii. p. 225 _et
seq._]

[Footnote 387: Lartet, Christy, and 'Reliquiæ Aquitanicæ.' London,
1865 _et seq._]

[Footnote 388: 'Monuments mégalithiques du Vivarais,' p. Oll. de
Marchand; Montpellier, 1870.]

[Footnote 389: 'Époques antéhistoriques du Poitou,' P. A.
Brouillet; Poitiers, 1865.]

[Footnote 390: This list must be taken as only tentative. All I
have done was to take the Atlas Joanne, and count the number of
names as well as I could. I feel far from confident that I have
counted all; and, besides, the scale of the maps is too small to
feel sure that all, or nearly all, are there. It is, however,
sufficient for present purposes of comparison. If it is thought
worth while to pursue the investigation farther, it must be done
on the 80,000 scale map of France, which would be work of great
labour.]

[Footnote 391: Delpon, 'Statistique du Département du Lot,' i. p.
383.]

[Footnote 392: In the Ordnance Maps, 1-inch scale, the termination
_ac_ occurs at least 38 times in this corner, though in these maps
always spelt with an additional _k_, as Botallac_k_, Carnidjac_k_;
although this is by no means the usual or ancient spelling of the
district.]

[Footnote 393: The whole of these churches are described in more or
less detail by Félix de Verneilh in his 'Architecture byzantine en
France,' 4to. Paris, 1851. Several of them are also illustrated in
my 'History of Architecture,' i. 418-441.]

[Footnote 394: The argument, which it is not necessary to enter
on here, has been well summed up by Dr. Schmitz, in Smith's
'Dictionary of Greek and Roman Geography,' _sub voce_ Cimbri.]

[Footnote 395: The existence of this line of dolmens and of a
separate people, all the way from Brittany to Narbonne, may serve,
perhaps, to explain the mode in which the tin of Britain found
its way across France to the Mediterranean Sea. That the Veneti
traded from the Côtes-du-Nord and the Morbihan to Cornwall and
the Cassiterides, no one, probably, will dispute. Their vessels,
according to Cæsar's account, were fully equal to carrying to
France all the metal this country could produce. The road by which
it reached Marseilles across France was always the difficulty. In
later times, the Celtic trade-route across France was apparently
up the Rhone, but on its left bank, and down the Seine, or on its
right bank; passing then through Celtica, but round the Aquitania
of Augustus, and reaching Britain through the country of the
Morini, which was the route Cæsar followed. This does not, however,
appear to have been the line which was taken by the trade in tin.
It followed, so far as we know, the central line of the dolmen
country; and the fact of one people and one language prevailing
throughout the whole of that region takes away any improbability,
and removes all the difficulties that have hitherto impeded the
adoption of that hypothesis.]

[Footnote 396: My intention was to have spent last autumn in
travelling through the southern departments of France with this
intent; but the war rendered the position of an exploring and
sketching foreigner so undesirable that I was forced to desist. Had
this book been a "statisque" of the subject, as it was originally
intended, I should have been obliged to defer its publication till
I had accomplished this journey, or till the monuments had been
illustrated. As, however, it has now assumed more the form of an
"argument," this is of comparatively little consequence.]

[Footnote 397: In a paper on the 'Monuments mégalithiques de
l'Auvergne,' by M. Cartheilhac, in the Norwich volume of the
Prehistoric Congress, he gives drawings of ten as types. Five of
these, or one-half, are dolmens on tumuli, which is, however,
probably more than a fair proportion. One has already been given,
woodcut No. 8.]

[Footnote 398: 'Statistique monumentale de la Charente,' 141.
Richard, 'France monumentale,' p. 677. 'Mém. de la Société royale
des Antiquaires de France,' vii. 26.]

[Footnote 399: The woodcuts are copied from Michon, 'Statistique
de la Charente.' In describing it, he quotes the Edict of the
Council at Nantes with regard to the destruction of these
"venerated stones." He (p. 141) gives the date of this council as
A.D. 1262, which would almost make it appear that this was
one of the stones against which the decree was fulminated. This
date, however, appears to be a mistake. The true date I believe to
be 658, as given above, p. 24.]

[Footnote 400: 'Rev. archéologique,' ix. 400.]

[Footnote 401: 'Essai sur les Dolmens,' p. 38.]

[Footnote 402: Paper read by S. Ferguson, Q.C., before the R. I. A.
14th Dec. 1863. See also pamphlet by René Galles (Vannes, 1863),
describing the exploration.]

[Footnote 403: 'Congrès préhistorique,' vol. de Paris, 1868, 42.]

[Footnote 404: All these are represented in Gailhabaud's
'Architecture ancienne et moderne,' ii. plates 7 and 8.]

[Footnote 405: The woodcut is from a publication privately printed
by Dr. Blair and Mr. Ronalds.]

[Footnote 406: Gailhabaud, 'Arch. anc. et mod.' i.]

[Footnote 407: Renouvier, 'Monuments de Bas-Languedoc.' No numbers
to plates.]

[Footnote 408: See one published by Sir R. Colt Hoare, 'Modern
Wiltshire,' iv. p. 57.]

[Footnote 409: 'Kilkenny Journal,' third series, vol. i. p. 40 _et
seq._]

[Footnote 410: I have not seen the monument myself, nor do I know
any one who has, but I cannot believe it to be a pure invention.
Too much stress must not, however, be laid upon it.]

[Footnote 411: There is a woodcut in Bonstetten's work (p. 25)
which, being taken endways, explains more clearly how, the
cap-stone resting on two points only, it can be understood to
oscillate. It is, however, much less correct as a representation.
]

[Footnote 412: Delpon, 'Statistique du Dép. du Lot,' i. p. 388.]

[Footnote 413: 'Ptolemæi Geo.' Amstel. 1605, p. 47.]

[Footnote 414: The only survey of this monument which has
been published, and can be depended upon, is that made by Mr.
Vicars, a surveyor of Exeter, for the Rev. Dr. Bathurst Deane.
It was published by him on a reduced scale in vol. xxv. of the
'Archæologia,' and re-engraved, with the principal parts on the
original scale, by Dr. Blair and Mr. Ronalds, in the work before
alluded to, but unfortunately never published. The original map, on
a scale of 440 feet to 1 inch, is still in Dr. Deane's possession,
at Bath, and is so valuable a record of what the monument was
thirty-two years ago that it is hoped it may be preserved by some
public body. Sir Henry Dryden and the Rev. Mr. Lukis have been
employed for some years past exploring and surveying in that
neighbourhood, and have brought back perfect plans, on a large
scale, of all the principal monuments; and if these were published,
they would leave little to be desired in that respect. Meanwhile
nothing can exceed Sir Henry's kindness and liberality in allowing
access to his treasures, and the use of them by any one who desires
it; and I am indebted to him for a great deal of the information
in this chapter. The general plans here published are from Messrs.
Blair and Ronalds' work, which is quite sufficiently correct for my
scale or my present purpose.]

[Footnote 415: The form of this enclosure, as will be seen from
the plan, is not an exact square, and some of the angle-stones
being removed, it is difficult now to ascertain its exact form.
Sir Henry Dryden makes it curvilinear. Messrs. Blair and Ronalds
make the east side quite straight; the south and west were slightly
curvilinear, but the whole figure is quadrangular; which is my own
impression of its form.]

[Footnote 416: Sir Henry Dryden counts ten rows. Mr. Vicars'
survey, from which the woodcut is copied, makes only eight. Their
irregularity makes it difficult to feel certain on such a point.]

[Footnote 417: 'Journal of Archæological Association,' vol. xxiv.
pp. 40 _et seq._]

[Footnote 418: _Ante_, pp. 163 _et seq._]

[Footnote 419: It is so difficult to realise these similarities,
except by representation, that I give here a woodcut of that at
Rodmarton. Allowing for the difference of drawing and engraving,
the openings are identical, and it is so peculiar in form that the
likeness cannot be accidental. If it does not occur anywhere else,
or at any other time, it proves, as far as anything can prove, that
the French and English long barrows were erected under the same
inspiration. If one is post-Roman, so, certainly, is the other; or
if one can be proved to be prehistoric, the other must follow.

]

[Footnote 420: These were exhibited in the inn in the village when
I was there. Where they are now, I do not know.]

[Footnote 421: 'Revue archéologique,' xii. p. 17.]

[Footnote 422: 'Proceedings of Royal Irish Academy,' vol. viii.
1864, p. 298 _et seq._]

[Footnote 423: 'Revue archéologique,' vol. x. 1864, pl. iv.]

[Footnote 424: Woodcuts No. 145 and 146 are copied from Mr.
Ferguson's paper in the 'Proceedings of Royal Irish Academy,' viii.
398 _et seq._]

[Footnote 425: These dimensions are from Richard; other authorities
make it 18 feet by 12 feet.]

[Footnote 426: The existence of the plume is doubted by Sir Henry
Dryden, and he is so accurate that he probably is right; but as
others say they have seen it, and nothing depends upon it, I have
allowed it to remain.]

[Footnote 427: It was in a very ruinous state when I saw it five
years ago; and there is an ominous silence regarding it among
subsequent tourists. The measurements here quoted are from Richard,
'France monumentale.']

[Footnote 428: The plan here given is reduced from one by Sir Henry
Dryden, and may be perfectly depended upon as far as the smallness
of the scale will allow.]

[Footnote 429: Sir Henry drew all these sculptures first on
the spot, and afterwards corrected his drawings from the casts
at St.-Germain. They are the only drawings existing which can
thoroughly be depended upon.]

[Footnote 430: A plan of the first-named alignment was published
by Freminville, 'Finistère,' part ii. pl. i., but the above
particulars and the woodcut are taken from a diagram by Sir Henry
Dryden in the last number of the 'Journal of the Anthrop. Inst.' He
has perfect plans of the whole.]

[Footnote 431: 'Revue archéologique,' new series, ix. pp.
81 _et seq._ I may mention that almost every other name in
their neighbourhood ends in _ac_. See 'Joanne Atlas,' dép.
Ille-et-Vilaine.]

[Footnote 432: Delpon, 'Statistique du Dép. du Lot,' i. 384.]

[Footnote 433: "C'est en 465 que Vannes reçut pour premier évêque
l'Armoricain St. Patern, qui mourut peu d'années après chez les
Francs, où les Goths l'avoient forcé de se réfugier. Modestus en
511 mit tout en œuvre pour repandre le Christianisme parmi
les Pagani de son diocèse, mais son zèle ne fut pas recompensé,
car plus de trente ans après la mort de Patern les habitans de
la Vénétie étoient encore presque tous païens. 'Erant enim tunc
temporis Venetenses pene omnes Gentiles.'--_Ap. Boll._ 'Vita St.
Melan.' vi. Jan. p. 311."--_Courzon_, 'Chartulaire de l'Abbaye de
Redon,' cxliii.]

[Footnote 434: The authority for these events will be found at
length in Gibbon, chap. xviii., and are too familiar to need
quoting here.]

[Footnote 435: Daru's 'Histoire de la Bretagne,' vol. i. p. 58.]

[Footnote 436: _Ibid._ p. 112.]

[Footnote 437: Dom. Bouquet, 'Recueil des Hist. des Gaules,' i.
p. 629. "Exuperantius anno circa 416 Armoricos qui a Romanis
defecerunt ad officium reducere tentavit."]

[Footnote 438: Daru, i. p. 112.]

[Footnote 439: "Gradlonus gratia dei rex Britonum necnon ex
parte Francorum."--_Chartulaire de Landevenec_; quoted by P.
Lobineau, ii. 17. And further: "'Pervenit Sancti (Wingaboei)
fama ad Grallonum regem Occiduorum Cornubiensium, gloriosum
ultorem Normannorum qui post devictas gentes inimicas sibi
duces subduxerat.'--_Gurdestan, Moine de Landevenec_, 'Vie de
St.-Wingabois.'"--_Daru_, i. p. 69.]

[Footnote 440: The information in this table must be received with
great limitation. In the first place, What is a dolmen? Do the
alignments at Carnac count as two, as seven, or as 700? Many also
are mere estimates of local antiquaries. It is, for instance, very
doubtful if Finistère contains more monuments than the Morbihan;
and subsequent information may introduce great modifications into
many of the numbers.

The value of the _ac_ distinction does not come out clearly: first,
because of the imperfect mode in which it has been obtained, but
more because it does not make it clear that there are in France
twenty-nine departments in which there are no dolmens, and no
_ac_-terminations; in fact, the negative evidence which does not
appear here is stronger than the positive.]