"But their safety," observes Mr. Wylie,* "was probably owing to their living in the middle of the lake, (Greek) en mese te limne, whereas the ancient Swiss settlers were compelled by the rapidly increasing depth of the water near the margins of their lakes to construct their habitations at a short distance from the sh.o.r.e, within easy bowshot of the land, and therefore not out of reach of fiery projectiles, against which thatched roofs and wooden walls could present but a poor defence."

(* W.M. Wylie "Archaeologia" volume 38 1859, a valuable paper on the Swiss and Irish lake-habitations.)

To these circ.u.mstances and to accidental fires we are probably indebted for the frequent preservation, in the mud around the site of the old settlements, of the most precious tools and works of art, such as would never have been thrown into the Danish "kitchen-middens," which have been aptly compared to a modern dusthole.

Dr. Ferdinand Keller of Zurich has drawn up a series of most instructive memoirs, ill.u.s.trated with well-executed plates, of the treasures in stone, bronze, and bone brought to light in these subaqueous repositories, and has given an ideal restoration of part of one of the old villages (see Plate 1 above),* such as he conceives may have existed on the lakes of Zurich and Bienne.

(*Keller "Pfahlbauten, Antiquarische Gesellschaft in Zurich"

Bd. 12 and 13 1858-1861. In the fifth number of the "Natural History Review" January 9, 1862, Mr. Lubbock has published an excellent account of the works of the Swiss writers on their lake-habitations.)

In this view, however, he has not simply trusted to his imagination, but has availed himself of a sketch published by M. Dumont d'Urville, of similar habitations of the Papuans in New Guinea in the Bay of Dorei. It is also stated by Dr. Keller, that on the River Limmat, near Zurich, so late as the last century, there were several fishing-huts constructed on this same plan.*

(* Keller "Pfahlbauten, Antiquarische Gesellschaft in Zurich" Bd. 9 page 81 note.)

It will be remarked that one of the cabins is represented as circular.

That such was the form of many in Switzerland is inferred from the shape of pieces of clay which lined the interior, and which owe their preservation apparently to their having been hardened by fire when the village was burnt. In the sketch (Plate 1), some fishing-nets are seen spread out to dry on the wooden platform. The Swiss archaeologist has found abundant evidence of fishing-gear, consisting of pieces of cord, hooks, and stones used as weights. A canoe also is introduced, such as are occasionally met with. One of these, made of the trunk of a single tree, fifty feet long and three and a half feet wide, was found capsized at the bottom of the Lake of Bienne. It appears to have been laden with stones, such as were used to raise the foundation of some of the artificial islands.

It is believed that as many as 300 wooden huts were sometimes comprised in one settlement, and that they may have contained about 1000 inhabitants. At w.a.n.gen, M. Lohle has calculated that 40,000 piles were used, probably not all planted at one time nor by one generation. Among the works of great merit devoted specially to a description of the Swiss lake-habitations is that of M. Troyon, published in 1860.*

(* "Sur les Habitations lacustres.")

The number of sites which he and other authors have already enumerated in Switzerland is truly wonderful. They occur on the large lakes of Constance, Zurich, Geneva, and Neufchatel, and on most of the smaller ones. Some are exclusively of the stone age, others of the bronze period. Of these last more than twenty are spoken of on the Lake of Geneva alone, more than forty on that of Neufchatel, and twenty on the small Lake of Bienne.

One of the sites first studied by the Swiss antiquaries was the small lake of Moosseedorf, near Berne, where implements of stone, horn, and bone, but none of metal, were obtained. Although the flint here employed must have come from a distance (probably from the south of France), the chippings of the material are in such profusion as to imply that there was a manufactory of implements on the spot. Here also, as in several other settlements, hatchets and wedges of jade have been observed of a kind said not to occur in Switzerland or the adjoining parts of Europe, and which some mineralogists would fain derive from the East; amber also, which, it is supposed, was imported from the sh.o.r.es of the Baltic.

At w.a.n.gen near Stein, on the Lake of Constance, another of the most ancient of the lake-dwellings, hatchets of serpentine and greenstone, and arrow-heads of quartz have been met with. Here also remains of a kind of cloth, supposed to be of flax, not woven but plaited, have been detected. Professor Heer has recognised lumps of carbonised wheat, Tritic.u.m vulgare, and grains of another kind, T. dicocc.u.m, and barley, Hordeum distichum, and flat round cakes of bread; and at Robbenhausen and elsewhere Hordeum hexastichum in fine ears, the same kind of barley which is found a.s.sociated with Egyptian mummies, showing clearly that in the stone period the lake-dwellers cultivated all these cereals, besides having domesticated the dog, the ox, the sheep, and the goat.

Carbonised apples and pears of small size, such as still grow in the Swiss forests, stones of the wild plum, seeds of the raspberry and blackberry, and beech-nuts, also occur in the mud, and hazel-nuts in great plenty.

Near Morges, on the Lake of Geneva, a settlement of the bronze period, no less than forty hatchets of that metal have been dredged up, and in many other localities the number and variety of weapons and utensils discovered, in a fine state of preservation, is truly astonishing.

It is remarkable that as yet all the settlements of the bronze period are confined to Western and Central Switzerland. In the more eastern lakes those of the stone period alone have as yet been discovered.

The tools, ornaments, and pottery of the bronze period in Switzerland bear a close resemblance to those of corresponding age in Denmark, attesting the wide spread of a uniform civilisation over Central Europe at that era. In some few of the Swiss aquatic stations a mixture of bronze and iron implements has been observed, but no coins. At Tiefenau, near Berne, in ground supposed to have been a battle-field, coins and medals of bronze and silver, struck at Ma.r.s.eilles, and of Greek manufacture, and iron swords, have been found, all belonging to the first and pre-Roman division of the age of iron.

In the settlements of the bronze era the wooden piles are not so much decayed as those of the stone period; the latter having wasted down quite to the level of the mud, whereas the piles of the bronze age (as in the Lake of Bienne, for example) still project above it.

Professor Rutimeyer of Basle, well-known to palaeontologists as the author of several important memoirs on fossil vertebrata, has recently published a scientific description of great interest of the animal remains dredged up at various stations where they had been embedded for ages in the mud into which the piles were driven.*

(* "Die Fauna der Pfahlbauten in der Schweiz" Basel 1861.)

These bones bear the same relation to the primitive inhabitants of Switzerland and some of their immediate successors as do the contents of the Danish "kitchen-middens" to the ancient fishing and hunting tribes who lived on the sh.o.r.es of the Baltic.

The list of wild mammalia enumerated in this excellent treatise contains no less than twenty-four species, exclusive of several domesticated ones: besides which there are eighteen species of birds, the wild swan, goose, and two species of ducks being among them; also three reptiles, including the eatable frog and freshwater tortoise; and lastly, nine species of freshwater fish. All these (amounting to fifty-four species) are with one exception still living in Europe. The exception is the wild bull (Bos primigenius), which, as before stated, survived in historical times. The following are the mammalia alluded to:--The bear (Ursus arctos), the badger, the common marten, the polecat, the ermine, the weasel, the otter, wolf, fox, wild cat, hedgehog, squirrel, field-mouse (Mus sylvaticus), hare, beaver, hog (comprising two races, namely, the wild boar and swamp-hog), the stag (Cervus elaphus), the roe-deer, the fallow-deer, the elk, the steinbock (Capra ibex), the chamois, the Lithuanian bison, and the wild bull. The domesticated species comprise the dog, horse, a.s.s, pig, goat, sheep, and several bovine races.

The greater number, if not all, of these animals served for food, and all the bones which contained marrow have been split open in the same way as the corresponding ones found in the sh.e.l.l-mounds of Denmark before mentioned. The bones both of the wild bull and the bison are invariably split in this manner. As a rule, the lower jaws with teeth occur in greater abundance than any other parts of the skeleton--a circ.u.mstance which, geologists know, holds good in regard to fossil mammalia of all periods. As yet the reindeer is missing in the Swiss lake-settlements as in the Danish "kitchen-middens," although this animal in more ancient times ranged over France, together with the mammoth, as far south as the Pyrenees.

A careful comparison of the bones from different sites has shown that in settlements such as w.a.n.gen and Moosseedorf, belonging to the earliest age of stone, when the habits of the hunter state predominated over those of the pastoral, venison, or the flesh of the stag and roe, was more eaten than the flesh of the domestic cattle and sheep. This was afterwards reversed in the later stone period and in the age of bronze.

At that later period also the tame pig, which is wanting in some of the oldest stations, had replaced the wild boar as a common article of food. In the beginning of the age of stone, in Switzerland, the goats outnumbered the sheep, but towards the close of the same period the sheep were more abundant than the goats.

The fox in the first era was very common, but it nearly disappears in the bronze age, during which period a large hunting-dog, supposed to have been imported into Switzerland from some foreign country, becomes the chief representative of the canine genus.

A single fragment of the bone of a hare (Lepus timidus) has been found at Moosseedorf. The almost universal absence of this quadruped is supposed to imply that the Swiss lake-dwellers were prevented from eating that animal by the same superst.i.tion which now prevails among the Laplanders, and which Julius Caesar found in full force amongst the ancient Britons.*

(* "Commentaries" lib 5 chapter 12.)

That the lake-dwellers should have fed so largely on the fox, while they abstained from touching the hare, establishes, says Rutimeyer, a singular contrast between their tastes and ours.

Even in the earliest settlements, as already hinted, several domesticated animals occur, namely, the ox, sheep, goat, and dog. Of the three last, each was represented by one race only; but there were two races of cattle, the most common being of small size, and called by Rutimeyer Bos brachyceros (Bos longifrons, Owen), or the marsh cow, the other derived from the wild bull; though, as no skull has yet been discovered, this identification is not so certain as could be wished.

It is, however, beyond question that at a later era, namely, towards the close of the stone and beginning of the bronze period, the lake-dwellers had succeeded in taming that formidable brute the Bos primigenius, the Urus of Caesar, which he described as very fierce, swift, and strong, and scarcely inferior to the elephant in size. In a tame state its bones were somewhat less ma.s.sive and heavy, and its horns were somewhat smaller than in wild individuals. Still in its domesticated form, it rivalled in dimensions the largest living cattle, those of Friesland, in North Holland, for example. When most abundant, as at Concise on the Lake of Neufchatel, it had nearly superseded the smaller race, Bos brachyceros, and was accompanied there for a short time by a third bovine variety, called Bos trochoceros, an Italian race, supposed to have been imported from the southern side of the Alps. (Caesar "Commentaries" lib 5 chapter 12.) This last-mentioned race, however, seems only to have lasted for a short time in Switzerland.

The wild bull (Bos primigenius) is supposed to have flourished for a while in a wild and tame state, just as now in Europe the domestic pig co-exists with the wild boar; and Rutimeyer agrees with Cuvier and Bell,* in considering our larger domestic cattle of northern Europe as the descendants of this wild bull, an opinion which Owen disputes.**

(* "British Quadrupeds" page 415.)

(** "British Fossil Mammal." page 500.)

In the later division of the stone period, there were two tame races of the pig, according to Rutimeyer; one large, and derived from the wild boar, the other smaller, called the "marsh-hog," or Sus scrofa pal.u.s.tris. It may be asked how the osteologist can distinguish the tame from the wild races of the same species by their skeletons alone.

Among other characters, the diminished thickness of the bones and the comparative smallness of the ridges, which afford attachment to the muscles, are relied on; also the smaller dimensions of the tusks in the boar, and of the whole jaw and skull; and, in like manner, the diminished size of the horns of the bull and other modifications, which are the effects of a regular supply of food, and the absence of all necessity of exerting their activity and strength to obtain subsistence and defend themselves against their enemies.

A middle-sized race of dogs continued unaltered throughout the whole of the stone period; but the people of the bronze age possessed a larger hunting-dog, and with it a small horse, of which genus very few traces have been detected in the earlier settlements--a single tooth, for example, at w.a.n.gen, and only one or two bones at two or three other places.

In pa.s.sing from the oldest to the most modern sites, the extirpation of the elk and beaver, and the gradual reduction in numbers of the bear, stag, roe, and freshwater tortoise are distinctly perceptible. The aurochs, or Lithuanian bison, appears to have died out in Switzerland about the time when weapons of bronze came into use. It is only in a few of the most modern lake-dwellings, such as Noville and Chavannes in the Canton de Vaud (which the antiquaries refer to the sixth century), that some traces are observable of the domestic cat, as well as of a sheep with crooked horns and with them bones of the domestic fowl.

After the sixth century, no extinction of any wild quadruped nor introduction of any tame one appears to have taken place, but the fauna was still modified by the wild species continuing to diminish in number and the tame ones to become more diversified by breeding and crossing, especially in the case of the dog, horse, and sheep. On the whole, however, the divergence of the domestic races from their aboriginal wild types, as exemplified at w.a.n.gen and Moosseedorf, is confined, according to Professor Rutimeyer, within narrow limits. As to the goat, it has remained nearly constant and true to its pristine form, and the small race of goat-horned sheep still lingers in some alpine valleys in the Upper Rhine; and in the same region a race of pigs, corresponding to the domesticated variety of Sus scrofa pal.u.s.tris, may still be seen.

Amidst all this profusion of animal remains extremely few bones of Man have been discovered; and only one skull, dredged up from Meilen, on the Lake of Zurich, of the early stone period, seems as yet to have been carefully examined. Respecting this specimen, Professor His observes that it exhibits, instead of the small and rounded form proper to the Danish peat-mosses, a type much more like that now prevailing in Switzerland, which is intermediate between the long-headed and short-headed form. (Rutimeyer "Die Fauna der Pfahlbauten in der Schweiz"

page 181.)

So far, therefore, as we can draw safe conclusions from a single specimen, there has been no marked change of race in the human population of Switzerland during the periods above considered.

It is still a question whether any of these subaqueous repositories of ancient relics in Switzerland go back so far in time as the kitchen-middens of Denmark, for in these last there are no domesticated animals except the dog, and no signs of the cultivation of wheat or barley; whereas we have seen that, in one of the oldest of the Swiss settlements, at w.a.n.gen, no less than three cereals make their appearance, with four kinds of domestic animals. Yet there is no small risk of error in speculating on the relative claims to antiquity of such ancient tribes, for some of them may have remained isolated for ages and stationary in their habits, while others advanced and improved.

We know that nations, both before and after the introduction of metals, may continue in very different stages of civilisation, even after commercial intercourse has been established between them, and where they are separated by a less distance than that which divides the Alps from the Baltic.

The attempts of the Swiss geologists and archaeologists to estimate definitely in years the antiquity of the bronze and stone periods, although as yet confessedly imperfect, deserve notice, and appear to me to be full of promise. The most elaborate calculation is that made by M.

Morlot, respecting the delta of the Tiniere, a torrent which flows into the Lake of Geneva near Villeneuve. This small delta, to which the stream is annually making additions, is composed of gravel and sand.

Its shape is that of a flattened cone, and its internal structure has of late been laid open to view in a railway cutting 1000 feet long and 32 feet deep. The regularity of its structure throughout implies that it has been formed very gradually, and by the uniform action of the same causes. Three layers of vegetable soil, each of which must at one time have formed the surface of the cone, have been cut through at different depths. The first of these was traced over a surface of 15,000 square feet, having an average thickness of 5 inches, and being about 4 feet below the present surface of the cone. This upper layer belonged to the Roman period, and contained Roman tiles and a coin. The second layer, followed over a surface of 25,000 square feet, was 6 inches thick, and lay at a depth of 10 feet. In it were found fragments of unvarnished pottery and a pair of tweezers in bronze, indicating the bronze epoch.

The third layer, followed for 35,000 square feet, was 6 or 7 inches thick and 19 feet deep. In it were fragments of rude pottery, pieces of charcoal, broken bones, and a human skeleton having a small, round and very thick skull. M. Morlot, a.s.suming the Roman period to represent an antiquity of from sixteen to eighteen centuries, a.s.signs to the bronze age a date of between 3000 and 4000 years, and to the oldest layer, that of the stone period, an age of from 5000 to 7000 years.

Another calculation has been made by M. Troyon to obtain the approximate date of the remains of an ancient settlement built on piles and preserved in a peat-bog at Chamblon, near Yverdun, on the Lake of Neufchatel. The site of the ancient Roman town of Eburodunum (Yverdun), once on the borders of the lake, and between which and the sh.o.r.e there now intervenes a zone of newly-gained dry land, 2500 feet in breadth, shows the rate at which the bed of the lake has been filled up with river sediment in fifteen centuries. a.s.suming the lake to have retreated at the same rate before the Roman period, the pile-works of Chamblon, which are of the bronze period, must be at the least 3300 years old.

For the third calculation, communicated to me by M. Morlot, we are indebted to M. Victor Gillieron, of Neuveville, on the Lake of Bienne.

It relates to the age of a pile-dwelling, the mammalian bones of which are considered by M. Rutimeyer to indicate the earliest portion of the stone period of Switzerland, and to correspond in age with the settlement of Moosseedorf.

The piles in question occur at the Pont de Thiele, between the lakes of Bienne and Neufchatel. The old convent of St. Jean, founded 750 years ago, and built originally on the margin of the Lake of Bienne, is now at a considerable distance from the sh.o.r.e, and affords a measure of the rate of the gain of land in seven centuries and a half. a.s.suming that a similar rate of the conversion of water into marshy land prevailed antecedently, we should require an addition of sixty centuries for the growth of the mora.s.s intervening between the convent and the aquatic dwelling of Pont de Thiele, in all 6750 years. M. Morlot, after examining the ground, thinks it highly probable that the shape of the bottom on which the mora.s.s rests is uniform; but this important point has not yet been tested by boring. The result, if confirmed, would agree exceedingly well with the chronological computation before mentioned of the age of the stone period of Tiniere. As I have not myself visited Switzerland since these chronological speculations were first hazarded, I am unable to enter critically into a discussion of the objections which have been raised to the two first of them, or to decide on the merits of the explanations offered in reply.

IRISH LAKE-DWELLINGS OR CRANNOGES.