Chapter XIII: Part 13

The locality was almost surrounded by the adjacent sluggish waters, and so wet that two pumps had to be kept going before the men could carry out the necessary excavations. From the sedimentary character of the deposit, as well as the abundance of fresh-water shells, there could be no doubt that formerly the place had been occupied by a lake. There was, first of all, a bed of peat about five feet thick, and under this came a deposit of mud and sand. On the south side of the space being excavated there was observed at a depth of nearly 12 feet a layer of greenish stuff, mixed with bones, impregnated with vivianite, and through this layer the piles were found to have penetrated to the sand underneath. It was in the muddy deposit immediately beneath the peat that the tops of the piles appeared, and they were arranged sometimes in parallel rows, and sometimes without any apparent regularity. Some were of oak, and others of soft wood. There was also much timber lying transversely, and many of the beams showed signs of charring.

The relics were collected between and around these piles, and uniformly all over the area. They consisted of a remarkable series of bronze implements and weapons, together with a few of stone and horn. There were also found the bones of tame and wild animals, a human brachycephalic skull and some other human bones, a portion of a canoe, and a very small quantity of pottery of an indeterminate character. The bones were very much broken, but, notwithstanding, they were identified as belonging to the following animals, viz.:--stag, roe, hare, bear, ox, horse, pig, and dog. It is noteworthy that the reindeer and elk were both unrepresented.

_Relics._--It is, however, the relics that distinguish this lacustrine find from others in North Germany, and these I shall now describe shortly:--three swords with handles (=Fig. 97=, Nos. 8, 9, and 10), one sword-blade attached by rivets (No. 11); an ornamented _commandostab_ (No. 18), and a small button-like object, ornamented with a running scroll of double spiral; three daggers have rivet-marks and one has a tang (Nos. 5, 6, 7, and 12); one dagger, still in its bronze handle (No. 13), has its butt end ornamented with concentric circles and spirals characteristic of the Scandinavian archæological area; two lance-heads with sockets (Nos. 4 and 17), one of which is ornamented with lines and crossbars (No. 17); one socketed celt (No. 3); five paalstabs (Nos. 1 and 2); and a piece of bronze wire.

Of stone objects there were two round grindstones or polishers (No. 15) and some sharpening stones, a polished perforated stone (No. 16), a portion of a hammer of greenstone, and a round stone ball 4 inches in diameter, like a cannon-ball.

Five staghorn axe-heads, a disc of horn perforated, a portion of a horn spear, five bits of rough unornamented pottery, and a large perforated ball of clay. Fragments of a canoe showing a length of 10 feet. Report goes that an iron implement, and a portion of a dish like earthenware of the twelfth century, were found; but probably they had no connection with the bronze objects above described. It is noteworthy that all the relics are of a military character, there being among them no spindle-whorls, combs, hair-pins, fibulæ, bracelets, or any other objects that can be said to belong to domestic life. For this reason this lacustrine abode is generally supposed to have been a military fort like La Tène. (B. 384 and 396.)

(C) POSEN AND POLAND.

OBJEZIERZE.--In the province of Posen there are several localities to be recorded which have yielded unequivocal indications of lake-dwellings, two of which, viz. Objezierze and Czeszewo, are supposed by local archæologists to date as far back as the Stone Age. The former existed in a swamp now filled up with peat, which has thus preserved and concealed piles and other remains recently brought to light by peat-cutters. The relics collected from this place are now deposited in the Posen Archæological Museum, and among them I have noted the following:--A few remarkably fine knife-flakes of flint, one of which is 7½ inches long, a perforated bead, four large clay rings (=Fig. 98=, No. 8), and three flint celts of the Scandinavian type (No. 7). In the same turf moor and in the vicinity of the lake-dwelling was found a large bronze torque ornamented as shown in No. 9.[56]

CZESZEWO (KR. WAGROWICE).--Although for many years the existence of piles in a particular spot in this lake was known to fishermen, it was not till 1871 that their true nature was recognised by Professor Lepkowsky of Krakow. The lake was surrounded by marshy borders and peat bogs, and at various times its level was lowered, which thus considerably reduced its area. Firewood being scarce in the district, the fishermen were in the habit of pulling out the piles, and in this way the structures were greatly damaged before a competent archæologist saw them. The site of the Pfahlbau was near a large tumulus constructed close to the original lake margin. The area occupied by the piles was in the form of a segment of a circle, the base of which was 250 paces long and its greatest breadth 75. Transverse beams from 10 to 12 feet long were found interspersed among the uprights, which were supposed to have bound the latter together, though neither wedges nails, nor mortises were detected. It was observed that the uprights had their tops charred and that the portions remaining were longer the farther they were placed from the shore, and hence it was supposed that the dwelling had been destroyed by a conflagration.

The remains of human industry collected from this station are now deposited partly in the Jagellon Museum at Krakow, and partly in the Archæological Museum at Posen. Among them are fragments of pottery, one being part of a dish perforated with small holes (=Fig. 98=, No. 5), perforated stone and horn hammers and axes (Nos. 1 and 2), one or two plain celts (No. 6), some fragments of clay rings (No. 3), two fragments of human skulls, and a large quantity of the osseous remains of different animals. One object of bronze is said to have been found on this station, and one of the stone implements is only partially perforated, the operation having been unfinished. The stone objects are made of dark granite.

Messrs. Kohn and Mehlis have published a small photographic view of the objects in the Krakow Museum, but the more interesting ones, though fewer in number, are at Posen, from which the illustrations here shown are taken. Notices of this lake-dwelling have been published by Count Przezdiecki (B. 156 and 195), by Kohn and Mehlis (B. 338), and by Ossowski of Krakow (B. 361)--the last being in Polish and French.

GROSSESEE.--At Alt-Gortzig, in the Grossesee, there was a small island which became visible on the lowering of the lake some 10 feet, around which were found piles and the usual _débris_ of a lacustrine dwelling, among which were pottery with parallel and wavy lines, charcoal, and an iron axe-head, together with numerous osseous remains. (B. 228 and 352.)

PAWLOWICE.--Mr. Schwartz, of Posen, describes what he considers to have been a lake-dwelling near Pawlowice. Here, in a turf-moor which had formerly been a lake, he found, at a depth of five feet, bits of clay plaster, hearth-stones, fragments of cooking vessels, etc.[57] Also at Komorowo, in the Bythinersee, indications of lake-dwellings have been found.[58]

LAGIEWNICKI.--Another interesting locality, discovered a few years ago, is at Lagiewnicki (Posen). Here the tops of oak piles were found at a depth of five feet in the peat, and associated with them were fragments of two kinds of pottery--one rough, like that used in the manufacture of urns, and the other of the Burgwälle type. Among the relics were a wooden mallet, a perforated bone implement, some flint flakes, the pin of a bronze fibula of La Tène type (=Fig. 98=, No. 11), and a silver necklace (No. 10), terminating at one end in a raised button which clasped with an eye at the other when fastened. (B. 430.)

KWACZALA.--At the request of the Academy of Sciences of Krakow, Mr. Adam Kirkor, curator of the Archæological Museum at Wilna, investigated, in the summer of 1873, a peat-moor near the village of Kwaczala, said to contain pile-dwellings. Mr. Kirkor found beams and piles in several spots pointed out by the proprietor, where the peat-cutters were said to have formerly encountered woodwork. Water came upon his trenches at a depth of three feet. Both upright and transverse beams of oak were found, some being over a yard in circumference and 8½ yards long. The area containing structural remains of woodwork was 70 yards long by 40 broad. The foundation was of horizontal beams, spread out in all directions, which he concluded to have been arranged after some kind of architectural principle. There was a large quantity of rude pottery, some showing linear or punctured ornamentation. Two perforated axe-hammer heads of stone, and about 300 bits of worked flint were collected among the _débris_, as well as some bones of the horse. Altogether, this primitive habitation appears to have been of a peculiar kind. (B. 338.)

BIALKA (LUBLINER KR.).--In the moor of Bialka, formerly covered with water, there is a small island about 100 paces in diameter, on which tradition says there was once an enchanted castle. Professor Joseph Przyborowski, of Warsaw, made some excavations on the island, and found on the surface some tiles and modern implements, which so far confirmed the tradition of the ancient castle; but upon digging he came upon wooden piles at a depth of four feet. His excavations extended some twenty feet long and nine feet wide, and in the whole of this area he found numbers of piles, as well as cross-beams. Associated with these wooden structures there was also a relic-bed, entirely distinct from the superficial layer, on which he found two well-formed flint arrow-heads, a portion of a perforated axe of serpentine, some flint implements, and broken bones of edible animals. If this site were properly investigated the author prognosticated results of considerable scientific value. (B. 338.)

Professor Ossowski, in his "Carte Archéologique" (B. 361), gives the following sites of lake-dwellings, none of which, however, have been carefully investigated:--(1) _Warlubie_ (Kr. Swieć). This is a vast peat deposit from which neolithic implements and staghorn hammers have been extracted from time to time. It was visited by Ossowski in the year 1878, who found some fragments of pottery and charcoal. But these merely strengthened the suspicion that the antiquities were due to Pfahlbauten. (2) Similar indications were found at _Kowalewo_, in the district of Tornú. (3) At _Wabrzeźno_, in the district of Chelmno, there is a small lake, in which were found a primitive vase, a bronze fish-hook, a stone hammer, and an implement of staghorn. (4) _Lankorsz_, district of Lubawa.

(D) EAST PRUSSIA AND LIVLAND.

In the eastern districts of Prussia lake-dwelling remains have been discovered in the following places, which have been more or less investigated and described in various archæological publications, especially in the _Altpreussiche Monatsschrift_:--

ARYSSEE (KR. LÖTZEN).--_A. M._, vol. iv. p. 667; xii. p. 89;
xiv. p. 181. _Zeit. für Ethn._, vol. xix.,
_Verhand._, p. 491.
CZARNISEE (KR. LÖTZEN).--_A. M._, vol. xiv. p. 181; vol. xv. p. 481.
KOCKSEE (KR. RÖSSEL).--_A. M._, vol. xxii. p. 169;
_Zeit. für Ethn._, vol. xvi.,
_Verhand._, p. 560.
PROBCHENSEE (KR. RÖSSEL).--_A. M._, vol. xxii. p. 169
QUERTZ (KR. HEILSBERG).--_A. M._, vol. xxii. p. 169.
BONSLACK (KR. WEHLAU).--_A. M._, vol. xxii. p. 485.
TULEWOSEE (KR. LYCK).--_A. M._, vol. v. p. 750.
SZONTAGSEE (KR. LYCK).--_A. M._, vol. xxiv. p. 488.
KOWNATKENSEE (KR. NIEDENBURG).--_A. M._, vol. xxiv. pp. 168 and 496.
LONKORRECKERSEE (CULMERLANDE).--_A. M._, vol. x. p. 579.
GESERICHSEE--_Phy. Ok. Gesel._, 1874, _Verhand._, p. 14.

ARYSSEE.--The existence of the _débris_ of a remarkable lake-dwelling in the Aryssee became known in 1863, in consequence of the discontinuance of a mill which had its motive power supplied by the surplus water from this lake, and the subsequent deepening of its outlet, which had the effect of lowering its level about seven feet. Its remains have been investigated and described by various persons, notably Professor Heydeck, of Königsberg, who has made plans and models of its peculiar structure. These, as well as a large collection of relics, are now deposited in the Prussia Museum at Königsberg. It appears to have been a kind of _Packwerk_. There were, first of all, two or three layers of round timbers lying transversely to each other on the bottom of the lake in the form of rectangles, after which their sides only were continued upwards by single beams, laid successively on each side, thus leaving empty spaces above. These horizontal beams were kept in position by numerous uprights, which here and there flanked them on both sides, as well as by deep cuts towards their extremities where they overlapped each other, precisely similar to the plan adopted at Persanzig and elsewhere. This understructure had a thickness of three to four feet, and over it was laid a wooden platform, above which the huts of its inhabitants were constructed. Clay floorings were found over these platforms, with evidences here and there of fire-places. The relics were found both on the platform and in the originally empty spaces, which, of course, were now filled up with _débris_. Upon its first appearance there was a layer of from 1½ to 2 feet of mud over the woodwork, but after its exposure for some time the mud dried and became greatly contracted. The central area of this structure measured 72 by 36 feet, and was surrounded by three rows of piles. A bridge or gangway, also constructed on a triple row of piles, extended to the shore, a distance of about fifty yards.

At first, and for several years after its discovery, no metal objects were found, and hence it was supposed to belong exclusively to the Stone Age; but this is no longer the case, as latterly it has furnished both iron and bronze objects.

Among the relics are the following:--fragments of an iron socketed lance-head 4 inches long, a large bronze button 1¾ inch in diameter, and a portion of cast bronze. Wooden hooks, like those from Robenhausen, and perforated square bits, supposed to have been floats for nets. Fragments of pottery and some whole dishes, the largest being 18 inches high, showing nail and finger marks (=Fig. 99=, No. 11), perforated rims (No. 10), and sometimes handles. A few flakes, arrow-points, and scrapers of flint (Nos. 8 and 9). Fragment of a perforated axe-hammer of diorite and some mealing stones. Six perforated horn axes, the largest 6½ inches long; pointers, pins, two needles, and various other objects of bone (Nos. 1 to 7). Arrow-points of bone are 3 to 4 inches long, and scrapers 1½ to 3½ inches. Portions of roofing thatch of rushes, clay flooring, etc.

CZARNISEE AND TULEWOSEE.--These two lakes are in the vicinity of the Aryssee, and each contained a lake-dwelling similar to that in the latter, both in structure and in the character of its relics. On the station in the Czarnisee were found a blue glass bead; a partially perforated stone axe, 3⅛ inches long, with the core still remaining; and two socketed iron lance-heads.

The KOWNATKEN Pfahlbau was also a Packwerk formed of round and split stems. It extended along the margin of the lake for about seventy paces, and had a breadth of twelve at the east end, which became reduced to seven or eight at the west end. Some of the pottery from this station, of which fragments of twelve vessels were found, was ornamented with finger marks as well as string marks (_Schnurornament_). Among the relics are pointers of bone, one supposed to be a skate (=Fig. 99=, No. 14); some round sling-stones; a well-formed stone hatchet (No. 12); worked flint flakes (No. 13); and sharpening stones. Among the bones were those of the stag, roe, pig, horse, ox, and portions of reindeer horn.

The lacustrine dwelling in the KOCKSEE was discovered on the lowering of the lake in the autumn of 1882. When its level had sunk 4½ feet, the structure appeared above the water. It was 34 yards long by 15 broad, and stretched lengthways along the shore, with which it was connected by a bridge. The woodwork appeared to have been cut by metal tools, although none of them have hitherto been discovered in the _débris_. The relics consisted of fragments of pottery and a few stone and bone objects.

In the neighbouring PROBCHENSEE a lake-dwelling, in all respects similar to that in the Kocksee, has recently come to light.

At BONSLACK transverse beams were observed, tied to the uprights by means of birch thongs (_Birkengeflecht_). From this station, some pottery, perforated like a sieve, and a mallet of wood, are recorded.

In the SZONTAGSEE there was also a lake-dwelling of the same class as those above described, the exact details of which have not yet been published. From it there are several interesting objects in the Prussia Museum at Königsberg, among which I have noted bone pointers and spatulæ, a well-formed needle of bone with the eye at one end, and a large bronze button with a raised eye.

As to the other localities in East Prussia where indications of lake-dwellings have been observed, the discoveries hitherto made on their sites are too indefinite to merit a detailed notice here, and I shall content myself with the references already given as to where such observations have been recorded.

ARRASCHSEE (LIVLAND).--In 1876 Count Sievers announced the discovery of a lake-dwelling in the Arraschsee, which was subsequently visited by the indefatigable Professor Virchow. This was a small circular island, covered with birch trees and bushes, which, on examination, turned out to have been an artificially-constructed island, like our own crannogs. Like them, also, it was surrounded by piles, and its interior was constructed of layers of wooden beams laid transversely over each other. Its structure was ascertained by digging two large square holes in different parts of the island, and in one nine layers of wood were counted, and in the other six. The chief relics collected were a bronze ring-pin, seven inches long; a bronze fibula (_eine lettische Fibel_); portion of a mould; a few clay beads; a pointed bone implement; bits of string and rolls of birch-bark; also fragments of grey and black pottery, with rude knobs and finger-marks, and without handles. From marks on the woodwork it was inferred that iron tools were used. The osseous remains belonged to the horse, ox, pig, and beaver. (B. 292.)

General Remarks on the Lake-Dwellings of North Germany.

Professor Virchow, as early as 1869, published an excellent thesis on the lake-dwellings of North Germany (B. 165), in which he maintained that all of them, with perhaps one or two exceptions, belonged to a much more recent period than those of Switzerland and South Germany. This opinion he founded on the following considerations:--

(1) Though many objects of stone and bronze were found on the former, yet in almost every case they were associated with others of a more recent type, including iron implements, etc.

(2) The food refuse contained most commonly the bones of the ordinary domestic animals, those of wild animals, such as reindeer, wild boar, stag, wild goat, and beaver, being but rarely met with.

(3) Many of the lake-dwellings were synchronous with the Burgwälle, a fact which was conclusively proved by their possessing the characteristic pottery of the latter, as was notably the case in the Persanzig lake-dwelling. Moreover, Virchow showed that some of the Burgwälle had direct communication with adjacent lake-dwellings, as in the Dabersee, Soldinersee, and Kloppsee.

Referring to this subject at a later date (1877), at the eighth Congress of the German Anthropological Society (B. 306), Virchow, while reviewing the further discoveries of lake-dwellings in North Germany, maintained the general correctness of his previous conclusions. These northern Pfahlbauten, according to him, were due to the immigration into the country of the Slavish people, and bear the same relation to the Burgwälle that the pile-dwellings in Italy do to the terremare. "Ich denke," says he, "wir werden uns entschliessen müssen, ganz im Gegensatze zu den süddeutschschweizerischen Pfahlbauten, die Einführung der nördlichen Pfahlbauten an die Einwanderung des Slavo-lettischen Stammes anzuknüpfen."

Virchow's opinion is not, however, universally accepted, as many of the local archæologists maintain that there are several lake-dwellings which have yielded relics that can only be explained on the supposition that they were founded during the earlier prehistoric ages. The chief examples relied on in support of their contention are those at Wismar, Spandau, Czeszewo, Objezierze, and Aryssee.

After carefully examining the relics from all these stations I must admit that much could be written on both sides of this controversy. Notwithstanding the number of typical objects of the Stone Age from Wismar and Gägelow, Dr. Lisch records that along with them were portions of querns. Now, querns are never found among the remains of the Swiss lake-dwellings, nor am I aware of their existence in any prehistoric remains in northern or western Europe prior to Roman times. The station at Spandau, if it be considered a true Pfahlbau, was undoubtedly of the Bronze Age. Czeszewo and Objezierze have yielded a considerable quantity of Stone Age relics, with scarcely any of the succeeding ages. Only one bronze object is said to have been found on the former, and from the latter there is in the Museum of Posen a bronze torque (=Fig. 98=, No. 9), which was found at a little distance from the lake-dwelling. As regards the Aryssee and its neighbouring lakes of Czarni and Tulewo, with their respective lake-dwellings, all of which are of the _Packwerk_ type, Professor Heydeck relies largely on the presence of pottery with string and finger ornamentation, as a proof of their antiquity, in addition to the ordinary stone celts, horn clubs, etc.; but yet along with them were found iron lance-heads and a blue glass bead (Czarnisee). Similarly in the _Packwerk_ in the Kownatkensee, polished stone celts, pottery with finger marks and _Schnurornament_, and portion of a reindeer horn, were found associated with a bone skate, and the osseous remains of the domestic animals, as the horse, pig, ox, etc. In attempting, therefore, to estimate the chronological range of these lake-dwellings from an examination of their contents, which (being unnoticed in the early annals of the country) is the only available means, the mixed character of these relics presents a considerable difficulty--a difficulty which, as we shall afterwards see, is equally applicable to the Scottish and Irish crannogs. But, whatever doubts may be cast on their antiquity and early origin, there can be none as to the comparatively late occupancy of many of them. A bone skate and a comb made of square bits bound together by cross pieces, and showing that the teeth were cut by a saw after the pieces were put together, precisely as may be seen in the combs from the Scottish crannogs and the terp-mounds of Holland, were found by Virchow in the Dabersee Pfahlbau. Iron hatchets (Dabersee, Persanzigsee, and Alt Friesack), horseshoes, and other iron implements (Soldinersee), pottery of Slavish type (Bonin, Kloppsee, Persanzig, etc.), leather (Bonin), and even armour and bricks of the thirteenth century (Lübtowsee), leave no doubt as to their almost mediæval character.

Reindeer horns were found at Butzow, Soldinersee, and Kownatkensee; but these objects do not necessarily indicate great antiquity, as this animal, though not referred to in the early annals of North Germany, is stated to have been an inhabitant of the country in the time of Cæsar.

The undoubted contemporaneity of many of these lake-dwellings with the Burgwälle opens up a field of research of considerable importance to European archæology; but their exact chronological relationship still remains an obscure problem, owing chiefly to the mystery which surrounds the latter.

_Burgwälle_ or _Rundwälle_ is the general name given to the remains of a remarkable class of prehistoric constructions found scattered over the larger portion of middle and north-western Europe, embracing the southern parts of Russia around the shores of the Black Sea, Roumania, Bulgaria, Transylvania, Hungary, Austria, Bohemia, Poland, North Germany, France, Great Britain, and the southern parts of Scandinavia. Their foundations now only remain and these show that the structures were generally circular or oval, but sometimes square and semicircular. They may be divided into three kinds, according to the materials of which their foundations are constructed, viz.:--earth, stones or stones in vitrified condition (_Erd-, Stein-und Schlackenwälle_). Their sizes vary from 20 to 100 paces in diameter, and their height from 10 to 30 feet, and they contained one, two, or sometimes three walls. Those made of earth were circular and generally situated in swampy land, or in countries where stones were not readily accessible. The Steinwälle were in hilly districts and varied in form according to the nature of the ground. Sometimes they assumed the irregular outline of a promontory or peninsula in a lake, at other times, especially when placed on an overhanging cliff, they were mere semicircles. Those of vitrified materials are of special interest to Scottish archæologists owing to the number of vitrified forts in Scotland. They are not very numerous on the Continent, Saxony and Bohemia containing the largest number. In the former country eight are known, viz.:--Schafberg by Löbau, Rothstein by Sohland, Stromberg by Weissenberg, Landeskrone by Görlitz, Brandwall by Blumberg, Koschütz near Dresden, Burgberg by Lichtenberg, and Vorberg by Kirchberg. According to Jelinek, Bohemia is rich in Schlackenwälle, those best known being near Katovic, Bukovec, Litoradic, Hradiste von Hostem, Hradiste bei Strakonic, Hradec bei Domanic Burgberg, Vladar, etc. ("Schutz-und Wehrbauten." p. 102). Instances also occur in Silesia, Thuringian Forest, Rhine district, Brittany, and Normandy.

The Burgwälle, like their analogues in the British Isles, have not yet been systematically investigated. From the character of the relics found in those that have been explored in North Germany they are divided into _Vorslavisch_, _Slavisch_ and _Spätslavisch_, a distinction which has been suggested by the unique character of Slavish pottery. These Slavish dishes are always without handles, but of well-burnt pottery, and when ornamented the ornamentation is in wavy lines running parallel to the rim forming the characteristic _Wellenlinie_.

Many of these remains have, of course, now entirely disappeared in the interests of agriculture, but their number still remaining is very great. In Eastern Germany Dr. R. Behla describes and tabulates no fewer than 1,100. They are more numerous in the fertile districts. In Oberlausitz, in one district measuring 9 miles long by 3 to 16 broad, they number 100, and in the neighbourhood of Bautzen within a one mile circle 20 can be counted.[59]

It is probable that the material used in the upper structures of the Burgwälle was wood, which, of course, has now completely decayed, except in some special conditions, as in swampy ground where wooden piles were used in their foundations. This is another point of contact between these buildings and the lake-dwellings which has not been overlooked by archæologists. Virchow describes the Burgwall of Potzlow, and that of Zahsow near Cottbus, as constructed over former Pfahlbauten;[60] and, indeed, the town of Cottbus seems to have been altogether built over piles, as, wherever diggings have been made, piles are met with, and in this way a finely ornamented quern was found.[61] Wooden substructures, in the form of a platform or _Packwerkbau_, have also been observed and recorded in many places, as at Schlieben, Gross Topola (Posen), the Labenzsee, Westpreussen.[62] Moreover, those in boggy places were approached by means of wooden gangways, the remains of which have been frequently met with in the form of a double row of piles.[63]

ANCIENT MARINE DWELLINGS ON THE COASTS OF HOLLAND AND WESTERN GERMANY.

Notwithstanding the striking and singular appearance the Swiss lake-dwellings must have presented to foreigners and strangers, it is a remarkable fact that Roman writers are entirely silent about them. Nor can this silence be accounted for on the supposition that the lake-dwellings had entirely come to an end prior to Roman times, as several of them have furnished antiquities whose Roman origin cannot be mistaken. Some archæologists think they recognise in the representation of a Dacian village on the Column of Trajan a true pile-village (B. 164); but this is doubtful, and, even if true, it is but a very meagre evidence of the custom, and leaves the problem of the lake-dwellings as mysterious as ever. Such reticence on the part of classical writers does not, however, extend to the class of ancient remains I am now about to describe.

Pliny very distinctly states that the Chauci (Frisians and other races along the coast of the German Ocean) were in the habit of constructing artificial mounds, on which they built their houses so as to be beyond the influence of the waves and tides. The following passage from his "Natural History"[64] will be read with interest in relation to the recent discoveries that have been made in the localities referred to.

"I have myself personally witnessed the condition of the Chauci, both the Greater and the Lesser, situate in the regions of the far north. In these climates a vast tract of land, invaded twice each day and night by the overflowing waves of the ocean, opens a question that is eternally proposed to us by Nature, whether these regions are to be looked upon as belonging to the land, or whether as forming a portion of the sea?

"Here a wretched race is found, inhabiting either the more elevated spots of land, or else eminences artificially constructed, and of a height to which they know by experience that the highest tides will never reach. Here they pitch their cabins; and when the waves cover the surrounding country far and wide, like so many mariners on board ship are they; when, again, the tide recedes, their condition is that of so many shipwrecked men, and around their cottages they pursue the fishes as they make their escape with the receding tide. It is not their lot, like the adjoining nations, to keep any flocks for sustenance by their milk, nor even to maintain a warfare with wild beasts, every shrub, even, being banished afar. With the sedge and the rushes of the marsh they make cords, and with these they weave the nets employed in the capture of the fish; they fashion the mud, too, with their hands, and drying it by the help of the winds more than of the sun, cook their food by its aid, and so warm their entrails, frozen as they are by the northern blasts; their only drink, too, is rainwater, which they collect in holes dug at the entrance of their abodes; and yet these nations, if this very day they were vanquished by the Roman people, would exclaim against being reduced to slavery! Be it so, then--Fortune is most kind to many, just when she means to punish them."

Notwithstanding the preciseness of Pliny's description and the fact that for several centuries, since the great sea-dykes were erected, the scattered remains of these mounds have been accessible on dry land, they have only quite recently attracted the attention of archæologists. I consider their investigation important, not only for the large amount of industrial remains they contain, but for supplying a missing link in the evidence of continuity in the European habit of constructing pile-dwellings.

TERPEN (WEST FRIESLAND).

Before the construction of the great sea-dykes in Holland nearly the whole of West Friesland would have been in that hybrid condition described by Pliny in which it was difficult to say whether it belonged to sea or land (_dubiumque terræ sit, an pars maris_). At the present time, however, these lands are richly cultivated and look as if they were a dead level. It is only on close inspection that the monotony is relieved by certain elevations of considerable extent called _Terpen_, whose summits rise to about the level of the larger dykes. These mounds are situated at more or less regular intervals, so that if the tides by any calamity had free scope, they would appear as so many islands scattered over the country. It is on such elevations that modern churches and villages are generally built, and, till they accidentally attracted the attention of agriculturists, nobody seemed to think anything about their origin. A few years ago it was discovered that their interior was composed of a rich ammoniacal deposit which agriculturists found valuable as a fertilising agent when spread over their fields. The excavation of this substance for manuring purposes now forms an important industry, and any landed proprietor who happens to own a workable _terp_--_i.e._ one free of buildings--is on the way to realise a small fortune. When a _terp_ is found suitable for being excavated they generally commence by digging a canal close up to its base, sufficiently large to admit of the passage of good-sized boats. The boats are then easily loaded with the stuff and so it is conveyed to all parts of the country. As the workings advance the canal is also advanced, so that the boats are always in close proximity to the diggings. In the course of these operations, bones and horns of various animals, pottery, and other relics of human industry, were occasionally turned up.

By degrees these repeated discoveries attracted the attention of antiquaries, and Dr. Pleyte, of Leyden, is now publishing a large illustrated work on the antiquities of Holland (B. 301), in which a conspicuous place is given to the terp-mounds and their contents. It is, however, to some of the office-bearers of the Museum of the Friesch Genootschap at Leeuwarden, more especially Mr. Corbelijn Battaerd, its conservator, that I am indebted for much of my information on the subject. In this museum are stored up most of the objects hitherto found in the terp-mounds, and the collection, already unique of its kind, is daily and rapidly increasing, as orders have been issued in regard to many of them that no relics are to be disposed of without being, in the first place, submitted to the authorities of the museum.

Like most countries, the early traditions of Holland have been forgotten or ignored, and in its annals little mention is made of the _terpen_. In explanation of the origin and early use of the word, Dr. Pleyte quotes from Ocko van Scharl a passage to the effect that one of the ancient kings of Friesland, named Adgillus, who reigned towards the end of the sixth century, had caused, on account of the ravages of an inundation which took place four years prior to his accession, a large number of elevated places to be formed, so as to give shelter to man and beast in the event of a recurrence of this danger. These mounds were then called _Terpen_.

Mr. Dirks, president of the Friesch Genootschap, as early as 1871 characterised these mounds as analogous to the terramara beds of North Italy ("_ce sont des terramares historiques_");[65] but it remained to Professor Pigorini of Rome to show that they were identical as regards internal structure. This he did in 1881 (B. 372c), after a visit to one at Aalzum which was then being excavated, when he showed that there was a circumscribing dyke, and, although no actual piles were then visible, he was informed by the proprietors that such wooden structures had been occasionally met with. Prior to his visit, it appears that no special attention was directed to these structural remains. From all he could learn, however, on this point, and especially from a consideration of the stratified arrangements of the _débris_, Pigorini concluded that the deposits were due to pile-dwellings, and had accumulated under precisely similar conditions to the terremare, in regard to which he is such a distinguished authority.

The _terp_ at Aalzum is still being systematically excavated, and, though only as yet partially cleared off, its results, from an archæological point of view, are now second to none of the kind in Holland. Moreover, the excavations are conducted on an extensive scale, and the locality is readily accessible. I can, therefore, conceive of no better means of conveying to you some knowledge of the nature and structural phenomena of these remarkable deposits, than by detailing the facts which came under my own cognisance during a visit I made this summer to the same spot under the guidance of my excellent friend, Mr. Battaerd.

The _terp_ lies about a mile to the north of the town of Dokkum, some twelve miles from Leeuwarden, and four or five from the seashore. In approaching the locality from Dokkum there was little to attract special notice beyond the usual Dutch scenery--canals, rich meadows, herds of splendid cattle, and here and there some well-cultivated cornfields. In front of us a slight elevation could be discerned, crowned by a small church in the midst of a clump of trees, the surroundings of which were neatly hedged meadows and cornfields. As we advanced towards this church, and within a few hundred yards of it, we entered on a sloping road, as if raised on a dyke, but on each side the land was perfectly flat and bearing a splendid crop: here a field of magnificent beans, and there an equally promising one of wheat. These fields, said Mr. Battaerd, were formerly part of the _terp_-mound from which the fertilising stuff has already been removed, but this road was left undisturbed, so that we are now actually walking on a portion of its surface. By-and-by we came in sight of heaps of clayey stuff, the tops of which sparkled with reflected light, and in their midst were to be seen the masts and rigging of three boats. Those whitish clay-like heaps, said Mr. Battaerd, formed the surface soil, which, being of no commercial value, had to be wheeled off before the saleable deposits could be got at. At last the actual workings were reached, and we found ourselves in front of a perpendicular section some 15 or 18 feet high, from which men and women were busily engaged in loading the boats. Uppermost in my thoughts was the paramount question of the existence of upright piles, which, it will be remembered, Pigorini had not actually seen. Great was my delight when, at the very first glance, my eye detected an undoubted pile of oak just in face of the cutting. Close by it I soon found another and as we moved along numbers were observed, some soft and yielding, scarcely offering any resistance to the spade; and others of oak very hard in the centre, but more decayed and ragged-like than those I have been in the habit of seeing among the lake-dwelling remains. Those seen in this section differed considerably in size; and I observed that some penetrated deeper than others. At a little distance lay a heap of oak beams which had recently been removed from the trenches--one of which I measured and found it to be four yards in length, and from six to eight inches thick. Upon inquiry, I ascertained that these beams lay horizontally, and about half way down, in the stratified stuff.

Those who, like Professor Pigorini, are acquainted with the structural features of the terremare of Northern Italy, will not be surprised at the comparative rarity with which piles are met with in the _terpen_, because of the rapidity with which timbers, when buried in dry earth, decay and disappear altogether, leaving in many instances no traces whatever behind them. This fact was strikingly shown by Chierici, who produced positive evidence of the former existence of piles in the upper strata of some of the terremare, by showing that the holes left by the piles, after the woody fibre had completely disappeared by decomposition, had become subsequently filled up by dust and infiltrated material, which ultimately became hardened, and so retained the actual form of the original piles. (See page 248.) In short, natural casts of the original piles were accidentally formed, which thus disclosed a knowledge of their former existence, which otherwise might never have been suspected. To the soundness of this deduction I have myself unconsciously contributed by an observation which I made some years ago, while digging at the crannog of Lochspouts, and having recorded it I may perhaps be allowed here to repeat my words. "One day I was greatly puzzled by finding what was evidently a portion of a birch tree, from 6 to 9 inches in diameter, quite flat, and with scarcely any wood left inside the thick bark. In no instance previously had I seen the evidence of pressure on logs of this size; but after carefully considering the point it was ascertained that such effects occurred only in the upper portion of the mound, and above the log pavement, where the wood had been exposed to atmospheric influences, so that when the woody fibres rotted away the flattening of the bark was easily produced. All the logs found buried in water or mud retained their original dimensions and showed no trace of having yielded to superincumbent pressure."[66]

The absence of piles and wooden structures from many of these mounds is, therefore, no proof that they have not formerly existed; and, indeed, it is difficult to account for the horizontality and regularity of the beds on any other hypothesis.

While I wandered about amidst the various sections presented by the progressive stages of the excavations, wondering at the distinctness of the strata, or picking up stray objects from the _débris_, such as mussel shells, bits of bone, fragments of pottery, etc., which were to be found here and there sticking in the face of the cuttings, my friend, Mr. Battaerd, was deeply occupied in examining a heap of bones, which lay weathering in a sunny corner. Having joined him in his osteological study, I found that the chief point of attraction was the head of a urus (_Bos_ _primigenius_) of great size, and with splendid horn cores--the finest example, according to Mr. Battaerd, that had yet found its way to the museum.

The land close to the brink of the section, and extending over a considerable portion of the mound, was occupied by growing corn, and hence its dimensions can only be approximately stated. The proprietors, Messrs. W. and J. Bierma, obligingly accompanied us, and one of them assured me it could not be less than from three to four hundred yards in diameter. Its greatest height above the water in the canal was 18 feet, but of course the level of the canal water is considerably lower than that of high tide in the open sea. The commercially valuable stuff commenced some 3 or 4 feet below the surface, and continued without interruption to within a few feet of the canal water. It was in this intermediate portion that the relics were found: but their exact position, especially that of the smaller objects, was seldom determined, as it was generally after the stuff had become partially broken up during transport that they were found.

The stuff _in situ_ was distinctly stratified, forming layers of various thicknesses, from a finger breadth up to 3 or 4 inches, or sometimes more, which in some instances could be continuously traced for long distances. Sometimes they shelved out altogether, and others commenced. Here, a bed of fibrinous matter, in which quantities of the partly decomposed fibres of flax could be readily recognisable; there, a thickish deposit of a brownish glutinous stuff like peat. Charcoal and ashes permeated the whole, and showed themselves sometimes as distinct layers. Clay and sand were also largely mixed with these deposits, and occasionally assumed the form of distinct and separate beds.

Having so far satisfied ourselves as to the structural arrangements of the mound, and the disposition of its contents, we walked up to the church, which is but a short distance from the workings. This small edifice is surrounded by a burying ground, and among the gravestones are some ancient-looking ones. Mr. Battaerd informed me that it dates as far back as the eleventh century.

It is calculated that there are altogether about 150 of these mounds in West Friesland alone, and that of these about the half have been more or less examined, some being now entirely cleared away. They are also to be found in the province of Gröningen and some other parts of Holland. Dr. Dirks states that the town of Leeuwarden is built over two terp-mounds;[67] and Dr. Pleyte informed me that he has reason to believe that the town of Leyden also reposes on similar deposits.

RELICS (=Fig. 100=).--The relics of human industry collected from the _terpen_ are very varied and numerous. Of these the following notes and illustrations, taken chiefly from the large assortment in the Leeuwarden Museum, will serve to convey some general idea of the social economy which prevailed among the occupiers of these singular settlements, as well as of the period in which they flourished.

_Prehistoric._--The prehistoric remains, commonly so-called, such as cutting implements of stone, are only feebly represented, but occasionally they do turn up, in which respect the _terpen_ resemble the Scottish and Irish crannogs.

_Clay Objects._--Perforated loom-weights, both conical and flat; spindle-whorls in great numbers, and often ornamented with finger marks or grooved lines (Nos. 2 and 3). Some flat and triangularly-shaped objects of clay (No. 22) are perforated with three holes, one at each angle, which are sometimes perpendicular and sometimes parallel to the surface; in bulk and composition they correspond with the loom-weights.

_Pottery._--Pottery is, as a rule, coarse but abundant, and represents vessels of various shapes and sizes, generally with ears, but a few with handles (Nos. 20 and 23). Samian ware is represented by many fragments of bowls and dishes. A few vases, apparently home-made, have some traces of coloured patches; and there are lids with raised handles and ornamented with hollowed dots.

_Bone and Horn._--Bone and horn implements are very abundant, consisting of combs (Nos. 1 and 16 to 19) of varied forms, and constructed of plates riveted together with iron rivets, and ornamented with consecutive circles, lines, dots, and curvilinear figures; among them are also a few combs with very long teeth (No. 30).[68] There are also pins (Nos. 26 and 28), needles (No. 29), buttons (No. 25), dice (No. 21), finger rings (No. 12), knife handles, pointers, etc. (Nos. 10, 11, 13 and 27). Many so-called skates made from the long bone of the horse's leg. Two or three short bones (foot of the ox) are covered with concentric circles, apparently for ornamentation. A curious bone object (No. 7) is supposed to have been used in making twine or ropes.

_Glass._--Beads, blue, green and variegated; also glass slag.

_Metal Objects._--A few bronze dishes (No. 5), one a tripod with projecting handle (No. 15). Figurines of men and animals; the hand of a Roman statue, apparently a female and about full size; Roman fibulæ; some three or four double spirals; a small pair of shears (No. 8), and a few bracelets with clasping-hooks. All these are of bronze. Among objects of iron are shears, hammers, bridle-bits (=Fig. 101=), slag, etc. A leaden bar or pig weighing 17 kilogrammes and marked with three crosses, so, "XXX", was found at Achlum.

_Coins._--Anglo-Saxon coins very abundant: at Hallum 180 _sceattæ_ were found in a jar; Byzantine money in gold; Roman imperial money, generally in silver, but sometimes in gold; Frankish coins. The proprietors of Aalzum found a few silver coins in this terp with the following inscription: "+ HLOTHARIVS. IMP. DORE STATVS MON (_eta_)," which defines their date to be between 840 and 855 A.D.[69]

_Wooden Objects._--Small spades precisely similar to those used by children while amusing themselves by digging the sand on the seashore. Numbers of large casks the staves of which are kept together by three iron hoops. In diameter these casks are not more than an ordinary herring barrel, but in length they are from six to seven feet, and about one-third from the top there is a small square hole 4 or 5 inches in diameter. The ends of the staves at the top rim of some are much decayed, but the rest is perfectly sound, and for this reason they are supposed to have stood in water with only the upper parts exposed. They have been found in almost all the _terpen_ examined, usually at regular distances, and deeply buried. One, 6 feet high, was found resting inside a vat 3 feet deep, and its highest point was over two yards below the surface of the mound. Canoes and small paddles may also be mentioned as occasional relics.

_Nondescript Objects._--Cock spurs; egg-shells of the domestic fowl and goose, some of which, singularly enough, were, when found, still unbroken; shells of various kinds of sea-urchins, star-fishes, and mussels; amber beads, also this material in the unworked form; amorphous vivianite; large quantities of the _débris_ of flax; one curious object is a flute made of the shank bone of a small animal; one small fictile dish has four feet, and a few others are in the form of three cups attached. At Aalzum, on the occasion of my visit, among the articles purchased by Mr. Battaerd were a mitten and some sort of head-dress like a felt wide-awake. The mitten had only one stall, for the thumb.

In the _terp_ called Beetgum there was found an urn, like those from the dolmens of the Drenthe, containing some burnt bones. Human bones are sometimes found, but they are supposed to have belonged to secondary burials. At Aalzum a grave was found containing a body and along with it was a fibula of the Merovingian period, with a flat back containing a beautiful mosaic pattern of variegated glass and amber.

_Fauna._--Osseous remains representing the following animals:--Horse, ox (several varieties--_Bos taurus_, _primgenius_, _longifrons_, _brevicornis_), cat, dog, sheep, wild boar, deer, roe, and fallow deer. Among the skulls of these animals (of which there are many) are one or two of the four-horned sheep. It may be of interest to note that the osseous remains of this animal were among those identified by Sir W. R. Wilde as coming from the crannog of Lagore (page 351).

WARFEN (EAST FRIESLAND).

In 1879 Dr. Tergast, of Emden, published a short account of the prehistoric antiquities of East Friesland,[70] in which he takes notice of the existence of certain mounds, in the low-lying regions, called "Warfen," which he believes to be the remains of very ancient settlements constructed for the protection of their inhabitants against floods and the fluctuations of the surrounding waters. The author does not give many details about these mounds. It would appear, however, that they are to be met with in considerable numbers, as he suggests that it would be of the highest interest to archæological science to have a map constructed showing their local distribution. Nor do they appear to have been subjected to much practical investigation, as only three objects from them are illustrated in Dr. Tergasts book. These are a bone implement (so-called skate or cloth-polisher), a necklace of glass and amber beads, and an iron arrow-point. He also figures a comb (six inches long) similar to that from the _terpen_ (=Fig. 100=, No. 30), but without specifying the locality where it was found. All these are precisely similar to objects found in the terp-mounds of Holland. Every indication, therefore, points to the conclusion that the _Terpen_ and _Warfen_ are quite analogous to each other and belong to the same period of time.

WURTHEN (DITHMARSCHEN).