Lake-Bottom Archaeology

Lake-Bottom Archaeology Mountain lakes rom France to Ausria and rom Germany to Italy attracted many prehistoric setlers. Underwater archaeologists are now investigating sites that once were only dredged by colectors by Aime Bocquet O ne of the most familiar pictures of how people lived in prehis­ toric times is that of European lake dwellers: Neolithic and Bronze Age farmers who lived in houses built on pil­ ings over the water. This picture is false, and has been known to be so for dec­ ades. The evidence is that these villagers often built their houses at the side of a lake but rarely, if ever, built them over the water. The reason it appeared they had done so is that since that time the level of the lakes has risen, submerging the remains of the villages and leaving them some distance from the shore. Archaeologists may regret the per­ petuation of the lake-dweller myth, but they can be grateful for the numer­ ous submergences that gave rise to it. Drowning has preserved many of the re­ mains that would long ago have been destroyed by natural causes at any dry­ land site. For example, wood objects and textiles that would weather and de­ cay at dry-land sites can be almost per­ fectly preserved in wetland and under­ water locations. Equally important, the same is true of both wild and domesti­ cated plant materials. Only a decade ago few investigators paid much attention to the clues such materials ofer to the var­ ied environments of prehistoric times. Today, however, archaeologists set great store by such information. It can reveal the extent of prehistoric man's environmental adaptation, and the more (or less) eicient prehistoric man's adap­ tation to his environment was, the more (or less) assured was his survival. A knowledge of prehistoric environ­ ments is also a prerequisite for assessing man's technical progress and his inlu­ ences on the world around him. For ex­ ample, a forest environment ought to inspire the invention of the woodsman's ax. When an archaeologist excavates an early forest dweller'S shelter, it is good to ind the axes the woodsman wielded. It is even better, however, to discover what kind of forest the woodsman in­ habited by analyzing the plant remains. It is better still if the analysis reveals the woodsman's impact on his environ­ ment, for example his selective cutting of certain species of trees. Studies of this kind are facilitated by the preservation of plant materials at wetland and sub­ merged sites. The eastern reaches of France, partic­ ularly those areas adjacent to the north­ ern face of the Alps, are abundant in lakes and lakeshores that provided vil­ lage sites from as long ago as Neolithic times up to the Middle Ages. Most of these villages were later drowned; many were irst discovered more than a centu­ ry ago during an exceptionally severe drought. In 1853 and 1854 the level of lakes fell throughout Europe. At Lake Geneva, for example, many long-sub­ merged shore dwellings on the French side of the lake were exposed to view. Today the number of such sites discov­ ered at the mountain lakes of France has grown to a total of 38. Similar drowned villages were exposed in western and central Switzerland, southwestern Ger­ many, Austria and northern Italy. n the years between the mid-1850's and 1935 rich collections of "lake dweller" artifacts were assembled. They are now to be found in private collec­ tions and in museums throughout Eu­ rope, but with a few rare exceptions the original collecting was far from scientif­ ic. The practice was merely to drag up to the surface various objects that were scattered on the lake bottom or were shallowly buried in the bottom silt. The collectors often worked with ingenious dredging tools in order to reap this spe­ cial harvest. I Much of the work was done at such Swiss sites as Auvernier and Cortaillod on Lake Neuchatel and Meilen and Wollishofen on Lake Zurich; the French contribution was mainly from Lake Bourget in the Savoy, halfway between Grenoble and Geneva. As if to fore­ shadow a more scholarly approach, in 1929 the drowned site of Sipplingen, on the German shore of Lake Constance, was pumped dry a little at a time, with the aid of caissons, thereby allowing the methods of dry-land archaeology to be applied during excavation. With the invention of practical free­ diving gear in the 1940's an entirely new means of underwater investigation became available to archaeologists. A number of amateur divers began to col­ lect "lake dweller" antiquities with the aid of the new apparatus. A. Favre, who worked at Lake Annecy, a little to the northeast of Lake Bourget, was one such early worker. These divers' work was not, strictly speaking, excavation, but it was a considerable advance over the earlier dragging and dredging. True lake-bottom archaeology began in 1952, and its pioneer was Raymond Laurent of the Centre de recherches ar­ cheologiques lacustres du Dauphine­ Savoie. Inluenced by the teachings of an eminent French prehistorian, An­ dre Leroi-Gourhan of the College de France, Laurent brought to his under­ water work a comprehensive grasp of how to organize a lake-bottom site for the precise recording of the horizontal and vertical distribution of artifacts. He too worked at Lake Bourget and Lake Annecy and he also worked at Lake Aiguebelette, a little to the southwest of Lake Bourget; his underwater excava­ tions between 1960 and 1969 paved the way for today's fully disciplined lake­ bottom work. By 1963 Ulrich Ruof was conducting equally sophisticated inves­ tigations in Switzerland at the Lake Zu­ rich site of Kleiner Hafner. At the same UNDERWATER EXCA V ATOR at Baigneurs, a drowned Neolithic village in eastern France, is seen at work in the photograph on the opposite page near one apex of a triangular duralumin frame. The equilateral triangle has sides ive meters long. Red and white ribbons subdivide the area into 25 smaller equilateral riangles, each enclosing an area of 430 square centimeters. In front of the excavator is what remains of an upright timber that was once a corner post in a Neolithic farmhouse. To the excavator's left is the collecting bucket that will carry to the sur­ face the house-loor debris he has gathered after noting its horizontal and vertical coordinates. 56 © 1979 SCIENTIFIC AMERICAN, INC 57 © 1979 SCIENTIFIC AMERICAN, INC It is my intention here to describe lake-bottom archaeology as it stands to­ day. taking as my example a Neolithic lake-bottom site in the northern Dau­ phine. southwest of the Savoy. The site was singled out for salvage archaeology in 197 1. The lake whose shores once at- time work of a more traditional kind continued. For example. between 1969 and 1972 a number of sites in the Auver­ nier Bay section of Lake Neuchatel were excavated by divers. and other bay sites were diked and pumped dry to allow conventional excavation. GERMANY F RANCE ' '_ DIJONe ./ . / / . . _ ( ( BASEL .. - LAKE - B1ENNE J/' (. . < LAKE /' eZURICH LAKEZURICH� . � �- , SWITZERLAND . \ . l : O USTRIA '-'-oj LI ECHTENSTEIN NEUCHATEL ::, ? ... '. . LAKE < ANNECY/)..--'/ LAKE BOURGET � �,. LAKE AIGUEBELETTE � . ITALY eMILAN ( eTURIN SIX-NATION SECTOR of alpine and subalpine Europe contains the many large and small lakes that were selected by Neolithic and Bronze Age farmers for lakshore settlements. A rise in lake levels concealed the deserted villags until a drought in the mid-19th century exposed them, bringing about the myth of the Alpine lake dwellers. The laks named here contain many of the principal Neolithic and Bronze Age sites; only a few have been scientiically excavated. LAKE PALADRU, in the Dauphine southeast of Lyons, givs rise to a tributary of the Isere River. Near Charavins at the outlet of the lake is the drowned Neolithic settlement of Bai­ gneurs. The author and his colleagues have undertaken rscue archaeology there since 1972 while at the same time conducting a workshop in the techniques of underwater archaeology. 58 © 1979 SCIENTIFIC AMERICAN, INC tracted Neolithic settlers is called Pala­ dru; it is a little less than 45 kilometers northwest of Grenoble on the route to Lyons. and the speciic drowned shore­ line site we investigated. some two to three meters underwater. is known as Baigneurs. The lake-bottom strata in which the Neolithic remains are found were deposited some 5.000 years ago. They are thin and readily broken up. so that they call for the most painstaking excavation. ur salvage operation at Baigneurs was based on the Laurent system of lake-bottom topographic analysis. the key requirement of which is that the location of all inds should be recorded quite precisely. Also implicit in the Lau­ rent system is that all the artifacts in each lake-bottom stratum. no matter how small or fragile. are salvaged. Our own contribution to the Laurent system was to make sure that literally all the objects-not just the man-made ones-were collected. Ever since my colleagues (Fran�oise Ballet. Patrick Grandjean. Christian Orcel and Alain Cura) and I began our work some seven years ago that has been a prime objec­ tive. Only this kind of collecting makes it possible to reconstruct the environ­ mental shifts of Neolithic times. The irst problem confronting any lake-bottom archaeologist is limitations in visibility. Even without taking into account the clouds of particles that rise from the 1ere act of digging in the lake bottom. the range of the diver's vision can be only a few centimeters, depend­ ing on the surface winds, the lake cur­ rents and particularly the season. Dur­ ing much of the year lake waters are naturally clouded with seasonal growths of algae and plankton. That is one reason the grid used to control dry-land excavations (a checker­ board of one-meter squares) is virtually useless underwater. It is diicult enough under conditions of limited visibility to lay out such a grid with geometric preci­ sion, and it is even more diicult to mea­ sure the actual location of an object within a particular square. One must run two perpendiculars from the object to two adjacent sides of the square and re­ cord both the precise intersection point and the precise length of each perpen­ dicular. Laurent's solution for the grid prob­ lem was simple and inspired: instead of squares he used equilateral triangles. To locate an object found within one of the triangles in such a grid it is necessary only to measure the distance to the ob­ ject from all three of the triangle's apex­ es. This is done by means of tapes at­ tached to each apex in the grid. To build the grid itself one begins with a single known point and three small girders of equal length. At each apex the girders are interlocked by a collar that its onto a vertical tube sunk into the lake bot- LOW-LEVEL AERIAL PHOTOGRAPH of Lake Paladru has been righs that formd a palisade on the inland perimeter of the vllage. (white) locats one large house (cicled uprighs) at the (broken annotated by the excavatos to indicate the extent of the drownd Annotation Neolithic village. In preparation for the photograph the dives set northwest corner of the village near the ancient lakeshore white metal plates on top of each upright house timber they had locat­ line) and two other houss nearby. The palisade and is gate are also ed on the lake bottom and set red plates on the smaller timber up- annotated. Additional nprights and debris are still being investigated. RECONSTRUCTION suggess the probable appearance of the rst setlement at 8aigneurs after the third farmhouse was built there. The village livestock when the animals were not herded into nerby mead­ ow and forest to forage. Fishing on the lake was a regular activity, as palisade was evidently not defensive but was a means of conining the is shown by the recovery of net weights and a shhook from the ruins. 59 © 1979 SCIENTIFIC AMERICAN, INC .. , LINE '" WATE� ' . ... . . TO P UMP . �.'.:' ' . �. . .. ., .- .. �'. . . . . . ' . ' :. . . ' , " " � '." ' " .�..-.. :. . . .:.. .. . ':. ·i:· ·· ··· ·· . �, .. ... : . . ,- . �.; .. :.) .. r.': " ;:.' , '.. . ", ' : : ";> ..... . , '. � . . ... ::., . . :.�, ' " :, ."�:'�"'"'' .. •.. . . . > .. ..... : . ". . ... : .. . . . . ...... : . .... : ' . . .-,: ..: . . ... .. ' .; .. . . .. :" . ,�.:, ' . . : ..-. . . . -- . . . : .. . . .... :.:::: :, . ," -7···. :·.·: . ... ', . ,. . 0', . . . . . ..... . .. . .. , . . . ...... ... . ::Q'�'."'" "WATER CURTAIN," a device developed by the Swiss diver Ulrich Ruof in 1963, is used at Baigneurs to improve lake-bottom visibility. In the diagram the water-curtain tube has been secured along one side of a ive-meter triangle; the lake botom coverd by the six small tri­ angles nearest the tube (color) s being excavated. A submergd pump supplies 60 cubic meters of water per hour to the tube; the water emerges through a series of bullet-size holes spaced 20 (black arrows) creates a local current (colored arrows) centimeters apart. The low of water that serves to carry away the particles of silt that cloud the water in the diver's work area. - ,,------ -- - - - - - --"- -- --- - - -- - I " / / I /\ \ \ / I / / / '-- /\ \ - -- -" - - ---- ---- -- - - /\ -\ / '\ \ \, , , , ' --------' ! !\ / / / / / / / / / / / / , \--------, \ , , \ \ / \ , , , ----- I I I I / , , , \, , --\ \ , \ T /\ I I - ____ �yl LIMIT OF EX i \ \ , \ \ I I \ \ I " \, I" - __ ________ ___ __ L _________________ " I I " I I I I / \ / 1 I I / / I I / I I / I / / / tAVATION / o 5 METERS I UNDERWATER GRID, used to determine the exact position of excavated materials, consists of interlocking equilateral triangles. The diagram shows the part of the site excavated so far (gray line) and the inuer set of six triangls that the divers were able to bring together to form a (gray tone). The cumulative error in closing the hexa­ hexagon with a 30-meter circumference gon was only seven centimetes. Additional riangls, erected on the core of the central six, embrace the entire excavation area; all but one of them extend beyond its perimeter. The sub­ division of a ive-meter triangle into 25 smaller triangles tom. Once the irst triangle has been set up the grid can be extended indeinitely by the addition of girders that are inter­ locked in the same way. The accuracy of the system is well within the neces­ sary limits. Using ive-meter duralumin girders to form successive triangles we have been ab'e to "close" a 30-meter hexagon made up of six contiguous tri­ angles with a cumulative error of only seven centimeters. Although the interlocking metal tri­ angles provide an excellent reference grid for both horizontal and vertical measurements, the area within each tri­ angle (nearly 11 square meters) is much too large for a single mapping unit. We therefore use ribbon or string to subdi­ vide each large triangle into 25 small triangles that measure one meter on a side. The area within each sm,ll triangle is much more manageable: it is less than half the area of the square-meter unit of the conventional dry-land grid. Once the reference points are estab­ lished underwater, excavation can be­ gin. The divers may, depending on sea­ sonal factors, be handicapped by the naturally poor visibility, but they no longer need to work in a cloud of silt of their own making. The reason is that an apparatus developed by Ruof in the early 1960's establishes artiicial water currents to carry away the ine particles that are stirred up by the digging, much as a breeze carries away smoke. We call the current created by the Ruof appara­ tus a water curtain. (color) is shown here in one location. 60 © 1979 SCIENTIFIC AMERICAN, INC he excavators working on the lake bottom bear the primary collect­ ing responsibility. but those working on shore to process the sediments collected by the divers make their own vital con­ tribution. Their sieving extracts from the sediments the largest possible quan­ tity of material for analysis. It is not for nothing that we refer to our kind of ar­ chaeology as two-story excavation. The diver's work is far more demand­ ing than might be supposed by someone who has dived recreationally in clear. warm water. Even dry-land excavation calls for dexterity and an ability to rec­ ognize the changes in soil texture and color that mark individual strata. Exer­ cising these same capacities when one is submerged in cold. murky water is pos­ sible for only limited periods of time. Just as one's body grows increasingly numb, so do one's wits lose their quick­ ness. To make even a small mistake is to risk the loss of irreplaceable infor­ mation. The ideal underwater excava­ tor combines physical and mental en­ durance with technical competence and what might be considered a moral quali­ ication: scientiic ethics. The top layer of lake-bottom sedi­ ment at Baigneurs. from 10 centimeters to a meter and a half thick, contains nothing of archaeological interest. This "sterile" sediment. however. has helped CLUSTER OF TIMBERS, some upright and some Hat on the lake the small triangles of ribbon that subdivide the COrner of the lrger botom, appears in this underwater photograph. The horizontal and triangle partly visible here. Analysis of the wod still protected by vertical positions of each timber will be recorded with reference to bark at Baigneurs reveals the time of year when the trees were felled. to preserve the archaeological materi­ als lying under it. Before beginning the painstaking work of excavation the div­ er removes the overlying sediment from his section of the bottom with a suction pump. Then. using only his bare hands. he gently lifts from the exposed surface one thin horizontal section of material after another. Only in this way can he feel any pieces of wood. bone. cloth or pottery the uncompacted debris may contain and take steps to protect them against destruction. If the diver's handling indicates that a horizontal section holds no sizable arti­ facts. he places the material in a bucket for transfer to the sievers on land. If he touches an artifact. he carefully uncov­ ers it and makes a sketch of it in situ before removing it from the section. If the water is suiciently clear and there is enough light. he also makes a photo­ graph of the object in situ. Of the objects we uncovered at Baigneurs we recorded in this manner most of the pottery. all the lint tools except the smallest ones. all the wood objects (such as spoons. combs. ax handles and small planks). animal remains (such as deer antlers) and two lint daggers with their wood handles still in place. When even more fragile materials. such as textiles. were uncovered. the divers were able to avoid disturbing them further by cutting out an entire block of the bottom and raising the block with the material to the surface in one piece. We used the same technique on a larger scale. removing blocks of lake bottom weighing from 10 to 100 kilograms. when we wanted to preserve entire stratigraphic sections or to obtain quantities of material for paleobotani­ cal analysis. Our excavators also took from the lake bottom numerous deep cylindrical "cores" to get samples of pollen for analysis of the ancient plant communities and samples of mollusk shell for analysis of the prevailing an­ cient temperatures. These cores were obtained from a loat by plunging into the bottom a plastic tube eight centime­ ters in diameter. from two to three me­ ters long and heavily weighted at the upper end. The upper half of our two-story exca­ vation difers in several ways from dry­ land archaeology. The kind of sieving we do is conventional, but many of the objects our sievers retrieve are organic. for example wood artifacts and plant materials such as seeds. Submersion has preserved these objects; to let them dry out would be to destroy them. After the objects are numbered. weighed and re­ corded on a chart they must be sealed in plastic bags to keep them wet. Even fragments of pottery. which are normal- ly indestructible. may need to be bathed in polyvinyl acetate to keep them from crumbling when they dry out. hen samples of wood are present an archaeological deposit. as they are at Baigneurs. there is the possi­ bility of establishing the deposit's abso­ lute age. In recent years much progress has been made in correlating actual cal­ endar dates with dates determined by sequences of tree rings and the decay of carbon 14. In the New World the tree­ ring year count has now been reliably extended back beyond 5000 B.C. In the Old World the best such record. the Chronology of Treves. goes back only to about 800 B.C. Tree-ring sequences also exist for earlier periods. such as the Neolithic. but they are not yet linked up with the more recent chronologies. Hence in general the only estimates of age for these "loating" chronologies are carbon-14 ones. If several loating counts can be corre­ lated. however. the absolute chronologi­ cal diferences between them can be ex­ pressed in actual solar years rather than carbon-14 years B.P. (before the pres­ ent). This kind of work is in progress at the Dendrochronology Laboratory of Neuch:hel under the direction of my colleagues Orcel and Lambert. The tree­ ring data from Baigneurs wood are now W in 61 © 1979 SCIENTIFIC AMERICAN, INC FLINT DAGGER, deftly pressure-laked and itted with a handle made of wood, was one of two such wood-hafted artifacts found in the compacted plant litter that covered house loors. being integrated into the expanding tree­ ring chronology for western Switzer­ land, even though the work faces such obstacles as the fact that much of the Swiss material is based on samples of oak and ours is based on samples of ir. The overall result is that we can talk about the Baigneurs settlements with a chronological precision that would have been impossible a few decades ago. Among the samples of wood at Bai­ gneurs are house posts and beams with the bark still on them, which makes pos­ sible even greater precision. The se­ quence of tree rings establishes the year a tree was felled. Bark, when it is pres­ ent, protects the latest ring, a fragile outer growth. Viewed under the micro­ scope, this developing growth provides evidence on whether the tree was felled in winter or summer, spring or fall. Ide­ ally all the house timbers that are more than seven or eight centimeters in diam­ eter should be sampled for their ring se­ quence and all the sequences should be interrelated. We have done this since 1974, and by the end of the 1978 season our 150 square meters of excavation had yielded the material for 8 10 individual tree-ring samples. In addition to the contribution to dat­ ing made by these ancient posts and beams other plant remains provide clues to the life of the Neolithic farmers who cleared the forest here. They felled the ir and ash trees but left the oaks stand­ ing, presumably because they valued the harvest of acorns. They also protected other nut trees, clearing away the near­ by growth so that the sun could reach them. As a result the nuts that were har­ vested from the protected trees were twice the size of those of untended neighbors. Our intensive collection of plant re­ mains has also enabled us to demon­ strate diferent uses of diferent materi­ als. For example, the inhabitants made beds of ir boughs and illed the chinks in their house walls with moss. We can even trace the lines of the house walls by the accumulation of hazelnut shells along the inside of the wall. Analysis of the charcoal from hearths indicates that for irewood the settlers preferred beech and oak. The Neolithic inhabitants of Bai­ gneurs selected as their building site a headland near the outlet of Lake Pala­ dru. The settlement occupied an area of some 1,500 square meters running down to the lakeshore. On the land side the WOOD LOOM COMB, one of several artifacts indicating that the farmers of Baigneurs made their own cloth, s a further example of how objecs that might soon have disintegrated at an open-air site may be preserved in wetland and underwater locations. Once uncovered, how­ ever, wood objects must be kept wet until laboratory techniques for preservation are applied. 62 © 1979 SCIENTIFIC AMERICAN, INC village was enclosed by a small palisade fence that included a gateway. A virtual forest of posts at the center of the enclo­ sure reveals a pattern of adjacent small rectangular houses aligned along nar­ row alleys. The village site was occupied not once but twice. The irst occupation level is separated from the second by a layer of silt that was deposited during a tempo­ rary rise in the level of the lake. The irst settlers arrived on the scene in about 2900 B.C. They were farmers who had apparently lived in a nearby village; the grassy headland with its thick forest of ir trees adjacent evidently caught their fancy. They came in winter, and in preparation for their return they felled several small irs with trunks ranging from 10 to 14 centimeters in diameter. These they stripped of their branches and stacked. T he visitors came back the following winter. They chopped down more ir trees and built their irst house, using these trees and the ones they had felled the preceding year. The house, 12 me­ ters long and four meters wide, was ori­ ented with its long axis running east­ west. It consisted of a single room with a clay hearth in the center and a door in the north wall. The ir-tree trunks formed the frame of the house; they were set in holes from three to four me­ ters deep. Trunks smaller in diameter formed the horizontal beams; they were lashed to the top of the uprights with ropes and vines. The roof covering this sturdy frame was probably thatched with reeds; the walls were made of slen­ der vertical poles, from two to three cen­ timeters in diameter, their butts set in a shallow trench. Although r was chosen for the uprights, many of the wall poles were hazel branches. The chinks be­ tween the poles were illed with moss and reeds. One can assume that the housebuilders burned over the area they had cleared to prepare the ground for cultivation, but there is no way of know­ ing whether they farmed the area the following spring. A second house, identical with the irst, was built the following winter. It was south of the irst house and was sep­ arated from it by an alley only 1.2 me­ ters wide. The door of the house, in the north wall, opened on this alley. Today the prevailing wind at Lake Paladru is from the north; the east-west orientation of the Neolithic house suggests that 5,000 years ago the prevailing wind was from the east. With the building of the second house the Neolithic settlement seems to have been irmly established. It was now destined to be occupied for some 30 years. The numerous artifacts we have brought up from the lake bottom enable us to reconstruct with some conidence the villagers' daily lives. Stone tools are less common than a 30-year occupation would suggest. Made from locally avail­ able nodules of lint, they are for the most part shaped roughly, although a few of them, including blades and scrap­ ers, are elegantly pressure-laked; so are the two lint daggers we recovered with their wood handles intact. Certain of the other lint tools also showed traces of wood handles. Most of the containers that have sur­ vived are clay pots, simple in shape and not too well ired. Bits of woven basket indicate that the villagers did not de­ pend for storage on pottery alone. The fragments of textiles provide evidence that weaving was practiced, as do nee­ dles, many balls of thread, wood 100m combs and wood spindles suitable for the spinning of wool and lax. The varie­ ty of these artifacts makes it clear that the villagers did their own weaving rath­ er than importing cloth. Among the other artifacts made of wood are spoons that could easily be mistaken for the wood spoons of today. The villagers' capability as woodsmen, clearly apparent in the construction of their houses, is further indicated by a number of long ax handles. One of these still held a blade of polished stone. Oth­ er handles were evidently itted with sleeves made of antler to help cushion the shock of chopping. W hat did these people eat? Al­ though they were farmers, they were by no means entirely dependent on YEARS CULTURE RADIO· CARBON SOLAR NEUCHATEL (B.C.) (B.P) 4000 5000 4028 B.C. AUVERNIER 379 3 B.C. 0 SWISS GERMAN 4000 B.C. DANUBE7 2827 B.C. 3500 NEOLITHIC 3018 B.C. AUVERNIER 2776 B.C. 4500 STRONG DEVIATION STRONG DEVIATION 4000 60 3000 2500 3146 B.C. AUVERNIER A 2776 B.C. 2751 B.C. AUVERNIER 2584 B.C. (±50) EARLY BRONZ E AG E 3500 2782 B.C. AUVERNIER B 2498 B.C. (±50) 2000 292 5 B.C. MIDDLE BRONZE AGE DANUBE3/10 1605 B.C. 1500 ,� LATE B ZE 3000 1265 B.C. AUVERNIER 841 B.C. D IRON AGE 500 2000 AVE 0 �---4 --- r --- �� ---- �1 000- -� 2 500 1215 B.C. ZURICH 8 84 B.C. 7 ��HEs A.D. 71 419 B.C. LA TENE 96 B.C. J �----- CHRONOLOGY OF TREVES 717 B.C. (TO A.D. 700 ) . TREE-RING CHRONOLOGIES for Europe extend from about 4000 B.C. to the present but do not form an unbroken sequence, The three longest sequences are the Chronology of Treves, which extends approximately from 750 B.C. to A.D. 700, and wo overlapping Danube chro­ nologies that run from 4000 D.C. to about 1500 B.C. The tree-ring dates at Baigneurs correlate with a Swiss chronology, Auvernier, that etends for some Z50 years starting in about 3000 B.C. For much of the period between 5000 and 1000 B.. age estimats based on carbon 14 are later than tree-ring dates; for example, sampls of 3000 B.C. (some 5,000 years ago) yield carbon-14 dates of about 4400 yeas D.P. (before the present). The erratic line at left shows carbon-14 discrepancy calculated by Hans E. Suess of University of California at San Diego. 63 © 1979 SCIENTIFIC AMERICAN, INC domesticated foodstufs. The animal bones we have recovered indicate that they regularly hunted deer and on occa­ sion even bears. Little lint projectile points suggest they also hunted small game. but the bones of such animals have not survived. The villagers ished in the lake; we have found stone net weights and bits of netting (and a single copper ishhook. even though 3000 B.C. is long before the age of metal began in this part of the world). From their domestic animals. and perhaps those of their neighbors. the villagers supple­ mented their meat diet with beef. mut­ ton. pork and goat meat. From the villagers' domesticated plants-wheat and barley-they milled a coarse lour with which they baked lat cakes; the milling was done with mill­ stones of granite. They also cultivated lax. and although they did so mainly to get iber. they may have prized the oily lax seeds. They gathered numerous wild plant foods. In addition to the acorns and hazelnuts I have mentioned they gathered beechnuts. blackberries. wild plums and apples. They must sure­ ly have baked their cakes on the clay hearths and broiled their meat over open lames. but they also knew the art of stone boiling. For this kind of cook­ ing one partly ills a pot with water and heats a number of stones in the ire until they are almost red-hot. The hot stones and the food are then put in the pot to­ gether and the stones bring the water to a boil. The sudden quenching often shatters the stones. These people used quartzite pebbles as their boiling stones. and their hearths became littered with bits of quartzite. In the ninth winter of their occupation the villagers completely rebuilt the irst house. Nine years later they rebuilt both houses. The house plan remained the same: an oblong with a central hearth. Over those 18 years the level of Lake Paladru may have luctuated slightly in response to variations in the climate and on occasion could even have wet the of wood house at a shoreline site is only 15 years. At the same time our tree-ring records indicate that the disaster came sometime after the 19th year of the set­ tlement; we have recovered a house post with growth rings extending to that year. evidently a late addition to one of the houses. The character of the plant community that reinvaded the deserted settlement was diferent from that of the primeval stand of ir. The forest was opener to sunlight; among the irs grew alders. elms and ashes. As the forest advanced the lake level rose. Eventually the water stood perhaps 1. 5 meters above its for­ mer level, and the burned ruins were covered by a thin layer of silt. The silt layer was then colonized by beds of reeds growing along the new shoreline. loors of the houses. There was no major looding; the settlement was eventually destroyed not by water but by ire. We cannot tell exactly when the hous­ es burned down. The ire could scarcely have been later than some 30 years after the founding of the settlement. By that time there would have been little left to burn: the maximum lifetime of this kind ome 60 years after the founding of the irst settlement. at a time when Lake Paladru had dropped back to its earlier level. a second Neolithic farming group settled in exactly the same place near the lake outlet. The style of their pottery and their lint artifacts was the same as that of the earlier group; the newcomers may even have been descendants of the original settlers. They did. however. build houses that were smaller and more nearly square. Surviving clay hearths were used by the newcomers. but in gen­ eral their houses on the headland were more haphazardly located. If there is any diference between the two groups other than the size of their houses. it is that the early settlers loored their hous­ es with r boughs and ferns and their successors sometimes loored them with bark. When it came to means of sub­ sistence and way of life. the settlements were the same. What subdivision of European Neo­ lithic culture is represented at Bai­ gneurs? Studies of the pottery and the lint artifacts indicate that these lake­ shore people were members of a widely distributed population whose cultural tradition lourished in western Switzer­ land. in the Jura Mountains. in the val­ ley of the Saone. in the Savoy and in northern Dauphine. To European pre­ historians the culture is known as the Saone-Rhone civilization. The western members of this population had close relations with the Neolithic farmers to the south in France; those closer to Switzerland and within it had similar ties with the Neolithic farmers of cen­ tral Europe. Carbon- 14 analysis places the Baigneurs settlements in an early Saone-Rhone phase. dated at about 2900 B.C. Established by chance at the end of Lake Paladru. the two settle­ ments spanned some 90 years before the second settlement was abandoned for no apparent reason. Thereafter a rise in water level transformed the lakeshore into lake bottom and the ruins of Bai­ gneurs remained untroubled for ive millenniums. STRATIGRAPHIC SECTION of lake botom at Baigneus sbows, from top to botom, an upper layer of sedimens that provd to be sterile, or entirely lacking in artifacs (light area), a layer containing the debris of the second Neolithic occupation (thick dark layer), a second ster­ (light), a very (dark) and inally the sterile lake botom below it. ile layer of sediments separating tbe second Neolithic occupation from tbe st thin irst stratum of occupational debris Wedged between the lake bottom and the irst stratum is a lat milling stone made of granite. 64 © 1979 SCIENTIFIC AMERICAN, INC S Out of photographic technology but not at all photographic KODAK EKTACHEM Clinical Chemistry Slides. Actually, nobody handles them or looks at them. At several hospitals in recent months, simpler the manipulation outside the laboratory tests for blood glucose factory. So it is to be in clinical chem­ (GLU) and blood urea nitrogen (BUN) istry, we intend. To wit: have been done by a new method based on little squares of ilm, 16 mm x 16 mm, in plastic mounts as though for a projector. But they bear no images. / � serum drop � ! ,,",- In health care, society can ill aford novelty for the sake of novelty. To justi­ fy our massive multidisciplinary efort, we must irst tell you that the popular image of the clinical laboratory, source quantitatively thinking physicians do their thinking, is out of date. The bub­ work can't handle the volume of routine GLUCOSE ASSAY tests routinely ordered. Nowadays little test tubes move through very complex bubbled, scanned, etc. in mechanical imitation of the chemists who were reaching reagent layer. REAGENT LAYER Contains 1. glucose oxidase, enzyme that catalyzes oxidation of glucose to gluconic acid, releasing H202; 2. peroxidase, enzyme that catalyzes coupling, in pro- tions bling glassware representing science at machines as liquids are added, with­ LAYER whitened with Ti02 Precision metering by isotropic porosity instead of fancy plumbing. A variation of 10% in drop volume would make only 1% difference in serum per unit area portion to HP2 present, of 3. 7-hydroxy-1-naphthol to 4. 4-aminoantipyrine to make dye; 5. buffer to maintain pH near optimum for both reac­ of the numbers with which today's drawn, stirred, iltered, heated, cooled, 80% air SPREADING � .c," � "mn /; � called in when the doctors stopped re­ lying on just the gross appearance of TRANSPARENT SUPPORT dye to be formed will be reference through which density of measured against white SPREADING LAYER Same as for GLU REAGENT LAYER Loaded with urease, enzyme which releases N H3 and NH4 + from urea. Buffered to pH 8 to push equilibrium toward NH4. BARRIER LAYER Cellulose acetate butyrate passes only nonionic sub­ stances, bars buffer and OH- which would cause dye to form if they penetrated. the patient and the patient's contents. Our approach restores simplicity. No INDICATOR LAYER pipes, no squirts. The only liquid is the serum under assay. It works because we know how to coat layer upon tissue­ . thin layer of compositions within which chemical and physical events and inter­ BUN ASSAY actions can be controlled with a nicety "" that seemed preposterous when irst Cellulose acetate carrying an indicator that gives ceior weakly with NH3 and not at all with N H4. (Otherwise the range of color density would be too wide for convenient reading because the corresponding BUN range of medical interest-4 to 130 mg%-is so wide.) TRANSPARENT SUPPORT through which dye density will be measured proposed for color photography long ago. In those days color photography For answers to the kind of questions Now wait till you hear about the little meant three-in-one cameras and ini­ that are prompted by more than just chip of non photographic ilm for assay­ nite patience. Photography has become passing curiosity, write to R. Barnes ing triglyceride in serum. It contains what it is today because the more that's Parsons, packed into the layers of the ilm, the 14650. Kodak, Rochester, N .Y. ATP, the storage stuf for biological energy, and four enzymes, including peroxidase from horseradish and an oxidase we prepare from The KODA. EKTACHEM GLU/ BUN Analyzer is our own version of an instrument which accepts undiluted serum samples and blank K K EKTACHEM Clinical Chemistry Slides, measures concentration of GLU or BUN or both, and accumulates the used slides for disposal. It receives push-button instructions to set itself for GLU or BUN, applies a 10-! drop to the appropriate type of slide with no carryover from one sample to the next, incubates for the proper number of minutes at 37°C, and uses the Ti02impregnated spreading layer of the slide as a background for a photomultiplier to read relection density through a narrow-band ilter at 540 nm for GLU or 670 nm for BUN . . The precision required of that relectometer unit was the kind of challenge to which optical engineers in the color photography business have become accustomed. The microprocessor ii the unit even corrects for nonlinearity of relection density with dye concentration! ODA © 1979 SCIENTIFIC AMERICAN, INC faecium. Streptococcus Or our ilm for amylase assay. Or the one for bilirubin. We could go on and on. In fact, we have. © Eastman Kodak Company, 1979 © 1979 SCIENTIFIC AMERICAN, INC