C-14 Dates for Sites in the Mediterranean Area Author(s)- Ellen L. Kohler and Elizabeth K. Ralph

8/3/2019 C-14 Dates for Sites in the Mediterranean Area Author(s)- Ellen L. Kohler and Elizabeth K. Ralph

1/12

C-14 Dates for Sites in the Mediterranean AreaAuthor(s): Ellen L. Kohler and Elizabeth K. RalphSource: American Journal of Archaeology, Vol. 65, No. 4 (Oct., 1961), pp. 357-367Published by: Archaeological Institute of AmericaStable URL: http://www.jstor.org/stable/502212 .

Accessed: 16/06/2011 08:33

Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at .

http://www.jstor.org/page/info/about/policies/terms.jsp. JSTOR's Terms and Conditions of Use provides, in part, that unlessyou have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you

may use content in the JSTOR archive only for your personal, non-commercial use.

Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at .http://www.jstor.org/action/showPublisher?publisherCode=aia. .

Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed

page of such transmission.

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of

content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms

of scholarship. For more information about JSTOR, please contact [email protected].

Archaeological Institute of America is collaborating with JSTOR to digitize, preserve and extend access to

American Journal of Archaeology.

http://www.jstor.org
http://www.jstor.org/action/showPublisher?publisherCode=aiahttp://www.jstor.org/stable/502212?origin=JSTOR-pdfhttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/action/showPublisher?publisherCode=aiahttp://www.jstor.org/action/showPublisher?publisherCode=aiahttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/stable/502212?origin=JSTOR-pdfhttp://www.jstor.org/action/showPublisher?publisherCode=aia


2/12

C - 1 4 D a t e s f o r S i t e s in t h e Mediterraneanr e aELLEN L. KOHLER ANDELIZABETHK. RALPH

The C-14laboratory t the Universityof Penn-sylvaniawas initiatedby FroelichRaineyin 1951,andis sponsored ointlyby the UniversityMuseumand the Departmentof Physics.From the start,datingeffortshave been devoted o the determina-tion of seriesof C-I4 dates for five regionsof theworld-namely,theMediterranean,earEast,Cen-tralAmerica,SouthAmerica,and the Arctic.Re-sults which are of special nterestto prehistoriansandhistorians f the Mediterraneanreaarebroughttogether n the accompanyingists. This Mediter-raneandatingprogramwas made possibleby G.RogerEdwardswho obtained amples romtheex-cavators.The emphasishas been to date seriesrather hanisolated amples n orderto minimizethe inclusionof errorsfrom non-representativeamples.Theseerrors,which we shall term "physical," re apartfromC-I4methodological roblems, ndfrequentlypresent he greatesthazard n the determinationfan accurateC-I4 date, one which representsanarchaeologicalvent.A briefexplanation f the C-I4 dating principlemayhelpto clarify his"physical" roblem.NaturalC' (written as C-14 for convenience),which ismanufacturedn theupperatmospherey thebom-bardment f N" atomswith neutrons componentsof cosmicrays),formsa very smallbut essentiallyconstantpartof ouratmosphericarbondioxide.Allliving organisms uch as treesand grassescontainthis equilibriumamountof C-I4, but when deadcease to take in C-I4, so that the radioactivedis-integration ate (as dictatedby the half-life) thenbecomesparamount nddetermineshe decreasenthe amountof C-14 n the deadmaterial.This pro-videsthe meansfor age determination. rom theseconsiderationst is apparent hat only the outergrowth ayersof a treewill representhetime of the

cuttingof the treeandusually he date of construc-tion of a building; he innerlayersof a tree areal-ready depleted n C-I4.This point has been em-phasizedherebecauseorsomeof the dates n theseseries, rrorsof ioo or moreyears(too early) fromthispossiblepitfallbecome mportant.For thisrea-son, too,samplesof grains,cloth,hide,etc. are dealbecause heywereused at the time thattheyceasedto incorporateC-i4.'Anotherphysicalproblem,also one of the truecontemporaneityf the sample, s thatof so calledhumic contamination.This may occur from theadsorptionfyounger inelydivided, lightly olubleorganicmatterby the sample,andis mostlikelytobe suspectedwhen a samplehas been exposedtocontinualwashing by ground waters.It has notbeen found to be a seriouscontaminant xceptinvery old age ranges,that is, tens of thousandsofyears.Sincesomeof thesampleson these ists,how-ever, were composedof finely dividedcharcoal,agood adsorber,his possible ntrusionwas checkedwheresamplesizeswereadequate.Samples reatedforremovalof humiccontaminant re labeled"pre-treatedwith NaOH."A considerationquallyvital in theseseriesanda possiblecause of errorfalls in the categoryofmethodological roblems.Since the productionofnaturalC-I4 is dependentupon the numbersofneutronspresentin the upper atmosphere,t isnecessary o assume that the cosmicray intensityhas been constantduring past eras, and that theequilibriumbetween atmosphere, ceans,etc. hasremained teady.Libbydemonstratedn I9492 hatthisassumptions correctwithin io%,butwith theincreasedprecisionof the technique, hereare in-dications3hatthe C-I4tontent of the atmospheremay have differed n some past time-intervals yI to 2 orpossibly %and/or thatthepreviously c-1 Informationconcerning sample requirements is available inthe C-I4 laboratoryat the University of Pennsylvania.2 Libby, W. F., RadiocarbonDating (2nd ed., Chicago 1955)

o10.3Ralph,E. K. and Stuckenrath,R., "Carbon-14Measurementsof Known Age Samples," Nature i88 (Oct. 15, 1960) 185-87.Willis, E. H., Tauber, H., and Miinnich, K. O., "Variationsin the Atmospheric RadiocarbonConcentration over the Past

1300 Years," American Journal of Science Radiocarbon Sup-plement 2 (1960) 1-4-Broecker, W. S., Olson, E. A., and Bird, J., "RadiocarbonMeasurementson Samples of Known Age," Nature I83 (June6, 1959) 1582-84.Vries, HI. de, "Variations in Concentrationof Radiocarbonwith Time and Location on Earth,"Proc. Kon. Ned. Akad. vanWetenschappen,B 61 (1958) I-9.


3/12

358 ELLEN L. KOHLER AND ELIZABETH K. RALPH [AJA 65cepted half-life value (5568+?30 years)4 is not thetrue one. A brief summary of this situation is asfollows:

A.D. 1850 to 500 B.C. Indications of relativelyshort-term variations, of approximately ioo years'duration,not greaterthan 2% in C-I4 contents (ap-proximately 200 years when converted to ages).These are predominantly in the upward directionwhen compared with the Libby (5568- 30 years)half-life.500 B.C. to 2000 B.C. General increasing trend

of C-i4 contents (going back in time) startingwithless than 0.5% to a maximum of approximately4%. Maximum error in the region of 2000 B.c. isapproximately200 years.2000 B.C. to 4000 B.C. Possibility of errorsof theorder of 500 years in the younger direction. Thislarger discrepancyis based upon comparisons withthe Egyptian chronology.The questions of changes in the C-I4 inventoryin the atmosphere and the accuracyof the half-lifedeterminationare interrelated.Short term variationsare due to the former, Ibutsince these are almostall in the upward direction, that is, younger (asdetermined by C-I4 measurements of samples ofknown ages), they are minimized if a higher half-life value is assumed. This reasoning is now sup-

ported by the new measurement of the half-life byW. B. Mann and W. F. Marlow5 at the NationalBureau of Standards with a value of 5760 +70years.However, becauseof the dual nature of this prob-lem and the fact that two other careful determina-tions of the half-life are in process, the ages in un-derlined type in this article have been calculatedwith an estimated "effective"half-life value of 5800years. This value has been obtained from the C-I4measurements of 18 samples (plus standard cali-bration samples) of known age6 which range inage from A.D.I00 to 1900 B.C. Since 1961 is an in-terim period in regard to the solution of these me-thodological problems, and our known age data areincomplete, we wish to emphasize that this is ourpresent best estimate of a reasonable value, butthat it is in closer agreement with the true ages ofwood samples and is supported additionally by the

new half-lifedetermination erformedby the Na-tional Bureauof Standards.For the sakeof con-formitywith other C-I4 laboratoriesnd publica-tions in the AmericanJournalof ScienceRadio-carbon upplementn i96i, thedatescalculatedwiththeLibbyhalf-lifevalue(5568years)are istedalso.Since it is difficult o evaluatethe accuracyofthis estimated5800half-lifeat this time, the half-life errorhas not been includedwith the statisticalcountingerrorsquotedwith the C-I4 dates.It isobvious,however,romthediscrepancyetween hetwo values,that a largesourceof errormay haveexisteddueto this half-lifeuncertainty.n addition,the possiblevariationsn cosmicrayintensityhavenot yetbeendetermined ccuratelynoughfor cor-rections o be made.Therefore,t mustbe kept inmind that the C-I4 datesare subject o these twomethodological in additionto statistical)errorswhichcould,butprobably o not,addupto a maxi-mum of approximately % (of the B.P., beforepresent,ages). It seemsappropriate,owever,forour laboratoryo make this realistic hangeto the5800half-life(rather han 5568) at this time sincethe C-I4 dateswhich we publishedpriorto 1959were based upon a similarseriesof known agemeasurements,hat s,an "effective"alf-lifewhich,within the greateruncertaintyof C-I4 measure-ments of severalyearsago, is in agreementwithourpresentvalue.We have illustrated his methodological orrec-tion graphically or the Lerna series in ill. i inwhich C-I4 dates, calculatedwith both half-lifevalues,areplotted.In the precedingparagraphs,arioussourcesoferror(for whichone mustbe cautionedconcerningtheinterpretationf dates)havebeendiscussed.Asis apparent rom the specificcomparisons f con-temporaneousamples,many of the "physical"r-rors,especially umicacidcontamination,renegli-gible in the Mediterraneaneries.Therestill exist,of course, he statistical ncertainties-the tandardstatisticaldeviationsquoted with each C-I4 date.The latteraredue to the randomnatureof radio-activedecayand areunavoidable. hey havebeenreduced, however, by counting each sample twoor more times and, in some cases,by counting dupli-cate portions of samples. Only the averages of the

4 Libby, W. F., op.cit. 36.5 Marlow, W. F., private communication (Jan. io, 1961).6 Thirteen of these samplesare describedby Ralph and Stuck-enrath, op.cit. Five consisted of selected sequoia tree-ring dated

sections, the dates of which have not yet been published.Ralph wishes to acknowledge with gratitude the financial sup-port of the National Science Foundation (grant G-14094) forour known-age dating program.


4/12

1961] C-14 DATES FOR SITES IN THE MEDITERRANEAN AREA 359several counts appear in the lists. The standardstatistical deviation (frequently called one sigma)must also be used with caution, for it indicates thatit is probableonly two times out of three that thetrue date lies within the range of the error quoted.If doubled (two sigma), then there is only onechance in twenty-two that the true value lies out-side this range.The temptation to extractindividualdates from a seriesmay be avoided if these probabil-ities are taken into consideration. For example,among three dates in a chronological series, oneseems not to fit. If it is realized, however, that inthree chances, one would be expected to fall outsidethe one sigma tolerance, then this apparent dis-crepancymay correctlybe attributed to chance.The samples in these lists were counted by thepure carbon dioxide proportionalgas counting tech-nique with equipment similar to that described byG. J. Fergusson.7All samples were pretreated withhydrochloric acid for removal of inorganic carboncompounds. Some, as noted, were also pretreated

with sodium hydroxide for removal of possiblehumic contaminants. Corrections for variations innatural fractionation as determined by C"/C' ra-tios have not been made, and are believed to beunimportant.

The sites are presented in geographicalorder, ap-proximately south to north through Anatolia, andnorth to south through Greece, ending with Chios.It so happens that this preserves an approximatelychronological order within the two major groups.The sites dated are as follows:

Anatolia:HacilarBeycesultanGordion

Greece:EutresisLernaPylosChios

HACILARThe British Institute of Archaeology at Ankara began excavations in 1957 on one of the habitationmounds near the village of Hacilar, 25 km. west of Burdur, in southwestern Turkey. James Mellaart,the Assistant Director of the Institute, was field director. As dated by ceramic and other culturalparallels,Levels I (at top) to V areconsideredto be Early Chalcolithic,and Levels VI to IX (at bottom),

Late Neolithic. The samples were collected by Mr. Mellaart in 1958 and 1959. Preliminary excavationreports have appeared annually in Anatolian Studies beginning with Volume VIII (1958).In the Hacilar series the portions of samples labeled "A" were pretreated with NaOH for removalof possible humic acid contamination. The difference between C-4 dates of portions was significantonly for sample P-3I3. It may, therefore, be assumed that humic acid contamination, if present, was ofslight degree. For samples P-314and P-3I5, the dates listed are the averagesof the two portions.Christian Calendar AgeCalculated with Half-life:*

5568 5800P-314. Level IX, Late Neolithic 5393- 92 B.C. 5687? 96 B.C.Charcoal fragments in rubbish deposit. The dates of the in-dividual portions, calculated with 5568half-life, are as follows:

P-3I4 5298 ? 143 B.C.P-3I4-A 5487-- 115 B.C.P-3I3-A. Level VI, Late Neolithic 5399 - 79 B.c. 5693 ?_ 82 B.c.Charcoal from a hearth in use before a fire. The date of por-tion P-313, calculated with half-life 5568, is 5196 + 93 B.c.

7 Fergusson, G. J., "RadiocarbonDating System," Nucleonics13 (Jan. 1955) 18-23.* Please note that errorsquoted for C-14 dates are statisticaldeviations only and do not include the uncertainty of eitherhalf-life value.


5/12

360 ELLEN L. KOHLER AND ELIZABETH K. RALPH [AJA 65Christian Calendar AgeCalculated with Half-life:*

5568 5800P-3I6-A. Level II, Early Chalcolithic 5219 ? 131B.c. 5506 ?_ 136 B.C.Charcoalfrom roof beam in Early Chalcolithic settlement, areaN, room 4. Since the excavations of 1958,it has become clearthat there are two phases of Level II. This sample would thendate the construction of the later phase (IIb). The first portionof this sample (P-316) was not processed.P-315. Level Ia, Early Chalcolithic 5037 - 119 B.C. 5317 ? 124 B.C.Charcoalfrom a roof beam in Early Chalcolithicfortress,room

5. Roof beam representsconstruction date of the fortress. Thedates of the individual portions, calculated with 5568 half-life,are as follows:P-3i5 5097 ? 22I B.C.P-3i5-A 4976_ 90 B.C.

BEYCESULTANBeycesultan is located in the upper Maeander River Valley, 5 km. south of the modern village ofgivril, in southwestern Turkey. Excavations, begun in 1954, were sponsored by the British Instituteof Archaeology at Ankara; Seton Lloyd was the director and James Mellaart,his assistant.Samples werecollected by Mr. Mellaart during the 1958season. Cf. AnatSt 8 (1958) 93-125and 9 (i959) 35-50.

P-297.8Level XXVIII 2740 ? 54 B.c. 2928 L 56 B.C.Lump of charcoalfrom sounding "SX,"referableto "ca.middleof Late Chalcolithic sequence."P-298. Level XXXVI 3014 - 50 B.c. 3213 -- 52 B.C.Charcoal, from fifth building level above virgin soil.GORDION

Since 1950 the University Museum, Philadelphia, has been excavating at Yassihiiyiik-Gordion, thesite of the capital of the Phrygian Empire. It lies on the Sangarius (modern Sakarya) River, ca. 70 milessouthwest of Ankara. The City Mound contains strata dating from the Chalcolithic to the Galatianperiods. Burials were in earth tumuli which date from the earliest (so far opened), ca. 740 B.c., andcontinue down to Hellenistic. There is also a cemetery of earth burials and cremations of Hittite andlater times under and near Tumulus H. Samples were collected in i95o, 1951, 1955, 1957 and 1959 byRodney S. Young, the Field Director. Interim campaign reports by him or G. Roger Edwards, AssistantField Director, have appearedin University Museum Bulletins 16-1 (I95I) and 17-4 (1952) and in AJAeach year since 1955-Interval between Hittite and Phrygian on the City MoundP-I37. Level "PhrygianIVa" in Deep Cut, Trench NCT II55 ? 125 B.c. I280 ? I30 B.C.Charcoalsample from a fallen beam lying on the floor of LevelIVa. Estimated age: ioth-9th c. B.c.

* Please note that errors quoted for C-14 dates are statisticaldeviations only and do not include the uncertainty of eitherhalf-life value.8 Samples P-297 and P-298 were both divided into portions;

one of which was pretreatedwith NaOH. Since the differencebetween the two resultswas statisticallynot significant,the dateslisted are the averagesof the two portions.


6/12

1961] C-14 DATES FOR SITES IN THE MEDITERRANEAN AREA 361Christian Calendar AgeCalculated with Half-life:*

Pre-CimmerianPhrygian Tumuli 5568 5800P-363. Tumulus W' 766 ?- 45 B.c. 875 ? 47 B.C.Roof beam fragments: wood slivers and strips (large piecesshowed slight signs of rot on edges, considerabledust present,either ashy material or rotted wood). Estimated age: 740-700

B.C.P-I27. Tumulus MM (Royal Tomb)"o 743 - 90 B.c. 85 -- 94 B.C.

Juniperus drupaceafrom log of outer wall of tomb. Estimatedage: 725-700B.c. Recent examinations of the tree-ring patternof this log have revealed that some shaping of the log was per-formed which may have removed approximately ioo growthrings.

P-275. Tumulus MM (Royal Tomb) 612 ? 52 B.C. 715 ? 54 B.C.Acer pseudoplatinus, part of the top of a 3-legged plain tablefound collapsed on floor of the tomb.

P-128. Tumulus MM (Royal Tomb) 673 ? 90 B.C. 778 ? 94 B.C.Textiles from coverlet on bed, NW corner of tomb.P-i34. Tumulus MM (Royal Tomb) 648 ?117 B.c. 752 ? 122 B.C.

Greasy substance (food?) contained in bowl #148 in tomb.Pre-CimmerianPhrygian Habitation Level on City Mound"P-I35. Megaron 312 665 119 B.c. 770 ? 124 B.C.Charredwood from outer surfaceof roof-beam. Estimated age:

700 ? 25 B.C.P-I36. Clay Cut, Burnt Phrygian House, North Room 732 ? 120 B.C. 840 ? 125 B.C.Charcoal. Estimated age: 700 ? 25 B.C.P-99. Clay Cut, Burnt Phrygian House, North Room 713 114B.c. 820 _ 1i9 B.c.Charcoallumps contained in red-polished pitcher on floor. Es-timated age: 700 + 25 B.C.Post-Cimmerian Phrygian TumulusP-221. Tumulus S, Chamber 710 ? 54 B.c. 817 ? 56 B.C.

Wood, reddish in color, very light in weight. Estimated age:700-600B.C.

Iron-Age Burial, Date UnknownP-352.Tumulus H Cemetery, Burial at U 436 ? 53 B.c. 532 ? 64 B.C.

Wood, red, crumbling; no pots or other finds for dating. Es-timated age: 750-500B.c.?Burials of Persian PeriodP-222. Tumulus H Cemetery. Burial A-I6I 527 ? 38 B.c. 626 ? 40 B.C.Charcoal and wood chips. Estimated age: 550 ? 25 B.c.

* Please note that errors quoted for C-14 dates are statisticaldeviations only and do not include the uncertainty of eitherhalf-life value.9 Young, AJA 64 (1960) 227-32 and pl. 55. This sample mayhave been from a squared beam. See discussion of inner wood

problem at the end of this section.10 Young, AIA 62 (1958) 147-54 and pls. 23-27.11 Cf. plan, AJA 64 (1960) pl. 57, fig. 16.12AJA 64 (1960) 237-40.


7/12

362 ELLEN L. KOHLER AND ELIZABETH K. RALPH [AJA 65Christian Calendar AgeCalculated with Half-life:*

5568 5800P-3o. Tumulus F, Cremation 470 ? 51 B.c. 567 ? 53 B.C.Charcoallumps from grave. Estimated age: 525 ? 25 B.c.P-256.Tumulus I, Cremation 563 ? 53 B.c. 664 ? 63 B.C.Charred wood (2 pieces of bark and small twigs or branches)from pyre. Estimated age: 525 _ 25 B.c.Hellenistic-Galatian Habitation Level on City MoundP-257. Trench NCT-W, House A, Room I 172 ? 51 B.c. 257 - 53 B.C.Large lumps of clean charcoal (sample taken from outer io

rings of piece showing 40 to 45 growth rings from center). Es-timated age: 200 -- 25 B.C.Several samples of wood from Gordion yielded unexpectedly old dates; these, upon analysis of the

ring structuresof some of the samples, could be inner wood and therefore may not date the buildingof the structures from which they came. They are arranged here in chronological order according tothe excavator'sestimates of their dates. A general commentary follows the list.P-I33 came from Tumulus MM which contained the tomb datable 740-700B.c., and was wood froma squaredbeam of Pinus nigra pallasianaused for the inner wall of the tomb chamber. The sample wastaken from the outer surface of the beam, but it is suspected that these trees grew large enough fora hundred years or so of growth rings to have been removed in the shaping of this squared beam bythe builders of the tomb. The C-14 date obtained was (981 + 122 B.c.) 1099 - 127 B.c.13P-358,from Tumulus N, was a sample from among the beams lining the burial chamber. The tombhas a possibly long range of date, as the contents were not very distinctive, but should fall between 700

and 6o00B.c.The wood was reddish and in crumbling condition. The C-I4 date obtained was (937 ? 81B.c.) 1053 - 84 B.C.P-220was collected in 1951from inside Tumulus J (datable approximately 650 ? 50 B.c.) and couldhave come from either the floor or a collapsedroof beam of the chamber. The wood was dry and crumbledinto chips, and a few pieces were slightly charred; several rootlets were removed. The C-I4 date obtainedwas (800 ? 39 B.c.) 910 41 B.C.P-356 was a sample of crumbling red wood from the chamber of Tumulus S2 with estimated date of650-550B.C.according to R. S. Young. The C-I4 result was (970 - 33 B.c.) io87 - 34 B.c.From the Persian level on the City Mound three samples, P-353, P-223, and P-218, were submitted.All were from buildings which must have been constructed in the interval, 540-500B.c. P-353 consisted

of three large pieces from the burned threshold of Building A, which was constructed ca. 540 B.c.It yielded a C-14 date of (962 - 43 B.C.) 1079 ? 45 B.c. P-223 came from the Persian Gateway, andconsistedof powdery red wood from inside the "leaning tower" of this construction. The C-I4 date was(882 + 34 B.c.) 996 ? 35 B.c. P-218 was a piece of wood from the fallen doorpost in the Gateway andconsisted of a portion of a log, partiallypowdery. The C-I4 date was (946 ?f 41 B.c.) Io62 ? 43 B.c.For most of the above samples, several portions were processed,and for all except P-358, the portionswere counted several times. This additional processing and extra care in counting was performed beforeit was realized that the growth rings in these woods were exceptionally narrow. Since these samples

* Please note that errors quoted for C-I 4 dates are statisticaldeviations only and do not include the uncertainty of eitherhalf-life value.13For the non-representativesamples from Gordion, C-14

ages calculatedwith the 5568 half-life are shown in parentheses;those which are calculated with the half-life of 58oo years areunderlined.


8/12

1961] C-14 DATES FOR SITES IN THE MEDITERRANEAN AREA 363consisted of crumbling dry wood or of charcoal of indeterminate structure, cross sections of Juniperusdrupacea were examined. Two large sections with diameters of approximately 40 cm. from TumulusMM contained, at least, 375 and 620 growth rings respectively. One small section of 15 cm. diameterfrom the rubble which supported the Persian Gate construction contained 140 growth rings. (Fordendrochronologicalpurposes,Juniperusis not ideal,but these sections were the only ones availablehere.)It has not yet been demonstrated that the rings represent annual growth. However, the erroneouslyold C-14dates indicate that this assumptionis plausible.On the basis of C-14dates reportedfor TumulusMM, it seems reasonable to assume that these well-preserved sections, which were examined, wererepresentativeof the wood which was used at Gordion during this time interval. For example, the C-I4dates for the Juniperus drupacea (P-127) sample which was an outer semi-unfinished log, and fortextiles (P-128), and for food (P-I34) are in reasonable agreement with the estimated archaeologicaldate of 725 B.c., whereas that for a squared beam of the inner wall (which consisted of Pinus nigrapallasiana, P-I33) was (981 ?+ 122 B.C.) IO99 I127B.c. Examinations of the construction of this tomb14have revealed that it was 'builtwithin a short period of time; that is, the outer growth layers, if present,of the inner and outer walls should ibe contemporaneous.It therefore seems reasonable to conclude that the C-I4 dates for these eight samples are not repre-sentative of the dates of constructionof the structures.The samples consisted of the fragments of remain-ing wood which antedated the cutting of the trees by several hundred years. These botanical observa-tions, in conjunction with the tree-ring analyses, may contribute useful information in regard to thechanges that took place at this site. The patterns of slow growth indicate that the forests at Gordionmay alreadyhave been marginal in Phrygian times, and might have died out naturally even if they hadnot been cut extensively. Vitruvius' comments (2.1) are particularly ' propos. He stated that "... thePhrygians, who live in an open country, have no forests and consequently lack timber."EUTRESIS

Eutresis is located in Boeotia, southwest of Thebes. The excavation, supplementary to that of MissHetty Goldman (1924-1927),was conducted by J.L. and Elizabeth Caskey for the American School ofClassicalStudies at Athens in September, 1958.The report was published in Hesperia 29 (1960) 126-67.For specific comments on samples P-307 and P-3o6, see p. 164. Samples were collected from stratifieddeposits.Christian Calendar AgeCalculated with Half-life:*

5568 5800P-307. Early Helladic I (Eutresis, Group III) 2492 ? 57 B.c. 2670 - 59 B.C.Charcoal from habitation deposits, Trench A, West, 1.17 m.above datum. Pretreatedwith NaOH.P-3o6. Early Helladic I (Eutresis, Group IV) 2496 - 69 B.C. 2673 + 72 B.C.Charcoal from habitation deposits, ca. 1.50om. above datum.Pretreatedwith NaOH.P-317. Early Helladic II (Eutresis, Group VIII) 2262 ? 56 B.c. 2431 ? 58 B.c.Charcoalfrom floor deposit in House L, Room III, ca. 2.50 m.above datum.LERNA

Lerna is situated on the western shore of the Argolic Gulf, opposite Nauplia. The site is beside theLernaean Spring, at the modern village of Myloi. It was occupied in the Neolithic and Bronze Ages,* Please note that errors quoted for C-14 dates are statisticaldeviations only and do not include the uncertainty of eitherhalf-life value.

14See note to supra.


9/12

364 ELLEN L. KOHLER AND ELIZABETH K. RALPH [AJA 652400B.C.

2300-B.C. p-316z2200 -

200-Houoe of T iB.C. "theTile o -B.C. N900 \Bothroso

B.C. RoomT D

000I N 4(L-3k -1 9 0 0 ft

1600 -

B.C.

60 K o 4 ---

1500B.C.

>,N \ ,', \ . cN ,w 3J j1700- N

ILL.I. Carbon-14 dates for Lerna samples. The stepped horizontal line represents the lower limits of Caskey'sarchaeological dating, based upon his excavational evidence, and his overall Lerna periods are indicated on theright of the figure. The C-i4 dates are represented by ovals with vertical bar extensions to delineate the 'limitsof the standard statistical deviations. The numbered ovals and heavy bar extensions are the C-i4 dates calcu-lated with our estimated "effective" half-life value of 5800 years. The small dots 'with thin line extensionsrepresent the same samples with ages calculated with the Libby half-life value of 5568 years.


10/12

1961] C-14 DATES FOR SITES IN THE MEDITERRANEAN AREA 365and, to some extent, in later times. Excavations 'were conducted by the American School of ClassicalStudies at Athens in annual campaigns, 1952-1958,under the direction of J. L. Caskey. Samples werecollected from levels assigned on evidence of stratigraphy to the Early and Middle Bronze Ages.Samples from the Neolithic levels proved to be too small for analysis. Preliminary reports on the exca-vations have been published in Hesperia 23 (i954) and 28 (i959). Christian Calendar AgeCalculated with Half-life:*

5568 5800P-321.Late Early Helladic II (Lerna Period III) 1946 ? 58 B.c. 2102 + 60 B.C.Charcoal from floorof burnt room, CA, in Trench GK; 4.70m.above sea level; precedes the House of Tiles.P-321-A Same 2031 ? 64 B.c. 2191 + 67 B.c.Pretreated with NaOH.1'P-312.Late Early Helladic II (Last Phase, Lerna Period III) 1887- 65 B.c. 2041 - 68 B.c.Charcoal from burnt debris in Corridor IV of the House ofthe Tiles.P-318.Late Early Helladic II (Last Phase, Lerna Period III) 2120 - 65 B.C. 2283 ? 68 B.C.Charcoal from burnt debris (presumably woodwork) in Cor-ridor IV near Door E, House of the Tiles.P-319.Late Early Helladic II (Last Phase, Lerna Period III) 2027 ? 59 B.c. 2186 -+-61 B.C.Charcoal from burnt debris near east end of Corridor IV,House of the Tiles (with slight possibility of contaminationfrom a rubbish pit of Lerna Period IV).P-320.Late Early Helladic II or Early Early Helladic III 1978 ? 58 B.c. 2136 ?+ 60 B.C.

(Late Phase,Lerna Period III or EarlyPhase,LernaPeriod IV)Charcoal from Area BB, burnt debris over northwest part ofthe House of the Tiles, about i m. above the floor. Probablycontemporary with the building, but possibly affected by ac-tivity in the first phase of Period IV.P-3oo.Early Early Helladic III (Early Phase, Lerna Period IV) 1919+ 53 B.c. 2073 + 55 B.C.Charcoal from Area BC, west of the House of the Tiles, justabove the burnt debrisof the building. Pretreatedwith NaOH.P-299. Early Helladic III (Third Building Phase, Lerna Period IV) I803 - 93 B.c. 1952 + 97 B.C.Charcoalfrom Area BH, a rubbishpit (Bothros 8), containingrefuse and broken pottery. Pretreated with NaOH.P-303. Middle Helladic (First Phase, Lerna Period V) 1568 + 51 B.c. 1708 + 53 B.C.Burnt material in large storage jar in Room 45, Area BE.P-303-A. Same 1798 Io08 B.c. 1948 - 117 B.C.Pretreated with NaOH.

An examination of the dates from Lerna, shown in ill. 1,16 indicates, at first glance, a lack of statisticalconsistency. For example, the idifference n age between the two portions of sample P-303 appearsto besignificant. When it is realized, however, that in a group of ten C-I4 dates, at least two might be* Please note that errors quoted for C-I4 dates are statisticaldeviations only and do not include the uncertainty of eitherhalf-life value.15 The second portion of this sample is listed separately in

order to illustrate graphically (in ill. I) that the difference isnot significant.16 We are indebted to Dr. Caskey for supplying the originalsketch for ill. I.


11/12

366 ELLEN L. KOHLER AND ELIZABETH K. RALPH [AJA 65expected to fall outside of the one sigma tolerance, then this apparent discrepancy may be attributedto chance.The change in C-I4 dates due to the half-life shift is also shown graphicallyin ill. I.

PYLOSPylos is located on a hill called Epano Englianos, 4 miles north of Navarino Bay in Western Messenia.Excavations begun in 1952 are being conducted by the Archaeological Expedition of the University ofCincinnati in cooperation with the Greek Archaeological Service. The Palace seems to have been builtand occupied in the period when pottery of Late Helladic, or Mycenaean, III B was in use. At the presentstage of excavation, the site comprises two administrative and residential buildings as well as a Work-shop and a Wine Magazine." The southwestern unit was probably erected first and the main centralstructure soon thereafter. All parts of the Palace as well as the lower town were destroyed by fire atthe end of Mycenaean III B. In 1958and 1959carbon samples from the Main Building, the Workshop,the Wine Magazine, and from neighboring areas on the acropolis were collected by C. W. Blegen andMarion Rawson. Preliminaryreportson the excavations have been published each yearsince 1953 n AJA.

Christian Calendar AgeCalculated with Half-life:*5568 5800

P-254.Southeastof Wine Magazine, "probably arlier than the Palace" 1499 + 54 B.c. 1636 ?L 56 B.C.Charcoal,possibly from earlier building cut through by foun-dations of Wine Magazine. Pretreated with NaOH.P-270.Southeast of Wine Magazine, "probably arlier than the Palace" 1385? 41 B.C. 1519? 43 B.C.Charcoallumps in black burnedearthabovepavement of build-ing earlier than the Wine Magazine. Pretreated with NaOH.P-340o.outheast of Wine Magazine, earlierthan Wine Magazine, but 1371 ? 48 B.c. 1504? 50 B.C.possibly contemporarywith Main Building of Palace

Charcoal.Pretreated with NaOH.P-326.Palace of Nestor, Main Building, middle Late Helladic III B'8 1500 + 58 B.C. 1638? 60 B.C.Charcoal (and rootlets) from a squared door jamb of SE doorof Queen's Hall.P-33o.Palaceof Nestor, Main Building, middle of Late Helladic III B 1405 ? 40 B.C. 1540 ? 42 B.C.Charcoal,from a horizontal squared beam above dado behindsentry stand, NE side of the doorway leading from portico tovestibule of the Megaron.P-329.Palaceof Nestor, Main Building, middle of Late Helladic III B 1306 ? 55 B.c. 1437 - 57 B.C.Charcoal (and rootlets) from a horizontal squaredbeam above

dado in E corner of inner portico of Propylon.P-328.Palace of Nestor, Main Building, middle of Late Helladic III B 1265 + 57 B.c. I394 - 59 B.C.Charcoal (and rootlets) from a horizontal squaredbeam abovedado in NE anta of portico of the Megaron.P-332.Palace of Nestor, Wine Magazine, late in Late Helladic III B 1185 + 53 B.c. 1311 - 55 B.c.Two charcoalsamples were combined from (a) W door jambof door to vestibule, and (b) beam-slot in W wall of vestibule.Pretreated with NaOH.

* Please note that errors quoted for C-14 dates are statisticaldeviations only and do not include the uncertainty of eitherhalf-life value.17 See latestplan, AJA 64 (196o) pl. 39.

18Samples P-326, P-33o, P-329, P-328, and P-337 were toosmall for pretreatment with NaOH for removal of possiblehumic acid contamination.


12/12

1961] C-14 DATES FOR SITES IN THE MEDITERRANEAN AREA 367Christian Calendar AgeCalculated with Half-life:*

5568 5800P-34I. Palace of Nestor, Workshop, late in Late Helladic III B 1055 ? 48 B.c. I1176? 50 B.C.Charcoallumps, with red and black burned earth, debris fromthe destruction of the Palace. Pretreated with NaOH.P-337.Northeast of Workshop, late in Late Helladic III B 1076 + 53 B.c. I197 - 55 B.C.Charcoal, with debris fallen from NE wall of Workshop;thrown down at time of destruction of Palace.

Samples as noted were pretreated for removal of possible humic acid contamination. As can be seenfrom comparisons of samples from adjacent levels (for example, compare P-341 and P-337), humicacid contamination was not prevalent in these samples.In this series, it is apparent, however, that some samples (see P-254 and P-326) may be erroneouslyold, of the order of Ioo years or more, because they may not have been taken from the outer growthlayers of the trees that were used for the constructions. This supposition is supported by the fact thatsamples P-326,P-33o, P-329 and P-328were obtained from beams which had possibly been squared. Weassume that a squaredbeam could have been fabricatedfrom a tree at least 200 years old, from which 100or more growth rings were removed in the process of shaping.It is interesting to note that the C-I4 dates calculated with the 5800 half-life value place all samplesclose to or earlier than 1200 B.c., the date estimated by Blegen and Rawson for the destructionof the site.EMPORIO, CHIOS

Excavations were carried on by the British School at Athens from 1952to 1955under the directorshipof M. S. F. Hood, assistedby J. Boardman and others, at Emporio on the E side of the southern tip ofthe island of Chios. Samples were collected by Mr. Hood during the 1954 season. Cf. JHS 74 (I954)162-64; 75 (I955) Arch. Suppl. p. 20; Archaeology 8 (1955) 245-51.Christian Calendar AgeCalculated with Half-life:*

5568 5800P-273.Settlement of Early Bronze Age (Troy I) Period 2025 ? 92 B.c. 2183 - 96 B.C.Charcoal from remains of wooden beams in one room of ahouse in the Troy I settlement, destroyed by fire. About 2 m.from surface. Pretreated with NaOH.P-292. Late Roman or Byzantine"1 A.D.1027 - 32 A.D. 990 -? 33Charcoal lumps from the bottom of a concrete-linedcistern infortified Roman settlement on top of acropolis.

UNIVERSITYUSEUMPHILADELPHIA* Please note that errors quoted for C-14 dates are statisticaldeviations only and do not include the uncertainty of eitherhalf-life value.19Sample P-292 was divided into two portions; one of which

was pretreatedwith NaOH. Since the difference between thetwo results was statisticallynot significant, the date given is theaverageof the two.

C-14 Dates for Sites in the Mediterranean Area Author(s)- Ellen L. Kohler and Elizabeth K. Ralph

Documents

Transcript of C-14 Dates for Sites in the Mediterranean Area Author(s)- Ellen L. Kohler and Elizabeth K. Ralph