1
Is i possible o iden i y empo al di e ences among combus ion ea u es in middle
palaeoli hic palimpses s? The a chaeomagne ic e idence: a case s udy om le el O
a he Ab ic Romaní ock-shel e (Capellades, Spain).
Á. Ca ancho1, J.J. Villalaín2, J. Vall e dú3,4, E. Ca bonell3,4
1. Á ea de P ehis o ia, Depa amen o de CC. His ó icas y Geog a ía, Uni e sidad de
Bu gos (UBU), Bu gos, Spain. Co esponding au o : [email p o ec ed]
2. Depa amen o de Física, Escuela Poli écnica Supe io , Uni e sidad de Bu gos
(UBU), Bu gos, Spain.
3. Ins i u Ca alà de Paleoecologia Humana i E olució Social (IPHES), Ta agona,
Spain,
4. À ea de P ehis ò ia, Depa amen d´His ò ia i His ò ia de l´A , Facul a de Lle es.
Uni e si a Ro i a i Vi gili (URV), Ta agona, Spain
Abs ac
A chaeomagne ic da ing is p obably one o he mos known applica ions o magne ic
me hods o a chaeology bu he e a e o he s s ill unde u ilized and o pa icula in e es
o Palaeoli hic a chaeology. He e, we epo a no el applica ion o a chaeomagne ism as
a echnique o de e mine empo al diach onies among combus ion ea u es om he
same su ace wi hin palaeoli hic palimpses s. The app oach is based on he sub le
di ec ional changes o he Ea h´s magne ic ield h ough ime (secula a ia ion, SV)
and on he abili y o bu ned ma e ials o eco d such a ia ions unde ce ain condi ions.
Th ee middle palaeoli hic hea hs om le el O (ca. 55 ka BP) a he Ab ic Romaní
ock-shel e (NE Spain), we e a chaeomagne ically in es iga ed. The s udied su ace
(black homogeneous ca bonaceous acies), eco ded he magne ic enhancemen
p oduced by i e wi h a en old inc ease in concen a ion-dependen magne ic
pa ame e s in he uppe mos cen ime e wi h espec o i s unbu ned o deepe
coun e pa s. Pseudo-single domain (PSD) Ti-low i anomagne i e was iden i ied as he
main emanence ca ie . The i e e sibili y o he momagne ic cu es sugges s ha
hese samples did no unde go enough high empe a u es as o eco d a ull
he mo emanence (TRM). Addi ionally, he occasional occu ence o maghaemi ized
magne i e is in e p e ed as an indica ion o a he mochemical emanen magne iza ion
*Manusc ip
Click he e o iew linked Re e ences
2
(TCRM), making hese samples unsui able o absolu e palaeoin ensi y de e mina ions.
Two well-de ined (α95 < 5º) and s a is ically indis inguishable a chaeomagne ic
di ec ions we e ob ained wi h hei mean di ec ions wi hin hei espec i e con idences
ci cles a he 95 % le el. The lack o di ec ional changes and he simila i y in he
magne ic p ope ies sugges ha hese hea hs eco ded simul aneously o closely
con ined in ime he Ea h´s magne ic ield di ec ion a he ime o cooling. These
esul s ag ee well wi h a chaeological e idence which indica es a synch onic occupa ion
o his ac i i y a ea. The possibili y o de e mining empo al di e ences among
combus ion ea u es in p ehis o ic si es a ises as a p omising ool in palimpses
dissec ion s udies and may help o econs uc occupa ion pa e ns o p ehis o ic g oups.
The p ac ical limi s o he me hod a e discussed as well as i s po en ial o iden i y pos -
deposi ional mechanical al e a ion p ocesses.
Keywo ds: a chaeomagne ism, diach ony, hea h, middle Palaeoli hic, Neande hals,
secula a ia ion.
1. In oduc ion
One o he majo opics in cu en a chaeological esea ch ocuses on palimpses
dissec ion in Middle Palaeoli hic si es. Unde s anding he way in which Neande hal
g oups o ganised hei li ing space in ca es and ock-shel e s la gely depends on
de ining p ecisely he spa ial and empo al ela ionships o a i ac s, eco ac s and o he
ea u es (Hen y 2012). Middle Palaeoli hic palimpses s usually consis o occupa ional
su aces wi h mul iple combus ion s uc u es, densely s a i ied, o en pa ially
o e lapping and con aining many li hic and aunal emains de ining domes ic ac i i y
a eas (Vaque o and Pas ó 2001; Bailey 2007, Hen y 2012). Thei deg ee o complexi y
is a iable depending on he numbe , ype and size o emains as well as hei spa ial
dis ibu ion (Bailey 2007). Howe e , he key issue is he di icul y o isola e and
quan i y indi idual episodes o ac i i y p ese ed in hese occupa ional su aces and
es ablish empo a y ela ionships be ween hem.
Beyond he a ie y o na u al o cul u al p ocesses in ol ed in he o ma ion o
palimpses s (see Hen y 2012), mul iple pos -deposi ional p ocesses may ac dis o ing
and e en des oying he a chaeological eco d. This complica es he in e p e a ion o he
iming and use o hese hea h- ela ed assemblages esul ing in he so-called “palimpses
p oblem” (e.g. Bailey 2007; Hen y 2012; Machado e al. 2013, 2015). O e ecen
yea s, dissec ion o palimpses s is being pe o med h ough combined analysis o aw
3
ma e ial uni s (RMUs), aunal and li hic e i s, a chaeos a ig aphy, oo h mic owea
analysis (e.g. Machado e al. 2013, 2015; Chacón e al. 2015; Ri als e al. 2009a;
Vaque o e al. 2015) and high esolu ion geoa chaeological echniques such as soil
mic omo phology, FTIR, e c. (e.g. Mille e al. 2013; Mallol e al. 2013; Cabanes e al.
2010). These app oaches a e p o iding aluable in o ma ion o econs uc si e
o ma ion p ocesses and he dynamics o occupa ion o Mous e ian g oups.
The in e p e a ion o hese con ex s especially h ough spa ial analysis equi es
exca a ion a eas wide enough o co e spa ial a iabili y. The s udy o spa ial
a angemen o hea hs, li hic and aunal emains has e ealed how Neande hals’ use o
space and hea h unc ion changed h ough ime. This has been shown by he s udies
ca ied ou a di e en le els in he Ab ic Romaní ock-shel e , NE Spain (e.g.: Pas ó e
al. 2000; Vall e dú e al. 2005, 2010; Vaque o e al. 2004, 2012a,b; Rosell e al.
2012a,b) and o he middle Palaeoli hic si es such as To Fa aj (Hen y 2012) o Pay e
(Moncel e al. 2007; Ri als e al. 2009a), among o he s.
Some ac i i ies as li hic ool ecycling by la e occupa ions also imply ime which needs
o be de e mined (Vaque o e al. 2004, 2015; Machado e al. 2015). In he ield one
usually obse e an amalgam o o e lapping elemen s di icul o di e en ia e bo h a he
spa ial and empo al scale. They may appea con empo aneous based on mac oscopic
ield obse a ions bu may also ep esen mul iple “sho - e m” occupa ions (e.g.
Vall e dú e al. 2005; Ri als e al. 2009a; Machado e al. 2013;Sánchez-He nández e
al. 2014) o al e na i ely, longe ones (e.g. Thiébau e al. 2009; Ri als e al. 2009b;
Vaque o e al. 2012a). Es ablishing he con empo anei y among ma e ial emains and
hea hs is a di icul ask and much o he p oblem lies in he deg ee o esolu ion o he
echniques a ailable. The e o e, explo ing me hodological op ions which can be used as
empo al ma ke s using single occupa ion episodes as he basic analy ical uni
(Machado e al. 2015; Chacón e al. 2015) and de e mine hei deg ee o
con empo anei y a e main goals in palimpses esea ch.
He e, we epo an applica ion o a chaeomagne ism as a echnique o de e mine
whe he di e en hea hs exposed on a li ing loo we e bu ned in he same momen o
con e sely, we e sepa a ed in ime (synch onous s. diach onous). Ou hypo hesis is
ha i se e al mean a chaeomagne ic di ec ions a e ob ained om a ious hea hs om
he same su ace and hey a e di e en (s a is ically dis inguishable), i could be
assumed ha hey we e ca ied ou in empo ally dis inc momen s being he e o e no
synch onous. This in o ma ion is o g ea alue o econs uc he occupa ion pa e ns o
4
Neande hal g oups and i has been es ed s udying se e al hea hs om le el O (ca. 55
ka BP) a he Ab íc Romaní ock-shel e . The echnique also allows assessing
mechanical pos -deposi ional al e a ion p ocesses in ca e i es (Ca ancho e al. 2012),
so hei p ima y posi ion is also e alua ed as a ool o s udy li ing loo ´s in eg i y. The
a chaeological implica ions and limi s o he me hod a e discussed.
2. Ma e ials and me hods
2.1 Ab ic Romaní ockshel e . Sampled ma e ials
The Ab ic Romaní si e (41º 32´ N; 1º 41´ E; 280 m abo e sea le el) is loca ed in he
own o Capellades, 50 km wes o Ba celona, NE Spain (Fig. 1a-b). The si e is
ockshel e opened in a a e ine cli called “Cingle a del Capelló”, in a ka s
landscape a he wes bank o he Anoia i e . The s a ig aphic sequence consis s o 20
m o well-s a i ied a e ine pla o ms da ed by U-se ies and adioca bon analysis o
be ween 40 and 70 ka BP (Vaque o e al. 2013). The a chaeological le els (a leas 25)
appea in e bedded be ween he a e ine pla o ms and co espond o pe iods o low o
no wa e inside he ock shel e . Excep le el A which belongs o he Au ignacian, all o
he a chaeological uni s co espond o he Middle Palaeoli hic and i een le els ( om
le el B o P) ha e been exca a ed (Fig. 1c).
The si e is pa icula ly sui ed o palimpses dissec ion analyses o i s mul idisciplina y
exca a ion me hodology and because is a sedimen a y con ex o high- esolu ion. The
mean sedimen a ion a e o he en i e sequence is es ima ed a ound 0.6 cm/y (Bischo
e al. 1988). The Ab ic Romaní has an ou s anding a chaeological eco d wi h housands
o li hic a e ac s, aunal and palaeobo anical emains and e en wood implemen s
(Ca bonell and Cas o-Cu el, 1995). Howe e , he si e is pa icula ly known o being a
key si e o s udy p ehis o ic i e wi h almos 200 combus ion ac i i y a eas exca a ed
(Vall e dú e al. 2010). E e y combus ion ac i i y a ea is eco ded ollowing a de ailed
ield-based desc ip ion (see Vall e dú e al. 2012). I comp ises sedimen a y acies
analyses including measu emen s o geome y, size and hickness o e e y combus ion
ea u e. In si u combus ion s uc u es a e iden i ied by ubi ied sedimen o e lain by
mix u es o ca bonaceous and ash acies. Ca bonaceous acies a e ca bon- ich
sedimen s which a e classi ied as “he e ogeneous” i unbu ned and bu ned sedimen a y
componen s a e mixed o as “homogeneous” i he ma ix exhibi s uni o m he mal
modi ica ion wi h > 40 % in cha coal con en (Vall e dú e al. 2012)
5
The hea hs s udied he e co espond o le el O da ed by means o U-se ies a a ound 55
ka BP (Bischo e al., 1988; Vaque o e al., 2013). I was exca a ed be ween 2004 and
2011 o e an a ea o abou 271 m2 wi h almos hi y combus ion s uc u es iden i ied
(Gabucio e al. 2014; Fig. 2). Mos o hem a e simple and la al hough some display
conca e o basin-like o ms. This s udy ocused on he domes ic ac i i y a ea O10 (g id
squa es UW/51-53; Fig. 3), a hea h- ela ed assemblage wi h a dense concen a ion o
li hic and aunal emains, many o hem wi h e idences o calcina ion (Chacón e al.
2015; Gabucio e al. 2014). A he ime o sampling, a da k homogeneous ca bonaceous
su ace (~ 3 m leng h) and amo phous geome y was obse ed. Acco ding o ield
obse a ions, a leas h ee di e en combus ion ocuses we e dis inguished by he
a chaeologis s. Following hei guidelines, h ee o ien ed hand-blocks we e collec ed
using Plas e o Pa is om each one o he h ee combus ion ocuses. In he lab, he
hand-blocks we e consolida ed in E hyl silica e and le o d y du ing 4 weeks and
subsequen ly subsampled aking special ca e o sample dep h. A o al o 50 cubic (10
cm3) specimens we e ob ained (Table 1).
2.2. Me hods
All palaeomagne ic and ock-magne ic analysis we e ca ied ou a he labo a o y o
Palaeomagne ism o Bu gos Uni e si y (Spain). The na u al emanen magne iza ion
(NRM) was measu ed using a 2G SQUID magne ome e (noise le el 5 × 10−12 Am2).
Low- ield suscep ibili y was measu ed wi h a KLY-4 Kappab idge (AGICO, noise le el
3×10−8 S.I.) a oom empe a u e ini ially and a e each he mal demagne iza ion s ep o
moni o possible magne o-chemical al e a ions. The NRM di ec ional s abili y was
analysed by s epwise p og essi e al e na ing ield (AF) and he mal (TH)
demagne iza ion. AF demagne iza ion was pe o med in 23 s eps up o a maximum peak
ield o 100 mT wi h he 2G magne ome e AF demagne iza ion uni . TH
demagne iza ion was ca ied ou in 15 s eps up o 585 ºC wi h a TD48-SC (ASC)
he mal demagne ize . Cha ac e is ic Remanen magne iza ion (ChRM) di ec ions we e
calcula ed by linea eg ession o he componen ha linea ly con e ges owa ds he
o igin o he o hogonal NRM demagne iza ion plo s using he Remaso so wa e
(Chadima e al 2006). Mean di ec ions and associa ed s a is ical pa ame e s we e
calcula ed using Fishe ´s (1953) s a is ics.

6
In addi ion, di e en ock-magne ic analyses we e ca ied ou in o de o iden i y he
main emanence ca ie s and hei domain s uc u e. By using a Va iable Field
T ansla ion Balance (MM_VFTB) p og essi e iso he mal emanen magne iza ion
(IRM) acquisi ion cu es, hys e esis loops (± 1 T), back ield coe ci i y cu es and
he momagne ic cu es up o 700 ºC in ai we e pe o med on ep esen a i e sample (~
450 mg) o each hea h. Cu ie poin de e mina ion was done ollowing he wo- angen
me hod o G ommé e al. (1969) and hys e esis pa ame e s –co ec ed o he
dia/pa amagne ic con ibu ion– we e calcula ed using he RockMag Analyze so wa e
(Leonha d 2006).
3. Resul s
3.1. Magne ic p ope ies
The con en o e omagne ic mine als (s.l.) among he s udied samples is a he
a iable. Many samples a e cha ac e ized by noisy diag ams whe eas o he s –mos ly
es ic ed o he uppe mos cm- a e su icien ly in e p e able as o cha ac e ize hei
magne ic p ope ies. Fig. 4(a-c) illus a es some ep esen a i e examples o
he momagne ic cu es, all co esponding o samples om he 1s uppe cm. The
in ensi y o magne iza ion a ies up o 2 o de s o magni ude and all cu es show
i e e sible beha iou . All hea ing cu es con ain a phase wi h a Cu ie poin (TC)
es ima ed be ween 530 ºC and 580 ºC indica ing ha Ti-low i anomagne i e is he main
magne iza ion ca ie (Fig. 4a-c). Occasionally, a phase wi h TC ex ending o 600 – 610
ºC was also iden i ied (Fig. 4a), which migh be ela ed wi h sligh ly oxidized magne i e
(magne i e pa ially maghaemi ized). Some samples also display in lec ions in he ange
o 350 – 370 ºC (Fig. 4a), p obably due o maghaemi e in e ing o less magne ic
haema i e du ing hea ing o a highly isomo phous subs i u ed spinel phase. The
occu ence o e omagne ic sulphides canno be uled ou bu , o he bes o ou
knowledge, is e y a e in bu ned a chaeological ma e ials. In o he cases, an inc ease in
magne iza ion is obse ed om 360 ºC indica ing he c ea ion o magne i e p obably
om he ans o ma ion o some pa amagne ic mine al (Fig. 4b). Mo eo e , seconda y
magne i e is also c ea ed as e ealed by he inc ease in magne iza ion in mos cooling
cu es (Fig. 4a-b). O e all, hese esul s indica e ha mos o hese samples did no
expe ience hea ing empe a u es as high as hose used in his expe imen (700 ºC).
O he wise, hey would be much mo e e e sible.
7
IRM p og essi e acquisi ion cu es (no shown he e) a e almos sa u a ed a ound ∼ 150
– 200 mT, indica ing ha he emanence is ca ied by a low-coe ci i y e omagne ic
mine al. Hys e esis measu emen s e ealed ha he samples a e cha ac e ized by low
coe ci e ields (Bc = 5.5 - 14.25 mT). The emanen coe ci i ies (Bc ), de e mined
sepa a ely om he back ield cu es, a y be ween 18 and 40 mT. Hys e esis a ios
ange om 0.10 < M s/Ms < 0.21 and 2.20 < Bc /Bc < 3.80 which acco ding o Dunlop
(2002) heo e ical mixing lines indica es he dominance o pseudo-single domain (PSD)
magne i e pa icles (Fig. 5).
In o de o s udy he a ia ion o magne ic p ope ies in dep h, one o hese hea hs
(GV4) was subsampled e e y cm (Fig. 6). The in ensi y o magne iza ion dec eases
almos an o de o magni ude below he uppe mos cm as shown in he he momagne ic
cu es (Fig. 6a-c-e) and hei co esponding hys e esis loops (Fig. 6b-d- ). This
indica es ha he 1s cm was he mos hea ed and below ha dep h, he samples a e
noisie and magne ically weake . This would be in acco dance wi h he colou a ia ion
om blackish a he op o yellowish a he base. A h ee poin smoo hing was applied o
he he momagne ic cu es in o de o make hem mo e amenable o in e p e a ion. A
single phase wi h a TC o a ound 585 – 600 ºC was obse ed poin ing o Ti-low
i anomagne i e as he main ca ie showing aces o maghaemi iza ion.
3.2 A chaeomagne ic di ec ions
Ini ial na u al emanen magne iza ion (NRM) alues ange om 6.03 x 10-6 o 6.78 x
10-8 Am2kg-1 whe eas magne ic suscep ibili y (MS) oscilla es be ween 2.08 x 10-8 and -
3.09 x 10-9 m3kg-1. 38 % o samples showed ini ial diamagne ic (nega i e) MS alues,
indica ing ha he concen a ion o e omagne ic mine als is poo . The highes alues
o bo h pa ame e s co espond o he uppe mos (1s cm) supe icial samples in
ag eemen wi h he esul s obse ed in he mic o-p o ile in dep h (Fig. 6).
The NRM s abili y o he s udied samples is qui e ep oducible among samples om he
GV3 and GV4 hea hs. Howe e , GV2 samples exhibi ed anomalous beha iou s as
explained below. A e emo al o a small seconda y componen (< 15 mT o 250 ºC),
AF demagne iza ion diag ams a e mos ly de ined by a s able single componen o
no mal pola i y almos demagne ized a 100 mT (Fig. 7a-c). In he case o TH
demagne iza ion, he ChRM di ec ion was de e mined be ween 250 – 300 ºC and 585 ºC
8
(Fig. 7b-d). In con as , samples om GV2 hea h a e e y he e ogeneous. Some
specimens show a uni ec o ial and s able no mal pola i y componen (Fig. 8a) while
o he s display mul icomponen s uc u e o magne iza ion wi h anomalous di ec ions
(Fig. 8c-d). Mos likely, his beha iou e lec s some ype o mechanical ewo king o
he sedimen a e bu ning. The mean a chaeomagne ic di ec ions de e mined om each
hea h and hei co esponding associa ed s a is ics a e shown in Fig. 9 and Table I,
espec i ely.
4. Discussion
The pu pose o his s udy is no he use o a chaeomagne ism as da ing me hod.
S anda d a chaeomagne ic da ing equi es o egional secula a ia ion cu es (SVCs)
composed o well-da ed and high-quali y (TRM) da a desc ibing he Ea h´s magne ic
ield a ia ions h ough ime. Da ing is ob ained by compa ing he di ec ion and/o
in ensi y de e mined om an a chaeological si e wi h he SVC a ailable o he e i o y
and pe iod conce ned. In Eu ope o example, mos SVCs “only” co e he las 2-3
millennia (e.g. Galle e al. 2002; Schnepp and Lanos 2005; Gómez-Pacca d e al. 2006;
Tema e al. 2006; Zanani i e al. 2007) and excep ionally some eco ds each he las 8
ky (Ko ache a e al. 2014; Tema and Kondopoulou 2011, Ca ancho e al. 2013).
Howe e , we a e dealing he e wi h mid-Palaeoli hic hea hs o ~ 55 ky and he e is no
any SV eco d o his ch onology. Ou app oach is based on he compa ison o he
mean a chaeomagne ic di ec ions ob ained om di e en hea hs exposed in he same
a chaeological su ace. I hese i eplaces we e bu ned a di e en imes (enough ime
o dis inguish di ec ional a ia ions o magne ic no h -o de o decades o mo e-),
di e en o s a is ically dis inguishable mean di ec ions should be eco ded p o iding
empi ical e idence o diach ony. The in e p e a ion is no so s aigh o wa d since
se e al ac o s come in o play. The age o hese hea hs is wi hin he no mal pola i y
B uhnes Ch on (las ~780 ka), so all in si u samples should display no hwa d
di ec ions. This means ha compa ison be ween di ec ions is based on small ( ew
deg ees) di ec ional changes. Accu acy in sampling o ien a ion and lab analysis is
impo an as e o s may imply signi ican sca e and high s a is ical unce ain ies.
Some imes is he in insic samples’ beha iou which p e en s a p ope in e p e a ion
ei he by di icul ies isola ing he ChRM di ec ion o due o mine alogical al e a ions.
A ailable SV eco ds o he las millennia o mid la i udes as he Ibe ian Peninsula
(Gómez-Pacca d e al. 2006), show ha magne ic declina ion luc ua es wi hin a
9
dispe sion ange o app oxima ely ± 20º o he p esen ield and inclina ion be ween 40º
o 65º in an chao ic, no p ede ined pa e n and mo e impo an ly, wi h a ying SV
a es. Tha is, he ield a ies apidly a ce ain imes and slowly in o he s. We a e awa e
ha i is emp ing o a chaeologis s o look o e idence o empo al di e ences
be ween mul iple combus ion ea u es om he same su ace. Howe e , o he gi en
easons, he compa ison o di e en mean di ec ions om he same occupa ional
su ace canno be di ec ly in e p e ed in e ms o du a ion.
The s udied i eplaces all show simila magne ic p ope ies wi h PSD Ti-low
i anomagne i e as main emanence ca ie . NRM in ensi ies a e qui e a iable wi h he
highes alues cons ained o he uppe mos cm. I should be kep in mind ha he
o iginal subs a e whe e he hea hs we e ca ied ou was a e ine, mos ly composed
o ca bona es and e y poo in e omagne ic (s.l.) mine als. Tha explains why below
he 1s cm - he mos hea ed pa - he magne ic signal is weak wi h noisy diag ams due o
a dominan diamagne ic beha iou . Two ou o he h ee s udied hea hs ha e yielded
well-de ined geomagne ic ield di ec ions, wi h good s a is ical pa ame e s (α95 < 5º)
and low sca e in he s e eog ams (Fig. 9 and Table 1). Samples om GV3-4 hea hs
displayed a s able, no mal pola i y ChRM componen indica ing ha hey success ully
eco ded he Ea h´s magne ic ield di ec ion a he ime o cooling. On he con a y,
some ype o mechanical dis u bance (e.g. ampling, bio u ba ion) a ec ed GV2 hea h
a e bu ning since mos o i s samples exhibi mul icomponen NRM beha iou and
anomalous di ec ions. The use ulness o he palaeomagne ic echnique o assess pos -
deposi ional al e a ion p ocesses in ca e i es has al eady been demons a ed (Ca ancho
e al. 2012).
Ideally, a chaeomagne ic s udies a e ca ied ou on in si u, well-hea ed a chaeological
ma e ials ca ying a he mal emanen magne iza ion (TRM). TRM is by a he mos
e icien eco ding mechanism o geomagne ic ield a ia ions. Howe e , i he ma e ial
does no each empe a u es high enough o eco d a ull TRM (e.g. > 580 °C, magne i e
Tc; Dunlop and Özdemi , 1997) o ehea ings a mild empe a u es occu , a pa ial
he mo emanen magne iza ion (pTRM) migh be eco ded pa ially o e lapping he
o iginal magne iza ion. The pTRM would eco d he Ea h´s magne ic ield di ec ion o
he las hea ing, bu we did no ind e idence o i he e. Al e na i ely, mine alogical
al e a ions as phase changes o g ain g ow hs may also imply he acquisi ion o
16
Dunlop, D., Özdemi , Ö., 1997. Rock Magne ism. Fundamen als and F on ie s.
Camb idge Uni . P ess, New Yo k. 573 pp.
Eighmy, J.L., Ha haway, J.H., 1987. Con empo a y A chaeomagne ic Resul s and he
Accu acy o A chaeomagne ic Da es. Geoa chaeology 2(1): 49-61.
Fishe , R., 1953. Dispe sion on a sphe e. P oceedings Royal Socie y London A 217,
295-305.
Gabucio, M. J., Cáce es, I., Rosell, J., Saladié, P., and Vall e dú. J., 2014. F om small
bone agmen s o neande hal ac i i y a eas: he case o Le el O o he Ab ic Romaní
(Capellades, Ba celona, Spain). Qua e na y In e na ional 330, 36-51.
Galle , Y., Gene ey, A., LeGo , M., 2002.Th ee millennia o di ec ional a ia ion o
he Ea h’s magne ic ield in Wes e n Eu ope as e ealed by a chaeological a e ac s.
Physics o he Ea h and Plane a y In e io s. 131, 81–89.
Gómez-Pacca d, M., Chau in, A., Lanos, P., McIn osh, G., Ose e, M.L., Ca anza i i, G.,
Ruiz-Ma inez, V.C., Núñez, J.I., 2006. Fi s a chaeomagne ic secula a ia ion cu e
o he Ibe ian Peninsula: compa ison wi h o he da a om Wes e n Eu ope and wi h
global geomagne ic ield models. Geochemis y, Geophysics, Geosys ems, 7, Q12001.
doi:10.1029/2006GC001476.
G ommé, C.S., W igh ,T.L., and Peck, D.L., 1969. Magne ic p ope ies and oxida ion
o i on- i anium oxide mine als in Alae and Makaopuhi la a lakes, Hawaii. J. Geophys.
Res., 74, 5277-5294.
Hags um, J.T., Blinman, E., 2010. A cheomagne ic da ing in wes e n No h Ame ica:
An upda ed e e ence cu e based on paleomagne ic and a cheomagne ic da a se s.
Geochemis y, Geophysics, Geosys ems 11 (6), Q06009.
Hen y, D.O., 2012. The palimpses p oblem, hea h pa e n analysis, and Middle
Paleoli hic si e s uc u e. Qua e na y In e na ional 247, 246-266.
Kappe , K., Anesin, D., Donadini, F., Angelucci, D., Ca ulli, F., Ped o i, A. and Hi ,
A., 2014. Linking si e o ma ion p ocesses omagne ic p ope ies. Rock- and
a chaeomagne ic analysis o he combus ion le els a Ripa o Gaban (I aly). Jou nal o
A chaeological Science, 41, 836–855.
Ko ache a, M., Kos adino a-A amo a, M., Jo dano a, N., Lanos, P., Boyadzhie , Y.,
2014. Ex ended and e ised a chaeomagne ic da abase and secula a ia ion cu es
om Bulga ia o he las eigh millennia. Physics o he Ea h and Plane a y In e io s
236, 79-94.

17
Lengyel, S.N., Eighmy, J.L., Van Bu en, M., 2011. A chaeomagne ic esea ch in he
Andean highlands. Jou nal o A chaeological Science 38, 147-155.
Leonha d , R., 2006, Analyzing ock magne ic measu emen s: The RockMagAnalyze
1.0 so wa e. Compu e s & Geosciences, 32, 1420–1431.
Machado, J., He nández, C.M., Mallol, C. and Gal án, B., 2013. Li hic P oduc ion, Si e
Fo ma ion and Middle Palaeoli hic Palimpses Analysis: In Sea ch o Human
occupa ion Episodes a Ab ic del Pas o S a ig aphic uni IV (Alican e, Spain). Jou nal
o A chaeological Science 40 (5): 2254-2273.
Machado, J., Mallol, C., and He nández, C.M., 2015. Insigh s in o Eu asian Middle
Paleoli hic Se lemen Dynamics: The Palimpses P oblem. . In: Cona d, N. J.,
Delagnes, A. (Eds.), Se lemen Dynamics o he Middle Paleoli hic and Middle S one
Age, Vol. IV. Tübingen Publica ions in P ehis o y. Ke ns Ve lag, pp. 361-382 .
Mallol, C., He nández, C.M., Cabanes, D., Sis iaga, A., Machado, J., Rod íguez, Á.,
Pé ez, L. and Gal án, B., 2013. The Black laye o Middle Palaeoli hic combus ion
S uc u es. In e p e a ion and A chaeos a ig aphic Implica ions. Jou nal o
A chaeological Science 20, 2515-2537.
Mille , C. E., Goldbe g, P. and Be na, F., 2013. Geoa chaeological in es iga ions a
Diepkloo ock Shel e , wes e n cape, Sou h A ica. Jou nal o A chaeological Science
40: 3432-3452
Men ze , S. M., 2009. Bone as a uel sou ce: The e ec s o ini ial agmen size
dis ibu ion. In I. Thé y-Pa iso , S. Cos amagno, & A. Hen y (Eds.), Ges ion des
Combus ibles au Paleoli hique e au Mesoli hique: Nou eaux Ou iles, Nou elles
In e p e a ions. UISPP P oceedings o he XV Wo ld Cong ess (Lisbon, 4–9 Sep embe
2006) (pp. 53–64). Ox o d: A chaeop ess (B i ish A chaeological Repo s In e na ional
Se ies 1914).
Moncel, M. H., Chacón, M.G., Bou eaux, A., Julien, M.-A. and Pa ou-Ma his, M.,
2007. Ocupación en ab igo du an e el es adio iso ópico 5: El ni el d del yacimien o de
Pay e (A dèche, F ancia). Ges ión del e i o io y ac i idades den o del yacimien o.
T abajos de P ehis o ia 64: 65-86.
Pas ó, I., Allué, E., and Vall e dú, J. 2000. Mous e ian hea hs a Ab ic Romaní,
Ca alonia (Spain). In C. B. S inge , R. N. E. Ba on, & J. C. Finlayson (Eds.),
Neande hals on he edge: Pape s om a con e ence ma king he 150 h anni e sa y o
he Fo bes’ Qua y disco e y, Gib al a (pp. 59–67). Ox o d: Oxbow Books.
18
Ri als, F., Moncel, M.H. and Pa ou-Ma his, M., 2009a. Seasonali y and In a-Si e
Va ia ion o neande hal occupa ions in he Middle Palaeoli hic locali y o Pay e
(A dèche, F ance) using den al wea Analyses. Jou nal o A chaeological Science 36:
1070-78
Ri als, F., Schulz, E. and Kaise , T.M., 2009b. A new Applica ion o den al wea
analyses: Es ima ion o du a ion o hominid occupa ions in A chaeological locali ies.
Jou nal o Human E olu ion 56, 329-39.
Rosell, J., Cáce es, I., Blasco, R., Bennàsa , M., B a o, P., Campeny, G., Es eban-
Nadal, M., Fe nández-Laso, M.C., Gabucio, M.J., Hugue , R., Ibáñez, N., Ma ín, P.,
Ri als, F., Rod íguez-Hidalgo, A., Saladié, P., 2012a. A zooa chaeological con ibu ion
o es ablish occupa ional pa e ns a Le el J o Ab ic Romaní (Ba celona, Spain).
Qua e na y In e na ional 247, 69-84.
Rosell, J., Blasco, R., Fe nández-Laso, Mª.C., Vaque o, M. and Ca bonell, E., 2012b.
Connec ing A eas: Faunal e i s as a diagnos ic Elemen o Iden i y Synch onici y in he
Ab ic omaní A chaeological Assemblages. Qua e na y In e na ional 252: 56-67
Sánchez-He nández, C., Ri als, F., Blasco, R. and Rosell, J., 2014. Sho , bu epea ed
Neande hal isi s o Teixone es Ca e (MIS 3, Ba celona, Spain): a combined analysis
o oo h mic owea pa e ns and seasonali y. Jou nal o A chaeological Science 49, 317-
325.
Schnepp, E., Lanos, P., 2005, A chaeomagne ic secula a ia ion in Ge many du ing he
pas 2500 yea s. Geophysical Jou nal In e na ional 163, 479–90.
Shipman, P., Fos e , G.F., Schoeninge , M., 1984. Bu n bones and ee h: an
expe imen al s udy o colou , mo phology, c ys al s uc u e and sh inkage. Jou nal o
A chaeological Science 11, 307-325.
S e nbe g, R. and Lass, E. H.E., 2007. An a chaeomagne ic s udy o wo hea hs om
Keba a Ca e, Is ael. In: (O e Ba -Yose and Liliane Meignen, eds.), Keba a Ca e, M .
Ca mel, Is ael: The Middle and Uppe Paleoli hic A chaeology, Pa I. Ame ican
School o P ehis o ic Resea ch, Bulle in 49, Peabody Museum o A chaeology and
E hnology, Ha a d Uni e si y, Camb idge, 123- 130.
Tema, E., Hedley, I., Lanos, P., 2006. A chaeomagne ism in I aly: a compila ion o da a
including new esul s and a p elimina y I alian secula a ia ion cu e. Geophysical
Jou nal In e na ional, 167, 1160–71.
19
Tema, E., Kondopoulou, D., 2011. Secula a ia ion o he Ea h’s magne ic ield in he
Balkan egion du ing he las eigh millennia based on a chaeomagne ic da a.
Geophysical Jou nal In e na ional 186, 603–614.
Thé y-Pa iso , I., 2002. Fuel managemen bone and wood du ing he lowe Au ignacian
in he Pa aud Rock Shel e Lowe Palaeoli hic, Les Eyzies de Tayac, Do dogne, F ance.
Con ibu ion o expe imen a ion. Jou nal o A chaeological Science 29, 1415-1421.
Thiébau , C, Meignen, L. and Lé êque, F., 2009. Les de niè es occupa ions
mous é iennes de Sain césai e (Cha en e-Ma i ime, F ance): Di e si é des echniques
u ilisées e compo emen s économiques p a iqués. Bulle in de la Socié é P éhis o ique
F ançaise 106 (4), 691-714.
Vall e dú, J., Allué, E., Bischo , J. L., Cáce es, I., Ca bonell, E., Ceb iá, A., e al.,
2005. Sho human occupa ions in he Middle Palaeoli hic le el I o he Ab ic Romaní
ock-shel e (Capellades, Ba celona, Spain). Jou nal o Human E olu ion, 48, 157–174.
Vall e dú, J., Vaque o, M., Cáce es, I., Allué, E., Rosell, J., Saladié, P., Chacón, Mª.G.,
Ollé, A., Canals, A., Sala, R., Cou y, M.-A. and Ca bonell, E., 2010. Sleeping ac i i y
a ea wi hin he si e s uc u e o A chaic human g oups: E idence om Ab ic Romaní
Le el N combus ion ac i i y a eas. Cu en An h opology, 51(1), 137–145
Vall e dú, J., Alonso, S., Ba galló, A., Ba olí, R., Campeny, G., Ca ancho, Á., e al.,
2012. Combus ion s uc u es o a chaeological le el O and Mous e ian ac i i y a eas
wi h use o i e a he Ab ic Romaní ockshel e (NE Ibe ian Peninsula). Qua e na y
In e na ional, 247, 313–324
Vaque o, M. and Pas ó, I., 2001. The de ini ion o spa ial uni s in Middle Palaeoli hic
Si es: The hea h ela ed assemblages. Jou nal o A chaeological Science, 28 (11), 1209–
1220.
Vaque o, M., Rando, J.M., Chacón, G., 2004. Neande hal spa ial beha io and social
s uc u e: hea h- ela ed assemblages om he Ab ic Romaní middle Palaeoli hic si e.
In: Cona d, N.J. (Ed.), Se lemen Dynamics o he Middle Paleoli hic and Middle S one
Age, ol. 2. Ke ns Ve lag, Tübingen, pp. 367-392.
Vaque o, M., Chacón, M.G., Ga cía-An ón, M.D., Gómez, R., Ma ínez, K., Cua e o,
F., 2012a. Time and space in he o ma ion o li hic assemblages: he example o Ab ic
Romaní Le el J. Qua e na y In e na ional 247, 162-181.
Vaque o, M., Chacón, M.G., Cua e o, F., Ga cía-An ón, M.D., Gómez de Sole , B.,
Ma ínez, K., 2012b. The li hic assemblage o le el J. In: Ca bonell, E. (Ed.), High
20
Resolu ion A chaeology and Neande hal Beha io . Time and Space in Le el J o Ab ic
Romaní (Capellades, Spain). Sp inge , Do d ech , pp. 189-311.
Vaque o, M., Allué, E., Bischo , J.L., Bu jachs, F., Vall e dú, J., 2013. En i onmen al,
deposi ional and cul u al changes in he Uppe Pleis ocene and Ea ly Holocene: he
Cingle a del Capelló sequence (Capellades, Spain). Qua e nai e 24 (1), 49-64.
Vaque o, M., Ba galló, A., Chacón, M.G., Romagnoli, F., Sañudo, P., 2015. Li hic
ecycling in a Middle Paleoli hic expedien con ex : E idence om he Ab ic Romaní
(Capellades, Spain). Qua e na y In e na ional 361, 212-228.
Zanani i, I., Ba , C. M., Lanos, P., Ta ling, D. H., and Lin o d, P., 2007,
A chaeomagne ic secula a ia ion in he UK du ing he pas 4000 yea s and i s
applica ion o a chaeomagne ic da ing, Physics o he Ea h and Plane a y In e io s, 160
(2), 97–107.
Figu e cap ions
Fig. 1 (a-b) Geog aphic loca ion o Capellades (Ba celona, Spain) and majo
physiog aphical uni s in he NE Ibe ian Peninsula. Legend: Pi, Py enees; CE, Eb o
Basin; Cp , P e-coas al anges; P, Penedès g aben; V, Vallès g aben; CL, Coas al ange;
MP, P ades massi ; Ll, Lleida; B, Ba celona; T, Ta agona. (c) Li hos a ig aphic
column o he Ab ic Romaní Co e a No d p o ile. The s a ig aphic column con ains he
empo al posi ion o he a chaeological le el in acco dance wi h he ch onology o he
basal bounda ies o he Dansgaa d-Oeschge e en s in he GISP2 empo al scale model
(Blunie y B ook, 2001). Legend o he li hological column: 1, o ganomine al g ey
ho izon; 2, ed siliciclas ic and calci ic sil y sand; 3, yellow calci ic sand; 4, yellow
u aceous- a e ine g a el and calci ic sand; 5, pla y g a els o c ys alli ic a e ine
and calci ic sand and sil ; 6, speleo hems; 7, cemen ed sands and a e ines; 8, dias em;
9, pa acon o mi y o e osi e uncon o mi y; 10, a chaeological bed. Legend o he
commen columns: a, ock- all o a e ine blocks and megablocks; b, le e s o he
a chaeological beds; e, sedimen a y sequences; d, he lowe bounda y ch onology o he
Dansgaa d-Oeschge e en s in he empo al scale model o he GISP2 co e (Blunie y
B ook, 2001).
Fig. 2. Si ua ion o combus ion s uc u es in Le el O o Ab ic Romaní wi h indica ion
o he s udied a ea 010 (dashed line). “Pou Romaní” e e s o a su ey pi and he
s a ig aphic es imonial is an unexca a ed a ea used as a s a ig aphic p o ile. A
21
bimodal dis ibu ion o combus ion s uc u es in his le el a ound 2.5 and 5.5 m om
he wall o he ockshel e can be dis inguished (see Vall e dú e al. 2012).
Fig. 3. Field eco d o combus ion s uc u e U-V / 50 - 53, which o ms pa o zone
O10 (Chacón e al., 2015), used o d aw up he Ab ic li ho acies map and de e mine he
mic os a ig aphy o Ab ic Romaní le el O. Legend: 1, ou line o he cha coal- ich
deposi s. 2, ou line o he eddened deposi s. a, eddened a e ine sands and g anules
in s a i ied ho izon al beds. b, a e ine sands and g anules in black s a i ied
ho izon al beds. c, a e ine sands and g anules in g ey and whi ish (milky) g ey
s a i ied ho izon al beds. d, a e ine sands and g anules s a i ied in yellow ho izon al
beds. e, cemen ed sands and a e ines s a i ied in yellow o ed dome-shaped o
ho izon al beds. , hea hs sampled o a chaeomagne ic analysis.
Fig. 4. (a-c) Rep esen a i e he momagne ic cu es (magne iza ion s. empe a u e) o
h ee samples om he 1s cen ime e o dep h. Hea ing (cooling) cycles a e plo ed in
ed (blue) wi h hei espec i e a ows. Sample code and magne iza ion in ensi y alues
a e indica ed.
Fig. 5. M s/Ms s. Bc /Bc loga i hmic plo -Day diag am- (Day e al. 1977) o
ep esen a i e samples om he hea hs s udied. The dashed lines ep esen mixing
cu es aken om Dunlop (2002) o mix u es o single-domain (SD) wi h mul idomain
(MD) o supe pa amagne ic (SP) magne i e pa icles.
Fig. 6. Va ia ion wi h dep h o he momagne ic cu es (a-c-e) and hei co esponding
hys e esis loops (b-d- ) o he GV4 hea h subsampled e e y cen ime e. Sample code,
dep h and magne iza ion in ensi y alues a e indica ed o each panel. Magne iza ion
in ensi y o hea ing cycles a 20 ºC is indica ed in he momagne ic cu es. Hys e esis
loops (± 1 T) a e co ec ed o he dia/pa amagne ic ac ion and exp essed on a mass-
speci ic basis. The main hys e esis pa ame e s and a ios a e also included.
Fig. 7. Rep esen a i e o hogonal NRM demagne iza ion plo s, s e eoplo s and
no malized decay in ensi y cu es o samples om (a-b) GV3 and (c-d) GV4 hea hs.
Solid (open) ci cles show p ojec ions o ec o endpoin s on o he ho izon al ( e ical)
plane. The in ensi y and sample code a e indica ed o each sample. A.F. (al e na ing
ield); TH (The mal).

22
Fig. 8. (a-d) Rep esen a i e o hogonal NRM demagne iza ion plo s, s e eoplo s and
no malized decay in ensi y cu es o samples om GV2 hea h. Legend and symbols as
in Fig. 7.
Fig. 9. Equal-a ea p ojec ions o all ChRM di ec ions oge he wi h he mean di ec ion
and α95 o each o he s udied hea hs. See also Table 1.
Fig. 10. Equal-a ea p ojec ions wi h he mean di ec ions ob ained acco ding o he
legend. The a ea is blown-up on he le o deno e how GV3 and GV4 mean di ec ions
a e con ained wi hin hei espec i e con idence ci cles. S a ig aphic ske ch a he base
showing he loca ion o he hea hs.
Figu e 1
Click he e o download high esolu ion image
Figu e 2
Click he e o download high esolu ion image
U
V
53 52 51 50
730
740
750
720
10 500 cm
B'
B
730
740
750
10 500 cm
C'
A’
C
A
A'
A
730
740
750
720
760
GV2
GV2
GV3
GV3
GV4
GV4
BB’
C
C'
[729]
[728]
[726]
[730]
[730]
[734]
[731]
[728]
[731]
[732]
10 500 cm
12
a b
c d
e
Figu e 3
N
GV2
GV3
GV4
730
740
750
720
760
GV2 GV3 GV4
Figu e 10

Table 1. A chaeomagne ic di ec ional esul s. F om le o igh : hea h code, N/N´(numbe o
specimens conside ed o calcula e he ChRM di ec ion / numbe o specimens p ocessed), Dec.
(magne ic declina ion), Inc. (magne ic inclina ion), k (concen a ion pa ame e ) and α95
(con idence limi o mean di ec ion a he 95 % le el, om Fishe s a is ics)
Hea h
N / N´
Dec.
Inc.
k
α95
GV2
4/10
10
46.1
107.8
8.9
GV3
15/19
2.1
52.2
109.8
3.7
GV4
14/21
5.6
52.6
97.0
4.1
Table 1
Answe s o e iewe ´s commen s
Please, ind below de ailed answe s o e e y poin aised by e iewe .
…
Some poin s may equi e mo e de ailed explana ion
- My majo conce n is he in e p e a ion o possible exis ence o he mochemical
emanen magne iza ion in some samples. I in i e o au ho s do discuss in g ea de ails
whe he he c ea ion o TCRM may a ec he p ima y di ec ions. This is e y impo an
poin since some e y ine changes in a chaeodi ec ions a e in e p e ed by he au ho s as
he esul o a ia ion o geomagne ic declina ion and inclina ion.
The ques ion is no i he eco d o a TCRM a ec s he p ima y di ec ion. A TCRM is a
ype o emanence acqui ed simul aneously du ing cooling below he Cu ie empe a u e,
so he di ec ional eco d o he Ea h´s magne ic ield is o ally us wo hy (Dunlop and
Özdemi 1997: 409). The p oblem he e is o p o e whe he he emanence is ac ually a
TCRM o a CRM. A CRM acqui ed ime a e he las bu ning may dis o he p io
TRM di ec ion. In single-phase oxida ion p ocesses such as he magne i e o
maghaemi e oxida ion (o maghaemi iza ion), pa icula ly o single-domain o small
PSD g ains, is gene ally accep ed ha he CRM inhe i s o p ese es he o iginal TRM
di ec ion (see Dunlop and Özdemi and e e ences he ein). Fo la ge (MD) g ains o
mul iphase oxida ion p ocesses (e.g. in e g own o maghaemi e-haema i e om
i anomagne i e) he ield ac ing du ing oxida ion may in luence he CRM di ec ion
e en lying in in e media e di ec ions o no palaeomagne ic signi icance (c . Heide and
Dunlop 1987). Clea ly, ha is no ou case because we ha e no e idence o haema i e
in e g owns in hese samples. We only obse ed occasionally aces o
maghaemi iza ion o pa ially maghaemi ized magne i e. These explana ions a e qui e
ha d o an a chaeological audience bu he basic ideas a e included in he discussion o
help unde s anding. Anyway, di ec ional ideli y in a TCRM is absolu ely ou o doub
as we claim because i is acqui ed du ing ini ial cooling and he ield does no change.
This is b ie ly explained in he main ex (see pages 9 and 10).
- The au ho s a gue ha 'The i e e sibili y o he momagne ic cu es… may indica e
he c ea ion o TCRM a he han a TRM ( he mo emanen magne iza ion). This is
p obably w ong since many o he ac o s (and no only occu ence o TCRM) may
cause he i e e sibili y obse ed on con inuous he momagne ic cu es.
Yes, he e iewe is igh . The lack o e e sibili y does no necessa y imply a TCRM.
This may be caused by many o he easons such as a ia ions in magne ic domains,
changes in he opology o he sample, phase al e a ions induced by hea ing (e.g.:
ans o ma ion o pa amagne ic mine als, phyllosilica es o e en o he e omagne ic
mine als). In any case, he i e e sibili y o he momagne ic cu es canno be used as
c i e ion o ecognize a TCRM. I has been modi ied in he ex acco dingly (see
abs ac and conclusions).
- The ideal case o a chaeomagne ic in es iga ion on bu ned ea u es consis s o
de e mine absolu e in ensi y oge he wi h a chaeodi ec ions. Please discuss in mo e
de ails why hese samples a e unsui able o such expe imen s (classical Thellie o
Mul ispecimen app oach)?
Tha ’s ue. The ideal si ua ion is o de e mine he ull geomagne ic ec o including
bo h di ec ions (declina ion and inclina ion) and a chaeoin ensi y alues. Howe e , he
success a e in palaeoin ensin y expe imen s is usually e y low (< 30 %) because he
ma e ials o en do no ul ill he necessa y equi emen s. These a e ha he emanence
*De ailed Response o Re iewe s
mus be p e e ably ca ied by non-in e ac ing single domain (SD) pa icles (pa icula ly
ue o he Thellie me hod; no so c i ical o he mul ispecimen me hod), he sample
mus exhibi a high he momagne ic e e sibili y and abo e all, he p ima y di ec ion
mus be a he mo emanence (TRM).
Fi s , he s udied ma e ials a e pseudo-single domain (PSD) magne i e pa icles,
implying ha hey con ain a mix u e o single-domain and mul idomain pa icles.
Howe e , he domain s a e is no so c i ical depending on he palaeoin ensi y me hod
used. Reliable da a can be ob ained wi h he mul ispecimen me hod ega dless o he
dominan domain s a e. In he Thellie me hod, he p esence o mul idomain pa icles is
a p oblem, because i gene a es conca e “A ai” diag ams.
Second, hese samples do no show a high he momagne ic s abili y gi en he
i e e sibili y o he momagne ic cu es. I s ongly indica es ha hey do no ca y a
ull TRM because p obably hey did no unde go e y high empe a u e hea ing. The
lack o e idence sugges ing a TRM as he p ima y emanence makes hese samples
unsui able ma e ial o absolu e palaeoin ensi y analysis. This is al eady indica ed in he
main ex (page 10).
- Domain s a e es ima ion using oom empe a u e hys e esis pa ame e s in e ms o
he plo o magne iza ion a io s coe ci i y a io has no esolu ion o mos o na u al
ocks and bu ned a chaeomagne ic ma e ials.
Based on he expe imen al s udy o he chemically well-iden i ied syn he ic
i anomagne i es, Day (1977) p oposed an empi ic ela ion be ween he domain
s uc u e and he hys e esis pa ame e s, which has been widely used in esea ch pape s
in paleo and ock-magne ism. Howe e , na u al ocks, almos always plo on he
pseudo-single-domain beha io judging om hei hys e esis pa ame e alues. This is
ue o his s udy oo. Please use Dunlop's (2002) in e ace o disc imina e be ween
hys e esis a ios and hei ela ionship wi h he domain sa e.
The hys e esis pa ame e s o his collec ion we e al eady ep esen ed in he Day plo
including he mixing heo e ical cu es o Dunlop (2002) o mix u e o single-domain
(SD) wi h mul idomain (MD) o supe pa amagne ic (SP) magne i e pa icles (see Fig.
5). I is indica ed in he legend o Fig. 5. The hys e esis esul s a e also discussed in he
main ex (end 2nd pa ag aph page 6).