ArticlePDF Available

Mammuthus meridionalis from Madonna della Strada (Scoppito, L’Aquila): Diagnostics and restoration

Authors:

Abstract and Figures

The skeleton of Mammuthus meridionalis (Nesti, 1825), found in 1954 at Madonna della Strada (Scoppito, L’Aquila, central Italy) and exhibited since 1960 in the Spanish Fortress in L’Aquila, has recently undergone a complex diagnostic and restoration project funded by the “Guardia di Finanza” as a contribution to the reconstruction of the city of L’Aquila after the earthquake of 6 April 2009. The specimen was only slightly damaged by the seismic event. The restoration based on a highly integrated multidisciplinary approach, began with a preliminary diagnostic analysis in order to acquire information on the state of preservation of the skeleton and on the dynamic behaviour of the metal supporting frame-skeleton assembly. The restoration was carried out by combining the most common techniques applied in Palaeontology with the most advanced techniques used for the restoration of cultural artefacts. The painstaking restoration enabled the original shape of the skeleton to be restored, revealing the actual shape of the skull and its pathological modifications. Before reassembling the skeleton, a new, correct posture was also defined.
Content may be subject to copyright.
Bollettino della Società Paleontologica Italiana, 56 (3), 2017, 359-373. Modena
ISSN 0375-7633 doi:10.4435/BSPI.2017.31
Mammuthus meridionalis from Madonna della Strada (Scoppito, L’Aquila):
diagnostics and restoration
Maria Adelaide , Silvano , Maria Rita , Ivana , Salvatore ,
Filippo  , Elena , Federica , Gianmario , Paolo , Cristina ,
Lisa , Giorgio   & Emanuela 
M.A. Rossi, Soprintendenza archeologia, belle arti e paesaggio dell’Abruzzo, Via degli Agostiniani 14, I-66100 Chieti, Italy;
mariaadelaide.rossi@beniculturali.it
S. Agostini, Soprintendenza archeologia, belle arti e paesaggio dell’Abruzzo, Via degli Agostiniani 14, I-66100 Chieti, Italy; silvano.agostini@beniculturali.it
M.R. Palombo, Dipartimento di Scienze della Terra, Sapienza Università di Roma, Piazzale Aldo Moro 5, I-00185 Roma; CNR-IGAG, Via Salaria km
29.300, I-00015 Monterotondo (Roma), Italy; mariarita.palombo@uniroma1.it
I. Angelini, Dipartimento dei Beni Culturali, Università degli Studi di Padova, Piazza Capitaniato 7, I-35139 Padova, Italy; ivana.angelini@unipd.it
S. Caramiello, Soprintendenza archeologia, belle arti e paesaggio dell’Abruzzo, Via degli Agostiniani 14, I-66100 Chieti, Italy;
salvatore.caramiello@beniculturali.it


F. Marano, Dipartimento di Scienze della Terra, Sapienza Università di Roma, Piazzale Aldo Moro 5, I-00185 Roma; federica.marano86@gmail.com
G. Molin, Dipartimento dei Beni Culturali, Università degli Studi di Padova, Piazza Capitaniato 7, I-35139 Padova, Italy; gianmario.molin@unipd.it
P. Reggiani, Paleostudy, Via Zabarella 21, I-35028 Piove di Sacco (PD), Italy; paleostudy@libero.it
C. Sangati, AR Arte e Restauro srl, Viale Navigazione Interna 49, I-35129 Padova, Italy; sangati@arpadova.it
L. Santello, Dipartimento di Geoscienze, Università degli Studi di Padova, Via Gradenigo 6, I-35131 Padova, Italy; lisa.santello@unipd.it
G. Socrate, AR Arte e Restauro srl, Viale Navigazione Interna 49, I-35129 Padova, Italy; socrate@arpadova.it
E. Vinciguerra, AR Arte e Restauro srl, Viale Navigazione Interna 49, I-35129 Padova, Italy; emavinci@hotmail.it
KEY WORDS - Mammuthus meridionalis, mounted skeleton, L’Aquila, restoration, diagnostics, dynamic behaviour.
ABSTRACT - The skeleton of Mammuthus meridionalis (Nesti, 1825), found in 1954 at Madonna della Strada (Scoppito, L’Aquila, central
Italy) and exhibited since 1960 in the Spanish Fortress in L’Aquila, has recently undergone a complex diagnostic and restoration project
funded by the “Guardia di Finanza” as a contribution to the reconstruction of the city of L’Aquila after the earthquake of 6 April 2009. The
specimen was only slightly damaged by the seismic event. The restoration based on a highly integrated multidisciplinary approach, began with
a preliminary diagnostic analysis in order to acquire information on the state of preservation of the skeleton and on the dynamic behaviour
of the metal supporting frame-skeleton assembly. The restoration was carried out by combining the most common techniques applied in
Palaeontology with the most advanced techniques used for the restoration of cultural artefacts. The painstaking restoration enabled the
original shape of the skeleton to be restored, revealing the actual shape of the skull and its pathological modications. Before reassembling
the skeleton, a new, correct posture was also dened.
RIASSUNTO - [Mammuthus meridionalis di Madonna della Strada (Scoppito, L’Aquila): diagnostica e restauro] - Lo scheletro di
Mammuthus meridionalis (Nesti, 1825), sul quale venne istituita la sottospecie M. m. vestinus, fu rinvenuto nel 1954 in una cava di argilla in
località Madonna della Strada (Scoppito, L’Aquila). Il reperto, esposto dal 1960 nella Fortezza Spagnola dell’Aquila, leggermente danneggiato
dal sisma del 6 aprile del 2009, è stato recentemente oggetto di un complesso progetto di diagnostica e restauro nanziato dalla “Guardia di
Finanza” come contributo per la ricostruzione del patrimonio culturale della città. Un’accurata indagine diagnostica multidisciplinare è stata
condotta al ne di conoscere lo stato di conservazione dello scheletro, i prodotti utilizzati nei precedenti restauri e il comportamento dinamico
dell’insieme telaio metallico-scheletro. L’intervento di restauro è stato eseguito combinando le comuni tecniche applicate in paleontologia
con quelle più avanzate utilizzate per le opere d’arte. I risultati delle indagini diagnostiche hanno guidato l’intervento conservativo. Lo
scheletro è stato riportato al suo aspetto originale; il cranio, in particolare, è stato rimodellato sulla base della nuova supercie corticale
portata alla luce. Nell’alveolo di sinistra il premascellare ha rivelato segni di un importante evento traumatico. Inne, attraverso l’uso di un
modello tridimensionale è stata denita una nuova postura, che ha guidato le modiche del telaio e il rimontaggio dello scheletro.
INTRODUCTION
After the strong L’Aquila earthquake of 6 April
2009, the Italian “Guardia di Finanza” made a tangible
       ’s
earnings for the restoration of the local cultural heritage.
The choice fell on an important symbol of the L’Aquila
community, the Mammuthus meridionalis skeleton from
Madonna della Strada. The “Direzione Regionale per i
Beni Culturali e Paesaggistici dellAbruzzo” (MIBACT)
therefore launched a project to restore the skeleton and
reorganise the display area in the east bastion of the
Spanish Fortress where the mammoth has been displayed
since 1960.
The nearly complete specimen in a fairly good state
of preservation was discovered in 1954 in a sandy level
of an early Pleistocene fluvial-lacustrine succession
cropping out in the Santarelli clay quarry at Madonna
della Strada (coordinates WGS84: 42.348080, 13.256035)
(Scoppito, LAquila) (Maccagno, 1958, 1962; Mancini et
al., 2012; Agostini et al., 2014), together with fragmentary
remains of a small-sized rhinoceros (Stephanorhinus
aff. S. hundsheimensis Toula, 1902), hippopotamus
(Hippopotamus antiquus Desmarest, 1822) and a large
S. P. I.
S
O
C
I
E
T
A
'
P
A
L
E
O
N
T
O
L
O
G
I
C
A
I
T
A
L
I
A
N
A
Bollettino della Società Paleontologica Italiana, 56 (3), 2017
360
deer (Maccagno, 1962; Magri et al., 2010). From the
clays with lignite overlying the Santarelli quarry sands,
some remains of amphibians (Triturus sp.; Discoglossus
cf. D. pictus Otth, 1837; Bufo cf. B. viridis Laurenti, 1768;
Rana dalmatina Fitzinger, in Bonaparte, 1838) and reptiles
(Emys orbicularis Linnaeus, 1758) and a few remains of
the arvicolid Mimomys savini Hinton, 1910 have also been
reported (Kotsakis, 1988), together with terrestrial and
freshwater molluscs (Esu et al., 1992, 1993).
Palynological data coming from the clayey lignitic
deposits overlaying the sandy layer suggest an age of about
1.3 Ma, older than the Jaramillo Subchron and younger
than MIS 40, giving a chronological constraint for the
southern mammoth skeleton found in the underlain sandy
deposits (Magri et al., 2010).
Maccagno (1962) regarded the elephant as a typical
M. meridionalis, close to the Upper Valdarno specimens,
whereas Azzaroli (in Ambrosetti et al., 1972) considered
the Madonna della Strada specimen as an advanced
representative of the M. meridionalis lineage and erected
the new subspecies Mammuthus meridionalis vestinus
based on this individual (cf. Azzaroli, 1977). The taxon
was described as having a larger overall size with more
advanced cranial morphology, including a shorter deeper
skull, higher cranium with more caudally displaced vertex,
more concave forehead and long narrow tusk alveoli.
After a revision of the material ascribed to this subspecies,
Ferretti (1999) demonstrated that the dental characteristics
of M. m. vestinus fall within the morphological and
dimensional range of those from the rich Upper Valdarno
sample. The most characteristic features of M. m. vestinus
were considered to be the distinctive shape of the skull and


Palombo & Ferretti, 2005 for a discussion).
The diagnostic and restoration activities carried out
between 2013 and 2015 represented a unique opportunity
to improve conservation of the specimen and gain a better
understanding of its distinctive anatomical characteristics.

a multi-analytical investigation protocol was followed
involving detailed photographs of the skeleton taken
in both normal and ultraviolet light; a 3D survey of the

shift laser scanner technology; radiographs of parts of the
skeleton; themed mapping; mineralogical, textural and
chemical analyses to characterize the original materials
(bone and tusk) and the products used in previous
restorations; taphonomic analysis; structural analysis of
the dynamic behaviour of the supporting frame-bones
assembly, and monitoring of the microclimate (humidity,
temperature, dew point).
The choice of the technical and methodological
approaches to apply during the restoration performed
directly in the exhibition hall, was facilitated by thorough
knowledge of the characteristics of the specimen and of the
taphonomic processes undergone by the elephant skeleton.
TAPHONOMIC PROCESSES
A better understanding of the degradation processes
taking place at the time of the fossilisation was achieved
thanks to taphonomic analysis based on the description
of the fossiliferous layer, study of further, recently found
photographic documentation of the skeleton’s excavation
(Fig. 1), and new data on the depositional facies, patinas
and micro-morphologies present on the bones. The
skeleton was buried not far from the lake shore near a
lacustrine estuary mouth. The skeleton came to rest on a

environment. Sands, still present in the spongy tissue of
the bones (particularly those of the appendicular skeleton),
consist largely of quartz, feldspars, mica, calcite and
minute fragments of calcareous rocks.
During the first burial phase in shallow-water
environments (swamp and palustrine facies), oscillations
of the lake level alternated with moments of subaerial

oxide precipitation in hydromorphic conditions. The

context, while the bones on the right side remained
exposed for a longer time. Above the fossiliferous layer,
there is a distinct change of facies, characterised by strata
of slightly organic dark grey clay, covered by strata of
clay with layers of black and brown lignite deposited in
a high marsh environment. The top of the sedimentary
    
sands, with limited erosion of small channels in a higher
energy environment.
       
phase of burial, the surface of the bones underwent limited
Fig.1 - Madonna della Strada, 1954 excavation. The stratigraphic
section described in Maccagno (1962) can be seen in the
background. The geometries of the layers under and englobing the

361
M.A. Rossi et alii - Mammuthus meridionalis from Scoppito (AQ): diagnostics and restoration

emulsion of water, mud and sand. The abrasion traces
are short, discontinuous with various orientations and
not longer than a centimetre. In addition, the alternation
of reducing and oxidising conditions produced the light
brown colour of the bones and more or less extensive
areas (particularly on the surface of the limbs) covered
by orange red patinas caused by the presence of iron
hydroxides formed by oxidation of the newly formed
pyrite. A number of holes of about a millimetre in size,
concentrated in irregularly shaped areas smaller than 10
cm², have been interpreted as associated with the action
of small plant roots in the damp or marshy ground. The
orange red patina is overlaid by a discontinuous black
patina showing in some cases lobed margins, which may
have been produced by contact with clays with a moderate
organic content and/or direct contact with algal mucilage.
In short, the almost complete presence of the entire
skeleton and the modest degree of erosion of the bone
indicate that burial occurred in a low-energy environment.
Moreover, with the exception of the skull, post-burial

fractures of the skeleton bones.
ANALYSIS OF PREVIOUS RESTORATIONS
After recovery, the skeleton was restored and studied
at “La SapienzaUniversity of Rome (Maccagno, 1958,
1962), then exhibited in the east bastion of the Spanish
Fortress in LAquila (Fig. 2). The restoration as described
in the Maccagno report (1958) included emptying of
sand from the cavities, internal reinforcement with a
structure of iron bars and consolidation using mastic. As
far as possible, the missing parts were reconstructed with
reference to the symmetrical bone and comparison with
homologous structures. The report does not, however,
mention either the type of mastic, or the type of material
used to reconstruct the missing parts. Each reconstructed
part was made to resemble the original as closely as


In 1987, the skeleton (with the exception of the skull
and pelvis) was again disassembled and transported to the
University of Florence for restoration, concluded in 1991.
The restoration involved cleaning and removal of part of
the mastic applied during earlier restoration, consolidation
with polyvinyl acetate and bonding with reversible glues
as described in a similar operation (Borselli et al., 2002).
Work on the cranium and pelvic bones was carried out on
site. Here too a special masticwas used to reconstruct
the missing or erroneous reconstructions, but again no
information is given concerning the type of mastic.
The right tusk (weighing about 150 kg) was remodelled
and reproduced with epoxy resins and polyurethane foam
(weighing 10 kg), then located in place of the original tusk,
which was rested at the animal’s feet (Fig. 3).


easy to scratch with scalpels, widely used in fossil

by a fragment of newspaper dated 11 December 1905
soaked in this material and found in the humeral epiphysis
of a fossil elephant from the Balzi Rossi site (Ventimiglia,

        
plaster-of-Paris and 2% rosin (Fabio Cozzini, personal
communication). It was prepared by melting the beeswax

and zinc oxide. This material was also used to restore the
limb long bones and tusks of a mammoth discovered in
1973 at San Giovanni di Valdobbiadene (Vidor, Treviso)
(Reggiani & Sala, 1992). As in Maccagno (1958), also in
this case the preparation involved removing the inner part
of the bone to create space for insertion of a metal armature
        
give the fossil solidity (Borselli, 1989).
MAPPING
In line with the general principle of non-invasive
pre-restoring diagnosis (Brando, 2000), the adopted
        
surfaces to identify the current state of all the osteological
Fig. 2 - First display of Mammuthus meridionalis in the Spanish
Fortress in L’Aquila in 1960.
Fig. 3 - The specimen after the second restoration in 1991.
Bollettino della Società Paleontologica Italiana, 56 (3), 2017
362
remains. Photographs in visible (RGB) and ultraviolet
(UV) light highlighted critical areas on the periosteum
(particularly fragmented elements, previous anatomical
reconstructions, pre-fossilization mineral absorptions,
etc.) and allowed the internal osteological structure to
be characterised using X-ray images. The qualitative
maps produced before the restoration thus involved a
combination of detailed optical inspection, high resolution
photos and X-ray plates.
The outcome also guided the following phase of
chemical and physical diagnostics and defined the
restoration plan, identifying areas at risk, breakage
included. The maps were also used while dismantling,
moving and reassembling the bones and for structural
evaluations. They were updated and enhanced constantly
as the amount of recorded data increased, especially when
the coat of paint was removed to reveal new evidence and
with the results of the diagnostic analyses.
All the analyses were performed using 156 photographs,
in RGB and in UV light, and 45 X-ray plates. Mapping
was carried out using AutoCAD-LT-2013. Detailed
information about the methodological approach is
provided in the Supplementary Online Material.
Results
The flat surfaces of the bones and especially the
vertebrae and coxae were covered with a hard coat of dust.
Bones are impregnated with iron minerals accumulated
as a consequence of the concentration of these minerals
in the soil in the peri-depositional context. The scattered
spots are visible over the entire skeleton and on the
diaphysis of the limbs and are particularly noticeable on
the subscapular fossa. Corresponding to these mineralized
areas, there were broken, although stable, bone surfaces,
most frequent on the articular processes and on thin bones
such as the tusk alveoli and the scapula. The trabecular
area near the epiphyses was stable and associated with
post-depositional events and the age of the mammoth at
the time of death.
      
materials used for the reconstructions (Fig. 4). UV light
   
improving remodelling of the lost anatomies, and the
 

for the reconstruction of the mammoth and to reproduce
its phalanxes.
The body is known to have been lying on its left side
when discovered and the right side was probably partially
destroyed (Maccagno, 1958). As a consequence, this side
of the animal is deteriorated with more reconstructions:
the right occipital bone and part of the skull, the right
hemimandible, part of the iliac fossa, the proximal
and distal epiphyses of the femur, humerus and ulnae
and most of the scapula and ribs. The left side is better
preserved, although pivots and metal armatures were
inserted to support the largest bones (skull, mandible,
scapula, long bones, coxa and some vertebrae), on
fragile structures (left zygomatic arch) and to support
the diaphysis (Fig. 5).
Finally, the fossils were partly covered with organic-
synthetic material (glue and epoxy resin), principally
corresponding to the reconstructions and showing on the
UV images as intense yellow and cyan blue (Tab. S1, Figs
S1, S2 in the Supplementary Online Material).
Remarks
X-ray analysis is commonly used in palaeontology
and palaeoanthropology to investigate trabecular structure
and osteological resistance and to recognize pathologies,
traumatic events or bone growth (Franzen et al., 2009;
Blondiaux et al., 2012; Leden et al., 2012; Sami & Ghezzo,
2015; Germonpré et al., 2016). UV images have been
widely used on pre-Quaternary fossils in a sedimentary
matrix. These techniques are used, for example, to
recognize colour patterns on shells (Hendricks, 2015),
different materials in fossils and previous surface
restorations (presence of organic glues or fakes) (Hone
et al., 2010; Dal Sasso & Maganuco, 2011; Stone, 2010).
They were recently applied to Jurassic fossils, revealing
the microstructure of feathers in non-avian theropods
and their evolutionary trends (Tischlinger, 2002, 2005;
Tischlinger & Unwin, 2004; Hone et al., 2010).
Both techniques are less frequently used in
palaeontology for pre-restoration analysis and UV
Fig. 4 - (color online) Right femur in frontal-lateral view. a) UV
photo (grey scale / RGB); b) mapping showing the coating of dust
(dark grey / light blue), impregnations (black and oblique lines /
brown and red) and surface fractures (simple black / fuchsia); c)
reconstructions (white / pink) and metal pins revealed by X-ray
(black / black rectangles) (see Tab. S1 in the Supplementary Online
Material for more details on the mapping).
Fig. 5 - Left mandible (left) and X-ray of the mandible (right) with
metal supports and reinforcement of the ascending ramus (as listed
in Tab. S1 in the Supplementary Online Material), with complete
reconstruction of the apical articulation (condyle).
363
M.A. Rossi et alii - Mammuthus meridionalis from Scoppito (AQ): diagnostics and restoration
photography is rarely used on relatively recent remains. In
the case of the mammoth discovered in 1954 at Madonna

areas to guide sampling for mineral and chemical analysis
(a method previously used by Riquelme et al., 2009).
Mapping generally proved to be useful in acquiring
information on the surface condition of the skeleton,
identifying all criticalities, including reconstructions,
fractures and the presence of sporadic organic phase

COMPOSITIONAL AND
TEXTURAL INVESTIGATIONS
Before proceeding with the restoration, a campaign

the mineralogical and chemical characteristics of the
constituent materials of the specimen.

phase, following meticulous morphological observation, a
wide range of surface samples considered representative
of the specimen as a whole were taken. The composition
and texture of the original materials (bones and tusk) were
determined, together with those of the materials used in
the previous restorations (reconstructions, resins, patinas
and paints) as their composition was not fully known

was aimed at acquiring information on the complexity of
the specimen to ensure that, throughout the restoration,
all actions and methodological choices were guided
        
acquired results, a second phase was carried out to acquire
detailed information on the composition and texture of the
three long bones chosen as representative (humerus, tibia
and ulna). This involved core sampling the bones. The
aims were to provide the structural engineer with data to
evaluate the mechanical resistance of the bones and the
restorer with information to guide the conservation and
consolidation work, including evaluation of whether or
not to replace the synthetic materials applied previously.
The analyses were performed generally on small samples
measuring about 0.5 mm3. Preliminary stereomicroscopic
study of the samples was followed by detailed study by
optical microscopy (OM). Furthermore studies were
performed using complementary diffractometric and
spectrometric techniques: X-ray powder diffraction
(XRPD) and infrared (FTIR and µ-FTIR) and µ-Raman
spectroscopy. Elemental analysis was carried out by
X spectroscopy using a scanning electron microscope
(SEM) coupled to an energy dispersive spectrometer
(EDS). Textural analyses, particularly important in light
of the project’s objectives, were carried out with both
OM and SEM.
The methodologies adopted are provided in the
Supplementary Online Material whereas the principal
results obtained are summarised below, in order to
contextualise the study campaign prior to the restoration
of the mammoth.
Results
 - As a whole, the histologic characteristics
of the original bones are more or less well preserved.
Porosity can be attributed to the texture of the bone
tissue, found to consist largely of carbonate-apatite,
generally not (or only slightly) replaced by precipitation
of other mineral phases during diagenesis (Fig. 6). In a
number of areas, “mechanical replacement” was found
to have occurred during fossilisation, with cementation
of angular arenaceous clasts from the sediments of the
fossil bed, almost like a “natural integration”. In those
cases, the presence of quartz, calcite, dolomite, feldspar
and ferrous mica granules was observed, often cemented
by calcite. Importantly, pyrite (FeS2) and iron oxides and
hydroxides (haematite, goethite and limonite, sometimes
pseudomorphs on pyrite) deriving from alteration of the
pyrite in a damp oxidising subaerial environment were
found in the pores and inter-tissue spaces (Fig. 7 and
Fig. S3 of the Supplementary Online Material). Pyrite is
therefore the principal mineral forming during diagenesis.
      
its oxidation and hydration products in the bone tissue

demonstrates that the carcass was buried and preserved
in an anoxic depositional environment characteristic
of lacustrine and paludine basins. The presence of
         
(xNicols). Thin section of the core of the right ulna: detail of an
osteon in the well-preserved bone tissue. Scale bar corresponds
to 1 mm.
         
Thin section of a bone sample from the right tibia. The apatite bone
tissue is embedded in iron oxide and hydroxide, responsible for the
orange colour. Secondary depositions of these phases are visible in
the central large pore. Scale bar corresponds to 1 mm.
Bollettino della Società Paleontologica Italiana, 56 (3), 2017
364
pyrite crystals also indicates that the relative process of
alteration through oxidation and hydration is still active
and requires accurate evaluation, given the risks for
long-term conservation (Turner-Walker, 2009). These
transformations, in fact, result in a notable increase in
volume (leading to the disintegration of the specimen),
with the creation of a markedly acid environment (due to
production of sulphurous-sulphuric acid) which in turn
encourages the transformation from calcite to gypsum with
a further increase in volume. The products applied during
previous restorations, such as polyvinyl acetate (PVAc),

  The examined samples revealed a compact,
well-organised structure of dentine fibres (ivory)
   
and texture of the internal sample were excellent, less
so the external sample where the presence of interstitial
calcite precipitated during diagenesis was found. The

of minimal quantities of iron oxides/hydroxides, as also
found on the bones.
 Unfortunately,
no detailed description remains of the composition and
location of the numerous reconstructions added to the
bones in previous restorations. A study therefore had to


reconstructions and dark reconstructions. Both were found
to have moderate plasticity and to contain a number of
minerals, with bassanite, gypsum, anhydrite and calcite
predominating. They also frequently contained white
pigments such as zincite and to a lesser extent barite.
Importantly, the organic component consisted mainly of a
polyester type compound in the case of the dark-coloured
       
case of the light-coloured reconstructions and patinas.
Five samples with a “resinous” appearance were also
taken, considered as representative of the variable textures
and colours observed for this type of material. Their
composition was found to be very varied and multiphasic,

had been involved in previous restorations. They range,
in fact, from a calcite and gypsum based filler with
organic aliphatic binder used for the tusk, to bassanite
with beeswax binder for the reconstructions on the right
tibia. PVAc and phenoxy and epoxy resins revealed also
to be present.
 During previous restorations, the reconstructed
areas had been painted to rectify their whiteness and blend
them in with the basically ochre colour of the bones. Once
       
were observed, generally involving application of a
mineral fraction consisting of a light-coloured gypsum-
calcite-barite based dispersant and rutile and titanite based
pigments. The organic fraction was found to consist of

 Core sampling
enabled the texture and composition of the three selected
bones (tibia, ulna and humerus) to be studied in detail,
together with the nature of the consolidants used in
previous restorations and the outcome of consolidation. As
regards the macroscopic texture, all the bones were found
to have retained the original histologic characteristics.
The periosteum of the bone was found to be compact,
becoming progressively more spongy towards the
endosteum. The microtexture is perfectly preserved, with
well-organised osteons and the Haversian canals generally
     
oxides-hydroxides deriving from alteration of the pyrite
deposited in considerable quantities during diagenesis.
By far the most predominant mineral component is
carbonate-apatite, evidence of the low degree of mineral
replacement during fossilisation. Of particular importance
      
which both give the specimen its ochre colour and also

found in pores and cracks (Fig. 7) can be attributed to
the presence of quartz and feldspars deriving from the
surrounding sands.
In previous consolidations, epoxy resin (araldite) had
been used, including to reconstruct the missing internal
parts. Study of the core sampled bones showed that this
may not have involved all the bones in the skeleton and
that the degree of penetration of the resin into the porous
bone was not homogeneous.
THE DYNAMIC BEHAVIOUR OF THE SKELETON
The bearing structure consists of an iron frame,
moulded in the 1950s by a local blacksmith. This can be
roughly divided into “primary” elements (leg supports,
hollow pipes supporting the spine, a pole supporting the
skull) and “secondary” elements (hooks connecting the
bones to the primary elements).
The on-site survey, consisting in ambient dynamic

showed the great deformability of the metal frame, with
main periods in the order of 1.8 to 1.3 s. It was, in fact,
possible to move the whole supporting frame virtually

direction, demonstrating physically that this was the
predominant mode shape in the dynamic behaviour
(Casarin et al., 2015).
The Finite Elements (FE) model of the mammoth
was constructed using the 3D geometric model obtained
with the laser scanner technique (Figs S4 and S5 in the
Supplementary Online Material). The FE model was
calibrated with the results of the dynamic tests. Four/
    
       

the frame (Figs S6 and S7 in the Supplementary Online
Material). The dynamic identification tests clearly
       
      
and longitudinal directions.
Linear static analysis was carried out in order to assess
the stress levels in the elements of the metal frame. The
only force considered for this analysis was self-weight.
All details of the applied methodology are provided
in the Supplementary Online Material.
365
M.A. Rossi et alii - Mammuthus meridionalis from Scoppito (AQ): diagnostics and restoration
Results
Analysing the axial stresses of the beam elements
of the frame, the most stressed elements were found to
be the vertical supports of the structure (leg supports)
which transmitted the entire weight of the skeleton to the
ground. The skeleton was not found to be subjected to
evident stresses (except in a number of contact points due
to singularity of the model), indicating that the stresses
were correctly transferred to the steel frame.
Maximum stresses - measured in the pole below the
skull - were approximately 40 N/mm2. The axial forces
on the frame elements are presented in Fig. 8.
Frequency analysis indicated the presence of well-

related to a participating mass of 84.5% in the x direction,
namely transversal to the mammoth, and 82.9% in the y,
or longitudinal, direction (Tab. S2 in the Supplementary
Online Material, Fig. 9).
Spectral analysis was carried out using two real elastic
response spectra (5% damped) derived from the main
shock acceleration record measured at the INGV seismic
station AQU at the same time as the earthquake at 03.32
on April 6, 2009 (Çelebi et al., 2010). The two spectra,
in orthogonal north-south and east-west directions,
were combined with the self-weight action (linear static
analysis) to simulate the earthquake of April 6, 2009 and
assess the most stressed frame elements.
       

allowed the frame freedom of motionto mitigate the
horizontal seismic force, which contained peaks in the
response spectrum of approximately 0.8 g at a frequency
of 10 Hz and measured PGA of 0.31 g.
Seismic analysis therefore indicates that the steel

the relatively limited inertial forces involved. However,
as regards the pole supporting the skull, the section has

and axial loads, according to the Italian national standard
(NTC, 2008).

frame and skeleton emerging from the analysis (Fig. 10),
in the range of 0.2 m or more in a transversal direction and
demonstrated indirectly after the 2009 earthquake, have
in reality put a considerable strain on the bones.
Remarks
In the case of the mammoth preserved in the Spanish
Fortress in L’Aquila, an area with high seismic activity,
the structural stability of the skeletal support frame plays a
very important role, and it was therefore decided to study
       
to the dynamic behaviour of the frame in relation to the
earthquake of 2009.
Considering both numerical simulations and
experimental activities, the studies highlighted the peculiar
structural feature which had “saved” the mammoth from
the earthquake, namely the high deformability of the metal
     
attained during the seismic motions, and this presumably
caused damage to the vulnerable bones of the mammoth.
THE RESTORATION
       
restoration lies in the application of practices typical of the
conservative restoration of works of art to the restoration
of a palaeontological specimen, respecting the intentions,
while at the same time enhancing them with the sensitivity
and technical approach of the art restorer.
As with restoration of a cultural artefact where
movement is complicated by dimensions, weight and
fragility, the following were temporarily installed in the
bastion room transformed into a fully equipped restoration

with all the operational necessities of the restoration,
with wheels to facilitate movement, movable shelves
and a top crossbeam with electric hoist; the equipment
and instruments required to perform all phases of the
restoration in safety; a lift platform with double pulley
adequate for moving all the bones; an extraction system

work stations for the various phases to rationalise and limit
risky handling of the bones (Fig. 11).
Fig. 8 - (color online) Axial forces on the supporting frame.

modes (in uniform grey the undeformed shape).
Bollettino della Società Paleontologica Italiana, 56 (3), 2017
366
All details of the applied methodology are provided in
the Supplementary Online Material. The chronologically
ordered list and a synthetic description of the operations
performed are provided below.
1. The bone elements were disassembled in three

support systems were designed and modulated according
to the anatomical characteristics and fragility of the
various elements (Fig. S8 in the Supplementary Online
Material). Modular systems were used for the limbs, with

S9 in the Supplementary Online Material).
2. Removal of paint from the surfaces of the
reconstructions and bones with solvent (acetone).
3. Cleaning of the original parts with removal of
the previous protective agent from the surface: the pure
solvents and solvent mixtures were tested. The analytical
results (FTIR) showed that cleaning with acetone
completely removed the PVAc present in considerable
quantities on the bone surfaces, without signs of
“transport” dynamics in depth.
4. Removal of widespread plaster residues not
removed after recovery of the skeleton in 1955 and mastic
residues left from previous restorations necessitated the
use of a precision micro-sandblaster (Fig. S10 in the
Supplementary Online Material).
5. Removal of excess filler and identification of
the contact surfaces between the original elements and
reconstructions.
6. Pre-consolidation of a number of particularly
fragile portions and consolidation by immersion (Fig.
S11 in the Supplementary Online Material). The choice
fell on suitable acrylic products (Keene, 1987; Shelton
& Chaney, 1994; Linares Soriano & Carrascosa Moliner,
      
weight, minimum molecule size, low viscosity and high

chosen was Degalan LP 64/12, an acrylic polymer which,
unlike the more common Paraloid B72, had all the above
characteristics, while Butyl acetate solvent was preferred
to acetone for its lower volatility.

gluing.
   
the supporting frame: XYZ displacements bending in transverse
direction.
Fig. 11 - The exhibition hall “transformed” into a restoration laboratory.
367
M.A. Rossi et alii - Mammuthus meridionalis from Scoppito (AQ): diagnostics and restoration
8. It was decided to modify a number of the anatomical
parts reconstructed during previous restorations as
these did not perfectly correspond to the anatomy of
the animal. For these new reconstructions, a light two-
component epoxy-based putty, reversible in polar solvents,
commercially known as Balsite was used.

addition of 5% weight of a mixture of natural coloured

10. Colouring of the reconstructed portions. After

were used.
Results
Observation of the bones after their removal from the
metal structure revealed the presence of numerous cracks,
particularly in jointed elements and above all in the sacrum

the earthquake. The degradation was found to be worse on
bones where the spongy part was exposed, due to a lack
of cortex. Parts of the skull and pelvis in direct contact

were in a very critical state of preservation.
Cleaning of the bones revealed widespread plaster
residues, previously covered with a thin layer of colour.
Removal of these with a precision micro-sandblaster
      
morphology of the bones, in respect of the original details
and colours. The cemented sand from the stratum in which
fossilisation had taken place was not, on the other hand,
removed.
Lowering of the edges of the reconstructions covering
large portions of the bone and tusk allowed more of the
original surfaces to be exposed.
There were numerous criticalities associated with the
pelvis which required considerable pre-consolidation
before proceeding with the consolidant bath.
Numerous tests were carried out on the extensive
reconstructed areas of the cranial vault, revealing the
presence of underlying bone material. The mastic, varying
in thickness from about 5 to 15 cm, was found to cover a
greatly altered and fragile cortical surface. Precautionary
pre-consolidation had therefore to be performed on all
      
mapped previously.
After consolidation by immersion, there was still
     
and study, these were held to be stable and basically
       
liquid epoxy resin and insertion of two 3 mm diameter

For the anatomical reconstructions of the front wing
of the left scapula, olecranon of the left and right ulnae,

chapter.
The chromatic treatment of the reconstructions was
aimed at enhancing the appearance of the skeleton as
a whole. The stippling technique adopted recreated an

respecting the basic restoration principle according to
which each additional part must be distinguishable from
the original, without disturbing the appearance of the
artefact as a whole (Carta CNR, 1987).
Remarks
With respect to previous interventions, this conservative
   
of the bone/reconstruction contact points in order to
increase the percentage of exposed original surfaces. The
restoration gave back the bones their original appearance
       
enhanced the artefact, in respect of the restoration
principles (Carta CNR, 1987; Borselli et al., 1998).
GETTING BACK THE ORIGINAL SHAPE
OF THE BONES: THE MAIN CHANGES
Even though the Madonna della Strada skeleton
had been accurately restored and the broken/missing
parts accurately integrated under the supervision of
A.M. Maccagno (Maccagno, 1954, 1962), the shape of
a number of skeletal elements appeared unusual with
respect to the shapes characterising southern mammoths
and proboscideans in general. Among the reconstructed
bones, the ulnae appeared the most anomalous. At the
top of the proximal epiphysis, above the olecranon
process, the outline of the ulna was not, in fact, gently
curved, as in nearly all proboscideans and certainly in
the Elephantinae, but characterised by a marked bulge,
nearly simulating a further olecranon (Fig. 12a). When
the material added during the restoration was removed
from the right ulna, it became clear that the erroneous
reconstruction had been based on the presence of a long
       
during the diagenetic process. In the left ulna, the entire
proximal part had been completely reconstructed. Both
ulnae were therefore brought back to their original,
correct shape (Fig. 12b).
However, the most important changes concerned
      
implications. As reconstructed by Maccagno (1958,

than the M. meridionalis skulls from Valdarno (see
Azzaroli, 1966, 1977; Ambrosetti et al., 1972), the medial

anteroposteriorly and the nuchal fossa is narrow and quite

with the morphology required to support the insertion
      
and nuchae and lateralis ligaments (e.g., Marchant &
Shoshani, 2007), as required to support the weight of
the large tusks, and balance the skull when a tusk was
lost (see below). This observation, coupled with the fact
that the skull dome had been partially preserved when
the skeleton was discovered (as documented by pictures
taken during excavation and recovery; Fig. 13), while after
restoration it appeared completely reconstructed, led to
the decision to remove the mastic to reveal the real shape
of the preserved skull apex. The result was surprising, as
the nearly completely preserved left side indicated that
the skull was much shorter than previously reconstructed
     

Balsite was used to reconstruct the missing part of the
right side of the skull, taking special care not to cover the
original fossil portions.
Bollettino della Società Paleontologica Italiana, 56 (3), 2017
368

Scoppito skull and the Valdarno skulls is the more concave
forehead (although this was reconstructed by juxtaposing
a number of bone fragments whose actual position is
unknown) and the very long premaxillary bone, apparently
slightly converging downwards and with a deep furrow
(Azzaroli, 1966, 1977; Ambrosetti et al., 1972). It is,
however, worth noting that the tusk alveoli of the Scoppito
M. meridionalis are clearly asymmetric, even considering
the incompleteness of the one on the right. The left is
narrow, thicker and latero-medially compressed, whereas
at the distal aperture the external portion of a broken tusk
is still preserved inside the alveolus. The peculiar shape
suggested removal of the thick layer of mastic covering
the premaxillary bone dorsally to investigate the original
shape and possible presence of deformities. When the
original surface was brought to light, a large hole (Pl. 1,
       
suggesting a pathological condition or traumatic injury
causing deformation of the alveolus, a reduction in the
width of the medial furrow between the alveoli and
possibly loss of the tusk (Della Salda et al., 2016).
The new information collected during the restoration
and new shape obtained for the skull, much closer to the
real one than that proposed by the previous restoration,
further support the hypothesis that the validity of M.
meridionalis vestinus as a sub-species separate from the
type species M. meridionalis should be considered with
great caution (see e.g., Palombo & Ferretti, 2005).
LOOKING FOR A CORRECT POSTURE
Elephants have generally been seen as the archetype
for columnar graviportal animals(see Ren et al., 2008 and
references therein), with long pillar-like limbs remaining
nearly extended even during gait, while the body axis
remains more or less horizontal. During the last couple
of decades, new studies on the locomotor kinematics of
elephants have questioned the idea that the functioning

substantially from the more flexed limbs of running
mammals (e.g., Hutchinson et al., 2006; Weissengruber et
al., 2006; Ren et al., 2008; Genin et al., 2010). Although
the actual possibility of the movement of limb segments
and joints during locomotion in extant elephants is still
an open question (see e.g., Hutchinson, 2009; Paul, 2009
and references therein), it is undeniable that in elephants
Fig. 12 - Right ulna before (a) and after (b) the last restoration: the anomalous reconstruction of the top of the proximal epiphysis (olecranon
process) has been brought back to its original shape.
EXPLANATION OF PLATE 1
Comparison of the skull of Mammuthus meridionalis before and after the last restoration. Scale bars correspond to 10 cm.
Fig. 1a, c - Anterior and left lateral view of the skull before the last restoration, as it was reconstructed by Maccagno.
Fig. 1b, d - Anterior and left lateral view of the skull after the last restoration. The bone lesion in the left tusk alveolus is clearly detectable.
Fig. 1e - The nuchal fossa, as reconstructed by Maccagno, was narrow and quite deep.
Fig. 1f - Posterior view of the skull, showing the new reconstruction of the nuchal fossa, wider and less deep.
369
M.A. Rossi et alii - Mammuthus meridionalis from Scoppito (AQ): diagnostics and restoration Pl. 1
Bollettino della Società Paleontologica Italiana, 56 (3), 2017
370
the acropodia (manus and pes) can neither be positioned
too far apart from the sagittal plane, as was the case in the

2), nor in line with the glenohumeral joint or closer to the
sagittal planes as they were positioned after the restoration.
We therefore analysed the position of the feet in the
largest African and Asian elephants and painstakingly
evaluated the accuracy of joint manoeuvrability while
positioning each limb segment to ensure the Scoppito
mammoth was positioned in a correct posture. The position
of the metapodials and phalanges was established taking
into account the angled foot structure of elephant feet,
due to the presence of large subcutaneous cushions. These
     
filled with adipose tissue, allowing the body weight
and mechanical forces to be absorbed (Weissengruber
& Forstenpointner, 2004; Hutchinson et al., 2006).
They represent an advanced feature of subunguligrade
elephantiforms with respect to the most primitive
plantigrade proboscideans (Hutchinson et al., 2011).
Basing on these observations, a number of potential
        
models in order to correctly make any change necessary on
the metal structure supporting the skeleton. The armature

spinal column was lengthened by 6 cm, the limb joints were
repositioned to keep the articular surfaces in an anatomical
position and the angle between the scapula and humerus
on the left forelimb was reduced (Fig. 14).
Some improvements were also made to the metal
structure. In particular, polyethylene shock absorbers
(Pasiuk, 2004; Del Favero, 2007) were inserted to separate
the vertebral endplates and between the support elements
and bones, the cervical vertebrae supports were lightened

the thoracic and lumbar vertebrae and hooks supporting


Fig. 13 - The skull as it appeared in 1954, just after its recovery.
Fig. 14 - The Mammuthus meridionalis from Madonna della Strada as it appears after the 2015 restoration.
371
M.A. Rossi et alii - Mammuthus meridionalis from Scoppito (AQ): diagnostics and restoration
At the end of the complex reassembly operations, the
static and dynamic behaviour of the skeleton-armature

CONCLUSION
The numerous and complex issues emerging during the
restoration design phase necessitated the involvement and

of diagnostic methods typically applied to cultural
artefacts improved our understanding of what had been
done in the past and led to a more precise evaluation
of the actual state of preservation of the specimen. The
most suitable restoration techniques were chosen on the
basis of the results obtained from the diagnostic studies.
Particular attention was paid to the bone parts hidden
by the mastic. Exposing them improved knowledge of
       
and revealed interesting pathological aspects. Through
suitable chromatic treatment of the reconstructions, the
conservative restoration further enhanced the specimen.
Analysis of the dynamic behaviour of the frame-
skeleton assembly showed the metal structure to be
highly deformable, a characteristic which had prevented

earthquake of 6 April 2009. To avoid and mitigate the

were inserted between the support elements and bones
and between the vertebrae.
In conclusion, the project, implemented in respect
of the principles of restoration, was based on accurate
diagnostics and the use of advanced techniques allowing
optimum conservative restoration and improving the
appearance and visibility of both the original and
reconstructed parts of the specimen.
SUPPLEMENTARY ONLINE MATERIAL
All the Supplementary data of this work are available
on the BSPI website at http://paleoitalia.org/archives/
bollettino-spi/
ACKNOWLEDGEMENTS
         
project, the “Direzione Regionale Beni Culturali per l’Abruzzo”
for promoting and managing the restoration, the “Soprintendenza
Archeologia, belle arti e paesaggio per la città dell’Aquila e i comuni
del cratere” now responsible for administering the project.

work meetings at the beginning of the restoration, to Fabio Cozzini for
the information provided to P. Reggiani concerning the material used
during the restoration performed in Florence, to Gianfranco Callori and
Cecilia Santarelli for the historic photograph of the site and to Mauro
Vitale for the photographs taken during the restoration. Finally our
thanks to the Reviewers Marco Ferretti and Emanuel Tschopp and the

REFERENCES
Agostini S., Rossi M.A. & Tallini M. (2014). Geologia e
paleontologia del Quaternario nel territorio aquilano. Atti del
Convegno “Nei dintorni di L’Aquila. Ricerche archeologiche
nel territorio dei Vestini Cismontani prima e dopo il terremoto
del 6 aprile 2009”, Roma, 12-13 Febbraio 2010. Collection de
l’Ecole Francaise de Rome, 494: 7-19.
Ambrosetti P., Azzaroli A., Bonadonna F.P. & Follieri M. (1972).
A scheme of Pleistocene chronology for the Thyrrhenian side
of central Italy. Bollettino della Società Geologica Italiana,
91: 169-184.
Azzaroli A. (1966). La valeur des caractéres craniens dans la
   Eclogae Geologicae Helvetiae,
59: 541-564.
Azzaroli A. (1977). Evolutionary patterns of Villafranchian
elephants in Central Italy. Atti dell’Accademia Nazionale dei
Lincei, Memorie Classe Scienze Fisiche, 14: 149-168.
Blondiaux J., Fontaine C., Demondion X., Flipo R.M., Colard T.
Mitchell P.D., Buzon M. & Walker P. (2012). Bilateral fractures
of the scapula: possible archeological examples of beatings
from Europe, Africa and America. International Journal of
Paleopathology, 2: 223-230.
Bonaparte C. L. J. L. (1838). Amphibiorum Tabula Analytica. Nuovi
Annali delle Scienze Naturali. 1: 394-397.
Borselli V. (1989). Il mammut di Vidor. In Gasparini D. (ed.), Due
villaggi della collina trevigiana Vidor e Colbertaldo. Comune
di Vidor: 161-181.
Borselli V., Confortini F., Dal Sasso C., Malanni M., Muscio G.,
Paganoni A., Simonetto L. & Teruzzi G. (1998). Carta del
restauro dei fossili. Museologia Scientica, 15: 215-226.
Borselli V., Cozzini F. & Landucci F. (2002). Il restauro
dell’elefante fossile (Elephas antiquus italicus) esposto nel
museo di storia naturale di Genova. Museologia Scientica,
17: 75-86.
Brando C. ed. (2000). Teoria del restauro.154 pp. Piccola Biblioteca
Einaudi, Nuova Serie.
Carta C.N.R. (1987). Carta 1987 della conservazione e del restauro
degli oggetti d’arte e di cultura. Roma, Consiglio Nazionale
delle Ricerche.
Casarin F., Beccaro E., Fattoretto M., Girardello P. & Caldon M.
(2015). The dynamic behavior of mammoth in the Spanish
fortress, L’Aquila, Italy. MATEC Web of Conferences, 24,
2015 - 05001, EVACES’15, 6th International Conference
on Experimental Vibration Analysis for Civil Engineering
Structures: 207-214.
Çelebi M., Bazzurro P., Chiaraluce L., Clemente P., Decanini L.,
De Sortis A., Ellsworth W., Gorini A., Kalkan E., Marcucci
S., Milana G., Mollaioli F., Olivieri M., Paolucci R., Rinaldis
D., Rovelli A., Sabetta F. & Stephens C. (2010). Recorded
motions of the 6 April 2009 Mw 6.3 L’Aquila, Italy, earthquake
and implications for building structural damage: overview,
earthquake spectra. Earthquake Engineering Research Institute,
26: 651-684.
Dal Sasso C. & Maganuco S. (2011). Scipionyx samniticus
(Theropoda: Compsognathidae) from the Lower Cretaceous
of Italy. Osteology, ontogenetic assessment, phylogeny, soft
tissue anatomy, taphonomy and palaeobiology. Memorie della
Società Italiana di Scienze Naturali e del Museo Civico di Storia
Naturale di Milano, 37: 1-281.
Del Favero L. (2007). La conservazione preventiva delle collezioni
geopaleontologiche. Museologia Scientica n.s., 1: 57-67.
Della Salda L., Rossi M.A., Agostini S., Cuomo A., Palombo M.R.
& Fornaciari G. (2016). Paleopathological study of Mammuthus
meridionalis of Madonna della Strada (L’Aquila). II Meeting
Nazionale Gruppo Italiano di Paleopatologia, L’Aquila 31
ottobre 2015. Patologica, 107: 215-216.
Desmarest A.G. (1822). Mammologie ou description des espèces
de mammifères. Mme Veuve Agasse imprimeur édit., 2: 277-
555.
Esu D., Girotti O. & Kotsakis T. (1992). Molluschi e vertebrati di
alcuni bacini continentali dell’Appennino centrale: indicazioni
Studi Geologici Camerti v.s.
1992/2 CROP 11: 295-299.
Bollettino della Società Paleontologica Italiana, 56 (3), 2017
372
Esu D., Girotti O. & Kotsakis T. (1993). Palaeobiogeographical
observations on Villafranchian continental molluscs of Italy.
Scripta Geologica, 2: 101-119.
Ferretti M.P. (1999). Mammuthus meridionalis (Mammalia,
Proboscidea, Elephantidae) from the ‘‘Sabbie Gialle’’ of Oriolo
(Cava La Salita, Faenza, Northern Italy) and other European
late population of southern mammoth. Eclogae Geologicae
Helvetiae, 92: 503-515.
Franzen J.L., Gingerich P.D., Habersetzer J., Hurum J.H., von
Koenigswald W. & Smith H. (2009). Complete primate skeleton
from the Middle Eocene of Messel in Germany: morphology
and paleobiology. PloS ONE 4: e5723.
Genin J.J., Willems P.A., Cavagna G.A, Lair R. & Heglund N.C.
(2010). Biomechanics of locomotion in Asian elephants. Journal
of Experimental Biology, 213: 694-706.
       
Sablin M.V. , Latham K. & Räikkönen J. (2016). Spondylosis

site: comparison with other canid populations. International
Journal of Paleopathology, 15: 83-91.
Hendricks J.R. (2015). Glowing seashells: diversity of fossilized
coloration patterns on coral reef-associated cone snail
(Gastropoda: Conidae) shells from the Neogene of the
Dominican Republic. PloS ONE 10: e0120924.
Hinton M.A.C. (1910). A preliminary account of the British voles
and lemmings; with some remarks on the Pleistocene climate
and geography. Proceedings of the Zoological Society, 1:
102-107.
Hone D.W.E., Tischlinger H., Xu X. & Zhang F. (2010). The
extent of the preserved feathers on the four-winged dinosaur
Microraptorgui under ultraviolet light. PloS ONE 5: e9223.
Hutchinson J.R. (2009). Response: of ideas, dichotomies, methods,
 
of other large animals? Journal of Experimental Biology, 212:
153-154.
Hutchinson J.R., Delmer C., Miller C. E., Hildebrandt T., Pitsillides

ontogeny, function, and evolution of elephant “sixth toes”.
Science, 334(6063): 1699-1703.
Hutchinson J.R., Schwerda D., Famini D.J., Dale R.H., Fischer
M.S. & Kram R. (2006). The locomotor kinematics of Asian
and African elephants: changes with speed and size. Journal of
Experimental Biology, 209: 3812-3827.
Keene S. (1987). Some adhesive and consolidants used in
conservation. The geological Curator, 1: 421-425.
Kotsakis T. (1988). Biostratigraphy of Plio-Pleistocene Arvicolids
(rodents) of Italy. Modern Geology, 13: 163-175.
Laurenti J.N. (1768). Specimen medicum, exhibens synopsin
reptilium emendatam cum experimentis circa venena et antidota
Reptilium austriacorum. J.T. Nob. de Trattner. 215 pp. Vienna.
Leden I., Forslind K., Svensson B. & the BARFOT study group
(2012). Ankylosis of wrist and small joints of he hand occurs in
rheumatoid arthritis: diagnostic implication in paleopathology.
International Journal of Paleopathology, 2: 249-251.
Linares Soriano A. & Carrascosa Moliner B. (2016). Consolidation
of bone material: chromatic evolution of resins after UV
accelerated aging. Journal of Paleontological Techniques, 15:
46-67.
Linnaeus C. (1758). Systema naturae per regna tria naturae,
secundum classes, ordines, genera, species, cum characteribus,
 th edition. 824 pp.
Laurentius Salvius, Holmiae.
Maccagno A.M. (1958). Relazione sulla tecnica di scavo, restauro
e montaggio dell’elefante fossile rinvenuto presso l’Aquila.
Annuario delle Istituzioni di Alta Cultura sorte nella città
dell’Aquila dal 1948 al 1957, 2: 124-137.
Maccagno A.M. (1962). L’Elephas meridionalis Nesti di
Contrada “Madonna della Strada” Scoppito (L’Aquila). Atti
dell’Accademia di Scienze Fisiche e Matematiche di Napoli,
Memoria, 4: 1-132.
Magri D., Di Rita F. & Palombo M.R. (2010). An Early Pleistocene
interglacial record from a intermontane basin of central Italy
(Scoppito, L’Aquila). Quaternary International, 225: 106-113.
          
Sardella R. (2012). 
biochronology in a Pleistocene intramontane basin: the case of
western L’Aquila Basin (central Apennines, Italy). Quaternary
International, 267: 62-77.
Marchant G.H. & Shoshani J. (2007). Head muscles of Loxodonta
africana and Elephas maximus with comments on Mammuthus
primigenius muscles. Quaternary International, 169: 186-191.
Nesti F. (1825). Sulla nuova specie di elefante fossile del Valdarno
all’Illustrissimo Sig. Dott. Prof. Ottaviano Targioni Tozzetti
(Lettere sopra alcune ossa fossili del Valdarno non per anco
descritte). Nuovo Giornale di Lettere, 11: 195-216.
NTC (2008). Norme Tecniche per le Costruzioni. Decreto
Ministeriale 14.01.2008.
Otth A. (1837). Beschreibung einer neuen europäischen
Froschgattung, Discoglossus. Neue Denkschriften der
Allgemeinen Schweizerischen Gesellschaft für die Gesammten
Naturwissenschaften, 1: 1-8.
Palombo M.R. & Ferretti M.P. (2005). Elephant fossil record from
Italy: knowledge, problems, and perspectives. Quaternary
International, 126: 107-136.
Pasiuk J. (2004). Safe plastics and fabrics for exhibit and storage.
Conserve O Gram, 1: 1-7.
Paul G. (2009). The nearly columnar limbs of elephants are very

mammals and birds. Journal of Experimental Biology, 212:
152-153.
Reggiani P. (2013). Relazione sul restauro dei resti di elefante
pleistocenico conservati presso il Museo Preistorico dei Balzi
Rossi. Archivio Soprintendenza Archeologica, Belle Arti e
Paesaggio per la città metropolitana di Genova e le provincie
di Imperia, La Spezia e Savona: 1-2.
Reggiani P. & Sala B. (1992). I mammut del Veneto. Memorie della
Società Geologica, 44: 171-191.
Ren L., Butler M., Miller C., Paxton H., Schwerda D., Fischer
M.S. & Hutchinson J.R. (2008). The movement of limb
segments and joints during locomotion in African and Asian
elephants. Journal of Experimental Biology, 211: 2735-2751.
Riquelme F., Ruvalcaba-Sil J.L. & Alvarado-Ortega J. (2009).
Palaeometry: non-destructive analysis of fossil materials.
Boletín de la Sociedad Geológica Mexicana, 61: 177-183.
Sami M. & Ghezzo E. (2015). Mammiferi del Pleistocene superiore
nelle Grotte dei Gessi di Brisighella e Rontana. In Lucci P.
& Piastra S. (eds), I gessi di Brisighella e Rontana. Studio
Multidisciplinare di un’area carsica nella vena del gesso
romagnola. Memorie dell’Istituto Italiano di Speleologia, II,
28: 101-112.
Shelton Y.S. & Chaney D.S. (1994). An evaluation of adhesives
and consolidants recommended for fossil vertebrates. In Leiggi
P. & May P. (eds), Vertebrate Paleontological Techniques, 1,
Cambridge University Press: 35-43.
Stone R. (2010). Altering the past: China’s faked fossils problem.
Science, 330: 1740-1741.
Tischlinger H. (2002). Der Eichstätter Archaeopteryx i m
langwelligen UV-Licht. Archaeopteryx, 20: 21-38.
Tischlinger H.(2005). Neue Informationen zum Berliner
Exemplar von Archaeopteryx lithographica H. v. Meyer 1861.
Archaeopteryx, 23: 33-50.
Tischlinger H. & Unwin D.M. (2004). UV-Untersuchungen
des Berliner Exemplars von Archaeopteryx lithographica
H. v. Meyer 1861 und der isolierten Archaeopteryx-Feder.
Archaeopteryx, 22: 17-50.
Toula F. (1902). Das Nashorn von Hundsheim Rhinoceros
(Ceratorhinus Osborn) hundsheimensis nov. form. mit
Ausführungen über die Verhältnisse von elf Schädeln von
Rhinoceros (Ceratorhinus) sumatrensis. Abhandlungen der
Kaiserlich-Königlichen Geologischen Reichsanstalt, 57: 445-454.
373
M.A. Rossi et alii - Mammuthus meridionalis from Scoppito (AQ): diagnostics and restoration
Turner-Walker G. (2009). Degradation pathways and conservation
strategies for ancient bone from wet, anoxic sites. Proceedings
of the 10th Triennial Meeting of the ICOM-CC Working Group
for Wet Organic Archaeological Materials. Huisman H.D.J. &
Strætkvern K. (eds) 10-15th September 2007: 659-674.
Weissengruber G.E. & Forstenpointner G. (2004). Musculature of
the crus and pes of the African elephant (Loxodonta africana):
insight into semiplantigrade limb architecture. Anatomy and
Embryology, 208: 451-461.
Weissengruber G.E., Fuss F.K., Egger G., Stanek G., Hittmair
K.M. & Forstenpointner G. (2006). The elephant knee joint:
morphological and biomechanical considerations. Journal of
Anatomy, 208: 59-72.
Manuscript received 3 March 2017
Revised manuscript accepted 6 December 2017
Published online 30 December 2017

Bollettino della Società Paleontologica Italiana, 56 (3), 2017. Modena
Mammuthus meridionalis from Madonna della Strada (Scoppito, L’Aquila):
diagnostics and restoration
Maria Adelaide , Silvano , Maria Rita , Ivana , Salvatore ,
Filippo , Elena , Feder ica , Gian mario , Paolo , Cristi na ,
Lisa , Giorgio   & Emanuela 
M.A. Rossi, Soprintendenza archeologia, belle arti e paesaggio dell’Abruzzo, Via degli Agostiniani 14, I-66100 Chieti, Italy;
mariaadelaide.rossi@beniculturali.it
S. Agostini, Soprintendenza archeologia, belle arti e paesaggio dell’Abruzzo, Via degli Agostiniani 14, I-66100 Chieti, Italy; silvano.agostini@beniculturali.it
M.R. Palombo, Dipartimento di Scienze della Terra, Sapienza Università di Roma, Piazzale Aldo Moro 5, I-00185 Roma; CNR-IGAG, Via Salaria km
29,300, I-00015 Monterotondo (Roma), Italy; mariarita.palombo@uniroma1.it
I. Angelini, Dipartimento di Beni Culturali, Università degli Studi di Padova, Piazza Capitaniato 7, I-35139 Padova, Italy; ivana.angelini@unipd.it
S. Caramiello, Soprintendenza archeologia, belle arti e paesaggio dell’Abruzzo, Via degli Agostiniani 14, I-66100 Chieti, Italy;
salvatore.caramiello@beniculturali.it


F. Marano, Dipartimento di Scienze della Terra, Sapienza Università di Roma, Piazzale Aldo Moro 5, I-00185 Roma; federica.marano86@gmail.com
G. Molin, Dipartimento di Beni Culturali, Università degli Studi di Padova, Piazza Capitaniato 7, I-35139 Padova, Italy; gianmario.molin@unipd.it
P. Reggiani, Paleostudy, Via Zabarella, 21, I-35028 Piove di Sacco (PD), Italy; paleostudy@libero.it
C. Sangati, AR Arte e Restauro srl, Viale Navigazione Interna 49, I-35129 Padova, Italy; sangati@arpadova.it
L. Santello, Dipartimento di Geoscienze, Università degli Studi di Padova, Via G. Gradenigo 6, I-35131 Padova, Italy; lisa.santello@unipd.it
G. Socrate, AR Arte e Restauro srl, Viale Navigazione Interna 49, I-35129 Padova, Italy; socrate@arpadova.it
E. Vinciguerra, AR Arte e Restauro srl, Viale Navigazione Interna 49, I-35129 Padova, Italy; emavinci@hotmail.it
SUPPLEMENTARY ONLINE MATERIAL
METHODS
1 - Mapping
A total of 156 photographs (size 7360×4912 pixel, 300
dpi) were obtained using a Nikon D800 digital camera
and Nikon Nikkor AF-S 24-70 mm f/2.8G Ed lens (used
for most photos). Of these, 113 were taken in RGB at
from ISO100 to ISO125 and 43 images were taken in
UV light at ISO400. The latter were obtained by applying
     
radiation source with a 400 watt gas-discharge lamp with
an emission peak of about 360 nm (Fig. S1). The x-ray
Fig. S1 - Lateral-dorsal view of the skull in (a) RGB with colour calibration target (x-Rite release) and (b) UV. In b), the reconstructed portions
in dark-blue and consolidants or glues along the temporal fossae (cyan) are clearly evident.
plates consist of 45 impressed images showing the inner
trabecular structure of the bones and where we assumed
that armatures, pins and metal structures had been inserted.
The x-ray plates were obtained using an x-ray tube with a
maximum capacity of 300 KV. The rays were impressed on

times were calibrated according to the osteological
structures. The resulting plates were digitalized with a
high resolution scanner and the grey scale was corrected
to improve the visibility of details.
Mapping was performed using AutoCAD-LT-2013,
with homologous overlapped RGB and UV photos and
S. P. I.
S
O
C
I
E
T
A
'
P
A
L
E
O
N
T
O
L
O
G
I
C
A
I
T
A
L
I
A
N
A
Bollettino della Società Paleontologica Italiana, 56 (3), 2017
ii
x-ray plates where available. The vector images were

      

four sets: surface characterization, considering the initial
status of the bones and their taphonomy; reconstructions
and paint from previous restorations; areas used as
preliminary markers (x-ray, chemical samples), and
       
chromatic response to UV light (Tab. S1, Fig. S2).
2 - Compositional and textural investigations
The analyses were performed on small samples
measuring about 0.5 mm3. In just a few selected
cases, thin sections were prepared from samples
generally measuring about 1.0-1.5 cm. Preliminary
stereomicroscopic study of the samples was followed
by detailed study by optical microscopy (OM), with

Based on these observations, studies were then performed


the mineralogical and crystallographic characteristics of
the crystal phases present and infrared (FTIR and µ-FTIR)
and µ-Raman spectroscopy to characterise the inorganic
and organic phases. Elemental analysis was carried out
by X spectroscopy using a scanning electron microscope
(SEM) coupled to an energy dispersive spectrometer
(EDS) (Fig. S3). Textural analyses, particularly important
in light of the project’s objectives, were carried out with
both OM and SEM.

compositions were taken from the surfaces of the
artefact (6 representative samples of the ulna, rib, radius
and humerus); tusk (2 samples, internal and external);
reconstructions (4 samples); surface patinas (2 samples);
resins (5 samples) and paints (4 samples).
For the second phase involving 3D study of the long
bones, minimal sized core samples were taken from the
epiphyses of the right ulna, right tibia and left humerus.
Polished thin sections were obtained from the median-
longitudinal plane of the core samples for textural and
spectroscopic study (OM, SEM-EDS, FTIR and µ-FTIR
and µ-Raman); the lateral portions provided the powder
for the XRPD study.
3 - The dynamic behaviour of the skeleton
The Finite Elements model of the mammoth was
constructed using the 3D geometric model obtained
with the laser scanner technique. The geometric “shell”
elements externally defining the overall geometry
       


all the defects of shape (overlapping faces, multiple faces,
etc.), rendering the model suitable for transformation into
a solid mesh. The 2D plate elements were then transformed
      


iii
M.A. Rossi et alii - Supplementary Online Material
the cavities in the model to obtain a completely solid 3D
model (Fig. S4).
In a second phase, the steel frame was manually
shaped using beam-type one-dimensional elements. The
model of the frame is independent from the structure of
the mammoth and linked at points corresponding to the
beam elements distributed along the frame (Fig. S5). This
modelling strategy was functional to correct transfer of
the entire weight of the bones to the frame.
Knowledge of the weight of the skull led to the
following properties being used for the bones:
Mass density: 13000 N/m3
As regards the elastic modulus, a relatively low value
was used (elastic modulus = 2.58e + 008 N / m2) as a lower

mass resulting from the mammoth skeleton. Type S235
steel was used for the frame structure.
The FE model was calibrated with the results of
the dynamic tests. Four/five acquisition setups were
        
seismic high resolution piezoelectric accelerometers
 
      
two frequencies (0.56 Hz, 0.75Hz), representing the
    
directions.
Linear static analysis was carried out in order to
assess the stress levels in the elements of the metal
frame. The only force considered for this analysis was
self-weight.
4 - The restoration
In chronological order, the operations were as follows:
1. disassembly; 2. removal of paint from the surfaces of the
reconstructions and bones; 3. cleaning of the original parts
with removal of the previous protective agent from the
surface; 4. removal of plaster and mastic residues using a


elements and reconstructions; 6. pre-consolidation of a
number of particularly fragile portions and consolidation
by immersion; 7. structural consolidation with resin
1) 
includes everything noted on the bone surface before conservative restoration
brown colour Dark brown colour of the bones. Area sampled for chemical and mineral analyses
red colour Red colour of the bones. Area sampled for chemical and mineral analyses
hard dusty coat (light blue) Dusty substance distributed over the bone surface and adhering to the periosteum
fractures/cracks
Areas with visible cracks as the consequence of peri-depositional events (presumably stable) or post-recovery events

restorations.
microholes The presence of holes was limited to the anterior diaphysis of the right femur
trabecular bone Inner trabecular structures or areas without periosteum
2) 
(paints - reconstructions): includes everything added after recovery of the skeleton
light brown paint Area with light brown paint
dark brown paint Area with dark brown paint
metal pins The position of the pins was obtained from the x-ray plates


x-ray 
rst sampling 


bright light blue Probable consolidant from fractures and broken parts
purplish blue response Corresponding to reconstructed and painted parts
red response 
light blue response 
uorescent yellow response 
     
UV photos and x-rays.
Mode nr. Frequency
(Hz)
Mass X
(%)
Mass Y
(%)
1st mode  84.2 
2nd mode   
3rd mode   
4th mode   
th mode   
        
mode shapes.
Bollettino della Società Paleontologica Italiana, 56 (3), 2017
iv
dismantling, a protective net was mounted under the
skeleton. In the case of the largest bones, suitable lift and
support systems were designed and modulated according
to the anatomical characteristics and fragility of the
various elements. The most complex movements involved
       
and lifting simulations (Fig. S8). Modular systems were

pelvis and scapulae (Fig. S9). The supports on which the
various elements rested transmitted the weights evenly,
avoiding dangerous load concentrations.
Before removing the pelvis, a safety bandage with
Paraloid B72 was applied, as the 160 kg weight, considerable
size, minimal thickness of the ilium wings and sponginess
of the ilium and pubis placed this anatomical element at
considerable risk. After protection of the surfaces with
cyclododecane in mineral solvent, a protective layer of
Japanese tissue was applied and plaster was poured into
the two ischiopubic openings to improve stability during
handling. A metal stabiliser tie rod was inserted into the
central cavity. Weighing about 500 kg, the skull was moved

consisting of metal tie rods, straps and manual hoists, with the
insertion of rubber protection. Use of the top metal crossbeam

2. The light brown paint was removed from the
surfaces of the reconstructions and bone material with
solvent (acetone) applied on pads and probes.

colouring of the reconstructed portions.
1. The bone elements were disassembled in three
       
operations in order to rationalise and reduce handling of
the bone elements. This operation was supported on site
by wheeled multi-shelved tables where the removed bones
were placed after cataloguing. To improve safety, before
Fig. S3 - Chemical maps based on SEM-EDS analyses of a thin section of a bone sample of the right tibia.
Fig. S4 - The Finite Element model of the Mammuthus meridionalis.
v
M.A. Rossi et alii - Supplementary Online Material
3. Before cleaning, the pure solvents and solvent
mixtures were tested. The efficacy of cleaning and
their actual capacity to remove the PVAc residues
  
microsamples before and after cleaning and on powder
residues. The solvent tests were carried out using acetone,
butyl acetate, dimethyl sulphoxide and Dowanol, pure and
mixed in various percentages.
The analytical results (FTIR) showed that cleaning
with acetone completely removed the PVAc present in
considerable quantities on the bone surfaces, without signs
of “transport” dynamics in depth. The volatility of the
acetone was also a further advantage for cleaning. Small
bones were cleaned by partial controlled immersion in
suitable tanks containing the solvent, using soft bristle
brushes to completely remove the PVAc residues. In the
case of large bones, the acetone was applied by brush
and poultice.
4. The presence on the bones of widespread plaster
residues not removed after recovery of the skeleton in
1955 and mastic residues left from previous restorations
necessitated the use of a precision micro-sandblaster. The
inert substance used was Hydrosoft HDO micronised
calcium carbonate with a grain size of 140 µm and a
rounded shape, allowing the action to be precisely targeted
by the operator (Fig. S10).
5. Lowering the edges of the reconstructions covering
extensive portions of bone enabled the contact surfaces
between the original bone and reconstructions to be
     

6. The consolidant was chosen bearing in mind that
the bones had already been treated previously. It was
decided not to experiment with either nanotechnologies
or ethyl silicate as there was a lack of supporting
literature. The choice therefore fell on suitable acrylic
products (Keene, 1987; Shelton & Chaney, 1994; Linares
Soriano & Carrascosa Moliner, 2016) with the following

Fig. S5 - The Finite Element model of the metal structure supporting
the mammoth.
Fig. S6 - View of some of the acceleration sensors.
Fig. S7 - Points of application of the acceleration sensors.
Bollettino della Società Paleontologica Italiana, 56 (3), 2017
vi
size, low viscosity and high adhesive capacity. After
      
LP 64/12, an acrylic polymer which, unlike the more
common Paraloid B72, had all the above characteristics.
The decision was made after testing the two consolidants

methyl ketone) at concentrations of 5%, 7% and 10%.
Butyl acetate solvent was preferred to acetone for its lower
volatility and application by immersion was chosen as the
most suitable procedure.
Chemical stratigraphic and micro-FTIR analyses were
performed on polished sections of small core samples
taken from the bone elements treated with immersion in
Paraloid B72 and Degalan LP 64/12, both in a 5% solution
in butyl acetate. With the sample treated with Degalan LP
64/12, penetration reached about 20 mm, better than the
15 mm penetration achieved in the sample treated with
Paraloid B72.

elements (2 points on the phalanx, 2 on the radius and 1
on the vertebra) showed good reinforcement of the spongy
tissue trabeculae. After further testing on site, the immersions
were performed in a 10% solution of Degalan LP 64/12 in
butyl acetate for a period of between 2 and 4 hours depending
on the size of the anatomical element (Fig. S11).
Particularly fragile anatomical elements, or parts of
these, were consolidated with localised injections of 10%
Degalan LP 64/12 in butyl acetate. Before consolidation,
nylon-lined boxes of a size and shape corresponding
to those of the bones were prepared, with the aim of
optimising and limiting as far as possible the quantities
of solvent used for the immersions. The solutions were
used for a number of cycles as long as the liquid remained

After the consolidant bath, the bones were dried slowly


vii
M.A. Rossi et alii - Supplementary Online Material

Fig. S10 - The restorer using a micro-sandblaster on the skull to remove residues of plaster and mastic left from previous restorations.
Bollettino della Società Paleontologica Italiana, 56 (3), 2017
viii
and characteristics of the individual bones, from 3-4 weeks
for the smallest to more than 2 months for the skull and
pelvis.
Immersion of the skull was particularly complex due
to its size and the 700 litres of solvent used. For this
purpose, a box was constructed around the bone element
and its support.
      
  
structural consolidation was performed by injections of
a 10% solution of the consolidant Degalan LP 64/12 in
butyl acetate, or a liquid two-component epoxy resin.

consolidated, then glued using a two-component epoxy

8. It was decided to modify a number of the anatomical
parts reconstructed during previous restorations as
these did not perfectly correspond to the anatomy of
the animal. For these new reconstructions, a light two-
component epoxy-based putty, reversible in polar solvents,
commercially known as Balsite was used. This product was
chosen for its positive characteristics, including ease of
working, good adhesion to the fossil surfaces and elasticity
(Ciocchetti & Munzi, 2007). This epoxy resin is very light
and tends to adhere to the tools used for shaping and the
nitrile gloves. Shaping could therefore only be performed
correctly by moistening the tools and gloves with absolute
alcohol. Thanks to its particular formula, it responds
coherently with the support to the stresses generated by
thermohygrometric variations and it is therefore indicated
Fig. S11 - Handling of a humerus after immersion in consolidant solution.
ix
M.A. Rossi et alii - Supplementary Online Material
for reconstruction of the missing parts of fragile specimens
(Ciocchetti & Munzi, 2007). Balsite has already been used
successfully in the restoration of other fossil vertebrates
(Agostini et al., 2010; Reggiani & Ghezzo, 2015)

was found to have the best characteristics in terms of
appearance, workability and speed of execution. Use
of Balsite with the addition of 5% weight of a mixture
of natural coloured earths, and walnut husk to obtain
     
reconstructions of all bones. For larger reconstructions,
only the junction with the bone was restored.
     
integrated chromatically. The choice of the method
respects the basic restoration principle according to
         

diluted in turpentine oil were used, giving a stippled

reconstructions made from additioned Balsite. For small

of a brush, while for larger surfaces it was achieved by
spattering using a round brush with hard bristles soaked
in the paint.
SUPPLEMENTARY REFERENCES
Agostini S., Caramiello S., Di Canzio E., Rossi M.A., Berardinelli A. & Letta S. (2010). Problematiche di recupero e restauro dei resti fossili
di Mammuthus meridionalis rinvenuti a L’Aquila località Campo di Pile. Giornate di Paleontologia X, Università della Calabria - Rende
(CS), Maggio 2010 :57.
Ciocchetti C. & Munzi C. (2007). La Balsite®: un nuovo materiale per il risanamento dei supporti lignei e per la realizzazione di parti mancanti.
Bollettino ICR n.s., 15: 19-37.
Linares Soriano A. & Carrascosa Moliner B. (2016). Consolidation of bone material: chromatic evolution of resins after UV accelerated
aging. Journal of Paleontological Techniques, 15: 46-67.
Keene S. (1987). Some adhesive and consolidants used in conservation. The geological Curator, 1: 421-425.
Reggiani P. & Ghezzo E. (2015). Dal sequestro al completo recupero: il restauro della Lince della “Grotta del Gattopardo” (Savona).
Museologia Scientica n.s., 9: 62-68.
Shelton Y.S. & Chaney D.S. (1994). An evaluation of adhesives and consolidants recommended for fossil vertebrates. In Leiggi P., May P.
(eds), Vertebrate Paleontological Techniques, 1, Cambridge University Press: 35-43.
... (Stephanorhinus aff. S. hundsheimensis Toula, 1902 according to some authors, or Stephanorhinus etruscus Falconer, 1868 according to others), large deer, possibly Eucladoeros giuli Kahlke, 1997, and Hippopotamus antiquus desmarest, 1822 (see Maccagno 1962;Magri et al. 2010;Mancini et al. 2012;Rossi et al. 2017). The mammoth skeleton and associated faunal element can be chronologically constraint to an age of 1.3 Ma, obtained via palynological data from a lignitic deposit just above the fossiliferous level (Magri et al. 2010;Rossi et al. 2017). ...
... S. hundsheimensis Toula, 1902 according to some authors, or Stephanorhinus etruscus Falconer, 1868 according to others), large deer, possibly Eucladoeros giuli Kahlke, 1997, and Hippopotamus antiquus desmarest, 1822 (see Maccagno 1962;Magri et al. 2010;Mancini et al. 2012;Rossi et al. 2017). The mammoth skeleton and associated faunal element can be chronologically constraint to an age of 1.3 Ma, obtained via palynological data from a lignitic deposit just above the fossiliferous level (Magri et al. 2010;Rossi et al. 2017). ...
... From 2013 to 2015, a new restoration project has been launched and carried out thanks to the consistent financial support of "Guardia di Finanza" (contribution to the reconstruction of the city of L'Aquila) and directed by the "direzione Regionale per i beni culturali e Paesaggistici dell'Abruzzo" (MibAcT), as part of the reconstruction projects following the great earthquake that struck L'Aquila in 2009. The skeleton was disassembled again, with a new restoration and detailed study conducted thanks to the latest innovative methodologies and technologies available in the paleontological and restoration fields (see Rossi et al. 2017). The skeleton was then reassembled by changing the supporting structure, in order to restore a general posture more appropriate from a biomechanical point of view (Rossi et al. 2017) (Fig. 1). ...
Article
Full-text available
In this contribution we present an in vivo reconstruction and volumetric body mass estimate for the mounted skeleton of Mammuthus meridionalis on exhibit at the east bastion of the Spanish Fortress at L’Aquila (Abruzzo, Central Italy). The reconstruction has been obtained starting from a 3D photogrammetric model of the skeleton acquired via a micro-drones and by digitally adding a percentage of soft tissues according to the conditions observed in wild specimens. By applying to the volume the density range proposed in literature for extant proboscideans we obtain an estimate of the body mass in the adult male specimen ranging from 11.3 t to 11.5 t, with average body mass equal to 11.43 t. In addition, we compare the volumetric BM estimate with the BM predictive values obtained by means of traditional regression equations based on long bones linear dimension and shoulder height. The results confirm that the volumetric method is always preferable if sufficiently complete mounted skeletons are available, since application of regression formulas to single bony element can lead to an underestimation or overestimation up to 130%. As a general indication, weight estimates in extinct tetrapods based on single measures and single bones should be totally avoided, especially in groups morphologically and phylogenetically distant from extant reference taxa.
... As a case study, we chose the skeleton of Mammuthus meridionalis, mounted at the east bastion of the Spanish Fortress at L'Aquila (Abruzzo, Central Italy) ( Figure 1). The skeleton has recently been restored, correcting and improving its general structure and posture (Rossi et al. 2017). ...
... The M. meridionalis skeleton was found in 1954 within a sandy layer of beach and bar deposits at the bottom of a lacustrine clayey sequence with lignite level and mud sandy fan outcropping at Santarelli clay quarry (Madonna della Strada, Scoppito, L'Aquila) (Maccagno 1958(Maccagno , 1962Magri et al. 2010;Mancini et al. 2012;Agostini et al. 2014;Rossi et al. 2017). Based on palynological data collected from a lignitic deposit just above the productive sandy one, an age of about 1.3 Ma has been proposed to chronological constrain the mammoth skeleton ( Associated with the skeleton, other faunal remains were found, referable to Hippopotamus antiquus (Desmarest 1822) and ...
... (by some identified as Stephanorhinus aff. S. hundsheimensis Toula 1902, by others as Stephanorhinus etruscus Falconer 1868), and a large deer, possibly Eucladoeros giulii (Kahlke 1997) (see Maccagno 1962;Magri et al. 2010;Mancini et al. 2012;Rossi et al. 2017). ...
Article
Unmanned aerial vehicles (UAVs), have seen tremendous development in the last decade, with numerous applications in civil and research fields. Drones' success, particularly in the field of research, is due to a number of factors, including rapid technological advancement, tool versatility, and prices that are becoming increasingly affordable even for small research groups or individuals. Given the versatility and ongoing development of micro drones, we tested the use of micro-drones in vertebrate palaeontology to reconstruct mounted skeletons using the photogrammetry method. The experiment was carried out on a massive specimen of Mammuthus meridionalis (Nesti 1825) from Madonna della Strada, which is on display at the east bastion of the Spanish Fortress in L'Aquila (Abruzzo, Central Italy), comparing the results with those obtained using the traditional method using a digital camera. Even though both the traditional digital camera and the drone methods produced a high-resolution 3D model of the skeleton, the results obtained, indeed, lead us to consider the use of micro-drones in museum structures as a very interesting and promising new field of application. Drones provide a simple, fast, and non-invasive system for the study, monitoring, and enhancement of cultural heritage in all of its possible manifestations.
Article
Full-text available
Spondylosis deformans is a common degenerative condition of the spinal column, especially in modern domestic dogs. The presence and severity of lesions are related to age and physical activity, but they can be influenced by genetics, with some modern breeds being particularly predisposed. Spondylosis deformans also has been reported in prehistoric dogs. Here, we describe three affected vertebrae, likely from three individuals, in a large canid assemblage from the Gravettian Předmostí site, Czech Republic. We compare the proportion of affected individuals from the Předmostí assemblage with that of affected individuals among recent wild Northern wolves, inbred wolves, captive wolves, and recent Northern dogs. The proportion of affected individuals among the captive wolves differs significantly from the other wolf groups. The proportion in the Předmostí assemblage does not differ significantly from that of the wild wolf groups.
Article
Full-text available
New finds of Mammuthus meridionalis from 'La Salita di Oriolo' quarry (Faenza, Emilia Romagna, Italy) are described. The 'Sabbie gialle' outcropping at Oriolo yielded several mammal species pointing out an early Galerian age (late Early Pleistocene) for the whole assemblage. A comparison with the type population of M. m. meridionalis from Upper Valdarno (Tuscany; Italy; Tasso Faunal Unit; early Early Pleistocene) and younger samples from Italy and other European localities allows to draw a possible scenario of the late evolution of the southern mammoth. On the basis of the systematic study two distinct lineages of southern mammoth are recognised. One, currently known only from Italy, is represented by several findings, among which are the Oriolo specimens and those referred to M.m. vestinus, characterized by large size, specialized skull and primitive dentition. A second group of forms, making its first occurrence in the early Galerian, presents instead derived dental morphology and is here referred to as M.meridionalis ex gr. tamanensis. During the middle Galerian (early Middle Pleistocene) the latter is replaced throughout Eurasia by the more evolved steppe mammoth M.trogontherii, a probable off-shoot of this group, while the forms belonging to the M.m.vestinus group became extinct without descendants.
Article
The Berlin example ol Archaeopteryx and the isolated Archaeopteryx feather were investigated and documented photographically using long-wave ultra violet light together with an established filtering technique. The current condiiion of the fossil slabs together with important phases of preparation and restoration work that began in 1876 are described. Many morphological details of the skeleton could be more precisely.resolved in ultra violet light than in visible light. ln the skull region significant anatomical details, such as the construction of the preorbital region and the cheek region, were brought to light. ln the vertebral column the number, length and form of what were formerly indistinct elements was significantly clarified and the presence of pneumatic openings in the cervical series was confirmed. A problematic calcified structure of the shoulder girdle could be treated as the right coracoid on the basis of its positional relationships; but by contrast the findings of the UV light study suggest that it was a poorly ossified sternum largely constructed from cartilage. Further, formerly controversial details of the skeletal morphology of the pelvic girdle and limbs were resolved. The UV light study also supported the idea that the Berlin specimen belongs lo Archaeopteryx lithographica.The exceptionally well preserved feather remains have usually been described as relief impressions. Under UV light it was clear that at least the distal region of the flight feathers is preserved as a thin filmy substance. Further soft part finds such as body feathers and keratinised structures are discussed. The single isolated feather of Archaeopteryx is better seen under UV than visible light and dark areas are possible relicts of what was originally colour patterning.