Full text of DOI:10.1163/156853800507471

Distribution, natural history and morphometrics of the
Coluber hippocrepis
critically endangered
populations of
Sardinia: a review, with additional data and conservation
implications
Claudia Corti1, Luca Luiselli2, Ernesto Filippi3, Massimo Capula4
1
Dipartimento di Biologia Animale e Genetica, Università degli Studi di Firenze, via Romana 17, I-50125
Firenze, Italia
e-mail: claudia.corti@uniŽ.it
2
Centro di Studi Ambientali “Demetra”, via dei Cochi 48/B, I-00133 Roma, Italia; and F.I.Z.V., via Cleonia 30,
I-00152 Roma, Italia; and Museo Civico di Storia Naturale, piazza Aristide Frezza 6, I-00030 Capranica
Prenestina, Roma, Italia
e-mail: lucalui@iol.it
3
via Gabrio Casati 43, I-00139 Roma, Italia
e-mail: e.Ž lippi@iol.it
4
Museo Civico di Zoologia, Sezione di Erpetologia, via Ulisse Aldrovandi 20, I-00197 Roma, Italia
e-mail: m.capula@comune.roma.it
Abstract. The horseshoe snake (Coluber hippocrepis) is one of the most threatened Italian snakes, especially in
Sardinia where its present occurrence was unclear. This paper reviews the current distribution of C. hippocrepis
in Sardinia, and conŽ rms that this species is still found in a limited number of sites of Cagliari province. Its actual
presence in Oristano province, on the contrary, is still in doubt. Statistical analyses on several morphometric
measurements taken from museum specimens suggest that snakes from the central Mediterranean area (Sardinia,
Pantelleria, and Tunisia) are distinguished from those of the western Mediterranean area (Portugal and Morocco)
by having signiŽ cantly higher numbers of ventrals. A few natural-history notes on Sardinian specimens are
presented, and a rationale for an urgent conservation plan is given.
Introduction
The snake fauna of Sardinia island (Mediterranean Sea) is quite depauperate in terms of
speciŽ c diversity (Bruno and Maugeri, 1990; Poggesi et al., 1995). Nevertheless, Sardinia
Coluber
, and the horseshoe snake
Coluber
Coluber Haemorrhois hippocrepis
houses two species of snakes of the genus
Hierophis viridiavus
, the western whip snake
(
)
(
)
,
®c
Koninklijke Brill NV, Leiden, 2000
Amphibia-Reptilia 21: 279-287

280
Claudia Corti, Luca Luiselli, Ernesto Filippi, Massimo Capula
which are normally allopatric in other European regions (Bruno and Maugeri, 1990;
Schätti, 1993).
Coluber
This coexistence of two
species is relevant both in ecological and conservation
terms. In ecological terms, it deserves attention because of the presence of all the basic
conditions for strong interspeciŽ c competition, as both these potential competitors are
large predators, phylogenetically related, similarly sized (up to 2 m long, Bruno and
Maugeri, 1990), terrestrial and good climbers, prey-limited, and with similar natural
history traits (Bruno and Maugeri, 1990; Schätti, 1993). In terms of conservation, this
C. hippocrepis
, is critically
case of coexistenceis also important as one of the two species,
endangered in Italy (Filippi and Luiselli, 2000), where, according to some authorities, it
was passively introduced during the historical age (Lanza, 1983; Gruber, 1989). In Italy,
C. hippocrepis
is nowadays known from only a limited number of sites in Sardinia and
in Pantelleria island (Bruno and Maugeri, 1990). However, although it has been ofŽ cially
protected by law since 1982 in the whole Italian range (Berne Convention), up to now
nothing has been done to functionallyprotect its remnant populations.
C. hippocrepis
, in this report we offer:
With regard to Sardinian
(1) a review of its present distribution, with an emphasis on its potential distribution
in relation to the species-speciŽ c ecological characteristics and to the concurrent
C. viridiavus
presence of
;
(2) a preliminary morphometricanalysis;
(3) a preliminary survey of its natural history;
(4) a concise deŽ nition of the guidelines of an immediate working project which is
thought to be essential for saving it.
Materials and methods
This study is based on both literature and original data, collected mainly from museum
specimens but also from free-ranging individuals. Bibliographic research was conducted
over several hundreds of contributions concerning the Sardinian herpetofauna (in part
summarized by Poggesi et al., 1995), and was devoted mainly to collate any reliable
C. hippocrepis
distribution records and natural history information of
in Sardinia.
C. hippocrepis
Detailed morphometric data on Sardinian
are not available in the litera-
ture, givenalso the extreme rarity of such specimensin worldwide museum collections(see
also Schätti, 1986). Thus, we measured all the specimens in the collectionsof the Zoologi-
cal Museum “La Specola”, Florence(MZUF), which is likelyto store thehighestnumber of
¢
Sardinian specimens available in the world s public collections. The examined specimens
were labelled as follows: MZUF 446, MZUF 2924 (both from Morocco), MZUF 29056,
MZUF 28855, MZUF 29057(all from Portugal), MZUF 7984, MZUF 7985, MZUF 7884,
MZUF 7885, MZUF 7886, MZUF 7887, MZUF 7888 (all from Sardinia), MZUF 19533,
MZUF 35543, MZUF 35094, MZUF 7998, MZUF 7997, MZUF 7996, MZUF 7999 (all

Coluber hippocrepis in Sardinia
281
from Pantelleria Island), MZUF 2626, MZUF 2628 (both from Tunisia). Every specimen
±
±
was measured to snout-vent-length(SVL, to 0.5 cm), tail length (tL, to 0.1 cm), head
±
±
length (HL, to 0.1 cm), head width (HW, to 0.1 cm), interorbital length (INT-ORB,
±
to 0.1 cm), and number of ventrals. Tails of several specimens were broken (a common
C. hippocrepis
event also in free-ranging
, see Capula et al., 1997), and thus tail length was
removed from any analyses.
Field research was conducted throughout several discrete survey periods (March and
August from 1983 to 1986, April 1990, April and June 1992, March, April, and June 1993,
October 1996, May 1998, March, October, and November 1999), devoted to study the
C. hippocrepis
reptile ecology in areas potentiallyinhabitedby
Poggesi et al., 1995).
(Capula and Luiselli, 1994;
Statistical analyses were done by using SPSS for Windows PC package, with alpha set
at 5% and all tests being two tailed.
Results and discussion
Distribution in Sardinia
Satellite islets. Coluber hippocrepis
is known only from Isola di San Pietro (Poggesi
et al., 1995), where it was originally collected by Stefani (1971). In this islet, it was found
multiple times during recent years (G. Serra, in Poggesi et al., 1995), but never by other
searchers (Bruno and Maugeri, 1990; Entomosarda s.n.c., 1993; Societas Herpetologica
Italica, 1996; Corti et al., unpublished). According to Bruno and Maugeri (1990), the
C. viridiavus
reliability of this record is doubtful. Isola di San Pietro is inhabited by
(Entomosarda s.n.c., 1993; Poggesi et al., 1995; Societas HerpetologicaItalica, 1996).
Mainland Sardinia. Coluber hippocrepis
was known to occur in southwestern Sardinia
a long time ago (Gené, 1834, 1839; Camerano, 1891), but was thought to be extinct by
C. hippocre-
Mertens and Wermuth(1960). However, Bruno and Hotz (1976)rediscovered
pis
in sixteen localitiessituated between Cagliari and Oristano, and conŽ rmed that it inhab-
ited primarily coastal sites with ponds, water bodies, and lagoons(see also Gené, 1834). A
few mountain sites in central Sardinia were recorded.
C. hippocrepis
to be present in coastal southwestern Sardinia, but
More recently, the distribution of
C. hippocrepis
in Sardinia remained a dispute. Puddu
et al. (1988) stated
¢
seemed to roughly follow Bruno and Hotz s (1976) data. Bruno and Maugeri (1990)
conŽ rmed its presence in some of the sites where it was observed previously (Bruno and
C. hippocrepis
Hotz, 1976). Lanza (1993) did not account
as a species of Sardinia, but
failed to express any reliable reason for this deletion. Also Societas Herpetologica Italica
(1996) was not able to report its presence in any Sardinian locality after the year 1982.
¢
However, the species was certainly found in the early 90 s in at least nine different areas of
the province of Cagliari (Entomosarda s.n.c., 1993).

282
Claudia Corti, Luca Luiselli, Ernesto Filippi, Massimo Capula
Figure 1. Current distribution of C. hippocrepis in Sardinia. Black squares indicate recent records (years 1990
to 1999, data from Entomosarda s.n.c., 1993), black circles indicate old records still to be conŽ rmed (Bruno and
Hotz, 1976), and asterisks indicate old records from mountainous sites which are considered unreliable by the
present study.
For practical reasons, it seems better to examine separately the three main areas of this
¢
species supposeddistribution,i.e. (i) the provinceof Oristano,(ii) the provinceof Cagliari,
and (iii) the internal mountainoussites of central Sardinia. The recent records (after 1990)
C. hippocrepis
of
in Sardinia are presented in Ž gure 1, and compared with old distribution
records.
Province of Oristano. Coluber hippocrepis
is known from coastal lakes and ponds in
the surroundings of Oristano (Stagno di Cabras, Stagno di Sale Porcus, Stagno di Pauli
Maiori, see Bruno and Hotz, 1976; Bruno and Maugeri, 1990; Schätti, 1993), where it was
discovered a long time ago (Gené, 1834). No vouchers from this area are stored in MZUF.
Moreover, we were unable to Ž nd any specimen on Oristano area, neither are we aware of
C. hippocrepis
any recent record. However,
was recently captured in the surroundings of
S. Antonio di Santadi(Entomosarda s.n.c., 1993), a town which, althoughbelongingto the
province of Cagliari, is situated in the southern part of the Gulf of Oristano.

Coluber hippocrepis in Sardinia
283
Province of Cagliari.
The knowledge of its distribution in the province of Cagliari is
better (Ž g. 1). If we compare data presented by Bruno and Hotz (1976) with those of
Entomosarda s.n.c. (1993), a remarkable similarity in the local distribution patterns could
C. hippocrepis
be observed.
was found in 9.8% of the “Unità CartograŽ che di Base” (total
n
= 92) of the province of Cagliari, and the sites of capture were situated along the
coast (e.g. surroundings of Cagliari) as well as in the inner territories (e.g. Villacidro),
and on both  at areas and in the Sulcis-Iglesientemassif (Ž g. 1; Entomosardas.n.c., 1993).
C. hippocrepis C. viridiavus
In every site of ascertained presence of
,
was also found
C. hippocrepis
(Entomosarda s.n.c., 1993). During October 1999, two
(an adult male and
a juvenile, both of them with blood sampled for future analyses) were captured by one of
¢
us (MC) in the surroundingsof Cagliari (Elmas), thus deŽ nitely conŽ rming Bruno s view
C. hippocrepis
that
is still present in Sardinia.
Internal mountainous sites of central Sardinia.
C. hippocrepis
Up to now,
was known
from a few sites: Giara di Gesturi and Fonni, both in the province of Nuoro (Bruno and
Hotz, 1976). The latter locality was considered unreliable by some authorities (Bruno,
1977; Bruno and Maugeri, 1990), but not by Schätti (1993) because of the existence of a
voucher(MZUF 7985 — Conte Ugolino Martelli legit, 16. VIII. 1894; this specimen was
also examined by us). Two of us (MC and LL) and their associates surveyed carefully both
C. hippocrepis
. Thus, we
these areas (e.g. Capula and Luiselli, 1994), but did not Ž nd any
are led to agree with Bruno (1977) that these localities of capture are unreliable, or that, at
least, the species is actually extinct there.
C. hippocrepis
It is noteworthy that
is known for most sites with single individuals,
which suggests a very low populationdensity at each site (Bruno and Maugeri, 1990).
Morphometry
C. hippocrepis
A total of 21
was examined (see table 1 for summarized morphometric
measurements). Since SVL was signiŽ cantly correlated with tL, HL, HW, and INT-ORB
r
P
(in all cases at least > 0.87, < 0.0001), residual scores from the general regressions of
Table 1. Summary of the morphometric and meristic characteristics of 21 specimens of Coluber hippocrepis
stored in the collections of the “Museo Zoologico La Specola”, Florence. Tail length is not presented because
several specimens had broken tail. All measures are expressed in mm. Symbols: MR = Morocco, PO = Portugal,
SA = Sardinia, PT = Pantelleria, TU = Tunisia; SVL = snout-vent length, HL = head length, HW = head width,
INT-ORB = interorbital length, VTR = numbers of ventrals.
Parameter
MR (n = 2)
PO (n = 3)
SA (n = 7)
PT (n = 7)
TU (n = 2)
±
±
±
±
±
x¯ s
x¯
s
x¯
s
x¯
s
x¯
s
±
±
±
±
±
94.0 56.6
SVL
HL
HW
INT-ORB
VTR
93.5 30.4
68.6 24.5
69.1 30.5
86.2 16.4
±
±
±
±
±
2.5 0.0
2.4 0.7
1.9 0.5
2.0 0.6
2.4 0.3
±
±
±
±
±
1.9 0.1
1.8 0.4
1.5 0.4
1.8 0.6
1.9 0.3
±
±
±
±
±
1.0 0.2
1.1 0.2
1.0 0.2
1.0 0.2
1.1 0.0
±
±
±
±
±
234.5 4.9
223.0 0.0
231.3 2.9
236.1 4.6
237.8 4.9

284
Claudia Corti, Luca Luiselli, Ernesto Filippi, Massimo Capula
these parameters to SVL were used to test for size-corrected intergroup differences. Once
SVL was taken into account, the various populationalgroups(Portugal, Morocco, Tunisia,
Sardinia, and Pantelleria) did not differ signiŽ cantly in any of the examined parameters
F
P
(MANOVA,
= 1.079,
= 0.279). SVL was not correlated with the number of
r2
4,16
r
F
P
= 0.716, > 0.4), and
ventrals ( = 0.190, adjusted
= - 0.014, ANOVA
1,19
P
t
sexes did not differ in terms of numbers of ventrals( > 0.4, Student -test). The various
populational groups differed signiŽ cantly in terms of numbers of ventrals (ANOVA,
F4,16
P
= 0.0079), and unequal sample Tukey HSD post-hoc test indicated
= 5.044,
that specimens from Pantelleria, Sardinia, and Tunisia had signiŽ cantly higher numbers
of ventrals than specimens from Portugal and Morocco. These data suggest that specimens
from central Mediterraneanarea(Sardinia, Pantelleria, and Tunisia) are distinguishedfrom
those of western Mediterranean area (Portugal and Morocco). A much larger sample of
both sexes is required to conŽ rm this preliminary evidence, also in consideration of the
fact that, contrary to our limited central Mediterranean sample, in Iberian specimens the
females had higher numbers of ventrals than males (Feriche et al., 1993).
Natural history
C. hippocrepis
Data on Sardinian
are extremely scanty, and, in practice, we can only avail
ourselves of the few comments by Bruno and Hotz (1976), Bruno (1977, 1984) and Bruno
and Maugeri(1990). According to these sources, these snakes inhabit only coastal lagoons
and ponds, where they feed upon frogs, lizards, and rodents. However, all this information
is anecdotal, and quantitative data are not presented. According to Bruno (1977, 1984),
and Bruno and Maugeri (1990), the reasons of this unusual habitat use should be (i) the
C. viridiavus
C. hippocrepis
to refuge into quite
claimed competition with
constraining
inhospitable sites, and/or (ii) the claimed passive introduction of this species in Sardinia
(Lanza, 1983), which was thus constrained to colonize suboptimal habitats.
Based on original data (from both museum vouchers and free-ranging specimens),
information from local zoologists, and data from Entomosarda s.n.c. (1993), it appears
C. hippocrepis
that Sardinian
inhabit not only ponds and coastal lagoons, but also dry sites
with evergreen Mediterraneanmaquis and sandy soil along the coasts, and rocky and stony
areas with scattered spiny vegetation in the hills of Iglesiente-Sulcis massif. The fact that
this species is also observed around ponds and coastal lagoons does not imply necessarily
that it shows unusual habitat preferences in Sardinia, as previously supposed (Bruno and
Hotz, 1976). In fact, we think that these snakes, being quite sensible to habitat alteration,
could become extinct in dry sites which are strongly altered by human activities(industry,
agricultural practices, and tourism), but persist in “natural refuges” around water bodies,
where also many other rare vertebrates(e.g. birds) tend to concentrate.
Two specimens had three undetermined murid rodents in the guts, thus suggesting a
C. hip-
rodent-based diet for these animals. This is in agreement with data available from
pocrepis
of Pantelleria(Capula et al., 1997) and Spain (Pleguezuelosand Moreno, 1990).

Coluber hippocrepis in Sardinia
285
Conclusion and a conservation project rationale
C. hippocrepis
From all this evidence, it is clear that Sardinian
should be regarded
among the most critically endangered snakes of Europe, and thus requires an immediate
conservation project by the pertinent authorities. We suggest that this project should work
throughoutsome major phases of conservationoriented research that could be summarized
as follows:
(1) evaluation of the macroecologicaldeterminants of its local distributionin Sardinia,
by recording vegetation and landscape characteristics of a large amount of sites
C. hippocrepis
(with known
presence or absence) within its potential range, and
then applying a rigorous predictive statistical model (logistic regression, see Hos-
mer and Lemeshow, 1989) on these sites, to predict which macroenvironmentalpa-
rameters may in uence its presence/absence (Luiselli and Angelici, 2000; Luiselli
et al., 2000).
(2) evaluation of the microecologicaldeterminants of its local distribution in Sardinia,
by radiotrackingsome wild specimens, and thus deŽ ning their activity patterns and
microhabitat selection in a conservation perspective (Nilson et al., 1999; Angelici
et al., 2000).
(3) evaluation of the genetic divergence of Sardinian populations from conspeciŽ cs
from elsewhere, by applyingmoleculartechniquesto bloodsamplesof free-ranging
individuals(Lenk and Joger, 1994; Lenk et al., 1999).
(4) evaluation of its extinction probability in Sardinia, by applying stochastic models
already used for predicting amphibian populationsdecline.
After having collected solid data for points (1) and (2), it will arguably be possible to
identifypriorityareas to be protected,as already suggested for other threatenedpopulations
of European snakes (Corbett, 1989; Nilson et al., 1999). Moreover, the collection of solid
data on point (3) will permit to recognize whether snake specimens from elsewhere could
be used to “import new blood” to the Sardinian populations. We have already started to
collect such data, but require that the Societas Europaea Herpetologica will act as soon as
possibletowards the Italian governmentand the Sardinianauthoritiesin order to effectively
C. hippocrepis
preserve the remnant populationsof Sardinian
.
Acknowledgements. We are gratefully indebted to M. Poggesi (MZUF), who permitted examination of
specimens stored in the collections under her care, and to L. Rugiero, G. Paoloni, D. Persichetti, and F.M. Angelici
who co-worked in the Ž eld. F.M. Angelici, D. Capizzi, G. Gollmann, C. Jäggi, L. Rugiero, and an anonymous
referee critically commented on previous drafts.
Note added in proof
Coluber hippocrepis
A recent record of a probable
was communicated to one of the au-
thors(CC, who is the responsible for Sardinia of the “Atlas of Amphibians and Reptiles of

286
Claudia Corti, Luca Luiselli, Ernesto Filippi, Massimo Capula
Italy” in prep. and edited by the Societas Herpetologica Italica) for a northern locality of
the island, characterised by a wet environment.
References
Angelici, F.M., Inyang, M.A., Effah, C., Luiselli, L. (2000): Analysis of activity patterns and habitat use of
radiotracked African burrowing pythons, Calabaria reinhardtii. Isr. J. Zool. 46: in press.
¢
Bruno, S. (1977): Rettili d Italia: Serpenti. Florence, Aldo Martello Editore.
¢
Bruno, S. (1984): Guida ai Serpenti d Italia. Florence, Giunti Martello.
Bruno, S., Hotz, H. (1976): Coluber hippocrepis auf der Insel Sardinien. Salamandra 12: 69-86.
¢
¢
Bruno, S., Maugeri, S. (1990): Serpenti d Italia e d Europa. Milan, Editoriale Giorgio Mondadori.
Camerano, L. (1891): MonograŽ a degli OŽ di italiani — Parte seconda: Colubridi e MonograŽ a dei Cheloni
italiani. Mem. r. Accad. Sci. Torino 41: 403-480.
Capula, M., Luiselli, L. (1994): Resource partitioning in a Mediterranean lizard community. Boll. Zool. 61: 173-
177.
Capula, M., Luiselli, L., Rugiero, L., Evangelisti, F., Anibaldi, C., Trujillo Jesus, V. (1997): Notes on the food
habits of Coluber hippocrepis nigrescens from Pantelleria Island: a snake that feeds on both carrion and living
prey. Herpetol. J. 7: 67-70.
Corbett, K. (1989): Conservation of European Reptiles and Amphibians. London, Cristopher Helm.
Entomosarda s.n.c. (1993): AnŽ bi e Rettili. In: Atlante Faunistico della Provincia di Cagliari (Autori Vari, Eds).
Assessorato Tutela e Ambiente, Provincia di Cagliari, Cagliari.
Feriche, M., Pleguezuelos, J.M., Cerros, A. (1993): Sexual dimorphism and sexing of Mediterranean colubrids
based on external characteristics. J. Herpetol. 27: 357-362.
Filippi, E., Luiselli, L. (2000): Status of the Italian snake-fauna and assessment of conservation threats. Biol.
Conserv. 93: 219-225.
¢
¢
Gené, J. (1834): Description d un reptile mal connu et d un poisson nouveau. Mem. regia Accad. Sci. Torino 1,
37: 299-305.
Gené, J. (1839): Synopsis reptilium Sardinae indigenorum. Mem. regia Accad. Sci. Torino 2, 1: 1-31.
Gruber, U. (1989): Die Schlangen Europas. Stuttgart, Kosmos Naturfuehrer.
Hosmer, D.W., Lemeshow, S. (1989): Applied Logistic Regression. New York, Wiley & Sons.
Lanza, B. (1983): Ipotesi sulle origini del popolamento erpetologico della Sardegna. Lavori Soc. it. Biogeogr. 8
(1980): 723-744.
Lanza, B. (1993): AnŽ bi e Rettili. In: Vertebrata — Checklist delle specie della fauna italiana. Minelli, A., Ruffo,
S., La Posta, S., Eds, Bologna, Calderini.
Lenk, P., Herrmann, H.-W., Joger, U., Wink, M. (1999): Phylogeny and taxonomic subdivision of Bitis (Reptilia:
Viperidae) based on molecular evidence. Kaupia 8: 31-38.
Lenk, P., Joger, U. (1994): Genetic relationships between populations and intraspeciŽ c subdivision of Elaphe
longissima (Laurenti, 1768) as suggested by plasma protein electrophoresis and DNA Ž ngerprinting.
Amphibia-Reptilia 15: 363-373.
Luiselli, L., Angelici, F.M.(2000): Ecological relationships in two Afrotropical cobra species (Naja melanoleuca
and Naja nigricollis). Canad. J. Zool. 78: 191-198.
¢
Luiselli, L., Angelici, F.M., Akani, G.C. (2000): Large elapids and arboreality: The ecology of Jameson s green
mamba, Dendroaspis jamesoni, in an Afrotropical forested region. Contrib. Zool. 69: in press.
Mertens, R., Wermuth, H. (1960): Die Amphibien und Reptilien Europas. Frankfurt am Main, Kramer.
Nilson, G., Andrén, C., Ioannidis, Y., Dimaki, M. (1999): Ecology and conservation of the Milos viper,
Macrovipera schweizeri (Werner, 1935). Amphibia-Reptilia 20: 355-375.
Pleguezuelos, J.M., Moreno, M. (1990): Alimentación de Coluber hippocrepis en el SE de la peninsula Ibérica.
Amphibia-Reptilia 11: 325-337.
Poggesi, M., Agnelli, P., Borri, M., Corti, C., Finotello, P.L., Lanza, B., Tosini, G. (1995): Erpetologia delle isole
circumsarde. Biogeographia 18: 583-618.

Coluber hippocrepis in Sardinia
287
Puddu, F., Viarengo, M., Erminio, C. (1988): Animali di Sardegna: gli AnŽ bi e i Rettili. Cagliari, Edizioni Della
Torre.
Schätti, B. (1986): Morphologie und Systematik von Coluber algirus und C. hippocrepis (Reptilia: Colubridae).
Bonn. Zool. Beitr. 37: 281-293.
Schätti, B. (1993): Coluber hippocrepis (Linnaeus, 1758) — Hufeisennatter. In: Handbuch der Reptilien und
Amphibien Europas, Band III/1. Schlangen (Serpentes), p. 115-130. Boehme, W., Ed., Wiesbaden, Aula
Verlag.
Societas Herpetologica Italica (1996): Atlante provvisorio degli anŽ bi e dei rettili italiani. Ann. Mus. Civ. St. Nat.
“G. Doria” 91: 95-178.
Stefani, R. (1971): Ricerche zoologiche e botaniche nelle isole sarde di S.E. e S.O. In: Relazione preliminare
delle ricerche sulle popolazioni insulari compiute nel triennio 1965-1968(Pasquini, P., Ed.). Quaderni de “la
Ricerca ScientiŽ ca”, C.N.R., Roma 73: 30-36.
Received: November 10, 1999. Accepted: January 7, 2000.