SHORT NOTES HERPETOLOGICAL JOURNAL, Vol. 11, pp. 79-82 (2001) Malpolon monspessulanus (Pleguezuelos, 1997), on an island without lacertids and with introduced rat OBSERVATIONS ON THE NATURAL populations may be of interest. It provides an excellent HISTORY AND MORPHOMETRICS opportunity to test for the ecological adaptations of OF THE MONTPELLIER SNAKE, colubrids to Mediterranean environments characterized MALPOLONMONSPE SSULANUS, ON by a relatively small surface area and limited food re­ LAMPEDUSA ISLAND sources. Therefore, we decided to start an ecological project on the snakes of Lampedusa, also in considera­ (MEDITERRANEAN SEA) tion of the factthat this island is currently under heavy pressure fromtourism forthe whole of the summer sea­ CLAUDIA CORTI1, LUCA LUISELLI2 AND MARCO A. L. ZUFFI3 son, and its snake populations may likely be exposed to serious conservation threats (Corti & Luiselli, 200 1). 1 Dipartimento di Biologia Animale e Genetica, In this report we offer (!) a preliminary Universita degli Studi di Firenze, via Romana 17, I- morphometric analysis of Lampedusa specimens of the 50I25 Florence, Italy Montpellier snake (Malpolon monspessulanus) with 1Istituto di Studi Ambientali "Demetra " (E. N.I. S.p.A.), comparisons with museum vouchers from other geo­ via dei Cochi 48/B, I-OOI Roma, Italia; and F.I.Z. graphic regions; and (2) some notes on this species' 33 V., natural history. via Cleonia 30, I- 00152 Rome, Italy; and Department of Biological Sciences, Rivers State Un iversity of Science Detailed morphometric data on M. monspessu/anus and Technology, P.M.B. 5080, Port Harcourt (R ivers specimens from Lampedusa are not available in the lit­ State), Nigeria erature, given also the extreme rarity of such specimens in museum collections (but see Lanza Bruzzone, 3Museo di Storia Naturale e de/ Te rritorio (sez. Zoologia & I 960). Consequently we measured all the specimens e Anatomia Comparata), Universita di Pisa, via Roma 79, Calci (P isa), Italy stored in the collections of the Zoological Museum "La I-56011 Specola'', Florence (MZUF), which is likely to store the Key words: snake, diet, morphology, conservation highest number of Lampedusa specimens available in the world's public collections. In addition, we exam­ Lampedusa is an offshore island of Sicily, approxi­ ined specimens in the collections of MSNG (Museo I). mately 20.2 km2 in surface area (Fig. From a Civico Storia Naturale, Genoa) and of ZFMK geological point of view it is part of the African plat­ (Zoologisches Forschungstinstitut und Museum A. form. Once covered by Mediterranean woodlands, the Koenig Bonn). The specimens examined were labelled whole surfaceof this island is now entirely deforested, as follows: MZUF 585, I 1352, 11353, 32640, 35108, and characterized by stony and rocky formations ( ei­ 36591, 38063, 38066, 38067 (all from Lampedusa); ther man-made or natural) on grasslands, and MZUF 128, 7935, ZFMK 48836, 4883 9, 48840 (all uncultivated land (mainly degraded Mediterranean from Cyprus); MZUF 1328, 1330, 6983, 6984, I 0563, maquis). This small island is very important in ecologi­ MSNG 48613 (all from France or north-western Italy); cal terms because of its rare and diverse wildlife MZUF 9054, 23876, 33026, 36577, 38068 (all from the (Meschini & Frugis, 1993), including several endemics Iberian peninsula); MZUF 1249, 1250, 11337, 11338, (Massa, 1995 and literature cited therein). With regard 31677, 31678, 38064 (all from Croatia and north-east­ to herpetofauna, Lampedusa is noteworthy because of ern Italy); MZUF 1254, 2614, 12483, 19935, 29754; some important peculiarities, including the factthat it is MSNG 306 19, 30620, 31582, 36538, 37852, ZFMK (I) an important reproduction site for marine cheloni­ 23042, 23044, 23047 (all from Tunisia, Egypt, and Is­ ans (Bruno, 1986; Jesu, 1995); (2) one of the few rael). Mediterranean islands without any representative of the Every museum specimen was measured (to ± 0.1 family Lacertidae (Corti, Lo Cascio, Vanni, Turrisi & cm) forsnout-vent-length (SVL), tail length (TL), head Vaccaro, 1997); and (3) is inhabited by only two spe­ length (HL), head width (HW), interorbital length cies of snakes, i.e. the colubrids Ma croprotodon (INT-ORB), and number of ventrals. cucullatus and Ma/po/on monsp essulanus (Bruno & Field research was conducted throughout several Maugeri, 1990; Corti et al. , 1997). short-term survey periods by Massimo Capula, Natural history information on Lampedusa snake Giovanni Di Claudio, Lorenzo Rugiero (who kindly populations is very limited (Bruno Maugeri, 1990). & . provided us with their unpublished observations) and The presence of one species with a lizard-based diet, Luca Luiselli between spring I 984 and autumn 1989; i.e. Macroprotodon cucullatus (e.g. see Bruno & and by Claudia Corti and Stefano Vanni during spring Maugeri, 1990), and one with an ontogenetic shiftfrom I 99 1. These periods ranged from three to seven days a lizard-based diet to a bird and mammal-based diet, i.e. each, from March to October, and involved searching for snakes along randomly selected paths. Correspondence: C. Corti, Dipa�imento di Biolog a . � Free-living specimens were individually marked by Animale e Genetica, Universita degh Stud1 d1 Firenze, via Romana 17, 1-50125 Florence, Italy. E-mail: scale-clipping, measured for SVL (to ± 0.5 cm) and [email protected] processed to determine their diet. The snakes were pal- 80 SHORT NOTES 12"30' E General MANOVA models indicated no effects on the �35:l(rN regression ofTL against SVL (independent variable) of both sex and locality (sex: F=3 .97, df=3 ,3 l, P=0.140; locality: F=3.19, df=3,3 1, P=0.1 85). This means that males and females do not differ in terms of body pro­ portions (i.e. tail length relative to body length), contrary to most snake species studied to date (e.g. cf. Shine, 1994). Neither is there any specific geographic variation in this general pattern.In this respect, our re­ sults fully agree with literature data (e.g. see Pleguezuelos, 1997, and literature cited therein). However, there were important geographical varia­ tions in other morphometric and meristic traits. For FIG. I. Position of Lampedusa island in the western instance, both sex and locality had significanteffects on Mediterranean area. The line represents 500 km; the enlarged the regression of HL against SVL (general MANOVA - circle represents geographic co-ordinates of the studied sex: F=8.37, df=3,40, P=0.050; locality: F=36.66, island, from the small circle. df=3,40, P=0.0062), and Tukey honestly significant post-hoe test (Sokal and Rohlf, 1981) indicated that ( 1) pated in the abdomen until regurgitation of the ingested males had larger heads than females forthe same body food or defecation occurred. The prey items were iden­ length, and (2) there were differences between areas in tified to the lowest taxon possible. The snakes were relative head size. Nevertheless, Lampedusa specimens then forced to reingest the disgorged prey. However, were not distinctive (relatively to the average of the vouchers of some of the disgorged prey items are de­ whole sample examined) in terms of head size patterns. posited in the herpetological collections of M. Capula With regard to the number of ventral scales, there and L. Rugiero (both collections in Rome). was no significant correlation between SVL and Dorsal col oration of Lampedusa specimens > 1 OOO number of ventrals (Spearman 's r=0.226, ANOV A: mm SVL was normally uniform green or grey-green, F=2 .42, df=l,45, P=0.126). The mean number of whereas smaller specimens had many dark spots on a ventrals in relation to geographic area is presented in grey-greenish background. Moreover, fe males tended Table 1. There was a very strong effect of locality on to be more spotted than males. These colour patterns the numbers of ventrals (ANOV A: F= l 7.84, df=5, 41, were consistent in both the living (n=l 7) and preserved P<0.0001), and general MANOVA models indicated (n=9) specimens examined by us. that both locality and sex (as the covariates) signifi­ For morphometric analyses, museum voucher speci­ cantly affected the regression of numbers of ventrals mens were grouped into six geographical groups: (1) against SVL (locality: F=48.53, df=3 ,42, P=0.0040; Lampedusa, (2) Cyprus, (3) France + NW Italy, (4) Ibe­ sex: F= l 13.36, df=3,42, P=0.0011), with the interac­ rian peninsula, (5) Croatia + NE Italy, (6) north Africa. tion effe ct between sex and locality being also SVL's of museum vouchers were not significantly dif­ statistically significant (F=77.58, df=2,42,P= 0.012). A ferent between sexes (Kruskal-Wallis ANOVA: Tukey honestly significant post-hoe test indicated that F=0.30, df=l,45, P=0.586) or among geographical specimens from Spain, France and Lampedusa had sig­ groups (Kruskal-Wallis ANOVA: F =1.61, df=l,41, nificantly higher ventral counts for the same body P=0.177). Since SVL was significantly correlated with length than specimens from North Africa, Cyprus and TL, HL, HW, and INT-ORB (in all cases at least Croatia. The number of ventral scales is generally as­ Spearman's r> 0.6, P<0.0001), residual scores from the sumed to be an indirect way of counting vertebrae general regressions of these parameters to SVL were number, in. both viperid (Saint Girons, 1978; Luiselli & used to test for size-corrected intergroup differences. Zuffi,20 01; Zuffi unpubl.), and colubrid species (Corti, Zuffi & Luiselli, 2000; Zuffi, 2000; Zuffi et al. , TABLE I. Mean number of ventral scales of museum unpubl.). Significant differences may well lead to the vouchers of Malpolon monspessulanus in relation to critical re-evaluation of the taxonomic position of the geographic area.