Hereditas 129: 187-194 (1998) On the mtDNA restriction patterns variation of the Iberian wood mouse (Apodemus sylvaticus). Comparison with other west Mediterranean populations J. R. MICHAUX!, R. LIBOIS!, M. G. RAMALHINHO2 and C. MAUROIS! 1 Unité de recherc/zes zoogéographiques, Institut de Zoologie, Universitéde Liège, Liège, Belgium 2 Centro de Biologia ambiental da Universidade de Lisboa, Lisboa, Portugal Michaux, J. R., Libois, R., Ramalhinho,M. G. and Maurois, C. 1998. On the mtDNA restriction patterns variation of the lberian wood mouse (Apodemus sylvaticus). Comparison with other west Mediterranean populations.-Hereditas 129: 187-194. Lund, Sweden. ISSN 0018-0661. Received April 23, 1998. Accepted November 9, 1998 . ln previous studies, the presence of three main lineages of wood mice mtDNA was describedin western Europe: a first one distributed. from the Pyrenees to Scandinavia, a Thyrrenian one occurring in peninsular Italy, Elba, Corsica and Sardinia and a third one restricted to Sicily and Marettimo. Do the Iberian wood DÙce belong to one of these lineages? ln order to answer this question, animais were trapped ail over the Iberian peninsula (Il sites) as weil as in three of the Balearic islands. Comparisonswith specimens from the above mentioned lineages were made. From 158 animais trapped in 30 sites, 78 different mtDNA restriction patterns were obtained and compared using the NEI and LI index (Nei M and Li WH, (1979). Mathematical model for studying genetic variation in terms of restriction endonucleases. Froc. Natl. Acad. Sci. USA 76: 5269-5273) of nucleotide divergence. A neighbour-joining tree, with a specimen of Apodemus ftavico/lis as outgroup, was then computed from the similarity matrix. Apodemus ftavicollis is weil separated from ail the A. sylvaticus and ail the lberian restriction patterns are clustered with those of continental France, showing a great similarity level betweenthe Iberian and north-west European animais. This group is weil separatedfrom the Sicilian and from the Tyrrhenian ODes. These results suggestthat the Pyrenees are not a biogeographic barrier for the wood mouse and that the postglacial recolonisation of western Europe by that species has its origin in populations which, during the latest Ice age, were living in refuges situated in southern France or in the Iberian peninsula. Becausethe Balearic restriction patterns are clusteredin a lineagejoining Iberian patterns at a low divergence leveI, we suggest that their origin is continental Spain or southern France. R. Libois, Unité de recherches zoogéographiques, Institut de Zoologie, Université de Liège, quai Van Beneden, 22, B-4020 Liège, Belgium. E-mail: [email protected]mtDNA lineages,has prevented the ltalian animaIs to move northwards after glaciation. We wonder if the lberian wood mice are distinct from the north- westernODes and if the Pyrenees played the samefoIe as the Alps at the end of the Iast glaciation. Two distinct subspecies are currently reported in the lberian peninsuIa: Apodemus sylvaticus dichrurus (Rafinesque, 1814) considered as the Mediterranean subspecies and A. s. callipides (Cabrera, 1907) sup- posed to inhabit the Cantabric and the Pyrenean chains and even to extend towards the Massif Central (MILLER 1912; SAINT GIRONS 1966, 1973). ln their morphological study of Portuguese speci- mens,RAMALHINHO and MADUREIRA (1982) did not find any crucial difference between the north Por- tuguese animaIs thought to belong to callipides and the south Portuguese onesascribedto dichrurus. Nev- ertheless, animaIs are sIightIy bigger in the south than in the north. This fact is highlighted by SANS-COMA et aI. (1987) in Spain too. Despite that thesemorpho- 10gicaI differences are slight, the existence of two distinct fonDS of the wood mouse in the lberian peninsula cannot be excluded. ln a previous study (MICHAUX et al. 1996), two lineages of wood mice (Apodemus sylvaticus) mtDNA were described in western Europe. The fust one (northwestern) extendsfrom the Pyrenean mountains to Scandinavia and the second one is distributed aIl over the Italian peninsula and in the Tyrrhenian islands (Elba, Corsica, Sardinia). A third group is recognised in Sicily and in Marettimo (MICHAUX et al. 1998). Till now, no information about the genetic vari- ability of the Iberian wood mouse mtDNA is avai1ab1e. As the wood mouse is mainly a forest dweller whose altitudinal and latitudinal distribution boundaries are close to the cree limit, it is conceivable that during the 1atest Ice age, its distribution was restricted to the remaining forest areas of southern Europe e.g. in southern France and in some parts of the Italian and Iberian peninsulas. Consequently, north-west Europe must have been reached by wood mice originating from these forest zones. Certain1y, the alpine chain which constitutes the north-occidental boundary between the two main
Transcript
Hereditas 129: 187-194 (1998)
On the mtDNA restriction patterns variation of the Iberian woodmouse (Apodemus sylvaticus). Comparison with other westMediterranean populationsJ. R. MICHAUX!, R. LIBOIS!, M. G. RAMALHINHO2 and C. MAUROIS!1 Unité de recherc/zes zoogéographiques, Institut de Zoologie, Université de Liège, Liège, Belgium2 Centro de Biologia ambiental da Universidade de Lisboa, Lisboa, Portugal
Michaux, J. R., Libois, R., Ramalhinho, M. G. and Maurois, C. 1998. On the mtDNA restriction patterns variation of thelberian wood mouse (Apodemus sylvaticus). Comparison with other west Mediterranean populations.-Hereditas 129:187-194. Lund, Sweden. ISSN 0018-0661. Received April 23, 1998. Accepted November 9, 1998 .
ln previous studies, the presence of three main lineages of wood mice mtDNA was described in western Europe: a firstone distributed. from the Pyrenees to Scandinavia, a Thyrrenian one occurring in peninsular Italy, Elba, Corsica andSardinia and a third one restricted to Sicily and Marettimo. Do the Iberian wood DÙce belong to one of these lineages?ln order to answer this question, animais were trapped ail over the Iberian peninsula (Il sites) as weil as in three of theBalearic islands. Comparisons with specimens from the above mentioned lineages were made. From 158 animais trappedin 30 sites, 78 different mtDNA restriction patterns were obtained and compared using the NEI and LI index (Nei M andLi WH, (1979). Mathematical model for studying genetic variation in terms of restriction endonucleases. Froc. Natl.Acad. Sci. USA 76: 5269-5273) of nucleotide divergence. A neighbour-joining tree, with a specimen of Apodemusftavico/lis as outgroup, was then computed from the similarity matrix. Apodemus ftavicollis is weil separated from ail theA. sylvaticus and ail the lberian restriction patterns are clustered with those of continental France, showing a greatsimilarity level between the Iberian and north-west European animais. This group is weil separated from the Sicilian andfrom the Tyrrhenian ODes. These results suggest that the Pyrenees are not a biogeographic barrier for the wood mouseand that the postglacial recolonisation of western Europe by that species has its origin in populations which, during thelatest Ice age, were living in refuges situated in southern France or in the Iberian peninsula. Because the Balearicrestriction patterns are clustered in a lineage joining Iberian patterns at a low divergence leveI, we suggest that their originis continental Spain or southern France.
R. Libois, Unité de recherches zoogéographiques, Institut de Zoologie, Université de Liège, quai Van Beneden, 22, B-4020
mtDNA lineages, has prevented the ltalian animaIsto move northwards after glaciation. We wonder ifthe lberian wood mice are distinct from the north-western ODes and if the Pyrenees played the same foIeas the Alps at the end of the Iast glaciation.
Two distinct subspecies are currently reported inthe lberian peninsuIa: Apodemus sylvaticus dichrurus(Rafinesque, 1814) considered as the Mediterraneansubspecies and A. s. callipides (Cabrera, 1907) sup-posed to inhabit the Cantabric and the Pyreneanchains and even to extend towards the Massif Central(MILLER 1912; SAINT GIRONS 1966, 1973).
ln their morphological study of Portuguese speci-mens, RAMALHINHO and MADUREIRA (1982) did notfind any crucial difference between the north Por-tuguese animaIs thought to belong to callipides andthe south Portuguese ones ascribed to dichrurus. Nev-ertheless, animaIs are sIightIy bigger in the south thanin the north. This fact is highlighted by SANS-COMAet aI. (1987) in Spain too. Despite that these morpho-10gicaI differences are slight, the existence of twodistinct fonDS of the wood mouse in the lberianpeninsula cannot be excluded.
ln a previous study (MICHAUX et al. 1996), twolineages of wood mice (Apodemus sylvaticus) mtDNAwere described in western Europe. The fust one(northwestern) extends from the Pyrenean mountainsto Scandinavia and the second one is distributed aIlover the Italian peninsula and in the Tyrrhenianislands (Elba, Corsica, Sardinia). A third group isrecognised in Sicily and in Marettimo (MICHAUX etal. 1998).
Till now, no information about the genetic vari-ability of the Iberian wood mouse mtDNA isavai1ab1e.
As the wood mouse is mainly a forest dwellerwhose altitudinal and latitudinal distributionboundaries are close to the cree limit, it is conceivablethat during the 1atest Ice age, its distribution wasrestricted to the remaining forest areas of southernEurope e.g. in southern France and in some parts ofthe Italian and Iberian peninsulas.
Consequently, north-west Europe must have beenreached by wood mice originating from these forestzones. Certain1y, the alpine chain which constitutesthe north-occidental boundary between the two main
distinctions, however. give no information about theactual origin of the animais.
ln this paper, we intend ta study the mtDNAvariability of Iberian wood rnice, including animaisfrom .the Balearic is1ands and to compare theirmtDNA restriction patterns with others from differ-ent continental areas. ln this way, we hope to getmore information about the influence of the Pyreneeson the postglacial recolonisation of Europe by thewood mouse, to know ta what extent the lberianpopulations of this species are differentiated and tounderstand the origin of its settlement in the Balearicarchipelago.
MATERIAL AND METHODS
The wood mice and the yellow necked mouse weretrapped in 'Manufrance' or in 'Sherman' traps previ-ously baited. The sampling localities (two lines of 40traps at 5 m. interval and distant of about 200 m.from each other) are spread aIl over Spain and Portu-gal as well as in Ibiza, Mallorca and Menorca. Un-successful attempts were also made on Formentera. Aspecial effort of capture was made in" the vicinity ofLa Corona (locus typicus of the subspecies A. s.callipides) and also on the French slopes of thePyrenees as weIl as in the Massif Central. Additional
As far as the Balearic islands are concemed, it isweIl known that the endemic mammal fauna of thearchipelago bas been completely replaced by man-in-troduced species. Most of these introdûctions tookplace during the Holocene. From archaeozoologicaldata, it is established that the presence of the woodmouse dates back to at least 6000 years B.P. inMaIlorca and to 3300 y B.P. on Menorca (REUMERand SANDERS 1984; VIGNE and ALCOVER 1985),when human exchanges were already quite commonbetween the archipelago and either the European orAfrican coasts (GUILAINE 1976; CAMPS 1990). Con-sequently, different hypotheses about the geographicorigin of the species in these islands can be inferred,especia11y if we consider that some smaIl mamma1insu1ar parasites (fieaS) undoubtedly have an Africanorigin whereas others have clearly European orEurasiatic affinities (BEAUCOURNU and ALCOVER1984).
Owing to its overall bigger sile and heavier weight,compared to continental references (ALCOVER andGOSALBEZ 1988), the wood mouse of Ibiza has re-ceived a distinct subspecific status: A. s. eivissensisALCOVER 1977. Still heavier and bigger than those ofIbiza, the wood mice of Formentera were also de-scribed as a separate subspecies: A. s. frumentariaeSANS-COMA and KAHMANN 1977. These taxonomic
~"
~
iof~~!!
,!
Fig. 1. Geographic distribution of the wood mice trapping localities. The codes refer to Table
Hereditas 129 (1998)
February 93May 93May 94May 94April 94October 93June 95October 95February 96
. The localities with an . are the Saille as in MICHAUX et a
animaIs from France and Italy were included in theanalyses to allow useful com~arisons. Fig. 1 showsthe geographic distribution of the sampling pointsand Table l gives information about the sample sizeand the precise location of each trapping site.
Trapped animaIs were maintained in the laboratoryand killed to allow the isolation of mitochondriasfrom fresh tissue (heart, spleen, liver and kidneys) bydifferentiated centrifugations, according to LANSMANet al. (1981). mtDNA was then isolated and purifiedby alkaline lysis and ether-phenol extraction (PAL VAand PAL VA 1985) and finally digested with two re-striction enzymes: Hae III (GGCC) and Rsa 1(GTAC) (Boehringer-Mannheim or BRL). One to 3~l mtDNA solution containing 20-40 ng DNA weredigested for 1-2 h in a 10 ~l reaction volume contain-ing one enzyme unit. The mtDNA fragments were
1. (1996) and are numbered in accordance.
isolated by electrophoresis on 4% PAA gels accordingto TEGELSTROM (1986) and polymorphism visualisedby the silver staining protocol of GUILLEMETTE and
LEWIS (1983). AlI distinctive mtDNA restriction frag-ment patterns were assigned a number when pro-duced by Rsa 1 or an alphabetical code whenproduœd by Hae III. ln this way, each animal was
assigned an alphanumeric code. AIl specimenssharing the SaUle composite restriction pattern wereconsidered as belonging to the same mtDNA matri-
lineal clone.The level of similarity between two individuals was
computed using the NEI and LI (1979) index. Aneighbour-joining tree was cvnstructed from the simi-
larity matrix using one individual of yellow-neckedmouse (A. fiavicollis) as outgroup. The robustness of
the inferences was assessed through bootstrap analY4
J. R. Michaux et al. Hereditas 129 (1998)190
.2,sis (1000 repeats) (Treecon programme, versionVAN DE PEER and DE WACHTER 1996).
RESULTS
A total of 159 animaIs (158 A. sylvaticus and 1 A.flavicollis) representing 31 populations have beenanalysed using ,two restriction endonucleases. Foreach animal, we obtained a total of 42-46 fragments,a number which is appropriate for a reliable estima-tion of divergence levels among samples (FERRIS etal. 1983a,b). The tables of the different fragment sizesobtained after digestion by Hae III and Rsa lareavailable upon request from J .R.M.
We obtained 79 different haplotypes from the ani-maIs examined. Some of them are illustrated in Fig. 2(Hae III) and Fig. 3 (Rsa 1). No difference wasobserved in the totallength of the mtDNA as recal-culated after interpretation of each restriction pat-tern, a potential problem raised by TEGELSTROM andJAAROLA (1989). ln Table 2, a haplotype diversityindex (Shannon H' with log in base 2: SHANNON1948) has been calculated. The obtained values are instrong correlation with sample size: H' = 1.920 +0.0504 x nb of animaIs (r = 0.999). This is not thecase with the regularity index (H' IH max) which isalways very high, indicating that, for any sample size,the animaIs are quite uniformly distributed betweenthe different haplotypes. A neighbour-joining tree(Fig. 4) shows that the yellow-necked mouse is verydivergent of aIl the woodmice samples (mean level ofgenetic divergence: 6.27%). The node separating these
two species is very robust (BP of 100%). Moreover,inside the woodmice group, it can be seen that aIl theIberian clones inc1uding the Balearic ones are clus-tered in a single subgroup including the F~ench sam-pIes whereas Italian animaIs are separated in a secondsubgroup and Sicilian individuals in a third one. Asexpected from our previous study (MICHAUX et al.1996), the IÏiean level of the nucleotide sequenœdivergence between these three groups is relativelyhigh (between 2.6 and 3.7%). This separation is veryweIl supported with respective bootstrap values of 90and 97%. The divergence within the Iberian/Frenchgroup is low (1.1%), approximately the same as calcu-lated previously for the 'north-western' group (0.8%)(MICHAUX et al. 1996).
Interestingly, quite different restr1ction patterns arefound for animaIs caught at one locality, the diver-gence level between them being sometimes larger than1% (e.g. patterns N26, L25, L27, 025, R24 in sam-pling point F2 or BB63, AZ58, AZ63, AZ70 andBA58 in point 83). AnimaIs caught very far fromeach other sometimes share the same restriction pat-tern [e.g. pattern 829 in œntral France (point F6) andin Catalunya (point S5), or AQ 62 present in thecentral French Pyrenees(F8) and near Valencia (S7)].
Considering the islands, it can be seen that aIl theisland clones are clumped together in a same sub-group and that the dominant pattern for Rsa 1 is thesame (79) on Mallorca and Menorca. ln fact, theisland patterns areverysimilar toeach other, .1eadingto a particularly low 1evel of genetic divergence be-tween the insular animals (mean p= 0.3%).
5 6 7 8 9 103 L 11 122, 4
-c 1650
1650 --c790
-562790-
_366
562-
366-
Fig. 2. Restriction patterns after Hae III endonuclease digestion of wood mouse mtDNA from 1: 15 (pattern A); 2: Ib1(pattern V); 3: F7 (pattern Ag); 4: Il (pattern A); 5: S7 (pattern Aq); 6: F8 (pattern Aq); 7: S3 (pattern Az): 8: Ib1 (patternV); 9: Me1 (pattern Bj); 10: F4 (pattern An); Il: F5 (pattern Al); 12: F7 (pattern Ag); 13: Me1 (pattern Bj); 14 and 15:Ma2 (pattern BI). The lanes marked L contains Lambda phage DNA digested with Restriction endonuclease Bg! 1 topro duce fragment size markers (size in BP indicated on the right and on the left).
-1650
_790
_366
-267
Fig. 3. Restriction patterns after Rsa l endonuclease diges-tion of wood mouse mtDNA from 1: Mal (pattern 80); 2:Mel (pattern 79); 3: Ibl (pattern 53); 4: 83 (pattern 63); 5:85 (pattern 29) and 6: Il (pattern 1). The size scale in BP isindicated on the right.
DISCUSSION AND CONCLUSIONS
The mean leveJ of $equence divergence of nucleotidesbetween .Iberian and French animaIs is quite low andtherefore Iberian animaIs can be considered as a partof the 'north-western' group whose distribution areaextends to Scandinavia (fEGELSTRÔM and JAAROLA1989; VAN ROMPAEY 1989; MICHAUX et al. 1996).These data strongly suggest that the post-glacial re-colonisation of north-west Europe is the consequenceof the spreading of a clade which found refuge in theIberian peninsula or in the very southern Franceduring the latest Ice age. This colonisation route isidentical for the wood mouse, the brown bear (Ursusarctos) and some white oaks (Quercus spp.) haplo-types (fABERLET et aI., 1998) but differs for otherspecies like the grasshopper Chorthippus parallelus
(COOPER et al., 1995) for which the Pyrenees are asuture zone between two groups of populations.
So, the Pyrenees do not constitute a serious obsta-cle for the wood mice north-south dispersaI move-ments. The north-south orientation of the mainvalleys and the low altitude of many passes betweenthe southern and the northern parts of the roountainchain leads to a continuity in the forest cover andperhaps favours thc wood mouse settlement. More-over, either near the Atlantic coast or near the Med-iterranean border, this forest cover is alsouninterrupted between France and Spain due to suffi-cient rainfalls, to suitable edaphic conditions and tothe low altitude of the mountains. The situationprevailing in the Alps is contrasting, at least in thewestern part, because the summit line is higher, thealtitude of only a few passes being just under or nearthe tree line. Near the Mediterranean sea, the pied-mont is very dry and the soil is stony, leading toecological conditions unsuitable for the wood mouse.ln the pas t, the ice sheet was more developed in theAlps th an in the Pyrenees, probably preventing thenorthward expansion of the mice populations isolatedin the ltalian refugium. This Alpine discontinuity hasbeen evidenced in many other species including mam-mals (Sorex araneus, Arvicola terres tris: TABERLET etal., 1994; TABERLET et al. 1998), amphibians (Tritu-rus sp.: WALLIS and ARNTZEN 1989), fish (Salmotrutta: BERNATCHEZ et al. 1992) or insects{Chorthip-pus parallelus: COOPER et al. 1995; Apis mellifica:GARNERY et al. 1992).
As far as the divergence level between the Iberiansamples is very low, it can be assumed that there is noevidence for two distinct wood mice groups withinthe Iberian Peninsula. This conclusion is in completeaccordance with the results of the morphologicalstudy of RAMALHINHO and MADUREIRA (1982).Moreover, a multivariate (pCA, multiple discrimi-nant analysis) craniometric study (MICHAUX 1996)shows that the lberian wood mice are hardly distin-guished from those trapped in France, Belgium andGermany.
ln conclusion, we suggest that the existence ofApodemus s. dichrurus in Spain and in southem
Table 2. Woodmouse haplotype diversity in the difJerent regions
No. haplotypes No. animaIs Diversity index Equitability index
_j'lg. 4. Neighbour joining tree of the 79 Apodemus haplotypes. The Sicilian and ltalianb\lbgroups are presented as single branches. Number over branches indicate bootstrapp\lpport (1000 replicates).
j;'rance has to be rejected since this taxon is described rated from both the Italian and the north-westerntrom Sicily and because the Sicilian mtDNA restric- ones at a high divergence level (MICHAUX et al. 1998(ion patterns form a. particular lineage, which is sepa- and this study). Furthermore, the animals
Hereditas 129 (1998)
trapped near the loCtIS lypicus of A. s. callipides revealrestriction patterns which are very similar to those ofother Spanish or French animais. 'fherefore, andfollowing CORBET (1978), we propose to consider A.s. ca/lipides as an invalid subspecies and to put it insynonymy with the nominal one.
The restriction patterns round in the Balearic ani-maIs are clearly of the 'north-western' 'type, indicat-ing that the origin of introduced animaIs Is theSpanish or French mainland, provided simi:larmtDNA restriction patterns are not present ln north-ern Africa. The genetic similarity observedwithin andbetween the populations of neighbouring islands asweIl as their grouping in a separate branch of thecladogram suggests a small founder population.
ACKNOWLEDGEMENTSThis research has been supported by an IRSIA-FRIAfellowship to the first author, by grants of the FNRScovering field collects in Spain and by a FRFC grant (n°2.4547.89). For their help in trapping or in keeping theanimaIs in captivity, we are grateful to J.P. Clara, R. Fons,C. Hallet, Th. Kervyn, S. Kinet, S. Mas-Coma,M.L.Mathias, J. Ninane F. Ronveaux, M. Santos-Reis, andb. Sirugue. We are particularly indebted to R. F. Matagnefor his invaluable logistic help. Thanks are also due to A.Haidon for herhelp in the revision of the English languageand two anonymous referees for their helpful comments.
mtDNA variabi/ity of fberianwood mouse(Apodemus sylvaticus) 193
Ferris SD, Sage RD, Prager EM, Ritte U and Wilson AC,(1983b). Mitochondrial DNA evolution in mice. Genet-ics 105: 681-721.
Garnery L, Cornuet lM and Solignac M, (1992). Evolu-tionary history of the honey-bee Apis mellifera inferredfrom mitochondrial DNA analysis. Mol. Ecol. 1: 145-154.
Guilaine J, (1976). Premiers bergers et paysans de"l'Occi-dent méditerranéen. Mouton, Paris, La Haye.
Guillemette JG and Lewis PN, (1983). Detection of sub-nanogram quantities of DNA and RNA on native anddenaturing polyacrylamide and agarose gels by silverstaining. Electrophoresis 4: 92-94.
Lansman RA, Sade RO, Shapira JF and Avise JC, (1981).The use of restriction endonucleases to measure mito-chondrial DNA sequence relatedness in natural popula-tions. III Techniques and potential applications. J. Mol.Évol. 17: 214-226.
Michaux JR, (1996). Biogéographie du mulot sylvestre(Apodemus sylvaticus) dans le bassin méditerranéen oc-cidental: étude génétique de l'origine des peuplements etmicroévolution en milieu insulaire. Ph.D. ZoologicalInstitute, UIÙversity of Liège, Liège (Belgium).
Michaux JR, Filippucci MG, Libois RM, Fons R andMatagne RF, (1996). Biogeography and taxonomy ofApodemus sylvaticus (the woodmouse) in the Tyrrhe-nian region: enzymatic variations and mitochondrialDNA restriction pattern analysis. Heredity 76: 267-277.
Michaux JR, Sara M, Libois RM and Matagne RF, (1998).Is the woodmouse (Apodemus sylvaticus) of Sici1y rea11ya separate species? Belgian J. Zool. 128: 209-212.
Miller GS, (1912). Catalogue of the mammals of westernEurope (Europe exclusive of Ru~sia) in the collections ofthe British Museum. British Museum, Natural History,London.
Nei M and Li WH, (1979). Mathematical mode1 for study-ing genetic variation in terms of restriction endonucle-ases. Proc. Natl. Acad. Soi. USA 76: 5269-5273.
Palva TK and Palva T, (1985). Rapid isolation of anima!mitochondrial DNA by alkaline extraction. FEBS Lett.192: 267-270.
Ramalhinho MG and Madureira ML, (1982). Contributionto the knowledge of Apodemus Kaup in Portugal(Mammalia, Rodentia). Arq. Museu Bocage 1: 415-433.
Reumer JWH and Sanders EAC, (1984). Changes in verte-brate . fauna of Menorca in Prehistoric and classica1times. Zeitschr. f. Saug~tierkde 49: 321-325.
Saint Girons MC, (1966). Etude du genre Apodemus Kaup,1829, en France J. Mammalia 30: 547-600.
Saint Girons MC, (1973). Les Mammifères de France et duBenelux. Doin, Paris.
Sans-Coma V and Kahmann H, (1977). Die Waldmaus derPityuseninsel Formentera. - Saugetierkdl. Mitt. 19: 363-
365.Sans-Coma V, Rosado LM and Gosâlbez J, (1987). Un
estudio de la morfometria y deI comienzo de la actividadreproductora de Apodemus sy1vaticus (L., 1758) en laPeninsula Ibérica basado en la consideraciôn de lospe1ajes y las mudas coma indicadores de la edad. ln:Mamiferos y He1mintos (eds V Sans-Coma, S Mas-Coma y J Gosalbez) Ketres, Barcelona, p. 89-98.
Shannon CE, (1948). A mathematical theory of communi-cation. Bell. System Technol. J 27: 379-423, 623-656.
Taberlet P, Fumagalli L and Hausser J, (1994). Chromoso-mal versus mitochondria1 DNA evolution: cracking theevolutionary history of the southwestern European pop-
REFERENCES
Alcover JA, (1977). The long tailed field mouse, Apodemussylvaticus (L.,1758) from the island of Ibiza. Pityusics.Saugetierkdl. Mitt. 25 (3): 204-213.
Alcover JA and Gosalbez J, (1988). Estudio comparado dela fauna demicromamiferos de las islas Baleares yPitiusas. Bull. Ecol. 19: 321-328.
Beaucournu JC and Alcover JA (1984). Siphonaptera fromsmail terrestrial mammals in the Pityusic is1ands. ln:Biogeography and Ecology of the Pityusic islands, (edsH Kuhbier, JA Alcover and C Guerau d'Arellano Tur)W Junk, The Hague, p. 377-392.
Bernatchez L, Guymard R and Bonhomme F, (1992).DNA sequence variation of the mitochondrial controlregion among geographically remote European browntrout Salmo trotta populations. Mol. Ecol. 1: 161-173.
Camps G, (1990). Navigation et migrations en Méditerra-née de la Préhistoire à nos jours. Collioure, 1983- Sète1985. CNRS, Paris, p. 138-156.
Cooper SJ, Ibrahim KM and Hewitt GM, (1995). Post-glacial expansion and genome subdivision in the Eu-ropean grasshopper Chorthippus parallelus. Mol. Ecol.4: 49-60.
Corbet GB, (1978). The mammals of the Palearctic region:a taxonomic review. British Museum (Natural History)Cornell University Press, London.
Ferris SD, Sage RD, Huang CM, Nielssen JT, Ritte U andWilson AC, (1983a). Flow of mitochondrial DNAacross a species boundary. Proc. Nat!. Acad. Sci. USA80: 2290-2294.
J. R. Michaux et al.194
Van Rompaey J, (1989). Variatie en differentiatie van mi-tochondf'ial DNA ,in Clethrionomys glareolus (Schre-ber, 1780), Apodemus sylvaticus (Linnaeus, 17$8) enApodemus fiavicQllis (Melchior, 1834) (Mammalia, Ro-dentia) in West Europa. Ph.D. Thesis, University ofAntwerp.
Vigne JD and AlcQver lA, (1985). Incidences des relationshistoriques entre l'homme et l'animal dans la composi-tion actuelle du peuplement amphibien, reptilien et,mammalien des îles de Méditerranée occidentale. ActesllUe Congrès nat. Soc. Savantes (Montpellier, i 985).Sect. Sci. 2: 19-91.
Wallis GP and Amtzen GW, (1989). Mitochondrial-DNAvariation in the crested newt superspecies: limited cyto-pl;ismic gene fiow among species. Evolution 43: 88-1()11
ulations of the Sorex araneus group (Mammalia, Insec-tivora). Evolution 48: 623-6J6. ..
Taberlet P, Fumagalli L, Wust-Saucy AG and Cosson !TF,(1998). Comparative phylogeography and ,postglacialcolonization routes in Europe. Mol. Ecol. 7: 453-464.
Tegelstrôm H, (1986). Mitochondrial DNA in Raturai pop-ulations: an improved routine for the screening of thegenetic variation based on sensitive silver-staining. Elec-trophoresis 7: 226-229.
Tegelstrôm H and Jaarola M, (1989). Genetic divergence inmitochondrial DNA between the wood mouse (Apode-mus sylvaticus) and the yellow necked mouse (Apode-mus flavicollis). Hereditas Ill: 49-60.
Van De Peer Y and De Wachter R, (1996). Construction ofevolutionary distance trees with TREECON for Win-dows: accounting for variation in nucleotide substitutionrate among sites. Comput. Appl. Biosci. 13: 221-230.
