Page 5 - MICHAUXetal1998

Hereditas 129 (1998) mtDNA variabilitv of Iberian wood mouse (Apodemus sylvaticus) 191

123456 (COOPERet al., 1995) for which the Pyrenees are a
suture zone between two groups of populations.
4 1650
So, the Pyrenees do not constitute a serious obsta-
4 790 cle for the wood mice north-south dispersal move-
ments. The north-south orientation of the main
4 366 valleys and the low altitude of many passes between
the southern and the northern parts of the mountain
4 267 chain leads to a continuity in the forest cover and
perhaps favours the wood mouse settlement. More-
Fig. 3. Restriction patterns after &a I endonuclease diges- over, either near the Atlantic coast or near the Med-
tion of wood mouse mtDNA from 1: Ma1 (pattern SO); 2: iterranean border, this forest cover is also
Me1 (pattern 79); 3: Ibl (pattern 53); 4: S3 (pattern 63); 5: uninterrupted between France and Spain due to suffi-
S5 (pattern 29) and 6: I1 (pattern 1). The size scale in BP is cient rainfalls, to suitable edaphic conditions and to
indicated on the right. the low altitude of the mountains. The situation
prevailing in the Alps is contrasting, at least in the
DISCUSSION AND CONCLUSIONS western part, because the summit line is higher, the
The mean level of sequence divergence of nucleotides altitude of only a few passes being just under or near
between Iberian and French animals is quite low and the tree line. Near the Mediterranean sea, the pied-
therefore Iberian animals can be considered as a part mont is very dry and the soil is stony, leading to
of the ‘north-western’ group whose distribution area ecological conditions unsuitable for the wood mouse.
extends to Scandinavia (TEGELSTR~aMnd JAAROLA In the past, the ice sheet was more developed in the
1989; VAN ROMPAEY1989; MICHAUXet al. 1996). Alps than in the Pyrenees, probably preventing the
These data strongly suggest that the post-glacial re- northward expansion of the mice populations isolated
colonisation of north-west Europe is the consequence in the Italian refugium. This Alpine discontinuity has
of the spreading of a clade which found refuge in the been evidenced in many other species including mam-
Iberian peninsula or in the very southern France mals (Sorex araneus, Arvicola terrestris: TABERLETet
during the latest Ice age. This colonisation route is al., 1994; TABERLEeTt al. 1998), amphibians (Tritu-
identical for the wood mouse, the brown bear (Ursus rus sp.: WALLISand ARNTZEN1989), fish (Salmo
arctos) and some white oaks (Quercus spp.) haplo- trutta: BERNATCHEeZt al. 1992) or insects (Chorthip-
types (TABERLETet al., 1998) but differs for other pus parallelus: COOPERet al. 1995; Apis melliJica:
species like the grasshopper Chorthippus parallelus GARNERYet al. 1992).

As far as the divergence level between the Iberian
samples is very low, it can be assumed that there is no
evidence for two distinct wood mice groups within
the Iberian Peninsula. This conclusion is in complete
accordance with the results of the morphological
study of RAMALHINHOand MADUREIRA(1982).
Moreover, a multivariate (PCA, multiple discrimi-
nant analysis) craniometric study (MICHAUX1996)
shows that the Iberian wood mice are hardly distin-
guished from those trapped in France, Belgium and
Germany.

In conclusion, we suggest that the existence of
Apodemus s. dichrurus in Spain and in southern

Table 2. Woodmouse haplotype diversity in the different regions

No. haplotypes No. animals Diversity index Equitability index

Sicily 5 6 2.252 0.969
Italy 15 33 3.555 0.91
France 21 41 4.047 0.921
Balearic islands 10 23 3.029 0.912
Iberian Peninsula 30 55 4.686 0.955

Id. +France 48 96 5.304 0.95