Nature © Macmillan Publishers Ltd 1998 8 scientific correspondence 734 NATURE | VOL 394 | 20 AUGUST 1998 Of the ten species of European truffle (fungi of the genus Tuber, phylum Ascomycota), some have economic value because of their organoleptic properties (taste and per- fume), in particular the black truffle (Tuber melanosporum Vitt.) and the summer and burgundy truffles 1,2 . The black truffle is mainly found in Spain, France and Italy (Fig. 1a), and it shows variation in several traits, including in its famous organoleptic properties, across this geographical range. Here we show that this variation probably results from environmental, rather than genetic, influences. In an attempt to explain the variation in T. melanosporum across its geographical range and to study the distribution of genetic variability within and among populations, we analysed fruiting bodies (ascocarps) from different populations in France and Italy for random amplified polymorphic DNA (RAPD) and microsatellite polymorphism. We found an extremely low level of poly- morphism over the whole study area for both types of marker (Fig. 1b). This pattern contrasts dramatically with that of the sym- patric summer truffle (T. aestivum Vitt., which forms a species complex with the bur- gundy truffle, T. uncinatum Chatin), which has distinct internal transcribed spacer alleles 3 and highly variable RAPD patterns 4 . The absence of heterozygotes in T. melanosporum, shown by study of microsatellites, is consistent with a very closed mating system, such as homothal- lism or even exclusive selfing. As the two putative haploid genomes from an ascocarp are very similar, we considered each variable RAPD band to be a distinct locus exhibiting two alleles (presence or absence of a band). The genetic models used subsequently showed no increase in genetic differentia- tion with geographic distance (Fig. 1c). The low level of genetic diversity of the black truffle probably cannot be explained by its present large population size. In France, 1–3ǂ10 4 kilograms of black truffles are now officially sold each year, and more than 10 6 kilograms per year were traded in the last century (the average weight of an ascocarp is about 30–40 grams). A population bottleneck probably occurred during the last, and coldest, glacia- tion, when the broadleaved forest of Europe was considerably reduced and restricted mainly to the Mediterranean coastal zone 5 . The black truffle ripens in winter (Novem- ber–February), which probably contributed to its drastic reduction in population size and restriction to its southernmost limits during the glaciation, as ascocarps of con- temporary T. melanosporum are susceptible to frost. The present low level of genetic variability in black truffle populations is consistent with such a bottleneck occurring 10,000 years ago, followed by a rapid colo- nization of southwestern Europe, which would also explain the absence of phylo- geographic signals in the few polymorphic markers found. The ‘glaciation hypothesis’ would be strengthened if more southerly populations (in Spain or Italy) were found to show greater genetic diversity. The seasonal behaviour of the summer and burgundy truffles, which ripen in spring and autumn, respectively, would have allowed them to sustain a larger geo- graphical range and population size during the last glaciation, explaining the present high level of genetic variability of this species complex. Moreover, their current geographical range, extending further to the east and north (for example, to Poland and Sweden 6 ), shows that they are more tolerant of colder climates. Our results show that the morphologi- cal and organoleptic differences seen over the geographical range of the black truffles can probably be explained by environmen- tal variation rather than by genetic factors. Research is needed to identify the environ- mental variables that affect the black truf- fle’s perfume and taste, which are the objects of intense human interest. G. Bertault, M. Raymond, A. Berthomieu Institut des Sciences de l’évolution (UMR 5554), Laboratoire Génétique et Environnement, Cc 065, Université Montpellier II, 34095 Montpellier, Cedex 05, France. e-mail: raymond@isem.univ-montp2.fr G. Callot UFR Sciences du Sol, Institut National de la Recherche Agronomique, Place Viala, 34060 Montpellier, Cedex 01, France D. Fernandez Laboratoire de Phytopathologie Tropicale, ORSTOM, BP 5045, 34032 Montpellier, Cedex 01, France 1. Henrion, B., Chevalier, G. & Martin, F. Mycol. Res. 98, 37–43 (1994). 2. Amicucci, A. et al. Biotechnol. Lett. 18, 821–826 (1996). 3. Guillemaud, T. et al. Mycol. Res. 100, 547–550 (1996). 4. Gandeboeuf, D. et al. Can. J. Bot. 75, 36–45 (1997). 5. Blondel, J. Biogéographie, Approche écologique et évolutive 1–297 (Masson, Paris, 1995). 6. Chevalier, G. & Frochot, H. La Truffe de Bourgogne (Tuber uncinatum Chatin) 1–257 (Pétraque, Levallois-Perret, 1997). 7. Rousset, F. Genetics 145, 1219–1228 (1997). 8. Raymond, M. & Rousset, F. J. Hered. 86, 248–249 (1995). rn r n n r r rp n n r r r r n p r .a, w rn r p n rp rn ( r ) r n p r n p n ( r ); n r r r p wn n b . r (n ǃ 0) wr n n r x p n p w w n n r ‘r ’.b, x p p rn n w 4 pr r ( p rn n ) r r r r n (n r r r a); rrw n p y rp n . z n r n p r. c, n y n y n rn r .7 n x ( n r y pr r 4 n ) n ny w p y rp r nn ( ) 4 n ( ) 9 wn w . p n w r p r r w n p p r n . (r . ). n r n n rn n w rp n w n n n ( n 0 p r n 0.0). 1 2 3 4 1 23 4 M 21 5 2 0.56 0.9 -0.5 a b c 0 3 2.5 2 1.5 1 0.5 Fst/(1-Fst) Log geographic distance (km) kbp * * * * * * * * * * * * 0 1 2 3 4 5 6 7