The amount of genetic variation in the rare perennial herb Gentiana pneumonanthe L. was determined to explore its relation to population size. Differences in isozyme variation between maternal plants and their offspring were used to investigate the relationship between population size and outcrossing rate. In 25 populations in The Netherlands, differing in size from 1 to more than 50,000 flowering individuals, 16 allozyme loci were analyzed on leaves of maternal plants and offspring grown in a greenhouse. Population size was significantly positively correlated with the proportion of polymorphic loci, but only marginally with heterozygosity and the mean effective number of alleles. Most of the studied populations were characterized by a complete absence of rare alleles, and F‐statistics suggest relatively high levels of genetic differentiation among populations and thus a low level of gene flow. Leaf samples (maternal) were mostly in Hardy‐Weinberg equilibrium, while several offspring samples showed an excess of homozygotes, which suggests selection favoring heterozygotes. Because most small populations consist only of adult survivors from formerly larger populations, this may partly explain the absence of a clear relationship between genetic variation of the maternal plants and population size. A significant positive correlation was found between the level of cross‐fertilization and population size. From these results, we conclude that, to some degree, small populations have a reduced level of genetic variation, while their present isolation in nature reserves has resulted in a very limited interpopulational gene flow level. At present a higher level of inbreeding in small populations contributes to a further loss of genetic variation and may also result in reduced offspring fitness.

No abstract

Especially for rare species occurring in small populations, which are prone to loss of genetic variation and inbreeding, detailed knowledge of the relationship between heterozygosity and fitness is generally lacking. After reporting on allozyme variation and fitness in relation to population size in the rare plant Gentiana pneumonanthe, we present a more detailed analysis of the association between heterozygosity and individual fitness. The aim of this study was to test whether increased fitness of more heterozygous individuals is explained best by the 'inbreeding' hypothesis or by the 'overdominance' hypothesis. Individual fitness was measured during 8 months of growth in the greenhouse as the performance for six life-history parameters. PCA reduced these parameters to four main Fitness Components. Individual heterozygosity was scored for seven polymorphic allozyme loci. For some of these loci (e.g. Aat3, Pgml and 6Pgdh2) heterozygotes showed a significantly higher relative fitness than homozygotes. To test the inbreeding model, regression analyses were performed between each Fitness Component and the number of heterozygous loci per individual. Multiple regressions with the adaptive distance of five loci as independent variables were used to test the overdominance model. Only the inbreeding model was a statistically significant explanation for the relationship between heterozygosity and fitness in G. pneumonanthe. The number of heterozygous loci was significantly negatively correlated with the coefficients of variation of three of the six initially measured fitness parameters. This suggests a lower developmental stability among more homozygous plants and may explain the higher phenotypic variation in small populations of the species observed earlier. The importance of the results for conservation biology is discussed.