- Top of page
- Materials and methods
Over recent decades, many plant species have been affected by both deterioration and fragmentation of their habitats, due to factors such as increasing atmospheric deposition of nitrogen and sulphur compounds, changes in land-use and lack of proper management. As a consequence, populations of many species are small, fragmented and isolated, and are believed to suffer from higher extinction risks (Menges 1991, 1992; Jennertsen et al. 1992; Vitousek 1994). The probability of such stochastic extinctions is determined by environmental, demographic and genetic processes, which are all strongly related to population size, fragmentation and isolation (Schaffer 1981, 1987). Environmental stochasticity initially was considered to be the most important threat to population persistence (Boyce 1992). However, many recent studies have shifted their attention towards genetic processes.
Demographic processes affecting the dynamics and genetic composition of small populations can be influenced by population size and may therefore be detrimental for population persistence. The Allee effect (Lande 1988) further implies that there is a threshold population size below which populations cannot recover. For example, larger populations appear to attract more pollinators, resulting in higher pollination success, whereas small populations may suffer from low visitation rates (Ågren 1996), reducing the level of outcrossing and increasing the rate of self-fertilization (Jennersten 1988; Lamont et al. 1993; van Treuren et al. 1994).
Habitat deterioration can also be a strong determinant of population viability and persistence. This factor is independent of population size and can cause detrimental effects on population persistence by inhibiting growth, reducing seed set and decreasing offspring performance in small populations (Dueck & Elderson 1992; Widén 1993; Oostermeijer et al. 1994; Pegtel 1994), but its effects may be difficult to separate from those of genetic and demographic processes.
For conservation purposes it is important to assess how, and to what extent, habitat quality interacts with genetic and demographic processes and how this affects population persistence and performance in small and isolated populations. Studies assessing plant performance have generally considered the effects of either genetic or demographic processes in declining populations. Genetic variation, population size and habitat quality can, however, all affect performance, and may affect not only rare, endangered species, but also common species suffering from fragmentation and habitat deterioration. Despite the potential value of studies including the effects of these processes on plant populations and individual plant performance, we know of none where all three are considered.
We therefore studied 17 populations of the perennial Succisa pratensis from the Netherlands that differed widely in population size. Genetic variation, population size, soil conditions and performance of these populations were measured and analysed. A path-analytical model was used to separate and quantify the relative direct and indirect contributions of each of the three factors on plant performance.
- Top of page
- Materials and methods
We investigated the relation between plant performance and population size, genetic variation and habitat quality in Dutch populations of the perennial Succisa pratensis. Although it has suffered a serious decline in distribution area, it is still quite a common species in the Netherlands, with reasonable high genetic variation in the larger populations. Nevertheless, we found reduced observed heterozygosity and increased inbreeding coefficients in small populations.
Most positive relationships between population size and plant performance have been ascribed to effects of reduced genetic variation, inbreeding depression or accumulation of detrimental mutations (Ouborg et al. 1991; Lande 1995; Oostermeijer et al. 1995; Ouborg & van Treuren 1995), although a few studies have shown that small population size can lead to genetic deterioration and subsequently to reduced plant performance (Ouborg 1993; Oostermeijer et al. 1994). Furthermore, besides genetic processes, demographic processes and deteriorating habitat quality could also be detrimental for reduced plant performance in small populations (Schaal 1984; Roach & Wulff 1987; Widén 1993). As far as we know, no studies have been carried out examining the influence of all possible processes on plant performance. As a consequence, it remains unclear whether the measured variation in plant performance is affected by reduced genetic variation or driven by other stochastic and/or non-stochastic processes, such as demographic processes or deterioration in habitat quality.
Earlier studies investigated the relative importance of some demographic processes, such as pollen limitation, on plant performance (Ågren 1996; Fischer & Matthies 1997; Oostermeijer et al. 1998; Morgan 1999). These studies clearly indicated that genetic factors might not be the sole determining factors for positive relationships between population size and plant performance. Furthermore, deteriorating habitat quality has also been suggested as a causative factor of reduced plant performance (Roach & Wulff 1987; Dueck & Elderson 1992; Pegtel 1994; Oostermeijer et al. 1998). For the rare perennial Arnica montana, it has already been found that populations suffer as a result of increased habitat acidification and eutrophication (de Graaf et al. 1998). The strong significant correlation in our data clearly indicates that habitat quality may well affect population size. Considerably lower concentrations of NO3wa and NH4ex were measured in the soil of larger populations, suggesting that a low eutrophication level (high habitat quality) may support larger populations. This indicates that environmental factors, in addition to genetic and demographic factors, might also be responsible for the reduced plant performance in smaller populations of Succisa pratensis, as these populations are found under less suitable conditions.
In order to separate and quantify the relative direct and indirect contributions of genetic variation and soil conditions on population size and plant performance, we used a path-analytical model. This model shows strong significant correlations between population size and genetic variation, between soil conditions and population size, between soil conditions and plant performance, and a weak non-significant correlation between population size and plant performance. It appeared that both genetic variation and soil conditions seriously affect plant performance. This emphasizes the importance of genetic erosion and habitat deterioration; genetically eroded populations may perform less well, and reduced habitat quality may also have detrimental effects on population fitness. Furthermore, the model showed that indirect population size effects (via genetic factors) are stronger than the direct demographic effects on plant performance, which were rather weak (and, as a direct effect, not significant). This does not contradict earlier studies that have attributed a strong effect of population size on plant performance (Petanidou et al. 1991; Lamont et al. 1993; Olesen & Jain 1994; van Treuren et al. 1994; Matthies et al. 1995; Jennersten 1988) as the total effect of population size on plant performance includes genetic variation. This result again highlights the importance of genetic variation for conservation biology. The direct effects of soil conditions on plant performance were not as strong as the direct genetic effects. However, the total effects of soil conditions were stronger than the total effects of genetic variation. These results support the assumption that habitat quality and environmental stochasticity are of more immediate importance than genetic and especially demographic processes in determining population persistence (Boyce 1992).
Overall, path-analysis reveals Succisa pratensis as a species that is vulnerable to habitat deterioration, which leads to reductions in population size and plant performance. However, genetic effects such as genetic drift, genetic erosion and inbreeding are also very important for population persistence and plant performance, even in the short term. Survival of small and declining populations suffering from unfavourable habitat conditions may not be promoted by habitat restoration, as the negative effects of genetic erosion still continue. The fact that effects of reduced population size, genetic erosion and habitat deterioration on plant performance can be measured, even in relatively common species, such as Succisa pratensis in the Netherlands, is of considerable concern for conservation management.