Woody plant species are able to produce potential embryonic meristems on different parts of the plant body. In temperate broad-leaved forests, the most effective system of clonal growth is displayed by root suckering (e.g. members of the species Populus, Salix and Prunus). Similarly to herbaceous plants, these meristems often behave as opportunistic organs securing colonization of unoccupied sites or increasing competitive power of the species within the community [13]. In this study we are particularly interested in the effects of two different forest management strategies on sexual vs. asexual reproduction and genetic diversity of the forest tree species Prunus avium L. (wild cherry). We are, however, aware of the problems with classical statistical methods of testing for clonal propagation that result from questionable assumptions on how sexual reproduction determines the association of genes in genotypes. Particularly for the known characteristics of the reproductive system of wild cherry, these assumptions seem to be unrealistic and may therefore lead to biased assessments of the role of clonal propagation in this species.
Two stands were chosen for our analysis, one about 70 years old (Roringen) that is managed as a HFS (a mixture of shelterwood and selective thinning), and a second stand (Wibbecke) that is about 100 years old and was formerly managed as CWS. The results of differrent man-made ecological conditions and, consequently, different establishment histories of the two stands are reflected in typical species compositions. Although the CWS is largely out of practice since the beginning of the last century, species such as Quercus petraea (sessile oak), Carpinus betulus (hornbeam) and Tilia cordata (basswood) are still dominating the forest of Wibbecke. According to [14] these species are better adapted to late (spring) frost and drought, since the density of the canopy was irregular under coppice regimes, covering only a small proportion of the area. Furthermore, these species are capable of sprouting (stump shooting) after (man-made) injury and, in the case of wild cherry, are even able to produce new individuals (ramets) by root suckers. They also contribute to forest tree species diversity.
In contrast, in the HFS, shade-tolerant species such as Fagus sylvatica (common beech) are the dominating species and by this reduce forest tree species diversity. Light-demanding species such as P. avium mainly colonize early successional stages and are thus dependent on gaps in the forest canopy. Such gaps (up to 60 m in diameter) are, for example, provided in shelterwood forests, where a group selection method of cutting is applied.
The conditions for establishment of wild cherry are thus distinctly different between the two management systems and can be expected to affect the mode of reproduction. In fact, our observations on clonal reproduction lend support to this expectation in special ways. The initially posed question as to the formation of more than one local generation can be answered differently for the two management systems. The individual cloning success as well as the degree of cloning at the stand level are distinctly higher in the CWS (N/G is doubled and C is even seven times larger). This indicates that in HFS, as compared to CWS, the chances are strongly reduced to reach the canopy in a second local vegetative generation. There are clear indications that this is different in the CWS, particularly in view of the considerable spatial spread and intermingling of several genets (see Figure 2). Distances of 80 m and more between ramets of the same clone can hardly be explained by first generation asexual establishment. Thus, CWS may provide the conditions for the formation of three local generations starting with a sexual generation followed by at least two asexual generations.
These aspects were not taken into account by [4] and [15] because almost young trees (10–30 years) were analysed. Further, these findings rely heavily on the applied methods of clonal analysis. As [4] emphasized, markers more variable than isozymes are required to describe wild cherry clone sizes. Our study, however, demonstrates that high variability alone may not be a sufficient condition for reliable estimates of clonal propagation. What is needed in addition is a well reasoned assumption on the amount of gene association. Based on our newly developed "Type one error method", it turned out in our study that even small deviations from the assumption of the absence of gene associations may imply sizable errors in the identification of clones. This goes along with differences for gene association among genotypes. The apparently small values of gene association inferred from our observations constitute lower limits of association. It can therefore not be ruled out that the actual degrees of gene association are distinctly larger.
Clonal propagation is also likely to affect sexual reproduction if it occurs at sizable proportions (in terms of C-values). In this case mating and thus seed set may be strongly reduced as a consequence of the gametophytic incompatibility system. It is interesting to note in this concern that the stand of Wibbecke with its distinctly larger C-value was observed to show only sporadic seed set since the last eight years.
The effect of clonal propagation on genetic variation is analysed with the help of measures of genetic diversity in both the gene and the hypothetical gametic pool. At a first site it may be surprising that the stand with the higher C-value (Wibbecke) also showed higher gene as well as hypothetical gametic diversity. Replacement of genetically variable individuals by copies of a single genotype can reasonably be expected to reduce overall haplotype variation. This should describe the situation in Wibbecke, since spontaneous repeated seedling recruitment is rarely observed within a stand of wild cherry. Because this species is extremely affected by browsing, quick recovering ability of root suckers from sublethal damage by herbivores may lead to mainly vegetative reproduction within local stands of P. avium [2]. Particularly within clumps of wild cherry occurrences, a genotype that consists of dozens of ramets is more likely to have some ramets escape herbivory than a genotype that consists of a single stem [16].
Accordingly, in a review of [1] surprise is expressed about the fact that populations of clonal plants can exhibit considerable levels of genetic variation, comparable with those found for populations of non-clonal plants [17-20]. Especially in stable habitats, any difference in cloning success of particular genets as well as random drift should lead to a decrease in genet number over time and dominance of a few clones, such that genetic variation decreases [1]. This situation can be assumed to be realized in CWS and could thus apply to our observations in P. avium.
The difference in size between the two stands and random events during establishment of the stands can be drawn upon for an explanation of the observations on genetic diversity. Indeed, there are fewer individuals in Roringen, and this corresponds to its smaller haplotype (gene and gamete pool) diversity. Yet, this contrasts with the fact that the number of genotypes in Roringen is twice the number of genotypes in Wibbecke. A closer look at the putative clones in Wibbecke reveals that the clone with the by far largest number of ramets has the second largest degree of heterozygosity. This explains the larger haplotype diversity in Wibbecke. In a more general context, the present observation suggests heterozygote advantage in clonal propagation as an interesting hypothesis to test in future studies, when asexual reproduction is associated with unusually large degrees of haplotype diversity (see also [21]).
Answers to all your Biology Questions
