Figure 1.
Huckleberry habitat and life cycle diagram.
Huckleberry is sparsely distributed in primary successional sites (A), and densely distributed in secondary (B) successional sites. (C) Life cycle diagram for huckleberry. The Greek letters denote the transition rates between the consecutive steps (see Table 1).
Table 1.
Reproductive vital rates for primary (PS) and secondary (SS) successional huckleberry at Mount St. Helens.
Figure 2.
The effects of consumers and mutualists differ between primary and secondary successional huckleberry.
In the primary successional population (PS), herbivory (ρ), lack of pollination (βπ), and post-dispersal seed predation (μ) reduced the number of seeds the most, which corresponded to strongly negative reproductive effects (inset: log response ratio, interannual mean and standard error; PS, black bars; SS, gray bars). The production and fate of huckleberry seeds start with one flowering adult plant and end with 5.7 primary (PS, solid line) and 4.2 secondary (SS, dashed line) successional huckleberry seedlings one year later. The Greek letters denote the transition rates between the consecutive steps (see Table 1), with the biotic interactions marked with asterisks (*). Overwinter seed survival (σS) and seedling establishment (ε) rates were estimated from the literature (see Methods).
Figure 3.
Fates of primary and secondary successional huckleberry.
Pie charts show the fates of potential flowers (A,C) and potential seeds (B, D) in PS (A, B) and SS (C, D). All flowers that do not develop into dispersing berries are lost due to either to grasshoppers, insufficient pollination, fungal infections or pre-dispersal predation. All potential seeds that do not establish as seedlings are lost due to herbivory, insufficient pollination, post-dispersal predation, unviability or over-winter mortality, or remain dormant in the seed bank.
Figure 4.
The integrated reproductive effect of consumers was greater than that of mutualists.
Mean (±s.d.) number of adults on the primary successional Pumice Plain after simulating the 1985–2005 period as a function of seedling survival (σL) using a stochastic model (see Methods section). All simulations started with 25 survivors, used the PS species vital rates, and had 10 coyote scats arriving annually from the nearby secondary successional population. The four scenarios were: ‘control’ = all species interactions as observed, ‘no pollinator limitation’ = no reduced berry (β = 1) or seed (π = 1) production due to insufficient pollination, ‘no consumers’ = no losses due to grasshoppers (ρ = 1), fungal infections (υ = 1), pre-dispersal predation (χ = 1) or post-dispersal predation (μ = 1), and ‘no pollinator limitation, no consumers’ = neither losses due to insufficient pollination nor due to antagonists. The arrow denotes our basic scenario for which plant survival (e.g. σL = 0.05) and growth rates have been set to match the observed trends in adults and juveniles over the 1985–2005 period.
Figure 5.
The number of adults and juveniles on the primary successional Pumice Plain after simulating 20 years of colonization.
Simulations were started with either 25 flowering plants as a source of local seeds (long-distance and local source: A,B) or 0 adults (long-distance source only: C,D). The stochastic population model (see Methods section) reflected either community interactions in the primary (A,C) or secondary (B,D) successional habitat. The number of coyote scats (containing 5,000 seeds each) arriving on the Pumice Plain from the nearby secondary successional population (long-distance seed source) are varied on the x-axis. The arrow denotes our basic scenario for which plant survival and growth rates have been set to match the observed trends in adults and juveniles over the 1985–2005 period.