Big-seed, durable construction syndrome

"Across species, a negative correlation has commonly been found between the seed mass and the potential relative growth rate (RGR) of seedligns grown under favourable conditions (e.g., Grime and Hunt, 1975; Shipley and Peters, 1990, Jurado and Westoby, 1992; Maranon and Grubb, 1993; Rincon and Huante, 1993; Osunkoya et al., 1994). However, there is no known mechanism through which larger seed mass might directly cause lower potential RGR. Rather, low potential RGR and large seed mass appear to be part of a trait syndrome also involving sturdy tissue construction (low-SLA and specific root length) (Reich et al., 1998; Wright and Westoby, 1999...) and low rates of tissue turnover (Bonger and Popma, 1990; Seiwa and Kikuzawa, 1991). Conversely, smaller-seeded species generally have higher potential RGR under near-optimal conditions, which is due in part to the seedlings being constructed of thinner or lower-density tissue with higher turnover rates."

Westoby et al. (2000) [see full citation in previous post]

Uniform seed size predicted

The seed size/number trade-off concept formalized in the Smith-Fretwell model predicts that: "if a mother plan is in a position to allocate more resources to seed output, it should produce more seeds of the same size...[assuming] some minimum size for a seed to have any chance of establishing and...diminishing returns at some stage as seed mass increases further. The curvature of the Smith-Fretwell function ensures that, if resources are reallocated such that one seed has higher seed mass than the optimum while another has less, the gain in fitness in the augmented seed is smaller than the loss in fitness in the diminished seed."

The mother plant's "...physiological machinery of seed provisioning " should have been selected to produce the maximum number of uniform, optimally sized seed, rather than simply producing larger seeds.

However, seeds vary within a species and even within an individual plant. Explanation?

"The moderate observed variation in seed mass within a species can be attributed either to the machinery of seed provisioning having limited capacity to deliver a completely standardized seem mass or to variability in the Smith-Fretwell function that seedlings are exposed to."

Source: Leishman, M.R., Wright, I.J., Moles, A.T., Westoby, M. (2000) The evolutionary ecology of seed size. In Seeds: the ecology of regeneration in plant communities.Fenner, M., ed. CABI Publishing, Wallingford: p. 31-57.

Seed size/number tradeoff summarized

"Seed size represents a fundamental trade-off, within the strategy of a species, between producing more small seeds versus fewer larger seeds from a given quantity of resource allocated to reproduction. The trade-off and its consequences were formalized in the model by Smith and Fretwell (1974). There is always selection pressure to produce more seeds, since more seeds represent more offspring (although there may be a lower limit to the seed size that permits a functional seedling to be produced (Raven, 1999)). On the other hand, larger, better-provisioned offspring have a greater chance of successful establishment, described by the Smith-Fretwell function....The best outcome from the mother's point of view is to maximize the ratio of seedling establishment chance to provisions invested in each seed..."

Weak positive correlation in Aquilegia pyrenaica

"Correlation analysis showed that only a few pairs of variables were significantly correlated (Table 3). No trade-off seems to exist between the number of seeds per carpel and the mean seed weight in a fruit, as the two variables are, if anything, positively correlated (r = 0.22, P = 0.06)."

M.C. Castellanos, M. Medrano, and C.M. Herrera. (2008) Subindividual variation and genetic versus environmental effects on seed traits in a European Aquilegia.

No seed size-seed number tradeoff in Lupinus perennis

"Within plants, this study detected no relationship between maternal parents’ seed number and seed size (weighted linear regression, b 5 3.8 3 1026, t 5 1.04, P 5 0.304) or between seed number and the coefficient of variation in seed size (weighted linear regression, b 5 23.8 3 1025, t 5 1.06, P 5 0.295; Fig. 2). Therefore, this study did not detect a trade-off between seed size and seed number, or a change in seed size variation with increasing seed number."

STACEY L. HALPERN (2005) SOURCES AND CONSEQUENCES OF SEED SIZE VARIATIONIN LUPINUS PERENNIS (FABACEAE): ADAPTIVE ANDNON-ADAPTIVE HYPOTHESES. American Journal of Botany 92(2): 205–213.