Dissecting components of population-level variation in seed production and the evolution of masting behavior


  • Walter D. Koenig,

  • Dave Kelly,

  • Victoria L. Sork,

  • Richard P. Duncan,

  • Joseph S. Elkinton,

  • Mikko S. Peltonen,

  • Robert D. Westfall

W. D. Koenig, Hastings Reservation, Univ. California Berkeley, 38601 E. Carmel Valley Rd., Carmel Valley, CA 93924, USA (wicker@uclink4.berkeley.edu).
D. Kelly, Dept Plant and Microbial Sciences, Univ. of Canterbury, Private Bag 4800, Christchurch 8001, New Zealand.
V. L. Sork, Dept of Organismic Biology, Ecology, and Evolution, and Institute of the Environment, UCLA, Los Angeles, CA 90095-1786, USA.
R. P. Duncan, Ecology and Entomology Group, PO Box 84, Lincoln Univ., Canterbury, New Zealand.
J. S. Elkinton, Dept of Entomology. Univ. of Massachusetts, Amherst, MA 01003, USA.
M. S. Peltonen, USDA Forest Service, 180 Canfield St., Morgantown, WV 26505, USA.
R. D. Westfall, Institute of Forest Genetics, PSW Res. Station, P. O. Box 245, Berkeley, CA 94701, USA.


Mast-fruiting or masting behavior is the cumulative result of the reproductive patterns of individuals within a population and thus involves components of individual variability, between-individual synchrony, and endogenous cycles of temporal autocorrelation. Extending prior work by Herrera, we explore the interrelationships of these components using data on individual seed production in 59 populations of plants from 24 species spanning a large range of annual variability, from species exhibiting strong masting to others with little annual variability in seed production. Estimates of population and individual variability were not biased by sample size or average overall seed production when based on untransformed seed production values, but these values declined as log-transformed seed production increased. Population variability was more strongly correlated with individual variability (r=0.86) than individual synchrony (r=0.73). These latter two components were also significantly correlated (r=0.45), but randomizations confirm that they need not covary closely. Thus, selection can act separately on inter-annual variability and between-individual synchrony. We illustrate the potential for such fine-tuned selection on seed production patterns by discussing several examples and by demonstrating significant differences in components of population-level variation in seed production among species related to their life-history.