Ten year study of the annual variation in berry and seed production in a population of Vaccinium corymbosum L.

INTRODUCTION

In their review of the demography of plants, Harper and White (1974) dismiss shrubs rather cavalierly as having a "reasonably defined and circumscribed variation" between life span and the age of the plant when it sets seed for the first time. After germination, the surviving seedlings of shrubs begin to fruit at about five years and persist fruiting for about 50 years, during which time they have a fecundity schedule and periodicity not much different than trees. A period of increasing fruit production is followed by a plateau of yield and a subsequent decline in reproduction all within an ambit of alternating "mast years." In short, Harper and White (1974) assume that the fecundity schedule of shrubs is just a shortened version of the long records of periodicity in seed production observed in Pinus, Abies and Fagus. The dynamics of shrub populations have been largely passed over because the evidence is either anecdotal or derived from cultivars in plantations.

Recruitment in shrubs is subject to dispute: Grime (1979) argued that woodland shrubs have either an escape mechanism, i.e., diaspores are widely dispersed in order to exploit spatially unpredictable patches where seedlings could become established, or a "bank of persistent seedlings." In contrast, Eriksson (1989) has some evidence that shrubs are dependent on seasonal regeneration or on a persistent seed bank.

An additional caveat, especially relevant to long term demographic studies, is the probable demise of the tagged population being studied. Indeed, in his review of population dynamics of plants, Crawley (1990) discovered that long-term studies of plant populations are typically set up around existing mature plants and that empty quadrats are ignored. Such a bias has two undesirable consequences: (1) tagged plants drift inexorably towards extinction, and (2) recruitment may occur in empty rather than occupied quadrats.

Our objective in this 10-year study of the reproductive capacity of Vaccinium corymbosum L., a long lived xenogamic shrub (Vander Kloet, 1988), was two-fold: (1) to measure the annual variation in seed production in a cohort of 50 tagged plants in order to test some of the assumptions made by Harper and White (1974) about periodicity within the ambit of age related factors in seed set in shrubs and (2) to assess recruitment strategies, i.e., does this population of V. corybosum have a persistent seed bank or a bank of persisting seedlings or does it exploit spatially unpredictable patches?

MATERIALS AND METHODS

During the winter of 1979, all V. corymbosum plants along Lake George and Leap Frog Lake, Yarmouth County, Nova Scotia, were located and marked. In all, 385 plants were found and these occurred either in bogs or shrubberies along the lake shores or adjacent Picea-Abies-Acer rubrum woods. Since lumbering, wind-throw and secondary succession were observed in these woods, they were classified as disturbed plant communities (U) and the bogs and bog margins as more or less stable (S). Next, 50 of the 385 plants were selected at random and given permanent aluminium tags. From 1980 until 1990 these 50 shrubs have been visited at least five times during the fruiting season (August-early September) and the berries on each shrub counted. In addition at each visit, six fully-ripe berries were collected at random from each shrub. The seeds were washed from each berry and air dried. Any collapsed or poorly formed seeds, which both Darrow (1941) and Bell (1957) found inviable, were separated …