A Comparative Demographic Analysis of Clonal Reproduction in a Temperate Soft Coral

In order to evaluate the relative importance of asexual and sexual reproduction to the fitness of a clonal organism, matrix projection models were used to quantify the contributions of each of these reproductive modes to population growth of the soft coral, Alcyonium sp., a species that undergoes frequent colony fission. The demographic and fitness consequences of eliminating either asexual or sexual reproduction from the life cycle of this species were examined by sensitivity analysis and by altering selected entries in the transition matrices to simulate changes in reproductive allocation. Four populations of Alcyonium sp. were monitored photographically for two years to record colony growth, mortality, fission, sexual reproduction, and larval recruitment, Despite high turnover, population densities remained reasonably constant. The 24—52% mortality was matched approximately by recruitment of daughter colonies produced by fission. Sexual reproduction was infrequent, and no larval recruitment was observed. The frequency of both fission and sexual reproduction increased with increasing colony size, while mortality decreased with increasing size. Size—class transition matrices constructed from the demographic data were analyzed by the methods of Caswell (1986, 1989). In both years of the study the size distributions of colonies observed in the field did not differ from the stable size distributions predicted from the projection models. Eliminating sexual reproduction from the life cycle did not alter the predicted stable size distributions; eliminating fission shifted the size distributions of all four populations towards the larger size classes. Reproductive value increased with increasing colony size. Damping ratio, °, a measure of the rate of convergence to the stable size distribution, increased with increases in both fission and sexual reproduction, suggesting that rapidly growing populations are more stable than slowly growing populations. Projected growth rates (=relative fitness, ‐l) of the four populations ranged from 0.66 to 1.15. Sensitivity analyses indicated that sexual reproduction contributes < 1% of the value of ‐l, while > 40% of l is accounted for by transitions between the upper size classes in the population, either by fission or growth. Eliminating sexual reproduction from the life cycle had a negligible effect on fitness (‐l); eliminating fission greatly reduced ‐l and suggested rapid extinction for all populations. This result held even when sexual reproductive output was increased by an order of magnitude. The results of these simulations suggest that selection for increased @l will increase fission at the expense of sexual reproduction in this species.