Coral Reefs(1993) 12:127-137 Coral Reefs 9Springer-Verlag 1993 / ~ ~ Hurricanes and coral reefs: the intermediate disturbance hypothesis revisited C. S. Rogers Virgin Islands National Park, P.O. Box 710, St. John, VI 00830, USA Accepted 17 June 1993 Abstract. A review of research on the effects of hurricanes on coral reefs suggests that the intermediate disturbance hypothesis may be applicable to shallow reef zones domi- nated by branching or foliaceous coral species that are especially susceptible to mechanical damage from storms. Diversity (H') increases because of an increase in evenness following destruction or removal of the species that was monopolizing the space. The intermediate disturbance hypothesis as presented by Connell focuses on changes in number of species, but should be expanded to include diversity (H') and evenness. It should also be modified to incorporate changes in living cover and the time elapsed since disturbances of varying intensities. This hypothesis predicts that when cover is high, diversity will be low. However, research on coral reefs does not consistently demonstrate an inverse correlation of coral diversity and coral cover. An increase in cover and decrease in diversity with depth would also be expected because deeper reef zones generally experience less disturbance. However, higher diversity (both H' and species richness) is often associated with deeper zones. The effects of hurricanes on coral reefs will depend on the temporal and spatial scales under consideration, the life history characteristics and morphology of the dominant species, the depth of the reef zone, the ecological history of the site, and the influence of any additional natural or human stresses. Introduction Ecologists are intrigued by the high diversity of tropical rain forests and coral reefs-a diversity which in the broadest sense is a reflection not only of the number of species of trees, corals and other organisms, but of their relative abundance, their different morphologies, and their arrangement in time and space. When we consider these ecosystems, we think of their magnificent physical struc- tures, one system dependent on the growth of solitary trees that sometimes reach over 50 m in height and the other on the growth of colonial, invertebrate animals that deposit limestone skeletons up to several meters thick. The intermediate disturbance hypothesis Two of the central questions in ecology are, "What is the effect of disturbance on diversity?" and "What factors promote or maintain high diversity?" The intermediate disturbance hypothesis (Paine and Vadas t969; Horn 1975; Connell 1978) is one of many advanced to explain or explore patterns of diversity in tropical forests and coral reefs (e.g., Grigg and Maragos 1974; Connell 1978, 1979; Condit et al. 1992). As presented by Connelt (1978, 1979), this hypothesis suggests that the highest number of species of trees or corals will be reached at intermediate levels (frequency and size) of natural disturbance. Lower diversity results if disturbances are either (1) too frequent or too infrequent, or (2) too large or too small. Connell (1978, 1979) explicitly states that he is examining variations in diversity within local areas and with respect to the number of species ("species richness") only. The simple conceptual model he presents incorporates the following key elements to help explain changes in diversity over time for a tropical rain forest or a coral reef: number of species, frequency of disturbance, and size of disturbance (Fig. 1). Here I will examine this hypothesis with reference primarily to the effects of hurricanes (= cyclones) on Caribbean and Pacific coral reefs, with limited comparison to tropical forests. Disturbance and diversity: some definitions Disturbance can be defined as "any relatively discrete event in time that disrupts ecosystem, community, or population structure and changes resources, substrate availability, or the physical environment" (Pickett and White 1985). "Diversity" refers to the number of species (species richness) or to a diversity index based on the number of species weighted by their relative abundance