International tourism is increasing at an unprecedented rate. Understanding the variety of national and local impacts of this increase is of importance to a growing number of governments. Butler's resort cycle model (1974, 1980, 1991) provides for several long-term possibilities as to the relationship between crowding and growth. McElroy, de Albuquerque, and Dioguardi (1993) focus on one of those possibilities. Specifically, using their penetration ratio, they predict that as tourist crowding continues for a group of Caribbean islands, the appeal of these islands decreases in the eyes of potential tourists and that, as time increases, the growth rate of the affected islands, actually decreases. Our article indicates that such a simple, straight-line relationship between increased crowding and a decrease in the rate of change may not be inevitable; indeed, diseconomies of scale may be avoided. The use of a curvilinear regression function reveals how both positive and negative scale economies existed in the Caribbean during the years 1992 through 1996.