relative distribution of species in each type of tropical forest, the rate of change in the area of each type of tropical forest, and the relationship between change in forest area and change in species numbers.
Most published projections of species extinctions resulting from deforestation in the tropics do not include the basis for their estimates in ways that can be examined independently (Table 6–1). Exceptions are the estimates of Love joy (1980) in the Global 2000 Report, that of Ehrlich and Ehrlich (1981) in their classic book on extinctions, and the recent paper by Simberloff (1986).
Estimating the total species richness of the tropical biome is probably beyond the means of scientific endeavor at this time. Total species inventory of a single tropical ecosystem does not even exist. Insufficient information handicaps any effort to estimate the number of species extinctions. Myers (1979) discussed the problems of estimating species numbers and concluded that of the 3 to 10 million species that exist globally, approximately 70% occur in the tropics. The World Resources Institute and the International Institute for Environment and Development (WRI and IIED, 1986) reported between 3.7 and 8.7 million species in the tropics (the actual number depending on whether the world has 5 or 10 million species), of which 0.6 million are known to science. Taxonomists estimate that only 1.5 to 1.7 million species are presently known to science (Raven, 1977; WRI and IIED, 1986). Clearly, scientific understanding of total numbers of species is still fragmentary. For this reason, it is best to use relative distributions of species in different forest types when making global estimates of species extinctions.
The rate of change in tropical forests of all kinds has been discussed in depth only by Lanly (1982), who made an effort to document the rate of increase in the area of secondary forests (by reforestation, afforestation, and natural regeneration; see Figure 6–1) as well as the rate of forest loss. Other attempts usually emphasize conversion or modification of mature forests with little or no analysis of recovery (Myers, 1980). Lanly’s data show that of the 11.3 million hectares of mature forest land deforested annually, 5.1 million hectares are converted to secondary forest fallow. He estimated that the total area of this forest type is 409 million hectares and that almost 1 million hectares of secondary forest is created annually on unforested land through natural regeneration or human intervention. Such large forest areas cannot be dismissed as irrelevant to the conservation of species diversity because they support an extensive biota (discussed below) and because under certain conditions, they are capable of supporting more complex biota than the mature system they replace (Ewel, 1983).
Lanly’s data also show that deforestation rates are higher in closed than in open forests (Figure 6–1). Within closed forests, a large fraction of the conversion involves logged forests—forests that have previously been modified by human activ-