Menzies, George, and Rowe C1973) define species diversity as a concept 

 in community ecology which refers to the heterogeneity (or lack of it) in 

 a community or assemblage of organisms. Thus, diversity is dependent 

 upon the nxomber of species present Cspecies richness, S) and the distribu- 

 tion of individuals among species (equitability or evenness) . Another 

 definition of diversity is simply the number of species found in a unit 

 area (Whittaker, 1972). Indices to measure diversity, species richness 

 and equitability are so niimerous that confusion exists (e.g., Hairston, 

 196A; Sanders, 1968; Hurlbert, 1971; Whittaker, 1972; Fager, 1972; Peet, 

 1974; Pielou, 1975; Smith, et al., 1979). The proliferation of indices 

 prompted Hurlbert (1971) and Peet (1974) to recommend discarding diversity 

 as a measure in ecological studies. However, placed in the proper perspec- 

 tive, diversity indices have been shown to be useful in "bioenvironmental" 

 studies (Boesch, 1972; Borowitzka, 1972; Swartz, 1972; Pearson, 1975; 

 Swartz, 1978). In this study the data analysis is restricted to the two 

 commonly used diversity indices: the Shannon-Weaver index (Shannon and 

 Weaver, 1963) and the Gini's index (Gini, 1912; Simpson, 1949). 



(1) The Shannon-Weaver Index of Diversity . This index is based on 

 information techniques, where diversity is equated to the amount of un- 

 certainty which exists regarding the species identity of an individual 

 selected at random from a community. The more species and the more evenly 

 their representation, the greater the uncertainty and hence, the greater 

 the diversity. The computational formula for Shannon's index is 



H' = C/N (N log N - E% log. „ n.) 

 1 ^i-j^ i 1 



where C = 2.3026 (for "nats", units expressed as natural logarithms), N is 

 the total number of individuals, and n^ the number of individuals in the ith 

 species. Lloyd, et al. (1968) presented the functions of "nlogj gn" for all 

 integers from n = 1 to n = 1050 to simplify the use of Shannon's index. 



(2) Gini's Index of Diversity . This index is a measure of the domi- 

 nance in a sample. Though it is usually insensitive to rare species, it 

 has been used commonly as a diversity index. The computational formula 

 for dominance diversity (Simpson, 1949) is 



DM = H^ .^ n (n-l)/N(N-l) 

 and complemental or actual diversity, d = 1 - DM (Gini, 1912). 



(a) Equitability . Equitability is considered a component of diversity 

 in that it provides an idea about the evenness of species distribution at 

 a site. Usually, a positive correlation exists between diversity and 

 equitability (De Jong, 1975), i.e., a high equitability would indicate a 

 high diversity. Traditionally, high equitability and diversity have been 

 considered to indicate a "healthy condition" of the fauna. Reduction of 

 equitability usually occurs with an increase in oligomixity (i.e., dominance 

 by few species). Pielou 's (1966) method of measuring equitability was used 

 in this study. The computational formula is 



J' = H'/log^s 



18 



