226 

 and 3.9 individuals/km^ in Late Secondary Forest with Gardens (Table 4-2). No coatis were sighted in 

 Combined/Early Secondary Forest. The plain chachalaca had a population density of 8.7 

 individuals/km^ in Late Secondary Forest wiUiout Gardens, 2.4 individuals/km" in Late Secondary 

 Forest with Gardens, and 20.3 individuals/km^ in Combined/Early Secondary Forest (Table 4-2). Due 

 to small sample sizes, it was not possible to compare densities statistically, however, it was possible to 

 compare average sighting frequencies. Average sighting frequencies were significantly different among 

 three forest successional stages for the coati (P = 0.0877) and the plain chachalaca (P = 0.0143; Table 

 4-1). Coatis were more frequently sighted in Late Secondary Forest with Gardens, whereas plain 

 chachalacas were more frequently sighted in Combined/Early Secondary Forest. 



Given the nature of these data, animal taxa were lumped into four categories of game versus 

 nongame birds and mammals in order to generalize about wildlife densities. For game mammals, 

 nongame mammals, and nongame birds, there were no significant differences in average sighting 

 frequencies among the three forest successional stages (P > 0.05). Only those taxa categorized as 

 game birds, primarily plain chachalacas, exhibited significant differences in average sighting 

 frequencies among the three forest successional stages (P = 0.0261; Table 4-1). This indicates that 

 even for broad groups of animals, there generally were no significant differences detected in average 

 sighting frequencies among forest successional stages. 



This general absence of significant differences in average sighting frequencies among forest 

 successional stages for bird and mammals species, taxonomic groups, and game and nongame birds and 

 mammals may be due to small sample sizes (e.g., too few censuses to conducted statistical analyses) 

 and experimental design problems (e.g., the transects may have been too close together). However, it 

 is highly possible that in fact there were no differences in wildlife densities between forest areas in the 

 vicinity of gardens and forest areas not in the vicinity of gardens, as had been proposed by Linares 

 (1976). This might be expected given the low human population density (ca. 3.0 people/km"), high 

 degree of habitat interspersion, and ecological characteristics of the wild animals that occur at the study 

 area (see Chapter 2). 



