per unit area, and index of species diversity (personal communication, 

 Hodgson, Moss Landing Marine Laboratories, 1972). The small area sampled 

 by the corers had a significant effect on estimates of the abundance of 

 certain larger and less common species, and on the total number of species 

 in the community. However, the other sampling techniques had disadvantages 

 of equal magnitude. Thus, the use of coffee can corers is adequate for 

 sampling when compared to other conventional sampling techniques. In 

 addition, the hand placement by divers permits good replication and precise 

 control of the area sampled. 



III. THE NUMBER OF REPLICATES 



A number of parameters were used to describe the benthic infauna: The 

 number of species, total number of individuals, number of individuals per 

 species, and species diversity. To determine the number of replicate corers 

 needed, the precision of estimating as many of these parameters as possible 

 should be maximized. 



Twenty-eight cores were taken from an area near the 20-meter control 

 station to examine the relationship between the parameters listed and the 

 number of replicate cores collected. The 28 cores were taken from the 

 three smaller areas in Figure A-1. This number of cores is well above the 

 maximum which would be taken at any one station in a routine sampling 

 program. Although sampling was not strictly random, individual cores were 

 chosen at random from the combined group (28 cores) for the following 

 analyses . 



1. Number of Species . 



Species were recorded in two ways: The total number of species col- 

 lected in all cores (species per sample), and the average number per core. 

 Both are sample values which grossly underestimate the actual number of 

 species in the community, but are very useful as relative parameters. 

 Averaging the number per replicate allows a measure of variance. However, 

 a single core collects fewer species than the total sample which covers 

 enough area to include most of the characteristic species though many of 

 these occur in low abundance. 



The slopes of the curves in Figure A-2 indicate the rate at which new 

 species are accumulated as the area sampled is increased (Jones, 1961). 

 They are averaged from three random orders of the 28 cores. Considering 

 all 28 cores as 100 percent, 16 cores represent 57 percent of the area and 

 contain 92 percent of the species, 8 cores are 30 percent of the area with 

 76 percent of the species, and 4 cores are 14 percent of the area and 57 

 percent of the species. Hodgson (personal communication, 1972) found the 

 corer was compared to two other devices which sample a much larger area; 

 however, almost identical ratios of percent area to percent species were 

 found for all three devices. 



To examine the relationship between the number of replicates and the 

 estimation of number of species per core, replicates were drawn at random 

 from the 28 cores; means and confidence limits were computed fcr 



66 



