centimeter coverage for species whose densities were estimated as 

 percent coverage) . 



Comparison of summary statistics on biomass and densities per- 

 mitted separation of associations in a subjective manner for the 

 intertidal associations (down to and including the raacrophytic algae 

 zone) . However, this approach was too arbitrary when it came to 

 identifying possible differences between similar associations on 

 different sides of the island or between associations grading into 

 one another on the same side. For these instances, a more rigorous 

 statistical test was necessary. Application of parametric statistical 

 tests requires that the data be normally distributed. This was not 

 the case for most of the data collected during quantitative sampling. 

 Also, it is unlikely that data transformation could be effectively 

 used to normalize the data. The nonparametric Wilcoxon "t" test 

 (Tate and Clelland, 1957) was applied to test differences between 

 densities of selected dominant species within potentially similar 

 associations and between dissimilar associations. An association on 

 the north side, which is dominated by the encrusting coralline alga, 

 Lithothamnium-Lithophyllum complex, was selected as the type associ- 

 ation against which most other associations were tested. 



7. Natural Bottom Survey . 



In addition to the above subtasks, ecological conditions in 

 nearby natural bottom habitats were investigated. This information 

 was to aid in interpreting the ecological changes induced by the 

 presence of the island. 



The composition of the epibenthic macrobiota (plants and animals) 

 on or just above the surface of the sediment or rock on the natural 

 bottom between the island and shore was surveyed along a transect lo- 

 cated away from the influence of the island and causeway (Fig. 3). 

 The transect survey was completed in two segments. The first segment, 

 over a depth of 13.7 meters MLLW near the island to a depth 6,1 

 meters MLLW toward shore, was surveyed by divers using Farallon 

 underwater propulsion units. The second segment, extending from 

 shore to the 6.1 -meter MLLW depth, was surveyed by divers entering 

 through the surf and swimming offshore. Triplicate sediment samples 

 for infauna (animals inhabiting the sediments) were taken at the 

 outer terminus of the transect at a 13.7-meter depth and at a point 

 midway in the transect at a depth of 10.7 meters MLLW (Fig. 3). The 

 samples were collected by pushing 3.13-liter lidless coffee cans into 

 the sediment and carefully sealing both ends of the cylinder with 

 plastic caps. Samples for grain-size analysis were collected by 

 pushing 0.2-liter jars 10 centimeters into the sediment. Infaunal 

 samples were sieved through 1-millimeter sieve screens and preserved 

 for later taxonomic analysis. 



19 



