The sediment size data were analyzed using Friedman's test (after Zar, 

 1974), a nonparametric test which requires only ordinal scaling of data. This 

 test was used to evaluate (1) differences in relative particle size 

 distribution among all six transects (data for all four stations on each 

 transect were combined for analysis) within each sampling period; and (2) 

 differences in relative particle-size distribution at station 1 among all six 

 transects within each sampling period. The percent ccwiposition values were 

 ranked within each particle-size category, and the ranked values were summed 

 for each transect to calculate: 



X^ = V. /^^ .^ E R? - 3b(a + 1) 



r ba(a + 1 ) 4^ 1 

 1=1 



where a is the number of treatments (columns), b the number of blocks, and R^ 

 the sum of the ranks squared in each column. Critical table values for combina- 

 tions of a and b were found in Zar (1974). 



5. Water Quality . 



At the sxirface of station 1 and at the surface and bottom of stations 2, 

 3, and 4 on each transect during each sampling period, water temperature and 

 dissolved oxygen concentration were measured with a YS1 Model 51B meter and 

 water samples to be used for determination of turbidity and suspended solids 

 were collected with a Van Dorn bottle. The samples were iced and stored in an 

 insulated container for analysis in the laboratory. Turbidity was measured 

 with an H F Instruments Ltd. Turbidimeter, Model 1000. The weights of 

 suspended solids were determined by filtering a known volume of each sample 

 under vacuum on a tared Whatman glass-fiber filter paper, drying the filter 

 paper at 40° Celsius for 24 hours and weighing the tared paper. 



6. Macrozoobenthos . 



Macrozoobenthos samples were collected with a Ponar grab. Three grab sam- 

 ples were collected at each station during each of the six sampling periods. 

 Previous macrozoobenthos studies (Schuytema and Powers, 1966) in the nearshore 

 waters of Lake Huron have indicated that three replicate grabs make up an ade- 

 quate sample. Each grab sample was washed through a standard No. 30 sieve (0.65- 

 millimeter mesh size) , and the benthic invertebrates (macrozoobenthos) retained 

 by the screen were placed in a labeled container, preserved in 10 percent form- 

 alin, and taken to the laboratory for processing. Organisms were identified to 

 the lowest practical taxonomic level (e.g., family, genus, or species) and the 

 criteria for assigning individuals to each such taxon were unchanged throughout 

 the study. Although grab sample volume varied, the number of organisms per 

 replicate grab remained relatively constant indicating that most of the 

 organisms were probably confined to the upper few centimeters of the substrate. 



Macrozoobenthos commvinities at each station were compared before and after 

 beach nourishment using Morisita's index of community similarity as modified by 

 Horn (1966). This index provides a measure of the probability that individuals 

 randomly drawn from each of the two communities will belong to the same 

 species, relative to the probability of randomly selecting two individuals of 

 the same species from one of the communities. Morisita's index values (CA) 

 were calculated as follows: 



