760 BECKETT 



The cause of these affinities in June (and not in August or 

 September) is most probably changing river discharge. Although the 

 colonization period for the June samplers had a mean dis- 

 charge (1670 m^/min) similar to that in August (1713 m-^/min), 

 the minimal flow during these periods is probably very important 

 since pollutional impact would be strongest during periods of least 

 dilution. The minimal flow for the August collection period (569 

 m^/min) was 27% lower than that of the June collection period (770 

 m^/min). In addition, flow over the last week before collection of 

 samples averaged 758 m^ /min for August compared to 1543 m^ /min 

 for June. Also, depressed dissolved-oxygen concentrations observed 

 in August at downstream stations were not apparent in June 

 (Table 2). The comparatively high flow for the June collection 

 period probably ameliorated pollutional influences, resulting in the 

 existence of markedly similar upstream and downstream communi- 

 ties. An ordering of stations along some type of gradient was not 

 apparent for the June data (Fig. 5). 



The August data disclosed no marked similarities between 

 communities upstream of station 8 (Fig. 4). All stations downstream 

 from and including 8 (with the exception of station 9) seemed to be 

 quite similar in terms of macroinvertebrate communities. The August 

 ordination (Fig. 5) showed a polarization tendency with upstream 

 stations at an opposite end of the graph from the downstream 

 stations (8, 10 through 14). Stations 5, 6, 7, and 9 were dissimilar 

 from both these upstream and far downstream communities. This 

 separation of the river into an upstream association and far down- 

 stream association is in contrast to the June ordination, which 

 shows a more random distribution of upstream and downstream 

 stations. Again, river flow is thought to be quite significant in 

 forming such a pattern. In August a gradient that correlated with 

 such disturbances as sewage-treatment plant outfalls, industrial 

 effluents, and/or thermal discharges was apparent (Fig. 5). 



The August ordination can be viewed as intermediary in a 

 progression caused by pollutional effects when compared to the 

 September ordination (Fig. 5). The separation of the river into an 

 upstream and downstream grouping of stations is accentuated in 

 September. The September similarity matrix diagram (Fig. 4) also 

 displayed this conspicuous clumping of communities. Stations 3 and 

 4 were quite similar to both stations 1 and 2 for this collection 

 period. The macroinvertebrates of station 5, the station immediately 

 below the Tait power-plant discharge, showed strong association with 

 the macroinvertebrates below both the Dayton sewage-treatment 

 plant (station 6) and the Moraine sewage-treatment plant (station 8). 

 Stations 10 through 14 continued to form a closely related cluster. 



