of distribution—that is, those increasing or de- 

 creasing in abundance in the sarae divisions of 

 the stream— V. ere considered associated, either with 

 each other or with similar or related ecological 

 factors. The association of several species was 

 measured statistically. 



The fish collections included a large pro- 

 portion of the fish population present in each part 

 of the stream censused, and so data on the indi- 

 vidual species lend themselves to statistical 

 treatment. The usual statistical measures of 

 association based on the presence or absence of 

 a species were of no significance in the treatment 

 of these data, because most of the abundant fishes 

 were taken in all 52 collections. Coefficients of 

 correlation (r) between figures involving distribu- 

 tion of the various species were calculated to give 

 indices of association. In species of the darter 

 family and in three species of minnows, the indices 

 were based on numbers of individuals, in all other 

 species on total weights. In order to reduce the 

 number of mathematical calculations and make the 

 size of the samples as nearly uniform as possible, 

 the analyses were based on the total collections 

 from each of the eight divisions rather than on 

 separate collections. Each of the correlation co- 

 efficients is simply a mathematical expression of 

 the degree of similarity between the distribution 

 patterns of the species being compared; the figures 

 do not explain or measure- any other relationship 

 between the species. 



Because the smallmouth black bass was the 

 fish most sought after by Jordan Creek anglers, 

 particular attention was given the fish communities 

 with which it was associated. Its distribution was 

 compared with that of each of several other abun- 

 dant species, table 11. 



Table 11. — Coefficient of correlation (r) for 

 the weight distribution of the smallmouth black 

 bass and that of nine other species of fishes in 

 the eight divisions of Jordan Creek, 1950. 



Species i 



Hog sucker 

 Hornyhead chub 

 Common shiner 

 White sucker 

 Stoneroller 

 Green sunfish 

 Longear sunfish 

 Golden redhorse 

 Creek chub 



0.85 



0.62 



0.56 



0.35 



0.04 



0.01 



-0.18 



-0.22 



-0.73 



A graph, fig. 5, was prepared to show the 

 relative abundance of the smallmouth and the other 

 main groups ot fishes in a series of 24 pools ranked 

 as good, mediocre, or poor on the basis of the 

 weight of bass they contained. A high bass pop- 

 ulation seemed to be associated with a high per- 

 centage of suckers anda low percentage of minnows. 

 The percentages of sunfishes (exclusive of small- 

 m.outh bass) and miscellaneous species were fairly 

 constant in this series of pools, dropping off only 

 slightly where the minnows were dominant. A 

 direct relationship between bass and suckers and 

 an inverse relationship between bass and minnows 

 probably reflected the habitat preferences of 

 these fishes. 



The smallmouth, fig. 6, and the hog sucker, 

 fig. 7, had a similar weight distribution, being most 

 abundant in Division 2 and generally decreasing in 

 their abundance upstream. The highest correlation 

 of the smallmouth (r =0.85) was with the hog sucker 

 and is significant at the 1 per cent level. These 

 species have similar distributions in Illinois. The 

 high degree of association in Jordan Creek probably 

 is best explained by the preference of both 

 for hard- bottomed pools. 



The hornyhead chub, fig. 8, had a weight 

 distribution pattern in the study section somewhat 

 similar to that of the smallmouth black bass, al- 

 though the weight of the chub was greatest in 

 Division 3 and the weight of the smallmouth was 

 greatest in Division 2. The correlation coefficient 

 of 0.62 between the weights of these species is 

 somewhat high but not significant. 



The abundance of the longear sunfish, fig. 9, 

 reached a peak in Division 5 and decreased at 

 stations upstream and downstream from Division 5. 

 Gerking (1949) suggested that the presence of the 

 longear sunfish was antagonistic to the smallmouth 

 black bass in three streams that he surveyed. 

 Antagonism between these species did not seem to 

 prevail in Jordan Creek. There was a negative 

 correlation (r =-0.18) between the weight of the 

 smallmouth and that of the longear sunfish, but it 

 is not significant. However, the negative correla- 

 tion (r =-0.73) between the weight of the small- 

 mouth black bass and the weight of the creek chub 

 is significant. This strong negative correlation 

 reflects the upstream increase of the chub and the 

 difference in habitat preference between the two 

 species. The creek chub, which is a headwater 

 form, showed a gradual increase in numbers in the 

 upstream stations, but in weight it had an irregular 

 distribution pattern, fig. 11. The graph for this 

 chub and that for the hornyhead chub indicate these 

 fish are of smaller average sizes progressively up- 



17 



