FISHERY BULLETIN: VOL. 81, NO. 4 



TABLE 7. — Summary of statistics from regression analyses for 

 Indian River fishes. R values are Pearson correlation coefficients; N 

 is the total number of fish collected; F values are for tests of regres- 

 sion significance by analysis of variance. 



Table 8. — Daytime abundance of fishes and fish species in three 

 seagrass beds in Indian River lagoon, Fla. Values are mean num- 

 bers per trawl sample ± SD (n — 7). * and + indicate mean values 

 that were not statistically different on a given date (ANOVA and 

 Duncan's multiple range test, P < 0.05; F values are provided). 



The number of species collected per trawl varied lit- 

 tle among the stations (Table 8); however, certain 

 compositional differences were evident in the collec- 

 tions from different sites. Orthopristis chrysoptera, 

 Hairdiella chrysoura, and Syngnathus louisianae in- 

 creased in abundance from Syringodium to Thalassia 

 to Halodule beds; however, because of wide variation 

 in the numbers collected in replicate trawls, differ- 

 ences in catch among the beds were not significant 

 statistically (ANOVA, P > 0.10). The gerreids, 

 Eucinostomus gula. and E. argenteus, were most 

 abundant in Halodule beds in October and January 

 (ANOVA, P < 0.05), the only months during which 

 they were caught in large numbers. Both species were 

 uncommon at the Thalassia bed. Code goby, 

 Gobiosoma robustum, was abundant only at the 

 Syringodium bed. Although other species were 

 collected in nearly equal numbers in the three 

 seagrass species, Halodule yielded mean abun- 



dances of non-pinfish higher than those from 

 Thalassia and Syringodium in January and July 

 (Table 8). Thalassia and Syringodium yielded statis- 

 tically similar numbers of non-pinfish in October, 

 January, and July. 



Despite major differences in the gross morphology 

 of the three seagrass species, blade density (Table 6) 

 proved to be a better predictor of fish abundance 

 than macrophyte biomass. Using all station- date 

 combinations, there was a significant positive cor- 

 relation between total number of fishes collected, 

 number of pinfish, and number of non-pinfish (Table 

 7). Despite low sample numbers (n = 3), collections 

 made in January, April, and July showed surprisingly 

 close, and statistically significant, correlations be- 

 tween total number of fishes collected at a given site 

 and macrophyte blade density (Table 7). 



Although the Halodule bed yielded a higher total 

 number of species (32) than the Thalassia and 

 Syringodium beds (26 and 24, respectively), the num- 

 ber of species collected was a direct function of the 

 number of individuals collected and may be artifac- 

 tual. Differences in number of species collected per 

 trawl were rarely significant (Table 8). 

 Length- frequency analysis of pinfish populations at 

 Indian River stations (Table 9) showed that each of 

 the three smallest size classes increased in total num- 

 bers from Syringodium to Thalassia to Halodule 

 beds, but relatively small populations of fish>80 mm 

 were uniformly distributed over the three habitats. 

 Within- site analysis indicated that nearly 95 % of the 

 pinfish population on the Syringodium bed were > 35 

 mm; numerically, smaller fish were much more im- 

 portant at the Thalassia and Halodule beds, making 

 up 14 and 23.19c of the total, respectively. Conse- 

 quently, mean pinfish size decreased from Syringo- 

 dium to Thalassia to Halodule beds, and numbers of 

 individuals increased. 



Night collections in Indian River lagoon yielded 

 much larger numbers of fish per trawl than daytime 

 collections in all three of the seagrass beds (Table 

 10); mean values for all fish species were between two 

 and five times greater at night. The same was true for 

 pinfish as well as non-pinfish. Although only two 



Table 9. — Distribution of Lagodon rhomboides at three seagrass 

 beds in Indian River lagoon, Fla., based on size class. SL is standard 

 length of fish; N is the total number of fish collected in a size 

 class. 



842 



