ALLEN: LITTORAL FISH ASSEMBLAGE 



tance of these factors to this assemblage. Indi- 

 vidual correlations between abiotic factors and 

 species abundances likewise emphasized the im- 

 portance of temperature and salinity. The corre- 

 lations between individual species abundances 

 and dissolved oxygen as well as distance into the 

 upper Newport Bay could be due to the intercor- 

 relations of both dissolved oxygen and distance 

 with temperature. 



Intercorrelations among factors can confound 

 the interpretation of relationships and introduce 

 redundancy in multivariate analyses. The rela- 

 tionship between dissolved oxygen and distance 

 into the upper Newport Bay is intuitive consider- 

 ing its shallow depths. The positive relationship 

 between temperature and dissolved oxygen was 

 probably due to photosynthesis by green algae 

 during the summer. Winter rainfall in the basic- 

 ally Mediterranean climate of southern Califor- 

 nia was responsible for the positive correlation 

 between temperature and salinity found in New- 

 port Bay. This relationship is by no means abso- 

 lute, as evidenced by the low salinities encoun- 

 tered during the tropical rains of September 

 1978 when temperatures were high. 



The results of the second canonical correlation 

 analysis indicate that interaction between tem- 

 perature and salinity explained most of the vari- 

 ability in species abundance in this system. The 

 correlation between these two abiotic factors 

 probably inflated the R 2 value slightly, but does 

 not negate the overall findings. Ordination of in- 

 dividual species by correlation coefficients with 

 temperature and salinity underscores the influ- 

 ences of these factors on individual species. Fur- 

 thermore, the substantial decrease in numbers of 

 A. affinis at station 1 and the somewhat smaller 

 decrease at station 3 during September rains 

 (low salinity) and relatively high temperatures 

 also illustrate this temperature-salinity inter- 

 action. 



I propose that an important consequence of 

 temperature-salinity influence found in the 

 present study is the transfer of biomass and, 

 therefore, energy from the littoral zone to the 

 adjacent channel and ultimately to local offshore 

 areas via migration of fishes. This mechanism 

 for energy transfer was best illustrated by the 

 apparent emigration of a large portion of the 0- 

 age class A. affinis from the littoral zone from 

 September to December 1978. The transfer also 

 included the biomass produced by essentially all 

 of the periodic species. Weinstein et al. (1980) 

 reached a similar conclusion in their study of the 



fishes in shallow marsh habitat of a North Caro- 

 lina estuary. An extensive mark and recapture 

 study should be planned to test this hypothesis in 

 the future. 



Seasonal fluctuations of temperate bay-estua- 

 rine fish populations may have several causes, 

 but temperature and salinity seem frequently to 

 be the underlying factors. The pattern of in- 

 creased number of species and individuals with 

 increased temperature in temperate bays and 

 estuaries has been reviewed by Allen and Horn 

 (1975). Recently the large-scale influence of sa- 

 linity on bay-estuarine fish populations has been 

 demonstrated by Weinstein et al. (1980) for Cape 

 Fear River Estuary, N.C. Unfortunately, any 

 salinity interaction with temperature was not in- 

 vestigated or discussed in the above study. 



Studies of subtropical estuaries (Amezcua- 

 Linares 1977; Warburton 1978; Quinn 1980) in- 

 dicate that salinity may have greater influence 

 on fish populations, since annual temperature 

 ranges are narrower than in temperate bays and 

 estuaries. In each of the above studies on sub- 

 tropical estuaries, increased abundances cor- 

 responded to the season of low rainfall and there- 

 fore high salinity. Blaber and Blaber (1981) con- 

 cluded that turbidity and not temperature and 

 salinity was the single most important factor to 

 the distribution of juvenile fishes in subtropical 

 Moreton Bay, Queensland. However, Blaber and 

 Blaber ( 1981) did not present statistical evidence 

 to support this contention. The most important 

 environmental factors influencing tropical estu- 

 arine (eelgrass) ichthyofaunas are more difficult 

 to identify (Weinstein and Heck 1979; Robertson 

 1980) and probably include biotic factors such as 

 prey availability, competitors, predators, as well 

 as abiotic factors. Biotic interactions are un- 

 doubtedly important in temperate estuarine sys- 

 tems including upper Newport Bay. However, 

 their overall influence on the system is probably 

 swamped by large fluctuations in the physical 

 environment. 



Fluctuations in rainfall and temperature re- 

 gimes during a year and from year to year can 

 have marked effects on the ichthyofauna of estu- 

 aries. Moore (1978) has identified long-term 

 (1966-73) fluctuations in summer fish popula- 

 tions in Aransas Bay, Tex. He found that diver- 

 sity values (H' range of 1.38-2.13) were quite 

 variable from year to year probably as a result of 

 major climatological changes (an unusually wet 

 year; a drought and two hurricanes). These 

 changes in diversity values were probably caused 



787 



