Printer Friendly

CONDITION-SPECIFIC COMPETITION: IMPLICATIONS FOR THE ALTITUDINAL DISTRIBUTION OF STREAM FISHES.

YOSHINORI TANIGUCHI [1,3]

SHIGERU NAKANO [2,4]

Abstract. The occupation of adjacent, nonoverlapping positions along environmental gradients by closely related and ecologically similar species has drawn considerable attention from many ecologists over the past decades. Condition-specific competition, wherein competitive superiority varies with the abiotic environmental gradient, has been proposed as the major structuring force behind such distributions. However, few studies have elucidated the underlying mechanisms, such as behavioral and demographic processes. We conducted laboratory experiments to examine the effects of temperature on interspecific competition between two stream salmonid fishes, Salvelinus malma and S. leucomaenis. The two species have a largely allopatric altitudinal distribution on Hokkaido Island, Japan, proposed to be the result of temperature-mediated competition. We tested predictions that at a higher temperature (12[degrees]C), S. leucomaenis would dominate over S. malma in aggressive interactions, foraging performance, growth, a nd survival, but become subordinate at a lower temperature (6[degrees]C). Indeed, S. leucomaenis initiated a greater number of aggressive acts, attained greater food intake and greater growth, and finally excluded S. malma at the higher temperature. Although the two species initiated a similar number of aggressive acts and foraged equally well at the lower temperature, S. leucomaenis achieved a higher growth rate than S. malma; however, the latter eventually became numerically dominant. Clear competitive release in allopatry occurred for S. malma only at the higher temperature, providing direct evidence of condition-specific asymmetric competition. The lower distribution boundary of S. malma in Hokkaido streams may therefore be determined by temperature-mediated condition-specific competition. However, mechanisms determining the upper distribution boundary of S. leucomaenis could not be fully explained by the competitive results at lower temperature, but required an understanding of how effects of competition interacted with species-specific physiological traits. Thus, species distributions along an environmental gradient cannot be solely explained by a simple model of condition-specific competition without considering mechanistic linkages among behavioral and physiological responses to the environment, resource use, and demographic processes.

Key words: altitudinal distributions; competition; growth rates; mechanistic approaches, condition-specific competition; size-dependent mortality; stream salmonids; survival analysis; temperature and competitive exclusion.

INTRODUCTION

Determinants of species' distribution patterns have long interested many ecologists, with the role played by competition having been hotly debated (Connell 1961, MacArthur 1972, Simberloff 1983, Abrams 1986, Moulton and Pimm 1986, Wiens 1989). Of particular interest are the abutting distributions of closely-related species along environmental gradients, since such species are often assumed to compete strongly (Diamond 1970, Terborgh 1971, Schluter 1982, Brown and Bowers 1984, Bull 1991). Where one species is absent or experimentally removed, niche expansion or density compensation observed for another species provides support for interspecific competition (Jaeger 1971, Terborgh and Weske 1975, Diamond 1978, Hairston 1980). Although abiotic factors can modify the outcome of competition along the environmental gradient, mechanisms underlying the interactions between biotic and abiotic factors have rarely been elucidated (Hutchinson 1961, Jaeger 1970, Grace and Wetzel 1981).

Condition-specific competition, wherein competitive superiority depends upon environmental conditions (Chesson 1986, Dunson and Travis 1991), has been proposed as the major structuring force for many communities (e.g., Hutchinson 1961, Jaeger 1970, Wilbur 1987). Such condition-specific competition has generally been examined by two alternative approaches, testing for effects on either demographic parameters or on characteristics of individuals. Warner et al. (1993) demonstrated that differing pH levels changed demographic processes between two species of anuran larvae, owing to changed competitive intensities, and thus successfully addressed the question of condition-specific competition at the population level. However, few studies have focused on the mechanisms behind such changes in demographic parameters. Several studies have examined how characteristics of individuals (aggressive behavior, feeding efficiency) may change with changes in physicochemical factors but without considering the competitive outc ome at the population level (Hartman 1966, Baltz et al. 1982, Reeves et al. 1987, De Staso and Rahel 1994). Therefore, a more thorough investigation linking both approaches can enhance our understanding of condition-specific competition.

During recent decades, the importance of mechanistic approaches has been emphasized for examining how individual differences in physiology and behavior are functionally linked to higher level processes such as population dynamics, multispecies interactions, and species distributions (Schoener 1986, Tilman 1987, Sutherland 1996). Experimental analyses of condition-specific competition are required to further of our understanding of the mechanisms responsible for the distributions of competing species. For instance, interspecific differences in physiological tolerance may influence resource acquisition and growth rates between competitors (Persson 1986, Reeves et al. 1987, Christie and Regier 1988). Differential growth may, in turn, have significant consequences at the population level, since slow growers may have higher mortality (see Elliott 1994), and this can lead to a cascade of consequences at the community level (Wellborn et al. 1996). Nevertheless, few studies have demonstrated effects of abiotic facto rs on both individual properties and population processes responsible for species distribution patterns along environmental gradients.

Many stream ecosystems worldwide are characterized by species replacement patterns along altitudinal gradients (Sheldon 1968, Vincent and Miller 1969, Schlosser 1982, Rahel and Hubert 1991). Because water temperature changes with altitude, temperature-mediated competition has been proposed as the major determinant of these altitudinal replacement patterns (Fausch 1989). We experimentally examined condition-specific competition between two native congeneric salmonid fishes, Salvelinus malma (Dolly Varden charr) and S. leucomaenis (white-spotted charr) that show altitudinal replacement in streams of northern Japan (Fausch et al. 1994). Our results show how biotic and abiotic effects operate synergistically to determine the distribution of these species.

STUDY SYSTEM

Salvelinus malma and S. leucomaenis are morphologically similar, closely related (Cavender 1989), and mostly allopatric in streams of Hokkaido, the northern island of the Japanese archipelago (Ishigaki 1987, Fausch et al. 1994). Most S. malma populations are entirely fiuvial and S. leucomaenis, though largely anadromous, live 2-7 yr in rivers, with landlocked forms common above waterfalls and man-made dams (Fausch et al. 1994). The two species have similar spawning (October-November) and emergence periods (March-April; Ishigaki 1987), and therefore juveniles may compete for shared resources in the zone of sympatry.

Where the two species occur in sympatry, both feed on drifting invertebrates in stream pools by maintaining focal points in the current and compete by interference for favorable foraging territories in interspecific sizestructured dominance hierarchies (Nakano and Furukawa-Tanaka 1994, Nakano et al. 1999b). Population densities of S. malma and S. leucomaenis in individual pools were negatively correlated and remained relatively constant for several years (Fausch et al. 1994; S. Nakano, unpublished data). Thus, interspecific competition most likely plays an important role in determining the relative local abundance of these species when occurring in sympatry.

The two species are known to have different thermal requirements. Foraging activities decline above 16[degrees]C for S. malma and above 22[degrees]C for S. leucomaenis (Takami et al. 1996). In sympatric streams, stream temperatures are coldest at allopatric S. malma sites (5-8[degrees] C), intermediate in areas of sympatry (7-10[degrees] C), and warmest at allopatric S. leucomaenis sites (10-15[degrees] C; Fausch et al. 1994, Taniguchi 1998). However, in streams where one species is absent, the other expands its thermal distribution range (Kitano et al. 1995, Nakano et al. 1996). Thus, the temperature ranges where the two species are found in sympatry are considerably narrower than those in allopatry, which should primarily represent their thermal tolerances (Nakano et al. 1996).

We constructed a verbal model to explain the distribution patterns of the two species. First, water temperatures may differentially restrict physiological performances (see Beamish 1964, Elliott 1976). Since the preferred temperature range of S. leucomaenis is likely higher than that of S. malma, S. leucomaenis should dominate in aggressive interactions at higher temperatures (see De Staso and Rahel 1994, Taniguchi et al. 1998), thereby acquiring energetically more profitable foraging microhabitats and achieving greater growth rates than the subordinate S. malma. Such growth differentials would eventually force subordinate S. malma to emigrate from the local habitat or die (see Elliott 1994). The reverse should be true at lower temperatures, with S. leucomaenis becoming subordinate. Hence, altitudinal distributions of these two species should result from an outcome of such condition-specific competition (i.e., competitive exclusion).

METHODS

Experimental design, laboratory stream, and test fishes

Twelve experimental stream reaches were used for the experiments, providing for both allopatry of each species and sympatry of the two species at two temperatures, 6[degrees]C and 12[degrees]C. We based these test temperatures on the thermal conditions in which thriving allopatric populations of S. malma (6[degrees]C) and S. leucomaenis (12[degrees]C) are commonly found, and expected that each species would perform better at the respective temperature. For each temperature treatment, three species combinations were used; allopatric populations of 50 S. malma, allopatric populations of 50 S. leucomaenis, and sympatric populations of 25 S. malma and 25 S. leucomaenis; each with two replicates. The stocking of 50 individuals in each experimental reach resulted in an initial density of 51 fish/[m.sup.2], which was 1.6 times higher than what is naturally realized for stream salmonids of this size (Grant and Kramer 1990). However, interpretation of the fish behavior and mortality under such condition should still be realistic because high initial densities are immediately reduced as a consequence of competition-induced mortality (see Elliott 1994). Indeed, such reductions occurred in the present study, particularly at the higher temperature.

The laboratory experiments were conducted at the Tomakomai Experimental Forest (TOEF) of Hokkaido University, Japan. The experimental reaches (each 2.80 m long, 0.35 m wide, and 0.12 m deep) were created by isolating portions of each of six stream tanks constructed of polyvinyl bicarbonate. Each experimental reach was blocked at its upstream and downstream ends by 1.5-mm mesh stainless steel screens. Observation windows (2.8 m long X 0.25 m wide) allowed observers, who sat in a darkened observation room next to each stream tank, to record fish behavior. Water temperatures were controlled ([pm] 0.2[degrees]C) by chilling and heating units using a thermostat. Flow was generated by an air-lift pump located at the most upstream point of each experimental reach. Water velocities ranged from 0.0 cm/s (measured at bottom) to 38.5 cm/s (at surface and mid-water column). The substrate was a mixture of gravel, 10-20 mm in diameter. The microhabitats thus created in the tanks were similar to those reported for young-of -the-year (age 0) charrs under natural conditions (Ishigaki 1987). The placement of stones ([approx]20 mm in diameter) on the stream beds so as to form 80 cells (9 cm wide X 14 cm long) facilitated the identification of each fish position. Filtration tanks (80 L) composed of sand and gravel were installed with each stream tank. In addition, the replacement of [sim]85% of the water in the stream tanks with filtered natural stream water every two weeks ensured good water quality throughout the experimental period. Overhead lighting was provided by two 100-W fluorescent bulbs 1.5 in above each stream tank from 0600 to 1800 (12h light: 12 h dark light regime) using a timer and diffused through a fine black mesh (0.5-mm mesh openings) to equalize intensities throughout the experimental reaches (range 220-260 lx at water surface).

All experimental fish were hatched from eggs of wild S. malma and S. leucomaenis resident in the Horonai Stream, a second-order stream running through the TOEF (see Nakano et al. 1999a). Eggs were obtained by stripping adult fish and artificially fertilized in November 1995; all fish hatched by January 1996. After yolk sac absorption, fry were fed commercial trout starter pellets (using a smaller particle size [400 [mu]m diameter and 0.05 mg] for the initial two weeks and a larger size [700 [mu]m and 0.1 mg] for the rest of the period until the experiments began) at 5% of their body mass adjusted for growth every two weeks. The two species were held separately in raceway tanks fed by a continuous input of filtered stream water at an ambient temperature fluctuating between 7[degrees]C and 13[degrees]C.

Fish were measured for fork length (FL) to the nearest 0.1 mm by digital caliper and weighed to the nearest 0.01 g. For S. malma, mean ([pm] 1 SE) initial FL and wet mass were 34.65 [pm] 0.07 mm and 0.32 [pm] 0.08 g, respectively. For S. leucomaenis, mean ([pm] 1 SE) FL and wet mass were 34.38 [pm] 0.06 mm and 0.31 [pm] 0.06 g. Lengths [greater than]30 mm FL were used to avoid unnecessary mortality due to marking. Mean lengths and masses did not differ between the two species in each of the four sympatric populations when the experiments began (P = 0.210 and P = 0.398 for the two sympatric tanks at 6[degrees]C, P = 0.107 and P = 0.430 for the two sympatric tanks at 12[degrees]C, by t test after [log.sub.10] transformation for all data). Thermal acclimation of the experimental fish was conducted prior to their individual markings while they were still held in the holding tanks (9[degrees]C), by either raising or decreasing the temperature 1[degrees]C/d and then holding fish at the constant test temperature for at least three days. Each fish was individually marked with elastomer fluorescent dye (Northwest Marine Technology, Shaw Island, Washington, USA), which proved to be fully biocompatible. Each group of marked fish was subsequently transferred to a designated experimental reach and acclimated to the stream and drifting prey for at least 5 d until the experiments began on 13 May.

To simulate natural invertebrate drift in the laboratory streams, commercial trout pellets (700 [mu]m and 0.1 mg) were released from an automatic pellet feeder located at the most upstream point of each experimental reach, ensuring continuous food input during the daylight period. Food was supplied to each reach at 4.5 particles per mm. Because the study objective was to observe competitive interactions, the daily input of food was limited to 2.0% of the initial fish body mass and remained unchanged throughout the experimental period (0.32 g/d supplied to each reach). Food particles released when fish were absent drifted downstream either on the water surface or in the water column, the majority (90%) reaching the downstream blocking screen and the rest (10%) apparently sinking to the bottom. When fish were present, however, virtually no food items were observed reaching the downstream screen or stream bed, all apparently having been consumed.

Fish observations and measurement

We recorded aggressive interactions and foraging activities during a 72-d period from 13 May to 19 July 1996. All of twelve experimental reaches were visited once within 5-d periods, resulting in 13 visits for each reach over the observation period. Observations for each reach were made in a random order with at least 3-d intervals between observations. For each visit, focal animal observations (2 mm per individual; Altmann 1974) were made for 10 and 20 randomly chosen fish for reaches of allopatry and sympatry, respectively, at various times during daylight hours (0700-1700). Within each visit, the same individual was not observed more than once, except for S. malma in the last four visits in sympatry at higher temperature when [less than]10 fish remained due to mortality. The behavior (aggression and foraging) and positions of both focal individuals and all other individuals with which the former interacted were recorded on 1:20 scaled maps.

Aggressive encounters where focal individuals initiated aggressive acts (cf. Nakano 1995a) were recorded. For those fish which initiated aggressive acts, aggressive distances (between the focal points of aggressors and snouts of recipient fish) were measured from the map records, and subsequently converted to actual distances. Foraging frequency was quantified by counting the number of foraging attempts for each individual over a 2-mm period and considered to reflect the energy intake rate in these experiments because all food particles were equal in size. Distance between upstream blocking screen and the fish snout (hereafter distance downstream) was also quantified for each focal individual from the maps. Such distances from food supply points is often quantified to describe the relative competitive abilities among the individuals (Fausch and White 1986, Hughes 1992) because locations near the pool inlets generally provide better access to drifting food items (Huntingford 1993, Nakano 1995a).

Fish lengths were measured five times at average intervals of 18 d; on days 1, 18, 38, 55, and 72; to construct length--frequency histograms that showed individual fish mortalities for periods 1-4. Individual specific growth rates in FL for each of the four 18-d periods were calculated according to Ricker (1979). For the measurements, fish were carefully netted with a dip net, anesthetized with 2-phenoxyethanol, and measured to the nearest 0.1 mm in FL with a digital caliper. Fish were allowed to recover in fresh water and released into their original reach in one hour. Because no mortality was observed during the following 24 h, handling stress was considered to be minimal.

We discontinued behavioral observation after 19 July 1996 (day 72) because the number of S. malma in sympatry at higher temperature was too low and the competitive ability of the species was no longer comparable to that of S. leucomaenis. Mortalities were, however, monitored until 15 November 1996 (day 191). The dead fish were immediately removed from the stream tanks and their body length measured. No disease or predation (or cannibalism) was observed throughout the 191-d experimental period.

Predictions and statistical treatment

We made the following predictions on how the two temperatures should mediate interspecific competition between the two species. At 12[degrees]C, S. leucomaenis will be competitively superior to S. malma by (1) showing higher aggressive frequency and longer aggressive distance, (2) holding shorter distances downstream and achieving a higher foraging frequency, (3) achieving higher specific growth rates during the 72-d period, and hence (4) displaying a higher survival rate throughout the 191 -d period. The opposite will hold at 6[degrees]C, where S. malma will be superior to S. leucomaenis. Moreover, slow growers (i.e., smaller fish) will die earlier than fast growers (i.e., larger fish), in both species at both temperatures. We also predicted that competitive release will occur for the poorer competitor in allopatry, i.e., it will perform better when released from inter-specific competition. Finally, levels of aggression, foraging, and growth were predicted to increase with temperature in allopatry as a phys iological response in both species.

We pooled the two replicates in all analyses because stream tank effects were not significant in preliminary analyses for aggressive frequency, aggressive distance, distance downstream, foraging frequency, and specific growth rate (t tests, P = 0.127-0.977). Each visit to experimental reaches during the 72-d period was treated as one replicate in behavioral analyses, yielding a total of 26 replicates. Statistical independence of the visits was considered plausible since time intervals between visits were long enough to assume the independence of behavioral bouts and observations were made on randomly chosen individuals within each visit. For the analyses of specific growth rates, an individual population in each stream tank was treated as one replicate.

All of the aggression- and foraging-related variables were analyzed separately for sympatry and allopatry by two-way ANOVAs, with temperature and species as the main effects. For these variables, one-way AN-OVAs followed by Fisher's protected least significant difference (PLSD) tests were conducted to test for differences in the species' means within and between low and high temperatures. Unpaired t tests were used to test for differences in the species means between the sympatric and allopatric populations at each temperature. For growth, after separate two-way repeated-measures ANOVAs for sympatry and allopatry, with species and temperature as main effects, a one-way repeated-measures ANOVA combining sympatry and allopatry was conducted. Multiple contrasts were also performed to detect differences in growth among groups. Raw data were [log.sub.10](x + 1)- or square-root(x + 0.5)-transformed, and all percentages arcsine square-root transformed to satisfy statistical normality and homogeneity of variance assumption s for ANOVAs and t tests. However, we present nontransformed data unless otherwise stated.

To examine size-dependent mortality, differences in mean FL were tested between dead and surviving individuals of each species, and between the two species, for each 18-d period using Mann-Whitney U tests. Tests were not conducted when sample sizes were less than five. Survival analysis was used to test for survival curve differences between the two species in sympatry and allopatry at both temperatures, and between sympatry and allopatry for each species. The Kaplan-Meier method of estimating survival functions and the nonparametric Mantel-Cox log rank test were used (Abacus 1994). Because stream tank effects were never significant ([[chi].sup.2] test, P = 0.119-0.616), the two replicates were pooled. Survival analysis has been regularly employed in medical science to analyze "incomplete" data recorded before the termination of the event of interest (Abacus 1994). Because many fish died before the present experiments were terminated, the incomplete data thus resulting could not be analyzed with traditional nonp arametric techniques (see Moore and Townsend 1998).

For all tests, [alpha] was set at 0.05, and all tests were two-tailed unless otherwise stated. StatView version 4.5 (Abacus 1994) was used for all statistical analyses.

RESULTS

Aggressive interactions

For aggressive frequency, in sympatry, two-way ANOVA revealed significant species and species X temperature interaction effects, with temperature effects not being significant (Table 1). Fisher's PLSD tests, after significant mean differences were found by one-way ANOVA, revealed that S. leucomaenis initiated more aggressive acts than S. malma at 12[degrees]C (P [less than] 0.001) despite comparable aggressive frequencies at 6[degrees]C (P = 0.602; prediction 1, Fig. 1). In S. malma, aggressive frequency at 6[degrees]C was higher than at 12[degrees]C (P [less than] 0.001), the opposite being true for S. leucomaenis (P = 0.002). In allopatry, the effects of species, temperature, and species X temperature interactions were all significant (Table 1, Fig. 1). Although aggressive frequency in sympatric and allopatric S. malma did not differ at 6[degrees]C (t = 0.02, P = 0.987), it was higher in the allopatric at 12[degrees]C (t = -5.07, P [less than] 0.001). In sympatric and allopatric S. Ieucomaenis, aggressive fr equencies did not differ at either 6[degrees]C (t = 0.34, P = 0.783) or 12[degrees]C (t =-0.96, P = 0.343).

Significant effects of species, temperature, and species X temperature interactions were also found for aggressive distance in sympatry (Table 1). Although significant differences were found among the means within sympatry (Fig. 1), subsequent Fisher's PLSD tests revealed that the two species did not differ in aggressive distance at 6[degrees]C (P = 0.436). However, S. leucomaenis had a longer aggressive distance than S. malma at 12[degrees]C (P [less than] 0.001; prediction 1). In S. malma, aggressive distances were similar at the two temperatures (P = 0.411), whereas in S. leucomaenis, it was longer at 12[degrees]C than at 6[degrees]C (P [less than] 0.001). In allopatry, the effects of species and temperature interactions were significant but not the species X temperature interaction (Table 1). Although aggressive distance in S. malma did not differ between sympatry and allopatry at 6[degrees]C (t = 0.84, P 0.406), it was greater in allopatry than in sympatry at 12[degrees]C (t =-2.90, P = 0.006). Aggressi ve distances of S. leucomaenis were also similar in sympatry and allopatry at 6[degrees]C (t = 0.90, P = 0.373), but was longer in sympatry than in allopatry at 12[degrees]C (t = 2.80, P = 0.007).

Foraging behavior

For the distance downstream in sympatry, significant effects were found for temperature and species X temperature interactions but not for species (Table 2). Although the distance downstream was similar between the two species at 6[degrees]C (P = 0.052), it was longer in S. malma than in S. leucomaenis at 12[degrees]C based on Fisher's PLSD tests after one-way ANOVA (P = 0.034; prediction 2, Fig. 2). Moreover, the distance downstream at 12[degrees]C was longer than that at 6[degrees]C in S. malma (P [less than] 0.001) but not so in S. leucomaenis (P = 0.859). In allopatry, species and temperature effects, but not species X temperature interactions, were significant (Table 2). In S. malma, in spite of similar distances in sympatry and allopatry at 6[degrees]C (t = 1.31, P = 0.198), the distance downstream in sympatry was greater than that in allopatry at 12[degrees]C (t = 2.54, P [less than] 0.014). Distances downstream for S. leucomaenis, however, were similar in sympatry and allopatry at both 6[degrees]C (t = 0.21, P = 0.838) and 12[degrees]C (t = -0.47, P = 0.641).

Two-way ANOVA indicated significant effects of temperature and the species X temperature interaction on foraging frequency in sympatry (Table 2). Fisher's PLSD tests indicated similar foraging frequencies for S. malma and S. leucomaenis at 6[degrees]C (P = 0.426), but a higher foraging frequency in S. leucomaenis at 12[degrees]C (P = 0.003; prediction 2, Fig. 2). Although the foraging frequency of S. leucomaenis at 12[degrees]C was higher than that at 6[degrees]C (P [less than] 0.001), no difference was found for S. malma between the two temperatures (P = 0.528). In allopatry, significant species' and temperature effects, but not species X temperature interaction, were apparent (Table 2). Foraging frequency of S. malma was similar in sympatry and allopatry at 6[degrees]C (t = 1.05, P = 0.300), but was higher in allopatry than in sympatry at 12[degrees]C (t = -2.76, P = 0.008). In contrast, foraging frequencies of S. leucomaenis were similar in sympatry and allopatry at both 6[degrees]C (t = 0.01, P = 0.994) and 12[degrees]C (t = 1.45, P = 0.153).

Growth and survival

The two-way repeated-measures ANOVA of specific growth rates revealed significant effects of species, temperature, time, and species X temperature interaction in sympatry (Table 3). However, in allopatry, none of these effects were significant except for time. In sympatry, one-way repeated-measures ANOVA with treatment as the main effect showed significant differences among the eight group means (Fig. 3). Multiple contrasts showed the growth of sympatric S. leucomaenis to be higher than that of S. malma, not only at 12[degrees]C (P [less than] 0.001) but also at 6[degrees]C (P = 0.002; prediction 4). Although the growth of sympatric S. malma was higher at 6[degrees]C than at 12[degrees]C (P [less than] 0.001), that of S. leucomaenis was greater at 12[degrees]C than at 6[degrees]C (P = 0.029). Moreover, although S. malma grew similarly in sympatry and allopatry at 6[degrees]C (P = 0.110), S. leucomaenis grew more in sympatry than in allopatry at that temperature (P = 0.002). Growth in S. malma was significant ly higher in allopatry than in sympatry at 12[degrees]C (P [less than] 0.001), but the opposite was true for S. leucomaenis (P [less than] 0.001).

Size-dependent mortality was evident in both species, in both sympatry and allopatry, in all of the periods that were tested (see Methods). As predicted, fish that died were smaller than survivors, as illustrated by S. malma in period 3 and S. leucomaenis in periods 3 and 4 in sympatry at 12[degrees]C (Fig. 4). The same trend was found in allopatry in periods 3 and 4. At 6[degrees]C, however, most data sets had fewer mortalities than could be tested. Only S. leucomaenis were tested in periods 3 and 4, when dead fish were again smaller than survivors. Sizes of fish that died were also compared between species. Tests were possible only for periods 3 and 4 at 12[degrees]C; S. leucomaenis that died were larger than mortalities of S. malma in both sympatry and allopatry (Fig. 4).

Survival rates during the 191-d period differed markedly between the two species, varying with temperatures and species combinations (Fig. 5). At the higher temperature, 2-10% of populations survived to day 191 in all species combinations except for sympatric S. malma, where no individuals survived beyond day 84. At the lower temperature, fewer mortalities were observed in both species, where 31-85% of populations survived irrespective of species combinations. Survival analysis revealed that S. malma in sympatry survived better than S. leucomaenis at 6[degrees]C (Mantel-Cox [[chi].sup.2] = 17.56, P [less than] 0.001; prediction 4). In contrast, the survival rate of S. leucomaenis was higher than that of S. malma at 12[degrees]C ([[chi].sup.2] = 20.20, P [less than] 0.001). The population of S. malma declined sharply at 12[degrees]C, becoming extinct on day 84, when 30% of S. leucomaenis still remained. In allopatry, S. malma survived better than S. leucomaenis at both 6[degrees]C ([[chi].sup.2] = 65.97, P [less than] 0.001) and at 12[degrees]C (Mantel Cox [[chi].sup.2] = 31.87, P [less than] 0.001; Fig. 5). Survival rates of both allopatric and sympatric S. malma ([[chi].sup.2] = 0.03, P = 0.853) and S. leucomaenis ([[chi].sup.2] = 3.65, P = 0.056) did not differ at 6[degrees]C (Fig. 5). At 12[degrees]C, however, allopatric S. malma had higher survival than the sympatric population ([[chi].sup.2] = 117.81, P [less than] 0.001), whereas survival of allopatric and sympatric S. leucomaenis was similar ([[chi].sup.2] = 1.94, P = 0.163).

DISCUSSION

Aggressive interactions and foraging behavior

In the present study, interspecific aggressive interactions between S. malma and S. leucomaenis were strongly mediated by water temperature (prediction 1). In allopatry, higher temperatures enhanced aggression in both species. However, for aggressive frequency, this temperature effect was reversed for S. malma in sympatry. This reversal appeared to be due to competitive suppression by sympatric S. leucomaenis, which initiated many more aggressive acts than S. malma, and from a greater distance, at the higher temperature. Thus, competitive ability of S. leucomaenis was enhanced at the higher temperature relative to S. malma's (see Takami et al. 1996). Similarly, the former species displayed a greater increase in aggressive frequency with increasing temperature in allopatry compared with the latter.

Distance downstream and foraging frequency, parameters reflecting access to drifting food and foraging return, respectively (Hughes 1992, Nakano 1995a), were clearly influenced by water temperature and generally reflected the overall trend observed in aggressive behavior. Salvelinus malma were apparently suppressed in obtaining favorable foraging positions at the higher temperature in sympatry, whereas they were clearly released from competition when S. leucomaenis were absent. In allopatry, S. leucomaenis were positioned further downstream than S. malma at the higher temperature, which was probably due to the stronger intraspecific interference competition in S. leucomaenis (see Discussion: Growth and survival). Salvelinus leucomaenis initiated more aggressive acts than S. malma, which forced subordinate conspecifics downstream. Although the foraging frequency in both species increased with water temperature in allopatry, this temperature effect disappeared in sympatric S. malma. In contrast, foraging frequ ency increased with temperature in sympatric S. leucomaenis (prediction 2). Thus, foraging by S. malma was suppressed at the higher temperature when in sympatry, but displayed a release from interspecific competition in allopatry. Foraging frequency generally reflected distance downstream with the species positioned further upstream achieving higher foraging frequency than the one further downstream. However, better access to a food source was not always followed by higher foraging return. Discrepancies resulted from the confounding effects of intensified competition that increased distance downstream and enhanced fish appetite at the higher temperature.

Substantial size differences that developed between the two species were not taken into account in the present analysis. However, the larger body size attained by sympatric S. leucomaenis at the higher temperature in the later periods (periods 3-4), compared to S. malma, were reflected in the behavioral dominance of the former over the latter species. Species-specific differences in physiological response and resultant differences in competitive ability are likely the reasons behind such growth differentials (Wootton 1990). Slight size advantages were probably gained early on by S. leucomaenis and were subsequently amplified by interference competition (i.e., growth depensation sensu Magnuson 1962). In contrast, although sympatric S. malma were slightly smaller than S. leucomaenis at the lower temperature, especially by period 4, the two species were generally equal competitors in terms of aggressive and foraging behavior. This discrepancy between body size and behavior supports the higher physiological aptn ess of S. malma than S. leucomaenis at the lower temperature.

Growth and survival

There were striking differences in the effects of water temperature on the specific growth rates between the two species, although aggressive and foraging behavior only partly accounted for such differences. In allopatry, the mean growth rates of both species were not influenced differently by water temperature. In contrast, growth of sympatric S. malma decreased with increasing temperature, whereas that of sympatric S. leucomaenis increased at the higher temperature (prediction 3). When sympatric, S. malma grew less than S. leucomaenis at both temperatures, but the difference was much greater at the higher temperature. Given that the growth rate of the sympatric S. malma at the higher temperature was considerably lower than that of the allopatric population, the latter were apparently released from interspecific competition with S. leucomaenis, as expected from the behavioral observations. Dominant S. leucomaenis attained energetically more profitable upstream sites in sympatry, assuring them of greater for aging success and suppressing S. malma (cf. Fausch and White 1986). That sympatric S. leucomaenis grew faster than S. malma at the lower temperature conflicted with prediction 3, as well as with the observation of equal competitive ability in foraging at that temperature. However, sympatric S. malma were not considered suppressed, since their growth rate was similar to that obtained in allopatry. Thus, temperature mediated the competitive influences on growth rates, as reported by Reeves et al. (1987) in their experiments on salmonid-cyprinid competition.

Similar growth of S. malma at the lower temperature in sympatry and allopatry may also suggest similar effects of inter- and intraspecific competition. In contrast, the intensity of intraspecific competition in S. leucomaenis was apparently greater than that of interspecific competition, as indicated by their lower growth rate in allopatry than in sympatry at both temperatures. In addition, similar growth rates in both species were observed at the two temperatures in allopatry, despite the higher foraging frequencies at the higher temperature. This was likely due to increases in both the basal and the active metabolism (see Elliott 1976).

Over the 72-d period, considerably greater size variation developed in both species at the higher than lower temperature, likely a result of greater differences in energy allocation among individuals at the higher temperature (see Elliott 1994). This process gradually increased the discrepancy between dominant and subordinate (Chapman 1962, Nakano 1995b). Such discrepancies in competitive abilities were likely accelerated at higher temperature, finally leaving several large despotic individuals in each stream tank in the last observation period.

Survival rates during the entire 191-d period differed markedly between the two species, varying with temperatures and species combinations, and reflecting the trends observed in growth differentials during the 72-d period. At the higher temperature, the sympatric population of S. malma became extinct within the 191-d period (whereas [greater than]30% of S. leucomaenis remained), despite the survival rate of allopatric S. malma being higher than that of allopatric S. leucomaenis at the same temperature, as expected from the observed competitive release of the former species. In contrast, S. malma survived better than S. leucomaenis at the lower temperature in sympatry (prediction 4). Because there was virtually no sign of disease or predation during the entire experimental period, all mortalities observed during the 191-d period were considered to have occurred in subordinate individuals by starvation due to the strong competitive suppression by the dominant fish as observed during 72-d period. Lower surviva l rates in S. leucomaenis than S. malma in allopatry irrespective of temperatures suggest the higher intraspecific competition in the former, partly reflecting the results in behavior and growth. Although effects of interspecific competition on population abundance have been reported in other stream salmonids under experimental conditions (Fraser 1969), behavioral mechanisms were not previously elucidated.

At the lower temperature, the higher survival rate of S. malma than S. leucomaenis in sympatry was not consistent with the growth rate or food acquisition results. These discrepancies were probably due to a difference in species-specific physiological traits, i.e., starvation resistance (cf. Shuter and Post 1990) being higher in S. malma than S. leucomaenis, because similar-sized individuals of S. malma experienced considerably fewer mortalities than S. leucomaenis in both sympatry and allopatry at both temperatures. At the higher temperature in sympatry, a lower starvation resistance in S. leucomaenis was probably masked by the effects of highly asymmetric competition. Thus, an interspecific difference in growth may not predict precisely which species will survive better, particularly at the lower temperature, although size-dependent mortality plays a critical role in population regulation irrespective of temperatures.

Implications for the species distribution

Streams with sympatric charrs in Hokkaido are characterized by S. malma at higher altitudes, S. leucomaenis in lower reaches, and a zone of sympatry in between (Fausch et al. 1994). Our experimental results for the higher temperature in sympatry, in which S. leucomaenis consistently dominated over S. malma, likely represent the outcomes of competitive interactions occurring in the contact zones and downstream reaches, i.e., the lower distribution boundary of S. malma is determined by interspecific asymmetric competition induced by the higher water temperature.

Our observation at the lower temperature showing that neither species dominated behaviorally and S. leucomaenis grew better than S. malma was at odds with our predictions and could not explain fully why S. malma often exists allopatrically in high elevation reaches of streams dominated downstream by S. leucomaenis (e.g., Ishigaki 1987, Fausch et al. 1994). It may be that competitive reversals occur at temperatures lower than that tested here (6[degrees]C), which represented thermal data in summer (the primary growth season). The mean annual stream temperature in many Hokkaido streams, in fact, can be far lower than this temperature (Anonymous 1987). Since we detected species X temperature interactions in all of the aggression-, foraging-, and growth-related variables in sympatry, using temperatures lower than 6[degrees]C might reverse the competitive outcome between the two species.

Secondly, as implied by the fewer mortalities of smaller individuals of S. malma, that species may be better able to resist starvation than similarly-sized S. leucomaenis, and could become dominant during long periods of low food availability at cold temperatures. Fish inhabiting headwater reaches often face stochastic events such as floods and droughts, which may cause severe depletion in food resources (Schlosser 1982). This would be the case in Hokkaido where dynamic changes in food resources are observed in a headwater stream (Nakano et al. 1999a; S. Nakano, unpublished data). Starvation resistance is linked to lower metabolic demands at cold temperatures, preventing the depletion of lipid reserves during prolonged winter conditions (see Shuter and Post 1990). Grime (1973, 1977; see also Connell 1978, Huston 1979) predicted that species found in undisturbed, resource-rich habitats should have high competitive abilities relative to species found in more disturbed or stressful habitats. We argue that S. ma lma inhabiting higher elevation sites is the more stress-resistant species, showing overall competitive superiority under lean resource conditions, whereas S. leucomaenis is the species with high competitive abilities under the resource-rich conditions (see Nakano et al. 1999a).

During the past decades, well-designed manipulated field and laboratory experiments have clearly demonstrated the existence and direction of competition (Connell 1983, Schoener 1983, Hairston 1989). Although there is also substantial evidence for condition-specific competition along environmental gradients across time and space (see Chesson 1986), the mechanisms responsible for its expression have rarely been elucidated because behavioral and demographic parameters were not always examined simultaneously. For example, although many investigators have reported that for pairs of stream fishes, competitive superiority in aggressive interactions or food consumption vary under different test temperatures (Gibson 1983, Cunjak and Green 1986, Glova 1986, Reeves et al. 1987, De Staso and Rahel 1994, Taniguchi et al. 1998), none of these studies incorporated interspecific differences in survival into their competitive equation. Importantly; our results indicate that the direction and magnitude of competitive effects extrapolated from a single variable do not always accurately predict community pattern. Therefore, such evaluation may not be able to elucidate all mechanisms involved in structuring the community and species replacement patterns along an environmental gradient may not be solely explained by condition-specific competition if one examines the isolated effects of an abiotic factor on the behavior, resource use, or demography of the component species. This is largely due to the complex interactions of competition with species-specific physiological traits. Indeed, incorporating such interactions, per se, is important in mechanistic approaches to predicting community patterns.

ACKNOWLEDGMENTS

We are sincerely grateful to K. D. Fausch, S. Higashi, D. C. Novinger, S. Ohdachi, F. J. Rahel, and H. Urabe for critical comments on an early draft of the manuscript, and express our sincere thanks to T. Iwakuma, T. Kohyama, M. Saneyoshi, and T. Takami for invaluable advice during the study. K. Ishigaki, who laid the groundwork for this study, must be acknowledged for sharing many discussions during the study. Additional thanks also go to the staff members of TOEF, and undergraduate and graduate students who helped conduct the experiments. Two anonymous reviewers and the editor, W. M. Tonn, greatly improved our manuscript. This research was part of a dissertation prepared by the first author in partial fulfillment of a Ph. D. degree at Hokkaido University and was partly supported by the Japan Science Society (grant 8-212 to Y. Taniguchi) and Japan Ministry of Education, Science, Sport, and Culture (grants 07740592, 09740571, and 09NP1501 to S. Nakano).

(1.) Graduate School of Environmental Earth Science, Hokkaido University, Sapporo, Hokkaido 060-0808 Japan

(2.) Tomakomai Research Station, Hokkaido University Forests, Tomakomai, Hokkaido 053-0035 Japan

(3.) Present address: Department of Life Environmental Science, Yamaguchi Prefectural University, 3-2-1 Sakurabatake, Yamaguchi, Yamaguchi 753-8502 Japan. E-mail: ytani@ws1.yamaguchi-pu.ac.jp

(4.) Present address: Center for Ecological Research, Kyoto University, Otsu 520-2113 Japan.

LITERATURE CITED

Abacus. 1994. Survival tools for StatView. Abacus Concepts, Berkeley, California, USA.

Abrams, P. A. 1986. Competitive exclusion principle: other views. Trends in Ecology and Evolution 1:131-132.

Altmann, J. 1974. Observational study of behavior: sampling methods. Behavior 49:226-267.

Anonymous. 1987. Report on the natural condition of rivers in Japan (Hokkaido) (in Japanese) Maeda Ippo Foundation, Akan, Hokkaido, Japan.

Baltz, D. M., P. B. Moyle, and N. J. Knight. 1982. Competitive interactions between benthic stream fishes, riffle sculpin, Cottus gulosus, and speckled dace, Rhinichthys osculus. Canadian Journal of Fisheries and Aquatic Sciences 39:1502-1511.

Beamish, F. W. H. 1964. Respiration of fishes with special emphasis on standard oxygen consumption. Canadian Journal of Zoology 42:177-188.

Brown, J. H., and M. A. Bowers. 1984. Patterns and processes in three guilds of terrestrial vertebrates. Pages 282-296 in D. R. Strong, Jr., D. Simberloff, L. G. Abele, and A. B. Thistle, editors. Ecological communities: conceptual issues and the evidence. Princeton University Press, Princeton, New Jersey, USA.

Bull, C. M. 1991. Ecology of parapatric distributions. Annual Reviews in Ecology and Systematics 22:19-36.

Cavender, T. M. 1989. Cytotaxonomy and interrelationships of Pacific basin Salvelinus. Pages 49-68 in H. Kawanabe, F. Yamazaki, and D. L. G. Noakes, editors. Biology of charrs and masu salmon. Special Volume 1. Physiology and Ecology Japan, Kyoto, Japan.

Chapman, D. W. 1962. Aggressive behavior in juvenile coho salmon as a cause of emigration. Journal of Fisheries Research Board of Canada 19:1047-1080.

Chesson, P. L. 1986. Environmental variation and the coexistence of species. Pages 240-256 in J. Diamond and T. J. Case, editors. Community ecology. Harper and Row, New York, USA.

Christie, G. C., and H. A. Regier. 1988. Measures of optimal thermal habitat for four commercial fish species. Canadian Journal of Fisheries and Aquatic Sciences 45:301-314.

Connell, J. H. 1961. The influence of interspecific competition and other factors on the distribution of the barnacle, Chthamalus stellatus. Ecology 42:710-723.

Connell, J. H. 1978. Diversity in tropical rain forests and coral reefs. Science 199:1302-1310.

Connell, J. H. 1983. On the prevalence and relative importance of interspecific competition: evidence from field experiments. American Naturalist 122:661-696.

Cunjak, R. A., and J. M. Green. 1986. Influence of water temperature on behavioral interactions between juvenile brook charr, Salvelinus fontinalis and rainbow trout, Salmo gairdneri. Canadian Journal of Zoology 64:1288-1291.

De Staso, J., III, and F. J. Rahel. 1994. Influence of water temperature on competitive interactions between juvenile brook trout and Colorado River cutthroat trout in a laboratory stream. Transactions of the American Fisheries Society 123:289-297.

Diamond, J. M. 1970. Ecological consequences of island colonization by southwest Pacific birds. I. Types of niche shifts. Proceedings of the National Academy of Sciences (USA) 67:529-536.

Diamond, J. M. 1978. Niche shifts and rediscovery of interspecific competition. American Scientists 66:322-331.

Dunson, W. A., and J. Travis. 1991. The role of abiotic factors in community organization. American Naturalist 138:1067-1091.

Elliott, J. M. 1976. The energetics of feeding, metabolism and growth of brown trout (Salmo trutta L.) in relation to body weight, water temperature and ration size. Journal of Animal Ecology 45:923-948.

Elliott, J. M. 1994. Quantitative ecology and the brown trout. Oxford University Press, Oxford, UK.

Fausch, K. D. 1989. Do gradient and temperature affect distributions of, and interactions between, brook charr (Salvelinus fontinalis) and other resident salmonids in streams? Pages 303-322 in H. Kawanabe, F. Yamazaki, and D. L. G. Noakes, editors. Biology of charrs and masu salmon. Special Volume 1. Physiology and Ecology Japan, Kyoto, Japan.

Fausch, K. D., S. Nakano, and K. Ishigaki. 1994. Distributions of two congeneric charrs in streams of Hokkaido Island, Japan: considering multiple factors across scales. Oecologia 100:1-12.

Fausch, K. D., and R. J. White. 1986. Competitions among juveniles of coho salmon, brook trout, and brown trout in a laboratory stream, and implications for Great Lake tributaries. Transactions of the American Fisheries Society 115:363-381.

Fraser, F. J. 1969. Population density effects on survival and growth of juvenile coho salmon and steelhead trout in experimental stream-channels. Pages 253-265 in T. G. Northcote, editor. Symposium on salmon and trout in streams. Institute of Fisheries, University of British Columbia, Vancouver, Canada.

Gibson, R. J. 1983. The behavior of juvenile Atlantic salmon (Salmo salar) and brook trout (Salvelinus fontinalis), with regard to temperature and to water velocity. Transactions of the American Fisheries Society 107:703-712.

Glova, G. J. 1986. Interaction for food and space between experimental populations of juvenile coho salmon (Oncorhynchus kisutch) and coastal cutthroat trout (Salma clarki) in a laboratory stream. Hydrobiologia 131:155-168.

Grace, J. B., and R. G. Wetzel. 1981. Habitat partitioning and competitive displacement in cattails (Typha): experimental field studies. American Naturalist 118:463-474.

Grant, J. W. A., and D. L. Kramer. 1990. Territory size as a predictor of the upper limit to population density of juvenile salmonids in streams. Canadian Journal of Fisheries and Aquatic Sciences 47:1724-1737.

Grime, J. P. 1973. Competitive exclusion in herbaceous vegetation. Nature 242:344-347.

Grime, J. P. 1977. Evidence for the existence of three primary strategies in plants and its relevance to ecological and evolutionary theory. American Naturalist 111:1169-1194.

Hairston, N. G. 1989. The experimental test of an analysis of field distributions: competition in terrestrial salamanders. Ecology 61:817-826.

Hairston, N. G. 1989. Ecological experiments. Purpose, design, and execution. Cambridge University Press, Cambridge, UK.

Hartman, G. F. 1966. Some effects of temperature on the behavior of underyearling coho and steelhead. Management Report No. 51, Fish and Wildlife Branch, Department of Recreation and Conservation, Victoria, British Columbia, Canada.

Hughes, N. F. 1992. Ranking of feeding positions by drift-feeding arctic grayling (Thymallus arcticus) in dominance hierarchies. Canadian Journal of Fisheries and Aquatic Sciences 49:1994-1998.

Huntingford, F. A. 1993. Can cost-benefit analysis explain fish distribution patterns? Journal of Fish Biology 43 (Supplement A):289-308.

Huston, M. 1979. General hypothesis of species diversity. American Naturalist 113:81-101.

Hutchinson, G. E. 1961. The paradox of the plankton. American Naturalist 95:137-146.

Ishigaki, K. 1987. Studies on the biology in the early stages of two types of chars in Hokkaido. Research Bulletins of the College Experiment Forests 3:1121-1141.

Jaeger, R. G. 1970. Potential extinction through competition between two species of terrestrial salamanders. Evolution 24:632-642.

Jaeger, R. G. 1971. Competitive exclusion as a factor influencing the distributions of two species of terrestrial salamanders. Ecology 52:632-637.

Kitano, F, S. Nakano, K. Maekawa, and Y. Ono. 1995. Effect of stream temperatures on longitudinal distribution of fluvial Dolly Varden and potential habitat loss due to global warming (in Japanese with English abstract). Wildlife Conservation, Japan 1:1-11.

MacArthur, R. H. 1972. Geographical ecology. Patterns in the distribution of species. Harper and Row, New York, USA.

Magnuson, J. J. 1962. An analysis of aggressive behavior and competition for food and space in medaka (Otyzias latipes), Pisces, Cyprinodontidae. Canadian Journal of Zoology 40:313-363.

Moore, M. K., and V. R. Townsend, Jr. 1998. The interaction of temperature, dissolved oxygen and predation pressure in an aquatic predator-prey system. Oikos 81:329-336.

Moulton, M. P., and S. L. Pimm. 1986. The extent of competition in shaping an introduced avifauna. Pages 80-97 in J. Diamond and T. J. Case, editors. Community ecology. Harper & Row, New York, USA.

Nakano, S. 1995a. Individual differences in resource use, growth and emigration under the influence of a dominance hierarchy in fluvial red-spotted masu salmon in a natural habitat. Journal of Animal Ecology 64:75-84.

Nakano, S. 1995b. Competitive interactions for foraging microhabitats in a size-structured interspecific dominance hierarchy of two sympatric stream salmonids in a natural habitat. Canadian Journal of Zoology 73:1845-1854.

Nakano, S., K. D. Fausch, and S. Kitano. 1999a. Flexible niche partitioning via foraging mode shift: a proposed mechanism for coexistence in stream-dwelling charrs. Journal of Animal Ecology 68:1079-1092.

Nakano, S., and T. Furukawa-Tanaka. 1994. Intra- and interspecific dominance hierarchies and variation in foraging tactics of two species of stream-dwelling chars. Ecological Research 9:9-20.

Nakano, S., F Kitano, and K. Maekawa. 1996. Potential fragmentation and loss of thermal habitats for two charr species in the Japanese Archipelago due to climatic warming. Freshwater Biology 36:711-722.

Nakano, S., H. Miyasaka, and N. Kuhara. 1999b. Terrestrial-aquatic linkages: riparian arthropod inputs alter trophic cascades in a stream food web. Ecology 80:2435-2441.

Persson, L. 1986. Temperature-induced shift in foraging ability in two fish species, roach (Rutilus rutilus) and perch (Perca fluviatilis): implications for coexistence between poikilotherms. Journal of Animal Ecology 55:829-839.

Rahel, F J., and W. A. Hubert. 1991. Fish assemblages and habitat gradients in a Rocky Mountain-Great Plains stream: biotic zonation and additive patterns of community change. Transactions of the American Fisheries Society 120:319-332.

Reeves, G. H., F H. Everest, and J. D. Hall. 1987. Interactions between the redside shiner (Richardsonius balteatus) and the steelhead trout (Salmo gairdneri) in western Oregon: the influence of water temperature. Canadian Journal of Fisheries and Aquatic Sciences 44:1603-1613.

Ricker, W. E. 1979. Growth rates and models. Pages 677-743 in W. S. Hoar, D. J. Randall, and J. R. Brett, editors. Fish physiology. Volume VIII. Academic Press, London, UK.

Schlosser, I. J. 1982. Fish community structure and function along two habitat gradients in a headwater stream. Ecological Monographs 52:395-414.

Schluter, D. 1982. Distributions of Galapagos ground finches along an altitudinal gradient: the importance of food supply. Ecology 63:1504-1517.

Schoener, T. W. 1983. Field experiments on interspecific competition. American Naturalist 122:240-285.

Schoener, T. W. 1986. Mechanistic approaches to community ecology: a new reductionism? American Zoologist 26:81-106.

Sheldon, A. L. 1968. Species diversity and longitudinal succession in stream fishes. Ecology 49:193-198.

Shuter, B. J., and J. R. Post. 1990. Climate, population viability, and the zoogeography of temperate fishes. Transactions of the American Fisheries Society 119:314-336.

Simberloff, D. 1983. Competition theory, hypothesis-testing, and other community ecological buzzwords. American Naturalist 122:626-636.

Sutherland, W. J. 1996. From individual behavior to population ecology. Oxford University Press, Oxford, UK.

Takami, T., F Kitano, and S. Nakano. 1996. High water temperature influences on foraging responses and thermal deaths of Dolly Varden Salvelinus malma and white-spotted charr S. leucomaenis in a laboratory. Fisheries Science 63:6-8.

Taniguchi, Y. 1998. Condition-specific competition in stream fishes: its consequences on demography and community patterns. Dissertation, Hokkaido University, Sapporo, Japan.

Taniguchi, Y., D. C. Novinger, F J. Rahel, and K. Gerow. 1998. Temperature mediation of competitive interactions among three fish species, brook trout, brown trout, and creek chub, that replace each other along longitudinal stream gradients. Canadian Journal of Fisheries and Aquatic Sciences 55:1894-1901.

Terborgh, J. 1971. Distribution on environmental gradients: theory and preliminary interpretation of distributional patterns in the avifauna of the Codillera Vilcabamba, Peru. Ecology 52:23-40.

Terborgh, J., and J. S. Weske. 1975. The role of competition in the distribution of Andean birds. Ecology 56:562-576.

Tilman, D. 1987. Secondary succession and the pattern of plant dominance along experimental nitrogen gradients. Ecological Monographs 57:189-214.

Vincent, R. E., and W. J. Miller. 1969. Altitudinal distribution of brown trout and other fishes in a headwater tributary of the South Platte River, Colorado. Ecology 50:464-466.

Warner, S. C., J. Travis, and W. A. Dunson. 1993. Effect of pH variation on interspecific competition between two species of Hylid tadpoles. Ecology 74:183-194.

Wellborn, G. A., D. K. Skelly, and E. E. Werner. 1996. Mechanisms creating community structure across a freshwater habitat gradient. Annual Reviews in Ecology and Systematics 27:337-363.

Wiens, J. A. 1989. The ecology of bird communities, Volumes 1 and 2. Cambridge University Press, Cambridge, UK.

Wilbur, H. M. 1987. Regulation of structure in complex systems: experimental temporary pond communities, Ecology 8:1437-1452.

Wootton, R. J. 1990. Ecology of teleost fishes. Chapman & Hall, London, UK.
COPYRIGHT 2000 Ecological Society of America
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2000 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:TANIGUCHI, YOSHINORI; NAKANO, SHIGERU
Publication:Ecology
Geographic Code:9JAPA
Date:Jul 1, 2000
Words:9027
Previous Article:THE ROLE OF DISPERSAL AND DISTURBANCE IN DETERMINING SPATIAL HETEROGENEITY IN SEDENTARY ORGANISMS.
Next Article:MODELING AND ESTIMATION OF STAGE-SPECIFIC DAILY SURVIVAL PROBABILITIES OF NESTS.
Topics:

Terms of use | Privacy policy | Copyright © 2019 Farlex, Inc. | Feedback | For webmasters