Impact of water calcium on the phenotypic diversity of alpine populations of Gammarus fossarum.
Knowledge of the contribution of the environment to intraspecific differentiation between neighboring populations is fundamental to understanding the ecological significance of geographic variation (for recent references, see Niewiarowski and Roosenburg 1993). A common approach is to identify differences between populations and to examine possible correlations between observed biological variation and environmental variation. In this respect, a synthetic approach that includes both comparative and experimental techniques plays an important role in identifying ecological sources of biological variation and their evolutionary significance (Niewiarowski and Roosenburg 1993). However, owing to the diversity of ambient parameters and their multiple interrelations, it is often difficult to identify a predominant factor responsible for phenotypic variation (Berven and Gill 1983).
Among freshwater crustaceans, Gammarus fossarum (Amphipoda, Gammaridae) has been considered a good model for studying intraspecific variation as it lives in clusters with less populated areas between (Siegismund 1988). Morphological and morphometrical characters were first used to differentiate neighboring populations of Gammarus fossarum (Dusaugey 1955, Roux 1970), as reported for other gammarid species (Bulnheim and Scholl 1981, Scheepmaker 1987, Kane et al. 1992). However, it soon became apparent that they were poor distinctive criteria because of their considerable amount of variability, not only between different populations, but also within the same population, going so far as to make the characterization of the species difficult (Pinkster 1983, Sheepmaker and Van Dalfsen 1989, Sheepmaker 1990). Moreover, identifying environmental sources of such morphological variation remained unsuccessful (Roux 1970). The diversity among geographic populations has been then considered at the ecophysiological level, leading to the concept of physiological races (Roux 1967). Of the ecophysiological traits investigated among different populations (Roux 1971), the calcium (Ca) requirement during molting is of key interest (Ormerod et al. 1988). Ca metabolism is a central parameter in the biology of gammarids (Graf 1974). As in other crustaceans, there is an important periodic Ca turnover through the cuticle during molt cycles throughout life (reviewed in Graf 1978, Greenaway 1985). Moreover, since Gammarus fossarum cannot store Ca before exuviation, as is the case for most gammarids (Vincent 1963, Graf 1965, Wright 1979, 1980), the Ca turnover during the molt cycle may depend upon the Ca availability in the water. Previous ecophysiological investigations at this level led to conflicting interpretations and a precise correlation between observed ecophysiological differences and clearly delimited environmental parameters was generally unclear (Roux 1971, Vincent 1971).
TABLE 1. Comparative analysis of ionic concentrations (Ca, Mg, Na, K, HC[O.sub.3], Cl, S[O.sub.4], and N[O.sub.3] [mg/L]) and pH in water from the different sites [ILLUSTRATION FOR FIGURE 1 OMITTED]. Ionic Site composition and pH B3 B4 C2 V3 V1 C1 Ca 12.45 14.85 75.80 78.5 83.50 95.70 Mg 2.20 1.50 3.60 4.30 1.10 3.70 Na 1.40 0.83 2.10 3.00 4.90 3.50 K 0.15 0.13 0.50 0.80 0.80 0.90 HC[O.sub.3] 36.60 41.50 221.00 206.00 261.00 293.00 Cl 0.60 0.30 2.70 1.40 8.20 4.40 S[O.sub.4] 8.00 7.20 12.00 11.00 6.00 9.80 N[O.sub.3] 1.20 1.50 0 0.80 4.20 2.70 pH 7.76 7.73 8.23 8.24 8.25 8.17
Recent advances in our knowledge of the molt cycle in gammarids (Graf 1986) and of the physiology of Ca turnover during molting (Graf 1979, Wright 1979, 1980, Meyran and Graf 1993) allow reconsideration of the impact of environmental Ca on the diversification between populations, at both comparative and experimental levels. In this respect, the Grenoble region (France) may offer a useful study site. First, in the upper reaches of the alpine streams near Grenoble, Gammarus fossarum has a high density and a patchy distribution (Dusaugey 1955, Roux 1963), as is generally observed in other countries (Roux 1982, Schrimpff and Foeckler 1985, Foeckler and Schrimpff 1985, Siegismund and Muller 1991). Also, there is an interesting geological contrast between the neighboring mountain ranges of the Belledonne (crystalline rock) and the Chartreuse-Vercors (limestone regions) [ILLUSTRATION FOR FIGURE 1 OMITTED] that leads to characteristic local differences in the Ca concentration of water (see Table 1).
Preliminary investigations have suggested that the diversity of the alpine populations of Gammarus fossarum may be related to the Ca concentration of the water (Meyran 1994). This hypothesis was further analyzed in this paper at both descriptive and experimental levels to examine the influence of environmental Ca on the geographic variability of Gammarus fossarum.
MATERIALS AND METHODS
Gammarus fossarum Koch is one of the four species of gammarids of the pulex group (Karaman and Pinkster 1977a, b, 1987), recognizable by the setation of the second antennae and the morphology of uropods and pleopods (Roux 1963, Goedsmakers 1981b). Nevertheless, morphological and morphometrical characteristics have great variability among different populations and also among individuals of the same population (Dusaugey 1955, Roux 1970). In the Grenoble region, Gammarus fossarum is the predominant species observed in the mountain torrents above an altitude of 800 m (Dusaugey 1955). It can reach high population densities, which makes collection of large samples possible.
To avoid the variability of multiple ambient parameters leading to difficult interpretations, study sites were selected for their comparable environmental characteristics, except for the water Ca concentration. All were located at an altitude of 900 m and within a 30-km radius of Grenoble. They had similar climatic conditions, to avoid temperature differences, low contrasted seasonal hydric flux, and comparable hydrodynamic characteristics, to allow field rearing. They also included numerous specimens in the populations for sufficient and uniform sampling.
For comparison between populations from the different geographic areas, two sites were selected in each area [ILLUSTRATION FOR FIGURE 1 OMITTED]. Sites from the crystalline Belledonne Massif were located on the Vorz torrent at "La Gorge" (B3), and on the Laval torrent at "La Boutiere" (B4). Sites in the limestone Chartreuse Massif were located on the Vence torrent at "Le Sappey" (C1), and on the Couzon torrent at "Saint-Pierre" (C2). Sites in the limestone Vercors Massif were located on the Furon torrent above "Engin" (V1), and on the Bruyant torrent at "Chateau-Bernard" (V3). Animals were collected either by turning stones and sweeping over the sediment with a fine-meshed hand net, or by shaking aquatic plants or decaying leaves in a bucket. Only adult males of maximum size were used.
At the beginning of each experimental period, when the animals were collected, water samples were taken using the technique of Foeckler and Schrimpff (1985), for ionic titration through atomic absorption spectrophotometry (Pinta 1971), after HN[O.sub.3] 1% addition. The water chemistry analysis (ionic titration and pH determination) was performed using standard methods (Franson 1992).
Specimen determination and dating
Gammarus samples were identified individually using morphological keys (Roux 1970, Kamaran and Pinkster 1977a, b, 1987). Then, they were subjected to a fine determination of their molt cycle stage using the exact staging method of Graf (1986). This was based upon the microscopic observation of the integumental morphogenesis of the dactylopodite and the propodite from a freshly cut pereiopod. This rapid method allows in vivo determination of the precise phase of the molt cycle for any specimen, without affecting the survival of the animal. It was thus possible to follow a given specimen during the whole sequence of the different phases of the molt cycle and to determine their specific duration.
According to Graf (1986), the molt cycle is divided into five periods from one exuviation (ecdysis = period E) to the next. During the postmolt (postecdysis = period A: early postmolt + period B: late postmolt), the new cuticle is progressively calcified. The cuticle is stable during the whole intermolt (period c), then is gradually decalcified during the premolt (preecdysis = period D).
Rearing and translocation experiments
For a comparative study of the duration of the molt cycle among geographic populations, dated specimens were reared individually, for at least 2 mo, in 6 x 4 x 4 cm transparent plastic perforated boxes containing aquatic plants or decaying leaves for food, and immersed in water. Animals were examined for molt cycle stage every 10 d during the period of experimentation.
In the field, the rearing boxes of control specimens were immersed in water from the collecting site (native water). Boxes of translocated specimens were transplanted to another site (alien water). To simplify the experimental design, not all combinations were performed (see Table 4). The average temperature of the water (11 [degrees] C) was determinated by periodic measurements in situ in all study sites during the period of experimentation.
At the laboratory, animals were reared in a climate-controlled room (constant temperature: 11 [degrees] C), with a daily cycle of 12 h of light. Control specimens were placed into water from their native site, whereas translocated specimens were reared in water from an alien site. The experimental design involved the same combinations as in the field experimentation (see Table 4). Water was changed weekly.
To further investigate the impact of the Ca concentration of the water on the duration of the molt cycle, some specimens from B3 and B4 populations were reared in their native water artificially enriched in Ca up to a concentration equivalent to that of C1 site by spectrophotometrically controlled addition of CaS[O.sub.4].
Both field and laboratory experiments were systematically performed on 600 animals divided into uniform samples each including 8 specimens of the same size, from all collecting sites. Experimentation occurred repeatedly and simultaneously in the field and in the laboratory from summer to autumn 1993 and from spring to autumn 1994.
Calcium balance analysis
At the end of each experiment, the total body Ca content of the dated specimens after 24 h of starvation was determined through atomic absorption spectro-photometry. [TABULAR DATA FOR TABLE 2 OMITTED] After measurement of their dry mass (Graf 1969), whole specimens were homogenized and then mineralized with [H.sub.2]S[O.sub.4] 6N (Meyran et al. 1987). The Ca concentration in homogenates was measured with a Varian apparatus at 422.7 nm after dilution in 0.2% lanthanum chloride. In order to avoid individual variation related to the animal's size, results were standardized to a constant dry mass of 10 mg, following Graf (1964). In the same way, the Ca balance during the successive phases of the molt cycle was determined on 400 field-collected specimens at all stages of the molt cycle, directly originating from all geographic populations.
Data were analyzed using the SAS statistical package (SAS 1988).
Environmental parameters: differences in water Ca concentrations
Successive measurements indicated no obvious difference in ionic concentration among water samples of the same study site from one period of experimentation to another. Data from Table 1, summarizing the first analysis, revealed that the most prominent differences among water samples from the different areas were Ca concentrations, which may be related to the geological differences of the leached substrata. High Ca titers were characteristic of waters from the limestone areas of Chartreuse and Vercors (from 75.8 mg/L in C2 to 95.7 mg/L in C1), whereas waters with low Ca concentration were localized within the crystalline area of the Belledonne (from 12.4 mg/L in B3 to 14.8 mg/L in B4). Differences in [Mathematical Expression Omitted] concentrations were also noticed, which may be responsible for the differences in pH between waters from the two different geological areas.
TABLE 3. Analysis of variance for the dry mass of animals and the mean duration of the molt cycle in reared control populations (NS = nonsignificant; * P [less than] 0.05; ** P [less than] 0.01; *** P [less than] 0.001) (localization of experiments: field or laboratory). Source df MS F Animal's dry mass Site of origin (SO) 6 63.73 23.98(***) Error 256 2.65 Mean duration of the molt cycle Site of origin (SO) 6 55.74 11.25(***) Localization of experiments (LOC) 1 13.7 12.77 NS SO x LOC 4 5.1 81.05 NS Error 109 4.95
1. Differences between the populations in situ. - Measurement of the dry mass of samples of the different populations at several times during the study period revealed that specimens from the limestone areas were on average heavier than those from the crystalline sites, regardless of the timing of sampling (Table 2). In the analysis of variance (Table 3, upper part), the effect "site of origin" (SO) is significant.
The comparative analysis of the duration of the molt cycle showed characteristic differences between populations (Tables 2-3). Whereas the duration of the molt cycle was not significantly different among populations from a given geologic area, it was [approximately equal to]8.5% longer in the populations from the limestone areas (C1, C2, V1, and V3) than in those from the crystalline sites (B3 and B4) (Table 2). In the analysis of variance (Table 3, lower part), the SO effect is significant. In fact, the duration of the molt cycle in smaller specimens from limestone populations was shorter than that observed for animals of the same size living in crystalline areas (data not shown). No significant differences were observed [TABULAR DATA FOR TABLE 4 OMITTED] between field and laboratory measurements on all control specimens (LOC effect: Table 3, lower part). In the same way, the mortality rate was between 3 and 12.7% in both natural and laboratory conditions, respectively.
Spectrophotometric analysis of the Ca balance in animals during the successive stages of the molt cycle revealed characteristic differences among populations according to habitat. These differences were observed in samples originating from the field and in animals exposed to experimentation [ILLUSTRATION FOR FIGURE 2 OMITTED]. Whereas the total body Ca content was comparable in any specimen at exuviation (period E, when the old cuticle is lost) and during intermolt (period C, when the calcified cuticle is stable), the Ca loss from the old cuticle to the water during premolt (period D) was more rapid in animals from limestone sites than in those from crystalline areas. In the same way, the Ca recovery from the water into the new exoskeleton during postmolt (period B) was faster in specimens from limestone areas than in the others.
2. Effects of translocation on the duration of the molt cycle and the Ca balance. - Results of the translocations performed are summarized in Table 4. In all cases, for a given experiment, laboratory results did not differ significantly from field results (P [greater than] 0.05). No significant differences (P [greater than] 0.05) were observed between controls and specimens with comparable Ca concentration from neighboring sites (from C1 to V1; from C2 to C1; from V1 to C1; from V3 to C1: Table 4).
Both field and laboratory data indicated a significant increase in the duration of the molt cycle in specimens translocated to water with a lower Ca concentration. This ranges from 30.6% in the translocation from C1 to B3 up to 53.1% in the translocation from C2 to B4. The in vivo chronological study of the phases of the molt cycle in those specimens revealed a typical increase in the duration of the early intermolt. In some extreme cases, from 20 to 80% of the population remained in intermolt before dying. Conversely, the duration of the molt cycle was significantly shortened in animals translocated to water with a higher Ca concentration (from 8.6% in the translocation from B4 to C1 up to 10.8% in the translocation from C2 to B3). In all cases, both effects SO (site of origin) and ST (site of translocation) are significant (Table 5).
TABLE 5. Analysis of variance for the mean duration of the molt cycle in animals translocated in natural water, performed on both field and laboratory data (NS = nonsignificant; * P [less than] 0.05; ** P [less than] 0.01; *** P [less than] 0.001). Source df MS F Mean duration of the molt cycle Site of origin (SO) 6 2 388.86 116.40(***) Site of translocation (ST) 6 2 502.97 121.97(***) SO x ST 18 390.69 19.04(***) Error 370 20.52
A decrease in the duration of the molt cycle similar to that previously observed was seen in specimens from B3 and B4 reared in their native water artificially enriched with Ca up to a concentration equivalent to that of C1 (from B3 to B3-equivalent C1; from B4 to B4-equivalent C1: Table 6). In the analysis of variance (Table 7), the effect "water Ca enrichment" (DOP) is significant; the SO effect is due to a significant difference between C1 and the other sites of experimentation. Spectrophotometric measurements comparing translocated and control specimens indicated no differences in the Ca balance for each given stage of the molt cycle [ILLUSTRATION FOR FIGURE 2 OMITTED].
The present study clearly relates phenotypic differences among alpine populations of Gammarus fossarum to the levels of environmental sources of Ca. The populations from limestone areas of Chartreuse and Vercors include specimens with a larger maximum size and a longer molt cycle than those from the crystalline site of Belledonne. The Ca balance of animals from limestone areas was different during both pre- and postmolt from that of specimens from crystalline sites. Experimental increasing of Ca in the environment can reduce the duration of the molt cycle and vice versa.
However, our experimental results appear to contradict the descriptive data. The comparison is biased at the descriptive level as the measurements were made only on adults of maximum size, which were largest in limestone populations. As previously stated, the duration of the molt cycle in smaller specimens from limestone populations was shorter than that observed for animals of the same size living in crystalline areas. Smaller animals have not been retained in our study as they are not representative for a rigorous comparison among different populations.
TABLE 6. Effects of translocation of B3 and B4 animals in their native water, artificially Ca-enriched up to a concentration equivalent to that of C1 (equivalent C1) on the mean duration (mean no. days, with standard errors in parentheses) of the molt cycle in laboratory experiments (N = number of observations) vs. the mean duration of the molt cycle in B3, B4, and C1 control populations. Average duration of the molt cycle Specimens trans- Originating Control specimens located in site in native water Ca-enriched water B3 39.08 (1.84) 34.92 (1.93) N = 23 N = 9 B4 38.78 (1.47) 34.14 (1.41) N = 25 N = 11 C1 42.10 (1.02) N = 5
The fact that populations of Gammarus fossarum may adapt to a large range of water Ca concentrations is not surprising since this has been observed in most gammarids (review in Roux 1971, Vincent 1971). Water Ca levels are not considered to be a limiting factor for the distribution of those populations (Okland and Okland 1985). More generally, it is assumed that populations of native species are not limited by minerals (Fielder 1986). However, as variations in water Ca levels may generate morphological and physiological differences between animals, this environmental parameter may be significant at least in the phenotypic differentiation of the populations.
Microgeographic variation in growth rates and maximum sizes has commonly been observed among gammarid populations (Goedmakers 1981a, b). The differences in gammarid size from site to site were usually correlated to size selective predation or to differences in maturity from one population to another (Goedmakers 1981a, b). To avoid this biasing factor in our comparative study, sampling was limited to mature males of maximum body size. Thus, the variation in maximum body size arising from our measurements may reflect a genuine difference between populations from limestone and crystalline areas. This is supported by the congruence of the results obtained from all experiments. Possible artifacts caused by the sampling may be avoided by the random periodic sampling procedure made at all study sites during the whole period of experimentation. Interestingly, such a correlation between the Ca content of the water and the maximum growth of Gammarus fossarum has also been reported for other gammarid species (Vincent 1971). Additional observations reveal that the differences in maximum body size observed among males are also true for the females (data not shown). As a positive correlation has been found between fecundity and body size by Goedmakers (1981a), the impact of such differences may be important at the level of population dynamics. Moreover, our field observations periodically performed during the entire experiment showed that the populations with the maximum density were located in waters that flow over limestone (J. C. Meyran, personal observations). Therefore, the populations living on limestone not only reach a maximum size but also have a maximum growth rate. However, other factors such as migration and predation rates may also be affecting the differences in density between low and high Ca sites. Since growth rates were correlated with the duration of the molt cycles in gammarids, as in other crustaceans (see references in Graf 1969), it is not surprising that our descriptive data indicated that the molt cycle was longer in control specimens from the limestone sites.
TABLE 7. Analysis of variance for the mean duration of the molt cycle in B3 and B4 animals translocated in their native water, artificially Ca-enriched to a concentration equivalent to that of C1 (equivalent C 1) (NS = nonsignificant; * P [less than] 0.05; ** P [less than] 0.01; *** P [less than] 0.001). Source df MS F Mean duration of the molt cycle Site of origin (SO) 2 49.73 9.32(***) Water Ca enrichment (DOP) 2 270.93 101.58(***) SO x DOP 1 0.78 0.30 NS Error 68 2.66
Previous data have only indicated inter- or intraspecific differences in water Ca tolerance among gammarids, with conflicting interpretations reviewed by Roux (1971). More recently, Roux (1971) concluded an indirect influence of water Ca levels on the geographical and ecological distribution of Gammarus fossarum. However, the existence of "physiological races" was not excluded. Comparable findings were reported by Vincent (1963, 1971, 1972) on other gammarid species, using experimental methods focused on the intermolt cuticular calcification and the postmolt Ca recovery.
The current investigations confirm the existence of significant ecophysiological variations in Ca metabolism during the molt cycle among different populations of Gammarus fossarum. These differences have been shown through a synthetic approach using both the measurement of the duration of the molt cycle by the staging method of Graf (1986) and the spectrophotometric determination of the Ca balance during the molt cycle. Moreover, it is important to compare results obtained under laboratory conditions with those obtained from animals in the field, since the situation in natural biotopes is usually characterized by various fluctuating factors (Koch-Kallnbach and Meijering 1977). Thus, our experimentation was conducted jointly in the field and in the laboratory, leading to congruent results.
Our measurements clearly show that, during molting, the environmental Ca availability appears to interfere directly with the dynamics of the Ca shifts between the animal and its surrounding medium. In this respect, the low Ca levels in the water may be responsible for increasing the duration of the premolt period in specimens from the Belledonne, contrary to that observed in animals from waters with a high Ca concentration from the Chartreuse and Vercors. However, the relative duration of the whole molt cycle may not be affected if we consider that the duration of the intermolt period is different in both types of populations, because of the different size of the specimens studied. This is supported by the results of our translocation experiments, which indicated changes in the duration of the intermolt period. Comparison of the duration of the premolt period from one population to another suggests that animals from limestone sites may be considered as favored compared to those from crystalline areas as they reach their cuticular Ca equilibrium more rapidly, which is of key interest for survival. Interestingly, specimens from crystalline sites show a higher mortality rate after exuviation than those living on limestone (data not shown).
In the same way, it appears from translocation experiments that increasing Ca in water can reduce the duration of the molt cycle, probably through an improvement of Ca recovery into the animal's new cuticle during postmolt (Vincent 1971). As an increase in the number of molt cycles for a given time may directly interfere with the growth rate, this environmental factor may be at least partly responsible for the differences in adult maximum size observed between populations from limestone and crystalline areas. Conversely, the poor Ca availability in water may be related to the smaller maximum body size of specimens from crystalline areas.
A clear demonstration of the genuine impact of the environmental Ca in the ecophysiological differentiation among geographic populations appears when animals from B3 and B4 are reared in their native water after artificial addition of Ca to a concentration equivalent to that of C1. The results indicate similar effects on the duration of the molt cycle for a given Ca concentration whether the specimens are translocated into alien or native water.
The ecopbysiological plasticity of Ca metabolism currently observed among alpine populations of Gammarus fossarum may be a determining factor in the dynamics of colonization of this species. Migration constitutes the central parameter of colonization processes in gammarids (review in Goedmakers and Pinkster 1981) but its relation with the environmental Ca availability is still controversial. Previous authors have considered Gammarus fossarum to be a preglacial inhabitant of the Alps (review in Scheepmaker 1990), largely represented in this area (Dusaugey 1955, Siegismund and Muller 1991), and rather independent from the Ca content of the water (Roux 1971). Such an independence may be limited, as suggested by our ecophysiological results and our field observations, which showed that the most flourishing populations were located in high Ca concentration areas. Moreover, current experimental data reveal that mortality rates increased in animals translocated to waters lower in Ca concentration. This relatively greater difficulty in adapting to lower environmental Ca levels, earlier observed in other gammarid species by Vincent (1971), allows us to suggest that the primitive populations of Gammarus fossarum probably lived in limestone areas and that colonization of the crystalline sites may have been secondary. This hypothesis agrees with the migration theory earlier proposed by Pacaud (1945a, b), which was based only upon biogeographic considerations.
The phenotypic variation among geographic populations of Gammarus fossarum observed in this study may not only reflect environmental diversity, as presently suggested, but may also have a possible genetic basis. In this respect, the concept of microgeographic races was proposed (Goedmakers 1980a, b, 1981a, b, Goedmakers and Pinkster 1981). Using enzymatic polymorphism analysis, Siegismund and Muller (1991) found that local populations of Gammarus fossarum show strong genetic differentiation compared to other gammarid populations. Such genetic differentiation is now being further considered at the molecular level through the comparison of mitochondrial DNA (Meyran and Taberlet 1996), as this genetic marker appears to evolve rapidly at the sequence level in crustaceans (Palumbi and Benzi 1991).
Finally, the present study clearly attests to the value of a synthetic approach including both comparative and experimental techniques in identifying environmental parameters responsible for biological variation among neighboring populations. Moreover, investigation of Ca metabolism during the molt cycle may provide a new powerful ecophysiological tool to supplement the criteria generally used to test the phenotypic variability among geographic populations of gammarids.
The author thanks A. Zganic and the Laboratoire Regional d'Analyse des Eaux, Universite J. Fourier, Grenoble, for the water chemistry analysis and the atomic absorption spectrophotometric dosage of calcium; Dr. I. Till-Bottraud for help in statistics, Dr. B. Serra-Tosio for helpful advice in the choice of the study sites, and L. Waits for help with the English. This work was funded by the Centre National de la Recherche Scientifique and the University J. Fourier (Grenoble, France).
Berven, K. A., and D. E. Gill. 1983. Interpreting geographic variation in life history traits. American Zoologist 23:85-97.
Bulnheim, H. P., and A. Scholl. 1981. Genetic variation between geographic populations of the Amphipod Gammarus zaddachi and G. salinus. Marine Biology 64:105-115.
Dusaugey, J. 1955. Etude morphologique et repartition des Gammares en Dauphine. Travaux du Laboratoire d'Hydro-biologie et de Pisciculture, Universite de Grenoble 42:9-18.
Fielder, P. C. 1986. Implications of selenium levels in Washington mountain goats, mule, deer, and Rocky Mountain elk. Northwest Science 60:15-20.
Foeckler, F., and E. Schrimpff. 1985. Gammarids in streams of Northeastern Bavaria, F.R.G. II. The different hydro-chemical habitats of Gammarus fossarum Koch, 1835 and Gammarus roeseli Gervais, 1835. Archiv fur Hydrobiologia 104:269-286.
Franson, M. A. 1992. Standard methods for the examination of water and wastewater. American Public Health Association, Washington, D.C., USA.
Goedmakers, A. 1980a. Population dynamics of three gammarid species (Crustacea, Amphipoda) in French chalk stream. Part I. General aspects and environmental factors. Bijdragen tot de Dierkunde 50:1-34.
-----. 1980b. Microgeographic races of Gammarus fossarum Koch, 1836. Crustaceana, Supplement 6:216-224.
-----. 1981a. Population dynamics of three Gammarid species (Crustacea, Amphipoda) in French chalk stream. Part II. Standing crop. Bijdragen tot de Dierkunde 51:31-69.
-----. 1981b. Population dynamics of three Gammarid species (Crustacea, Amphipoda) in French chalk stream. Part IV. Review and implications. Bijdragen tot de Dierkunde 51:181-190.
Goedmakers, A., and S. Pinkster. 1981. Population dynamics of three gammarid species (Crustacea, Amphipoda) in French chalk stream. Part III. Migration. Bijdragen tot de Dierkunde 51:145-181.
Graf, F. 1964. Etude de la variation du calcium total au cours du cycle d'intermue chez les Crustaces Amphipodes Gammarus pulex pulex L. et Orchestia gammarella Pallas. Comptes Rendus de l' Academie des Sciences Paris, serie III 259:2703-2705.
-----. 1965. Etude comparative de la variation du calcium total au cours du cycle d'intermue chez les Crustaces Amphipodes Niphargus virei Chevreux, Gammarus pulex pulex L. et Orchestia gammarella Pallas. Comptes Rendus de l'Academie des Sciences Paris, serie III 261: 819-821.
-----. 1969. Le stockage de calcium avant la mue chez les Crustaces Amphipodes Orchestia (Talitride) et Niphargus (Gammaride hypoge). These, Faculte des Sciences, Universite de Dijon 105:1-216.
-----. 1974. Quelques aspects du metabolisme du calcium chez les crustaces. Pages 13-22 in D. Pansu, editor. Physiologie comparee des echanges calciques. Simep, Villeurbanne, France.
-----. 1978. Les sources de calcium pour les crustaces venant de muer. Archives de Zoologie experimentale et Generale 119:143-161.
-----. 1986. Fine determination of the molt cycle stages in Orchestia cavimana Heller (Crustacea: Amphipoda). Journal of Crustacean Biology 4:666-678.
Greenaway, P. 1985. Calcium balance and moulting in the Crustacea. Biological Review 60:425-454.
Kane, T C., D.C. Culver, and R. T Jones. 1992. Genetic structure of morphologically differentiated populations of the amphipod Gammarus minus. Evolution 46:272-278.
Karaman, G. S., and S. Pinkster. 1977a. Freshwater Gammarus species from Europe, North Africa and adjacent regions of Asia (Crustacea-Amphipoda). Part I. Gammarus pulex-group and related species. Bijdragen tot de Dierkunde 47:1-97.
Karaman, G. S., and S. Pinkster. 1977b. Freshwater Gammarus species from Europe, North Africa and adjacent regions of Asia (Crustacea-Amphipoda). Part II. Gammarus roeseli-group and related species. Bijdragen tot de Dierkunde 47:165
Karaman, G. S., and S. Pinkster. 1987. Freshwater Gammarus species from Europe, North Africa and adjacent regions of Asia (Crustacea-Amphipoda). Part III. Gammarus balcanicus-group and related species. Bijdragen tot de Dierkunde 57:207-260.
Koch-Kallnbach, M. E., and M.P. D. Meijering. 1977. Duration of instars and precopulae in Gammarus pulex (Linnaeus, 1758) and Gammarus roeseli (Gervais, 1835) under semi-natural conditions. Crustaceana 4:119-127.
Meyran, J. C. 1994. Adaptative diversity of calcium metabolism in Gammarus fossarum populations. Comptes Rendus de l'Academie des Sciences Paris, Sciences de la Vie 317: 1043-1048.
Meyran, J. C., J. Fournie, and F. Graf. 1987. Carbonic anhydrase activity in a calcium mobilizing epithelium of the crustacean Orchestia cavimana during molting. Histochemistry 87:419-429.
Meyran, J. C., and F. Graf. 1993. Calcium turnover through a mineralizing-demineralizing epithelium in the terrestrial crustacean Orchestia during molting. Trends in Comparative Biochemistry and Physiology 1:121-137.
Meyran, J. C., and P. Taberlet. 1996. Mitochondrial DNA polymorphism among closely located populations of Gammarus fossarum (Crustacea, Amphipoda). Molecular Ecology, in press.
Niewiarowski, P. H., and W. Roosenburg. 1993. Reciprocal transplant reveals sources of variation in growth rates of the lizard Sceloporus undulatus. Ecology 74:1992-2002.
Okland, K. A., and J. Okland. 1985. Factor interaction influencing the distribution of the freshwater "shrimp" Gammarus. Oecologia 66:364-367.
Ormerod, S. J., K. R. Bull, C. P. Cummins, S. J. Tyler, and J. A. Vickery. 1988. Egg mass and shell thickness in dippers Cinclus cinclus in relation to stream acidity in Wales and Scotland. Environmental Pollution 55:107-121.
Pacaud, A. 1945a. Les Amphipodes de la faune nutritive des eaux douces francaises. Bulletin Francais de Pisciculture 136:105-120.
-----. 1945b. Donnees morphologiques et ecologiques sur les varietes de Gammarus (Rivulogammarus) pulex L. en France metropolitaine. Bulletin de la Societe Zoologique de France 70:57-67.
Palumbi, S. R., and J. Benzi. 1991. Large mitochondrial DNA differences between morphologically similar penaeid shrimps. Molecular Marine Biology and Biotechnology 1: 27-34.
Pinkster, S. 1983. The value of morphological characters in taxonomy of Gammarus. Beaufortia 33:15-28.
Pinta, M. 1971. Spectrometrie d'absorption atomique. Masson, Paris, France.
Roux, A. L. 1963. Donnees morphologiques et biologiques sur des gammares du groupe pulex recoltes dans le Massif de la Grande Chartreuse et le Bas Dauphine. Crustaceana 6:89-100.
-----. 1967. Les Gammares du groupe pulex (Crustaces, Amphipodes). Essai de systematique biologique. These, Faculte des Sciences, Universite de Lyon 8201: 1-159.
-----. 1970. Les Gammares du groupe pulex. Essai de systematique biologique. I Etude morphologique et morphometrique. Archives de Zoologie Experimentale et Generale 111:313-356.
-----. 1971. Les Gammares du groupe pulex. Essai de systematique biologique. II Quelques caracteristiques ecologiques et physiologiques. Archives de Zoologie Experimentale et Generale 112:471-503.
Roux, C. 1982. Les variations du metabolisme respiratoire et de l'activite de quelques invertebres dulcaquicoles sous l'influence de divers facteurs ecologiques. These, Faculte des Sciences, Universite de Lyon 8201:1-159.
SAS. 1988. SAS/STAT user's guide. Release 6.30 edition. SAS Institute, Cary, North Carolina, USA.
Scheepmaker, M. 1987. Morphological and genetic differentiation of Gammarus stupendus Pinkster, 1983 in the Massif de la Sainte Baume, France. Bijdragen tot de Dierkunde 57:1-18.
-----. 1990. Genetic differentiation and estimated level of gene flow in members of the Gammarus pulex-group (Crustacea, Amphipoda) in Western Europe. Bijdragen tot de Dierkunde 60:3-30.
Scheepmaker, M., and J. Van Dalfsen. 1989. Genetic differentiation in Gammarus fossarum Koch, 1835 and G. carpati Petre-Stroobant, 1980 (Crustacea, Amphipoda) with reference to G. pulex pulex in North-West Europe. Bijdragen tot de Dierkunde 59:127-139.
Schrimpff, E., and F. Foeckler. 1985. Gammarids in streams of Northeastern Bavaria, F.R.G.I. Prediction of their general occurrence by selected hydrochemical variables. Archiv fur Hydrobiologia 103:479-495.
Siegismund, H. R. 1988. Genetic differentiation in populations of the freshwater amphipods Gammarus roeseli and Gammarus fossarum. Hereditas 109:269-276.
Siegismund, H. R., and J. Muller. 1991. Genetic structure of Gammarus fossarum populations. Heredity 66:419-436.
Vincent, M. 1963. Le calcium total chez Gammarus pulex et la teneur en calcium de l'eau. Comptes Rendus de la Societe Biologique de France 157:1274-1277.
-----. 1971. Ecologie et ecophysiologie des Gammarides epiges du Centre-Ouest. These, Faculte des Sciences, Universite de Limoges 712:1-141.
-----. 1972. Preferendum ionique chez des amphipodes epiges du Centre-Ouest. Vie et Milieu 1:65-79.
Wright, D. A. 1979. Calcium regulation in intermoult Gammarus pulex. Journal of Experimental Biology 83:131-144.
-----. 1980. Calcium balance in premoult and post-moult Gammarus pulex (Amphipoda). Freshwater Biology 10: 571-579.
|Printer friendly Cite/link Email Feedback|
|Date:||Jul 1, 1997|
|Previous Article:||Canopy closure rate and forest structure.|
|Next Article:||Variations of wood delta 13C and water-use efficiency of Abies alba during the last century.|