Reproductive potential and lifetime potential fecundity of the freshwater amphipods Gammarus fossarunt and G. roeseli in Austrian streams and rivers

• 1 Fecundity of Gammarus fossarum populations at six stream sites (S1-S6) and G. roeseli populations at two sites (S5 and S6) was studied monthly during 1985–1988, distinguishing seven morphological stages of embryonic development. Mean survival of G. fossarum eggs was 65% (60–70%) in nature and 60% (57–63%) in the laboratory; egg survival of G. roeseli was 52% (46–58%) in nature and 41% (35–47%) in the laboratory. Regressions of egg numbers per clutch against body wet weight (WWT) decreased significantly for egg developmental stages 2 to 7 (hatching).
• 2 In stream populations, ovigerous females of G. fossarum were absent in October (six sites) and November (five sites); ovigerous G. roeseli were usually absent from October until March.
• 3 Overall mean egg volume (EV) increased significantly from 0,08 ± 0.001 mm³ and 0.075 ± 0.001 mm³ for stage 2 eggs, to 0.174 ± 0.012mm³ and 0.160 ± 0.013 mm³ for stage 6 eggs of G. fossarum and G. roeseli, respectively. The dry weights of stage 2 eggs, stage 6 eggs and neonates (stage 7) were not significantly different, within and between species; mean dry weight was 36 ± 3 μg.
• 4 Egg volumes and fecundity indices (FI, weight-specific number of eggs per clutch) were not significantly different between the same months of four successive years, so the data were pooled to give a single representative ‘year’. At most sites there were marked seasonal fluctuations in these variables. In both species, mean EV was largest for ‘winter’ eggs in December/January and smallest for ‘summer’ eggs in May, EV decreased during some months of the year when FI increased, and vice versa. Mean reproductive effort (RE = EV × FI) declined from high ‘winter’ values to low ‘summer’ values, and this difference was most marked at sites where the major foodstuff available was seasonally limited (autumn-shed tree leaves). Overall (‘annual’) mean RE was also lowest at these sites (S3 and S4) but was some 13% greater at sites where nutrients and food were present in the greatest quantifies (S5 and S6). Specific maximum potential body sizes and hence maximum potential fecundities (fifty-nine eggs for G. fossarum and eighty-seven eggs for G. roeseli) were attained only at S5 and 56. Mean RE was nearly identical for G. fossarum and G. roeseli from the same sites.
• 5 At experimental temperatures (T) in the range 3.8–20.2°C, females of the same body weight carried significantly different egg numbers per clutch. The relationship between FI and T was described by a second-order polynomial. The calculated optimum T was 12.1°C for G. fossarum and 16.3°C for G. roeseli. Interspecific differences were highly significant.
• 6 Theoretical lifetime potential fecundity was approximately 194 eggs for G. fossarum