Fig. 1. Force–frequency relationship of the swimming muscle in the medusa of P. penicillatus and the effect of caffeine. (A) A typical force trace of muscle contraction at stimulation frequencies of 0.1, 0.2, 0.4 and 0.8 Hz using 30 ms square pulses that were just suprathreshold. As stimulation frequency increased there was a stepwise increase in contractile tension. (B) Quantitative data for the force–frequency relationship. All maximal contraction forces were normalized to the mean force at 0.1 Hz in ASW. Increasing frequency of stimulation significantly increases the amplitude of contractile force (ANOVA test, P<0.05). The underline between 0.4 Hz and 0.8 Hz at the upper right corner marks the results of the Student–Neuman–Keuls multiple comparison test, indicating no significant difference between amplitudes of contractile force when muscles were stimulated at the rate of 0.4 Hz and 0.8 Hz. Data were collected from nine muscle strips. Values are means ± S.E.M. (C) A typical trace of the effect of 10 mmol l^-1 caffeine on the force–frequency relationship. Arrow indicates the beginning of 10 mmol l^-1 caffeine perifusion, which produced a transient increase in tension. Caffeine perifusion was continued to the end of the trace. After caffeine treatment there was no facilitation of tension at increased frequencies. (D) Quantitative data for the force–frequency relationship when 10 mmol l^-1 caffeine was present in the perifusate. The data for the ‘washed’ group were obtained at least 10 min after the removal of caffeine perifusate with ASW and after the contractile force had stabilised. All maximal contraction forces were normalised to the mean force at 0.1 Hz in ASW. Data were collected from three muscle strips. Caffeine significantly inhibited the maximal tension developed at all frequencies (ANOVA test, P<0.05, indicated by asterisks). Values are means ± S.E.M.