1. Intracellular recording techniques were used to study the effects of temperature on resting membrane conductances, electrical excitability and synaptic efficacy in fast-glycolytic (FG) skeletal muscle fibres from the lizard Dipsosaurus dorsalis.

  2. The conductance of the resting muscle membrane to chloride ions (gci) increased from 488 μS cm−2 at 15°C (pH7-8) to 730 μS cm−2 at 45°C (pH7·4), yielding a temperature coefficient (thermal ratio, Rio) of 1·14. Resting potassium conductance (gK) increased from 84μS cm−2 at 15 °C to 236 μS cm−2 at 45 °C (R10=1·41).

  3. Fibres bathed in Cl-free Ringer’s solution were hyperexcitable, and produced repetitive action potentials both during and following intracellular current injection. At the preferred body temperature of Dipsosaurus (near 40°C) the fibres also fired repetitively in response to single nerve shock.

  4. The electrical excitability of Dipsosaurus fibres decreased with increasing temperature. Threshold current, measured at endplate regions of fibres bathed in normal Ringer’s solution, was 146 nA at 15°C and 353 nA at 45°C (R10= 1·34).

  5. Despite the temperature-dependent change in threshold current, at both 15 and 45 °C all fibres examined had suprathreshold neuromuscular transmission in response to single nerve shock.

  6. The relative thermal independence of gCl in Dipsosaurus fibres may be an adaptation that contributes to a large safety factor for neuromuscular transmission at the high body temperatures preferred by this lizard species.

In vertebrate skeletal muscle, 50–90 % of the resting membrane conductance is carried by Cl, with the remaining conductance mostly attributable to K+ (Bretag, 1987). A large resting chloride conductance (gCl) is necessary for maintaining appropriate levels of electrical excitability; when gCl is reduced experimentally or as a consequence of disease, muscle fibres generate repetitive action potentials and contract tetanically in response to single stimuli (Bryant & Morales-Aguilera, 1971; Adrian & Bryant, 1974). Thus, it is important that skeletal muscle fibres possess adequate gCl.

Conversely, since resting membrane conductances largely determine the input resistance of a muscle fibre, and thereby influence its response to synaptic current, it is also important that gCl should not be too large. Excessively large ga would lower the input resistance of muscle fibres, raise the requirement for endplate current, and thus diminish the safety factor for neuromuscular transmission.

Temperature often has significant effects on membrane conductances. For skeletal muscle ga, the magnitude and direction of the effect vary widely among different species and classes of vertebrates. For example, gCl has a temperature coefficient of 1·3 in frog sartorius muscle (Sperelakis, 1969), 1·5 in teleost myotomal muscle (Klein & Prosser, 1985) and 2·0 in elasmobranch fin muscle (Hagiwara & Takahashi, 1974). In mammalian muscle, the temperature coefficient of ga ranges from 0·8 to 1·8, but is frequently less than 1·0 (Bretag, 1987).

Since membrane gCl is important in determining the electrical excitability of muscle fibres and is also temperature-dependent, the question arises of how temperature influences the electrical excitability of skeletal muscle. This is especially relevant for poikilothermic vertebrates such as reptiles that experience wide diurnal variations in body temperature. Little information is presently available concerning the chloride conductance of reptilian muscle or the effects of temperature upon the electrical excitability of vertebrate skeletal muscle. The goals of the present work were to measure gCl in reptilian skeletal muscle fibres, to determine its thermal dependence, and to assess directly the effects of temperature on electrical excitability. The experiments were done on fast-twitch, highly glycolytic muscle fibres from the lizard Dipsosaurus dorsalis. This species was selected because it is relatively thermophilic (field activity temperature 42°C; Norris, 1953) and thus might be expected to show interesting adaptations of membrane conductance and electrical excitability to high temperatures.

Male desert iguanas (Dipsosaurus dorsalis) [N = 22; mean (±S.E.M.) body mass 42 ± 5g (range 25–54g)] were collected near Palm Springs, CA, under California State Fish & Game permit no. 2171. Lizards were held in glass aquaria, provided with a photothermal gradient that allowed behavioural thermoregulation, and were fed chopped iceberg lettuce that was coated with pulverized Purina Puppy Chow and Special-K high-vitamin cereal. Water was available at all times. Lizards used in this study had been in captivity for at least 2 months and were in excellent condition.

Fibres from the white region of the iliofibularis muscle were used for all experiments described here. This type of fibre [fast-twitch, highly glycolytic (FG); Gleeson et al. 1980] makes up 80–90 % of the skeletal muscles of Dipsosaurus (Putnam et al. 1980). Pairs of iliofibularis muscles were pinned at 120% of the in situ length (taken as the in vivo length with the knee at 90° and the femur perpendicular to the vertebral column; range of muscle lengths 17–23 mm) in Sylgard-lined dishes. In some experiments, the motor nerve was placed in a separate chamber and stimulated (30 V, 0·1 ms pulses) via a pair of silver wires. For determination of passive electrical properties, one muscle from each lizard was covered with normal Ringer’s solution [145 mmol l−1 NaCl; 2 mmol l−1 KC1; 2·5 mmol l−1 CaCl2; 4 mmol l−1 Hepes; 100 nmol l−1 tetrodoxin (TTX); pH adjusted to 7·60 at 23°C with NaOH] and studied immediately. The contralateral muscle was equilibrated in Cl−-free Ringer (145 mmol l−1 sodium isethionate; 1 mmol l−1 K2SO4; 3·5 mmol l−1 CaSO4; 4 mmol l−1 Hepes; 100 nmol l−1 T1X; pH7-6 at 23°C with NaOH) for 1–2 h before measurements were made. The Ca2+ activity of the Cl-free Ringer was calculated to be 2·5 mmol l−1 (Pollard et al. 1977). For determination of electrical excitability, the muscles were bathed in normal Ringer without TiX. During all experiments the muscles were continuously superfused with the appropriate solution. Temperature was varied by heating or cooling the Ringer before it entered the muscle chamber and was recorded with a small glass-covered thermocouple placed within 1mm of the muscle. The sequence of temperature exposures was random for all but the highest temperature, which was tested last. The preparation was allowed to equilibrate at each newly selected temperature for at least 5 min before measurements were made. The pH of the Ringer’s solutions varied with temperature: normal Ringer had a pH of 7·8 at 15°C and a pH of 7·4 at 45°C; Cl-free Ringer had a pH of 7·7 at 15°C and a pH of 7·4 at 45°C. These pH changes with temperature are very similar to that of Dipsosaurus plasma (Bickler, 1981). Previous studies (Hagiwara & Takahashi, 1974; Palade & Barchi, 1977; Klein, 1985) have shown that variations in pH over this range do not affect the resting chloride conductance of skeletal muscle.

Intracellular recording techniques

Micropipettes having resistances of 10–15 MΩ (3 mol l−1 KC1) were used for injecting current and recording intracellular potentials. Membrane potentials were amplified using Getting model 5 preamplifiers set on wide bandpass (d.c. to 10 kHz). Reference electrodes were chlorided silver wire connected to the bath via a Ringer-agar bridge. Injected currents were monitored with a virtual ground circuit.

Measurement of resting membrane conductances

Passive electrical properties of Dipsosaurus FG muscle fibres were determined by the procedure described in Adams (1987) with the following modifications: (1) small depolarizing rather than hyperpolarizing current pulses were used, and (2) fibre diameters were calculated using previously determined values for sarcoplasmic resistivity in Dipsosaurus FG muscle fibres (217 Ωcm at 15 °C and R10 = 0·76; Adams, 1987). The current intensity was adjusted such that the change in membrane potential was less than 15 mV at the site of current injection. Impalements were made near the middle of the fibres. Measurements obtained at three or more electrode separations were used to calculate membrane constants. Data from a given muscle fibre were included in the analysis only if the initial and final resting potentials were at least −80 mV and −70 mV, respectively, and the least-squares linear regression of electrode separation versus log voltage offset had a correlation coefficient of 0·95–1·00. For each lizard, several fibres from one muscle were analysed in normal Ringer’s solution, and several fibres from the contralateral muscle were analysed in Cl “-free Ringer. The fibres were treated as infinite cables for the calculation of membrane constants; this procedure was judged to be appropriate because the fibres were long (17–23 mm) compared to their length constants (less than 1·5 mm in normal Ringer and 2·9 mm in Cl-free Ringer). Membrane conductance (gM) was calculated as the inverse of apparent membrane resistance (Rm), measured in normal Ringer’s solution. The membrane conductance attributable to K+ (gK) was measured as the apparent membrane conductance following equifibration of the muscle in Cl−_-free Ringer’s solution. Chloride conductance (gCl) was calculated as the difference between average gM and average gK at each temperature.

Measurement of electrical excitability

Electrical excitability was measured at endplate regions, visually identified by locating terminal nerve branches, and confirmed upon impalement by the presence of spontaneous miniature endplate potentials having a fast rising phase (<1 ms at 25°C). Two microelectrodes, one for recording membrane potential and one for injecting current, were inserted into the same muscle fibre, separated by 50 –100 μm. Usually, the resting potential depolarized by a few millivolts upon insertion of the second electrode. Data from a given muscle fibre were used only if the initial resting potential was at least −80 mV and did not depolarize by more than 5 mV following insertion of the second electrode. The membrane potential was held at the initial resting level by passing steady hyperpolarizing current. Square pulses of depolarizing current (2ms duration, rise-time 10–50 μs) were injected using a current-clamp circuit. The amplitude of the square pulse was gradually increased until an action potential was generated. Threshold current was defined as the minimum amplitude current pulse required to generate an action potential. Threshold potential was taken as the point at which membrane potential inflected and became regenerative, leading to an action potential. Input resistance was calculated as the maximum change in membrane potential in response to a 2 ms subthreshold current pulse divided by the pulse amplitude. The change in membrane potential used in determining input resistance was kept 10–15 mV below that eliciting an active membrane response. Data on electrical excitability were obtained from 36 different muscle fibres per temperature, sampled equally from six different lizards (six fibres from each lizard at each temperature).

Where appropriate, the experimental data were log-transformed prior to statistical analysis to correct for unequal variances at different temperatures. Temperature-dependence of the experimental variables was tested by one-way analysis of variance (ANOVA), followed by a Student-Newman-Keuls (SNK) multiple-range test. Threshold current was analysed by multiple regression with input resistance and critical depolarization as independent factors. Values in the text are reported as mean±S.E.M., with the number of observations in parentheses.

Temperature effects on resting membrane conductances

Fig. 1 shows characteristic membrane responses to depolarizing current injection of Dipsosaurus FG muscle fibres bathed in normal Ringer or in Cl-free Ringer. In normal Ringer (Fig. 1A) the membrane charged rapidly during current injection, and membrane potential levelled off at a stable value that lasted as long as the current pulse. During large depolarizations, membrane potential declined from an initial peak. This decline could reflect the activation of membrane chloride and/or potassium conductances. Observations made in Clfree solution suggest that activation of membrane chloride conductance is responsible for the decline seen in normal Ringer (see below).

Fig. 1.

Membrane responses to intracellularly injected, depolarizing current pulses in Dipsosaurus FG muscle fibres. (A) Fibre bathed in normal Ringer. (B) Fibre bathed in Cl-free Ringer. In both A and B, the resting membrane potential was −90mV and the temperature was 40 °C.

Fig. 1.

Membrane responses to intracellularly injected, depolarizing current pulses in Dipsosaurus FG muscle fibres. (A) Fibre bathed in normal Ringer. (B) Fibre bathed in Cl-free Ringer. In both A and B, the resting membrane potential was −90mV and the temperature was 40 °C.

In Cl-free Ringer (Fig. IB) the membrane charged much more slowly, and during large depolarizations lasting several hundred milliseconds, membrane potential continued to ‘creep’ upwards asymptotically. The creep probably reflects K+ accumulation in the transverse tubule system during the depolarizing pulse (Aimers, 1972). The extent of the creep depended upon absolute membrane potential, and was negligible at potentials negative to −75 mV. In some fibres, creep was large and interfered with the determination of membrane time constant (τ); data from such fibres were discarded. The decline of membrane potential from an initial peak, seen in fibres bathed in normal Ringer (Fig. 1A), was never observed in fibres bathed in Cl-free solution, suggesting that the decline reflects the activation of membrane chloride conductance.

Table 1 lists the passive electrical properties of Dipsosaurus FG muscle fibres, measured in normal Ringer or in Cl-free Ringer at temperatures from 15 to 45 °C. To summarize, length constant (L), input resistance (Rin), membrane time constant (τ), apparent membrane resistance (Rm) and apparent membrane capacitance (Cm) were all considerably larger in Cl-free Ringer than in normal Ringer. These differences in passive electrical properties all resulted from the greatly decreased membrane conductance of muscle fibres in the absence of Cl. Calculated fibre diameter (dcalc) was reduced in Cl-free Ringer to about 72 % of its value in normal Ringer, reflecting the loss of intracellular KC1 and water which occurs following equilibration of muscles in Cl-free solutions (Hodgkin & Horowicz, 1959; Hutter & Noble, 1960).

Table 1.

Passive electrical properties of Dipsosaurus FG muscle fibres, measured innormal and Cl-free Ringer’s solution

Passive electrical properties of Dipsosaurus FG muscle fibres, measured innormal and Cl−-free Ringer’s solution
Passive electrical properties of Dipsosaurus FG muscle fibres, measured innormal and Cl−-free Ringer’s solution

Increasing the experimental temperature caused substantial decreases in Rin, τ and Rm (Table 1). These results are similar to those obtained in a previous study of Dipsosaurus muscle fibres (Adams, 1987). Temperature effects on these cable properties were similar for fibres bathed in normal Ringer or in Cl-free Ringer.

The component membrane conductances of Dipsosaurus FG fibres and their thermal dependencies are summarized in Table 2. Depending on temperature, the membrane conductance attributable to Cl (gci) was 73–85% of the total membrane conductance (gM). Increasing the experimental temperature caused significant increases in gM (R10 = 1·19), gK (R10 = 1·41) and gCl (R10 = 1·44).

Table 2.

Component membrane conductances of Dipsosaurus FG muscle fibres

Component membrane conductances of Dipsosaurus FG muscle fibres
Component membrane conductances of Dipsosaurus FG muscle fibres

Chloride conductance is required for normal electrical excitability

Dipsosaurus fibres bathed in Cl-free Ringer’s solution were hyperexcitable, and generated repetitive action potentials in response to small (50–100 nA) pulses of intracellularly injected current. Unlike fibres bathed in normal Ringer (Fig. 2A), fibres bathed in Cl-free solution repolarized slowly or not at all following the current pulse, and continued to generate action potentials for hundreds of milliseconds after the current pulse had ended (Fig. 2B). Hyperexcitability in response to intracellular current injection was observed at both low (15–20°C) and at high (40–45°C) temperatures.

Fig. 2.

Active membrane responses of Dipsosaurus FG fibres in response to suprathreshold current pulses. Representative responses of fibre bathed in (A) normal Ringer’s solution and (B) Cl-free Ringer’s solution. In both A and B resting potential was −90 mV and temperature was 20°C. The zero potential level is indicated by horizontal lines.

Fig. 2.

Active membrane responses of Dipsosaurus FG fibres in response to suprathreshold current pulses. Representative responses of fibre bathed in (A) normal Ringer’s solution and (B) Cl-free Ringer’s solution. In both A and B resting potential was −90 mV and temperature was 20°C. The zero potential level is indicated by horizontal lines.

Fibres bathed in Cl-free Ringer also showed hyperexcitability following a single shock to the motor nerve. At 40°C, the initial nerve-evoked action potential was accompanied by an afterdepolarization that was quite large and usually gave rise to repetitive action potentials (Fig. 3A). Movement artefact made it difficult to measure the afterdepolarization in most fibres. However, in five fibres with no apparent artefact, the average afterdepolarization at 40°C was 39 ±2 mV. Interestingly, at 20°C fibres bathed in Cl-free Ringer had smaller afterdepolarizations (16 ±2 mV in five fibres with no apparent movement artefact) and usually produced only a single action potential in response to nerve shock (Fig. 3B).

Fig. 3.

Hyperexcitability of Dipsosaurus FG fibres bathed in Cl-free Ringer’s solution in response to nerve stimulation at (A) 40°C and (B) 20°C. Resting potentials were −87 mV (A) and −85 mV (B). The zero potential level is indicated by horizontal fines.

Fig. 3.

Hyperexcitability of Dipsosaurus FG fibres bathed in Cl-free Ringer’s solution in response to nerve stimulation at (A) 40°C and (B) 20°C. Resting potentials were −87 mV (A) and −85 mV (B). The zero potential level is indicated by horizontal fines.

Electrical excitability is temperature-dependent

The electrical excitability of Dipsosaurus muscle fibres was measured by injecting short current pulses (2 ms duration) intracellularly at endplate regions. Endplate regions were examined because they are the site of normal, nerve-evoked excitation, and because previous studies have shown that skeletal muscle fibres possess a greater Na+ channel density (Beam et al. 1985) and a greater electrical excitability (Thesleff et al. 1974) at endplate than at extrajunctional regions. Short current pulses were used to simulate the brief time course of the evoked endplate current.

Fig. 4 shows membrane voltage responses during a series of gradually increasing current injections. Current pulses of small amplitude, and the corresponding subthreshold voltage deflections, were used to calculate input resistance (see below). A just suprathreshold pulse, found by gradually increasing the current amplitude, was used to determine threshold potential and threshold current. Because the 2 ms pulses used here were considerably shorter than the membrane time constant of these muscle fibres (4–5 ms; Table 1), a single action potential was always generated at the end of the current pulse.

Fig. 4.

Membrane responses during determination of input resistance, threshold potential and threshold current in Dipsosaurus FG muscle fibres. The lower traces are membrane potential and the upper traces are injected current. Depolarizing current injections are shown as downward deflections of the current traces. Holding potential was −80 mV. Temperature = 15°C. The zero potential level is indicated by the horizontal line intersecting the rising phase of the action potential.

Fig. 4.

Membrane responses during determination of input resistance, threshold potential and threshold current in Dipsosaurus FG muscle fibres. The lower traces are membrane potential and the upper traces are injected current. Depolarizing current injections are shown as downward deflections of the current traces. Holding potential was −80 mV. Temperature = 15°C. The zero potential level is indicated by the horizontal line intersecting the rising phase of the action potential.

Fig. 5A shows the effect of temperature on threshold current, i.e. the minimum current needed to generate action potentials. Larger currents were needed toi excite the muscle fibres as temperature was increased. Average threshold current was 146 ± 6nA at 15°C, increasing to 353 ± 17nA at 45°C (R10 = 1·37).

Fig. 5.

Effect of temperature on the electrical excitability of Dipsosaurus FG muscle fibres. (A) Threshold current. (B) Input resistance, calculated from membrane responses to 2 ms subthreshold current pulses. (C) Critical depolarization; the open squares represent average threshold potential and the filled squares represent average resting potential. For A, B and C each point represents the average of 36 different muscle fibres. Bars indicate ± S.E.M. where this is larger than the symbol.

Fig. 5.

Effect of temperature on the electrical excitability of Dipsosaurus FG muscle fibres. (A) Threshold current. (B) Input resistance, calculated from membrane responses to 2 ms subthreshold current pulses. (C) Critical depolarization; the open squares represent average threshold potential and the filled squares represent average resting potential. For A, B and C each point represents the average of 36 different muscle fibres. Bars indicate ± S.E.M. where this is larger than the symbol.

Fig. 5B,C shows the effects of temperature on muscle fibre input resistance (determined using 2 ms current pulses) and critical depolarization (the difference between resting and threshold potential). These variables determine the Ohmic response of the fibres and strongly influence threshold current (see below). Input resistance decreased from 182 ± 9kΩ at 15°C to 114 ± 5 kΩ at 45°C (R10 = 0·86). Critical depolarization did not change between 15 and 25 °C, but increased from 28 ± 0·7 mV at 25°C to 40 ± 0·9 mV at 45 °C (R10 = 1·23). The change in critical depolarization resulted from opposing changes in resting and threshold potentials. Resting potential hyperpolarized with increasing temperature, with a least-squares linear regression slope (±S.E.M.) of −0·31 ±0·015 mV °C−1 (95% confidence interval 0·27–0·35). This change in resting potential is equivalent to 0·36–0·41 % °C−1 (0·33–0·35 % °C−1 is predicted by the Goldman-Hodgkin-Katz equation). Threshold potential did not change significantly between 15 and 35°C, but became more positive between 35 and 45°C (R10 = 0·87).

Threshold current is determined by Ohmic properties

The relationship between threshold current and Ohmic properties was statistically analysed using a step-wise multiple regression, with threshold current as the dependent variable and input resistance and critical depolarization as independent variables. Data from all temperatures were pooled, and input resistance was entered as the first element in the regression. The analysis revealed that threshold current was negatively related to input resistance and positively related to critical depolarization. Input resistance explained 71 % (P< 0·001) and critical depolarization 23 % (P < 0·001) of the variance in threshold current. The strong correlation between threshold current and input resistance illustrates the importance of resting membrane conductances (especially gCl) in determining the electrical excitability of these skeletal muscle fibres.

Safety factor is large in Dipsosaurus FG fibres

An important question is whether the increased threshold current at high temperatures (Fig. 5A) has any consequence for the efficacy of neuromuscular transmission. In particular, are the evoked endplate currents (EPCs) large enough to excite the muscle fibres at high temperatures? To examine this issue, voltage recordings were made from endplate regions of Dipsosaurus FG fibres bathed in normal Ringer’s solution at 45 °C. All fibres examined (N = 150 fibres from five different lizards) produced an overshooting action potential in response to a single, supramaximal nerve shock. No fibres were observed which produced only a subthreshold endplate potential and no action potential. These observations imply that the evoked EPCs were at least as large as the threshold current requirement for each muscle fibre. To calculate the safety factor for neuromuscular transmission, it was necessary to measure the evoked EPCs directly. Numerous attempts were made to voltage-clamp the endplate regions of Dipsosaurus FG fibres during nerve stimulation, using a two-microelectrode voltage-clamp. These attempts were uniformly unsuccessful, owing to the very large size and rapid time course of the unblocked currents. Thus, these results strongly suggest that Dipsosaurus FG fibres possess a large safety factor for neuromuscular transmission, even at 45°C where relatively large EPCs (>300nA; Fig. 5A) would be required to produce an action potential.

The data presented here demonstrate that gCl makes up a large percentage (approximately 80%) of the resting membrane conductance of Dipsosaurus FG muscle fibres. This value is well within the range reported for twitch fibres from other vertebrate animals (Bretag, 1987). In contrast, the thermal dependence of gCl in Dipsosaurus fibres (R10 = 1·14; Table 2) is somewhat lower than that found for twitch fibres from other poikilothermic vertebrates (R10 range 1·3–2·0; see Introduction). The possible physiological significance of this low thermal dependence of gCl is discussed below.

One striking result presented in Table 1 concerns the apparent membrane capacitance (Cm) of Dipsosaurus fibres. The values obtained in Cl-free Ringer are much higher (9·11 μF cm−2) than the values obtained in normal Ringer’s solution (3-4 pF cm−2). A similar finding was reported by Adrian & Aimers (1974) for frog sartorius fibres and by Dulhunty et al. (1984) for rat sternomastoid fibres. The greater apparent membrane capacity in Cl-free Ringer probably reflects a high chloride conductance of the T-tubule membrane in Dipsosaurus fibres. High gCl makes the space constant of the T-tubules short in normal Ringer’s solution, and leads to smaller values of effective membrane capacitance (Adrian & Aimers, 1974; Dulhunty et al. 1984). In Cl-free solution, the space constant of the T-tubules becomes significantly longer and larger values of effective membrane capacitance are obtained.

The repetitive and sustained firing of Dipsosaurus fibres in response to intracellular current injection (Fig. 2B) is very similar to that reported by Adrian & Bryant (1974) for mammalian skeletal muscle fibres bathed in Cl-free solution. In addition, the hyperexcitability in response to nerve stimulation (Fig. 3A) demonstrates that gCl is required for normal neuromuscular function of reptilian fast-twitch muscle. High gCl ensures that the muscle fibres fire only once in response to a single, suprathreshold EPC. It is interesting that hyperexcitability following single nerve shock was observed at 40°C (Fig. 3A) but not at 20°C (Fig. 3B). A similar effect of temperature on nerve-evoked repetitive activity was observed by Hutter & Noble (1960) in frog sartorius fibres. The tendency to fire repetitively in Cl-free Ringer is clearly related to the size of the afterdepolarization (Bryant & Morales-Aguilera, 1971; Adrian & Bryant, 1974), which in this study was much larger at 40°C than at 20°C. Thus, the differential response of the muscle fibres to nerve stimulation at the two temperatures suggests that gCl is more important for maintaining normal neuromuscular function near the preferred temperature of Dipsosaurus than at lower temperatures.

The present results also show that larger depolarizing currents are required to excite Dipsosaurus FG fibres as temperature is increased (Fig. 5A). The increased threshold current results primarily from decreased input resistance (Fig. 5B), which in turn results from increased resting membrane conductances (mostly gClTable 2). Because gci in Dipsosaurus FG fibres has a low temperature dependence, threshold current never increases enough to reduce the safety factor for neuromuscular transmission-Dipsosaurus fibres are all suprathreshold at 45 °C (see Results). However, if gci did not have such a low thermal dependence, threshold current might increase enough to reduce safety factor significantly. Thus, the low thermal dependence of gCl in Dipsosaurus FG fibres may be a specific adaptation for life at high temperatures, because it tends to minimize the effects of temperature on muscle fibre input resistance and threshold current, and thus helps to maintain a large safety factor for neuromuscular transmission at the high body temperatures preferred by this lizard species.

This work was supported by NSF grants PCM 81-02331 and DCB 85-02218 to Dr Albert F. Bennett, a Grant-in-Aid of Research from Sigma Xi, The Scientific Research Society, and by a University of California Chancellor’s Patent Fund Award to BAA. I am grateful to Drs A. F. Bennett, T. Bradley, R. Josephson, R. Miledi, I. Parker and Z. Eppley for their generous help and advice.

Adams
,
B. A.
(
1987
).
Thermal dependence of passive electrical properties of lizard muscle fibres
.
J. exp. Biol
.
133
,
169
182
.
Adrian
,
R. H.
&
Almers
,
W.
(
1974
).
Membrane capacity measurements of frog skeletal muscle in media of low ionic content
.
J. Physiol., Lond
.
237
,
573
605
.
Adrian
,
R. H.
&
Bryant
,
S. H.
(
1974
).
On the repetitive discharge in myotonic muscle fibres
.
J. Physiol., Lond
.
240
,
505
515
.
Almers
,
W.
(
1972
).
Potassium conductance changes in skeletal muscle and the potassium concentration in the transverse tubules
.
J. Physiol., Lond
.
225
,
33
56
.
Beam
,
K. G.
,
Caldwell
,
J. H.
&
Campbell
,
D. T.
(
1985
).
Na+ channels in skeletal muscle concentrated near the neuromuscular junction
.
Nature, Lond
.
213
,
588
590
.
Bickler
,
P. E.
(
1981
).
Effects of temperature on acid-base balance and ventilation in desert iguanas
.
J. appl. Physiol
.
51
,
452
460
.
Bretag
,
A. H.
(
1987
).
Muscle chloride channels
.
Physiol. Rev
.
67
,
618
724
.
Bryant
,
S. H.
&
Mo Rales-Aguilera
,
A.
(
1971
).
Chloride conductance in normal and myotonic muscle fibres and the action of monocarboxylic aromatic acids
.
J. Physiol., Lond
.
219
,
367
383
.
Dulhunty
,
A.
,
Carter
,
G.
&
Hinrichsen
,
C.
(
1984
).
The membrane capacity of mammalian skeletal muscle fibres
.
J. Muscle Res. Cell Mot
.
5
,
315
332
.
Gleeson
,
T. T.
,
Putnam
,
R. W.
&
Bennett
,
A. F.
(
1980
).
Histochemical, enzymatic, and contractile properties of skeletal muscle fibres in the lizard Dipsosaurus dorsalis
.
J. exp. Zool
.
214
,
293
302
.
Hagiwara
,
S.
&
Takahashi
,
K.
(
1974
).
Mechanism of ion permeation through the muscle fibre membrane of an elasmobranch fish, Taeniura lymma
.
J. Physiol., Lond
.
238
,
109
127
.
Hodgkin
,
A. L.
&
Horowicz
,
P.
(
1959
).
The influence of potassium and chloride ions on the membrane potential of single muscle fibres
.
J. Physiol., Lond
.
148
,
127
160
.
Hutter
,
O. F.
&
Noble
,
D.
(
1960
).
The chloride conductance of frog skeletal muscle
.
J. Physiol., Lond
.
151
,
89
102
.
Klein
,
M. G.
(
1985
).
Properties of the chloride conductance associated with temperature acclimation in muscle fibres of green sunfish
.
J. exp. Biol
.
114
,
581
598
.
Klein
,
M. G.
&
Prosser
,
C. L.
(
1985
).
The effects of temperature acclimation on the resting membrane of skeletal muscle fibres from green sunfish
.
J. exp. Biol
.
114
,
563
579
.
Norris
,
K. S.
(
1953
).
The ecology of the desert iguana Dipsosaurus dorsalis
.
Ecology
34
,
265
287
.
Palade
,
P. T.
&
Barchi
,
R. L.
(
1977
).
Characteristics of the chloride conductance in muscle fibres of the rat diaphragm
.
J. gen. Physiol
.
69
,
325
342
.
Pollard
,
H. B.
,
Creutz
,
C. E.
,
Pazoles
,
C. J.
&
Hansen
,
J.
(
1977
).
Calcium binding properties of monovalent ions commonly used to substitute for chloride in physiological salt solutions
.
Analys. Biochem
.
83
,
311
314
.
Putnam
,
R. W.
,
Gleeson
,
T. T.
&
Bennett
,
A. F.
(
1980
).
Histochemical determination of the fibre composition of locomotory muscles in a lizard, Dipsosaurus dorsalis
.
J. exp. Zool
.
214
,
303
309
.
Sperelakis
,
N.
(
1969
).
Changes in conductance of frog sartorius fibres produced by CO2, ReO4,), and temperature
.
Am. J. Physiol
.
217
,
1069
1075
.
Thesleff
,
S.
,
Vyskocil
,
F.
&
Ward
,
M. R.
(
1974
).
The action potential in endplate and extrajunctional regions of rat skeletal muscle
.
Acta physiol, scand
.
91
,
196
202
.