Intracellular recordings were obtained from growing tips of regenerating giant axons in the lamprey spinal cord, the recording sites verified by Lucifer yellow injection. In the presence of extracellular Ba+ + (3-6 mM), tetraethylammonium (10-15 mM), and 4-aminopyridine (4-6 mM), action potentials showed prolonged plateaus. The fast initial phase of the action potential, but not the plateau (Ba++spike), was blocked by tetrodotoxin (lop6 gm/ml). The Ba++ spike was associated with increased membrane conductance and could be terminated with hyperpolarizing current pulses. Normal axons did not generate similar Bat+ spikes. However, TTX-resistant. voltage-dependant conductance changes could be elicited in normal axons if much higher concentrations of Ba++ (18-30 mM) were used. Their rate of rise was slower than in regenerating axons (0.6 Vlsec vs 3.2 V/sec; n = 5 ) , and the response did not outlast the current pulse. The Ba++ responses in normal and regenerating axons were blocked by ions known to block voltage-gated Ca+ + conductances (Co++, Ni++, or Cd++). Therefore, these spikes probably represent Ba+ + entry through voltage-dependent Ca+ + channels, suggesting the presence of a higher-than-average voltage-dependent Cat + conductance in the growing axon. However, Ca++-dependent spikes could not be obtained under any conditions in either normal or regenerating axons. Simultaneous intracellular recordings from growth cones and axons indicated that the Ba+ + spike was initiated, in most cases, at the growth cone. The Ba++ spikes were recorded in regenerating axons for as long as 50 days following cord transection and were not correlatable with the "dying-back" phenomenon in cut axons, which usually is over before day 6.
The concept of a higher-than-average voltage-dependent"Ca+ + conductance in growing tips of regenerating axons is in agreement with the hypothesis that Cat+ is important in regeneration and that regeneration may be related to the process of chemical synaptic transmission.