50 J. C. ECCLES 



0-5 msec (Eidec/a/., I960). Ttwas further found by Eccles, Fattand Landgren 

 (1956) and by Eccles, Eccles and Lundberg (I960) that the Ta impulses 

 selectively excited interneurons in the intermediate nucleus (Fig. 2), and much 

 evidence indicated that these interneurons had properties that precisely fitted 

 them to be interpolated on the la inhibitory pathway, and so to be responsible 

 for the delay of 0-8 msec over and above the monosynaptic excitatory path. 

 For example they were selectively activated by group la volleys (Fig. 2a-c, e) 

 responding with brief latencies and high frequencies (Fig. 2d). Finally, R. M. 

 Eccles and Lundberg (1958) showed that spatial summation of la afferent 

 impulses was necessary in order for them to produce any i.p.s.p. of moto- 

 neurons, which can be explained only if there are interpolated interneurons 

 that require spatial synaptic activation in order to discharge impulses along 

 the final stage of the inhibitory pathway. 



It seemed as if the interneuron had been firmly established as an essential 

 link in the la inhibitory pathway, and hence that the simplest inhibitory 

 pathway was disynaptic as shown in Fig. 2f. However, Lloyd and Wilson 

 (1959) and Lloyd (1960) sought to re-establish the monosynaptic inhibitory 

 pathway by an extraordinary manoeuvre. Their attack on the disynaptic 

 hypothesis was based on the very brief central latency of the inhibition which 

 afferent volleys in the second or third sacral roots exert on contralateral 

 motoneurons; but in addition they radically extended their argument to the 

 la inhibitory action on motoneurons supplying hind-limb muscles. They 

 sought to undermine all the evidence from intracellular recording by making 

 two bold assertions: that the intracellularly recorded i.p.s.p. is not the primary 

 agent of la inhibitory action, but may be a secondary and later manifestation 

 thereof, occurring up to 1 msec later than the onset of inhibition of reflex 

 discharge; that the spike potential recorded intracellularly in the soma 

 cannot be used as a signal of the reflex discharge of an impulse along the axon. 

 Evidently, further experiments were needed in order to resolve this conflict, 

 and they should conform to the requirement of Lloyd and Wilson (1959); 

 namely, that the latency of inhibitory action must be measured on the impulses 

 discharged along the axons in the ventral root. In view of their assertions 

 results derived from intracellular recording are inadmissible. 



Figure 3 illustrates measurements of central latency that have been made 

 for the first time on the reflex spike discharge. This method depends on the 

 considerable range in the latency of the individual components of the mono- 

 synaptic reflex discharge (cf. Fig. 3i, j). If the inhibitory action is timed to 

 begin during the dispersed reflex discharges from motoneurons, as illustrated 

 in Fig. 3k, the later components of the complex spike discharge will be delayed 

 or suppressed, so causing a deviation of the inhibited spike from the control, 

 as illustrated in Fig. 3l. The onset of the deviation would be expected to 

 provide an accurate measure of the latency of inhibitory action, and this is 

 illustrated in Fig. 3a-f, where the interval between the entrance into the spinal 



