1. Group Ia EPSPs were recorded from lumbosacral motoneurones in anaesthetized cats after almost complete section of the relevant dorsal roots. The EPSPs were usually of small amplitude (median value of 230 muV) and an averaging device was used to improve the definition of their time course.2. From a total of over 500 averaged EPSPs a smaller number (342) were subjected to analysis. The other EPSPs were rejected either because they showed signs of multiple origin in the rising phase of their time course (see Methods) or because the resting membrane potential of the cell was less than 50 mV. All the selected EPSPs had their rise time (from the 10 to the 90% level) and half-width measured, and a semilogarithmic plot of their decay time course was made.3. 252 of the EPSPs showed an exponential decline in their later time course and the slope of this line was used to give an estimate of the membrane time constant. The range of the time constant for different motoneurones was 2.3-12.9 msec, with a mean value of 5.8 msec.4. In ten cells an EPSP was recorded which was judged to be generated exclusively by synaptic knobs located on the soma. On this assumption measurements of the normalized rise time, half-width and break point time were used to estimate alpha, rho(infinity) and L by the method suggested in Jack & Redman (1971b). The estimated value of alpha ranged from 18 to 65. A positive correlation was found between alpha and tau(m), indicating that for these EPSPs the duration of current injection was independent of the membrane time constant. The peak time of the wave form of current injection was between 0.1 and 0.25 msec. The estimates of rho(infinity) were not thought to be very accurate. A lower limit of 4 was assumed and the highest measured value was 12, but in three cells the time course of the EPSP could not be fitted even with a very high value of rho(infinity). Some possible explanations for this discrepancy are mentioned in the Discussion. The electrotonic length of the dendrites (L) was usually greater than 1.0 lambda and ranged between 0.75 and 1.5 lambda. Evidence for an open-circuit termination of the dendrites was found in some cells.5. The normalized values of the rise time and half-width were used to make an electrotonic distance allocation to the 246 EPSPs which were judged to be non-somatic. The method of allocation was not precise because individual values of rho(infinity) and L were not available for these motoneurones. Instead, a maximum possible range was assumed: for rho(infinity), 4-25; for L, 0.75-1.5. The range of alpha was also assumed, from 12 to 100. With these values the motoneurone model (Jack & Redman, 1971b) was used to set limits within which the normalized rise time and half-width of all EPSPs, generated by current at a single point, should lie. Twenty of the 246 EPSPs lay outside these boundary lines and hence they did not receive a distance allocation. The remaining 226 were assigned values between 0.2 and 1.6 lambda (in 0.2 lambda steps); the majority of the allocations (183) were to the proximal electrotonic part of the dendrites (0.2, 0.4 or 0.6 lambda). The relationship of these distance allocations to the histological results of Conradi (1969) is discussed.6. It is concluded that there is no good evidence against the view that the main time course of minimal Ia EPSPs can be explained by their generation by a brief pulse of synaptic current and subsequent passive spread.
Read full abstract