Motor-Unit Recruitment in the Decerebrate Cat: Several Unit Properties Are Equally Good Predictors of Order

Document Type


Publication Date



1. Recruitment order was studied in pairs of motor units of the medial gastrocnemius (MG) muscle of decerebrate cats with the use of dual microelectrode recording from intact ventral root filaments. Excitation was provided by stretch of MG, stretch of synergists [lateral gastrocnemius (LG), plantaris (PL), and soleus (SOL) muscles] or electrical stimulation of the caudal cutaneous sural (CCS) nerve. Motor units were characterized by axonal conduction velocity (CV), tetanic tension (Pmax), twitch contraction time (CT), and fatigue index (FI).

2. Consistent with the recruitment pattern described by others, most often in relation to either CV or Pmax, the first unit of a pair to be recruited by MG stretch was typically the one with the lower CV and Pmax, and the higher FI and CT. The proportion of pairs that agreed in rank order of each property and recruitment order was as follows: for CT, 94%; for CV, 87%; for Pmax, 84%; and for FI, 75%. With a single marginal exception (CT vs. FI), no motor-unit property proved to be significantly better than the others at predicting recruitment (G test; P > 0.05).

3. In all 11 tested pairs containing one slow (type S) and one fast (type F) unit, the S was more easily recruited by stretch. Type F units divided into groups with high (type FR), low (type FF), and intermediate (type FInt) values for FI were recruited in order from FR to FInt to FF in 8/11 pairs. Thus our findings were similar to earlier demonstrations that recruitment proceeds in order by type.

4. Stretch of MG synergists usually recruited units in the same order as MG stretch. In two S-S pairs, recruitment order was switched with synergist stretch.

5. Stimulation of the CCS nerve was generally excitatory to the MG units sampled. Most unit pairs were recruited by CCS stimulation in the same order as by MG stretch, but, for 6 of 39 pairs, CCS stimulation switched the order produced by stretch. Thus, whereas sural afferent input can preferentially excite some units over others as suggested by Kanda et al., that effect is not widespread or selective for unit type under these conditions.

6. Assuming that all MG motor units cooperate as a single functional pool in homonymous stretch reflexes, we support others in concluding that a motoneuron's recruitment threshold is not strictly determined by its size. However, our data do not distinguish other schemes that predict recruitment order more accurately than the size principle. Rather, they show that different synaptic inputs may introduce occasional shifts in recruitment order, but that the usual order applies generally over the ranges of motor-unit properties studied, irrespective of the source of synaptic excitation or the measure of motor-unit character.