1. The recruitment order of 64 pairs of motor units, comprising 121 type-identified units, was studied during centrally evoked muscle contractions of the cat medial gastrocnemius (MG) muscle in an unanesthetized, high decerebrate preparation. Motor units were functionally isolated within the MG nerve by intra-axonal (or intramyelin) penetration with conventional glass microelectrodes. 2. Graded stimulation of the mesencephalic locomotor region (MLR) was used to evoke smoothly graded contractions, which under favorable conditions was estimated to reach 40% of maximum tetanic tension of the MG muscle. With this method of activation, 100% of slow twitch (type S) units, 95% of fast twitch, fatigue-resistant (type FR) units, 86% of fast twitch, fatigue-intermediate (type FI) units, and 49% of fast twitch, fatigable (type FF) units studied were recruited. 3. Motoneuron size as estimated by axonal conduction velocity (CV) was correlated with muscle-unit size as estimated by maximum tetanic tension (Po). Although the correlation between these properties was significant among type S and FR units, no significant correlation was found for these properties among type FI and FF units. 4. Motor-unit recruitment was ordered by physiological type (S > F, 100% of pairs; S > FR > FI > FF, 93% of pairs). Although none of the motor-unit properties studied predicted recruitment order perfectly, motor-unit recruitment was found to proceed by increasing Po (89% of pairs), decreasing contraction time (79% of pairs), decreasing fatigue index (80% of pairs), and increasing CV (76% of pairs). These percentages were significantly different from random (i.e., 50%). Statistically, all four motor-unit properties were equivalent in predicting recruitment order. These results are similar to those reported by other investigators for motor-unit recruitment order evoked from other supraspinal centers, as well as from peripheral sites. 5. When, however, motor-unit recruitment within pairs of motor units containing two fast-twitch (type F) units was examined, Po was a significantly better predictor of recruitment order than CV (85% vs. 52% of pairs). One explanation for this observation is that the correlation between Po and CV is high among type S, type FR units, and possibly among the lower-tension type FF units, but not among the remaining higher-tension type FF units. 6. The reproducibility of recruitment order in multiple contractions was investigated in 16 motor-unit pairs. Recruitment order was found to be variable in only three motor-unit pairs, all of which contained units of similar physiological type and recruitment threshold. 7. Analysis of recruitment order by pair-wise testing confirms the general conclusion reached in human studies that the muscle force level at recruitment for a motor unit is highly correlated with its strength. As an additional confirmation, the whole-muscle force level at recruitment for 41 units was measured in a series of contractions in which the rate of rise of muscle tension was limited to rates <1,000 g/s. For these contractions, a significant correlation was found between muscle tension at recruitment and motor-unit Po.
ASJC Scopus subject areas