The present "loop analysis" study was undertaken to determine the extent to which the segmented electromyographic (EMG) responses of cat triceps brachii to imposed elbow displacements can be used as a model for those occurring in primate wrist musculature under similar input conditions, since previous workers have assumed homology between the two systems. To this end, the response properties of the EMG peaks and the range of latencies available to a "transcortical loop" were determined in the cat and compared with previous findings from primates.During chronic experiments, simultaneous recordings were made from motor cortical neurons (MCNs) and triceps EMG in response to step loads imposing flexion on the elbow joint. Subsequent acute experiments in the same animals used intraspinal microstimulation to determine the antidromic latencies to the area of cortex containing responsive MCNs, and orthodromic latencies to triceps EMG activation from the same loci.Following displacements imposed from flexed initial joint angles, the EMG was segmented into two or more peaks while, from extended angles, a single segment was followed by a "silent interval" and tonic activity. Segmentation into three peaks was only observed under limited conditions. Peak 1 showed a constant latency (8 to 12 msec) and resembled the primate Ml segment. A "silent interval," not present in primate wrist muscles, followed peak 1, and its duration varied with the average velocity of the imposed displacement. Peak 2 followed the "silent interval" at variable latency (34 to 52 msec), in contrast to the constant and shorter latency shown by the primate wrist M2. Peak 3 was variable in appearance and latency (69 to 85 msec when distinguishable) and its homologies remain unclear. Peak 1 was monotonically graded with average velocity of displacement, whereas peaks 2 and 3 showed variable responses.Summation of the afferent and efferent latencies for the "receiving zone" of the motor cortex gave modal values of 19 to 23 and 23 to 27 msec for the expected time of occurrence of EMG activity dependent upon MCNs responding to the imposed displacement as part of a "transcortical loop." This activity would therefore occur during the "silent interval," in contrast to the M2 segment in primate wrist muscles, where MCN activity seems to be appropriately timed to contribute to this segment.It is concluded that the cat triceps brachii EMG responses, together with the related motor cortical responses to imposed elbow displacements, cannot be utilized as a model to study the mechanisms underlying the longer latency EMG segments observed at the primate wrist.Invasive methods, including lesioning, microstimulaneural circuits underlying mammalian motor control. In tion, and single neuron recording techniques, carried out attempts to determine whether similar circuitry operates in the cat and in a few subhuman primate species have in man, one approach has been to compare electromyprovided information essential to the understanding of ographic (EMG) activity durin...