We have examined the organization of muscle afferent projections to motoneurons in the lumbosacral spinal cord of chick embryos between stage 37, when muscle afferents first reach the motor nucleus, and stage 44, which is just before hatching. Connectivity between afferents and motoneurons was assessed by stimulating individual muscle nerves and recording the resulting motoneuron synaptic potentials intracellularly or electrotonically from other muscle nerves. Most of the recordings were made in the presence of DL-2-amino-5- phosphonovaleric acid (APV), picrotoxin, and strychnine to block long- latency excitatory and inhibitory pathways. Activation of muscle afferents evoked slow, positive potentials in muscle nerves but not in cutaneous nerves. These potentials were abolished in 0 mM Ca2+, 2mM Mn2+ solutions, indicating that they were generated by the action of chemical synapses. The muscle nerve recordings revealed a wide-spread pattern of excitatory connections between afferents and motoneurons innervating six different thigh muscles, which were not organized according to synergist-antagonist relationships. This pattern of connectivity was confirmed using intracellular recording from identified motoneurons, which allowed the latency of the responses to be determined. Short-latency potentials in motoneurons were produced by activation of homonymous afferents and the heteronymous afferents innervating the hip flexors sartorius and anterior iliotibialis. Stimulation of anterior iliotibialis afferents also resulted in some short-latency excitatory postsynaptic potentials (EPSPs) in motoneurons innervating the knee extensor femorotibialis, though other connections were of longer latency. Afferents from the adductor, a hip extensor, did not evoke short-latency EPSPs in any of these three types of motoneurons. Short-latency, but not long-latency EPSPs, persisted during repetitive stimulation at 5 Hz, suggesting that they were mediated monosynaptically. Long-latency, fatigue-sensitive potentials were maintained in the presence of APV, picrotoxin, and strychnine, suggesting that polysynaptic pathways utilize non-NMDA receptors as well as NMDA receptors. We found no difference in the pattern of inputs to femorotibialis motoneurons between stage 37–39 and near hatching at stage 44, suggesting muscle afferent projections to these motoneurons are correct at stage 37, when the afferents first reach the lateral motor column in substantial numbers.