Endocannabinoid modulation of synaptic transmission in the mammalian CNS is well established, and generally occurs via cannabinoid CB1 receptors1. The endocannabinoid system (ECS) is now widely explored for its fundamental physiological interest, and its potential as a therapeutic target for treating neurodegeneration, pain and mental illness. However, actions at neuromuscular synapses in skeletal muscle are much less well characterised. If present, the ECS might also prove a useful target for treating neuromuscular diseases, such as Isaac’s disease2, congenital myasthenias and failing synapses in motor neurone disease. Recent studies show CB1-selective agonists, particularly arachidonylcyclopropylamide (ACPA), inhibit neuromuscular transmission at lizard and frog neuromuscular junctions (NMJs)3,4. Here, we ask if mammalian NMJs might have a similar system and if so, can CB1-mediated synaptic inhibition modulate nerve-evoked muscle contraction? Diaphragm strips from adult male MF1 mice (30-45gm) were isolated ex vivo after humane killing (Schedule 1, ASPA, 1986, EU modification) and suspended in Krebs solution at room temperature. After ensuring healthy neurotransmission (curare-sensitive nerve-evoked contraction >95% muscle evoked contraction), it was reduced by incubating in 0.5mM Ca2+ and sufficient Mg2+ (1.19-3mM) to reduce twitch and tetanic tension to ~50% of initial values. This provided a suitable bioassay to investigate both effects on synaptic transmission (increases or decreases in nerve-evoked contraction), and direct stimulation-evoked muscle contractility.0.1 – 10µM ACPA produced a dose-dependent inhibition (P<0.0001, Two-Way ANOVA) of both nerve-evoked twitch tension, by up to 48.7 ± 5.6% (mean ± SE), and tetanic (50Hz, 0.5s) tension, by 14.4 ± 1.3% (both n=4), vs vehicle control (DMSO, n=8) at 10µM. It reached significance at 0.2µM (P<0.03, Bonferroni post-hoc). Inhibition by another CB1-selective agonist, ACEA (arachidonyl-2′-chloroethylamide), was less for twitch (maximum 32.2±6.8% at 10µM; ANOVA P<0.01 vs ACPA), but not tetanic (maximum 14.7 ± 9.4% at 10µM, both n=6) tension over the same range. ACPA-induced inhibition of nerve-evoked twitch and tetanus (both n=10) was abolished by 10µM AM251 (CB1 antagonist/inverse agonist, n=9). Cannabinoids had no effect on direct muscle-evoked muscle tension.This is the first substantive evidence that CB1 receptors are functionally important at mammalian skeletal muscle NMJs. The data suggest that these receptors can strongly modulate neuromuscular transmission, and hence that the CB1 receptor might be a novel therapeutic target for agonists, inverse agonists or allosteric modulators to treat neuromuscular diseases involving such perturbed neurotransmission.