The in vitro uptake of [3H]inulin and horseradish peroxidase (HRP) has been studied in innervated and 6 days denervated extensor digitorum longus muscle of the mouse. Both markers were taken up at a higher rate in denervated muscle. The increase in uptake after denervation was, however, larger for HRP than for [3H]inulin. After 2 h incubation at 37 degrees C, pH 7.3, in the presence of equimolar concentrations of HRP and [3H]inulin (approx. 2.1 microM), the uptake of HRP was approx. 8 times as great as the uptake of [3H]inulin in the same innervated muscles. In denervated muscle the HRP uptake was approx. 19 times as great as the [3H]inulin uptake in the same muscles. Various possible explanations of these differences in uptake have been considered and tested experimentally. [3H]Inulin uptake in skeletal muscle has previously been shown to obey bulk kinetics. The present investigation shows the HRP uptake to obey saturation kinetics. The HRP uptake shows dependency on divalent cations and is reduced if incubation is carried out at pH 6.4. The uptake of HRP, when used at a low, non-saturating concentration (10 micrograms/ml approx. 0.25 microM), is inhibited greater than or equal to 60% by yeast mannan (0.1 mg/ml), ribonuclease B (0.1 mg/ml, approx. 7.4 microM), mannose (30 mM), monodansylcadaverine (1 mM), chloroquine (100 microM), trifluoperazine (25 microM) or maleic acid (2 mM). It is concluded that HRP is taken up in innervated and denervated skeletal muscle by a process of receptor-mediated endocytosis and that this uptake is under neurotrophic control.