In humans, the respiratory efficacy of the external intercostal muscle (EIC) is higher in the dorsal area and in the upper ribcage. As quantification of the electromyographic (EMG) activity is not possible to obtain for this muscle in man, a neuromechanical matching is difficult to demonstrate. Here, we studied this matching in the rabbit, species in which ribs are slanted caudally and ventrally like in man. The mechanical advantage was first evaluated by measuring the change in muscle length during passive inflation of the chest ([Delta]L/L[Delta]V_L). When the lung was inflated from functional residual capacity (FRC) to total lung volume, [Delta]L/L[Delta]V_L was -9.0 1.2, -4.20.6 and –3.01.0% of the length at FRC for dorsal, middle and ventral bundles, respectively. In the 1^st to 7^th interspace, [Delta]L/L[Delta]V_L was -7.31.1, -11.21.9, -8.20.7, -5.81.7, -0.31.2, 2.91.6 and 6.21.6 % of the length at FRC, respectively. Integrated EMG activity was then evaluated and expressed in percentage of the maximal activity during nerve stimulation. In the 3^rd interspace, the mean activity was 7.82.7 and 1.71.2%max. in the dorsal and middle bundles, respectively (P<0.05). The ventral bundle was always inactive. With resistive loading, the activity increased in the three bundles and averaged 34.713.8, 23.36.1 and 8.63.9%max., respectively. In the different interspaces, resting activity was in mean as follows: 2^nd: 4.60.9%, 3^rd: 6.72.0%, 4^th: 4.21.8%, 5^th: 0.80.3%max. The muscle was always inactive in the 7^th interspace. With resistive loads, the muscle remained silent in this interspace; but, for the 2^nd to 5^th interspace, the activity rose respectively to 22.65.5, 40.913.0, 22.110.3, 14.65.9%max. when breathing against the highest load. In the rabbit, the distribution of the EMG activity in the external intercostal muscle thus follows the topographic distribution of the mechanical advantage, confirming the neuromechanical matching observed in the dog.