Figure 1.

C. morosus legs assume constant, gravity-independent postures in the absence of leg muscle nerve activity. A, Photographs of an animal in head-up (A1) and head-down (A2) positions with the long axis of the body positioned parallel to gravity (i.e., in A1 the animal is oriented similar to a standing human). The left side of the animal was opened ventrally and all nerves innervating the middle leg muscles, including the protractor and retractor coxae muscles, were severed. The innervations of the muscles on the right side were left intact and EMGs recorded from the protractor and retractor muscles. Both legs remained almost perpendicular to the body's long axis regardless of animal position. Simultaneous recordings (A3) of activity in the protractor and retractor muscles, which control leg position in this plane, showed that maintenance of these postures was not associated with measurable muscle activity, which was absent at all head orientations and during rotations between orientations. B, Recordings of protractor and retractor activity during walking induced by touching the animal showed that the recording electrodes were functioning. C, Photographs of an animal ventral (C1) and dorsal (C2) side down. Femur–tibia joint (the femur is the portion of the leg lying relatively horizontally and the tibia the portion lying vertically, pointing downward, in C1) angle did not change when animal position relative to gravity was changed. Recordings from the flexor and extensor muscles (which control this joint) showed that this maintenance of femur–tibia joint angle was not associated with measurable muscle activity (C3). The last portions of the recordings were the activity that occurred when the animal was induced to walk, and show that the EMG electrodes were functioning. The recordings in C3 were from the same animal but are not simultaneous (first the extensor muscle was recorded from and the animal rotated 360° and then the flexor was recorded from and the animal again rotated 360°).