1. The structure and innervation of muscle spindles from normal, de-afferented and de-efferented muscles of the cat hind limb were studied. The spindles were either completely isolated by microdissection, or were serially sectioned transversely. 2. All spindles contain two distinct types of intrafusal muscle fibre, ‘nuclear bag fibres’ and ‘nuclear chain fibres’, which differ in structure and innervation. 3. Nuclear bag muscle fibres, usually two per spindle, are less than half the diameter of extrafusal fibres, and each contains numerous large nuclei packed together in the equatorial region of the spindle. Nuclear bag fibres practically never branch. The fibres contain numerous myofibrils uniformly distributed in cross-sections, and relatively little sarcoplasm; they atrophy very slowly after the ventral spinal roots are cut. Several small motor nerve fibres (y, fibres) enter each spindle and terminate in a number of discrete motor end-plates on the nuclear bag muscle fibres. These y x end-plates lie in a group at each spindle pole and long lengths of nuclear bag fibre are free of motor innervation. 4. Nuclear chain muscle fibres, usually four per spindle, are about half the length and diameter of nuclear bag fibres in spindles in the leg muscles. The nuclear chain fibres in spindles from the small muscles of the foot may, however, equal the nuclear bag fibres in length, and in diameter beyond the ends of the lymph space. Each nuclear chain fibre contains a single row of central nuclei in the equatorial region; the fibres occasionally branch, but often none of them do so. They contain fewer myofibrils per unit area, irregular in size and distribution, and relatively more sarcoplasm, than nuclear bag fibres. Nuclear chain fibres atrophy nearly as rapidly as extrafusal fibres after the ventral roots are cut. A number of very fine motor nerve fibres fibres) enter each spindle and terminate in a network of fine axons and small nerve endings (the network’) situated on the nuclear chain muscle fibres in most regions other than the nuclear region. 5. All spindles receive both y
1
xand y
2
innervation, fibres forming slightly more than half of the total number of motor fibres which varies from seven in simple spindles in phasic muscles to twenty-five in the most complex spindles in tonic muscles. Both y
1
and y
2
fibres remain intact after dorsal root transection and degenerate following ventral root transection. The histological evidence supports the view that the yj and y2 nerve fibres at the spindles are derived from two types of stem fibre, neither of which belongs to the a group. 6. Each spindle has one primary sensory nerve ending, supplied by one group 1
a
afferent nerve fibre, and from zero to five secondary sensory nerve endings, each supplied by one group II afferent nerve fibre. The primary sensory terminations lie on both nuclear bag and nuclear chain muscle fibres. The secondary sensory terminations lie predominantly on the nuclear chain muscle fibres. In spindles with several secondary sensory endings, their terminations may lie on the same region of nuclear chain fibres as motor endings of the y
2
network. 7. In general, spindles in tonic muscles have more secondary sensory endings and motor nerve fibres and endings than those in other muscles. Nuclear chain intrafusal fibres are probably functionally ‘slower’ than nuclear bag intrafusal fibres, while both types are ‘slower’ than extrafusal fibres. Both nuclear chain fibres and nuclear bag fibres, however, probably show a gradation in activity related to the nature of the muscle in which they lie. The reader is advised to study figure 33 and its legend first, at the same time studying the plate figures to which reference is made in figure 33
b
, then to read the portions of the Results in italics consecutively followed by the Discussion, finally studying the detailed Results. Further details of many of the illustrations and tables are available for reference in the Archives of the Royal Society.
The effects of exercise training versus intensive insulin treatment on skeletal muscle fibre content in type 1 diabetes mellitus rodents