1. An account is given of the innervation of the skin from selected regions of the body in man, the rhesus monkey, and the rabbit. It is based on a critical evaluation of methylene blue and silver preparations in skin which had been treated with the enzyme hyaluronidase before removal from the body and in skin which had not thus been treated.
2. The nerve fibres and terminals in preparations treated with hyaluronidase resemble those seen in fresh specimens of cornea under phase contrast conditions.
3. The numerous artifacts seen in the cornea following methylene blue staining or silver impregnation, the origin of which is known, resemble those seen in specimens of skin not treated with hyaluronidase.
4. It is shown that cutaneous nerves ultimately terminate in an arborization of fine naked axoplasmic filaments. The naked filaments always arise from ensheathed stem fibres of varying diameters and they all end freely and extracellularly in relation to a variety of tissue elements in the skin.
5. The actual terminal filaments wherever they end in relation to the epidermis, dermal connective tissue, blood-vessels, sweat glands, hair follicles, or within capsules of epithelial cells, cannot be distinguished from one another on the basis of inherent morphological differences.
6. Terminal filaments from neighbouring stem fibres are always intimately related to one another and to the tissues in which they lie.
7. The stem fibres and axoplasmic filaments to which they give rise in encapsulated endings and hair follicles are so arranged that the slightest deformation of capsule or of follicle will alter the relationship of the terminal filaments to their stem fibres.
8. Encapsulated nerve-endings may be served by more than one stem fibre but none of the terminals of these fuse with one another. Likewise, the terminals derived from individual stem fibres ending in relation to hair follicles do not enter into protoplasmic continuity with one another.
9. The number and length of the axoplasmic filaments varies greatly from stem fibre to stem fibre.
10. Our observations on the behaviour of the cutaneous nerve fibres and their terminals under different histological conditions provides us with a standard by which we can distinguish artifacts and thus re-evaluate the observations on cutaneous innervation which have been reported in the literature.
11. Evidence is brought forward which suggests that the naked axoplasmic filaments derived from sensory stem fibres may be responsible for ‘flare’ and other ‘nocifensor reactions’ (Lewis), in addition to and apart from their sensory functions.
12. The relationship of these observations to the mechanism of cutaneous sensibility is discussed.