NERVE ENDINGS IN MAMMALIAN SKIN

Abstract

SUMMARY To most authors the study of nerve terminations in mammalian skin has been an exercise in descriptive morphology. A single histological technique was usually considered adequate and presumably impeccable, for observations resulting from different techniques were not compared and the possibility of artefacts was not considered. The literature is thus excessively large and full of controversies over histological minutiae. It also contains theories concerning the organization of the nervous system which are at variance with the results of physiological observations. The work of a few authors is in striking contrast, but their observations have either been ignored or discounted. The recent discovery that nerve fibres can be seen in fresh specimens of cornea under phase‐contrast conditions has led to the development of new neuro‐histological techniques. These display nerve fibres and their terminals selectively in skin. They appear in an undistorted state and are virtually free from artefacts. The use of new techniques has shown that it is characteristic of all nerves entering mammalian skin to terminate in an arborization of fine (< 1μ in diameter), naked, axoplasmic filaments which probably end freely. Ensheathed stem fibres give rise to unencapsulated nerve endings in all skin strata, and terminals from neighbouring stem fibres overlap and interdigitate extensively. Axoplasmic filaments terminate in the cellular layers of the epidermis and in the dermis but in no specific relation to the capillaries. They are found in relation to the myoepithelial and gland cells of sweat glands, and in relation to the adventitia and media of blood vessels. In hairy skin there are in addition to the unencapsulated nerve endings nerves which end specifically in relation to hair follicles. Ensheathed myelinated stem axons give rise to two distinct and separate series of arborizations of fine, naked, axoplasmic filaments which lie at right angles one within the other. The outer series encircle the hair lying among the cells of the middle layer of the dermal coat. The inner series lie among the cells of the outer root sheath parallel to the hair shaft. No encapsulated nerve endings are seen in hairy skin. In glabrous skin and mucous membranes, such as the lip, anus and glans penis, there are, in addition to the unencapsulated nerve endings, numerous encapsulated nerve endings. They are of different sizes and shapes, and the ensheathed myelinated stem fibre or fibres which enter the capsule pursue a more or less tortuous course. They then give rise in all cases to an arborization of fine, naked, axoplasmic filaments which end freely among the capsular cells in planes roughly parallel to the surface layer of cells. In the light of these findings a re‐analysis of the observations and, in particular, the diagrams in the literature, forces one to the conclusion that the actual facts which have been reported by various authors have more in common than would have been supposed from a perusal of their summaries and conclusions. As the result of comparing observations on the innervation of skin following the use of numerous different techniques including those involving the use of hyaluronidase, it is now possible to reinterpret the observations in the literature. A standard for histological comparison now exists and it is also possible to determine what is and what is not an artefact, and what a distorted and disrupted nerve fibre or terminal really looks like. If the literature is reinterpreted in this way, the points concerning the innervation of mammalian skin which emerge correspond both to those described by a few authors whose work is usually ignored or discounted, and to those described by Weddell et al. (1954) using the improved neurohistological techniques. If this is so, then we are forced to the conclusion that there are not and never have been any purely histological grounds on which to erect theories of cutaneous sensibility based on the existence of four primary modalities, touch, warmth, cold and pain, operating within the ‘law of specific nervous energies’. Furthermore, there is not and never really has been convincing histological evidence for the commonly accepted statement that morphologically specific nerve endings subserve each of the primary modalities of cutaneous sensibility. Our observations suggest that the morphology of nerve terminals in mammalian skin should be studied from the point of view of their functions as transducers of stimuli (mechanical, thermal or chemical) into propagated action potentials rather than taxonomically. This work was made possible by a grant from the Rockefeller Foundation which is gratefully acknowledged. Our best thanks are also due to Miss C. Court for her careful and accurate execution of the illustrations.