And what is proved of nerve-medicines holds good still more obviously with medicines that act on the blood, and on blood disorders. Nearly all of these have been found to exist in the blood, and to pervade the whole mass of the circulation, wherever at first introduced. The glands of the body form a third case in which we require proof of actual local access. This matter will be discussed when we have to consider the subject of Eliminative medicines, when I shall attempt to show that the majority of those medicines actually pass through and are excreted by the glands which they affect. When Mercury is chemically detected in the secretions of the liver and bowels; Sulphur in that of the skin; Turpentine and Copaiba in that of the kidneys; it is evident that these substances must have reached bodily the glandular organs to which their action is directed. Astringents are medicines which from the very nature of their action--apparently a chemical one--cannot operate at all without touching the muscular fibre, which they cause to become contracted. But in laying down this rule of the necessity of local access for the production of the primary[30] effect of a medicine, we must be careful that we do not make it too absolute. In saying that no action at all can be propagated by the agency of the nervous system, we do not make proper allowance for the vital properties of nerve-fibre. The vital nature of common nerve-fibre is such that two actions can be conducted along it;--an impulse producing sensation, or an impulse producing muscular action, at a distant point. It is known that an impression on the terminal extremity of a sensory nerve is capable of producing either sensation or motion at a distance, by what is called _reflex nervous action_. Though this impression must pass through the brain or spinal cord, yet these centres are not appreciably affected by it. Now it is possible, though it does not often happen, that the action of a medicine on the extremity of a nerve may cause this distant action, without reaching the part at which it is manifested. It may be positively affirmed that in no other case can the direct action of a medicine be conducted along a nerve. Having made allowance for the physiological characters of nerve, we must also take notice of the vital properties of muscular fibre. We sometimes find, especially in the case of unstriped muscle, that when one part of a muscular organ is caused to contract, a wave of contraction is propagated along the fibres; and this action may even be extended to a neighbouring muscle, either by contact or sympathy. In one or two cases it seems that muscular contraction may occur in this way as the result of the action of a medicine. We will now consider separately the two exceptional cases.