The experiments already described have shown that curare does not paralyse the trunks of motor nerves (p. 148), nor the muscular substance (p. 148), and does paralyse the peripheral terminations of the motor nerves within the muscles: but they do not show what the exact part of the peripheral terminations is on which the drug exerts its action.

When a nerve enters a muscle it divides and subdivides dichotomously until the fibres become single, and, losing their myelin sheath, the axis-cylinders enter the muscular fibres. There they end in the nerve-plates, from which the ultimate branches pass to the muscular substance.

The paralysis produced by curare may be due to its action on:

(a) The single nerve-fibrillae before they completely lose their myelin sheath;

(b) The axis-cylinders;

(c) The end plates;

(d) The ultimate branches.

As curare acts so much more readily on the nerves passing to voluntary than on those passing to involuntary muscles, and the most marked anatomical difference between these two kinds of curarised muscle.

9 Harnack, Buchheim's Pharmacologic, 3rd ed. p. 615.

10 Sachs, Archivf. Physiol, 1877, p. 91: Schiffer, Dcutsch. med. Wochenschr., 1882, No. 28.

11 Several authors quoted by Guareschi and Mosso, Les Ptomaines. 1883.

12 Schroff, Wochcnblatt d. Ztschr. d. Aertzc zu Wicn, No. 14, 1866.

13 Eabuteau, Traite dementaire de The'rapcutique, 4me ed. p. 536 et scq.

14 Vulpian, Arch, de Physiologic, 1868.

muscles consists in the termination of the former in end plates, it is natural to suppose that curare acts upon these plates.

Fig. 55.   Curve showing the excitability in different parts of the sartorius of a frog in a normal and

Fig. 55. - Curve showing the excitability in different parts of the sartorius of a frog in a normal and

Fig. 56.   Shows the distribution of the nerves in the gastrocnemius of the frog and the curve of excitability in different parts of the muscle. It will be observed that the excitability is greatest in those parts where there are most nerve endings.

Fig. 56. - Shows the distribution of the nerves in the gastrocnemius of the frog and the curve of excitability in different parts of the muscle. It will be observed that the excitability is greatest in those parts where there are most nerve-endings.

Moreover, this supposition appears to receive confirmation from the observation of Kuhne - that the end plates undergo a certain alteration in poisoning by curare, their outlines becoming more distinct than in the normal condition. This slightly increased sharpness of outline may be regarded as indicating a slight physical change, which might, however, be associated with such profound chemical changes in the end plates as to destroy their power of conducting stimuli from the nerve to the muscle.

But recent researches by Kuhne and one of his pupils, Politzer, appear to render it probable that some of the nerve-structures within the muscle retain their functional activity even in profound poisoning by curare; and Politzer supposes that the part of the nerve which is acted on by curare is the nerve-fibril before it has quite lost its medullary sheath, and that the poison destroys the conducting power of the nerve by acting on the cement-substance at Ranvier's nodes. The grounds on which this supposition is based are that, even in profound poisoning by curare, those parts of the sartorius of the frog which contain nerve-endings are more irritable than those which contain none (Fig. 56), and that the irritability increases or diminishes in proportion to the number of nerve-endings, just as it does in the normal muscle, although the excitability of all the parts containing nerves is less than normal in curare-poisoning.

That this variation in irritability in different parts of the muscle is due to nervous structures, and not to variations in the muscular fibres themselves, is shown by the fact that, when the excitability of the nerve is depressed by throwing it into a state of anelectrotonus, these variations in the excitability of the muscle disappear.

It is just possible that the nervous structures which retain a certain amount of excitability in curare-poisoning may be the ultimate terminations which pass from the motor plate to the muscular fibre: but Politzer appears to throw this possibility aside, and considers that the amount of nervous excitability retained shows that all the parts beyond the last node of Ranvier still possess their functions.

Should Politzer's supposition - that curare paralyses motor nerves by acting on the cement at Ranvier's nodes - be correct, it may perhaps serve to explain, not only the difference between its action on motor nerves going to voluntary and those going to involuntary muscular fibre, but also the difference between the action of curare, or poisons having a similar action, and of atropine on the inhibitory fibres of the vagus.