Before we inquire into the mechanism by which this extension of the pain takes place, we ought in strictness to ask ourselves whether the essential heart-pain is felt only in the spinal sensory branches, or whether the sympathetic fibres are themselves capable of feeling pain. The latter supposition, notwithstanding all that has been argued in its favor from the supposed analogies of the pain of colic, gall-stone, etc., seems to me very doubtful. It would appear more probable that both the latter pains, and also those of angina, are really connected with branches either of the vagus or of other spinal nerves. And there is no need to invoke the sympathetic as a sensory nerve, to account either for the essential heart-pain of angina, or for its extension into arm, chest-wall, and neck. For the plexus cardiacus receives spinal branches, both from the vagus and also (through the medium of the sympathetic ganglia of the neck) from the whole length of the cervical and the uppermost part of the dorsal cord-centres. And, in this way, it would seem quite possible intelligibly to account for the pain radiating into intercostal, cervical, and brachial nerves, merely by extension of a morbid process essentially seated in the cord. Usually, however, one sees it explained not in this way, but by the inter-communications that exist outside the spine, between the branches from the cervical ganglia and the lower cervical and upper dorsal nerves; and the pain in the arm is especially explained by the connection (outside the spinal canal) of the inferior cervical ganglion, on the one hand with the lower cervical nerves, which go to the brachial plexus, and, on the other hand, with the heart itself. There remains to be explained, however, the singular tendency of the arm-pain to be one-sided (this happens in at least four cases out of five); and this explanation seems to me insuperably difficult, on the theory that the transference of morbid action to the brachial nerves takes place through external anastomoses. It appears greatly more probable that angina is essentially a mainly unilateral morbid condition of the lower cervical and upper dorsal portion of the cord; liable of course to be seriously aggravated by such peripheral sources of irritation as would be furnished by diseases of the heart, and especially by diseases of the coronary arteries; the latter affection probably involving constant mechanical irritation of the cardiac and the aortic plexuses. It is noteworthy that the arm-pain is sometimes (I do not know how often) accompanied by vaso-motor paralysis in the limb; this phenomenon could also certainly be more easily accounted for on the supposition of radiation from a spinal vaso-motor centre (to which the morbid process had extended from a posterior nerve-root) than on that of communication between painful sensory nerves and vaso-motor nerves; through either of the plexuses independently of the spinal centres.