304. The ultimate end for which the apparatus of the circulation is constructed, and for which all its action is exerted, is to convey arterial blood to the capillary arteries. These vessels are totally distinct in structure and in office from the larger arterial tubes. All the tunics of these minute vessels diminish in thickness and strength as the tubes lessen in size, but more especially the middle or the fibrous coat; which, according to Wedemeyer, may still be distinguished by its colour in the transverse section of any vessel whose internal diameter is not less than the tenth of a line; but that it entirely disappears in vessels too small and too remote to receive the wave of blood in a manifest jet. But while the membranous tunics diminish, the nervous filaments distributed to them increase: the smaller and thinner the capillary, the greater the proportionate quantity of its nervous matter; and this is most manifest in organs of the greatest irritability. The coats of the capillaries successively becoming thinner and thinner, at length disappear altogether, and the vessels ultimately terminate in membraneless canals formed in the substance of the tissues. "The blood in the finest capillaries," says Wedemeyer, "no longer flows within actual vessels; it is not contained in tubes whose parietes are formed by a membranous substance distinguishable by its texture and compactness from the adjoining cellular tissue: it is contained in the different tissues in channels which it forms in them for itself; and, under the microscope, the stream is seen easily and rapidly to work out for itself a new passage in the tissues which it penetrates." 305. Some of these fine capillaries, before they entirely lose their membranous tunics, communicate directly with veins. Of the capillaries which terminate by direct communication with veins, some are large enough to admit of three or four of the red particles of the blood abreast; the diameter of others is sufficient to admit only of one; while others are so small that they can transmit nothing but the serum of the blood. As long as the capillary is of sufficient magnitude to receive three or four of the particles abreast, it is evident that it possesses regular parietes; but by far the greater number, before they communicate with veins, lose altogether their membranous coats. There are no visible openings or pores in the sides or ends of the capillaries by means of which the blood can be extravasated, preparatory to its being imbibed by the veins. There is nowhere apparent a sudden passage of the arterial into the venous stream; no abrupt boundary between the division of the two systems. The arterial streamlet winds through long routes, and describes numerous turns before it assumes the nature and takes the direction of a venous streamlet. The ultimate capillary rarely passes from a large arterial into a large venous branch.