How is it, then, that the tendency of a vacuum to become refilled is unable to afford nourishment to one in such a condition? Because its rule is that only so much of the contiguous matter should succeed as has flowed away. Now this is sufficient for nourishment in the case of those who are in good condition, for, in them, what is _presented_[228] must be equal to what has flowed away. But in the case of those who are very emaciated and who need a great restoration of nutrition, unless what was presented were many times greater than what has been emptied out, they would never be able to regain their original habit. It is clear, therefore, that these parts will have to exert a greater amount of _attraction_, in so far as their requirements are greater. And I fail to understand how Erasistratus does not perceive that here again he is putting the cart before the horse. Because, in the case of the sick, there must be a large amount of _presentation_[228] in order to feed them up, he argues that the factor of "refilling"[227] must play an equally large part. And how could much _presentation_ take place if it were not preceded by an abundant _delivery_[229] of nutriment? And if he calls the conveyance of food through the veins delivery, and its assumption by each of these simple and visible nerves and arteries not delivery but _distribution_,[230] as some people have thought fit to name it, and then ascribes conveyance through the veins to the principle of vacuum-refilling alone, let him explain to us the assumption of food by the hypothetical elements.[231] For it has been shown that at least in relation to these there is no question of the refilling of a vacuum being in operation, and especially where the parts are very attenuated. It is worth while listening to what Erasistratus says about these cases in the second book of his "General Principles": "In the ultimate simple [vessels], which are thin and narrow, presentation takes place from the adjacent vessels, the nutriment being attracted through the sides of the vessels and deposited in the empty spaces left by the matter which has been carried away." Now, in this statement firstly I admit and accept the words "through the sides." For, if the simple nerve were actually to take in the food through its mouth, it could not distribute it through its whole substance; for the mouth is dedicated to the psychic pneuma.[232] It can, however, take it in through its sides from the adjacent simple vein. Secondly, I also accept in Erasistratus's statement the expression which precedes "through the sides." What does this say? "The nutriment being attracted through the sides of the vessels." Now I, too, agree that it is attracted, but it has been previously shown that this is not through the tendency of evacuated matter to be replaced. VI