There is a singular uncertainty in the effects of arsenic when applied to ulcerated surfaces. Some persons, like Roux’s patient, are obviously affected by a single application; while others have had it applied for a long time without experiencing any other consequences than the formation of an eschar at the part. Two causes have been assigned for these differences, and probably both are founded on fact. One, which has been assigned by Mr. Blackadder, is the relative quantity of arsenic applied. He says he never witnessed but one instance of its acting constitutionally, although he often applied it to sores; and he imputes this success to his having always used a large quantity. For he considers that by so doing the organization of the part is quickly destroyed, and absorption prevented,—but that if the quantity be small, as in the mode practised by Roux, it will cause little local injury and readily enter the absorbing vessels.[690] Another unequivocal cause is pointed out by Harles in his treatise on arsenic. While treating of its therapeutic properties, and noticing the controversy that prevailed last century throughout Europe respecting the propriety of its outward application, he remarks that it may be applied with safety to the abraded skin, to common ulcers, to wounded surfaces, and to malignant glandular ulcers, even when highly irritable, provided the part be not recently wounded, so as to pour out blood.[691] The reason of this is obvious; the application of the poison to open-mouthed vessels is the next thing to its direct introduction into a vein. It is some confirmation of Harles’s opinion, that Roux, whose patient was so easily affected, recommends that before arsenic is applied to an ulcer, a fresh surface be made by paring away the granulations; and that Küchler’s patient had an ulcer which did not discharge pus, but serum, and was easily made to bleed.