[91] That this is the case with the atmosphere at Martinico is now determined by a letter from Dr. George Davidson to Dr. Mitchell of New York, inserted in the Medical Repository, vol. ii, p. 279. With equal parts of nitrous and atmospheric air there was an absorption of 67 parts out of 100; but when two parts of atmospheric air were used to one of nitrous, the absorption was only from 52 to 58 parts; with a mixture of iron filings and sulphur, upwards of four tenths of the air were absorbed. These experiments were attested by a number of medical gentlemen who were present. In a letter subjoined from Dr. Chisholm, he says, that, having made a trial with iron filings and sulphur, the absorption was forty parts of an hundred, or exactly four tenths, with the eudiometer fifty-six. “It appears to me (says Dr. Chisholm) to be a singular circumstance, that, although the ground on which the Ordnance Hospital Hands is a perfect morass, partially drained, yet a result almost exactly similar to that given by the experiments made with the eudiometer at my house, should take place, with the same instrument and in circumstances very different. The proportion at the Ordnance Hospital, I think, has been 58 out of 100, and at your house, a situation less swampy, and nearer the sea, it has been 67. An explanation of so singular a result, in situations so different, is perhaps more to be wished than expected.” From this discovery it appears, that, whatever may be the cause of the frequency and violence of epidemics in warm climates, it _is not_ the want of oxygen. Nay, we should rather be tempted to think that they were produced by too great an abundance of it; and this the more especially when we know that animals confined in oxygen air are supposed to die of a burning fever; and it is likewise known that this kind of air is prejudicial to consumptive people, and even brings on the disease on those who had it not before. From the experiments mentioned in the note, it seems probable that there are but few even of swampy places in hot climates, where oxygen does not predominate; and in these the heat thrown into the blood must still be augmented by that produced from the quantity of vapour decomposed or condensed in the lungs, which, as the condensation depends upon unknown circumstances, can never be foreseen, or ever prevented, but by a removal from the place.