From the year 1701 to 1756 inclusive, a variety of observations were made with thermometers, and the following were the results. The greatest degree of heat, and of cold, which was experienced at Paris in each year was collected; a total of these was made, and it was found that the mean estimate, in all the thermometers, reduced to Rheaumur's division, was 1026, for the greatest heat in summer, that is 26 degrees above the freezing point; and that the mean degree of cold in winter, during those 56 years, was 994, or 6 degrees below the freezing point of water, whence we concluded that the greatest heat in our summers at Paris differs from the greatest cold of our winters only 1/32, since 994 : 1026 :: 31 : 32; and it was on this foundation that we stated the latter to be the relation of the greatest heat to the greatest cold. But it may be objected against the precision of this valuation, the defect of the construction of the thermometer, and Rheaumur's division (to which we have here reduced the scale of all the rest); and this defect is extending only 1000 degrees below that of ice, as if 1000 degrees were in fact, that of absolute cold, whereas absolute cold does not exist in nature; and that of the smallest heat should be supposed 10,000 instead of 1000, which would alter the thermometer's gradation. It may likewise be said that it is possible all our sensations between the greatest heat and the greatest cold are comprised in as small an interval as that of a unit on 32 of heat, but that the voice of judgment seems to be raised against this opinion, and tells us this limit is too confined, and that it is much easier to reduce this interval than to give it an eighth, or a seventh instead of a thirty-second.