(232)
Curious as much Information, as may be, in reference to Baroscopes; I shall venture to send you some Account of what I did but name (in my former Letter) to you.
Though by a Passage, you may meet with in the 19th and 20th Pages of my Thermometrical Experiments and Thoughts, you may find, that I did some years agoe think upon this New kind of Baroscope; yet the Changes of the Atmosphere's Weight not happening to be then such, as I wish'd, and being unwilling to deprive my self of all other use of the exactest Ballance *, * The Scales here meant were before competent Eye-witnesses made to turn manifestly with the thousandth part of a grain. that I (or perhaps any man) ever had, I confess to you, that successive avocations put this attempt for two or three years out of my thoughts; till afterwards returning to a place, where I chanc'd to find two or three pairs of Scales, I had left there, the fight of them brought it into my mind; and though I were then unable to procure exacter, yet my desire to make the Experiment some amends for so long a neglect, put me upon considering, that if I provided a Glass-buble, more than ordinary large and light, even such Ballances, as those, might in some measure perform, what I had tried with the strangely nice ones above-mention'd.
I caused then to be blown at the Flame of a Lamp some Glass-bubles as large, thin and light, as I could then procure, and choosing among them one, that seem'd the least unfit for my turn, I counterpoised it in a pair of Scales, that would loose their Æquilibrium with about the 30th part of a Grain, and were suspended at a Frame. I placed both the Ballance and the Frame by a good Baroscope, from whence I might learn the present weight of the Atmosphere. Then leaving these Instruments together; though the Scales, being no nicer than I have express'd, were not able to shew me all the Variations of the Air's weight, that appear'd in the Mercurial Baroscope, yet they did what I expected, by shewing me variations no greater, than alter'd the height of Quicksilver half a quarter of an Inch, and perhaps much smaller, than those: Nor did I doubt, that, if I had had either tender Scales, or the means of supplying the Experiment with convenient accommodations, I should have di-
