VOL. IV.] PHILOSOPHICAL TRANSACTIONS. 345 



day, added to the time of the clock, gives the true time by the sun. And when 

 after some days, you desire to know by the watch the time by the sun, add to 

 the time shown by the watch, the equation of that day ; and the aggregate will 

 be the time by the sun, if the watch has been well adjusted after the measure of 

 the mean days ; for the doing of which, this will be a convenient way : Draw a 

 meridian line on a floor, then hang up two plummets, each by a small thread or 

 wire, directly over that meridian, at about 1 feet from each other. When the 

 middle of the sun, the eye being so placed as to bring both the threads into 

 one line, appears to be in the same line exactly, then immediately set the watch, 

 not precisely to the hour of 12, but by so much less as is the equation of that 

 day by the table. Ex. gr. If it were the 12th of March ; the equation of that 

 day being, by the table, 8 min. 3 sec.; these deducted from 12 hours, the re- 

 mainder will be 11 hours 51 min. 57 sec; to which hours, minutes and seconds 

 set the index of the watch : Then after some days observe again in the same 

 manner, and likewise note the hour, min. and sec. of the watch ; to which add 

 the equation of these days taken out of the table ; and if the sum make just 12 

 hours, the watch is adjusted to the right measure; but if it differ, then divide 

 the minutes and seconds of that difference by the number of the days between 

 the two observations, to get the daily difference. Suppose this second observa- 

 tion to have been made the 20th of March, viz. 8 days after the first, and find- 

 ing that the middle of the sun being seen in the meridian in the same line with 

 the two threads, the watch points out llh. 51m. 7sec. ; now the equation of 

 the 20th of March, by the table, is 10m. 40 sec, which being added to the 

 time shown by the watch, gives 12h. Im. 47 sec If this had been just 12 

 hours, the watch would have been well adjusted, but being 1 min. 47 sec. more 

 than 12, it has gone so much too fast in 8 days : and these 1 min. 47 sec that 

 is 107 sec being divided by 8, there comes 134- sec. ^or the difference on every 

 24 hours ; which difference being known, if you want time, or have no mind 

 to take the pains to adjust the watch, note only the daily difference, and regu- 

 late yourself accordingly. But if you would adjust it better, remove the less 

 weight of the pendulum a little downwards, which will make it go slower ; and 

 then observe anew by the sun as before. If it had gone too slow, you must have 

 removed the weight a little upwards. And this is of such importance in find- 

 ing the longitude, that if it be not observed, you may sometimes in the space 

 of three months misreckon 7 degrees or more, and yet without any fault in the 

 watches ; which under the tropics will amount to above 400 English miles. 



Having shown how the watches may be adjusted at land, or how their daily 

 difference may be known, next follows how the same may be done, when a 

 vessel rides at anchor, it being hardly practicable when she is under sail. In 



VOL. I. X X 



