Newton’s experimental fudges

In the 1600s, estimates for the speed of sound in air ranged from 180 ms^-1 to 450 ms^-1. Newton set out to determine the speed of sound himself and made a measurement in the cloister (covered walk) of Trinity College, Cambridge, adjusting a pendulum until it swung in time with a returning echo. In the 63 m cloister, he found that the echo was slower than the swing of a 14 cm pendulum but quicker than a 20 cm pendulum. This placed the speed of sound between 280 and 331 ms^-1. However, based on a comparison of densities, Newton had also estimated the speed of sound in air to be 295 ms^-1. He subsequently amended his experimental data arguing the echo was slower than a 13 cm pendulum but faster than one of 18 cm. The historian of science, Richard Westfall, has noted that this is one of a number of occasions on which Newton fudged the data to fit his preconceptions.