Ernest William Brown FRS (29 November 1866 – 22 July 1938) was an English mathematician and astronomer, who spent the majority of his career working in the United States and became a naturalised American citizen in 1923.

After leaving school, he entered Christ's College, Cambridge, where he graduated with first-class honours as sixth Wrangler in mathematics in 1887.

[6] At Haverford, Brown continued with his studies of the lunar theory, and made a thorough review of the work of earlier researchers, such as Hill, de Pontécoulant, Delaunay and Hansen.

His mastery of the field was shown by the publication of his first great work, An Introductory Treatise on the Lunar Theory,[7] in 1896, when Brown was still less than 30 years of age.

In 1907, he was appointed Professor of Mathematics at Yale University, with which he secured an agreement for funding the massive task of calculating detailed tables of the Moon's motion, based on his lunar theory.

He also included perturbations due to the other planets (principally Jupiter and Venus) and also accounted for the more difficult problem of the non-spherical nature of the Earth and Moon.

Observations showed that Brown's tables were indeed superior to those of Hansen, which had been in use since 1857, but there was still a large unexplained fluctuation in the Moon's mean longitude of the order of 10 arcseconds.

The implication of this was that it was not the Moon that was speeding up – it was time (as measured in terms of Earth's increasingly long day) that appeared to be slowing down.

Brown devoted much study to this problem and proposed it should be attacked observationally, using lunar occultations to map the Moon's path more precisely.

However, he eventually concluded that Newcomb was right, and not only was the Earth's rate of rotation slowing, but also there were random, unpredictable fluctuations, and he published these findings in a paper in 1926.

After having fortified himself with strong coffee from a thermos bottle he set to work without leaving his bed, smoking numerous cigarettes.

This also gained some improvement in precision, since the tables had embodied some minor approximations, in a trade-off between accuracy and the amount of labour needed for computations in those days of manual calculation.

[13] By the middle of the 20th century, the difference between Universal and Ephemeris Time had been recognised and evaluated, and the troublesome empirical terms were removed.