ad been supposed.

Later, when Newcomb's investigations along this line had extended to the
major planets and their satellites, a curious anomaly in the moon's motion
made it necessary for him to look for possible observations made long
before those hitherto recorded. The accepted tables were based on
observations extending back as far as 1750, but Newcomb, by searching the
archives of European observatories, succeeded in discovering data taken as
early as 1660, not, of course, with such an investigation as this in view,
but chiefly out of pure scientific curiosity. The reduction of such
observations, especially as the old French astronomers used apparent time,
which was frequently in error by quarter of an hour or so, was a matter of
great difficulty. The ancient observer, having no idea of the use that was
to be made of his work, had supplied no facilities for interpreting it,
and "much comparison and examination was necessary to find out what sort
of an instrument was used, how the observations were made, and how they
should be utilized for the required purpose." The result was a vastly more
accurate lunar theory than had formerly been obtained.

During the period when Newcomb was working among the old papers of the
Paris Observatory, the city, then in possession of the Communists, was
beset by the national forces, and his studies were made within hearing of
the heavy siege guns, whose flash he could even see by glancing through
his window.

Newcomb's appointment as head of the Nautical Almanac office greatly
facilitated his work on the various phases of this problem of planetary
motions. Their solution was here a legitimate part of the routine work of
the office, and he had the aid of able assistants,--such men as G.W. Hill,
who worked out a large part of the theory of Jupiter and Saturn, and
Cleveland Keith, who died in 1896, just as the final results of his work
were being combined. In connection with this work Professor Newcomb
strongly advocated the unification of the world's time by the adoption of
an international meridian, and also international agreement upon a uniform
system of data for all computations relating to the fixed stars. The
former still hangs fire, owing to mistaken "patriotism"; the latter was
adopted at an international conference held in Paris in 1896, but after it
had been carried into effect in our own Nautical Almanac, professional
jealousies brought about a modification of the plan that