lish foot contains 304 mm., which is usually held to differ slightly from that in vogue in the United States. Until there was a great deal of national and international intercourse the need of some uniform standard of weights and measures was not seriously felt; consequently the efforts of physicists in the seventeenth and eighteenth centuries did not receive much encouragement. Absolute accuracy in matters of this kind is unattainable, but in practical affairs it is not particularly difficult. What the term "accuracy" means to a maker of instruments of precision is forcibly illustrated by an anecdote told of John A. Brashear, of Pittsburgh. A prospective customer once asked him what it would cost to have a bar of glass made that was absolutely straight. Mr. Brashear would not promise absolute straightness, but was willing to come as near as he could for two hundred thousand dollars. After listening to a lecture on absolute accuracy by the renowned mechanician the customer concluded that his needs would be supplied by a ruler that would be correct to the one sixty-fourth of an inch and costing about forty dollars.
Physicists became convinced long ago that the only fixed standard of linear measure is some portion of the earth's circumference. No intelligent Greek or Roman from the time of Plato had any doubts about the shape of the earth. But after the Bible had come to be recognized as an authority in science as well as in doctrine the belief was gradually abandoned and various theories took the place of the true one until the time of Copernicus. Archimedes, about 200 b.c. used an ingenious argument to prove the sphericity of our planet. As water always seeks the lowest level the ocean must be equally deep everywhere and the bottom equally distant from a central point. As this is possible only in the case of a sphere, the earth must be spheroidal in form. The first attempt to calculate the circumference of the earth was made by the celebrated savant Eratosthenes in the third century B.C. Observing that the difference of latitude between two points in Egypt, Alexandria and Syene, was 7° 12′ and supposing them to be on the same meridian, and having ascertained as best he could that they were about five thousand stades apart, he reckoned this to be the fiftieth part of the earth's circumference, which would accordingly be 250,000 stades. More than a century later Poseidonius estimated the distance between Rhodes and Alexandria, on the testimony of seamen, to be five thousand stades, or one forty-eighth part of the circumference. Putting the value of the stade at six hundred feet—authorities vary considerably on this point—both estimates must be considered a remarkably close approximation to the truth.
In 1525 Fernel measured the distance between Paris and Amiens with a wheel. Almost a century later Snellius discovered, or rather rediscovered, trigonometry, which greatly simplified geodesy. By this