true length for that temperature. Is has lately been shown, however, by the experiments of Professor Rogers, that if two steel bars, one of which is nickel-plated, be subjected to a gradual change of temperature, they will acquire their true length after the temperature has been maintained constant for about twelve hours; but, if the change be an abrupt one, it is not safe to compare them until after the lapse of from forty-eight to sixty hours.
Enough has been said to indicate what great precautions must be taken in order to obtain accurate copies of a given standard of length. We may now consider how the standard measures at present in use were originally obtained, and how they are related to each other. We will confine our attention to the measures of France and England, since these possess more interest for us than do the measures of other nations, with which we are less familiar.
It is quite generally supposed that the length of a yard exactly corresponds to the length of a pendulum beating seconds of time, in a vacuum, in the latitude of London, at the sea-level. This, however, is not the fact. The act of Parliament in relation to this matter has been generally misunderstood, for it does not declare the length of the yard to be absolutely that of the pendulum; in truth, these lengths are not the same. Parliament only provided that, in case the original standard should be lost, it could be restored by reference to the unit pendulum. The standard that was legalized was made by Bird, from Graham's scale, in the year 1760. It was named the "imperial standard yard."
According to experiments conducted at that time, it was found that the relation between the length of the standard imperial yard and that of a seconds-pendulum was in the proportion of thirty-six inches to thirty-nine inches and 10000 of an inch. On October 16, 1834, both Houses of Parliament were destroyed by fire, and, although the imperial yard was found in the ruins, it had become unfit for use as a standard. The problem of its restoration was then presented, but since the passage of the act of 1824, which declared the relations between the pendulum and the lost standard, it had been found that the data from which the relations were calculated were, in several respects, unreliable. It was finally decided not to attempt the restoration of the lost standard by means of the pendulum, but to work from the various standards which had been compared with it. For this purpose six different scales were found available, among which was the tubular scale belonging to the Royal Astronomical Society; but this scale was not the principal authority from which the new standard was constructed, although it is so asserted in both Appletons' and Johnson's Cyclopædias. The scales actually made use of were two by Shuckburgh, one by Kater, that belonged to the Royal Society, and two bars of the Ordnance Department. The work of renewing the standard was intrusted to Sir Francis Baily, but he died