spersed with others less luminous, indicating thereby the degree of force with which they respectively bear.
In getting up a surface of considerable extent, it is necessary to take into account the strain which the metal suffers from its own weight, and the length of time required to produce the full effect on the external form. It will be found, for example, that after a piece of metal has remained for some days in one position undisturbed, it assumes a form different from that which it had while undergoing preparation. Hence, it is desirable to provide for the work while in hand, similar support to what it will have when applied to its intended use.
Another disturbing cause is the unequal contraction of the metal in cooling, when originally cast. The mass assumes the curved form, and is pervaded by elastic forces counteracting each other. These continue in permanent activity, and any portion of metal, taken from any part, tends to disturb the balance previously established.
It remains to consider the second case proposed, viz., how to prepare an original surface. A brief description of the proper method will still further illustrate the case already considered, and will also show how surface plates are to be corrected.
