do, namely, the plodding repetition of a set operation. In other cases the man does the work because he must use his intelligence all the time to produce the ever-changing application which is called for.
A steam shovel will transfer sand or gravel from the bank at a rate to defy the competition of men, so long as the task is to transfer it in the same way, or within narrow limits; but a machine to apply steam-power to the excavation of a trench for a sewer, through hard soil and under constantly varying conditions, is hardly imaginable. Here, therefore, we find the human power almost unrivaled.
On the other hand, we find in mills and factories machines which, as the saying is, "can almost talk." They perform very complicated operations with perfection, provided only the task is to be repeated without limit of time, in some process of manufacture. Hence the machines and the power are all the time invading the domain of intelligence, wherever the task can be put in the mechanical form, and made to comply with the mechanical conditions; the man stands by and supplies the intelligence at the points where the intelligence is still indispensable. Some machines seem more intelligent than some men, but no machine can "feed itself" with new material. The operations of the machine are often immeasurably more worthy of intelligence than the operation of going after more material and feeding it into the machine, but not always so; for the arrangement of the machine and the material, that the machine may do its work well, is often no trifling demand on intelligence.
I believe, therefore, that it is a correct statement of the case that, where the task calls for brute force only, the steam or other engine supplants the man, and that
