your consideration: What actual laws of nature have been discovered by the method of least squares?
The Mechanical Instruments of Research
A few minutes were to have been given to the instruments employed by the scientific man to sharpen and amplify his natural senses and sensations—in a word, the tools furnished him by the mechanician. I am glad, however, both for your sake and mine, that this part of the subject was covered by an interesting paper presented at the previous meeting of the society. It was emphasized there that for the best results it is essential that the investigator be able to work with instruments so constructed as to permit him to control or renew the various adjustments without the necessity of returning the instrument to the maker. The principle at times employed, which assumes that when adjustments are once made they are to "remain put," is apt to prove a very pernicious one. A number of very interesting examples from my own experience in the purchase of magnetic instruments during the past ten years might be cited; but, as has been said, this part of the subject having already been covered, there is no need to dwell further upon it than to emphasize the injunction that the research worker, if he desires the best results, must know his instruments as thoroughly as himself.
Subjects of Research
We come next to a brief consideration of the subjects of research, though not specifically mentioned in the title of our paper, yet implied in it. The rapid progress made by a science as soon as it reaches the stage of experimentation has already been noted. A crucial experiment has at times furnished information which by mere observation of phenomena, running their natural and unmodified course, might either have never come into our possession or at best would have taken a considerably longer time than that of the decisive experiment. You are all familiar with such cases, for almost every science can furnish examples.
Now it is an extremely interesting and suggestive fact that the greatest experimental discoveries to-day are not made in the older, well-recognized sciences, but on their borderlands—in the "twilight zones" of more or less related sciences. I have but to mention the words "physical chemistry," "physical geology," "astrophysics," "biochemistry," etc., and you will readily grant the assertion made. In the overlapping regions there seem to be the greatest opportunities afforded for solid, thorough, and at the same time remarkably rapid, experimental achievements. And so we are having produced almost daily new specialties or new sub-specialties.
