This is the way in which much of our knowledge is collected. Edison found the right substance for his first carbon filament by sending for all sorts of materials from all over the world, carbonizing them, and trying them out. The right one proved to be a kind of bamboo. If Edison had hit on this at the first trial it would have been so “lucky” a chance as almost to be counted a miracle; as it was, he eliminated chance by multiplication. Nothing annoys an executive so much as to be told that the adoption of this or that course will result in a specified way, when no one has ever tried it. This was a common attitude in the time of Galileo, when the idea that anything could be found out by observation or experiment was regarded as a public scandal. That was the time when a man refused to look through the newly-invented telescope for fear that he might see something contrary to the teachings of Aristotle. These people are not all dead by any means. I have heard them assert that a proposed change would ruin the library and then object to trying it because they were afraid the result would be contrary to their own predictions. The medieval philosophers at least had Aristotle to fall back on; their modern successors would appear to be posing as Aristotles themselves. A housemaid recently said to her mistress “I’ve told everybody to-day ye weren’t at home; now don’t sit in the window and make me a liar.” No discovery; no falsehood, you see. So if we librarians can be prevented from trying experiments, the false predictions of some of our advisers will not be false in their own eyes, simply because they will not be exposed. My advice to librarians, and to everyone else is to keep on trying experiments. If you get a satisfactory result the first time, you may stop, and ascribe it, if you please, to your good luck. If the result is unsatisfactory, however, you need not stand pat on your ill luck. “If at first you don’t succeed Try, try again”. There is more philosophy in that than in all Aristotle. It is also a practical exposition of the doctrine of chances. Somewhere is the combination that you want. You will find it, if you only keep on long enough. Libraries that are afraid of being victimized by chance, or, as we may put it, becoming martyrs to bad luck, should ponder somewhat more closely the possibilities of relief from insurance. Of course here I am using the word “luck” in its simpler meaning of unforeseen occurrence. Take the case of the library that suffers from the fact that an influential member of the committee that fixes the amount of its annual appropriation has eaten something indigestible for breakfast. Such an unforeseeable occurrence, such a “piece of bad luck”, might cost a library anywhere from two to twenty thousand dollars, according to the usual size of its appropriation.