This, to be frank, was my initial attitude—one of cautious interrogation and alertness to find signs of credulity, failure to appreciate the possibilities of chance, or lack of data by which the calculus of chance coincidence could be determined. But as I read on and studied the reproductions of drawings it became more and more evident that something besides chance had operated, that the conditions of many of the experiments had been excellently devised, and that where the conditions were relaxed Mr. Sinclair had been quite aware of the fact and was candid enough to admit it. He stated that such relaxation did not increase the percentage of success, and it certainly so appeared from the examples given. He reported the total number of experiments, and estimated the percentages of successes, partial successes, and failures. In 290 experiments, he made these percentages: successes, approximately 23 per cent; partial successes, 53 per cent; failures, 24 per cent. He admitted that judges probably would not agree upon exactly the same ratios. In fact I personally think that certain examples which he did not publish are better than a few which he did, but have not yet found reason to quarrel with his general estimates. After considerable study of the book, becoming interested beyond any expectation, I wrote to Mr. Sinclair, stating that I had become favorably impressed, and making the somewhat audacious proposal that he should send me all the original materials for a fresh study by the individual standards and through the particular methods of a professional investigator. One can think of several reasons which might make the most honest and confident man hesitant to assent to such a proposal, coming from one whom he had never seen, and who might for all he knew have a set of prejudices which after all would cause him to make a lawyer’s argument against the case. I was really surprised that the bundle of materials was sent as quickly as it could be gotten together.[2]