I wrote and asked Mr. Sinclair if Mrs. Sinclair was told the fact that this and several other series of original drawings consisted of geometrical drawings, letters and figures, and he said that she was not so told, that he would have regarded this as a vitiation of the experiments. It would certainly increase the chance of getting drawings right by guess, but it would hardly have ruined the experiments. In fact, some people think that the most scientific experiments are those in which the range of chance guess is limited to an extent known to the percipient, as when the problem is to determine which of the 52 cards of a pack is being looked at, or which of only ten known diagrams. This opinion is probably based on the fact that then the ratio of success to chance expectation can be exactly calculated, though why it should be more satisfactory to know that the chance of a correct guess is exactly 1 in 10 than it is not to be able to tell exactly what the chance is but to be sure at least that it cannot be 1 in 100, I do not know. Unless I had carefully recorded at the time that there was no chance of the percipient having a hint that the drawings were now for a time to consist of geometrical designs, letters and figures, I would not dare to be certain of it after several years have passed. If Mrs. Sinclair had no inkling, the change in the general character of her drawings is a fact of great interest. But we will take cognizance only of whatever resemblance may or may not be found between the several reproductions and their originals. The first series of drawings by the secretary were seven in number, and, says Mr. Sinclair, “They brought only partial successes; Craig would get elements of the drawing, but would not know how to put them together.... There is some element right in every one.” Let us see. 1. _Agent’s_ drawing, a script B; _Percipient’s_ drawing, a figure very like a script 3, practically the B without its vertical line. 2. _Agt._, a script S; _Per._, a script J. As made, each has two balloon-like parts joined at the small ends, certain details of course different. 3. _Agt._, a hexagon; _Per._, two lines forming an acute angle, like two sides of the hexagon, also a capital E with a line drawn down at an acute angle to the left from the upper extremity of the vertical line. 4. _Agt._, script M made with a peculiar twist in its first line; _Per._, almost precisely that first line with its twist. 5. _Agt._, a thin, long, quadrilateral, like a shingle; _Per._, (1st drawing) what would be almost exactly the same quadrilateral, narrow and long, but its shorter sides are wanting, and (2nd drawing) a closely similar quadrilateral, with another and longer one attached to its side at a sharp angle.