England village, and none at all in a Prussian barrack. A familiar illustration of the restrictions on his experience of it is to be found in the rule which compels the calling of "experts" when there is a question as to any point of science or art. "The words science or art," says Mr. Fitzjames Stephen, "include all subjects on which a _course of special study or experience is necessary to the formation of an opinion_," and the opinion of such an expert is a "relevant fact." So that Dr. Taylor's "distinguished legal friend," if a good lawyer, would not, in spite of his proficiency in circumstantial evidence, undertake to dispute with Professor Huxley about the relation of the anchitherium, hipparion, and horse; and if Dr. Taylor offered himself for examination on such a point he would be laughed out of court. In none of our courts is the presentation allowed of _all_ the circumstances which strengthen or weaken a probability. A lawyer, therefore, though he might not be as ill fitted for a scientific discussion as a minister, is, _as such_, hardly more of an authority on the force and limits of that portion of scientific proof which is drawn from simple observation. Dr. Taylor's consulting one as a final authority as to the very nature of the argument on which he was himself about to sit in judgment is at the outset a suspicious incident. The definition of circumstantial evidence which he got from his legal friend was this: "The process of proof by circumstantial evidence consists in reasoning from such facts as are known or proved, and thence establishing such as are conjectured to exist. The process is fatally vicious, first, if any material circumstance from which we seek to deduce the conclusion depends itself on conjecture; and, second, if the known facts are not such as to exclude to a reasonable degree of certainty every other hypothesis." "Now, tried by these two tests," says Dr. Taylor, "the professor's argument was a failure." Taking this definition as it stands, however, we think it will not be difficult to show that Dr. Taylor is not competent to apply the tests, or to say whether the professor's argument is a failure or not.