As to civilized societies, our comparative ignorance of social history is such as to preclude inductive reasoning concerning the greater part of it. This cannot be too often or too emphatically stated. I have even gone so far as to say bluntly, in my first note, that researches based on historical evidence alone cannot claim to be exhaustive. In saying this, I am thinking particularly of social regulations and conventions; and I confess that I view with alarm the current habit, deplorably widespread among historians and antiquarians, of assuming that the regulations and conventions that prevailed in a century of which we have direct knowledge prevailed also in a preceding or in a succeeding century, of which we may have no direct knowledge at all. Whenever our knowledge is complete, we find that in any vigorous society the method of regulating the relations between the sexes was constantly changing; and, unless there is direct evidence, it is wrong to assume that in any such society social laws were ever static and unchanging, even for three generations.