All the generic and specific forms of life are governed by their own peculiar laws of interunion, and hybridism or hybridity is therefore a phenomenon of varying character, having, it is true, certain resemblances in those instances which approach each other, but absolutely different in all cases. Naturalists have sometimes made great blunders in this respect, for they have assumed that hybridism was governed by the same laws in all cases, and therefore sought its application or inferred its presence in instances the most remote and contradictory. The most extraordinary, and, indeed, inexcusable instance of the kind has been seen in the efforts made to confound the distinctions of race, and to pervert truth into the most shameful and what would seem to be the most palpable falsehoods. It has been assumed by naturalists of high character that different genera never produce offspring, that the _offspring_ of different species are incapable of reproduction, and that varieties are unlimited in their powers of virility. If, therefore, there were doubt in respect to the character of certain (supposed) genera, and it was found that offspring followed a conjunction of sexes, in this particular instance, it was inferred that they were merely different species. And if the product or progeny of these _species_ were found to be equally virile, then it was inferred that they were all originally of the same species, and nothing but varieties. This test, so simple that it can hardly be mistaken, serves with sufficient accuracy to determine the real character, and when the naturalist properly applies the laws of hybridity, that is, admits a modification of these laws in all cases or in all the different genera subjected to his examination, then he is armed with sufficient data to render his labors accurate and effective. But however pains-taking or correct in other particulars, when he assumes that hybridity is a unit, and rigidly applies this in all cases, or to families widely remote in other respects, his labors, from this defect, must be comparatively valueless.