There is evidence that the bad consequences of self-fertilization and close interbreeding may almost fail to appear under favourable conditions of life. In-and-in bred plants, when allowed enough space and good soil, frequently show little or no deterioration: whereas, when placed in competition with another plant, they often perish or are much stunted.[2020] Crampe’s experiments with brown rats proved that the breeding in-and-in was much less injurious, if the offspring of the related parents were well fed and taken care of, than if it was otherwise.[2021] And this is in striking accordance with Dr. Mitchell’s observations as to consanguineous marriages in Scotland. The results there appear to be least grave, and are frequently almost _nil_, if the parents and children live in tolerable comfort, without anxiety or much thought for the morrow, and easily earning enough to procure good food and clothing—in short, when they work, but do not struggle for existence. On the other hand, when they are “poor, pinched for food, scrimp of clothing, badly housed, and exposed to misery; when they have to toil and struggle for the bare necessaries of life—never having enough for to-day and being always fearful of to-morrow,”—the evil may become very marked.[2022] If this is the case, we must expect to find that consanguineous marriages are much more injurious in savage regions, where the struggle for existence is often very severe, than they have proved to be in civilized society, especially as it is among the well-off classes that such marriages occur most frequently.[2023] In England, according to Mr. G. H. Darwin, cousin-marriages among the aristocracy are probably 4½ per cent.; among the middle and upper middle class, or among the landed gentry, 3½ per cent.; but in London, comprising all classes, they are probably only 1½ per cent.[2024] He thinks that the slightness of the evils which he found to result from first-cousin marriages perhaps depends upon the fact that a large majority of Englishmen live under what are on the whole very favourable circumstances.[2025] We must also, however, remember that there has been a great mixture of races in Europe, and that this necessarily makes marriage of kinsfolk less injurious, so far as the evil results of such unions depend upon too great a likeness between the sexual elements. The conclusion that closely related marriages produce more destructive effects among savage than civilized peoples, derives perhaps, some additional probability from certain ethnological facts. These facts may, at least, serve to show that such marriages, and the experience of isolated communities, are not everywhere in favour of Mr. Huth’s conclusions. Several statements on the subject have, indeed, scarcely any value as direct evidence for the harmfulness of consanguineous marriages, but to two or three considerable weight must be attached.