against coincidence are indeed great, but the cases appear to be correspondingly rare.

Darwin acknowledges that many supposed instances of inherited mutilation may be due to coincidence; and there is apparently no more reason for attributing inherited scars, &c., to any special form of heredity than to the effect of the mother's imagination on the unborn babea popular but fallacious belief in corroboration of which far more alleged instances could be collected than of the inheritance of injuries.

As an instance of the coincidences that occur, I may mention that a friend of mine has a daughter who was born with a small hole in one ear, just as if it were already pierced for the earring which she has since worn in it. I suppose, however, that no one will venture to claim this as an instance of the inheritance of a mutilation practised by female ancestors, especially as such holes are not altogether unknown or

inexplicable, though very rarely occurring low down in the lobe of the ear.1

Many cases are known of the inheritance of mutilations or malformations arising congenitally from some abrupt variation in the reproductive elements. In such cases as the one-eared rabbits, the two-legged pigs, the three-legged dogs, the one-horned stags, hornless bulls, earless rabbits, lop-eared rabbits, tailless dogs, &c., if the father or the mother or the embryo had suffered from some accident or disease which might plausibly have been assigned as the cause of the original malformation, these transmitted defects would readily be cited as instances of the inheritance of an accidentally-produced modifica

tion.

The inheritance of exostoses on horses' legs may be the inheritance of a constitutional tendency

1 See pp. 179-182, Evolution and Disease, by J. Bland Sutton, to whom and to our mutual friend Dr. D. Thurston I am indebted for information on various points.

rather than of the effect of the parents' hard travelling. Horses congenitally liable to such formations would transmit the liability, and this might readily be mistaken for inheritance of the results of the liability. An apparent increase in this liability might arise from greater attention being now paid to it, or from increased use of harder roads; or a real increase might be due to panmixia and some obscure forms of correlation.

QUASI-INHERITANCE.

Of course artificially-caused ill-health or weakness in parents will tend in a general way to injure the offspring. But deterioration thus caused is only a form of quasi-inheritance, as I should prefer to call it. Semi-starvation in a newborn babe is not truly inherited from its half-starved mother, but is the direct result of insufficient

1 Variation of Animals and Plants under Domestication, ii. 290; i. 454.

nourishment. The general welfare of germs-as of parasites is necessarily bound up with that of the organism which feeds and shelters them, but this is not heredity, and is quite irrelevant to the question whether particular modifications.

transmitted or not.

are

Another form of quasi-inheritance is seen in the communication of certain infections to offspring. Not being transmitted by the action of the organism so much as in defiance of it, such diseases are not truly hereditary, though for convenience' sake they are usually so described.

A perversion or prevention of true inheritance is also seen in the action of alcohol, or excessive overwork, or any other cause which by originating morbid conditions in individuals may also injure the reproductive elements.

These forms of quasi-inheritance are, of course, highly important so far as the improvement of the race is concerned. So, too, is the fact that

improved or deteriorated habits and thoughts are transmitted by personal teaching and influence and are cumulative in their effect. But all this must not be confounded with the inheritance of ac

quired characters. Cases of quasi-inheritance may perhaps be most readily distinguished from cases of true inheritance by the time test. When a modification acquired in adult life is promptly communicated to the child in early life or from birth, it may rightly be suspected that the inheritance, like that of money or title, is not truly congenital, but is extraneous or even anti-congenital in its nature. Judged by such a standard, the inherited injuries in Brown-Séquard's guinea-pigs are only exceptional cases of quasiinheritance, and are not necessarily indicative of any general rule affecting true inheritance.