kindness to persons whom we have never seen or heard of before does exist. I should not scruple to charge any one who should deny this with mala fides, with prevaricating either to himself, or others. It is a maxim which these gentlemen seem to be unacquainted with that it is necessary to strain an hypothesis to make it fit the facts, not to deny the facts because they do not square with the hypothesis. It generally happens, that when a metaphysical paradox is first started, it is thought sufficient by a vague and plausible explanation to reconcile it tolerably well with known facts: afterwards it is found to be a shorter way, savouring more of a certain agreeable daring in matters of philosophy and dashing the spirit of opposition sooner, to deny the facts on the strength of the hypothesis. Independently however of all experimental proof, the reasoning as it is applied confutes itself. It is said that habit is necessary to produce affection. Now supposing this: in what sense is the principle true? If the persons, feelings and actions must be exactly and literally the same in both cases, there can be no such thing as habit: the same objects and circumstances that influenced me to-day cannot possibly influence me to-morrow. Take the example of a child to whose welfare the attention of the parent is constantly directed. The simple wants of the child are never exactly the same in themselves, the accidental circumstances with which they are combined necessarily varying every moment: nor are the sentiments and temper of the parent less liable to constant and imperceptible fluctuations. These subtle changes, however, this dissimilarity in subordinate circumstances, does not prevent the parent's affections for the child from becoming an inveterate habit. If therefore it is merely an extraordinary degree of resemblance in the objects which produces an extraordinary degree of strength in the habitual affection, a more remote
