Reviewing the foregoing examples, it appears to me, contrary to expectation, that in passing from the strongest resemblances to those that are fainter, the pleasure rises gradually in proportion. Can this be accounted for? or shall I renounce my taste as capricious? When I renew the experiment again and again, I feel no wavering, but the greatest pleasure constantly from the faintest resemblances. And yet how can this be? for if the pleasure lie in imitation, must not the strongest resemblance afford the greatest pleasure? From this vexing dilemma, I am happily relieved, by reflecting on a doctrine established in the chapter of resemblance and contrast, that the pleasure of resemblance is the greatest, where it is least expected, and where the objects compared are in their capital circumstances widely different. Nor will this appear surprising, when we descend to familiar examples. It raiseth not wonder in the smallest degree, to find the most perfect resemblance betwixt two eggs of the same animal. It is more rare to find such resemblance betwixt two human faces; and upon that account such an appearance raises some degree of wonder. But this emotion rises to a still greater height, when we find in a pebble, an aggat, or any natural production, a perfect resemblance to a tree or other organised body. We cannot hesitate a moment, in applying these observations to the present subject. What occasion of wonder can it be to find one sound resembling another, where both are of the same kind? It is not so common to find a resemblance betwixt an articulate sound and one not articulate; and accordingly the imitation here affords some slight pleasure. But the pleasure swells greatly, when we employ sound to imitate things it resembles not otherwise than by the effects produced in the mind. I have had occasion to observe, that to complete the resemblance betwixt sound and sense, artful pronunciation contributes not a little. Pronunciation therefore may be considered as a branch of the present subject; and with some observations upon it I shall conclude the section. In order to give a just idea of pronunciation, it must be distinguished from singing. The latter is carried on by notes, requiring each of them a different aperture of the windpipe. The notes properly belonging to the former, are expressed by different apertures of the mouth, without varying the aperture of the windpipe. This however doth not hinder pronunciation to borrow from singing, as a man sometimes is naturally led to do, in expressing a vehement passion. In reading, as in singing, there is a key-note. Above this note the voice is frequently elevated, to make the sound correspond to the elevation of the subject. But the mind in an elevated state, is disposed to action. Therefore in order to a rest, it must be brought down to the key-note. Hence the term _cadence_.