which it may be connected. Let us consult experience, the guide that ought always to be followed whenever it can be found.
The scheme of representation, as a substitute for a meeting of the citizens in person, being at most but very imperfectly known to ancient polity; it is in more modern times only that we are to expect instructive examples. And even here, in order to avoid a research too vague and diffusive, it will be proper to confine ourselves to the few examples which are best known, and which bear the greatest analogy to our particular case. The first to which this character ought to be applied, is the house of commons in Great Britain. The history of this branch of the English constitution, anterior to the date of Magna Charta, is too obscure to yield instruction. The very existence of it has been made a question among political antiquaries. The earliest records of subsequent date prove, that parliaments were to sit only, every year; not that they were to be elected every year. And even these annual sessions were left so much at the discretion of the monarch, that under various pretexts, very long and dangerous intermissions were often contrived by royal ambition. To remedy this grievance, it was provided by a statute in the reign of Charles II, that the intermissions should not be protracted beyond a period of three years. On the accession of William III, when a revolution took place in the government, the subject was still more seriously resumed, and it was declared to be among the fundamental rights of the people, that parliaments ought to be held frequently. By another statue which passed a few years later in the same reign, the term "frequently," which had alluded to the triennial period settled in the time of Charles II, is reduced to a precise meaning, it being expressly enacted, that a new parliament shall be called within three years after the determination of the former. The last change, from three to seven years, is well known to have been introduced pretty early in the present century, under an alarm for the Hanoverian succession. From these facts it appears, that the greatest frequency of
