It is certain, however, that such a method would open a door for party organisation and party action of a most formidable kind, by the means of settled and printed lists, giving an increased amount of influence to mere numbers acting together, which under the simpler system of regarding only the first, or first uncancelled name, and of appropriation according to localities, no party ticket could secure, but which, on the contrary, any such ticket would rather impair.
It will be interesting, however, to refer to the different result of this case, as it would be determined if the improved rule for obtaining the quota, examined in Appendix E, were applied. The last example there put (p. 307) shows that B would be returned with A and C, without any of the dangerous consequences of giving overwhelming power to numbers, which might be feared from the foregoing method of valuation.
Appendix E.
Among those,—Earl Russell, Mr. Andriæ, Mr. James Garth Marshall, Earl Grey, and others, who within the last ten years have sought a remedy for the defects in representative institutions which have made them the creatures of a number of dispersed majorities,—must be included the name of Mr. Droop, of Lincoln's Inn, and late Fellow of Trinity College, Cambridge. Mr. Droop, “on observing the rather peculiar phenomena of the general election in 1857, directed his attention to the problem how representatives might be elected who should best represent the whole constituency, and not merely the majority.” Perceiving that I had introduced contingent voting, which was also a part of his scheme, he has kindly commimicated to me his method of appropriation of every vote to the candidate for whom it is successively designed, and of arriving at a quota which should be exactly the nmnber of votes that ought to be retained for each successful candidate in order to ensure his election. The accurate determination of this quota, Mr. Droop remarks, would require the solution of an algebraical equation of a high degree, but there is no difficulty in approximating to it to any required degree of accuracy. One step of this process may be adopted with great advantage. It is the result of an observation that the quota proposed to be adopted cannot be greater than the quotient produced by dividing the aggregate number of votes polled for the number of candidates to be elected, who stand highest on the poll, by such last-mentioned number. Thus, if there were 2,000 candidates, all polling more or less votes, of whom only 654 can be elected, it is not necessary to take as the dividend the whole number of votes polled by the 2,000, but it is sufficient to take the number of votes polled by the 654 who stand highest on the poll. The quotient thus obtained, which will be necessarily far smaller in number than that obtained by the process pre-
