PROPORTIONAL REPRESENTATION.

I To THE EDITOR OF THE. " SPECTATOR.")

quite agree with your correspondent, "The Author of The Party Vote,'" that curious results may happen under any scheme of proportional representation that makes use of the arbitrary quota found by adding one to the quotient obtained on dividing the whole number of votes cast at the election by one more than the number of seats to be filled. It is quite true than any candidate who can command this quota of votes must be returned, but the error is in assuming that he cannot be returned by less. If there are one or two independent candi- dates on whom votes are wasted, the real quota is less than that given by the rule.

But may I point out to your correspondent that in any system—such as that you allowed me to -describe a few weeks back—under whioh the electors may vote for any one candidate, or for a set (or party), but not for several candidates of different parties, or under which, as in the example he puts, the voters all "give a strictly party vote,—that is, always give their transfer votes to members of their own pfixty,"—the arbitrary quota need not be employed ? It is the cross transfers that make it necessary. I may take his example :-- Three Liberal Candidates, with 6,000 votes between them.

Four Conservatives, „ 8,200 „ „ One Radical, „ 3,000 „

One Independent, „ 1,300 „

It is clear, on these figures, that by no arrangement could any candidate score more than 8,200 votes.

One Conservative candidate (if only one were

returned) could command 8,200 votes.'

Two (if two be returned) could command 4,100 each. Three (if three be returned) do. 2,733 each.

Four (if fonr be returned) do. 2,050 each.

One Liberal do. 6,000 votes.

Two Liberals do. 3,000 each.

Three Liberals do. " .2,000 each.

The one Radical can command 3,000 votes.

The one Independent do. 1,300 votes.

Arrange these numbers in descending orders of magnitude thus, —8,000, 6,000, 4,100, 3,000, 3,000, 2,733, 2,050, 2,000, and 1,300. There are just so many of them (nine) as there are candidates in the field, and they give the quotas to return one, two, three, four, five, six, seven, eight, or nine candidates. There is no difficulty in applying this mode of calculation where separate (as well as' joint) votes are allowed ; but I need not trouble you with further details, that belong specially to the particular system I described.

May I add that this system—that of allowing candidates to stand in quasi-partnership—is not open to the slur you suggest on the use of cast-off votes. The other objection you mention, to any system of proportional representation, that it may fail at a bye-election to supply the place of a minority Member, has never, I confess, had much weight to my mind. It has already appeared that, partly from a sense of fair-play, which the system helps to encourage, and partly from the difficulty of finding a decent candidate of the party of the majority willing to contest a seat he will probably forfeit at the next General Election, the seat will often remain with the minority. But suppose they do lose their representation for a time, is that a reason they should never have one ? Let us do justice in this imperfect world where we can.—I am, Sir, &c.,