In reality the business is far more complicated than this. In reality a country will elect all sorts of different people according to the electoral method employed. It is a fact that anyone who chooses to experiment with a willing school or club may verify. Suppose, for example, that you take your country, give every voter one single vote, put up six and twenty candidates for a dozen vacancies, and give them no adequate time for organisation. The voters, you will find, will return certain favourites, A and B and C and D let us call them, by enormous majorities, and behind these at a considerable distance will come E, F, G, H, I, J, K, and L. Now give your candidates time to develop organisation. A lot of people who swelled A’s huge vote will dislike J and K and L so much, and prefer M and N so much, that if they are assured that by proper organisation A’s return can be made certain without their voting for him, they will vote for M and N. But they will do so only on that understanding. Similarly certain B-ites will want O and P if they can be got without sacrificing B. So that adequate party organisation in the community may return not the dozen a naïve vote would give, but A, B, C, D, E, F, G, H, M, N, O, P. Now suppose that, instead of this arrangement, your community is divided into twelve constituencies and no candidate may contest more than one of them. And suppose each constituency has strong local preferences. A, B and C are widely popular; in every constituency they have supporters, but in no constituency does any one of the three command a majority. They are great men, not local men. Q, who is an unknown man in most of the country, has, on the contrary, a strong sect of followers in the constituency for which A stands, and beats him by one vote; another local celebrity, R, disposes of B in the same way; C is attacked not only by S but T, whose peculiar views upon vaccination, let us say, appeal to just enough of C’s supporters to let in S. Similar accidents happen in the other constituencies, and the country that would have unreservedly returned A, B, C, D, E, F, G, H, I, J, K and L on the first system, return instead O, P, Q, R, S, T, U, V, W, X, Y, Z. Numerous voters who would have voted for A if they had a chance vote instead for R, S, T, etc., numbers who would have voted for B, vote for Q, V, W, X, etc. But now suppose that A and B are opposed to one another, and that there is a strong A party and a strong B party highly organised in the country. B is really the second favourite over the country as a whole, but A is the first favourite. D, F, H, J, L, N, P, R, U, W, Y constitute the A candidates, and in his name they conquer. B, C, E, G, I, K, M, O, Q, S, V are all thrown out in spite of the wide popularity of B and C. B and C, we have supposed, are the second and third favourites, and yet they go out in favour of Y, of whom nobody has heard before, some mere hangers-on of A’s. Such a situation actually occurs in both Ulster and