By that standard, of course, very few governments in Canadian history could claim a mandate. The last time an Ontario party won more than 50 per cent of the popular vote was in 1934. It's not unknown for a party to win a majority of the seats with a smaller share than its nearest rival.

Such are the vagaries of the first-past-the-post electoral system, where the plurality, not the majority rules: In each riding, the winner is the candidate with the most votes, no matter how few. The relationship between the number of seats won and the overall popular vote will thus be tenuous. In a three-party system, in fact, it is theoretically possible to win a majority government with just over one- sixth of the popular vote: one-third of the vote in one-half of the seats. Even with the stipulation that each seat must be won with a majority of the votes cast in that riding, a party could form a majority government with about 25 per cent of the popular vote.

Such anomalies are often cited by supporters of the electoral system known as proportional representation, in which each party is allotted a number of seats equal to its share of the popular vote. The Conservatives, with 45 per cent of the vote, would get 45 per cent, the Liberals would get 31 per cent, the NDP 21 per cent, and the fringe parties would share the remaining 3 per cent. Members of the legislature would be drawn from party lists, rather than represent a particular constituency.

Almost no one, however, proposes a system of pure proportionality: not after the experience of Israel, where the smallest sliver of support entitled a party to representation in a Knesset that suffered increasingly from gridlock. Moreover, a purely proportional system leaves voters without local representation and members without local accountability.

So most schemes envisage a mix of proportional and constituency representation, such as the half-and-half model in use in Germany. In such a remodelled legislature, the Tories, having won 63 per cent of the constituency- based seats and 45 per cent of the proportional- representation seats (based on popular vote), would still have a clear majority.

The other contender among reform proposals is the transferable ballot, as practiced in Australia. Voters do not mark an x beside the name of one candidate, but rather rank all candidates numerically in order of preference. When the ballots are counted, the votes for the last-place candidate are redistributed among the remaining candidates according to the last-place voters' second choices. Then the next-to-last candidate is knocked out, and his votes redistributed in the same manner; then the next-lowest candidate, and the next, until one candidate has a majority.

How would this approach affect the party standings in Ontario? In exactly half of the Conservatives' seats, 41 in all, the party won more than 50 per cent of all votes cast. The Liberals won 8 seats by an absolute majority, while the NDP took 1. No change here. (By contrast, of the 74 seats the NDP won in 1990, just 25 were by a majority of the votes.)

In another 12 seats, the Tories were denied an outright majority only by the votes cast for the Family Coalition Party, the Confederation of Regions Party, the Freedom Party or the Libertarian Party, whose conservative supporters could be presumed to rank the Tories second on their ballots. So the Tories start from a solid base of 53 seats in the 130-seat house.

What would happen in ridings where it was more of a two-horse race and the reapportioning of the third-place candidate's votes could make a difference? Suppose that two-thirds of Liberal voters would make the Tories their second choice, while two-thirds of Tory votes would go to the Liberals - based on the percentage of the electorate that told pollsters they would not vote NDP under any circumstances. Assume further that all NDP voters would mark the Liberals as their second choice.

The Tories finished second to the NDP in seven ridings. In two of these, a reapportioning of the third-place Liberal vote under a transferable ballot would have given the Tories a win under these assumptions.

In addition, the Tories finished second in one riding to independent Peter North, a defector from the NDP. Presumably those who voted for the fourth-place NDP candidate would feel little amity towards Mr. North. In which case the second choices of the third-place Liberal and NDP voters would together be enough to put the Tories over the top here too.

In six more seats where the Tories finished first, but did not win a majority, the NDP finished second. Since the bulk of the third-place Liberal votes would likely have gone Tory, we can assume the Tories would have held onto these under a transferable ballot.

Finally, in three other seats where the Tories won but failed to secure an absolute majority, they nonetheless received more votes than the second-place Liberals and NDP combined. In two of the three, it is plausible to assign a win to the Tories, as there is no way of knowing which way the second choices of fringe voters in these ridings would have gone.

Adding it up, the Tories would in all likelihood have won at least 64 seats, possibly more, under a transferable ballot. A majority? Probably. A mandate? That's for you to decide.