It's not news that money matters in congressional elections, but on the basis of data elicited by the campaign finance reform legislation of the early 1970s, Jacobsen (Political Science, Univ. of California, San Diego) is in a position to demonstrate how: ""the challenger's spending is what matters,"" he concludes, ""what incumbents spend makes relatively little difference."" And consequently the extensions of public subsidies to congressional elections--recently rejected by Congress--would be to the advantage of nonincumbents. Jacobsen develops this argument through analyses of the results of the '72, '74, and '76 elections that will prove hard-going for those without a background in statistics; and the text is written in standard academese. But there are subsidiary findings that claim attention--if not always acquiescence--throughout. It's not surprising that Republican spending has a greater electoral impact than Democratic--after all, they have farther to go. More arguable is Jacobson's contention that small gifts can have no effect on the outcome, that it is therefore irrational to restrict individual contributions. He overlooks the relative strength of incumbents, the degree of acceptability of the challenger. And he ignores both the inherent ability of wealthy candidates to raise funds (from family and friends) and the fact that--especially in Senate races--they may lose anyhow. The limitation of his thesis overall is that, however valid it may be statistically, it cannot be applied wholesale to individual situations. But he scores here, for those with an interest in politics rather than in theory, in his close examination of seven 1974 Congressional races--to see how, in the wake of a crisis like Watergate, expectations alter results. The concluding chapters scrutinize the course of campaign finance regulation; the motivation of office-holders, unions, corporate, and public-interest groups; and the various suggested alternatives. It's methodical and thoroughgoing--and sure to be widely cited if not widely read.