This paper discusses the results of a series of 42 interviews with Chemists, Computer Scientists and Sociologists conducted in 2006-2007 at two Australian universities. All academics perform peer review with later career researcher usually taking a greater load. The amount and type of review undertaken differs between disciplines. In general, review of journal articles and conference papers is unpaid work although reviewing books (a much larger task) often results in at least an offer of a free book from the publishers. Reviewing of grant proposals and theses does attract an honorarium, but these are insignificant amounts. Most interviewees indicated that reviewing is part of what is expected in academia, and that it offers the benefit of early access to new research results. The competing requirements of an academic’s peer group and the institution at which they work has meant a sharp increase in the number of papers published over the past decade. This in turn has made finding referees difficult, and the fact the work goes unrecognised by the performance measurement process adds to the problem. The claim of certain conferences that their papers are refereed is met with some cynicism, even in Computer Science, which normally uses conferences as its main channel of peer reviewed communication. Overall these findings open the question of whether the amount of effort expended in peer review is justified.