Over recent years, online auctions have become increasingly popular. Fouryears ago we developed an online auction system, where bidders were allowedto bid for multiple units of a good. In this paper we experimentally comparethe performance of the English clock auction against our web-basedauction using traditional economists' criteria. The English clock auction,implemented offline, is an open-cry, uniform price auction, where biddersrespond to prices by indicating desired quantity. In contrast, our web-basedauction allows (but does not initiate) someprice discrimination. The bids are two-dimensional, consisting of priceand quantity combinations. In our experiments the two auction mechanisms,one implemented offline, the other online, were found to be (equally)highly efficient and generated approximately an equal amount of revenue forthe seller/buyer. The logical next step is to replicate the comparisonin field circumstances. It remains to be seen, but one can argue that manyof the advantages of web-based systems do not show in laboratory circumstances.We would also like to raise the question, whether the traditional economistscriteria, such as allocative efficiency and short-run revenue, thoughimportant, are sufficient to discriminate among traditional and web-basedauction mechanisms.