An important problem in the numerical study of pottery from excavations is the way in which quantities of pottery are measured, i e. the answer to the question how much pottery has been found.' Four common measures (sherd count, sherd weight, vessels represented and vessel equivalents) are examined in the light of the need to estimate the relative proportions of different types of pottery present in a context, and to compare the proportions in one context with those in another. Their advantages and disadvantages are discussed, leading to a grading of the measures according to their usefulness. Further criteria are set out and briefly discussed. Finally, the work needed to advance the theoretical basis of quantitative pottery studies is reviewed. -- AATA