More than 100 missense mutations in presenilin 1 and 2 are associated with early-onset dominant Alzheimer disease. These proteins span the membrane several times and are ostensibly the catalytic component of the gamma-secretase complex, which is responsible for producing the amyloid beta-peptide (Abeta) that deposits in the Alzheimer brain. A common outcome of Alzheimer-associated presenilin mutations is an increase in the ratio of the more aggregation-prone 42-residue form of Abeta to the 40-residue variant, which is often referred to as a presenilin 'gain of function'. An apparent paradox is that most of these mutant presenilins have reduced proteolytic efficiency, which forms part of the counter argument that presenilin 'loss of function' can cause the neuronal dysfunction and death that lead to the disease. In this review, a unifying hypothesis is presented that puts forward a biochemical mechanism by which slower less-efficient forms of the protease can result in a greater proportion of 42-residue Abeta.