According to several prominent philosophers, pleasure and pain come in measurable quantities. This thesis is controversial, however, and many philosophers have presented or felt compelled to respond to arguments for the conclusion that it is false. One important class of these arguments concerns the problem of aggregation, which says that if pleasure and pain were measurable quantities, then, by definition, it would be possible to perform various mathematical and statistical operations on numbers representing amounts of them. It is sometimes argued that such operations cannot be sensibly applied to pleasure and pain, and that sentences expressing such operations must be false or meaningless. The purpose of this paper is to present, explain, and rebut several versions of this argument. In the first section, I present a generic version of the argument. In the second section, I present a defense of its key premise based on a case involving comparisons of relief from pain, and explain why I think it fails. In the third section, I present and rebut another defense, based on a pair of analogies with temperature. In the final section, I present a third defense, based on an analogy with spatial distances. I then present my reasons for rejecting it. Along the way, I explain my reasons for thinking that pleasure and pain are amenable to interval measurement.