Using a data set in which a strong relation between market value (book-to-market) and return is known to exist, this paper fails to find a significant relation between average returns and four other measures of firm size. Furthermore, while market value (book-to-market) does provide significant additional explanatory power over the single beta model, the same result is not true of the other size measures. The conclusion is that there is no evidence that the size effect is due to the existence of a relation between expected return and firm size. Instead, the results are evidence in favour of the hypothesis that the size effect is due to the endogenous identity relating the market value of a firm to its discount rate.