Several methodological problems have interfered with the research discussed in this review. First, the correlational approach, of relating hormone levels to events going on at the same time may be meaningless in follicular maturation, because of long latencies in follicle growth; for ovulation correlations between hormone levels and the final event are excellent. Second, FSH and LH have been defined, purified, and measured on the ultimate criteria of bioassays which may not correlate with the "bioassays" of follicular maturation and ovulation; if the indices of discrimination of these two biological effects differ from standard bioassays, then purification and measurement of gonadotrophins may be inadequate at this time with reference to the functions we have been discussing. Third, the antiserum approach, which seemed so promising as a technique for separating endogenous FSH and LH functions, has been disappointing, particularly in the case of FSH. Antisera which can antagonize rat pituitary FSH on bioassays appear to have no effect on events during the cycle; is this because FSH has no necessary function during the cycle, or that the HCG-augmentation bioassays are meaningless in measuring acute FSH effects in adults, or that the antiserum does not combine, for some reason, with endogenously secreted hormone? Antisera to LH in general have higher binding affinities than those to FSH (Diczfalusy, 1969); can the problem be that the FSH-antibody complex dissociates more readily? Also, there is the phenomenon, which at least has been seen in young animals, of antisera appearing to enhance hormone action, possibly because the complex of hormone and antibody prolongs the circulation time of each. The issue of whether the active site of the hormone is acting as the antigen and providing the site of combination for antibody is another unanswered question.