Human GH represents a family of proteins rather than a single hormone. The circulation contains a bewildering array of GH forms, including several monomeric variants, their homo- and heteropolymers, fragments, and complexes with at least two BPs. The net biological activity of this mixture is difficult to predict, as the various molecular forms interact as partial agonists and/or antagonists at the receptor level. The number of GH forms that can be counted in plasma exceeds 100. Table 5 attempts to illustrate what is known and provide estimates for circulating variants. It does not include GH-V and its variants, which have to be added in pregnancy. Of note, what is commonly understood as "plasma GH," i.e. free monomeric 22K, represents only 21% of total immunoreactivity in plasma. In view of this complicated picture, it should be no surprise that different assays of plasma GH yield different results (107, 108, 290). While immunoassays are relatively unaffected by the BPs (291), receptor assays are seriously affected by the high affinity BP (261). Immunoassays, particularly of the monoclonal variety, are vulnerable to differential recognition of molecular variants depending on the unique epitope specificity of the antibody used. Polyclonal assays are more robust in this regard because of "epitope averaging" among the wide spectrum of epitope specificities present in the antibody population. Future work should aim at developing antibodies that are specific for individual GH variants. Such molecular probes will be helpful not only in standardizing immunoassays, but also in delineating the biological role of the various GH forms. The physiological significance of the numerous GH forms (or of the BPs) is still largely unknown. Progress in this area has been hampered, on the one hand, by the unavailability of pure GH variants in quantities sufficient for biological studies, and, on the other, by a certain lack of interest stemming from suspicions about artifacts. The recent resurgence of interest in GH and in its receptor and BPs should also refocus attention on the various molecular forms. Thus far, this interest has been largely confined to monomeric 22K, which is certainly effective for its original intended purpose, namely growth promotion. Whether 22K is sufficient for optimal growth and development, or whether it can fulfill all the functions of the GH family is unknown. It can be argued that evolutionarily conserved GH variants probably have biological importance.(ABSTRACT TRUNCATED AT 400 WORDS)