Sometimes, like a giant jigsaw puzzle, concepts in endocrinology take time to develop, to form a coherent picture of how endocrine systems are integrated to support physiological responses. Such, it can be argued, has been the case with respect to our understanding of the mechanisms underlying puberty. For more than 50 years, it has been known that mammals experience a period of agonadal restraint during development, in which reproductive function is held in check (1). As puberty approaches, this restraint is switched off and pulsatile GnRH secretion is activated. What caused the juvenile period to end remained a matter of conjecture, although a shift in steroid feedback was suggested by observations showing that immature rats had enhanced sensitivity to the negative feedback effects of estradiol (2), whereas the onset of ovarian cyclicity could be advanced by prepubertal injections of estradiol (3). In the past 5 decades, several other important pieces of the puzzle have been filled in. The brain was itself found to be a site of gonadal steroid synthesis (4, 5). Data emerged to indicate that neurosteroid synthesis might contribute to reproductive function (6). Dramatic advances were made in understanding the neurotransmitter and neuropeptide circuitry responsible for regulating the onset of reproductive function, in particular the kisspeptin system (7, 8). However, what actually triggered the change in feedback control of gonadotrophin secretion at puberty remained obscure. A shift from a predominantly inhibitoryhypothalamic environment toone inwhichexcitatory neurotransmitters activated the GnRH neurons was believed to occur (9, 10), but what caused this shift remained unknown. In the current issue of Endocrinology, Kenealy et al (11) develop the novel idea that in female primates, the control of GnRH secretion before and during puberty may actually operate in a relatively similar hypothalamic steroidal milieu: only the source of the estradiol is different. To determine hypothalamic steroid concentrations, perfusates were collected in vivo from the stalk-median eminence (S-ME) region of 32 female rhesus monkeys at different stages of prepubertal and pubertal development. The samples were assayed for estradiol, estrone, testosterone, androstenedione, and progesterone, using a sensitive and specific liquid chromatography-mass spectrometry-based method. Concomitantly, GnRH in the perfusates, as well as LH and steroid levels in circulating serum samples from the same animals, were also assayed. As expected, GnRH and LH levels followed a similar developmental pattern, levels at midpuberty being higher than either prepuberty or early puberty. Circulating estradiol levels also were higher in midpuberty. The steroid measurements in the S-ME perfusates, however, exhibited a strikingly different pattern. Although androstenedione, testosterone, and progesterone did not change significantly in the perfusates across the different stages of puberty, estrone and estradiol were found to be high in the S-ME perfusates from prepubertal animals, declining in midpuberty and early puberty, respectively. Because serum estradiol levels increased over the same period, the S-ME to serum ratio fell dramatically in the midpubertal animals, more than 5-fold in the case of estradiol and more than 60-fold in the case of estrone. These results suggest that although the prepubertal period is indeed characterized by a period of GnRH inhibition that does not depend on ovarian hormone secretion, it is not in fact steroid independent. Rather, it may be mediated at least in part by estradiol produced in situ,