Vitellogenin is the serum precursor of the yolk proteins α-lipovitellin, rβ-lipovitellin, and phosvitin. The precursor can be dissociated to produce the yolk proteins only by proteolytic enzymatic action, to which it is very susceptible. Denaturation in sodium dodecyl sulfate, combined with reduction of disulfide bridges and blocking of thiols, yields a complex with a molecular weight of 200,000 to 250,000. α-Lipovitellin contains three polypeptides, with molecular weights of about 135,000, 105,000, and 40,000, and rβ-lipovitellin is composed of two polypeptide chains with molecular weights of 135,000 and 30,000. The 40,000 subunit of α-lipovitellin and both rβ-lipovitellin subunits are phosphopeptides We tested RNA isolated from the liver of estrogen-treated roosters for mRNA activity in a cell-free reticulocyte system. The vitellogenin mRNA has a sedimentation coefficient greater than 28S and thus contains enough information to code for a long polypeptide chain. Estrogen administration to roosters induces the appearance of vitellogenin and a lowdensity lipoprotein, the syntheses of which are not coordinated. The course of vitellogenin synthesis was calculated from accumulation and turnover data, and it was found that from about 25 hr after estradiol-17rβ administration the rate of vitellogenin synthesis increases linearly for several days, paralleling an increase in vitellogenin-synthesizing polysomes. Thus, we estimate a constant translation rate of about 8 amino acids per ribosome per sec. A “memory” effect is observed when a second hormone dose is given some time after the vitellogenin induced by the first dose has disappeared from the blood. After the second dose vitellogenin synthesis is detected sooner, and its initial increase is more rapid, than after the first dose. Although the synthesis of vitellogenin starts 3 to 4 hr after the second as well as after the first injection, the rate of synthesis after the first injection increases much more slowly during the first 15 hr than during the subsequent period of linear accumulation, whereas after the second injection the linear increase in the rate of synthesis begins immediately after the lag period of 3 to 4 hr. The “memory” effect is undiminished even 50 days after the first hormone dose; thus, the causative factor either is very stable or is synthesized in great excess during the first stimulation. When the second injection is given during the descending part of the turnover curve, an increase in vitellogenin synthesis is observed within 3.5 hr. There are thus at least three different effects of estradiol; (i) the “memory” effect, which probably is due to commitment or differentiation of vitellogenin-synthesizing cells; (ii) the effect that causes the committed cells to give full response after the 3- to 4-hr lag period; and (iii) the effect that causes the immediate response. To explain these results we suggest that committed cells can synthesize vitellogenin mRNA only during a certain period of the cell cycle.