The relative stabilities of specific embryonic mRNAs that persist in Drosophila melanogaster larvae were determined using an approach that combined RNA density labeling with cell-free translation. Unlike the other methods commonly used to measure the decay of individual mRNAs, the density labeling approach does not depend on the use of transcriptional inhibitors or on the measurement of precursor pool specific activities. Using this approach, we have determined that different embryonic mRNA species persist for varying periods during subsequent development, with half-lives ranging from ~2 to ~30 h. The embryonic histone mRNAs are relatively unstable; they are no longer detectable by 9 h of larval development. By 41 h of larval development, 90% of the nonhistone mRNAs assayed have decayed considerably; computerized scanning densitometry of translation products indicates that these transcripts are not decaying as members of discrete half-life classes. The persisting mRNAs that remain are very long-lived; their in vitro translation products can still be detected after 91 h of larval development. We have tentatively identified the mRNAs that encode actin, tropomyosin, and tubulin as members of this stable mRNA population. Although embryonic mRNAs do fall into these three broad classes of stability, they appear to decay with a continuum of half-lives. Because the range of half-lives is so great, mRNA stability is probably an important factor controlling mRNA abundance during Drosophila development.The cytoplasmic abundance of a particular mRNA can be controlled at both the transcriptional and posttranscriptional levels. One of the most poorly understood posttranscriptional controls is that which defines the stability of specific mRNA sequences. The importance of turnover rate in controlling mRNA abundance, and thus gene expression, is best illustrated by the observation that many of the mRNAs that encode the most highly abundant cell type-specific polypeptides are preferentially stabilized over other mRNAs present within the same cytoplasm (6,21,25,29,44).The stability of specific eucaryotic mRNAs has most often been determined by assaying for the presence of particular mRNAs after treating cells with the transcriptional inhibitor actinomycin D (4,13,25,34,43,46). However, this kind of analysis is subject to several possible artifacts. In addition to a secondary effect on cellular protein synthesis (12,15,39,40) it has also been shown that actinomycin D treatment can both decrease the stability of total mRNA populations (41) and increase the half-life of specific mRNAs (9, 42). Also, if 1808 the actinomycin D "chase" does not begin immediately or is not completely effective, then the measured half-life of a mRNA species will appear longer than it actually is.The best experimental system to use for studying the differential stability of individual mRNAs would be one in which no transcriptional inhibition is necessary. For example, by measurement of the rate of radiolabel accumulation into specific mRNAs by usi...