In translational bypassing, a peptidyl-tRNA::ribosome complex skips over a number of nucleotides in a messenger sequence and resumes protein chain elongation after a ''landing site'' downstream of the bypassed region. The present experiments demonstrate that the complex ''scans'' processively through the bypassed region. This conclusion rests on three observations. (i) When two potential ''landing sites'' are present, the protein sequence of the product shows that virtually all ribosomes land at the first and virtually none at the second. (ii) In such a sequence with two landing sites, the presence of a terminator triplet in phase in the coding region immediately after the first landing site drastically reduces the efficiency of bypassing. (iii) Internally complementary sequences that can form a stable stem͞loop in the bypassed region significantly reduce the efficiency of bypassing. We analyze bypassing from a given ''takeoff'' site to ''landing sites'' at different distances downstream so as to derive estimates of the frequency of ribosome takeoff and of the stability of the bypassing complex.I n ''translational bypassing,'' a ribosome skips over a number of nucleotides in a messenger sequence so as to join the information from two noncontiguous ORFs into the sequence of a single, continuous polypeptide chain (reviewed in ref. 1). The prototypical case has been the translation of the phage T4 gene 60 mRNA coding for a topoisomerase subunit, which involves the bypassing of a 50-nt untranslated segment (or ''coding gap'') between a GGA ''takeoff'' triplet and a GGA ''landing site'' triplet. (In the text, we use boldface to show RNA triplets and normal type for triplets or other sequences in DNA.) This case of bypassing depends on very specific features of the sequence (1-3): (i) identity of the takeoff and landing site triplets, implying that peptidyl-tRNA is what makes the trip; (ii) optimum spacing of the takeoff and landing sites; (iii) a terminator triplet immediately 3Ј of the takeoff triplet; (iv) a stem-loop of optimal stability and placement in the region after the takeoff site; and (v) a particular amino acid sequence of the peptide encoded in a region 5Ј of the takeoff site. The role of these sequence elements and of participating factors has been investigated in detail (1, 4-6).The sequence rules governing this kind of bypass event seem very constraining. We have reported on a type of bypass event that seems to be subject to more relaxed rules and may therefore occur on a wide variety of sequences. This is a bypass induced by ribosome pausing at a ''hungry'' codon calling for an aminoacyltRNA in short supply (7,8). We showed that this kind of bypass shared with the gene 60 case only the requirement for synonymous takeoff and landing sites. Otherwise, it could be demonstrated over various coding gap distances (from 7 to at least 40 nt), and seemed largely independent of the sequence in the coding gap and upstream of it (7). In those experiments and the present ones, we inhibited the activation of isole...