psi are ubiquitous in ribosomal RNA. Eubacteria, Archaea, and eukaryotes all contain psi, although their number varies widely, with eukaryotes having the most. The small ribosomal subunit can apparently do without psi in some organisms, even though others have as many as 40 or more. Large subunits appear to need at least one psi but can have up to 50-60. psi is made by a set of site-specific enzymes in eubacteria, and in eukaryotes by a single enzyme complexed with auxiliary proteins and specificity-conferring guide RNAs. The mechanism is not known in Archaea, but based on an analysis of the kinds of psi synthases found in sequenced archaeal genomes, it is likely to involve use of guide RNAs. All psi synthases can be classified into one of four related groups, virtually all of which have a conserved aspartate residue in a conserved sequence motif. The aspartate is essential for psi formation in all twelve synthases examined so far. When the need for psi in E. coli was examined, the only synthase whose absence caused a major decrease in growth rate under normal conditions was RluD, the synthase that makes psi 1911, psi 1915, and psi 1917 in the helix 69 end-loop. This growth defect was the result of a major failure in assembly of the large ribosomal subunit. The defect could be prevented by supplying the rluD structural gene in trans, and also by providing a point mutant gene that made a synthase unable to make psi. Therefore, the RluD synthase protein appears to be directly involved in 50S subunit assembly, possibly as an RNA chaperone, and this activity is independent of its ability to form psi. This result is not without precedent. Depletion of PET56, a 2'-O-methyltransferase specific for G2251 (E. coli numbering) in yeast mitochondria virtually blocks 50S subunit assembly and mitochondrial function (Sirum-Connolly et al. 1995), but the methylation activity of the enzyme is not required (T. Mason, pers. comm.). The absence of FtsJ, a heat shock protein that makes Um2552 in E. coli, makes the 50S subunit less stable at 1 mM Mg++ (Bügl et al. 2000) and inhibits subunit joining (Caldas et al. 2000), but, in this case, it is not yet known whether the effects are due to the lack of 2'-O-methylation or to the absence of the enzyme itself. Is there any role for the psi residues themselves? First, as noted above, the 3 psi made by RluD which cluster in the end-loop of helix 69 are highly conserved, with one being universal (Fig. 2B). In the 70S-tRNA structure (Yusupov et al. 2001), the loop of this helix containing the psi supports the anticodon arm of A-site tRNA near its juncture with the amino acid arm. The middle of helix 69 does the same thing for P-site tRNA. Unfortunately, the resolution is not yet sufficient to provide a more precise alignment of the psi residues with the other structural elements of the tRNA-ribosome complex so that one cannot yet determine what role, if any, is played by the N-1 H that distinguishes psi from U. Second, and more generally, some psi residues in the LSU appear to be near the site of peptide-bond formation or tRNA binding but not actually at it (Fig. 2B) (Nissen et al. 2000; Yusupov et al. 2001). For example, position 2492 is commonly psi and is only six residues away from A2486, the A postulated to catalyze peptide-bond formation. Position 2589 is psi in all the eukaryotes and is next to 2588, which base-pairs with the C75 of A-site tRNA. Residue 2620, which interacts with the A76 of A-site-bound tRNA, is a psi or is next to a psi in eukaryotes and Archaea, and is five residues away from psi 2580 in E. coli. A2637, which is between the two CCA ends of P- and A-site tRNA, is near psi 2639, psi 2640, and psi 2641, found in a number of organisms. Residue 2529, which contacts the backbone of A-site tRNA residues 74-76, is near psi 2527 psi 2528 in H. marismortui. Residues 2505-2507, which contact A-site tRNA residues 50-53, are near psi 2509 in higher eukaryotes, and residues 2517-2519 in contact with A-site tRNA residues 64-65 are within 1-3 nucleotides of psi 2520 in higher eukaryotes and psi 2514 in H. marismortui. A way to rationalize this might be to invoke the concept suggested in the Introduction that psi acts as a molecular glue to hold loose elements in a more rigid configuration. It may well be that this is more important near the site of peptide-bond formation and tRNA binding, accounting for the preponderance of psi in this vicinity. What might be the role of all the other psi in eukaryotes? One can only surmise that cells, having once acquired the ability to make psi with guide RNAs, took advantage of the system to inexpensively place psi wherever an undesirable loose region was found. It might be that in some of these cases, psi performs the role played by proteins in other regions, namely that of holding the rRNA in its proper configuration. Confirmation of this hypothesis will have to await structural determination of eukaryotic ribosomes.