The correlation between ppGpp synthesis and the onset of antibiotic production in Streptomyces coelicolor A3(2) is reviewed. S. coelicolor A3(2) produces at least four antibiotics, two of which, actinorhodin and undecylprodigiosin (whose synthesis is determined by the act and red genes, respectively), are pigmented compounds. Conditions were established that gave dispersed and rapid exponential growth in liquid culture. Nitrogen limitation resulted in a short transition phase, during which a peak of ppGpp was observed, followed by stationary phase. Undecylprodigiosin production commenced in transition phase, while actinorhodin synthesis was detected 5∼6 hr later. Whereas the level of act and red biosynthetic transcripts increased markedly during transition phase, the level of the primary bldA transcript, which encodes the only tRNA of S. coelicolor A3(2) that can efficiently translate the rare leucine codon UUA and which is required for both actinorhodin and undecylprodigiosin production, decreased during early exponential phase; this decrease corresponded to an increased signal for the 5’ end of the mature tRNA, which then remained constant throughout growth. Nutritional shiftdown of exponentially growing cultures evoked immediate synthesis of ppGpp and a severe reduction in growth rate; transcription of act genes occurred shortly after, and was followed by actinorhodin production. However, there was no immediate stimulatory effect on red gene transcription, and undecylprodigiosin production did not occur. Addition of serine hydroxamate, a seryl tRNA synthetase inhibitor, to exponentially growing cultures led to ppGpp accumulation, but to lower levels than after nutritional shiftdown, and did not elicit antibiotic production. Thus only a partial correlation was observed between ppGpp synthesis and the transcription of antibiotic biosynthetic genes. S. coelicolor A3(2) DNA that hybridised to the Escherichia coli (p)ppGpp synthetase gene (relA) was cloned, but it did not encode a relA homologue, and polyclonal antibodies against RelA failed to reveal a homologue in S. coelicolor A3(2) cell extracts. A relA homologue of Serratia marcescens was isolated using the same approach, but failed to hybridise to S. coelicolor A3(2) DNA.