Numerous studies have shown that pre-mRNA processing — 5′capping, splicing and polyadenylation — takes place while the mRNA is being transcribed. Does this simply reflect the fact that these reactions get going before the polymerase (Pol II) has had time to transcribe the entire gene, or are the processes functionally coupled? In a Commentary onp. 3865, Karla Neugebauer reviews the evidence for functional coupling. Capping enzymes and components of the polyadenylation machinery bind directly to Pol II, and there is good evidence for coupling in these cases. Pol II has also been proposed to recruit splicing factors, but Neugebauer argues that any transcription-splicing coupling is more likely to involve interactions between splicing snRNPs and elongation factors; moreover, she notes that the recursive splicing of transcripts such as the Drosophila Ubx pre-mRNA rules out a strict coupling of splice site synthesis and binding of splicing factors to these transcripts. Finally, Neugebauer points out that mRNA processing is also functionally coupled to mRNA export, since improperly processed mRNAs are often retained at transcription units, which are known to recruit export factors.