We have recently shown that although pre-mRNA splicing in plants shares some features in common with splicing in vertebrates, there are some crucial differences. In plants there is a requirement for a general enrichment for A+U within the intron and there is no requirement for a 3 ' polypyrimidine tract (Goodall and Filipowicz, 1989). We suspect that the AU-rich intron sequences are recognized by specific RNA-binding proteins. In vertebrates several RNAbinding proteins appear to be involved in pre-mRNA splicing. Depletion of the hnRNP C protein from active in vitro splicing extracts abolishes splicing (Choi et al., 1986) and UV crosslinking studies have identified a protein of 62 kDa that binds specifically to the polypyrimidine tract of several introns (Garcia-Blanco et al., 1989). We have indirect evidence that in plants, as in vertebrates, proteins bind to nascent RNA during synthesis. Furthermore, we find that several proteins in plant nuclear extracts can be labelled by UV-crosslinking with intron-containing RNAs. On the other hand, we have not been able to isolate hnRNP particles from plants using methods that were successful in our hands for isolating hnRNP particles from HeLa cells and Western blots with the hnRNP C protein-specific monoclonal antibody 4F4 (Choi et al. , 1986 ) revealed the appropriate sized band from HeLa cell extracts, but did not identify any proteins in plant nuclear extracts. To further examine the RNA-binding proteins of plants, we have employed a cDNA cloning strategy. Many RNA-binding proteins contain a presumptive RNA binding domain, the "RNP domain", within which can be perceived two short consensus regions that are conserved from yeast to mammals. We have designed oligonucleotide primers based on these consensus sequences, and used them in PCR reactions on tobacco cDNA. The amplification products were subcloned and sequenced, revealing 15 different cDNA sequences with strong similarity to the RNP domain. To determine the sequences outside of the RNP domains, and thereby perhaps recognize whether we have cloned any plant homologues of identified RNAbinding proteins, we are using the cDNA sequences obtained to screen a N. plumbaginifolia cDNA library.