The debate continues on the issue of whether nuclear introns were present in eukaryotic protein-coding genes from the beginning (introns-early) or invaded them later in evolution (introns-late). Recent studies concerning the location of introns with respect to gene and protein structure have been interpreted as providing strong support for both positions, but the weight of argument is clearly moving in favour of the latter. Consistent with this, there is now good evidence that introns can function as transposable elements, and that nuclear introns derived from self-splicing group II introns, which then evolved in partnership with the spliceosome. This was only made possible by the separation of transcription and translation. If introns did colonize eukaryotic genes after their divergence from prokaryotes, the original question as to the evolutionary forces that have seen these sequences flourish in the higher organisms, and their significance in eukaryotic biology, is again thrown open. I suggest that introns, once established in eukaryotic genomes, might have explored new genetic space and acquired functions which provided a positive pressure for their expansion. I further suggest that there are now two types of information produced by eukaryotic genes--mRNA and iRNA--and that this was a critical step in the development of multicellular organisms.