Although we agree with William Martin and Eugene V. Koonin's point in Correspondence (“A positive definition of prokaryotes” Nature 442, 868; 2006 doi:10.1038/442868c) about the validity of the term 'prokaryote', a term that Norman R. Pace has proposed abolishing (“Time for a change” Nature 441, 289; 2006), they have lost sight of the organismic biology forest for the molecular biology trees. The main differences between prokaryotic and eukaryotic cells probably relate to the original symbioses from which eukaryotes evolved.

Eukaryotes — whether protoctists, fungi, animals or plants — routinely open their membranes to take in (or let out) nuclear genomes, whole cells or other large particles, in processes such as ingestion, fertilization and hybridization. They reseal their membranes and live happily ever after. All eukaryotic sexuality requires cell fusion. Nearly all eukaryotic cell phenomena involve microscopically visible intracellular motility that never happens in prokaryotes.

We need to reassess our understanding of the course of evolution by recognition of the differences between unidirectional transfer of genetic material as the basis of prokaryotic sexuality — genophore DNA, viruses, plasmids — and parental cell fusion in eukaryotes. Roger Stanier and Cornelius van Niel's concept of 'prokaryote' was brilliantly recognized in 1927 by Boris Kozo-Polyansky, who only wrote in Russian. The word 'procariotique' was independently coined in 1925 by Edouard Chatton for cyanobacteria and all other bacteria including archaebacteria (J. Sapp International Microbiology 9, 163–172; 2006). Because of the modern developments of biochemistry and molecular biology, electron microscopy and comparative genetics, the term 'prokaryote' is even more valid now than it was when first introduced.