Bacteria are not Lamarckian

Instructive influence of environment on heredity has been a debated topic for centuries. Darwin's identification of natural selection coupled to chance variation as the driving force for evolution, against a formal interpretation proposed by Lamarck, convinced most scientists that environment does not specifically instruct evolution in an oriented direction. This is true for multicellular organisms. In contrast, bacteria were long thought of as prone to receive oriented influences from their environment, although much was in favour of the Darwinian route (1). In this context Cairns et al. raised a passionate debate by suggesting that bacteria generate mutations oriented by the environmental conditions (2). Several independent pieces of work subsequently demonstrated that mutations overcoming specific defects arised as a consequence of cultivation on specific media (3-7). Two diametrically opposed interpretations were proposed to explain these observations : either induction of mutations instructed by the environment (e.g. by a process involving a putative reverse transcription) or selection of variants among a large set of mutant bacteria generated when stress conditions are present. The experiments presented below indicate that the Darwinian paradigm is the most plausible.