Sir

A large majority of the public supports the principle of animal experimentation to improve biological knowledge, human and veterinary health, nature conservation and, last but not least, animal welfare. This support, however, depends on strict adherence to the 3Rs (replace, refine, reduce) principle, described in a 2005 report by the Nuffield Council on Bioethics (see Nature 435, 392; 2005). This is aimed to minimize animal numbers, pain, suffering and lasting harm.

A recent set of News Features on animal research (“A matter of life and death” Nature 444, 807–816; 2006) identified considerable scope for advancing the 3Rs, which crucially depends on an effective spread of relevant knowledge and techniques. Several specialist journals — such as Laboratory Animals, Lab Animal and ALTEX — publish research aimed at advancing the 3Rs criteria, but their readership and impact are limited. Regulators and animal-care committees do not have the means to broadcast and implement novel techniques effectively. Many useful refinements are developed and applied by single labs, but do not get published because they were not a primary focus of the research or because they are not sufficient for a paper on their own. Much relevant information might thus never become widely available within and between these communities.

Most scientific journals require a statement of adherence to legal and institutional animal welfare guidelines. Some have their own codes of practice and ethical committees to guarantee high animal-welfare standards in published material (see, for example, Animal Behaviour at http://asab.nottingham.ac.uk/ethics/guidelines.php).

Journals could play a much more effective part, however, by including a 3Rs section in the methods section of published papers. First, this would allow authors of controversial papers to detail their measures to minimize pain, suffering and lasting harm. Second, it would let them describe novel tools or techniques used in the paper that serve the 3Rs. Journals could make this 3Rs section optional, and — depending on the significance and length of the material — could either include it in the print version of the paper or make it available online only as accompanying supplementary material.

Leading journals such as Nature could pioneer such a policy. This would be in the best interests of editors, scientists and the public, as well as to the benefit of the experimental animals.