Sir

Your News Feature “More than a cosmetic change” (Nature 438, 144–146; 2005) includes an emotive photograph showing the heads of six white rabbits, immobilized to have substances dropped into their eyes, with the caption “Tests that put chemicals into the eyes of rabbits have changed little since the 1940s”. This is not true, at least as far as Britain is concerned. The Home Office, which is responsible for regulating experiments on living animals in Britain, issued guidelines in 1987 for eye irritation/corrosion tests (the Draize test), designed to reduce the pain and injury the test may cause. For example, a substance expected from its chemical nature to be seriously painful must not be tested in this way; the test is permissible only if the substance has already been shown not to cause pain when applied to skin, and in vitro pre-screening tests are recommended, such as a test on an isolated and perfused eye. Permission to carry out the test on several animals is given only if the test has been performed on a single animal and a period of 24 hours has been allowed for injury to become evident.

The interesting News Feature in which this photograph appears is unduly dismissive of experiments on living animals. What are the alternatives? The possibilities are either to stop the development of new drugs for human and veterinary use, or to put new drugs on the market without testing them on living animals, or to test new drugs on humans without previous testing on other animals. Few people would be prepared to accept any of these.

To speak of in vitro tests as ‘alternatives’ to testing on living animals is misleading: both are necessary. It is impossible to imitate in vitro the unimaginable complexity of a human being or indeed of any mammal. In vitro tests on bacterial cultures and tissue cultures are necessary in the early stages of testing the very large numbers of substances that are synthesized in order to produce a single drug for use on humans. These tests eliminate all but a very few of those substances, and only those few are candidates for testing on living animals.