Raise a glass to the elected members of the European Parliament (MEPs). Without their intervention last week, the European Union (EU) directive on the protection of laboratory animals would have continued its tortured path through legislative procedures in a form that was thoroughly toxic to biomedical research.

The European Commission, the EU's executive arm, began working on the directive back in 2002.The draft that finally emerged last November was singularly uninformed. It should have balanced the undisputed duty to protect animals with the needs of biomedical researchers to understand disease and develop innovative therapies. Instead, it proposed restrictions that would have blocked whole areas of fundamental research while having no positive influence on animal welfare. In particular, it would have restricted research on non-human primates to “life-threatening or debilitating diseases” (which it did not define) without thought for the basic research required to understand such diseases biologically.

The draft would also have forbidden the reuse of animals in any procedure that could cause more than a “mild” (again, undefined) level of suffering. This would have ruled out the use of surgically implanted telemetric devices, which continuously monitor physiological aspects such as blood pressure or heart rate, save animals the stress of frequent handling, and allow for the testing of different compounds on a single animal. As surgery could hardly be classified as “mild”, an animal would have to be killed after just a single test.

Justifiably alarmed, researchers (and Nature, see 456, 281–282; 2008) added their voices to the powerful lobby of the European drug industry — and MEPs responded. In last week's vote, the European Parliament reversed most of the problematic clauses.

So why did things go so wrong at the commission? The legislation was handled in its environment directorate, which initially consulted with all the stake-holders, but then shut itself off from all influences except the powerful animal-welfare lobby. It even failed to consult on the text with the research directorate. Then, when the text was at last opened to the entire commission for comment last summer, there was little time to make substantial changes.

In parliament, by contrast, the procedure was transparent and professional. The draft directive was examined by three committees — agriculture, research and environment— which considered the interests of animal welfare and researchers with appropriate balance.

The process is far from over, however. According to Europe's elaborate co-decision process, not only the parliament, which is directly elected every five years, but also the European Council of Ministers, comprising representatives of each of the EU's 27 member states, must agree on the final text in two readings. The council will start work on the amended text during the Swedish presidency, which begins in July. The commission will then redraft the directive, taking into account the wishes of parliament and council before the second reading. Changes can be introduced at any stage. But in the final directive, which is likely to be approved during 2011, the interests of research will not be as neglected as they were at the outset.

The European Parliament is the only one of the three EU bodies that is elected and therefore directly answerable to EU citizens. This example shows how important it is to have research-savvy MEPs. The next election takes place early next month. Scientists in the EU would be well advised to consider their local candidates' attitude towards science and to cast their vote accordingly. Meanwhile, researchers and their organizations should keep their eye on the passage of the directive, and keep their campaign weaponry close at hand.