I was disappointed to read in your Editorial “Brown's budget briefing” (Nature 440, 581; 2006), discussing a future replacement for the UK Research Assessment Exercise (RAE), that you consider “external research income” as “a reasonable basis for departmental funding”.

I hope it is still true to say that no major scientific prizes or seats at the high tables of science have been awarded on the basis of an individual's research income. Although external research income has been a significant — some might say disproportionate — factor in previous RAEs, it should not become a substitute for scientific excellence.

For scientists, the only significant research outcome is the science and, primarily, its communication through publication. Scientists are all too aware of instances of poor accountability in the spending of highly competitive research grants — the result of grant-giving bodies supporting the idea and not the individual.

A scientific meritocracy based upon success in gaining research funding is wholly reversible. But there are no instances of scientists whose success has been measured by their output, on the other hand, being asked to leave their seats at science's high table, their Nobel prizes in envelopes marked “Return to sender”.

In other words, the measure of scientific success is scientific output — the elucidation of new knowledge and its dissemination through publication — and not science income.

Awarding science funding primarily upon the basis of the latter will only ensure a decline in the former.