Sir

Your unusual and perceptive News Feature ”Crop improvement: a dying breed“ (Nature 421, 568–570; 2003), on the importance of classical plant breeding, contains an error concerning intellectual property rights applied to plants. Plant breeders' rights, called 'plant variety protection' in the United States, allow and defend the right of plant breeders to use protected varieties as parents in further breeding, contrary to the impression given in your feature. The patent system, not plant breeders' rights, prohibits the use of protected varieties as parents in further breeding. Moreover, utility patents on plants became important only after public plant breeding had already declined.

That decline reflects a sea change in US public policy. From the Civil War until quite recently, agricultural research was seen as a public responsibility, creating public goods. Then, decades of sustained prosperity after the Second World War gave rise to a belief that this was no longer necessary. It was also believed that market forces would supply all the goods that were needed, and that the public sector would be competing with the private sector.

During the same decades, the formerly rural US population became suburban and urban. Public goods of concern to today's policy-makers are the ones that preoccupy urban citizens. These are the reasons behind the decline in support for public plant breeding (see K. J. Frey, National Plant Breeding Study, Iowa Agric. Home Econ. Exp. Station, Ames, Iowa; 1996), as correctly pointed out in your News Feature.

Much as we may consider agriculture to be passé — if we think about it at all — our urban lifestyle depends on a secure supply of abundant, inexpensive, safe and nutritious food. As pointed out in your News Feature, classical plant breeding is the only technology that is currently capable of delivering the secure harvests we require. This is because only plant breeding can manage many subtle traits at once, as needed to confront challenges of climate change, newly emerging pests and other uncertainties.

Public plant breeding is also a good investment. The rate of return to investment in public plant breeding is 35% for potatoes in the northwest United States, realized in reduced cost of production and improved food quality as well as in increased yield, according to A. A. Araji and S. Love (Am. J. Potato Res. 79, 411–420; 2002). These results are typical of studies in other crops.