We live in the age of genetic engineering. It seems that every week a newly engineered plant is announced; there's a good deal of concern over whether this is always a good thing. But before one even gets to question the potential threats posed by genetically modified crops, one has to answer a more basic question: do they work?

In other words, is there any evidence that genetically modified crops are making some farms more productive, or more reliable, or more easily run than before? Now, dramatic data from India answer this question with a strong, if tentative, yes.

The crop in question is cotton. A genetically altered form of the plant, known as Bt cotton, sports a toxin gene lifted from bacteria. The result is--hold your breath--cotton plants that grow their own pesticide.

It sounds like science fiction, but it's quite real. And in India, Bt cotton seems to be working. Smallish farms in that country have reported an eighty percent increase in their cotton yield since they switched to Bt cotton. That's a huge difference. Less dramatic improvements have been found for Bt cotton in Arizona crops.

This may be a significant development for regions of the world where pesticides are too expensive or too difficult to procure, and where insects routinely destroy crops. Of course, it's still unknown whether the genetically produced resistance to certain insects will last, or whether the insects will simply become resistant to the Bt toxin themselves. And will there be long-term consequences for crops, insects, or consumers? On that question, the jury is still out.