Within the bosom of our family, there is consensus that our dog is dim - dim even by dog standards, which takes some doing. Fortunately for him, I know of no doggy IQ tests that might confirm or deny our opinion.

Even if there were, I'm not sure that we would be allowed to use them. Tests designed to measure general intelligence, in widespread use for close to a century, are now so "out" that it is in bad taste even to talk about them. Our education system, which can be relied on to lead the nation in political correctness, is moving to an extreme in which tracking according to academic ability is "unfair," "undemocratic," and "demeaning to the children who are not selected." According to this logic, there will soon be no tryouts for track and field, while the football team will be chosen by random selection among all students.

That sounds ridiculous, but it is easy to see how we came to our present position. Intelligence testing in the past often seemed designed merely to confirm an opinion already held by the testers. "Intelligent" had the implicit meaning, "like me." Since the tests were designed and conducted by educated white males, educated white males - surprise, surprise - came out on top. Everyone else, from other races to uneducated people to women to immigrants, ranked somewhere lower down. And no wonder. What would the unschooled child of an impoverished field worker make of the test question: "Rearrange the letters of 'cart horse' to make something that you like to listen to?" (Try it. The answer is at the end of the column.)

And yet, despite the general mistrust of intelligence testing, and despite our inability to define exactly what we mean by intelligence, we all know that some people are smarter than others. Stephen Jay Gould wrote a whole book, "The Mismeasure of Man," to prove that the idea of IQ, a general "intelligence quotient," was misleading, misguided, and meaningless. Unfortunately he made his case with such eloquence and erudition that he demolished his own argument. Any reader of "The Mismeasure of Man" will conclude that this is the work of a person of superior intelligence.

The problem is, we don't have a good way to measure smartness. It's not that people, some of them very smart people, haven't tried. A hundred years ago, Charles Spearman took the results of many different tests of various areas of knowledge and ability. He sought some combination of them that measured overall intelligence. And, not surprisingly, he found such a combination. He called it "g" for "general intelligence." The idea of the "g-factor" that measures overall intelligence underlies such written tests as the Stanford-Binet IQ test, which together with its more recent variations has been in widespread use from the First World War, to the still-controversial "eleven- plus" examination in Britain.

The problem with measures of intelligence has been that any innate or genetic component seems to be hopelessly entangled with social, educational and environmental factors. Recent work in basic genetics suggests that this may be about to change - and you can bet that the change will produce a whole new round of arguments, accusations and moral dilemmas.

In 1997, Robert Plomin announced the results of an investigation involving a group of 12- to 14-year-olds. This was a particular set of very smart kids. In terms of the hotly debated historical measures, they had IQ's around 160. What was undeniable was that they were gifted and talented. Plomin took a look at their DNA, and discovered that part of it, on chromosome 6, often was different in this group from other people. On that chromosome is a lengthy section of DNA that forms a gene with the unrevealing name, IGF2R. Within that gene lies the sequence that was frequently different for the gifted and talented children.

This is certainly not a "genius gene." It is not sufficient, by itself, to account for the difference between the test group and other children. However, Plomin suggested that there may be at least ten more sites in the human genetic code correlating with higher intelligence. The cumulative effect of these small changes in DNA sequence could give a person a high IQ as measured by all the usual tests.

And now we are moving into very deep waters. Before Plomin's work, a person's best chance of being smarter than average was to have smart parents and grandparents. But suppose that you were to take the fertilized egg that in nine months will be delivered as your own baby, and insert the variant sequence of gene IGF2R. You would also presumably make the adjustments to any other genes that seemed to correlate with being gifted and talented. Wouldn't there be a temptation to do this, as part of your desire to give your child the best possible start in life? It may be expensive, but don't you already try to provide your children with the best educational aids, and send them to the best schools and colleges? Is this any different?

It is, profoundly so, because at this stage of our knowledge when we make "small" changes to an embryo's genetic composition we are not sure what else we may be doing. That does not mean that some parents would not try it. They would. We already have plenty of "stage moms" and "coach dads."

Also, I want to emphasize that such genetic games are not beyond today's science. Already, and almost routinely (though not in humans), genetic engineering permits us to replace a defective gene by a working form, or a normal form by a variation from some different organism. We can pop a mouse gene into a fruit fly, or a fruit fly gene into a mouse. The technique is known as "genetic rescue," which certainly sounds less ominous than "genetic tampering." And if we can be sure of one thing, it is that at some point the operations won't be restricted to fruit flies and mice.

Is there a variation of IGF2R in your unborn grandchild's future? If not, will she be at a competitive disadvantage?

Answer to the cart horse question: The arrangement of the letters that the tester likes to listen to is "orchestra." The tested child has probably never been to an orchestral concert.

"Borderlands of Science" is syndicated by: