A dozen years ago, when William Shockley first began talking about improving the stock of the human race, nobody knew that he was going to try to do it all by himself.

Back then, you may recall, the man who won a Nobel Prize for making transistors started talking about making babies. He believed that the world would be better off with more smart people than dumb people and that intelligence was inherited.

Therefore, he suggested that the best way to improve the world was to set up sperm banks for very smart men and impregnate very smart women. As an amateur geneticist, he sounded like an expert in making transistors.

Well, now it turns out that a 74-year-old California businessman, Robert K. Graham, has actually founded an exclusive sperm bank. In what sounds like a Woody Allen script, he solicits donations from Nobel Prize scientists only -- Peace Prize and literature prize winners need not apply. And guess who was among the first three donors? William-Father-of-the-Year-Shockley.

When I first heard this piece of news out of The Los Angles Times, I was stunned by the sheer conceit of the donors. The belief in the superiority of your sperm -- even 70-year-old sperm -- can only come from long nights spent re-reading the Stockholm speeches. If ego is carried along by the DNA, the three women impregnated so far in this program will give birth to a trio of miserably conceited little monsters.

The Repository for Germinal Choice (I did not make the name up) is a phallic symbol -- without the symbolism. But there is more here than meets the ovum.

The the past two weeks, we have had this bulletin about a sperm bank for propagation of elite in California and another tale about a program for sterilization of thousands of "misfits" from 1922 to 1972 in Lynchburg, Va. These are both entries in the historic annals of genetic control.

For over a century, this country has debated whether heredity or environment is most important in determining the physical, mental and "moral" health of a human being.

In mid-19th-century Massachusetts, the poor Irish who lived in slums with virtually no health care had a higher infant mortality rate than others. An extensive study conducted by Massachusetts Brahmins blamed it on their "weaker genetic stock." After World War I, when the country went through a fit of xenophobia, they used IQ tests at Ellis Island to "prove" that most of the Jewish, Hungarian, Italian and Russian immigrants were "feeble-minded."

In the same vein, an old study of Harvard College graduates lamented the fact that the educated Harvardians underreproduced themselves. Early birth-control advocates put it bluntly: "More children from the fit; less children from the unfit."

In short, the situation at Lynchburg was only a more efficient example of a fairly popular concept. People categorized rather blithely as retards and "misfits" -- including young women who were committed for being sexually active -- were to be prevented from polluting the genetic pool. No one ever proved that they had genetic diseases.

Well, in the 1930s, due to a well-known eugenicist named Adolf, a lot of these theories fell into disrepute. But they started percolating up through the surface of social reform again about 10 years ago when Arthur Jensen, Richard Herrnstein and, yes, "Papa" Shockley started putting forth theories that IQ was largely inherited.

Shockley endorsed not only the concept of "increasing the people at the top of the population" but also the concept of "reducing the tragedy of the genetically disadvantaged at the bottom." Not coincidentally, he believes that blacks are genetically less intelligent than whites. Here we go back to Ellis Island.

There are reasons why the notions of "genetic inferiority and superiority" become popular from time to time. When we lose confidence or interest in reforming the environment, we are more likely to blame genes.

If an underclass exists in a democratic society, we want to blane its "stock" rather than our system. When the economic times are hard, I think we are also more likely to think in terms of controlling people rather than helping them. When social programs seem messy and complicated and exhausting, we turn to efficient engineering of science.

It all sounds so logical. Cast genius sperm upon the world, like Johnny Appleseed, and you will get a crop of genuises. But genetically it just ain't so. Furthermore, the definition of a successful human life isn't as simple as that of a successful race horse. Genius is more than genes.

And if you're in the market for some Shockley sperm, think about this: you might end up with the genes of a lousy geneticist instead of a decent transistor.