Let me tell you how I made a big mistake about genetics, since it may warn you against judgments that seem reasonable but are based on too little knowledge.

Nobody is so dumb nowadays that he thinks it makes no difference what you inherit from parents, and hardly anybody thinks it makes no difference what the environment is.

Down's syndrome (mongoloidism as it used to be called, somewhat brutally) will set limits to a child's future. He is not going to be a track star and so on; and in this sense you could say biology is destiny.

Equally, if a child grows up without enough to eat, without cuddling and talk (without a good supply of yappy aunts and uncles, in other words, or their equivalent), without a chance to climb trees and chase dogs, then he is not going to be as wonderful as the rest of us. You could say that nurture is destiny.

We all know that much. What we don't know is the dimension of our ignorance in both genetics and environment, even though that ignorance may be more crucial than the knowledge we pride ourselves on.

Genetic engineering is now supplanting killer bees as something to worry about, and I would hate to see researchers lynched on grounds they are inventing monsters. Careful scientists are by definition careful, but it is expecting too much for anybody to be "careful" if he thinks he is on safe ground. A guy walking on quicksand is not careful if he thinks he's on a highway. So scientists should not get ahead of themselves too much.

And now with some embarrassment I shall report on my own expertise:

A great buddy of mine, the late Eddie Fox, used to breed garden irises, bright flowers of many colors. Every year he raised 3,000 new kinds in all colors and many variations of shape and stature.

Fortunately the iris is extremely simple compared to a human. All you want is a pretty face and you can forget brain chemistry and the rest.

Simple though the flower is, in comparison with humans, it is not all that easy to breed even pretty faces. You will get surprises. Even the simple iris genes along its 48 chromosomes are sufficiently complicated to wear you out if you raise them. From two rose-colored parents, say, you should raise at least 500 offspring to give yourself a fair chance to see what those parents are capable of producing. And to explore it more fully, you need to inbreed for several generations.

Statistically, a progeny of 500 should cover the genetic ground but there is no reason you should not grow 20,000 offspring and find the great one in the last seed you planted. If you raise only 30 you may get pretty faces, but you can hardly say you know the genetic potential of the parents.

All the same, if you spend years raising the dratted beasts you probably do know more about them than somebody who never looked at them. Eddie and I knew a lot, and we knew there are genes for the height of the flower stem. An iris with genes for a stem six inches high will not grow tall and you can pray otherwise till you're purple without changing anything.

One year we decided to select 20 irises with short stems from among the thousands of tall ones. We chose the best short ones. But knowing that trifling variations of weather can make enormous differences in the flowering performance of an iris, we were cautious. A late cold snap may produce a short stem on an otherwise tall iris, and the degree of injury may depend on the state of the plant at the time the cold hits. One plant may have its embryonic stalk further developed than others, and be hurt when they are not. So an experienced breeder does not make snap judgments but watches the plant over several years.

Our 20 shorties were uniform for three years running, though their siblings were tall. They had the same sun, same soil, same culture as the tall ones and we concluded, reasonably, they were short because they had genes that made them short and kept them short. Besides, if you know irises inside out (hear! hear!) you have a feel for these things.

So with full faith in God's shortness we moved our 20 plants to another garden where -- just here I go a bit pale -- every one of them turned tall, ranging from 38 to 42 inches.

It was reasonable to conclude the irises were genetically short, but the unreasonable truth is they were genetically tall and we experts were fooled on this. If you plant a short iris in another place and it grows to chest height you may not know why, but you do know it is not genetically short. That means something in their environment kept them from growing tall, even though two knowledgeable observers could not tell what that something was.

We were both expert, in the sense of knowing more than most people about irises. It's just that we could not have got worse results if we had flipped coins or asked the first guy on the street to do the selecting. Rude persons might say we were not all that expert after all.

I try to remember these irises when, say, someone in traffic performs like a natural-born jackass. If somebody drives like a banshee or drools or otherwise annoys us, clearly there is something serious the matter with him, and we attribute it to his mother, commonly. But it may not be the genes. Jerk probably had the wrong aunts or went to the wrong kindergarten or ate too much Brie.