At age ten, I had my first question about genetic inheritance. That was when I found out that former President Kennedy, who must of course have been the smartest man in the world, had a sister who was mentally retarded. How could that be? I thought kids were simply homogenized versions of their parents. I had never seen pictures of JFK's mom and dad, but with he and his two brothers so close in resemblance, how could it be that their sister dramatically wasn't?

The siblings roaming my own neighborhood were all obviously cut from the same cloth: The shape of the nose, the set of the eyes, the color of hair gave them away. The litter of kittens born to our cat Smoky in a dark corner of the basement were all charcoal gray. Except for the runt, they struck me as interchangeable.

My conclusion, that inheritance was a mix of the sum of parental parts, put me in good company. Linnaeus thought the same thing. So did Darwin. As I saw it, inheritance worked a bit like a Mister Softee ice cream machine: You put in chocolate and vanilla, and what came out was an even blend. I dismissed the example of Kennedy's sister as a case of the machine breaking down.

By junior high, however, other, more troublesome, holes started to tear in my theory. One spring, we salvaged a rusty incubator from a great aunt's barn, set it up on our ping-pong table back in suburbia, and hatched a half-dozen eggs. The contributing hen was black, the responsible rooster white. The chicks that emerged were . . . . black (2) and white (4). Not one of them was gray, or even speckled.

Equally puzzling were the differences that emerged between me and my siblings. I hadn't noticed these early on—maybe because we all wore the same brand of pajama. But by junior high people started to subtly point them out: Seeing my brother and me together at the Little League field, folks would ask which one of us was adopted. Uhhh . . . .neither. My sister and I got our father's olive skin and jet-black hair. My brothers were red-blond and fair, like our mom. What happened to blending?

The Mendel unit in seventh-grade science came just in time. The law of segregation, with dominant and recessive traits competing for expression, helped me understand why my brothers and I looked so un-alike. And the idea that dominance could be incomplete took care of what appeared to be blending in kittens and in playmates. Best of all, the grid-like Punnett squares used for crossing gene types promised statistical certainty. With a little extrapolation, I thought, I could pretty well predict how I was going to turn out.

The most important Mendelian trait, in this context, was the one for tallness, which in peas equals stem length. For the sake of my planned basketball career, I knew I needed above-average height. Encouragingly, tallness in Mendel's peas was a dominant trait. My parents, I thought, were both reasonably tall. Wanting to be conservative, I represented them each as Tt (not TT), and worked the square.

By its reckoning, I had a solid 75 percent chance of achieving the stature I hoped for. Reassuring odds. But what if I looked closer? My father's father had barely topped five feet. Grandma was not much bigger. Italians as a rule were short people. (My father's size was considered sufficiently unusual that some of his kinsmen regarded him as a near-giant.) The relatives on my mother's side, although not exactly short, were not noticeably tall either.

So, where had my parents' tallness come from? If tallness really was dominant, why hadn't it popped out more often on my family tree?

People aren't pea plants. It turns out that height in humans is what is called a polygenic trait. In plain terms, a whole bunch of genes—maybe two dozen, maybe more—work together to determine how tall you're going to be. Each one of these genes affects your height only a smidgen. It's their cumulative effect, the way they interact, that really matters.

Add to this the nurture half of the equation. In this case, the main environmental consideration—nutrition—seems pretty obvious. But nutrition has its gray areas. And if you look close enough, there may be even less tangible factors involved. (Emotional deprivation, studies show, can stunt development.)

Abnormality is a lot simpler: A person with Marfan syndrome, a “tallness disorder” that causes cardiovascular and skeletal problems as well as (usually) extraordinary height, has an overriding mutation. Its large effect makes it relatively easy to spot. But we may never fully understand the causes for variation within the normal range of something as “simple” as height, to say nothing of behavioral traits, or intelligence.

By some shuffle of the genetic deck I ended up a couple of inches taller than my father, and taller than each of my siblings—who ate the same diet, except they were not as finicky. I grew tall enough for basketball, and played in high school. Not remarkably well, however. I wonder how many genes figure into hand-eye coordination?