Fiddler on the Proof

I was rather slow at deciding to become a scientist. I play violin and viola and toyed with the idea of being a musician. Eventually I realized that I needed more talent and wasn't willing to practice hard enough to pursue music as a career, but I still play as a hobby. For many years I played in the Madison Symphony Orchestra.

I've always liked science. When I finished college, the path of least resistance was to apply for graduate school. I got a fellowship offer in genetics at the University of Texas, and I was panicked, so I accepted right away. It was the first offer that came. Later I got an offer in chemistry, and if those had come first, I'd be a chemist today.

I spent seven years on the faculty at Dartmouth. Most of those were war years. I didn't think that research in genetics was going to advance the war effort, so I decided to focus on teaching. I taught everything they asked me to, including parasitology, statistics, navigation, and calculus. Teaching mathematics meant that later I had techniques at my fingertips that otherwise I would have had to look up.

I have not been obsessed with research the way many people are. I enjoyed it, but I always taught a great deal. With teaching you have instant gratification. The students are there, they're enjoying it or they're not, but you get feedback immediately. With research, most days don't go very well, and what rewards you get come rather few and far between. They're great, but they don't come every day.

Most of my experimental work used Drosophila. For a while I studied resistance to insecticides. Then I switched to studying the impact of mutations on the population. A pleasant episode in my career was the idea that we could use people's names to learn about the population. If people with the same name marry, there's a reasonable chance it's because they're related. I worked out the theory of this, a modification of standard inbreeding theory. It wasn't very complicated, but it was fun. I liked to call this poor man's genetics. All you need is the public record. At the time I said that someday when we have good markers on the Y chromosome we won't need this method any longer. I think that's come true.

In the summer of 1948, I moved to Wisconsin and I've been there ever since. Much of the research I've done was mathematical. I've been blessed with many good mathematicians as students and postdoctoral fellows. Often I would think of the biological problem, and then they would solve it.

I talk rapidly when I'm teaching. I didn't want to slow down, but students were having a terrible time taking notes. So my first year at Wisconsin, I mimeographed notes of what I was planning to say. When the office staff found mimeographing burdensome, I published the notes as a book. Crow's Notes went through eight editions. The last one was about 20 years ago. Much of it is out of date now. Molecular biology has changed so much. I can't deny progress, and I revel in it, actually.

When I was still at Dartmouth, I was teaching students how to use the chi square test, and they had to look up in a table to see whether the results were significant or not. And it occurred to me that if I could make that into a graphical form, you could get a probability rather than just find out whether it was significant. I fiddled around and discovered a transformation that made these chi square lines essentially straight lines, making interpolation easy. I prepared this graph and had it published in the Journal of American Statistical Association. It wasn't my most important publication, but it's far and away the one that had the largest numbers of reprint requests. It was a very useful tool, but now it's been replaced by a computer.

Computers have just changed everything—including my own life. Problems that I couldn't do before because I couldn't handle the mathematics, I can do numerically nowadays.

One of my best students was a Japanese man named Motoo Kimura. We wrote a textbook on population genetics together, and for a while it was the standard. He moved back to Japan and used his royalties to build a greenhouse. He was an orchid breeder, and he named one of the orchids he'd developed after me. It turned out to be a prize winner. For a while and maybe even still, my name was better known among orchid fanciers than among geneticists.

My wife and I met while we were students at the University of Texas and we were both in the orchestra. We had three children and six grandchildren—two from each child. My wife died two days after our sixtieth wedding anniversary in 2001.

I retired in 1986, when I was 70. I've been retired for 20 years now, but I still go to my office every afternoon. Being surrounded by other faculty and their students keeps me from going totally stale. I just recently finish collaborating with a student on a paper about gene purging by inbreeding. I also do some historical and review writing. It would be very nice if you could have the happy life that I have without getting old.