He shot back: "Yeah, but remember that marathon runner, what's his name, who keeled over? When your time's up, your time's up."
I had a professional's stout faith that an exercise bicycle and an absence of well-marbled steaks from my diet would keep me young. He had a workingman's certainty that fate can knock us off our high horses any old time.
We were both right. And both wrong.
Michael Crawford says that how quickly people age and die depends about 45 percent on their genes and about 55 percent on their behavior.
That's the bottom line of a 20-year study of 1,200 Mennonite residents in Goessel and Newton in central Kansas and Henderson, Neb.
Mennonites made ideal study subjects, Crawford says. They're reliable, geographically close to Lawrence, don't smoke or drink. They're also genetically homogeneous.
Crawford is a Kansas University professor of anthropology. His academic study of the Mennonites is titled "Different Seasons: Biological Aging among the Mennonites of the Midwestern United States." Eighteen other scholars from 11 institutions also contributed.
Between 1979 and 1982, a bunch of measurements were taken of the 1,200 Mennonites. Records of age, height, weight and 33 other body features were logged. So were 25 blood chemistry measurements. Hand steadiness, reaction time, trunk flexibility, lung capacity -- even personality -- were measured.
Years later, the scientists returned to the communities to see who'd died. They used statistical methods to compare the biological and psychological profiles of those who lived and those who didn't.
Surprisingly, Crawford said, cholesterol levels, blood pressure and body mass index were not good barometers of who survived. Neither was personality. Whether you were a driven Type A or a laid-back Type B made little difference.
Of course the biggest factor in survivorship was the person's age when the study started. Beyond that, levels of certain kidney chemicals were the single most telling indicators of who died, Crawford said.
The finding that 45 percent of the difference between how quickly people age is accounted for by genes and about 55 percent by what they do is the "pearl" in the data, Crawford said. That pearl is being submitted for publication to the American Journal of Human Genetics.
Some of our traits are more genetically based than others, Crawford said, including hand steadiness, hand-eye coordination, reaction time and trunk flexibility. By contrast, blood chemistry, body mass index and personality type are more susceptible to environmental or behavioral modification.
Two questions then, Professor Crawford.
First, what advice would you give to those who want to die young and leave a good-looking corpse?
He said, "If you have genes that put you at risk for a disease and you abuse your body, you're not going to make it very long."
Second, when it comes to living longer, would you put your bet on genes or on behavior?
He said, "Pick your parents carefully. Some genes will do you in."
Fine. But I'm going to the gym at lunch today, anyway.
With the margin between genes and environment that close, I say hedge your bets.
-- Roger Martin is a research writer and editor for the Kansas University Center for Research and editor of Explore, KU's research magazine Web site, which can be found at www.research.ukans.edu. Martin's e-mail address is firstname.lastname@example.org.