Infant growth: a sporadic phenomenon.

Anthropologist Michelle Lampl, trained as a physician and growth researcher, recalls learning that healthy young children experience "a perfectly regular rate of growth, with no breaks or spurts." Indeed, she says, she recently listened to a growth researcher tell pediatricians that any child who does not grow over a 30-day period must be dead.

But Lampl's own findings challenge the notion of smooth, consistent growth, and instead suggest that children grow in sporadic fits and starts.

Conventional wisdom holds that very young children gain an average of about one-half millimeter in body length per day. But Lampl's measurements of 32 healthy infants and one adolescent indicate that growth occurred in a random series of roughly 1-centimeter spurts, each apparently lasting less than 24 hours. During the two to approximately 60 days that separated successive spurts, she says, absolutely no increases in body length occurred.

Lampl, of the University of Pennsylvania in Philadelphia, described her new study in Chicago this week at a meeting of the American Association for the Advancement of Science.

Over periods of four to 18 months, she visited the homes of the youngsters. With the help of a parent, she stretched out and measured each child -- four of them daily, 18 twice-weekly and 11 at weekly intervals. Her daily measurements "provide the most precise description of growth yet reported," she says.

The daily data document long quiescent periods of no growth, suddenly punctuated by a permanent lengthening of 0.5 to 1.8 cm over a 24-hour period. Lampl describes these nonperiodic, stepwise changes in the growth curve as "saltatory," or abruptly jumping. Because saltatory spurts of similar magnitude showed up in children who were measured weekly or twice-weekly, she suspects that these growth changes occurred over a 24-hour period.

While conceding she used a very small study population, Lampl says she saw no signs of a correlation between infant size and the total number of discrete growth episodes. However, she notes, "there was a distinct correlation between fussiness and [increased] hunger at the time of the growth episodes." Parents also reported signs of increased sleepiness right before growth spurts, she adds.

At the same meeting, Michael Hermanussen of the University of Kiel in Germany described a study of lower-leg length in healthy schoolchildren. He found evidence of weekly changes, with growth spurts following no-growth periods that sometimes lasted more than 60 days (including occasional intervals of shrinkage). "I was not aware of saltatory [changes in these data]," he says. However, he adds, "I'm aware that I might have missed them."

That's not surprising, says Lampl, because until now, growth researchers have lacked mathematical models for stepwise changes that are nonperiodic. Without such models, they have attempted to fit their growth data points -- usually collected weeks or months apart -- to a smooth curve. But Lampl found that such a curve didn't really fit her detailed data.

For help in finding a better curve, she turned to biophysicist Michael L. Johnson, handing over her data on a 13-year-old boy whose height she had measured on about 400 consecutive days. Johnson, of the University of Virginia in Charlottesville, reported at the Chicago meeting that a stepwise, saltatory model fits these data better than any previous model. Without Lampl's daily data on dormant periods and growth pulses, the flaws in the old approach remained unrecognized, he adds.

"I would never have imagined pulsatile growth," comments Mark L. Hartman of the University of Virginia School of Medicine, who studies factors affecting growth-hormone secretion and its relationship to human growth.

The body's pulsed secretion of growth hormone can trigger metabolic changes, such as increased protein synthesis, Hartman notes. And since his group has recently shown that growth-hormone pulses occur frequently throughout the day, sometimes at intervals of just 30 seconds, some of his colleagues suspected that these pulses contribute to slow, incremental daily growth. But Lampl's results certainly confound that picture, he adds.

"My mind is going 100 miles a minute trying to explain the new data," Hartman told SCIENCE NEWS. "I think I will have to talk to my colleagues and see if we can generate some new ideas to explain these new findings."