The rate of growth in height and the timing of adolescent growth events are analyzed for two samples of Guatemalan children. One sample includes Mayan school children, 33 boys and 12 girls between the ages of 5.00 to 17.99 years, living under poor conditions for growth and development. The second sample includes ladino children, 78 boys and 85 girls of the same age range, living under favorable conditions for growth. The Preece-Baines model I function is used to estimate mean values for rates and timing of childhood and adolescent growth events for the two groups. Significant statistical contrasts (t-tests) of these means show Mayan boys reach the age of "take-off" (TO; the onset of the adolescent growth spurt) 1.45 years later, achieve peak height velocity (PHV) 1.68 years later, and continue growing for about 2.0 years longer than do the ladino boys. Despite the Mayan boys' increased duration for growth they grow significantly more slowly than the ladinos. Mayan boys are 6.60 cm shorter than ladinos at the age of TO and are estimated to be 7.71 cm shorter than the ladinos at adulthood. Mayan girls reach the age of TO 0.93 years later than do the ladina girls, but the two groups do not differ in the age at PHV or the age at adulthood. The mean height of Mayan girls is significantly less than that of ladinas at the age of TO (6.5 cm), and this difference increases to an estimated 11.14 cm at adulthood. Possible causes of these ethnic and sex-related differences in amounts and rates of growth are discussed in relation to hypotheses about the genetic and environmental determinants of human development.

The Preece-Baines model I function, adapted for use with a personal computer, is applied to the longitudinal growth records of Guatemalan children and adolescents of high socioeconomic status. The fit of the Preece-Baines function to the Guatemalan data is compared with those of published analyses of the function fitted to the growth of British, Belgian, urban and rural Indian, Australian Aborigine, and African children. Guatemalan, British, and Belgian samples share generally favorable environments for human development and show few differences in the amount and velocity of growth, or in the timing of growth events. Urban Indians live under relatively good environmental conditions and are similar to Guatemalans in the timing of growth events, but grow more slowly and grow less than the Guatemalans, British, or Belgians. Rural Indian, Australian, and African samples live in environments that delay or retard growth, and these last-named three samples grow more slowly, delay the onset of the adolescent growth spurt, and achieve smaller adult height than the Guatemalans. Parameters of the Preece-Baines model are compared between all samples and show that there are several alternate paths in the rate of growth and the timing of adolescent growth events that may be taken from childhood to adulthood.