Sources of biological variation. Is sex really important?

Abstract

Metric characters are often discretely defined using means while variation of these traits is given less attention. Sex is but one of many factors contributing to individual variation. Although averages of various male and female characters may differ statistically significantly, there is wide overlap of male and female distributions. In paleoanthropological studies size differences between individual fossils are either interpreted as taxic or sex, while individual variation is largely ignored. Physical anthropologists have long argued that the amount of variation between different populations is so small in relation to the total human variation that the ‘race’ is not an important explanatory category whilst nobody questions categorizing data into male and female. We partitioned variance in cranial characters of extant humans into that resulting from sexual dimorphism, population affinity and from differences between individual males and individual females in the same population. Data used included body weight, face width, nose height, minimum frontal breadth and soft tissue depths in six locations on human faces. Typically, the largest portion of variance (>50%) in these characters resulted from individual differences, while sex and population affinity contributed only about 20% each. Why then do we stake so much on determining that a particular individual is a male or a female while differences between individuals of the same sex may be much greater?