Biology Reference
In-Depth Information
the United States in 1990 2 and the dawn of the genomic era. For a discussion of the value of
traditional morphometrics in an age of geometric morphometrics and ancient DNA analyses,
see Pietrusewsky (2008) .
Morphometric research within biological anthropology has been particularly fruitful.
Studies using cranial measurements, such as Howells' (1973) analysis of samples from
around the world and Jantz's (1973) analyses of regionally defined samples, clearly demon-
strated the capacity of these approaches for exploring morphological variation and inte-
grating evolutionary models. While measurements of postcranial elements were employed
to estimate sex and stature, explore secular changes (phenotypic differences in a population
through time), and evaluate health status among other things, the primary focus of morpho-
metric research in skeletal biology was the study of biological distance (biodistance), the
evolutionary relationships of groups based on phenotypic similarity or dissimilarity ( Buik-
stra et al., 1990; Stojanowski and Schillaci, 2006 ).
Initial applications of morphometrics to studies of biodistance primarily focused on two
areas. One was explanations of change in craniofacial morphology as a product of migration
versus in situ change due to evolutionary forces or secular change. The other was ancestry
estimation for application in modern forensic casework (i.e., Giles and Elliot, 1962 ). Many
of these studies employed canonical variates analysis (CVA) to explore variation among
groups and in some cases construct possible temporal sequences. This analysis is quite
appealing as it generates Mahalanobis distance (D 2 ) among groups and the canonical scores
can be plotted in two or three dimensions for visual interpretation of the degree of pheno-
typic similarity or dissimilarity present. In many cases Mahalanobis distances (D 2 ) were
used as biological distances between populations using the assumption that smaller Maha-
lanobis distances reflect greater phenotypic similarity and more closely related groups ( Buik-
stra et al., 1990; Stojanowski and Schillaci, 2006 ). Others used discriminant function analysis
(DFA) to classify an unknown specimen based on multivariate distance to reference group
centroids or a DFA score (e.g., Giles and Elliot, 1962, 1963 ). While these studies were infor-
mative, the inability to account for environmental influences on the phenotype limited inter-
pretations and permitted criticism by those who saw craniofacial morphology as containing
minimal genetic information.
In 1982, Relethford and Lees outlined the application of population genetic analyses to
quantitative traits (i.e., morphometrics) and began the integration of quantitative genetic
models into morphometric research designs. Relethford and Blangero (1990) developed an
extension of the model from Harpending and Ward (1982) for calculating genetic variation
based on multivariate quantitative traits such as morphometrics. The theoretical framework
for themodel holds that expected levels of heterozygosity are proportional to total phenotypic
variation in the population, and, as such, parameter estimates such as gene flow and genetic
drift can be made. The model also permits heritability to be less than 1. This means that the
amount of phenotypic variance in a trait within a population can be explained by partitioning
environmental and genetic transmission. For example, many studies have used an average
heritability for craniometric traits of 0.55 (after Devor, 1987), which means that, overall, the
additive genetic transmission to the total phenotypic variance for craniometric traits is on
2 Public Law 101-601; 25 U.S.C.
x
3001 et seq. Also see DiGangi and Moore (Chapter 1), this volume, for
a discussion.
Search WWH ::




Custom Search