Limits...
From Lucy to Kadanuumuu: balanced analyses of Australopithecus afarensis assemblages confirm only moderate skeletal dimorphism.

Reno PL, Lovejoy CO - PeerJ (2015)

Bottom Line: In addition to a number of isolated specimens, the sample for this species includes two small associated skeletons (A.L. 288-1 or "Lucy" and A.L. 128/129) and a geologically contemporaneous death assemblage of several larger individuals (A.L. 333).These have driven both perceptions and quantitative analyses concluding that Au. afarensis was markedly dimorphic.Prevention of the dominance of a single specimen's contribution to calculations of multiple dimorphism estimates confirms that Au. afarensis was only moderately dimorphic.

View Article: PubMed Central - HTML - PubMed

Affiliation: Department of Anthropology, The Pennsylvania State University , University Park, PA , USA.

ABSTRACT
Sexual dimorphism in body size is often used as a correlate of social and reproductive behavior in Australopithecus afarensis. In addition to a number of isolated specimens, the sample for this species includes two small associated skeletons (A.L. 288-1 or "Lucy" and A.L. 128/129) and a geologically contemporaneous death assemblage of several larger individuals (A.L. 333). These have driven both perceptions and quantitative analyses concluding that Au. afarensis was markedly dimorphic. The Template Method enables simultaneous evaluation of multiple skeletal sites, thereby greatly expanding sample size, and reveals that A. afarensis dimorphism was similar to that of modern humans. A new very large partial skeleton (KSD-VP-1/1 or "Kadanuumuu") can now also be used, like Lucy, as a template specimen. In addition, the recently developed Geometric Mean Method has been used to argue that Au. afarensis was equally or even more dimorphic than gorillas. However, in its previous application Lucy and A.L. 128/129 accounted for 10 of 11 estimates of female size. Here we directly compare the two methods and demonstrate that including multiple measurements from the same partial skeleton that falls at the margin of the species size range dramatically inflates dimorphism estimates. Prevention of the dominance of a single specimen's contribution to calculations of multiple dimorphism estimates confirms that Au. afarensis was only moderately dimorphic.

No MeSH data available.


Demonstration of the calculation for the template method.(A) Many Au. afarensis specimens can be used to judge the species’ skeletal size distribution, but only a few represent the same skeletal site (horizontal axis) and therefore cannot be simply compared with one another. However, many sites are also found in the Lucy skeleton (diamonds). (B) The relative size of each non-Lucy specimen can therefore be calculated as a simple ratio (vertical axis) of each specimen to the same site as preserved in Lucy. These (dimensionless) ratios can then be used to compute a CV or BDI for the species. Note that the template ratios all equal 1 for the template specimen, so the choice of metric used for this individual has no effect on the dimorphism calculation. (C) To convert ratios to “real dimensions” each can be “normalized” using Lucy’s FHD. This has no effect on the sample’s dispersion (i.e., compare B to C), and therefore has no impact on the value of the CV and BDI when Lucy (or other specimen) is used as a “template.” See Table 1 for variable definitions.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4419524&req=5

fig-1: Demonstration of the calculation for the template method.(A) Many Au. afarensis specimens can be used to judge the species’ skeletal size distribution, but only a few represent the same skeletal site (horizontal axis) and therefore cannot be simply compared with one another. However, many sites are also found in the Lucy skeleton (diamonds). (B) The relative size of each non-Lucy specimen can therefore be calculated as a simple ratio (vertical axis) of each specimen to the same site as preserved in Lucy. These (dimensionless) ratios can then be used to compute a CV or BDI for the species. Note that the template ratios all equal 1 for the template specimen, so the choice of metric used for this individual has no effect on the dimorphism calculation. (C) To convert ratios to “real dimensions” each can be “normalized” using Lucy’s FHD. This has no effect on the sample’s dispersion (i.e., compare B to C), and therefore has no impact on the value of the CV and BDI when Lucy (or other specimen) is used as a “template.” See Table 1 for variable definitions.

Mentions: While large samples exist of Au. afarensis postcranial fossils, relatively few represent the same anatomical loci. As such, much of the variation in the sample reflects size differences between anatomical parts (Fig. 1A). Fortunately, the existence of a relatively complete skeleton can serve as a template to account for the differences in size between these sites. To apply the template method, the metric from each fossil is converted to a simple size ratio by dividing each by the homologous metric from a template specimen (Lucy or Kadanuumuu) (Fig. 1B). This calculation preserves the relative dispersion of the individuals within each variable and enables these dimensionless size ratios to be compared across variables. To standardize size ratios and to facilitate inspection of the data, we multiplied each by a reference metric commonly used in size estimation: femoral head diameter (FHD) for the Lucy template and distal tibia breadth (DSTB) for Kadanuumuu (Table 1). This step has no effect on the estimate of dimorphism because it is simply multiplying each ratio by a constant leaving the dispersion of the sample unchanged (Fig. 1C). The sample is then used to compute the Coefficient of Variation (CV) and Binomial Dimorphism Index (BDI) as in previous analyses (Lovejoy et al., 1989; Reno et al., 2003) with each specimen providing an equal contribution to the dimorphism estimate.


From Lucy to Kadanuumuu: balanced analyses of Australopithecus afarensis assemblages confirm only moderate skeletal dimorphism.

Reno PL, Lovejoy CO - PeerJ (2015)

Demonstration of the calculation for the template method.(A) Many Au. afarensis specimens can be used to judge the species’ skeletal size distribution, but only a few represent the same skeletal site (horizontal axis) and therefore cannot be simply compared with one another. However, many sites are also found in the Lucy skeleton (diamonds). (B) The relative size of each non-Lucy specimen can therefore be calculated as a simple ratio (vertical axis) of each specimen to the same site as preserved in Lucy. These (dimensionless) ratios can then be used to compute a CV or BDI for the species. Note that the template ratios all equal 1 for the template specimen, so the choice of metric used for this individual has no effect on the dimorphism calculation. (C) To convert ratios to “real dimensions” each can be “normalized” using Lucy’s FHD. This has no effect on the sample’s dispersion (i.e., compare B to C), and therefore has no impact on the value of the CV and BDI when Lucy (or other specimen) is used as a “template.” See Table 1 for variable definitions.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4419524&req=5

fig-1: Demonstration of the calculation for the template method.(A) Many Au. afarensis specimens can be used to judge the species’ skeletal size distribution, but only a few represent the same skeletal site (horizontal axis) and therefore cannot be simply compared with one another. However, many sites are also found in the Lucy skeleton (diamonds). (B) The relative size of each non-Lucy specimen can therefore be calculated as a simple ratio (vertical axis) of each specimen to the same site as preserved in Lucy. These (dimensionless) ratios can then be used to compute a CV or BDI for the species. Note that the template ratios all equal 1 for the template specimen, so the choice of metric used for this individual has no effect on the dimorphism calculation. (C) To convert ratios to “real dimensions” each can be “normalized” using Lucy’s FHD. This has no effect on the sample’s dispersion (i.e., compare B to C), and therefore has no impact on the value of the CV and BDI when Lucy (or other specimen) is used as a “template.” See Table 1 for variable definitions.
Mentions: While large samples exist of Au. afarensis postcranial fossils, relatively few represent the same anatomical loci. As such, much of the variation in the sample reflects size differences between anatomical parts (Fig. 1A). Fortunately, the existence of a relatively complete skeleton can serve as a template to account for the differences in size between these sites. To apply the template method, the metric from each fossil is converted to a simple size ratio by dividing each by the homologous metric from a template specimen (Lucy or Kadanuumuu) (Fig. 1B). This calculation preserves the relative dispersion of the individuals within each variable and enables these dimensionless size ratios to be compared across variables. To standardize size ratios and to facilitate inspection of the data, we multiplied each by a reference metric commonly used in size estimation: femoral head diameter (FHD) for the Lucy template and distal tibia breadth (DSTB) for Kadanuumuu (Table 1). This step has no effect on the estimate of dimorphism because it is simply multiplying each ratio by a constant leaving the dispersion of the sample unchanged (Fig. 1C). The sample is then used to compute the Coefficient of Variation (CV) and Binomial Dimorphism Index (BDI) as in previous analyses (Lovejoy et al., 1989; Reno et al., 2003) with each specimen providing an equal contribution to the dimorphism estimate.

Bottom Line: In addition to a number of isolated specimens, the sample for this species includes two small associated skeletons (A.L. 288-1 or "Lucy" and A.L. 128/129) and a geologically contemporaneous death assemblage of several larger individuals (A.L. 333).These have driven both perceptions and quantitative analyses concluding that Au. afarensis was markedly dimorphic.Prevention of the dominance of a single specimen's contribution to calculations of multiple dimorphism estimates confirms that Au. afarensis was only moderately dimorphic.

View Article: PubMed Central - HTML - PubMed

Affiliation: Department of Anthropology, The Pennsylvania State University , University Park, PA , USA.

ABSTRACT
Sexual dimorphism in body size is often used as a correlate of social and reproductive behavior in Australopithecus afarensis. In addition to a number of isolated specimens, the sample for this species includes two small associated skeletons (A.L. 288-1 or "Lucy" and A.L. 128/129) and a geologically contemporaneous death assemblage of several larger individuals (A.L. 333). These have driven both perceptions and quantitative analyses concluding that Au. afarensis was markedly dimorphic. The Template Method enables simultaneous evaluation of multiple skeletal sites, thereby greatly expanding sample size, and reveals that A. afarensis dimorphism was similar to that of modern humans. A new very large partial skeleton (KSD-VP-1/1 or "Kadanuumuu") can now also be used, like Lucy, as a template specimen. In addition, the recently developed Geometric Mean Method has been used to argue that Au. afarensis was equally or even more dimorphic than gorillas. However, in its previous application Lucy and A.L. 128/129 accounted for 10 of 11 estimates of female size. Here we directly compare the two methods and demonstrate that including multiple measurements from the same partial skeleton that falls at the margin of the species size range dramatically inflates dimorphism estimates. Prevention of the dominance of a single specimen's contribution to calculations of multiple dimorphism estimates confirms that Au. afarensis was only moderately dimorphic.

No MeSH data available.