Volume 125 (2017) Issue 2 Pages 67-83
Homo erectus is among the best-represented fossil hominin species, with a particularly rich record in Indonesia. Understanding variation within this sample and relative to other groups of H. erectus in China, Georgia, and Africa is crucial for answering questions about H. erectus migration, local adaptation, and evolutionary history. Neurocranial shape is analyzed within the Indonesian sample, including representatives from Sangiran, Ngandong, Sambungmacan, and Ngawi, as well as a comparative sample of H. erectus from outside of Java, using three-dimensional geometric morphometric techniques. This study includes several more recently described Indonesian fossils, including Sambungmacan 4 and Skull IX, producing a more complete view of Indonesian variation than seen in previous shape analyses. While Asian fossils can be distinguished from the African/Georgian ones, there is not a single cranial Bauplan that distinguishes all Indonesian fossils from those in other geographic areas. Nevertheless, late Indonesian H. erectus, from sites such as Ngandong, are quite distinct relative to all other H. erectus groups, including earlier fossils from the same region. It is possible that this pattern represents a loss of genetic diversity through time on the island of Java, coupled with genetic drift, although other interpretations are plausible. A temporal pattern of diachronic change was identified within Indonesia for the posterior neurocranium such that younger Sangiran fossils more closely approached the Ngandong/Sambungmacan/Ngawi pattern, but there was not a linear trend of shape change from Sangiran to Sambungmacan to Ngandong, as has been suggested previously. The Sambungmacan 3 fossil, which often appears as a morphological outlier, fits the general pattern of late Indonesian vault shape, but has a more extreme expression of the shape trends for this group than other individuals.