On silk purses and pig's ears

This new paper by Jay Kelley (University of Illinois, Chicago) is about as close to a detective story that paleontologists get (via Palanthsci message board). Here's an excerpt from the introduction:

This paper concerns the regular misidentification for nearly 100 years of a number of non-primate upper canine teeth as belonging to the Miocene, Siwalik hominoid Sivapithecus. The same misidentification was repeatedly made by numerous paleontologists who collected in the Siwaliks, beginning with Guy Pilgrim. It went unrecognized by every hominoid expert who has either collected in the Siwaliks or analyzed the Siwalik hominoid collections, including me until recently (Kelley 2005:2).

The basic story is that the sample of Sivapithecus upper canine teeth has included many that actually belong to an extinct pig. The morphology of these canines is atypical for hominoid canines, but their identity as hominoid teeth has apparently not been questioned previously. Instead, analysts have suggested that the teeth represent evidence for taxonomic distinctions within the genus, such as a division into three species based on relative male canine size (Greenfield 1979), or a division into two temporal species based on a segregation of canine anatomy into earlier and later samples (Kelley 1986).

The article is a great description of comparative anatomy at work, and its taxonomic consequences. Here is part of the conclusion:

The most significant implication of removing the atypical canines from Sivapithecus is that there is no longer any clear morphological justification for recognizing S. indicus and S. sivalensis as time-successive species. Accepting the stated provenance of the two Chinji canines with a typical Sivapithecus morphology, GSI D. 238 and BMNH M34438, there are no discernable differences in hominoid upper-canine size or morphology between older and younger levels in the Siwaliks. While there are suggestions of other differences in the Sivapithecus samples from the Chinji and Dhok Pathan Formations, for example in tooth proportions (Kelley 1988), that might indicate the presence of different species, these have not been systematically assessed (Kelley 2005:8).

The other conclusion is that sexual dimorphism in Sivapithecus should be reassessed without the pig canines. According to Kelley, this is problematic because the upper canines of only a single female individual have been recovered, along with those of several males. Taking this single female as average, the level of canine dimorphism is consistent with orangutans or gorillas, again according to Kelley.

To me, this is the best kind of story -- it seems so obvious in retrospect, but somehow everyone missed it until the right person came along asking the right question.


Greenfield LO. 1979. On the adaptive pattern of "Ramapithecus." Am J Phys Anthropol 50:527-548.

Kelley J. 1986. Paleobiology of Miocene hominoids. Ph.D. Dissertation, Yale University.

Kelley J. 1988. A new large species of Sivapithecus from the Siwaliks of Pakistan. J Hum Evol 17:305-324.

Kelley J. 2005. Misconceptions arising from the misassignment of non-hominoid teeth to the Miocene hominoid Sivapithecus Paleontologica Electronica 8:16A. Paleontologica Electronica online