Greg Laden's Blog

... according to a new study published in the Proceedings of the National Academy of Sciences. The study compares skull measurements of Flores material with a wide range of other hominid data and concludes that Flores cannot be clustered with Homo sapiens. This is the first published study that takes into account how size affects shape. By correcting for size, this study makes, the authors claim, a more valid comparison between measurements taken on the Flores material and other comparative data.

We show that whether or not the effects of its small cranial size are accounted for, the external cranial morphology of the [Flores] LB1 cranium cannot be accommodated within a large global sample of normal modern human crania. Instead, the shape of LB1, which is shown by multivariate analysis to differ significantly from that of modern humans, is similar to that of Homo erectus sensu lato, and, to a lesser extent, Homo habilis. Our results are consistent with hypotheses that suggest the Liang Bua specimens represent a diminutive population closely related to either early H. erectus s. l. from East Africa and/or Dmanisi or to H. habilis.

As you probably know, there have emerged two distinct kinds of interpretation of the Flores hominid. One suggests that Flores were humans who were very different because of pathological effects. For instance, one recent study suggested that the Flores hominids "are myxoedematous endemic (ME) cretins, part of an inland population of (mostly unaffected) Homo sapiens. " (see this discussion). The alternative interpretation is that Flores was either a diminutive population of Homo erectus, or a separate species (still diminutive).

The present study collected (from various sources) six cranial variables for a very large samle of modern humans and 30 fossil hominid crania, to compare with the same measurements take on the LB1 Flores cranium. These data were then subjected to a series of fairly typical statistical processes, some controlling for size others not. The result of this is a number of statistical measures and graphics relating LB1 to other hominid data. Here is a typical example:

This is a plot of principle components (PC1, PC2 and PC3 on the graph) derived from these data. A principle component is the result of a statistical analysis that reduces a number of actual, measured (usually) data vectors into a smaller number, such that the first PC is a derived variable that accounts for a certain chunk of the variation in the data, the second PC a lesser chunk, the third PC an even lesser chunk, etc. This can allow, as you can see in this case, for a three dimensional picture of the relationship among data points.

In this case, the big blob of black dots is the modern human sample,
and the various colored circles represent a variety of other homind skulls. We can see that Neanderthal (a brown circle) is close to but not deeply within the modern human sample, and that variou searly homonids, including australopiths, H. erectus, and H. habilis form a group floating off in the distance.

The Flores cranium, the "white" or "clear" circle with an L in it, is out in this cluster. of early hominids. Several different analysis done in this study place Flores in a variety of locations nearer or father from modern humans or other hominids. The authors suggest:

This clustering pattern demonstrates that

• (i) although the six variables selected for this study do not sort hominin crania into species, crania are sorted in a manner that corresponds to conventional taxonomic interpretations;


• (ii) major morphological differences distinguish clusters; and


• (iii) when shape is considered in the absence of scaling patterns, LB1 appears most similar to the non-Asian H. erectus specimens D2700 (from Dmanisi, Georgia) and KNM-ER 3733 (from Koobi Fora, Kenya) and is distinct from later Homo.

Furthermore, [this analysis] shows that LB1 is closest to ...non-Asian H. erectus, followed by ... all H. erectus (including Sangiran 17)
... L

The authors claim that the various pathological models are not supported by this result. Furthermore, they indicate that while phylogeny may be difficult to determine at this point because of a lack of appropriate comparative materials with H. erectus, the most likely interpretation is that Flores is a group derived from some population of erectus, but that the possibility that Flores is more closely related to the habilines cannot be ruled out.

.

Gordon, A.D., Nevell, L., Wood, B. (2008). The Homo floresiensis cranium (LB1): Size, scaling, and early Homo affinities. Proceedings of the National Academy of Sciences, 105(12), 4650-4655. DOI: 10.1073/pnas.0710041105