Hominin Relations
This unit will explore recent developments and current thinking about how early hominins were evolutionarily related to one another. This subject—phylogeny—has always attracted the attention of anthropologists, often overshadowing the more basic questions of hominin biology, such as subsistence strategies and behavior.
During the first half of the twentieth century, scholars commonly assigned a new species name to virtually each new fossil unearthed. In this "splitting" paradigm, each variant in anatomical structure was taken as indicating a separate species. (See figure 22.1.) The result was a plethora of names in the hominoid record. In 1965, Elwyn Simons and David Pilbeam, both then at Yale University, rationalized this paleontological mess and reduced the number of genera and species to a mere handful (the "lumping" paradigm). (See figure 22.2.)
Lumping became the guiding ethic of anthropology. Taken to its extreme, it led to the "single-species hypothesis," which became popular during the 1960s and early 1970s (see unit 3). Although the single-species hypothesis is no longer considered valid, there is a persisting tendency to interpret anatomical differences as within-species variation rather than among-species variation. One reason for this trend is that, because of the nature of the system, no practical guide has been developed to explain how much anatomical difference between two fossils signals the existence of separate species. "The reason for this is, of course, that there is no direct relationship, indeed no consistent relationship at all, between speciation and morphological change," says Ian Tattersall, an anthropologist at the American Museum of Natural History (see unit 4).
In other words, a daughter species might sometimes diverge from the parental species but develop very little obvious anatomical difference, while considerable differences might arise in other cases. Unless the living animals are available so that you can observe their behavior, it is often impossible to know whether the individuals belong to one
fioure 22.1 Splitters and lumpers:
Louis Leakey (seated) was a keen splitter, reflecting the philosophy of his time; his son Richard Leakey was more cautious, reflecting changing times. (Courtesy of the L. S. B. Leakey Archives.)
ficure 22.2 Lumpers and splitters:
Different philosophical and methodological approaches yield different views of the species richness of the fossil record. In its early years, anthropology was dominated by splitters, which yielded a plethora of species. Sentiment then switched to lumping, which underestimated species richness. Recently, a swing away from lumping has occurred, but not a return to the previous excesses.
Lumpers
Splitters
Fewer species
More species
Anatomical variation seen as: /ntraspecific Interspecific species or two. As a result, it is obviously easier to subsume anatomical differences under within-species variation rather than to argue for separate species. This tendency has certainly become a tradition in anthropology. The result, argues Tattersall, "is simply to blind oneself to the complex realities of phylogeny." In other words, the true hominin family tree—the one that actually happened in evolutionary history —almost certainly is more bushy than the ones currently drawn by anthropologists.
Although most anthropologists would regard Tattersall's position as somewhat extreme, many are coming to accept that hominin phylogeny is more complex than it is usually portrayed. This view was emphasized by the rethinking provoked by the 1985 discovery of the "black skull" (KNM-WT 17,000), a robust australopithecine (Australopithecus aethio-picus) that did not immediately fit into the prevailing phylo-genetic picture (see unit 20), and by other recent discoveries (see unit 19). Cladistic methodology appears to offer the most promising approach for overcoming the problem of lumping (see unit 8).
Which data are the most REUAbLE phylo^^^ indicators?
Paleontologists reconstruct phylogenies from comparisons of anatomical similarities present in fossil specimens. As discussed in unit 8, only those similarities that result from a shared evolutionary history (homologies) can reliably lead to accurate phylogenies. Similarities that result from independent, parallel evolution (homoplasies) may lead to erroneous phylogenies. Most anthropologists now accept that homo-plasy has been common in hominin evolution but, as we will see later in this unit, less agreement has been reached regarding which traits are homoplasies between certain lineages and which are not. Again, cladistic analysis should, in principle, help resolve this issue.
A further obstacle to accurate phylogenetic reconstruction arises from the way in which different traits are treated. In anthropology, phylogenetic reconstruction is based almost exclusively on cranial traits, for the very good reason that postcranial fossils are much rarer. In one of the more complete cladistic analyses of hominin phylogenetics, Randall
Skelton, of the University of Montana, and Henry McHenry, of the University of California, Davis, employed 67 such traits. In their 1992 paper, Skelton and McHenry addressed the issue of the values assigned to these traits, identifying two problems: the independence of the traits, and sample bias.
If all 67 traits were independent, then they would provide information on 67 evolutionary transformations, forming a powerful body of evidence. Anatomical traits are not independent, however, but form parts of trait complexes. For instance, an important trend in early hominin evolution was toward heavy chewing in order to process tough plant foods. This development is seen, for instance, in an increase in the size of molar teeth and in the thickness and depth of the mandible. Bigger teeth and more powerful chewing also require a more robust mandible, changes in face structure, and possibly alterations in the mechanics of muscles that move the jaws. Changes in the size of molar teeth and the robusticity of the mandible are therefore linked as part of an evolutionary package and are not independent of one another.
Thus, phylogenetic analyses should logically group traits into functional packages, rather than treat them as independent. In their analysis of hominin phylogeny, Skelton and McHenry identified five such functional complexes among the 67 traits: heavy chewing (34 traits), anterior dentition (11 traits), basicranium flexion (11 traits), prognathism/ orthognathism (8 traits), and encephalization (3 traits). Grouped in this way, the 67 traits give phylogenetic information on just five evolutionary transformations—not 67. (See figure 22.3.) Even these five functional complexes are not completely independent, however, because the masticatory system involves many parts of the cranium. For instance, the evolution of traits associated with anterior dentition is linked in part to the evolution of heavy chewing, as is the shape of the face and certain cranial traits, such as the possession of a sagittal crest.
The second problem of bias sampling is evident from the traits listed above—namely, some aspects of anatomy are more widely represented than others in the fossil record. Traits associated with heavy chewing are obviously the most common, because teeth and jaws are the most resilient parts of the cranium and consequently become part of the fossil record much more frequently. For this reason, anthropologists
- figure 22.3 Interdependence of characters: Individual anatomical traits are typically parts of functional complexes and are not evolutionarily independent. These five functional complexes are associated with the hominin cranium.
have concentrated much of their work on teeth and jaws, including basing phylogenetic reconstruction on them. Teeth and jaws, however, are particularly susceptible to homo-plasy: species with similar diets will develop similar dentition through natural selection. Teeth and jaws, and their interpretation, may therefore receive more attention than their phylogenetic reliability justifies.
KEy questions in hominin phyLogEnv
Three key questions arise in a phylogenetic reconstruction of early hominins:
- The relationship of Australopithecus afarensis to earlier and later hominins.
- The relationships among the robust australopithecines (A. aethiopicus, robustus, and boisei).
- The origin of the genus Homo. (See figure 22.4.)
As we saw in unit 19, two decades after the first specimens of A. afarensis were discovered no consensus had been reached on whether they represent one extremely sexually dimorphic species or two less variable species (one large and one small). Until recently, the majority view held that just one species was present between 3.9 and 2.9 million years ago, and that this species was ancestral to all later hominins. The discoveries of more A. afarensis fossils from Ethiopia did not resolve this difference of opinion. (See unit 19.)
The recent discoveries of hominins earlier than A. afarensis proved false the often implicit assumption that A. afarensis was the founding species of the hominin clade. The likelihood that ramidus and anamensis, for instance, were part of a bushy phylogeny prior to afarensis, rather than being stages in a single, transforming lineage, impacts the status of afarensis. (The notion of a single, transforming lineage does have its supporters, however.) It is unlikely that a phylogenetically bushy clade would be reduced to a single species, which then gives rise to further bushiness. Unlikely—but not impossible. Further fossil finds in the period 5 to 3 million years ago will be necessary to resolve this issue.
The question of robust australopithecine relationships affects the placement of A. aethiopicus in the evolutionary tree: Is it ancestral to the two later robust australopithecines, or is it separate from them? The issue of the origin of the genus Homo concerns the identity of its direct ancestor: Is it A. afarensis, A. africanus, or some as yet unknown third species? These two questions will be considered through Skelton and McHenry's cladistic analysis, not because it is universally accepted (it is not, but it is widely respected), but because it offers a strategy for addressing some key problems, particularly that of homoplasy.
The Skelton/McHenRy analysis
Skelton and McHenry performed a cladistic analysis of the 67 cranial traits in several ways: they treated the traits as if they were independent; they compared the five functional complexes discerned; and they grouped the traits by anatomical region (face, anterior dentition, posterior dentition,
Relationship of A. afarensis to early and later hominins
Relationships among the robust australopithecines
Origin of the genus Homo
Relationship of A. afarensis to early and later hominins
Relationships among the robust australopithecines
Origin of the genus Homo figure 22.4 Key questions in early hominin evolution.
mandible, palate, basicranium, and cranial vault), which is another way of overcoming linkage between traits. They then compared the results from these various analyses. Their study was performed prior to the discovery of Ardipithecus ramidus and Australopithecus anamensis, and it took the conservative position that Australopithecus afarensis is indeed a single species. The analysis of the later hominins is unaffected by these recent discoveries. One of the most important, and controversial, conclusions of their work was that traits associated with heavy chewing in hominins are subject to homoplasy.
Mentioned earlier was the trend in early hominin evolution toward ever-heavier chewing. Traits associated with heavy chewing are least developed in A. afarensis and most strongly developed in A. boisei. The black skull, A. aethiopicus, also possesses large cheek teeth and a robust mandible, which many anthropologists interpret as indicating an ancestral relationship to A. boisei and the South African robust australopithecine, A. robustus. The anterior dentition of A. aethiopicus, however, is more similar to that of A. afarensis than to that of the other robust australopithecines. The degree of prognathism in A. aethiopicus resembles that in A. afar-ensis, while the other robust australopithecines are much less prognathic and more similar to Homo. The most parsimonious tree from a phylogenetic analysis using only traits related to the functional complex of heavy chewing gives a cladogram that links all three robust australopithecines as a clade. Analyses using posterior dentition, an anatomical region associated with heavy chewing, produce the same phylogenies.
By contrast, most other types of analysis (taking the 67 traits independently, and assessing the other functional and regional complexes, either independently or grouped) yield a different series of possible cladograms, with one being most common. (See figure 22.5.) In this tree, A. aethiopicus is not ancestral to the other australopithecines, but rather a large-toothed form of A. afarensis that became extinct with no descendants. The persistence of this particular cladogram is evidence of its strength, say Skelton and McHenry, which implies that the traits associated with heavy chewing shared by A. aethiopicus and the other two robust australopithecines are homoplasies—not the result of common ancestry. A second aspect of Skelton and McHenry's phylogeny that differs from phylogenies constructed by other workers is its proposal of a close link between the other robust australop-ithecines (A. boisei and robustus) and earliest Homo (discussed below). The proposed phylogeny requires three hypothetical ancestors—species that are as yet unknown, but are implied by the evolutionary transitions in the phylogeny.
Skelton and McHenry's phylogeny is as follows. Australopithecus afarensis is the most primitive early hominin after Ardipithecus ramidus and Australopithecus anamensis, from which it probably derived (see unit 19). They propose that afarensis gave rise to an as yet unknown species that was aethiopicus-like in some ways (in traits not related to heavy
- 0
- 0
| 3.0
- 0
- 0
e 3.0
- 0
- 0
Hadarl i 3 fl
■ Hadar and Laetoli
I anamensis
- ramidus Hypothesis one
- ramidus Hypothesis one e 3.0
- anamensis
ramidus Hypothesis three
- 0
- 0
- 0
- 0
- 0
- 0
- 0
- 0
- 0
- 0
- 0
- 0
- 0
- 0
anamensis
17000
17000
anamensis ramidus Hypothesis two ramidus Hypothesis two afarensis I
anamensis I ramidus I Hypothesis four afarensis I
anamensis I ramidus I Hypothesis four ficure 22.5 A forest of hominin evolutionary trees:
Numerous phylogenetic interpretations of hominin history have been proposed. Hypothesis 4 is based on Skelton and McHenry's analysis, and shows the hypothetical ancestors as open boxes. Hypothesis 3 shows the three robust australopithecines as being monophyletic.
chewing); this species was the common ancestor of aethio-picus on one hand, and gave rise to A. africanus, early Homo, and the later robust australopithecines on the other. Australopithecus aethiopicus is therefore viewed as a side branch that became extinct, while A. afarensis was ancestral to all later hominins (but was not their common ancestor). Australopithecus africanus is derived from the aethiopicus-like ancestor, and in its turn gave rise to another proposed africanus--like species; this species was the common ancestor of earliest Homo on one hand and the robust australopithecines (via a proposed robustus-like common ancestor) on the other. Many anthropologists agree that robustus-like anatomy is likely to be ancestral to boisei. The close relationship between Homo and A. robustus and A. boisei (they share a common ancestor to the exclusion of other hominins) is reflected in a more flexed cranial base, a deeper jaw joint, less prognathism, and greater encephalization compared with A. africanus.
This phylogenetic scheme, like other proposed alternatives, implies considerable homoplasy in hominin evolution, particularly in the heavy chewing complex. In contrast to Skelton and McHenry's proposal, other schemes have proposed that A. aethiopicus was ancestral to the other robust australopithecines, and that heavy chewing traits are homologous (not homoplasic). A recent cladistic analysis by David Strait and Frederick Grine, at the State University of New York, Stony Brook, strongly supports this view (the mono-phyly of the three robust australopithecines). This phylogeny shifts the requirement for homoplasy to other traits—namely, anterior dentition, basicranial flexion, encephalization, and prognathism/orthognathism—that A. aethiopicus shares with other species.
A second area of homoplasy appears in the evolution of Homo. The shape of the face and small cheek teeth superficially resemble those of A. afarensis. Thus, these traits in Homo must have resulted from the retention of primitive traits present in afarensis, in which case afarensis would be the direct ancestor of Homo, or via a reversal of the hominin trend, in which case africanus would be the ancestor. A study of the ontogeny of facial development reveals that the formation of facial anatomy in Homo is unique, not a primitive retention. The well-documented reduction in the size of cheek teeth later in the Homo lineage also leads to the conclusion that this trend began with early Homo, and thus was not a primitive retention at this stage. If, as Skelton and McHenry point out in their analysis, the face and dentition of Homo are indeed uniquely derived, then these traits provide no useful information about the large-toothed australopithecine (known or yet to be discovered) from which it evolved; other, shared traits, such as basicranial flexion and orthog-nathism, are necessary to link Homo to A. africanus.
Skelton and McHenry's preferred phylogeny is one of several that can been seen in the anthropological literature;
its strength, however, lies in its cladistic methodology and thoughtful treatment of potential biases. Many other schemes derive more than one lineage from A. afarensis, for instance, and designate A. aethiopicus as the ancestor of the other robust australopithecines. The most controversial aspect of the Skelton/McHenry phylogeny is its suggestion that the robust australopithecines are not monophyletic.
Post a comment