302
H.C. Harpending and E. Eller
would let us read this history of population relationships, migrations, and the genesis of human genetic diversity. In this paper we suggest that the new data from
neutral markers paint a rather dull picture of high levels of local gene flow everywhere and unremarkable correlations between genetic distances and geographic
distances between populations. In a sense the new wealth in data has been a disappointment.
When populations exchange neutral genes there is essentially blending of gene
frequencies, so red and white each become pink over time. Markers that do not
blend in this way may give us better information about population history. Languages, for example, do not blend like gene frequencies. Instead there is "majority
advantage" in which immigrants just learn the indigenous language (Renfrew 1987).
Prominent visible "racial" traits may also have enjoyed such majority advantage, so
that appearance can tell us more about ancient relationships than can gene frequencies.
2 Human Demographic History
2.1 Small Effective Size of Humans
Findings from many genetic systems suggest that the effective size of humanity is
on the order of 10000 breeding individuals. Since this estimate is wrong by a factor
of a million or so today, the implication is that the our ancestry is some specific
small isolated population of archaic humans (SaM) rather than the whole array of
Homo erectus relatives that occupied the temperate Old World for one to two million years (MRM, multi regional model).
Haigh and Maynard Smith (1972) suggested that there was a bottleneck in our
ancestry on the basis of the spectrum of substitutions in hemoglobin. Subsequently
the number 10000 bas become widely established as the summary effective breeding size of humans (Li and Sadler 1991). This could reflect either a popUlation that
was this small for a very long time else a severe transient bottleneck during which
the number of our ancestors was much less than this. Estimates from nuclear genes
(Harding et al. 1997; Hey 1997; Zietkiewicz et al. 1998), mtDNA (mitochondrial
DNA) (Rogers and Jorde 1995), the HLA system (Takahata and Satta 1998; in
press) and from human-specific alu insertions (Sherry et al. 1997) all converge on
a similar figure.
2.2 Expansion from Small Size
The tree of human mitochondrial DNA is star-like as if it is recording a major
population expansion in our history (Di Rienzo and Wilson 1991; Slatkin and Hudson
1991; Rogers and Harpending 1992; Harpending et al. 1993). The pattern in mtDNA
is clear, but it could be the result of selection as well as population expansion. The
expansion hypothesis has been in limbo for several years, since no such pattern is
H.C. Harpending and E. Eller
would let us read this history of population relationships, migrations, and the genesis of human genetic diversity. In this paper we suggest that the new data from
neutral markers paint a rather dull picture of high levels of local gene flow everywhere and unremarkable correlations between genetic distances and geographic
distances between populations. In a sense the new wealth in data has been a disappointment.
When populations exchange neutral genes there is essentially blending of gene
frequencies, so red and white each become pink over time. Markers that do not
blend in this way may give us better information about population history. Languages, for example, do not blend like gene frequencies. Instead there is "majority
advantage" in which immigrants just learn the indigenous language (Renfrew 1987).
Prominent visible "racial" traits may also have enjoyed such majority advantage, so
that appearance can tell us more about ancient relationships than can gene frequencies.
2 Human Demographic History
2.1 Small Effective Size of Humans
Findings from many genetic systems suggest that the effective size of humanity is
on the order of 10000 breeding individuals. Since this estimate is wrong by a factor
of a million or so today, the implication is that the our ancestry is some specific
small isolated population of archaic humans (SaM) rather than the whole array of
Homo erectus relatives that occupied the temperate Old World for one to two million years (MRM, multi regional model).
Haigh and Maynard Smith (1972) suggested that there was a bottleneck in our
ancestry on the basis of the spectrum of substitutions in hemoglobin. Subsequently
the number 10000 bas become widely established as the summary effective breeding size of humans (Li and Sadler 1991). This could reflect either a popUlation that
was this small for a very long time else a severe transient bottleneck during which
the number of our ancestors was much less than this. Estimates from nuclear genes
(Harding et al. 1997; Hey 1997; Zietkiewicz et al. 1998), mtDNA (mitochondrial
DNA) (Rogers and Jorde 1995), the HLA system (Takahata and Satta 1998; in
press) and from human-specific alu insertions (Sherry et al. 1997) all converge on
a similar figure.
2.2 Expansion from Small Size
The tree of human mitochondrial DNA is star-like as if it is recording a major
population expansion in our history (Di Rienzo and Wilson 1991; Slatkin and Hudson
1991; Rogers and Harpending 1992; Harpending et al. 1993). The pattern in mtDNA
is clear, but it could be the result of selection as well as population expansion. The
expansion hypothesis has been in limbo for several years, since no such pattern is