And Adam called his wife's name Eve;
because she was the mother of all living.
Genesis 3:20
"We have calculated the consensus sequence for human mitochondrial DNA using over 800 available sequences. Analysis of this consensus reveals an unexpected lack of diversity within human mtDNA worldwide. 
Not only is more than 83% of the mitochondrial genome invariant, but in over 99% of the variable positions, the majority allele was found in at least 90% of the individuals. In the remaining 0.22% of the 16,569 positions, which we conservatively refer to as “ambiguous,” every one could be reliably assigned to either a purine or pyrimidine ancestral state.
There was only one position where the most common allele had an allele frequency of less than 50%, but this has been shown to be a mutational hot spot.
On average, the individuals in our dataset differed from the Eve consensus by 21.6 nucleotides. Sequences derived from sub-Saharan Africa were considerably more divergent than average. Given the high mutation rate within mitochondria and the large geographic separation among the individuals within our dataset, we did not expect to find the original human mitochondrial sequence to be so well preserved within modern populations. With the exception of a very few ambiguous nucleotides, the consensus sequence clearly represents Eves mitochondrial DNA sequence.
The developed model stands in direct opposition to the Recent African Origins Hypothesis (RAO).
There are several critical assumptions on which RAO relies—all part of the “Standard Neutral Model” of Kimura (1968)—and most of the assumptions have been openly questioned in the evolutionary literature (Carter, 2007).
---These assumptions include the need for mutations to accumulate in all lineages at an equal rate (a molecular clock),
---that mtDNA undergoes no recombination,
---and that all new mutations are free from natural selection.
*If the molecular clock is violated, a reliable phylogenetic tree for worldwide mtDNA haplotypes cannot be built.
Tests for a molecular clock have failed in African L2 clades of mtDNA (Howell, Elson, Turnbull, & Herrnstadt, 2004; Torroni et al., 2001; ). This is a grave difficulty for RAO because haplogroup L2a is the most common haplogroup specific to Africa (Salas et al., 2004). Several recent studies have raised the specter of non-clock like evolution of mtDNA and have openly, but politely, questioned RAO theory (for example, Howell et al., 2004; Zsurka et al., 2007).
A second major assumption behind RAO is that mtDNA undergoes
no recombination. This has been debated often in the evolutionary literature but at least one of the newer studies seems to have found conclusive evidence for mitochondrial recombination (Zsurka et al., 2007).
---If true, many phylogenetic studies will need reassessment for it is the pure maternal inheritance of mtDNA (that is, no input from the paternal side and no recombination of mixed maternal lineages) that allows for a clear-cut phylogeny to be constructed.
This might be an explanation for at least some of the homoplasy (identical mutations occurring in parallel lineages) found in the mitochondrial family tree (Zsurka et al., 2007), especially for the African sequences.
***Most mtDNA phylogenies use chimpanzee mitochondrial sequences as an out group.
---Not only is this a product of circular reasoning,
---but it also skews the resulting trees towards the consensus of human and chimpanzee mtDNA sequences.
The human and chimp mtDNA sequences are substantially different, ---we do not know the ancestral chimp sequence,
---and we do not know the degree of degeneration that has occurred in chimp lineages.
*Each of these factors will affect placement of the root.
Without chimp, one is free to explore alternative root placement options. This is essentially what we have done in our consensus calculation.
The genetic facts, apart from the formulation of historical scenarios, are clear:
(a) There was a single dispersal of mankind with three main mitochondrial lineages interspersed within the clans.
(b) This dispersal either passed through, or originated within, the Middle East.
(c) These things happened in the recent past.
(d) The dispersion was essentially tribal in nature, with small groups pushing into previously-uninhabited territory.
In addition, genetic evidence indicates that male lineages are much
more geographically specific than female lineages, with female “migration rates” up to eight-fold higher than males (Seielstad, Minch, & Cavalli-Sforza, 1998; Stoneking, 1998)—a direct confirmation of the Babel account where the initial, well-mixed population split up and migrated according to paternal lineage.
While the Biblical model fits very well with the data collected by many evolutionary studies, the main difficulty comes from the mtDNA clades from sub-Saharan Africa.
These become much less problematic when given proper consideration. Simple visual analysis (see Nordborg, 2001) of any published mtDNA phylogenetic tree (for example, Torroni et al., 2006) indicates that
---the African clades have had different historical population 
histories, with the African clades forming a cascading pattern with deep branches and the non-African lineages forming a star-like pattern with short branches.
---The evolutionary explanation is that these groups have been in Africa for tens of thousands of years longer than the lineages that left Africa. However, there are a number of alternative explanations, all of which support the Biblical model.
---For instance, if the groups that eventually made up the African populations were restricted to smaller tribe sizes until recently, drift would have occurred more quickly and they would have diverged from the rest of the world, and from each other, at a higher rate. ---Likewise, if the African groups have a different DNA repair system than the others (either defective or differential), this would also explain their more rapid divergence.
While these are only a theoretical considerations, they serve to illustrate the large number of assumptions implicit in RAO theory. ---Generation time is another consideration.
Evolutionary models assume equal generation times among all subpopulations, but cultural and genetic factors could easily influence generation time.
--For example, if the average age of marriage in one population was 20 years old and the average age of marriage in another was 18 years old, a 10% difference in generation time results. Analysis of the life spans of the patriarchs shows how average age of marriage changed dramatically downward in the first generations after the Flood, but there is no indication that this change occurred at the same rate in all populations. Average lifespan differences among populations might also skew generation time differences.
Using the Revised Cambridge Reference Sequence (rCRS) (Andrews, Kubacka, Chinnery, Lightowlers, Turnbull, & Howell, 1999) as a template for nucleotide numbering and BioPerl (Stajich et al., 2002) for all calculations, a hash table was constructed that included all variant positions with sequence names and nucleotide positions as keys.
Human mitochondrial genetic history was modeled using Mendel’s Accountant (Sanford, Baumgardner, Brewer, Gibson, & ReMine, 2007a, b), with parameters designed to mimic the mitochondrial genome (for example, genome size = 16,500 nucleotides, one linkage block). A population of 1,000 individuals (a Biblically-reasonable size) was allowed to freely interbreed under realistic constraints for 150 and 10,000 generations. Our population size of 1,000 is considerably smaller than most evolutionary estimates of historic human population size, but recent data suggest an effective human population size of just a few thousand individuals (Tenesa et al., 2007). With an average generation time of 30 years (Tremblay & Vézina, 2000), there have only been c.a. 150 generations in the 4,500 years since the Flood. The 10,000-generation model run is more consistent with evolutionary models.
The Eve mitochondrial consensus sequence is unambiguous. Invariant positions made up 83.9% of the mitochondrial genome, and nearly half of the variable positions (43.8%) were due to the 
presence of private mutations (that is, at each of these positions, a single sequence in the database carried an alternate allele). Each variable site had more than one allele, but 99% of these sites had a primary allele frequency of 0.90 or greater.
That is, for nearly all variant sites, there was a strongly dominant allele. There were only 36 positions (0.22% of the 16,569 nucleotides in human mtDNA) that had a primary allele frequency of less than 0.90.
Many of these were not simple polymorphic nucleotides, but were “poly-x” sites.
---For the non-poly-x sites, the ancestral site was consistently and clearly either a purine (nearly all alleles were either “A” or “G”) or a pyrimidine (nearly all alleles were either “C” or “T”).
--The most variable position, 309, is a poly-C tract which is clearly a mutational hotspot and shows a high rate of heteroplasmy within individuals (Carter, 2007).
Thus, even the most variable position does not challenge the model of a single invariant human mitochondria in the recent past. This is strong evidence of a young mitochondrial genome.Pair-wise differences from Eve1.0 and among all sequences .... On average, individuals differed from the consensus at only 22.6 positions, with sequences from sub-Saharan Africa varying at up to 89 positions.
---If there were many positions with no dominant allele, we could not reasonably infer the original Eve sequence.
---Essentially, all mutant alleles are rare.
---The mitochondrial genome is subject to high mutation rates (as evidenced from the high degree of private mutations), but the lack of significant worldwide variation indicates a young mitochondrial genome."
Carter/Sanford/Criswell ICC/ICR