Tolja!

According to Science/AAAS:

"Variability in mitochondrial (mt)DNA is often used to infer population size, history, and diversity on the assumption that mtDNA is essentially evolutionary neutral. Bazin et al. (p. 570; see the Perspective by Eyre-Walker) compared a wide range of animal species for polymorphisms in allozymes, nuclear DNA, and mtDNA. Within-species allozyme and nuclear DNA variability correlated with expected species abundance and ecological variables, whereas essentially no difference was observed between a broad range of taxa in terms of mtDNA variability. Instead, mtDNA seem to have undergone recurrent fixation of beneficial mutations and loss of variability at linked loci. Thus, mtDNA is far from a neutral marker; its diversity is essentially unpredictable and may not reflect population history and demography." (Emphasis mine)

That's what I've been saying for years.

So whaddayagotta say now, hotshot?