Footnotes and extra appendixes transferred away from the PhD manuscript for the lack of space

Case studies of the problems associated with the evolutionary Molecular clock/Mutation Clock

Originally the Kimura’s neutrality theory (1968) predicted, as mathematically formulated from the hypothesis of Zuckerland & Pauling (1965), that molecular evolution behaves like a stochastic clock. Now that we have even too much of the data at hand it seems, however, that the number of amino acid replacement in a given protein do not appear to change linearly with time, after all.

This phenomenon is called erratic overdispersion, and when Rodriguez-Trelles et al (2001) studied glycerol-3-phosphate dehydrogenase, suproxide dismutase, and xanthine dehydrogenase from a total of 78 species of plants, fungi, and mamals, they ended up tabulating divergence times, respectively, as follows: 41, 46, and 53 Mryrs between Drosophila groups; 60, 60, and 60 Mryrs between Drosophila subgenera; 142, 74, and 65 Mryrs between drosophilid genera; 455, 110, and 90 Mryrs between dipteran families; 400, 105, and 38 Mryrs between mammalian orders; 3890, 374, and 364 Mryrs between animal phyla; 6699, 276, and 130 Mryrs between fungi; and 7045, 451, and 398 Mryrs between kingdoms. (When aplying the rate of Drosophila subgenera to other organisms.)

Regarding the clocks at the multicellular divergence dates, Michael Lee (1999) summarizes:

"Even if one makes the bold assumption that molecular clock models have little error, there seems little objective reason for accepting as sacrosanct a few fossil dates used in calibrations and rejecting as unreliable the much more numerous fossil dates that contradict the resultant molecular estimates… Unfortunately, molecular clocks studies have yet to provide a set of rigorous criteria for justifying which fossil dates are to be used in calibrations and which are to be treated with scepticism." Lee concludes his evaluation by stating: "Many molecular studies argue persuasively for the gross inadequacy of he fossil record, but it has been insufficintly acknowledged that their inferences, too, are ultimately based on this same record and should therefore be treated with similar caution." The assumption, really, is bold after tabulation of metazoan (common animal) divergence for 22 important nuclear proteins with a variation between 1,600-274 (weighted average 851± 81) Mryrs (based directly on mammal-bird divergence of 310 Mryrs on fossil record), or 1,375-594 (weighted average 830±55) Mryrs (based on metazoan-fungus divergence of 1,100 Mryrs, but ultimately only on the former one). Lee calibrates these already "internally" or "externally" calibrated published dates by four corrections: 0.93 (based on mammal-bird divergence of 288 Mryrs), 0.85 or 0.65 (based on primate-rodent divergence of 85 or 65 Mryrs), and 0.548 (based on metazoan-fungus divergence of 603 Mryrs) upon revised and "direct" evidence from the fossil record. Lee ends up at the figure of 791-528 Mryr for the metazoan divergence, because he rejects the metazoan-fungal fossil dates leading to 455 Mryr, which comes late to the dates with the scenario of Neoproterozoic-Cambrian radiation, popularly called as the cambrian explosion. Lee also acknowledges other data series starting at divergence of over 1,000 Mryrs, that he is able to recalibrate nearly by half.
Lee MSY (1999) Molecular Clock Calibrations and Metazoan Divergence Dates. J  Mol Evol 49, 385-91

Kimura M (1968) Evolutionary rate at the molecular level. Nature 217, 624-6

Kimura M (1983) The neutral theory of molecular evolution. Cambridge Univ. Press, UK.

Rodriguez-Trelles F, Tarrio R & Ayala FJ (2001) Erratic overdispersion of three molecular clocks: GPDH, SOD, and XDH. Proc Natl Acad Sci 98, 11405-10

Zuckerland E & Pauling L (1965) in Evolving Genes and Proteins, eds. Bryson V & Vogel HJ. Academic, New York pp. 97-166