factor of two when comparisons are made between the more diverged brachycerans and
nematocerans (23.4× 10 −^10 /site per year obtained assuming that brachycerans and
nematocerans split at 250 Ma). Assuming that Anopheles diverged from Aedes and Culex at
160 Ma, the rate between these two lineages is only 11.0× 10 −^10 /site peryear, much slower
than between Drosophilidae. Notice also the mammalian rate of 6.8× 10 −^10 , which is
several times slower than the rate of the contemporaneously evolving Drosophila
subgenera (35.5× 10 −^10 ).
XDH
When comparisons are made between different kingdoms, this enzyme evolves at a rate
(11.5× 10 −^10 /site per year) no greater than any other enzyme. Yet, XDH evolves very fast
in the Drosophilidae. As noted for other enzymes, the XDH rate variations displayed in
Table 1.3 mask a much greater variation between lineages at different times because the
rates given apply to largely overlapping lineages. Thus, the average number of
replacements between birds and mammals is 10.5× 10 –10/site per year. If we accept that
the lineage of mammals has evolved at an average rate of 17.1 (the mammal rate in
Table 1.3) since they separated from birds, to attain an average of 10.5 between mammals
and birds, the bird lineage must have evolved at a rate of only 3.9, ~8 times slower than
the Drosophila rate. Additional rate discrepancies have been pointed out by Rodríguez-
Trelles et al. (2001a).
The estimated rates of amino acid replacement shown in Table 1.2 (i.e. between
dipteran families and successively lower categories) differ from their correlates in
Table 1.3 because the rates assume different degrees of among-site rate variation: the
rates in Table 1.2 use α values obtained from dipterans, which are substantially smaller
than the values used in Table 1.3, derived from the global dataset (see above). Which of
the two sets of α values is more nearly correct is not easily decided (Nei et al. 2001). In
the absence of sampling bias, for a substitution process stationary with respect to among-
site rate variation (i.e. sites retain the same relative rates of change through-out the tree),
the estimates obtained from closely related species should be closer to the true value of α
(Zang and Gu 1998). If the process is non-stationary, however, using closely related
species to estimate α can be misleading, because different lineages can have disparate α
values. In the present case, it is apparent that the GPDH α value obtained from Diptera is
very low because this gene is extremely conserved in Drosophila; therefore, the rate of
GPDH amino acid replacement between Ceratitis and the drosophilids obtained with this
value of α is likely to be unduly large (i.e. the variation introduced by Ceratitis is
outweighed).
We have unveiled disparate differences in evolutionary rates among and within
lineages, which are inconsistent across genes. Doubtless, these differences reflect
biological processes, as illustrated by GPDH. In Drosophila, GPDH is subjected to
constraints which considerably restrict the number of sites that can accept amino acid
replacements and the particular amino acid replacement that can occur at each site (Ayala
1997). The question is, nevertheless, whether observed rate differences are detected by
standard statistical tests. We have conducted likelihood ratio tests of the molecular clock
hypothesis on the global datasets. Strictly speaking this comparison is valid only if the
20 FRANCISCO RODRÍGUEZ-TRELLES ET AL.