Four Years in Notothenioid Systematics: A Molecular Perspective
269
taken into account. As shown before, neither the placement of
Trematominae and Pleuragramminae together nor the existence of the
clade Pleuragramminae are supported. Similarly the taxonomic status of
the subfamily Nototheniinae is uncertain. At best, the vicari ant scenario
should be reevaluated in light of the new phylogenetic relationships
among nototheniids.
Phylogenetic trees based on sequence data can also give informations on
the tempo and mode of evolution of the species examined. For instance,
the low level of resolution at the interfamiliallevel seems not to be due to
the use of inappropriate molecular tools, since a reliable phylogenetic
signal is recovered for deep branching events (divergence of B. variegatus
and E. maclovinus) as well as for more recent ones. This might be,
instead, the phylogenetic mark of a radiation that occurred early on in the
evolution of notothenioids.
Moreover, genetic distances between species can be used to measure the
time since their divergence; this is possible when the rate of genetic
divergence is linear with time and homogeneous between taxa (molecular
clock hypothesis). If the clock is valid, molecular data provide a valuable
tool in timing evolutionary events. However, even accepting the clock
hypothesis, a problem still remains: calibrating the clock with absolute
timing. This calibration can be accomplished when geological records for
cladogenetic events are available; dates are usually provided by good
fossil records of the taxa under investigation and/or by specific, wellestablished geological events associated with separation of taxa [29].
With regard to our data, evolutionary distances were calculated on all
possible changes (since no mutational saturation is present either for
transitions or for transversions [16]). Divergence estimates were also
based on transversions (TV) only, since transversions seem to evolve
linearly with time. A few differences in the rate were detected by the
relative rate test. However, rates were considered sufficiently
homogeneous to be used in estimating time of divergence. Two different
estimates of divergence rates were applied to our data in order to calculate
time of divergence between notothenioid species. A rate of 0.3-0.7%
substitutions per million years (Myr) has been proposed by Rand [30] for
mtDNA of ectotherms, while a rate of 0.14% TV/Myr for ribosomal
mitochondrial genes (12S and 16S) was calculated in mountain newts [29]
and bovids [31]. When the separation of E. maclovinus and the rest of
nonbovichtid notothenioids is considered, our estimate of the divergence
time ranges between 23 and 22 Myr, which is roughly the time of the
formation of the Antarctic Convergence, suggesting that the Polar Front
possibly played a role as a vicariant factor, separating the lineage that led
to E. maclovinus from the ancestor to the rest of the notothenioid taxa.
269
taken into account. As shown before, neither the placement of
Trematominae and Pleuragramminae together nor the existence of the
clade Pleuragramminae are supported. Similarly the taxonomic status of
the subfamily Nototheniinae is uncertain. At best, the vicari ant scenario
should be reevaluated in light of the new phylogenetic relationships
among nototheniids.
Phylogenetic trees based on sequence data can also give informations on
the tempo and mode of evolution of the species examined. For instance,
the low level of resolution at the interfamiliallevel seems not to be due to
the use of inappropriate molecular tools, since a reliable phylogenetic
signal is recovered for deep branching events (divergence of B. variegatus
and E. maclovinus) as well as for more recent ones. This might be,
instead, the phylogenetic mark of a radiation that occurred early on in the
evolution of notothenioids.
Moreover, genetic distances between species can be used to measure the
time since their divergence; this is possible when the rate of genetic
divergence is linear with time and homogeneous between taxa (molecular
clock hypothesis). If the clock is valid, molecular data provide a valuable
tool in timing evolutionary events. However, even accepting the clock
hypothesis, a problem still remains: calibrating the clock with absolute
timing. This calibration can be accomplished when geological records for
cladogenetic events are available; dates are usually provided by good
fossil records of the taxa under investigation and/or by specific, wellestablished geological events associated with separation of taxa [29].
With regard to our data, evolutionary distances were calculated on all
possible changes (since no mutational saturation is present either for
transitions or for transversions [16]). Divergence estimates were also
based on transversions (TV) only, since transversions seem to evolve
linearly with time. A few differences in the rate were detected by the
relative rate test. However, rates were considered sufficiently
homogeneous to be used in estimating time of divergence. Two different
estimates of divergence rates were applied to our data in order to calculate
time of divergence between notothenioid species. A rate of 0.3-0.7%
substitutions per million years (Myr) has been proposed by Rand [30] for
mtDNA of ectotherms, while a rate of 0.14% TV/Myr for ribosomal
mitochondrial genes (12S and 16S) was calculated in mountain newts [29]
and bovids [31]. When the separation of E. maclovinus and the rest of
nonbovichtid notothenioids is considered, our estimate of the divergence
time ranges between 23 and 22 Myr, which is roughly the time of the
formation of the Antarctic Convergence, suggesting that the Polar Front
possibly played a role as a vicariant factor, separating the lineage that led
to E. maclovinus from the ancestor to the rest of the notothenioid taxa.