sites per 10^6 years (Table 6.1). The rRNA rates are also relatively stable in a family of
planktonic Foraminifera, the Globigerinidae (de Vargas et al. 1997). The mean
evolutionary rate in the Globigerinidae was shown to average 4.3 substitutions/1000 sites
per 10^6 years, with values ranging from 4.0–4.6 substitutions/1000 sites per 10^6 years in
pairwise comparisons, and from 3.2–6.1 substitutions/1000 sites per 10^6 years, when
they were calculated on individual lineages (de Vargas and Pawlowski 1998). The
existence of local molecular clocks was also observed by Smith et al. (1992) for LSU rRNA
in individual echinoid lineages, and by O’h Uigin and Li (1992) for numerous protein-
coding genes in rodents.
On the other hand, much higher SSU rRNA rate variations are observed in another family
of planktonic Foraminifera, the Globorotaliidae, where pairwise divergence rates were
shown to vary from 0.3–5.3 substitutions/1000 sites per 10^6 years (de Vargas and
Pawlowski 1998). Re-examination of the data in one of the rapidly evolving species,
Globorotalia truncatulinoides, by including additional species of Globorotaliidae and with a
revision of their divergence times, suggests even higher values of 19.7– 24.6 substitutions/
1000 sites per 10^6 years, about 20 times faster than the rates of the most slowly evolving
representatives of this family (de Vargas et al. 2001; see also Table 6.1). Interestingly, the
exceptional acceleration of the substitution rate in G. truncatulinoides seems to result from
a progressive increase of the rates in the phylogenetic lineage leading to this species, from
2.25 substitutions/1000 sites per 10^6 years in Globorotalia scitula to 4.1–5.2 substitutions/
1000 sites per 10^6 years in Globorotalia hirsuta, the two species the most closely related to
G. truncatulinoides (de Vargas et al. 2001).
Acceleration of the foraminiferan stem-lineageThe most striking example of rate variation in the Foraminifera is the acceleration of their
stem-lineage, as evidenced by analyses of 1175 unambiguously aligned positions of
complete SSU rRNA sequences. When sequences of Foraminifera are compared with
those of other eukaryotes, their stem-lineage appears as a very long branch leading to a
radiation of all foraminiferan lineages (Figure 6.1).
Fossil calibrationIt is impossible to calibrate precisely the foraminiferan stem-lineage acceleration because
no fossil data exist either for the divergence of the Foraminifera from their eukaryotic
ancestor or for the beginning of the radiation of extant foraminiferan lineages. We can
only propose a very conservative estimate of the time of the foraminiferan stem-lineage
evolution. The divergence between Foraminifera and their eukaryotic ancestor probably
occurred at less than 1000 Ma. This estimation corresponds to the radiation of ‘crown’
Eukaryotes, based on the palaeontological record (Knoll 1992). As the Foraminifera
apparently evolved from some filose sarcodinids within this radiation (Keeling 2001;
unpublished data), it seems reasonable to consider that their divergence occurred at
around this time.
112 JAN PAWLOWSKI AND CÉDRIC BERNEY