MOLECULAR CLOCKS 583

Kishino, H., Thorne, J. L., and Bruno, W. J., 2001. Performance of Thorne, J. L., Kishino, H., and Painter, I. S., 1998. Estimating the
a divergence time estimation method under a probabilistic model rate of evolution of the rate of molecular evolution. Molecular
of rate evolution. Molecular Biology and Evolution, 18, 352–361. Biology and Evolution, 15, 1647–1657.
Kodandaramaiah, U., 2011. Tectonic calibrations in molecular dat- Warnock, R. C. M., Yang, Z., Donoghue, P. C. J., 2012. Exploring
ing. Current Zoology, 57, 116–124. uncertainty in the calibration of the molecular clock. Biology
Marshall, C. R., 1997. Confidence intervals on stratigraphic ranges Letters, 8, 156–159.
with nonrandom distributions of fossil horizons. Paleobiology, Wilkinson, R. D., Steiper, M. E., Soligo, C., Martin, R. D., Yang, Z.,
23, 165–173. and Tavaré, S., 2011. Dating primate divergences through an
Müller, J., and Reisz, R. R., 2005. Four well-constrained calibration integrated analysis of palaeontological and molecular data. Sys-
points from the vertebrate fossil record for molecular clock esti- tematic Biology, 60, 16–31.
mates. Bioessays, 27, 1069–1075. Yang, Z., and Rannala, B., 2006. Bayesian estimation of species
Parham, J. F., Donoghue, P. C. J., Bell, C. J., et al., 2012. Best practices divergence times under a molecular clock using multiple fossil
for justifying fossil calibrations. Systematic Biology, 61, 346–359. calibrations with soft bounds. Molecular Biology and Evolution,
Peters, S. E., 2005. Geologic constraints on the macroevolutionary 23, 212–226.
history of marine animals. Proceedings of the National Academy Zuckerkandl, E., and Pauling, L., 1962. Molecular disease, evolution
of Sciences, 102, 12326–12331. and genetic heterogeneity. In Kasha, M., and Pullman, B. (eds.),
Peters, S. E., and Foote, M., 2001. Biodiversity in the Phanerozoic: Horizons in Biochemistry. New York: Academic, pp. 189–225.
a reinterpretation. Paleobiology, 27, 583–601.
Pyron, R. A., 2011. Divergence time estimation using fossils as ter-
minal taxa and the origins of Lissamphibia. Systematic Biology,
60, 466–481.
Rannala, B., and Yang, Z., 2007. Inferring speciation times under an
episodic molecular clock. Systematic Biology, 56, 453–466. MOLECULAR CLOCKS
Raup, D. M., 1972. Taxonomic diversity during the
Phanerozoic. Science, 177, 1065–1071. Simon Y. W. Ho
Reisz, R. R., and Müller, J., 2004. Molecular timescales and the fos- School of Biological Sciences, University of Sydney,
sil record: a paleontological perspective. Trends in Genetics,
20, 237–241. Sydney, NSW, Australia
Ronquist, F., Teslenko, M., van der Mark, P., Ayres, D. L., Darling,
A., Höhna, S., Larget, B., Liu, L., Suchard, M. A., and Definition
Huelsenbeck, J. P., 2012a. MrBayes 3.2: efficient Bayesian phy-
logenetic inference and model choice across a large model space. Molecular clock. A hypothesis that predicts a constant
Systematic Biology, 61, 539–542. rate of molecular evolution among species. It is also
Ronquist, F., Klopfstein, S., Vilhelmsen, L., Schulmeister, S., Mur- a method of genetic analysis that can be used to estimate
ray, D. L., and Rasnitsyn, A. P., 2012b. A total-evidence evolutionary rates and timescales using data from DNA
approach to dating with fossils, applied to the early radiation of or proteins.
the hymenoptera. Systematic Biology, 61, 973–999.
Rota-Stabelli, O., Daley, A. C., and Pisani, D., 2013. Molecular
timetrees reveal a cambrian colonization of land and a new sce- Introduction
nario for ecdysozoan evolution. Current Biology, 23, 392–398. The tempo and mode of evolution are central themes of
Sanderson, M. J., 1997. A nonparametric approach to estimating
divergence times in the absence of rate constancy. Molecular biological research. This places importance on the estima-
Biology and Evolution, 14, 1218–1231. tion of evolutionary timescales, which provide the back-
Sanderson, M. J., 2002. Estimating absolute rates of molecular evo- drop for our interpretations of evolutionary patterns and
lution and divergence times: a penalized likelihood approach. processes. Traditionally, such inferences were made from
Molecular Biology and Evolution, 19, 101–109. the fossil record, coupled with radiometric dating. Fossils
Sansom, R., Gabbott, S., and Purnell, M., 2010. Non-random decay can provide an estimate of when different lineages first
of chordate characters causes bias in fossil interpretation. Nature,
463, 797–800. appeared and when species diverged from each other. In
Smith, A. B., 2007. The shape of the Phanerozoic marine many cases, however, such data are unavailable, forcing
palaeodiversity curve: how much can be predicted from the sed- us to look elsewhere for a source of temporal information.
imentary rock record of Western Europe? Palaeontology, The “molecular clock,” proposed in the 1960s
50, 765–774. (Zuckerkandl and Pauling, 1962, 1965), allows evolution-
Smith, A. B., and McGowan, A. J., 2007. The shape of the Phaner- ary timescales to be estimated using genetic data. Molecu-
ozoic marine palaeodiversity curve: how much can be predicted
from the sedimentary rock record of Western Europe? lar clocks have provided insights into significant
Palaeontology, 50, 765–774. evolutionary events such as human evolution and the radi-
Strauss, D., and Sadler, P. M., 1989. Classical confidence intervals ations of birds, mammals, and flowering plants. Molecu-
and Bayesian probability estimates for ends of local taxon lar-clock methods continue to undergo improvement,
ranges. Mathematical Geology, 21, 411–427. which is crucial if we are to take advantage of the growing
Tavaré, S., Marshall, C. R., Will, O., Soligo, C., and Martin, R. D., wealth of genomic data (Kumar, 2005).
2002. Using the fossil record to estimate the age of the last com-
mon ancestor of extant primates. Nature, 416, 726–729.
The molecular-clock hypothesis posits that rates of
Thorne, J. L., and Kishino, H., 2005. Estimation of divergence times molecular evolution, as reflected in changes in DNA or
from molecular sequence data. In Nielsen, R. (ed.), Statistical protein sequences through time, are constant among line-
Methods in Molecular Evolution. New York: Springer, ages (but not across different regions of the genome).
pp. 233–256. Nucleotide mutations cause DNA sequences to change
584 MOLECULAR CLOCKS

over time, whereas mutations of amino acids lead to evo-

lution in proteins. These mutations occur with a constant
probability rather than at a constant frequency – that is,
the molecular clock is stochastic rather than metronomic
(Zuckerkandl and Pauling, 1965). The occurrence of
mutations in a DNA sequence is often modelled using
a Poisson process.
A corollary of the molecular-clock hypothesis is that
the genetic difference between any two species is propor-
tional to the time since they last shared a common ances-
tor. This is a useful relationship because it allows
evolutionary timescales to be estimated from genetic data,
provided that the rate of molecular evolution is known.
The molecular clock represents the only method for study-
ing the evolutionary timescales of organisms that have
failed to leave any trace in the fossil record, including
many groups of invertebrates, bacteria, and viruses. It also
allows us to estimate the timing of events that are too
recent to be resolved by fossil evidence, such as diver-
gences among conspecific populations (Arbogast et al.,
2002). Molecular Clocks, Figure 1 Plot of amino acid identity of
fibrinopeptides from various pairs of mammals, plotted against
time since divergence. The divergence times are based on
History of the molecular clock estimates from the fossil record (Data are from Doolittle and
Blomback (1964)).
The molecular-clock hypothesis was put forward by Emile
Zuckerkandl and Linus Pauling (1962), who assumed
a constant evolutionary rate in their analysis of globin pro-
teins from vertebrates. They observed about 18 differ- evidence to suggest that adaptive change occurred at
ences in amino acids between horse and human and a uniform rate and it seemed implausible that the complex
estimated the mutation rate by assuming that the diver- process of molecular evolution could be described by such
gence between these two species occurred 100–160 Ma a simple statistical model.
ago. Upon extrapolating this rate, Zuckerkandl and Pau- Meanwhile, there was growing evidence of high evolu-
ling (1962) estimated that humans diverged from gorillas tionary rates in various proteins, suggesting that a large
about 11 Ma ago. They also estimated that different copies proportion of the changes in amino acids must have
of globin genes first diverged from each other in the late a negligible impact on evolutionary fitness. As
Precambrian. When reporting these estimates, a response, Motoo Kimura (1968) proposed the neutral
Zuckerkandl and Pauling (1962) warned about the possi- theory of molecular evolution, which states that many
ble confounding effects of a number of factors, including mutations have such a small effect on the fitness of an
natural selection, variations in population size, and muta- organism that they can be considered as “neutral.” This
tional saturation. can be explained by the fact that many amino acids in
In the following years, further studies produced evi- a protein can be exchanged for other amino acids with
dence of clocklike evolution in other proteins. Doolittle similar biochemical properties, with negligible impact on
and Blomback (1964) found a simple relationship the overall function or structure of the protein. At the level
between sequence identity and time since divergence in of DNA, many mutations in protein-coding genes are
mammalian fibrinopeptides (Figure 1). A year later, “synonymous” because they do not result in any changes
Zuckerkandl and Pauling (1965) coined the term “molec- in the amino acid sequence of the protein. The fate of these
ular evolutionary clock.” It promised to be a useful tool “neutral” mutations is determined stochastically by
in biological research, as demonstrated shortly afterward a process known as genetic drift. The neutral theory was
by Sarich and Wilson (1967a, b) in their pioneering stud- proposed independently by Jack King and Thomas Jukes
ies of the evolutionary timescale of hominids and other (1969), who cited a range of evidence to support their
primates. hypothesis.
At the time, the idea of a constant substitution rate was One of the predictions of the neutral theory is that rates
controversial (Morgan, 1998). The molecular clock was of molecular evolution are constant among lineages. How-
received unenthusiastically by George Gaylord Simpson ever, this prediction refers specifically to the rate of
(1964), among others. Criticisms of the molecular clock genetic change per generation. As a consequence, we
were partly motivated by the apparent lack of uniformity expect to see a generation-time effect, whereby species
in the pace of morphological evolution, which was pre- with shorter generations tend to evolve more quickly per
sumably linked to molecular evolution. There was no unit of time. For example, a higher rate would be observed
 MOLECULAR CLOCKS 585

in rodents than in whales. This is based on the assumption Rates of molecular evolution
that most mutations occur during the replication of Absolute rates of molecular evolution vary among regions
germline DNA. Such a generation-time effect was of the genome because of differing constraints. Function-
observed in the analyses of non-coding DNA (Laird ally important genes often evolve at an extremely slow
et al., 1969; Kohne, 1970). This was in contrast with the pace, because many new mutations are detrimental to the
rate of protein evolution, which appeared to be indepen- organism and are rapidly removed by natural selection.
dent of generation time. As a consequence, such genes are often conserved even
In response to the shortcomings of the neutral theory, among distantly related organisms, making them useful
Tomoko Ohta (1972, 1973) proposed the nearly neutral genetic markers for studying deep evolutionary relation-
theory of molecular evolution. In this framework, there ships. For example, histones, which are proteins that play
is a large class of “nearly neutral” mutations that have an important role in binding and packaging DNA, have
small effects on an organism’s fitness. In contrast with a very low substitution rate. In contrast, non-coding
the results of the neutral theory, the nearly neutral theory DNA has fewer functional constraints and can evolve at
states that the population sizes of species have a higher rate than protein-coding DNA.
a significant influence on the molecular evolutionary pro- The mitochondrial genome evolves rapidly in animals
cess. Many mutations are slightly harmful and are gradu- and experiences about 10
8 substitutions per nucleotide
ally removed from the population. In large populations, per year, which is an order of magnitude higher than the
natural selection is effective at removing mutations that average rate in the nuclear genome. This is in contrast with
are detrimental to the organism’s fitness. In small the pattern of molecular evolution in plants, where nuclear
populations, natural selection tends to be ineffective and genomes evolve more quickly than either chloroplast or
most genetic change is driven by genetic drift. Generally, mitochondrial genomes. Rates of change in the genomes
drift acts more quickly in small populations than does of viruses are higher by several orders of magnitude. In
selection in large populations. However, a greater number rapidly evolving RNA viruses, such as influenza virus
of mutations appear each generation in large populations, and human immunodeficiency virus (HIV), mutation rates
simply because there are more individuals that can experi- can exceed 10
3 mutations per nucleotide per year, and
ence mutations. These two effects offset each other, lead- measurable genetic change can occur over a matter of
ing to a relatively constant evolutionary rate among weeks.
species per unit of time. There is substantial variation in rates of molecular
Through the ensuing decades, the molecular clock was evolution across the tree of life. This variation is possibly
used to study the evolutionary timescales of a range of driven by biological factors such as generation time, pop-
organisms. Some of the most prominent studies involved ulation size, longevity, and body temperature, as well as
analyses of the metazoan evolutionary timescale, with abiotic factors such as ultraviolet radiation. The relative
a focus on the divergences among animal phyla. In consid- importance of each of these factors is still poorly under-
erable conflict with the “Cambrian explosion” scenario stood (Bromham, 2009). Studies have found evidence
supported by the fossil record, estimates from the molecu- that molecular evolutionary rates are associated with lon-
lar clock suggested a protracted Precambrian timescale for gevity in mammals (Welch et al., 2008), with generation
basal metazoan divergences (e.g., Dickerson, 1971; time in invertebrates (Thomas et al., 2010), and with
Runnegar, 1982; Doolittle et al., 1996). Similarly, molec- height in plants (Lanfear et al., 2013). Research into the
ular evidence pointed to much deeper diversifications of factors governing rate variation among organisms is
avian and mammalian orders than those suggested by ongoing.
paleontological evidence (e.g., Cooper and Penny, 1997;
Springer, 1997). These discrepancies remain some of the
key sources of debate about the accuracy of the molecular Molecular clocks and phylogenetic analysis
clock (Bromham and Penny, 2003). Molecular clocks are typically used in phylogenetic ana-
Meanwhile, there was a growing body of evidence that lyses, which aim to reconstruct evolutionary trees that
pointed to rate variation among lineages. This came show the relationships among species of interest
partly from various statistical tests that had been devel- (Figure 2). Internal nodes in the tree represent evolution-
oped to test for clocklike evolution in genetic data ary divergence events. The timing of these events can be
(Wu and Li, 1985; Tajima, 1993). Departures from the estimated using molecular clocks. A number of statistical
molecular clock provided a challenge for studies of evo- methods are available for testing the molecular-clock
lutionary timescales. It was not until the late 1990s, hypothesis for a given set of DNA or protein sequences.
however, when molecular-clock methods were devel- When the molecular clock is rejected for a data set, one
oped that were able to account for variation in evolution- can use a statistical model to account for rate variation
ary rates (Sanderson, 1997; Thorne et al., 1998). Known when estimating evolutionary timescales (Welch and
as “relaxed” molecular clocks because they relax the Bromham, 2005).
assumption of rate constancy among lineages, these have When estimating evolutionary timescales in
become the standard techniques in molecular evolution- a phylogenetic analysis, the molecular clock needs to be
ary analysis. “calibrated.” This can be done by assigning an absolute
586 MOLECULAR CLOCKS

phylogeny, the ages of the remaining nodes can be esti-

mated using a molecular-clock analysis of the genetic
data. This process is referred to as “molecular dating” or
“divergence-time estimation.” There is a large range of
methods and models that have been developed for this
purpose, implemented in various statistical frameworks.
These methods are available in a range of computer pro-
grams, including most standard phylogenetic software.

Universal molecular clocks

There is abundant evidence of evolutionary rate variation
among species, dispelling any hope of a universal molec-
ular clock across the tree of life. However, some
researchers entertain the idea of homogeneous rates of
mitochondrial evolution within certain groups of organ-
isms, such as birds, mammals, and arthropods. This is
a convenient assumption because it allows evolutionary
Molecular Clocks, Figure 2 Phylogenetic tree showing the rates to be applied in molecular-clock analyses when fossil
relationships of species from three genera, A, B, and C. Genera or geological calibrations are otherwise unavailable.
A and B are found on the mainland, whereas species of Genus C Many studies of birds assume that the mitochondrial
are found on an offshore island that was formed by volcanic genome evolves at about 1 % per million years, a value
activity. The tree is drawn to an unspecified timescale, so that
the branch lengths are proportional to time. The tree illustrates that was initially based on a study of five species of geese
the use of two molecular-clock calibrations. First, a fossil taxon (Shields and Wilson, 1987). Various analyses of avian
has been assigned to the lineage leading to Genus A. This places mitochondrial DNA have supported this estimate, includ-
a minimum age constraint on the divergence of A from B and C. ing an extensive study involving 74 calibrations and
Second, a geological calibration is placed on the divergence genetic data from 12 orders of birds (Weir and Schluter,
between B and C. The split between these genera is assumed to 2008). The avian mitochondrial clock has been used to
coincide with the formation of the offshore island on which
Genus C is found. The age of the island is estimated using investigate key questions in the evolution of birds, includ-
radiometric dating. The uncertainty in this date estimate is ing the impact of Pleistocene glaciation on the diversifica-
incorporated into the calibration, as indicated by the black bar. tion of passerines. Some have questioned the reliability of
Gray error bars represent the uncertainty in the molecular-clock this clock, citing evidence of significant rate variation
estimates of lineage divergence times. among lineages and across timescales (García-Moreno,
2004; Ho, 2007). Despite this criticism, the avian mito-
chondrial clock is still widely used.
age to one or more nodes in the phylogeny, which can then In a similar fashion, a universal mitochondrial clock of
act as a reference point for estimating the ages of the 1 % per million years is often assumed for mammals. This
remaining nodes. Calibrations are often based on the fossil rate was initially based on a mitochondrial study of pri-
record, which can provide date estimates for the diver- mates and rodents (Brown et al., 1979), but it soon found
gence events in the phylogeny (Figure 2). If a fossil taxon support from estimates derived from other organisms
can be reliably assigned to one of the lineages in the tree, (Wilson et al., 1985). Comprehensive analyses of mam-
then the divergence of that lineage from its sister lineage malian DNA have demonstrated that there is substantial
must be older than the age of the fossil. Calibrations can variation in rates among species, partly driven by differ-
also be based on geological events, including the separa- ences in longevity (Nabholz et al., 2008; Welch et al.,
tion of continents or the emergence of islands, if they are 2008).
linked to evolutionary divergences. An example of this Studies of evolutionary timescales in arthropods have
might involve two sister genera, one of which is endemic often employed a rate of 1.15 % per million years, which
to the mainland and the other endemic to an offshore is based on an analysis of insects and crustaceans (Brower,
island that formed as a result of volcanic activity 1994). This arthropod mitochondrial clock is widely used,
(Figure 2). The timing of the divergence between the partly because of the relative paucity of reliable fossil cal-
two genera can be estimated by the age of the island ibrations for many invertebrate taxa. However, this clock
which, in turn, can be estimated using radiometric dating. was based on a very small number of data points, leading
In some cases, previous genetic estimates of dates are to its validity being questioned (Papadopoulou et al.,
employed as calibrations. These are known as “second- 2010; Ho and Lo, 2013). Moreover, there is strong evi-
ary” calibrations and are generally used when other cali- dence of mitochondrial and nuclear rate variation among
brating information is unavailable. invertebrate species (Thomas et al., 2010).
Once the phylogenetic tree is calibrated by fixing or The employment of “standard” mitochondrial clocks,
constraining the ages of one or more nodes in the such as those described above, has often been criticized.
 MOLECULAR CLOCKS 587

This is because the mitochondrial genome is subject to dif- Conclusions

fering degrees of natural selection among species, which The molecular clock has undergone considerable evolu-
can lead to heterogeneity in evolutionary rates. Similar tion during its long history. It is useful as a hypothesis in
reasoning applies to evolutionary rates in the nuclear molecular evolution and as a tool for estimating evolution-
genome. Given that rates can show substantial variation ary rates and timescales. New molecular-clock methods
even among closely related species, the use of standard are being developed in order to take advantage of the large
clocks has the potential to yield date estimates that are amounts of genomic data that are being generated. With
highly misleading. further refinement and development, the molecular clock
will continue to play an important role in understanding
the evolution of life on Earth.
Controversies
The molecular clock has had a long history of contro-
versy, beginning with the criticisms of the assumption Bibliography
of rate constancy and continuing with the debate over Arbogast, B. S., Edwards, S. V., Wakeley, J., Beerli, P., and
the relative merits of the neutral theory compared with Slowinski, J. B., 2002. Estimating divergence times from molec-
natural selection. Current debates include the use of cali- ular data on phylogenetic and population genetic timescales.
Annual Review of Ecology, Evolution, and Systematics,
brations in genetic dating analyses and the constancy of 33, 707–740.
evolutionary rates across timescales (Pulquério and Bromham, L., 2009. Why do species vary in their rate of molecular
Nichols, 2007). evolution? Biology Letters, 5, 401–404.
The choice of calibrations has a substantial influence on Bromham, L., and Penny, D., 2003. The modern molecular clock.
the outcome of a molecular-clock analysis. Inadequate Nature Reviews. Genetics, 4, 216–224.
modelling of calibrations can lead to highly misleading Brower, A. V. Z., 1994. Rapid morphological radiation and conver-
estimates of evolutionary timescales. Accordingly, recent gence among races of the butterfly Heliconius erato inferred
from patterns of mitochondrial DNA evolution. Proceedings of
work has focused on accounting for the uncertainty in fos- the National Academy of Sciences of the United States of Amer-
sil and geological calibrations (Ho and Phillips, 2009). ica, 91, 6491–6495.
There is also ongoing development of methods for model- Brown, W. M., George, M., and Wilson, A. C., 1979. Rapid evolu-
ling and quantifying the uncertainty in fossil calibrations tion of animal mitochondrial DNA. Proceedings of the National
(e.g., Marshall, 2008; Wilkinson et al., 2011). Some Academy of Sciences of the United States of America,
aspects of calibration methodology are poorly understood, 76, 1967–1971.
Cooper, A., and Penny, D., 1997. Mass survival of birds across the
such as the impact of using different methods for model- Cretaceous-Tertiary boundary: molecular evidence. Science,
ling the uncertainty in calibrations. One of the key points 275, 1109–1113.
of agreement is that best practice involves the use of mul- Dickerson, R. E., 1971. The structure of cytochrome c and the rates of
tiple calibrations throughout the phylogenetic tree. molecular evolution. Journal of Molecular Evolution, 1, 26–45.
The use of secondary calibrations in molecular dating Doolittle, R. F., and Blomback, B., 1964. Amino-acid sequence
has been the target of strong criticism (Graur and Martin, investigations of fibrinopeptides from various mammals: evolu-
tionary implications. Nature, 202, 147–152.
2004). Although the use of secondary calibrations is Doolittle, R. F., Feng, D.-F., Tsang, S., Cho, G., and Little, E., 1996.
sometimes unavoidable, good practice involves taking Determining divergence times of the major kingdoms of living
into account the uncertainty in previous molecular-clock organisms with a protein clock. Science, 271, 470–477.
estimates. Ignoring this uncertainty can lead to artificially García-Moreno, J., 2004. Is there a universal mtDNA clock for
precise estimates of evolutionary timescales. Instead, the birds? Journal of Avian Biology, 35, 465–468.
uncertainty should be incorporated into the dating analysis Graur, D., and Martin, W., 2004. Reading the entrails of chickens:
molecular timescales of evolution and the illusion of precision.
so that it can be included in the resulting estimates. This Trends in Genetics, 20, 80–86.
can readily be done in a Bayesian framework, in which Ho, S. Y. W., 2007. Calibrating molecular estimates of substitution
the user can choose prior distributions that reflect the rates and divergence times in birds. Journal of Avian Biology,
degree of uncertainty in the parameters and node times. 38, 409–414.
There is growing evidence that estimates of rates Ho, S. Y. W., and Lo, N., 2013. The insect molecular clock. Austra-
depend on the timescale of observation. Low rates of evo- lian Journal of Entomology, 52, 101–105.
lution are seen across long evolutionary time frames, such Ho, S. Y. W., and Phillips, M. J., 2009. Accounting for calibration
uncertainty in phylogenetic estimation of evolutionary diver-
as those analyzed in phylogenetic analyses of distant spe- gence times. Systematic Biology, 58, 367–380.
cies. In contrast, high evolutionary rates have been esti- Ho, S. Y. W., Lanfear, R., Bromham, L., Phillips, M. J., Soubrier, J.,
mated in studies of populations and pedigrees (Howell Rodrigo, A. G., and Cooper, A., 2011. Time-dependent rates of
et al., 2003). The exact causes of this disparity remain molecular evolution. Molecular Ecology, 20, 3087–3101.
unclear, although they are likely to include the effects of Howell, N., Bogolin Smejkal, C., Mackey, D. A., Chinnery, P. F.,
natural selection and inaccurate modelling of the molecu- Turnbull, D. M., and Herrnstadt, C., 2003. The pedigree rate of
sequence divergence in the human mitochondrial genome: there
lar evolutionary process (Ho et al., 2011). If the time is a difference between phylogenetic and pedigree rates. Ameri-
dependence of molecular rates is not taken into account, can Journal of Human Genetics, 72, 659–670.
evolutionary timescales can be under- or overestimated Kimura, M., 1968. Evolutionary rate at the molecular level. Nature,
by an order of magnitude. 217, 624–626.
588 MOLECULAR CLOCKS, HUMAN EVOLUTION

King, J. L., and Jukes, T. H., 1969. Non-Darwinian evolution. Sci- Wilkinson, R. D., Steiper, M. E., Soligo, C., Martin, R. D., Yang, Z.,
ence, 164, 788–798. and Tavaré, S., 2011. Dating primate divergences through an
Kohne, D. E., 1970. Evolution of higher-organism DNA. Quarterly integrated analysis of palaeontological and molecular data. Sys-
Reviews of Biophysics, 3, 327–375. tematic Biology, 60, 16–31.
Kumar, S., 2005. Molecular clocks: four decades of evolution. Wilson, A. C., Cann, R. L., Carr, S. M., George, M., Gyllensten,
Nature Reviews. Genetics, 6, 654–662. U. B., Helm-Bychowski, K. M., Higuchi, R. G., Palumbi,
Laird, C. D., McConaughy, B. L., and McCarthy, B. J., 1969. Rate S. R., Prager, E. M., Sage, R. D., and Stoneking, M., 1985. Mito-
of fixation of nucleotide substitutions in evolution. Nature, chondrial DNA and two perspectives on evolutionary genetics.
224, 149–154. Biological Journal of the Linnean Society, 26, 375–400.
Lanfear, R., Ho, S. Y. W., Davies, T. J., Moles, A. T., Aarssen, L., Wu, C.-I., and Li, W. H., 1985. Evidence for higher rates of nucleo-
Swenson, N. G., Warman, L., Zanne, A. E., and Allen, A. P., tide substitutions in rodents than in man. Proceedings of the
2013. Taller plants have lower rates of molecular evolution. National Academy of Sciences of the United States of America,
Nature Communications, 224, 1879. 82, 1741–1745.
Marshall, C. R., 2008. A simple method for bracketing absolute Zuckerkandl, E., and Pauling, L., 1962. Molecular disease, evolu-
divergence times on molecular phylogenies using multiple fossil tion and genetic heterogeneity. In Kasha, M., and Pullman, B.
calibration points. American Naturalist, 171, 726–742. (eds.), Horizons in Biochemistry. New York: Academic,
Morgan, G. J., 1998. Emile Zuckerkandl, Linus Pauling, and the pp. 189–225.
molecular evolutionary clock, 1959–1965. Journal of the His- Zuckerkandl, E., and Pauling, L., 1965. Evolutionary divergence
tory of Biology, 31, 155–178. and convergence in proteins. In Bryson, V., and Vogel, H. J.
Nabholz, B., Glémin, S., and Galtier, N., 2008. Strong variation of (eds.), Evolving Genes and Proteins. New York: Academic,
mitochondrial mutation rate across mammals – the longevity pp. 97–166.
hypothesis. Molecular Biology and Evolution, 25, 120–130.
Ohta, T., 1972. Evolutionary rate of cistrons and DNA divergence.
Journal of Molecular Evolution, 1, 150–157. Cross-references
Ohta, T., 1973. Slightly deleterious mutant substitutions in evolu- Gene Sequencing
tion. Nature, 246, 96–98. Molecular Clock Calibration
Papadopoulou, A., Anastasiou, I., and Vogler, A. P., 2010. Molecular Clocks, Human Evolution
Revisiting the insect mitochondrial molecular clock: the mid- Molecular Clocks, Relaxed Variant
Aegean trench calibration. Molecular Biology and Evolution, Molecular Dating of Evolutionary Events
27, 1659–1672. Molecular Rate Variation (Molecular Clocks)
Pulquério, M. J. F., and Nichols, R. A., 2007. Dates from the molec- Polymerase Chain Reaction DNA Amplification
ular clock: how wrong can we be? Trends in Ecology & Evolu-
tion, 22, 180–184.
Runnegar, B., 1982. A molecular-clock date for the origin of the ani-
mal phyla. Lethaia, 15, 199–205.
Sanderson, M. J., 1997. A nonparametric approach to estimating MOLECULAR CLOCKS, HUMAN EVOLUTION
divergence times in the absence of rate constancy. Molecular
Biology and Evolution, 14, 1218–1231.
Sarich, V. M., and Wilson, A. C., 1967a. Rates of albumin evolution Simon Y. W. Ho1 and Phillip Endicott2
1
in primates. Proceedings of the National Academy of Sciences of School of Biological Sciences, University of Sydney,
the United States of America, 58, 142–148. Sydney, NSW, Australia
Sarich, V. M., and Wilson, A. C., 1967b. Immunological time scale 2
Départment Hommes, Natures, Sociétés, Musée de
for hominid evolution. Science, 158, 1200–1203. l’Homme, Paris, France
Shields, G. F., and Wilson, A. C., 1987. Calibration of mitochon-
drial DNA evolution in geese. Journal of Molecular Evolution,
24, 212–217. Definition
Simpson, G. G., 1964. Organisms and molecules in evolution. Sci- Human evolution (molecular clocks). The timescale of
ence, 146, 1535–1538. human evolution and migration can be estimated from
Springer, M. S., 1997. Molecular clocks and the timing of the pla- genetic data that have been sampled from living, histori-
cental and marsupial radiations in relation to the Cretaceous-
Tertiary boundary. Journal of Mammalian Evolution, 4, cal, and ancient humans. This can be done using statistical
285–302. methods based on the molecular clock hypothesis.
Tajima, F., 1993. Simple methods for testing the molecular evolu-
tionary clock hypothesis. Genetics, 135, 599–607. Introduction
Thomas, J. A., Welch, J. J., Lanfear, R., and Bromham, L., 2010. The timescale of human evolutionary origins and migra-
A generation time effect on the rate of molecular evolution in
invertebrates. Molecular Biology and Evolution, 27, 1173–1180. tion across the globe has been a long-standing focus of sci-
Thorne, J. L., Kishino, H., and Painter, I. S., 1998. Estimating the entific research. Genetic studies of the human
rate of evolution of the rate of molecular evolution. Molecular evolutionary timescale were first performed in the 1960s,
Biology and Evolution, 15, 1647–1657. with the divergence time between humans and chimpan-
Weir, J. T., and Schluter, D., 2008. Calibrating the avian molecular zees being estimated using data from the albumin protein
clock. Molecular Ecology, 17, 2321–2328. (Sarich and Wilson, 1967). The complete sequence of
Welch, J. J., and Bromham, L., 2005. Molecular dating when rates
vary. Trends in Ecology & Evolution, 20, 320–327. the human mitochondrial genome, consisting of about
Welch, J. J., Bininda-Emonds, O. R. P., and Bromham, L., 2008. 16.5 thousand nucleotides, was published in 1981
Correlates of substitution rate variation in mammalian protein- (Anderson et al., 1981). Two decades later, the Interna-
coding sequences. BMC Evolutionary Biology, 8, 53. tional Human Genome Consortium (2001) released
http://www.springer.com/978-94-007-6303-6