The geologic time scale

Felix M. Gradsteina,* and James G. Oggb

a

Museum of Natural History, University of Oslo, N-0318 Oslo,

Norway; bDepartment of Earth and Atmospheric Sciences, Purdue
University, West Lafayette, IN 47907, USA
*To whom correspondence should be addressed (felix.gradstein@
nhm.uio.no)

Abstract
Construction and assembly of the Geologic Time Scale
involves: (a) constructing a relative (chronostratigraphic)
standard scale for key periods in the Earths rock record;
(b) identifying high-resolution linear age dates to calibrate
this relative scale in linear time; (c) astronomically tuning
intervals with cyclic sediments or stable isotope sequences
which have sufcient fossil or geomagnetic ties to be merged
in the standard scale, and increase its resolution; (d) interpolating for those relative time intervals where direct linear
age information is insufcient; and (e) estimating error bars
on the age of boundaries and on unit durations.

Time is an indispensable tool for all of us. The time kept

by innumerable watches and a great variety of clocks
regulates our everyday life, while the familiar calendar
governs our weekly, monthly, and yearly doings. These
eventually condense into the historical record of the
events over centuries. The standard unit of modern time
keeping is the second, defined by a precise number of
vibrations of the cesium atomic clock. The atomic second
is defined as the duration of 9.192.631.770 periods of the
radiation corresponding to the transition between two
hyperfine levels of the ground state of the Cesium 133
atom. This value was established to agree as closely as
possible with the ephemeris second based on the Earths
motion. The advantage of having the atomic second as
the unit of time in the International System of Units is the
relative ease, in theory, for anyone to build and calibrate
an atomic clock with a precision of 1 part per 1011 (or better). In practice, clocks are calibrated against broadcast
time signals, with frequency oscillations in Hertz being
the pendulum of the atomic time-keeping device.
The tick of the second paces the quick heart beat, and
traditionally was the 60th part of the 60th part of the
24th part of the 24-h day, with the minute and the hour

being convenient multiples to organize our daily life and

productivity. The day carries the record of light and dark,
the month the regularly returning shapes of the moon,
and the year the cycle of the seasons and the apparent
path of the sun. All is clear, and we have grown up with
the notion that time is a vector, pointing from the present to the future. Events along its path mark the arrow
of time, and the arrow is graded either in relative natural units, or in units of durationthe standard second
and its multiples, like hours and years, and millions of
years.

Geologic time and the sediment record

A majority of geologists consider time as a vector pointing from the distant past to the present. Instead of distant past, the term deep time has been coined in the
vernacular. What is exactly the concept of geologic time,
what are its natural units, how are they defined, and how
do we use these units properly? A good understanding of
geologic time is vital for every scientist who deals with
events in the Earth sediment and rock record, or with
the genetic record of evolution in living organisms, especially those who strive to understand past processes and
determine rates of change. This understanding takes
place in a framework called Earth Geological History, a
super calendar of local and global events. The challenge
to this understanding is reading, organizing, and sorting the Earths stone calendar pages. In the process, we
often have to reconstruct the content of missing pages.
Correlation of the rock record between regions is a vital
part of the reconstruction process.
One of the earliest reconstructions is by Nicolas Steno
(16311687) who made careful and original stratigraphic
observations. Based on these observations, Steno concluded that the Earths strata contain the superimposed
records of a chronological sequence of events that can be
correlated worldwide. Geological correlation formally is
expressed in terms of five consecutive operations (each is
followed by one or more examples):
(a) Rock units, like formations or well log intervals =
lithostratigraphic correlation
Kimmeridge Clay Formation of England

F. M. Gradstein and J. G. Ogg. The geologic time scale. Pp. 2634 in The Timetree of Life, S. B. Hedges and S. Kumar, Eds. (Oxford University
Press, 2009).

Hedges.indb 26

1/28/2009 1:24:44 PM

The Geologic Time Scale

(b) Fossil units, like zones = biostratigraphic

correlation
Turrilina alsatica benthic foraminifer zone
(c) Relative time units = geochronologic (Earth time)
correlations
Jurassic Period, Eocene Epoch, Oxfordian Age,
polarity chron C29r
(d) Rocks deposited during these time units =
chronostratigraphic (timerock) correlation
Jurassic System, Eocene Series, Oxfordian Stage,
polarity zone C29r
(e) Linear time units or ages = geochronologic
correlation
150 million years ago (Ma), 10,000 years ago (ka)
Without correlation to a global reference scale, successions of strata or events in time derived in one area are
unique and contribute nothing to the understanding of
Earth history elsewhere. The rules of hierarchy in geological correlation, from rocks and fossils to relative and
linear time, are carefully laid down in the International
Stratigraphic Guide. An abbreviated copy of this rule
book with further references may be found on the Web
site of the International Commission on Stratigraphy
(ICS) under www.stratigraphy.org.
Before we deal with linear geological time, a few words
are necessary about the common geological calendar
built from relative age units. This chronostratigraphic
scheme is not unlike a historical calendar in which societal periods, for example, the Minoan Period, the reign
of Louis XIV, the American Civil War, are used as building blocks, devoid of a linear scale. Archeological relics
deposited during these intervals (e.g., the Palace of Minos
on Crete, Versailles or spent cannon balls at Gettysburg,
respectively) comprise the associated physical and
chronostratigraphic record. A chronostratigraphic scale
is assembled from rock sequences stacked and segmented in relative units based on their unique fossil and
physical content. When unique local fossil and physical
records are matched with those of other rock sequences
across the globein a process known as stratigraphic
correlationa relative scale can be assembled that, when
calibrated to stage type sections, becomes a chronostratigraphic scale. The standard chronostratigraphic scale, in
downloadable graphics format, is available from the ICS
Web site. This time scale is made of successive stages in
the rock record, like Cenomanian, Turonian, Coniacian,
and so on, within the Cretaceous system.
Originally, each stage unit was a well-defined body
of rocks at a specific location of an assigned and agreed
upon relative age span, younger than typical rocks of the

Hedges.indb 27

27

underlying stage and older than the typical rocks of the

next higher stage. This is the concept of defining stage
units with type sections, commonly referred to as stratotype sections. The principles and building blocks of this
chronostratigraphy were slowly established during centuries of study in many discontinuous and incomplete
outcrop sections. Inevitably, lateral changes in lithology
between regions and lack of agreement on criteria, particularly in which fossils were characteristics of a relative unit of rock, have always resulted in a considerable
amount of confusion and disagreement on stage nomenclature and stage use. Almost invariably classical stage
stratotypes turned out to only represent part of stages.
Hence, a suite of global subdivisions with precise correlation horizons was required.

Global stratotype section and point

Now, relatively rapid progress is being made with definition of Global Stratotype Sections and Points (GSSPs)
to fi x the lower boundary of all geologic stages, using
discrete fossil and physical events that correlate well in
the rock record. For the ladder of chronostratigraphy,
this GSSP concept switches the emphasis from marking
the spaces between steps (stage stratotypes) to fi xing the
rungs (boundaries of stages).
Each progressive pair of GSSPs in the rock record also
precisely defines the associated subdivision of geologic
time. It is now 25 years ago that a golden spike struck
the first GSSP. This event of historic proportions for
the geologic time scale involved the boundary between
the Silurian and Devonian Periods, or rather the lower
limit of the Devonian, at a locality called Klonk in
Czechoslovakia.
The problem of the SilurianDevonian boundary and
its consensus settlement in the Klonk section hinged on a
century-old debate known as the Hercynian Question
that touched many outstanding geoscientists of the nineteenth century. The issue came to the forefront after 1877,
when Kaiser stated that the youngest stages (tages) of
Barrandes Silurian System in Bohemia correspond to
the Devonian System in the Harz Mountains of Germany
and other regions. Kaisers findings contrasted with the
conventional nineteenth century wisdom that graptolite
fossils became extinct at the end of the Silurian. Eventually, it became clear that so-called Silurian graptolites
in some sections occur together with so-called Devonian
fossils in other sections, leading to the modern consensus that graptolites are not limited to Silurian strata.
A bronze plaque in the Klonk outcrop shows the exact
position of the modern SilurianDevonian Boundary,

1/28/2009 1:24:47 PM

28

THE TIMETREE OF LIFE

At present over 55 GSSPs have been defined (Fig. 1;

see www.stratigraphy.org for details), but there are more
stages in the Phanerozoic Eon in need of base definition.
Fortunately, a majority of those now have target definitions, and are awaiting consensus on the best outcrop or
borehole section to place a golden spike. Thus, with the
definitions in place, we can proceed to scale the deep
time stage units linearly.
This brings us to Geochronology, referring to the geochronologic calendar of Earth events called the Geologic
Time Scale. While the chronostratigraphic scale is a
convention to be agreed upon rather than discovered,
calibration of the scale in seconds and (mega-) years
is a matter for discovery and estimation rather than
agreement. Like human time, linear geological time is
expressed in units of standard durationthe second and
hence (thousands or millions of) years.

Building a geological time scale

The ideal time scale is built from accurate radiometric
ages, taken precisely at stage boundaries throughout the
stratigraphic column in the Phanerozoic Eon. For more
detailed resolution, the exact number of orbitally tuned
sedimentary cycles is counted within each stage, such
that calibrations and correlation may be achieved within

orb

ital

tun
ing
sea
floo
spr
ead r
ing
dire
ct d
atin
g
det
aile
dat d di
rec
ing
t
pro
por
sub tion
al
zon
sca e scali
led
ng
stan com
pos
dar
ite
cub d
ic
ma splin
e
x.
erro likelih fitting
r es ood &
tim
atio
n

which is taken at the base of the Lochkovian Stage, the

lowest stage in the Devonian. The base of the Lochkovian
Stage is defined by the first occurrence of the Devonian
graptolite Monograptus uniformis in bed #20 of the
Klonk Section, northeast of the village of Suchomasty.
The lower Lochkovian index trilobites with representatives of the Warburgella rugulosa group occur in the next
younger limestone bed #21 of that section.
The concept of the GSSP has gained acceptance among
those stratigraphers who consider it a pragmatic and
practical solution to the common problem that conventional stage type sections inevitably leave gaps, or lead
to overlap between successive stages. The boundary stratotype very much relies on the notion that it is possible
to arrive at accuracy in correlation through the use of
events, like a geomagnetic reversal, a global change in a
stable isotope value, or the evolutionary appearance of
one or more prominent and widespread fossil taxa. Thus,
the limits of a stage can now be defined with multiple
event criteria that to the best of our current knowledge
are synchronous over the world. Delimiting successive
stages in a clear and practical manner enhances their
value as standard units in chronostratigraphy and ultimately in geochronology. Without standardized units
neither the (relative) stratigraphic scale nor the (linear)
time scale can exist.

Ma
0
Cenozoic
90

180

Cretaceous
Jurassic
Triassic

270

Permian
Carboniferous

360

450

540

Devonian
Silurian
Ordovician
Cambrian
Methods used to construct Geologic Time Scale 2004 (GTS 2004)

Fig. 1 Methods used to construct Geologic Time Scale 2004 (GTS2004) (1).

Hedges.indb 28

1/28/2009 1:24:47 PM

The Geologic Time Scale

20 thousand years for the last 540 million years or so . . . .

If this sounds too good to be true, let it rest. Back to
reality.

Geologic reality is schematically illustrated in Fig. 1,

providing a quick overview of the actual methodology applied to construct Geologic Time Scale 2004
(GTS2004) (1), the most recent standard time scale.
Before the Cambrian, the first period of the Phanerozoic
Era, the geologic time scale is less sophisticated, and
based only on sparse radiometric dates. The steps
involved in Phanerozoic time scale construction may be
summarized as follows:
(a) Construct a relative (chronostratigraphic) standard
scale for the key periods in the Earths rock
record
(b) Identify high-resolution linear age dates to calibrate
this relative scale in linear time
(c) Astronomically tune (see later) intervals with cyclic
sediments or stable isotope sequences which have
sufficient fossil or geomagnetic correlation ties to
be merged in the standard scale, and increase its
resolution
(d) Interpolate for those relative time intervals where
direct linear age information is insufficient
(e) Estimate error bars on the age of boundaries and on
unit durations.
The first step, integrating multiple types of stratigraphic
information to construct the standard chronostratigraphic scale, is the most time consuming; it summarizes
and synthesizes centuries of detailed geological research
and tries to understand all relative correlations and calibration to the standard.
The second and third steps, identifying which radiometric and cycle-stratigraphic studies to use as the primary constraints for assigning linear ages, are the ones
that have much evolved. Historically, Phanerozoic time
scale building went from an exercise with very few and
relatively inaccurate radiometric dates, as available to
the pioneer of the geologic time scale Arthur Holmes,
to one with many dates with greatly varying analytical
precision, as in the mid-1980s. Next, time scale studies
started to appear of selected intervals, like Paleogene,
Late Cretaceous, or Ordovician, that selected a small
suite of radiometric dates with high analytical precision
and relatively precise stratigraphic position.
At the same time, a high-resolution Neogene time
scale started to take shape, using orbital tuning of long
sequences of sedimentary and/or oxygen isotope cycles
in the Mediterranean region and in Atlantic and Pacific

Hedges.indb 29

29

pelagic sediments. The present trend for the pre-Neogene

is to incorporate radiometric dates that have very small
analytical and stratigraphic uncertainties, and pass the
most stringent tests.
The fourth step, interpolating the stratigraphic and
radiometric information, has much evolved. An early
method already constructed the basic two-way graph,
used until now. It plotted the cumulative sum of maximum global thickness of strata per stratigraphic unit
along the vertical axis and selected radiometric dates
from volcanic tuffs and other suitable layers along the
horizontal linear axis. This best fit line method interpolated ages to the stages, but is a far cry from methods used
today that scale stages along the vertical axis with composite standards of fossil zones. In the mid-1990s, Frits
Agterberg and Felix Gradstein started to apply mathematical/statistical error analysis to the time scale ages,
which, for the first time, allowed them to assign fairly
realistic error bars to ages of Mesozoic stage boundaries,
a trend that persists today for the whole of Phanerozoic
below the Neogene.
The following is a simplified introduction to the modern building tools depicted in Fig. 1.

Music of the spheres

Let us start with a brief outline of the principle of the
sedimentary cycles approach to time scale building, as is
now standard for the last 23 Ma (Neogene), and provides
superior resolution and precision. Gravitational interactions of the Earth with the Sun, Moon, and other planets cause systematic changes in the Earths orbital and
rotational system. These interactions give rise to cyclic
oscillations in the eccentricity of the Earths orbit, and
in the tilt and precession of the Earths axis, with mean
dominant periods of 100,000, 41,000, and 21,000 years,
respectively. The associated cyclic variations in annual
and seasonal solar radiation onto different latitudes alter
long-term climate in colder vs. warmer and wetter vs.
dryer periods that lead to easily recognizable sedimentary cycles, such as regular interbeds of limy and shaly
facies. Massive outcrops of hundreds or thousands of
such cycles are observed in numerous geological basins,
for example around the Mediterranean, and in sediment
cores from ocean-drilling sites.
Counting of this centimeter to meter thick cycles in
great detail over land outcrops and in ocean-drilling
wells, combined with the additional correlation aids
provided by magnetostratigraphy, oxygen isotope stratigraphy, and biostratigraphy, produced a very detailed

1/28/2009 1:24:48 PM

30

THE TIMETREE OF LIFE

Neogene cycle pattern. The critical step is the direct

linkage of each cycle to the theoretical computed astronomical scale of the 21,000, 41,000, and 100,000-year
paleoclimatic cycles. This astronomical tuning of the
geological cycle record from the Mediterranean and
Atlantic by earth scientists at Utrecht and Cambridge
Universities such as Luc Lourens, Frits Hilgen, and Nick
Shackleton led to unprecedented accuracy and resolution
for the last 23 million years (2). In New Zealand, Tim
Naish and colleagues have calibrated the upper Neogene
record to the standard Neogene time scale. Using the
high-resolution land-based cycle, isotope and magnetic
record in the Wanganui Basin, these authors thereby
transferred precise absolute ages to local shallow marine
sediments and demonstrated the link between sequence
and cycle stratigraphy.
Efforts are underway to extend the continuous astrochronologic scale back into Oligocene and Eocene by
applying a combination of cycle stratigraphy, improved
astronomical projections, oxygen isotope stratigraphy,
and magnetostratigraphy to the deep sea record.
A special application of orbitally tuned cyclic sediment
sequences is to rubber-band stratigraphically floating
units, like parts of Paleocene, Albian, and parts of Lower
Jurassic, skilfully executed by specialists like Ursula
Rohl, Tim Herbert, and Graham Weedon. A quantitative
estimation of the duration of all cycles within a stratigraphic unit allows estimates of their duration.

Decay of atoms
For rocks older than Neogene, the derivation of a numerical time scale depends on the availability of suitable
radiometric ages. Radiometric dating generally involves
measuring the ratio of the original element in a mineral,
like sanidine feldspar or zircon, to its isotopic daughter
products. The age of a mineral may then be calculated
by means of the isotopic decay constant. Depending on
the half-life of the element, several radiometric clocks
are available; 40Ar/39Ar and the family of U/Pb isotopes
are the most common suites nowadays applied to the
Phanerozoic, because of analytical precision and utility
with tuffaceous beds in marine or non-marine sequences.
Radiometric dating of sedimentary rocks follows several
geological strategies:
(a) Dating of igneous intrusions within sediments
records the time of primary cooling, when the
igneous rocks were emplaced and had cooled
sufficiently (to a few hundreds of degrees
centigrade) to set the radiometric decay clock in

Hedges.indb 30

action. Because of uncertainty in the relation of the

intrusion to the host sediment, such dates may be of
limited stratigraphic use.
(b) Dating of volcanic flows and tuffs as part of the
stratified sedimentary succession.
(c) Dating of authigenic sedimentary minerals, mainly
involving glauconite, found widespread in many
marine sediments. Mild heating or overburden
pressure after burial may lead to loss of argon, the
daughter product measured in the 40K/40Ar clock
in glauconite. Another problem is that glauconite
also contains an abundance of tiny flakes that allow
diff usion of Ar at low temperatures. The result is
that glauconite dates may be too young. Because
of such problems which may be difficult to detect,
modern geologic time scales avoid dates based on
glauconite.
Calibration of the decay constants or measurement
standards can be enhanced by intercalibration to other
radiometric methods, or by dating rocks of a known
age, for example a volcanic ash within an astronomically
tuned succession. Astrochronologic and interlaboratory
recalibration of the 40Ar/39Ar monitor standard indicates that many of the 40Ar/39Ar ages used in previous
Phanerozoic time scales are too young by about 0.5% to
1.0%. For example, the 65.0 Ma age, assigned 10 years
ago to the top Cretaceous, is now 65.5 Ma.
Radiometric dating techniques with less than 1%
analytical error are providing suites of high-precision
U/Pb and Ar/Ar dates for the Paleozoic and Mesozoic.
Surprisingly, perhaps, there are only seven direct age
dates on period or stage boundaries (Fig. 1), with a majority of the 200+ radiometric age dates used for GTS2004
floating at some level within a stage.
The integration of this level of chronometric precision
with high-resolution biostratigraphy, magnetostratigraphy, or cyclic scales is a major challenge to time scale
studies. Even the most detailed biostratigraphic scheme
probably has no biozonal units of less than 0.51.0 million
year (my) duration, not to speak of the actual precision
in dating a particular stratigraphic piercing point, for
which an U/Pb age estimate would be available with an
analytical uncertainty of 0.1 to 0.5 my. Similarly, combination of analytically less precise K/Ar dates with
much more precise Ar/Ar or U/Pb dates in statistical
interpolations creates a strong bias toward the latter, despite the fact that both may have equal litho-, bio-, and
chronostratigraphic precision.
Nevertheless, the combination of precise stratigraphic definitions through GSSPs and accurate

1/28/2009 1:24:48 PM

The Geologic Time Scale

radiometric dates near these levels is paving the way

for a substantial increase in the precision and accuracy of the Geologic Time Scale. The bases of Paleozoic,
Mesozoic, and Cenozoic Eras are bracketed by analytically precise ages at their GSSP or primary correlation
markers542.0 1.0 Ma, 251.0 0.4 Ma, and 65.5 0.3
Ma respectivelyand there are direct age dates for the
base Carboniferous, base Permian, base Jurassic, base
Aptian, base Cenomanian, and the base Oligocene. Most
other period or stage boundaries lack direct age control.
Therefore, the third step, linear interpolation, also plays
a key role for the time scale.

Interpolation and statistics

Despite the progress in standardization and dating, parts
of the Mesozoic and Paleozoic Eras have sparse radiometric records (see Fig. 2). Ideally, each of the 90+ stage
boundaries that comprise the Paleozoic, Mesozoic, and
Cenozoic Eras of the Phanerozoic Eon should coincide
with an accurate radiometric date from volcanic ash.
However, this coincidence is rare in the geological record.
The combined number of fossil events and magnetic
reversals far exceeds the total number of radiometrically datable horizons in the Phanerozoic. Therefore, a
framework of bio-, magneto-, and chronostratigraphy
provides the principal fabric for stretching of the relative
time scale between dated tiepoints on the loom of linear
time. For such stretching, interpolation methods that are
employed are both geological and statistical in nature.
Earlier, we mentioned the outdated method of plotting the cumulative global thickness of periods against
selected linear age dates. Among the modern geological
scaling methods, an assumption of relative constancy of
seafloor spreading over limited periods of time is a common tool for interpolating the Latest Cretaceous through
Paleogene relative scale. Magnetic polarity chrons, the
units of magnetochronology, can be recognized both
on the ocean floor as magnetic anomalies measured in
kilometers from the mid-ocean spreading center, and
in marine sediments as polarity zones that contain biostratigraphic events and can be linked to linear time.
Knowing the linear age of a few ocean crust magnetic
anomalies (earth magnetic reversals or magnetochrons)
allows interpolation of the ages of the intervening magnetic pattern, which in turn can be correlated to the
fossil record and geological stage boundaries. The subduction of pre-late Jurassic oceanic crust precludes such
an interpolation approach for older Mesozoic and the
Paleozoic strata.

Hedges.indb 31

31

A second geological method involves building a

zonal composite to scale stages. Several outstanding
examples are documented in GTS2004 built by a large
international team of scientists under the direction of
Felix Gradstein and James Ogg. For this scale, Roger
Cooper and colleagues have built a very detailed composite standard of graptolite zones from 200+ sections
in oceanic and slope environment basins for the uppermost Cambrian, Ordovician, and Silurian intervals.
With zone thickness taken as directly proportional to
zone duration, the detailed composite sequence was
scaled using selected, high-precision age dates. For the
Carboniferous through Permian, a composite standard of conodont, fusulinid, and ammonoid events
from many classical sections can now be calibrated to a
combination of U/Pb and 40Ar/39Ar dates. A composite
standard of conodont zones was used for early Triassic.
This procedure directly scales all stage boundaries and
biostratigraphic horizons.
The two-way graph of linear age vs. scaled stages requires
a best fitting method, and that is where statistics comes
into play, with cubic spline fitting and maximum likelihood interpolation most suitable. On the time scale chart
(late 2008 edition; Fig. 3), a majority of Phanerozoic stage
boundaries for the first time show error bars; an exception
is the Neogene Period where analytical errors are negligible. The error bars reflect both radiometric and stratigraphic uncertainty; in addition, error bars were calculated
on stage duration. Uncertainty in the duration of the age
units is less than the error in age of their boundaries.

TS-Creator
Now, Adam Lugowski, Ogg, and Gradstein are producing
an electronic version of the Geologic Time Scale with the
international standard bio-magneto-sequence time scale
charts. There are charts for Paleozoic, Mesozoic, and
Cenozoic Eras and for each period. This JAVA language
package, called TS-Creator, can be freely downloaded
from the ICS website (www.stratigraphy.org).
It contains tables of Cambrian through Holocene
stratigraphic events calibrated to GTS2004 ages. There
are nearly 15,000 biostratigraphic, sea-level, and magnetic zones and levels, plus a suite of geochemical
curves. Documentation of zonal definitions, relative age
assignments, and how these events were recalibrated
to GTS2004 was also compiled. This included updating cross correlations and enhancing detail for selected
stratigraphic methods using trilobites, conodonts, graptolites, ammonoids, fusulinids, chitinozoans, megaspores,

1/28/2009 1:24:48 PM

Stage/Age
boundaries

Epoch
Pleistocene

Holocene
Pliocene

L
E

Miocene
Oligocene

Eocene
Paleocene

M
E
L
E
L
M

Piacenzian/Gelasian
Zanclean
Messinian

Tortonian

Serravallian
Langhian

Burdigalian
Aquitanian
Chattian
Rupelian
Priabonian
Bartonian

Lutetian

Ypresian

L
M

Thanetian
Selandian

UP
Ar/ b
A
Rb r, K-A
-Sr
r

Cenozoic

Quatern.

Paleogene Neogene

Era Period

Danian
Maastrichtian
Campanian

Cretaceous

Late

Albian
Aptian

Early

Barremian
Hauterivian
Valanginian
Berriasian
Tithonian
Kimmeridgian
Oxfordian
Callovian

Late

Jurassic

Mesozoic

100

Santonian
Coniacian
Turonian
Cenomanian

Middle

Bathonian

Early

Pliensbachian

Bajocian
Aalenian
Toarcian

Sinemurian
Hettangian
Rhaetian

Triassic

200

Norian

Late

Carnian

Middle

Ladinian
Anisian
Olenekian

Early

Carboniferous

300

Guadalupian

Cisuralian

Induan
Changhsingian

Wuchiapingian
Capitanian
Wordian/Roadian
Kungurian
Artinskian
Sakmarian

PennMississippian sylvanian

Permian

Lopingian

Late
Middle
Early

Asselian
Gzhelian
Kasimovian
Moscovian
Bashkirian

Late

Serpukhovian

Middle

Visean

Early

Tournaisian

Devonian

Frasnian

Middle

Cambrian

Eifelian
Pragian
Lochkovian

Pridoli

Ludlow
Wenlock

Ludfordian/Gorstian
Homerian/Sheinwoodian
Telychian

Llandovery

Aeronian/Rhuddanian
Hirnantian

Late

Katian
Sandbian

Middle

Darriwilian
Floian

Early
Furongian

500

Givetian
Emsian

Early

Ordovician Silurian

400

Paleozoic

Famennian

Late

Series 3
Series 2
Series 1

Tremadocian
Stage 10
Stage 9
Paibian
Stage 7
Drumian
Stage 5
Stage 4
Stage 3
Stage 2

Legend
GSSP
Direct radiometric constraint
on GSSP rocks
Other radiometric date

Stage 1

20
0
10
# radiometric dates
Fig. 2 Geologic Time Scale 2004 showing which stage and period boundaries have a Global Boundary Stratotype Section and Point
(GSSP), which ones are dated directly and which other age dates were used. Intervals with sparse or no dates required interpolation.

Hedges.indb 32

1/28/2009 1:24:48 PM

Series
Epoch

Upper

Stage
Age

Berriasian

Valanginian

Hauterivian

Barremian

Aptian

Albian

Cenomanian

Turonian

Coniacian

Santonian

Campanian

Maastrichtian

Danian

Selandian

Thanetian

Ypresian

Lutetian

Bartonian

Priabonian

Rupelian

Chattian

Aquitanian

Burdigalian

Langhian

Serravallian

Tortonian

Messinian

Zanclean

Piacenzian

Gelasian

Calabrian

Ionian

Age
Ma
145.5 4.0

140.2 3.0

~ 133.9

130.0 1.5

125.0 1.0

112.0 1.0

99.6 0.9

93.6 0.8

~ 88.6

85.8 0.7

83.5 0.7

70.6 0.6

65.5 0.3

~ 61.1

58.7 0.2

55.8 0.2

48.6 0.2

40.4 0.2

37.2 0.1

33.9 0.1

28.4 0.1

23.03

20.43

15.97

13.82

11.608

7.246

5.332

3.600

2.588

1.806

0.781

0.126

0.0117

Bashkirian

Lower

Lower

Middle

345.3 2.1

328.3 1.6

318.1 1.3

311.7 1.1

307.2 1.0

303.4 0.9

299.0 0.8

294.6 0.8

284.4 0.7

275.6 0.7

270.6 0.7

268.0 0.7

265.8 0.7

260.4 0.7

253.8 0.7

251.0 0.4

~ 249.5

~ 245.9

237.0 2.0

~ 228.7

216.5 2.0

203.6 1.5

199.6 0.6

196.5 1.0

189.6 1.5

183.0 1.5

175.6 2.0

171.6 3.0

167.7 3.5

164.7 4.0

161.2 4.0

~ 155.6

150.8 4.0

145.5 4.0

Tournaisian 359.2 2.5

Visean

Upper Serpukhovian

Moscovian

Kasimovian

Gzhelian

Asselian

Sakmarian

Artinskian

Kungurian

Roadian

Wordian

Capitanian

Wuchiapingian

Changhsingian

Induan

Olenekian

Anisian

Ladinian

Carnian

Norian

Rhaetian

Hettangian

Sinemurian

Pliensbachian

Toarcian

Aalenian

Bajocian

Bathonian

Callovian

Oxfordian

Kimmeridgian

Middle

Upper

Cisuralian

Guadalupian

Lopingian

Lower

Middle

Upper

Lower

Middle

Upper

Stage
Age

Tithonian

Age
Ma

GSSP

Stage
Age

Floian

Dapingian

Darriwilian

Sandbian

Katian

Hirnantian

Rhuddanian

Aeronian

Telychian

Sheinwoodian

Homerian

Gorstian

Ludfordian

Lochkovian

Pragian

Emsian

Eifelian

Givetian

Frasnian

Famennian

Age
Ma
December 2008

478.6 1.7

471.8 1.6

468.1 1.6

460.9 1.6

455.8 1.6

445.6 1.5

443.7 1.5

439.0 1.8

436.0 1.9

428.2 2.3

426.2 2.4

422.9 2.5

421.3 2.6

418.7 2.7

416.0 2.8

411.2 2.8

407.0 2.8

397.5 2.7

391.8 2.7

385.3 2.6

374.5 2.6

359.2 2.5

Erathem
Era

Siderian

Rhyacian

Orosirian

Statherian

Calymmian

Ectasian

Stenian

Tonian

Cryogenian

Ediacaran

System
Period

Hadean (informal)

Eoarchean

Paleoarchean

Mesoarchean

Neoarchean

Paleoproterozoic

Mesoproterozoic

Neoproterozoic

~4600

4000

3600

3200

2800

2500

2300

2050

1800

1600

1400

1200

1000

850

~635

542

Age
Ma

Subdivisions of the global geologic record are

Tremadocian
formally defined by their lower boundary. Each unit
488.3 1.7
of the Phanerozoic (~542 Ma to Present) and the
Stage 10
base of Ediacaran are defined by a basal Global
~ 492 *
Standard Section and Point (GSSP
), whereas
Stage 9
Furongian
~ 496 *
Precambrian units are formally subdivided by
Paibian
absolute age (Global Standard Stratigraphic Age,
~ 499
GSSA). Details of each GSSP are posted on the
Guzhangian
ICS website (www.stratigraphy.org).
~ 503
Numerical ages of the unit boundaries in the
Series 3
Drumian
~ 506.5
Phanerozoic are subject to revision. Some stages
Stage 5
within the Cambrian will be formally named upon
~ 510 *
international agreement on their GSSP limits. Most
Stage 4
sub-Series boundaries (e.g., Middle and Upper
~ 515 *
Series 2
Aptian) are not formally defined.
Stage 3
~ 521 *
Colors are according to the Commission for the
Stage 2
Geological Map of the World (www.cgmw.org).
~ 528 *
Terreneuvian
The listed numerical ages are from 'A Geologic
Fortunian 542.0 1.0
Time Scale 2004', by F.M. Gradstein, J.G. Ogg,
This chart was drafted by Gabi Ogg. Intra Cambrian unit ages A.G. Smith, et al. (2004; Cambridge University Press)
with * are informal, and awaiting ratified definitions.
and The Concise Geologic Time Scale by J.G. Ogg,
Copyright 2008 International Commission on Stratigraphy
G. Ogg and F.M. Gradstein (2008).
Lower

Middle

Upper

Llandovery

Wenlock

Ludlow

Pridoli

Lower

Middle

Upper

GSSP
GSSA

Fig. 3 The International Stratigraphic Chart summarizes the set of chronostratigraphic units (geologic stages, periods) and their computed ages, which are the main framework for
Geologic Time Scale 2004. Uncertainties on ages are expressed as two-sigma (95% condence). This version incorporates changes made by the International Commission up to
December, 2008.

* Definition of the Quaternary and revision of the Pleistocene are under discussion. Base of the Pleistocene is at 1.81 Ma
(base of Calabrian), but may be extended to 2.59 Ma (base of Gelasian). The historic Tertiary comprises the Paleogene
and Neogene, and has no official rank.

Lower

Upper

Paleocene

Eocene

Oligocene

Miocene

Pliocene

Pleistocene

Holocene

GSSP

International Commission on Stratigraphy

GSSP

INTERNATIONAL STRATIGRAPHIC CHART

Series
Epoch

ICS

Phanerozoic

Eonothem
Eon
Erathem
Era
System
Period

Quaternary *

Phanerozoic
Cenozoic
Neogene

Paleogene

Mesozoic
Cretaceous

Eonothem
Eon
Erathem
Era
System
Period

Meso zoic
Jurassic
Triassic
Paleo zoic
Carboniferous
Permian

Series
Epoch

Pennsylvanian
Mississippian

Eonothem
Eon
Erathem
Era
System
Period

Devonian
Phanerozoic
Paleo zoic
Silurian
Ordovician
Cambrian

Eonothem
Eon

Precambrian
Archean
Proterozoic

Hedges.indb 33

1/28/2009 1:24:49 PM

34

THE TIMETREE OF LIFE

nannofossils, foraminifers, dinoflagellates, radiolarians,

diatoms, strontium isotope, and C-org curves.
Numerical ages are calculated within the database
using the calibrations; therefore, all ages can be automatically recomputed when control ages are improved in
future time scales. Regional scales of selected areas (e.g.,
Russia, China, North America, and New Zealand) are
also included.
TS-Creator automatically takes the reference database, gets instructions from the user on the stratigraphic
interval and stratigraphic information to be displayed,
and then generates both on-screen and scalable-vector
graphic (SVG) renditions that directly input into drafting programs. Next, the user can click on a value, zone,
or boundary in the charts on the computer screen, and
a window opens with an explanation of the calibration,
definition, and interpolated age. This hot-linked chart
suite is currently a back-looking reference to information
in the source tables, but in the future will also provide
links to other tables and text from the GTS2004 book,
images of stage-boundary outcrops and fossil taxa, and
the additional enhancements anticipated during the
major update for Geologic Time Scale 2010.

Additional information on geologic time

The goal of this brief synopsis was to introduce the basic

concepts involved in the construction of the geologic time
scale. Further details can be found elsewhere (118).

References
1. F. M. Gradstein, J. G. Ogg, A. G. Smith, A Geologic Time
Scale 2004 (Cambridge University Press, New York, 2004).

Hedges.indb 34

2.

3.
4.
5.

6.
7.
8.
9.
10.
11.
12.

13.
14.
15.
16.
17.
18.

L. Lourens, F. Hilgen, N. J. Shackleton, L. Laskar,

D. Wilson, in Geologic Time Scale 2004, F. M. Gradstein,
J. G. Ogg, A. G. Smith, Eds. (Cambridge University Press,
New York, 2004), pp. 409440.
S. C. Cande, D. V. Kent, J. Geophys. Res. Solid Earth 100,
6093 (1995).
S. A. Bowring, D. H. Erwin, Y. Isozaki, Proc. Natl. Acad.
Sci. U.S.A. 96, 8827 (1998).
R. M. Carter, T. R. Naish, Eds., The High-Resolution
Chronostratigraphic and Sequence Stratigraphic Record
of the Plio-Pleistocene Wanganui Basin (Institute
of Geological and Nuclear Sciences, New Zealand,
1999).
F. M. Gradstein, et al., SEPM Spec. Publ. 54, 95 (1995).
W. B. Harland et al., A Geologic Time Scale 1989
(Cambridge University Press, Cambridge, 1990).
F. J. Hilgen, W. Krijgsman, C. G. Langereis, L. J. Lourens,
EOS Trans. Am. Geophys. Union 78, 285 (1997).
A. Holmes, Trans. Geol. Soc. Glasgow 21, 117 (1947).
A. Holmes, Trans. Edinburgh Geol. Soc. 17, 183 (1960).
S. L. Kamo et al., Earth and Planetary Science Letters 214,
75 (2003).
A. Martinsson, The Siluro-Devonian Boundary. International Union of Geological Sciences Series A, 5
(Schweizerbartsche Verlagsbuchhandlung, Stuttgart,
1977).
P. R. Renne et al., Geology 22, 783 (1994).
N. J. Shackleton, S. J. Crowhurst, G. P. Weedon, J. Laskar,
Phil. Trans. Roy. Soc. Lond. A 357, 1907 (1999).
G. P. Weedon, H. C. Jenkyns, A. L. Coe, S. P. Hesselbo,
Phil. Trans. Roy. Soc. Lond. A 357, 1787 (1999).
T. D. Herbert, S. L. DHondt, I. Premoli-Silva, E. Erba,
A. G. Fischer, SEPM Spec. Vol. 54, 81 (1995).
J. D. Obradovich, Geol. Assoc. Canada Spec. Pap. 39, 379
(1993).
U. O. Rhl, J.G., T. L. Geib, G. Wefer, Geol. Soc., Spec.
Publ. 183, 163 (2001).

1/28/2009 1:24:49 PM