See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/328005106

The evolution of plate tectonics

Article  in  Philosophical Transactions of The Royal Society A Mathematical Physical and Engineering Sciences · November 2018
DOI: 10.1098/rsta.2017.0406

CITATIONS READS

44 4,720

1 author:

Bob Stern
University of Texas at Dallas
544 PUBLICATIONS   30,067 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Subduction Initiation and Tethyan Suturing-Zagros Orogenic Belt View project

Late Neoproterozoic Magmatic “Flare–up” along the Northern Margin of Gondwana View project

All content following this page was uploaded by Bob Stern on 24 April 2019.

The user has requested enhancement of the downloaded file.

 Downloaded from http://rsta.royalsocietypublishing.org/ on October 2, 2018

The evolution of plate tectonics

Robert J. Stern
rsta.royalsocietypublishing.org
Geosciences Department, University of Texas at Dallas, Richardson,
TX 75080, USA
RJS, 0000-0002-8083-4632
Research To understand how plate tectonics became Earth’s
Cite this article: Stern RJ. 2018 The evolution dominant mode of convection, we need to address
three related problems. (i) What was Earth’s tectonic
of plate tectonics. Phil. Trans. R. Soc. A 376:
regime before the present episode of plate tectonics
20170406. began? (ii) Given the preceding tectonic regime, how
http://dx.doi.org/10.1098/rsta.2017.0406 did plate tectonics become established? (iii) When
did the present episode of plate tectonics begin? The
Accepted: 23 June 2018 tripartite nature of the problem complicates solving
it, but, when we have all three answers, the requisite
consilience will provide greater confidence than if we
One contribution of 14 to a discussion meeting only focus on the long-standing question of when
issue ‘Earth dynamics and the development of did plate tectonics begin? Earth probably experienced
plate tectonics’. episodes of magma ocean, heat-pipe, and increasingly
sluggish single lid magmatotectonism. In this effort
Subject Areas: we should consider all possible scenarios and lines
plate tectonics, Solar System, of evidence. As we address these questions, we
should acknowledge there were probably multiple
extrasolar planets
episodes of plate tectonic and non-plate tectonic
convective styles on Earth. Non-plate tectonic styles
Keywords: were probably dominated by ‘single lid tectonics’
plate tectonics, lithosphere, rocky planets, and this evolved as Earth cooled and its lithosphere
single lid thickened. Evidence from the rock record indicates
that the modern episode of plate tectonics began
in Neoproterozoic time. A Neoproterozoic transition
Author for correspondence:
from single lid to plate tectonics also explains
Robert J. Stern kimberlite ages, the Neoproterozoic climate crisis and
e-mail: rjstern@utdallas.edu the Neoproterozoic acceleration of evolution.
This article is part of a discussion meeting issue
‘Earth dynamics and the development of plate
tectonics’.

1. Introduction
Plate tectonics is the central unifying theory for
geology and geophysics. The original definition of plate
tectonics [1] has recently been modified to include
a description of the driving force as ‘A theory of
global tectonics powered by subduction in which the
lithosphere is divided into a mosaic of plates, which
move on and sink into weaker ductile asthenosphere.

2018 The Author(s) Published by the Royal Society. All rights reserved.
 Downloaded from http://rsta.royalsocietypublishing.org/ on October 2, 2018

Three types of localized plate boundaries form the interconnected global network: new oceanic
2
plate material is created by seafloor spreading at mid-ocean ridges, old oceanic lithosphere sinks
at subduction zones, and two plates slide past each other along transform faults. The negative

rsta.royalsocietypublishing.org Phil. Trans. R. Soc. A 376: 20170406

........................................................
buoyancy of old dense oceanic lithosphere, which sinks in subduction zones, mostly powers plate
movements’ [2]. Plate tectonics is a special planetary convection style that is characterized by the
largely independent motion of lithospheric fragments, with new lithosphere formed at mid-ocean
ridges and consumed in subduction zones. This is what is meant by ‘the modern episode of plate
tectonics’ in the rest of the text.
We want to understand how plate tectonics became the defining convective style of our planet:
the evolution of plate tectonics. To do this, we need to address three related problems. (i) What
was Earth’s tectonic style before the modern episode of plate tectonics began? (ii) How did plate
tectonic begin? (iii) When did the modern episode of plate tectonics begin? The three-headed
nature of the problem complicates arriving at a solution but, when we have all three answers,
the requisite consilience will provide confidence that useful answers have been arrived at. In this
effort we should keep an open mind and consider all possible scenarios and lines of evidence.
For example, we should consider the possibility that there have been multiple episodes of plate
tectonics in Earth history and that the transition from one tectonic/convective mode to another
could have taken tens to hundreds of millions of years to accomplish.
This paper critically examines these three related topics. Although the structure of this paper
as outlined above is robust, many of the conclusions are likely to be controversial because I
interpret the physical constraints and the geological evidence to indicate that plate tectonics has
only operated for the last quarter or so of Earth history, beginning in Neoproterozoic time. In
addition, this essay also explores ‘other things explained’. Because the transition to plate tectonics
must have affected all Earth systems, including the biosphere and hydrosphere, answering the
three central questions should also provide additional, unexpected insights into Earth history.
Included among the ‘other things explained’ by this interpretation are the kimberlite record,
Neoproterozoic climate change and acceleration of biological evolution in Neoproterozoic time.
All these points are examined further below.

2. What was Earth’s tectonic style before plate tectonics?

We want to understand what was Earth’s tectonic and convective style before the modern
episode of plate tectonics was established. We can gain a useful perspective on the range of
Earth’s possible tectonic and magmatic styles by considering other active large (greater than
1000 km diameter), rocky (silicate) bodies in our Solar System. In 2015, mankind completed a first
examination of the 30 largest bodies orbiting the Sun, the four largest of which (Jupiter, Saturn,
Uranus and Neptune) are gassy or icy giants not considered here. The other 26 largest solid bodies
in the Solar System can be subdivided on a basis of density into eight rocky (ρ > 3000 kg m−3 ,
inferred to be mostly silicates) and 18 icy (ρ < 2200 kg m−3 ) bodies [3]. For the purposes of
this discussion we ignore the icy bodies because their very different rheologies and densities
complicates comparisons with Earth.
Next we should identify silicate bodies that are tectonically active and focus on these for
comparison with Earth.
Before doing this it is useful to consider the roles of lithosphere and asthenosphere. The
concept of lithosphere has evolved considerably since it was introduced by Barrell [4] to describe
the strong outer shell or lid of the Earth. Earth’s lithosphere has three useful definitions: thermal,
chemical and rheological/mechanical [5]. Thermal lithosphere has a conductive geothermal
gradient; chemical lithosphere has compositional and isotopic characteristics that reflect
prolonged isolation from well-mixed asthenosphere; and rheological (mechanical) lithosphere is
defined by strength. So defined, these different lithospheres have different thicknesses; generally
thermal and chemical lithosphere are subequal and much thicker than rheological lithosphere. For
tectonically active planets, lithosphere is underlain by asthenosphere, which is hotter and weak
enough to flow. Earth’s asthenosphere lies a few hundred kilometres below the surface but we
 Downloaded from http://rsta.royalsocietypublishing.org/ on October 2, 2018

the five active large silicate (density > 3000 kg m–3) bodies in the Solar System 3
more active less active

rsta.royalsocietypublishing.org Phil. Trans. R. Soc. A 376: 20170406

........................................................
to scale

BODY: Io Venus Earth Mars Europa

DIAMETER (km): 3600 12 100 12 700 4220 3100

DENSITY (kg m–3): 3500 5200 5500 3900 3000

TECTONIC single lid single lid plate tectonics single lid ice-covered
STYLE: (heat pipe) (vigorous) (sluggish) silicate tectonics
unknown

Figure 1. Images of the solid surfaces of the five large (greater than 1000 km diameter), active (TAI = 2 or greater; [3]), silicate
(density = 3000 kg km−3 or greater) bodies. Three of the five (Io, Venus and Mars) have some variety of single lid tectonics. We
do not know what is the tectonic style of the silicate interior of Europa because the planet is covered in ice. Only Earth has plate
tectonics.

Table 1. Active silicate planetary bodies (planets in italics).

diameter outer
body (km) mass (kg) density (kg m−3 ) TAIa lithosphere tectonic style plumes?
Venus 12 100 4.90 × 10 24
5.28 × 10 3
3 silicate single lid yes
..........................................................................................................................................................................................................

Earth 12 700 6.00 × 10 24

5.59 × 10 3
3 silicate fragmented lid yes
..........................................................................................................................................................................................................

Mars 6779 6.40 × 1023 3.92 × 103 2 silicate single lid yes
..........................................................................................................................................................................................................

Io 3600 8.90 × 1022 3.64 × 103 2 silicate single lid yes

..........................................................................................................................................................................................................

Europa 3100 4.80 × 10 22

3.08 × 10 3
2 ice fragmented (ice) lid yes
..........................................................................................................................................................................................................
a Tectonic activity index.

have few constraints on lithospheric thicknesses on other active silicate bodies. On Earth, mantle
convection via plate tectonics is reflected in deformation (folding and faulting of lithosphere) and
volcanism (asthenospheric melts that reach the surface) and similar features are expected on other
active silicate bodies: Venus, Mars and Io. Tectonically dead silicate bodies are easily recognized
by surfaces that are pockmarked by impacts, like those of Mercury and Earth’s moon.
We can use the above considerations to assess whether a silicate body is active or not.
The tectonic activity index (TAI) uses observed surface features to yield a three-point score
summarizing a body’s volcanism, faulting and density of bolide impacts. Combining density
information and TAI allows us to identify the large active silicate bodies [3]. Table 1 summarizes
the five active silicate bodies in our Solar System: three planets (Venus, Earth and Mars) and two
satellites of Jupiter (Io and Europa) (figure 1). Europa is classified as an active body because its icy
surface is faulted and has few impacts but because it is covered in ice we cannot assess whether
its silicate interior is active or not. For this reason it is omitted from further consideration and
focus is on the four active silicate bodies with observable surfaces (Venus, Earth, Mars and Io).
Among the four active silicate bodies listed in table 1, only Earth has plate tectonics. The other
three have a tectonic style where the lithosphere is not broken into largely independent mobile
plates but is a single lithospheric ‘lid’ that completely encloses the asthenosphere and the interior
of the body. This idea was originally advanced by Solomatov & Moresi [6] for Venus, who argued
 Downloaded from http://rsta.royalsocietypublishing.org/ on October 2, 2018

that convection in its interior was dominated by upwelling plumes similar to those observed
4
on Earth, and by cold ‘drips’ of lithosphere that sink into the asthenosphere. Solomatov &
Moresi [6] coined the term ‘stagnant lid’ and used it to characterize the present Venusian tectonic

rsta.royalsocietypublishing.org Phil. Trans. R. Soc. A 376: 20170406

........................................................
regime. Venusian lithosphere is deformed along lithospheric weak zones and is accompanied by
widespread magmatism (Venus-style squishy lid [7]). Similarly, Mars has a stagnant lid tectonic
regime, albeit more sluggish than the Venusian tectonomagmatic regime (e.g. [8]).
The idea that large silicate bodies tend to have an all-encompassing lithospheric lid finds
increasing acceptance by planetary scientists. Our understanding of this non-plate tectonic regime
is evolving rapidly and we are now exploring what are the possible styles of ‘lid tectonics’
for large silicate bodies. We recognize three active examples (Venus, Mars and Io) but we have
only glimpsed their surfaces. Names for the wide range of single lid tectonic behaviours are
proliferating rapidly. For example, O’Neill & Roberts [9] refer to ‘stagnant, sluggish, plutonic-
squishy, or heat pipe’ variants; in addition there is plume-lid tectonics [10]. ‘Sagduction’—the
vertical sinking of weak lithosphere—is another vigorous non-plate tectonic style [11]. Here I use
the term ‘single lid’ to describe all types of tectonic and magmatic activity in a convecting and
magmatically active silicate body where the lithosphere is not fragmented into plates. It must be
stressed that we know very little about the range of single lid behaviours [12], but the proliferation
of new terms for these indicates that we have begun the serious work needed to achieve this
understanding.
Given that Earth may have experienced one or more episodes of single lid behaviour, how
can we identify these in the rock record? We cannot do that with any confidence yet, but we
should keep in mind that some transitions that have been identified in the rock record—for
example, different trace element and isotopic ratios preserved in sediments and igneous rocks—
might mark changes in single lid behaviour and not necessarily the beginning of plate tectonics.
Part of the reason for this uncertainty is that we have an incomplete understanding of the
range of single lid behaviours, largely due to the fact that we cannot yet study in detail any
of the three single lid examples in our Solar System. (The Mars InSight mission (scheduled to
land on Mars in November 2018) promises to make the first measurements of an active single
lid silicate body interior.) We expect that mantle plume activity will be an important mode of
convective upwelling in active single lid silicate bodies and that loss of lithosphere via Rayleigh–
Taylor instabilities or ‘drips’ will be important modes of convective downwelling. Interactions
between plume-derived magmas and pre-existing continental crust may yield incompatible
trace element abundances that mimic those of convergent margin magmas (e.g. [13]). Styles
of plume upwellings and drip downwellings will change as the body cools and lithosphere
thickens. There is petrological evidence that Earth’s potential temperature—approximating the
temperature of the asthenosphere—has decreased by approximately 150–300°C over the past
2.5 Ga [14,15], and cooler asthenosphere implies thicker mantle lithosphere and thinner oceanic
(basaltic) crust. The expected changes in the thickness of oceanic crust and mantle lithosphere
over the life of a cooling planet should be manifested in lithospheric behaviour, not only in
terms of whether plate tectonics or some other tectonic style exists but also in terms of changing
styles of single lid behaviour. Thin mantle lithosphere at the beginning of a silicate body’s
evolution is likely to be associated with different types of upwellings and downwellings than
those associated with thicker mantle lithosphere later in the history of the same silicate body. A
cartoon of this based on the behaviour of hyperactive Io, active Venus and sluggish Mars is shown
in figure 2.
The planetary perspective indicates that plate tectonics is unusual. Given that single lid
tectonics is the most common convective mode of large, active silicate bodies in the Solar System,
plate tectonics may also be unusual in Earth history. A tectonic style that is unusual in space may
also be unusual in time. Earth-centric uniformitarianism suggests that plate tectonics has always
operated, but planetary uniformitarianism yields a very different conclusion. Based on what we
have learned from exploring the Solar System, it is very likely that Earth experienced one or more
episodes of single lid behaviour before modern plate tectonics began.
 Downloaded from http://rsta.royalsocietypublishing.org/ on October 2, 2018

lithosphere thickness and strength

5
none thin, weak ~1000 km

rsta.royalsocietypublishing.org Phil. Trans. R. Soc. A 376: 20170406

........................................................
time
early, hot late, cold
stagnant lid
early intermediate late
delamination terminal
magma ocean heat pipe drips and plumes and upwellings stagnant lid

crust
lithosphere
magma partial melt
drips Rayleigh–Taylor
drips asthenosphere delamination
plumes upwelling and
solid mantle melting

core
Io Venus Mars Moon, Mercury

crust
possible stages
lithosphere
in the tectonic
plate
evolution of a tectonics asthenosphere
silicate planet
subduction

Figure 2. Possible evolution of tectonic styles for large silicate bodies like the Earth. Plate tectonics requires certain conditions
of lithospheric density and strength to evolve and is likely to be presaged and followed by single lid tectonics. (Online version
in colour.)

3. How did plate tectonics begin?

In order to understand how plate tectonics began, we need to understand how plates form
and move today. By understanding the modern balance of driving and resisting forces, we can
think more clearly about when this first existed such that a single lid could break to form
plates and set these in motion; plate tectonics must have begun after that time. Geodynamic
considerations compel the conclusion that plate motions today are mostly due to the sinking
of oceanic lithosphere in subduction zones [16,17]. This is because modern oceanic lithosphere
more than 30 Myr old is slightly (approx. 1%) denser than underlying asthenosphere [18] so
old oceanic lithosphere will sink beneath lithosphere if possible. Such a density inversion was
less likely in early Earth history because higher mantle potential temperatures were likely to
be reflected in thinner mantle lithosphere (dense) and thicker mafic crust (buoyant). Given
the existence of such a density inversion, the great strength of oceanic lithosphere is the most
important impediment to starting and maintaining subduction zones (a global network of these
and associated spreading ridges and transform faults is required for plate tectonics, not just one
subduction zone). Today, oceanic lithosphere is generally strong but local zones of weakness exist
that are significant enough to allow new subduction zones to form. A key observation is that
 Downloaded from http://rsta.royalsocietypublishing.org/ on October 2, 2018

of Gurnis et al. [19], who noted that about a third of all active subduction zones formed in the
6
last 65 Ma, implying that subduction initiation must be a process that now occurs ‘easily and
frequently’. This process nevertheless requires a sufficiently long (>∼1000 km) lithospheric weak

rsta.royalsocietypublishing.org Phil. Trans. R. Soc. A 376: 20170406

........................................................
zone in order to nucleate a new subduction zone [20]. The great strength of undamaged oceanic
lithosphere is demonstrated by the observation that no new subduction zones have nucleated
along passive continental margins in the past 65 Ma (and longer) and that it is energetically more
favourable for India to continue converging with Asia than for a new subduction zone to form
to the south in the Indian Ocean [2]. Lithospheric strength at continent–ocean boundaries may
have been lower in the past, in association with higher mantle potential temperatures and thinner
lithosphere, as suggested by geodynamic modelling (e.g. [21]). A key point is that plate tectonics
itself creates long weak zones in the oceanic lithosphere in the form of transform faults and
their inactive equivalents, fracture zones, but such plate tectonic-induced lithospheric weaknesses
would not have existed before plate tectonics began.
With this caveat in mind, the general sequence of events leading to the formation of a new
subduction zone today is reasonably well understood. Sinking of old, dense oceanic lithosphere
adjacent to a long (approx. 1000 km or more) transform fault and/or fracture zone is thought
to have been where the present Izu–Bonin–Mariana (IBM) subduction zone nucleated. The IBM
subduction zone formed when the western Pacific plate—which encompasses some of the oldest
oceanic lithosphere on Earth [22]—began sinking approximately 52 Ma [23]. Initial subsidence
of old oceanic lithosphere eventually began down-dip motion (true subduction) to torque the
Pacific plate to move WNW and form the bend in the Emperor–Hawaii seamount/island chain
approximately 47 Ma [24].
Long lithospheric weaknesses like transform faults and fracture zones are unlikely to have
existed before plate tectonics began and are unlikely to have been exploited to form the first
subduction zone. Other types of lithospheric weaknesses must have existed in order to form
the first subduction zone and start plate tectonics. What processes might produce lithospheric
weaknesses of sufficient length to start a new subduction zone?
Impacts of large bolides could rupture the lithosphere sufficiently to start a new subduction
zone. Numerical modelling indicates that bolides with diameters greater than 500 km could
damage the lithosphere sufficiently to start subduction [25]. Such large impacts are not likely after
the first few hundred million years of Earth history and are thus unlikely to have been responsible
for starting plate tectonics in Archaean or younger times, which is when most workers think is
when plate tectonics started [26].
Another idea about how to develop lithospheric weaknesses in a single lid tectonic regime
was offered by Bercovici & Ricard [27]. They suggested that lithosphere could be sufficiently
damaged by a variety of non-plate tectonic processes such as mantle flow, transient subduction
and dripping. Such processes promote shear localization leading to the establishment of long-
lived weak zones that eventually—as Earth cools and oceanic lithosphere increases in thickness
and density—collapse to form true tectonic plates driven by subduction alone.
It is also likely that the sufficiently large, hot and long-lived heads of starting mantle plums
could rupture the lithosphere over a large enough region (greater than 1000 km diameter) to cause
lithospheric collapse around the margins to create the first subduction zone needed to start plate
tectonics [28]. Mantle plumes are ubiquitous features of active silicate bodies [3] and occurred on
Earth all through geological time [29].
Water plays an essential role in localizing the lithospheric weaknesses that are needed to
initiate and maintain subduction [30]. Water facilitates plate bending by penetrating down
along bending-induced normal faults into the upper mantle, where it transforms strong, olivine-
dominated peridotite into weak serpentinite [31], making it much easier to bend the plate. Water
continues to be important for the sustainability of subduction because a weak zone composed of
wet, weak sediments and serpentinites quickly forms on top of the sinking lithospheric slab. This
weakness lubricates the plate interface and allows the plate to mostly slide down-dip (although
approx. 10% of plate motion is vertical, resulting in trench rollback; [32]). Weakness above the
 Downloaded from http://rsta.royalsocietypublishing.org/ on October 2, 2018

subduction interface is enhanced due to the slab-released fluids and off-scraped sediments,
7
allowing the subducted lithospheric slab to develop and maintain the asymmetric subduction
zone configuration, which is required to pull the rest of the downgoing plate and drive plate

rsta.royalsocietypublishing.org Phil. Trans. R. Soc. A 376: 20170406

........................................................
motions [33]. Water is almost certainly necessary for plate tectonics to occur but the presence of
water alone is not sufficient.
A final key point is that making the first subduction zone is necessary but not sufficient for
starting plate tectonics. There must be large enough tracts of strong, gravitationally unstable
lithosphere that can be induced to rupture, sink and move, and in this way evolve a global plate
mosaic. Even when lithospheric conditions become suitable for plate tectonics, it is likely to take
some time to accomplish the transition from single lid to global mosaic, a point that is explored
below.

4. When did plate tectonics start?

In order to address this question, we must keep in mind several things. (i) The possibility that
multiple episodes of stagnant lid and plate tectonics may have transpired in Earth history. (ii)
When did the force balance needed to sustain plate tectonics first exist? (iii) Does continental crust
formation require plate tectonics? (iv) What is the geological evidence and how do we interpret
it? (v) Other phenomena explained by a Neoproterozoic start of plate tectonics. We consider these
five topics separately below.

(a) The possibilities of multiple tectonic episodes

Discussing when plate tectonics began is often unnecessarily forced into a dichotomy: that today
there is plate tectonics and at some time in the past this ‘switched on’ from some unspecified
non-plate tectonic mode. The assumption that Earth has only experienced two tectonic modes
(plate tectonics and pre-plate tectonics) impedes the full range of thinking about the problem
that is required to solve it. Reducing the pre-plate tectonic mode to a single behavioural regime
is inconsistent with what we know about other active silicate bodies in the Solar System, as
discussed above, where three distinct styles of single lid behaviour are readily recognized: ‘heat-
pipe’ single lid tectonics on Io, vigorous single lid tectonics on Venus and sluggish single lid
tectonics on Mars (figures 1 and 2). Framing the question as a temporal dichotomy of pre-
plate tectonics and plate tectonics can also lead to misidentifying significant changes in Earth
products as signalling the start of plate tectonics, when the change could have been some
transition in single lid behaviour. The danger of such misidentification is seen in discoveries of
significant change in some Earth product at a given time and this change is interpreted to be
when plate tectonics began. Examples of this ‘assumed duality’ approach include the appearance
of eclogitic inclusions in diamonds at approximately 3.0–3.2 Ga [34], the change to more felsic
continental crust as the ‘great thermal divergence’ inferred from chemical compositions of basalts
approximately 2.5 Ga [15] and changes in Ni/Co and Cr/Zn ratios in Archaean terrigenous
sedimentary rocks and Archaean igneous/metaigneous rocks interpreted as indicating the start
of modern plate tectonics [35]. The changes that these studies recognize may just as well reflect a
significant transition from one style of single lid behaviour to another, not the beginning of plate
tectonics. Such logic traps are easily avoidable if the likelihood of multiple styles of single lid and
multiple transitions from single lid to plate tectonics in Earth history is acknowledged.
There is increasing evidence that the Earth experienced multiple tectonic episodes. There
is general acknowledgement that very early Earth experienced a magma ocean stage [36] and
this cooled to form a protocrust. Earth’s magma ocean may have evolved into a heat-pipe
tectonomagmatic regime [37] whereby mafic and ultramafic lava flows without a mantle root
were progressively erupted, buried by younger flows, and ultimately remelted to erupt again,
as inferred for modern Io (figures 1 and 2). Heat-pipe or other post-magma ocean stages may
have evolved into a single lid tectonic style, with growth of mantle lithosphere beneath the
Hadean protocrust, and lithospheric thickening is likely to have continued into Archaean time.
 Downloaded from http://rsta.royalsocietypublishing.org/ on October 2, 2018

Geoscientists are starting to identify Earth’s single lid tectonic episodes: one is suggested for 2.8–
8
2.9 Ga [12] and between 2.45 and 2.2 Ga [38]. The ‘boring billion’ between 1.8 and 0.8 Ga may have
been another single lid episode.

rsta.royalsocietypublishing.org Phil. Trans. R. Soc. A 376: 20170406

........................................................
The boring billion was characterized by remarkable environmental, evolutionary and
lithospheric stability [39], with the formation of the Rodinia supercontinent near the end of this
interval. Cawood & Hawkesworth [40] list seven characteristics of the boring billion: paucity
of passive margins; absence of glacial deposits and iron formations; lack of significant seawater
Sr isotope spikes; lack of phosphate deposits; high ocean salinity; abundant anorthosites and
alkali granites; and limited orogenic gold deposits. Sr and C isotopic compositions of seawater-
proxy carbonates change little during this time. The boring billion was also a long time when the
complexity of life increased very slowly. These characteristics are very like that expected from a
sluggish single lid episode.

(b) When did the force balance needed to sustain plate tectonics first exist?
Plate tectonics could not have begun on Earth until three conditions were satisfied: (i) large
tracts of lithosphere became generally denser than underlying asthenosphere; (ii) large tracts of
lithosphere became generally strong enough to remain intact in subduction zones and pull the
attached surface plate; and (iii) lithosphere developed weak zones that were profound enough to
rupture and become new plate interfaces. When in Earth history were these three conditions first
satisfied? Let us consider lithospheric density first. The density of oceanic lithosphere (thermal
lithosphere of [5]) is the sum of crustal and mantle contributions. Unless there is significant
intraplate igneous activity, the crustal thickness is essentially established at the spreading ridge
but the mantle contribution increases with age as the plate cools and mantle lithosphere thickens
[18]. It is unlikely that oceanic lithosphere has always been as dense and as strong as it is
today; specifically, it is likely to have been weaker and more buoyant in the past when the
mantle was hotter. As noted in the previous section, the mantle has been cooling over Earth
history. If we accept the inference that Earth’s potential temperature was approximately 150–
200°C higher 2.5 Ga ago than it is today [14,15], then the mantle lithosphere is likely to have
been thinner than it is today [41]. Descent styles of thinner, weaker oceanic lithosphere have
been explored quantitatively as a function of mantle potential temperature. Eclogitic dripping
(Rayleigh–Taylor instabilities) is likely only at mantle potential temperatures (Tp ) greater than
1600°C under a single lid tectonic regime [42]. Geodynamic modelling for Tp = 1425–1600°C
indicates that lithospheric mantle will separate from the crust and sink, a process that Chowdhury
et al. [43] call ‘peeling off’. Once separated from buoyant upper crust, lithospheric peels will break
off and sink vertically into the mantle.
In addition to controlling lithospheric behaviour, higher asthenospheric potential temperatures
would have led to more melting and generation of thicker oceanic crust with a given amount of
decompression [44]. For the modern Earth, most of the strength of old oceanic lithosphere resides
in the mantle, especially for old (greater than 30 Ma) seafloor [45]. Thinner mantle lithosphere and
thicker oceanic crust expected for the Archaean would have resulted in oceanic lithosphere that
was less likely to sink though underlying asthenosphere and hold together as it sank [42].
The last statement is increasingly supported by geodynamic modelling. There is not space
here to summarize all of the significant contributions on this topic but a general sense of this
rapidly advancing field is useful for understanding when plate tectonics was likely to have begun.
Numerical modellers are increasingly attracted to the problem of early Earth tectonics, especially
Archaean tectonics. Their largely independent and progressively more realistic experiments
increasingly agree that plate tectonics (as described in the Introduction) could not have occurred
given the likely strength and density of oceanic lithosphere of those times. This does not mean
that subduction did not happen in the Archaean, only that weak, unstable subduction zones
were unlikely to persist long enough to ‘infect’ the rest of the lithosphere and generate a global
plate mosaic. Lithospheric buoyancy considerations were emphasized by van Thienen et al. [46]
in their critique of Archaean plate tectonics. Moyen & van Hunen [47] combined geochemical
 Downloaded from http://rsta.royalsocietypublishing.org/ on October 2, 2018

data and geodynamic models to infer that a short-term episodic style of subduction was a
9
viable style of early Earth tectonics. Their modelling results show how the low strength of
slabs in a hotter Earth led to frequent slab break-off events that would have prevented modern-

rsta.royalsocietypublishing.org Phil. Trans. R. Soc. A 376: 20170406

........................................................
style, long-lived subduction systems, resulting instead in frequent cessation and re-initiation of
subduction on a time scale of a few million years. Thébaut & Rey [48] used realistic rheological
and thermal parameters to conduct coupled thermo-mechanical numerical experiments to show
that the dominant Archaean tectonic regime was a vigorous single lid style called ‘sagduction’
in response to density inversions and thermal weakening due to the emplacement of thick
continental flood basalts onto thin, hot protocontinental crust. Sizova et al. [21] conducted two-
dimensional petrological–thermomechanical tectonomagmatic numerical experiments developed
for conditions appropriate to the hotter early Archaean lithosphere and used the results to infer
delamination and dripping of the lower primitive basaltic crust into the mantle, local thickening
of the primitive basaltic crust and small-scale crustal overturn. They envisioned a vigorous single
lid tectonic regime that was interrupted by short-lived subduction-like episodes. These results
increasingly impel us to consider that early Earth experienced multiple episodes and styles of
single lid behaviour.

(c) Continental crust formation: is it a plate tectonic signal?

Some geoscientists assume that, because continental crust is generated by plate tectonics today
(mostly above subduction zones at convergent plate boundaries), the existence of continental
crust of any age implies that plate tectonic processes must have operated in order to form
such crust. This is a logical fallacy, because if this assumption was correct then the existence of
continental crust as old as 3.8 Ga must indicate that plate tectonics began at least that long ago.
In fact, most geoscientists agree that plate tectonics began significantly after 3.8 Ga [26], so the
implicit linkage between formation of continental crust and plate tectonics should be critically
considered.
There is no empirical or experimental evidence indicating that plate tectonics is required in
order to generate continental crust. Formation of thickened crust of more felsic composition than
expected for mantle melts (basalt) only requires hydrous magmatic conditions, including hydrous
mantle melting and/or remelting of hydrous mafic crust. The importance of water for generating
felsic melts has been known for decades. Campbell & Taylor [49] were among the first to argue
that involvement of water in magmagenesis is required to make felsic continental crust. Hydrous
melting of mantle tends to generate more Si-rich magmas (e.g. [50]), including high-Mg andesite
and boninite, and such compositions can fractionate to form proportionately much larger volumes
of felsic magma than can dry basaltic magmas. In addition, mafic rocks metamorphosed under
hydrous conditions are likely to form amphibolite in the middle crust. Silica-rich magmas are
expected from partial melting of amphibolite in the crust (P ∼ 1 GPa) at T ∼ 750–1000°C (e.g.
[51,52]), conditions that were common in Archaean time. Subduction is an excellent way to
introduce water into the mantle to trigger hydrous melting and to generate and melt amphibolites
today but it is not the only way. All that is needed is that there be abundant surface water and a
way to move this deep into the crust and upper mantle where it can participate in partial melting.
We know that surface water has been present on Earth since at least 4.3 Ga [53]. Amphibole
will form wherever water interacts with mafic igneous rocks at T ∼ 400–600°C; such conditions
are characteristic of lower oceanic crust (e.g. [54]) and may reflect the root zones of seafloor
hydrothermal vents, which were probably very abundant in Archaean times, as demonstrated
by the abundant volcanogenic massive sulfide deposits of that age. Once water is fixed in mid-
lower crustal amphibolite, this can be melted in situ to generate felsic magmas or can sink into
the mantle via single lid processes like delamination or sagduction, where the water will be
released and delivered to the surrounding mantle. A cartoon illustrating these processes is shown
in figure 2. The point to be emphasized is that fractionation of hydrous mantle melts and/or
partial melting of crust containing abundant amphibole is all that is needed to generate felsic
melts and continental crust, and these processes do not require plate tectonics [7,42].
 Downloaded from http://rsta.royalsocietypublishing.org/ on October 2, 2018

total (37 830) 10

900 detrital ancient sediments (21 849) 2700
orogenic granitoids (8928)

rsta.royalsocietypublishing.org Phil. Trans. R. Soc. A 376: 20170406

800 detrital modern sediments (7053; 2x)

........................................................
700 1870

600
500 300 600
400 1000
little felsic crust preserved
300
200
100

0 500 1000 1500 2000 2500 3000 3500 4000 4500

age (Ma) Condie & Aster [56]

Figure 3. The detrital zircon age spectra, for orogenic granitoids, sedimentary rocks and modern river sediments, modified after
Condie & Aster [56]. See text for further discussion.

Recent studies add detail to these observations. Moyen & Laurent [55] noted that Archaean
granitic rocks are dominated by tonalites, trondhjemites and granodiorites (the TTG suite) formed
by partial melting of alkali-rich metabasalt and amphibolite. Moyen & Laurent [55] acknowledged
the arc-like trace element signatures of the TTG suite but cautioned that these signatures were
controlled by the nature of the source and the conditions of melting, concluding that such controls
and conditions are not directly linked to one particular tectonic setting. They noted that Archaean
TTG suites were likely to be generated under a range of melting depths, from approximately 15
to 70 km deep (∼5 to >20 kbar). They emphasized that mafic crust could be partially melted to
generate TTGs in a range of settings that did not require plate tectonics.
A slightly different perspective is provided by Rozel et al. [7]. These workers built on the
understanding that hotter Archaean mantle temperatures resulted in two distinct styles of single
lid behaviour: a heat-pipe regime dominated by volcanism (Io-like single lid) and the ‘plutonic
squishy lid’ tectonomagmatic regime dominated by intrusive magmatism (Venus-like single
lid). Rozel et al. [7] carried out numerical modelling of thermochemical convection to show
that both regimes would be capable of producing TTG suites and thus continental crust. These
numerical models showed that the volcanism-dominated heat-pipe tectonic regime was not able
to produce felsic continental crust whereas the plutonic squishy lid tectonic regime dominated
by intrusive magmatism was associated with hotter crustal geotherms and generated a variety
of TTG magmas. Rozel et al. [7] concluded further that the pre-plate tectonic Archaean Earth
operated in the plutonic squishy lid regime rather than in an Io-like heat-pipe regime, which may
have been more important in the Hadean.
Geological evidence from the detrital zircon record (figure 3) strongly suggests that Eo-
Archaean and Hadean crust was largely mafic [57]. Chowdhury et al. [43] addressed the dearth
of such old crust using thermomechanical modelling. They inferred that such remnants were
destroyed when continental crust began to stabilize in Late Archaean and Early Palaeoproterozoic
time (3–2 Ga). Their numerical models indicate that non-plate tectonic recycling via lower
crustal ‘peeling-off’ (delamination) was common during this interval and further inferred that
destruction of the early mafic crust may have resulted in felsic magmatism that accelerated
continental crust production by single lid magmatic processes.
One of the most important datasets that we have to constrain this discussion is the spectrum
of zircon U-Pb ages. Zircon is generated when silicate magmas cool and is most abundantly
 Downloaded from http://rsta.royalsocietypublishing.org/ on October 2, 2018

produced by slow cooling of felsic magmas, such as TTGs or granite. Zircon age peaks are
11
commonly interpreted either as crustal production peaks or as peaks of subduction-produced
crust selectively preserved during continent–continent collision. The zircon age spectrum mostly

rsta.royalsocietypublishing.org Phil. Trans. R. Soc. A 376: 20170406

........................................................
reflects the proportion of zircon produced of a certain age, which reflects the amount of plutonic
felsic rocks produced per unit time. There may have been quantitative loss of Hadean and Early
Archaean zircons as this crust was mostly destroyed, but, once continental crust began to be
preserved later in the Archaean, zircons of this age seems to have been largely preserved [58].
The record from about 3.0 Ga probably reflects crustal production, for the most part. Some felsic
magmas produce more zircon per unit of magma than others (zircon fertility factor (ZFF) of [59])
but all felsic plutonic igneous rocks crystallize zircon and yield considerable zircon when they are
eroded. The zircon age spectra for igneous rocks, sedimentary rocks of various ages and modern
river sediments show similar trends, leading to confidence that these records contain useful
information about the rate of felsic magma production and thus continental crust through time.
The zircon age spectra show two dominant peaks at 2700 and 1870 Ma (figure 3). The Late
Archaean peak in the compilation of Condie & Aster [56] is somewhat larger than the mid-
Palaeoproterozoic peak but they are broadly similar. The next largest peak approximately 1000 Ma
is far smaller; this peak may be larger than representative because this age granitoids have very
large ZFF ∼3.5. After this there are minor peaks at 600 and 300 Ma.
It is very difficult to explain the large 1870 and 2700 Ma peaks as having formed by plate
tectonic processes, these peaks are far too large by comparison with zircon/crustal production
rates of slightly older and younger ages. Plate tectonics creates and destroys crust at a relatively
constant rate, with some variation over the course of the supercontinent cycle [60]. In the
compilation of Condie & Aster [56], the approximately 600 Ma Greater Gondwanaland peak is
substantial but the approximately 300 Ma Pangea peak is not especially large. The 1870 and
2700 Ma peaks represent tremendous increases over zircon/crustal production just before and
just after the peaks. These huge differences are most consistent with non-plate tectonic causes, for
example mantle overturn and planetary resurfacing during variants of very vigorous single lid
behaviour.

(d) What is the geological evidence and how do we interpret it?

Here we briefly consider what the geological record says about when the modern episode of
plate tectonics began, and we should recognize that some kinds of evidence are more compelling
than others. Large-scale features like ophiolites and high-pressure metamorphic rocks should be
emphasized over small-scale lines of evidence like trace element ratios in sediments and igneous
rocks or inclusions in zircons and diamonds because changes in the latter may reflect changes
in single lid behaviour, as outlined above, whereas the former are diagnostic of plate tectonic
processes. Figure 4 summarizes the temporal distribution of lithologies formed by plate tectonic
processes in Cenozoic time and that certainly formed by plate tectonic processes in the past. The
most diagnostic rock assemblages are ophiolites (figure 4a), blueschist and related subduction
zone metamorphic assemblages (figure 4b), and ultra-high-pressure (UHP) metamorphic terranes
and the collision gemstone ruby (figure 4c). These diagnostic lithologies are concentrated in
Neoproterozoic and younger times, although a few 2.1–1.8 Ga ophiolites are also documented.
More data have been added to the compilation, and these new data continue to validate the
original conclusion of Stern [63] based on these three diagnostic assemblages that the modern
episode of plate tectonics began in Neoproterozoic time. Just as important is the absence of these
lithologies from ∼1.0 to ∼1.8 Ga.
It must be acknowledged that the geological record is incomplete and becomes more
incomplete with age. Crustal recycling and erosion remove evidence, but there are many well-
preserved outcrops of Archaean and Proterozoic rocks. The best way to overcome preservation
bias is to consider multiple lines of evidence that are affected differently by erosion. Ophiolites
commonly occur as the uppermost units of nappe stacks in orogenic belts and are especially
vulnerable to erosion; in contrast, metamorphic rocks like blueschists and UHP terranes are
 Downloaded from http://rsta.royalsocietypublishing.org/ on October 2, 2018

(a) plate tectonic indicators

12
seafloor ophiolites 20
spreading Archaean Palaeoproterozoic Mesoproterozoic Neo- Phanerozoic

rsta.royalsocietypublishing.org Phil. Trans. R. Soc. A 376: 20170406

and subduction

........................................................
proterozoic 10
initiation
indicator

no. occurrences
blueschists (sensu lato) and 15
(b) glaucophane-bearing eclogites 10
subduction
indicators lawsonite-bearing 15
metamorphic rocks 10
jadeitite (subduction gemstone) 3
(c) UHP metamorphic rocks 3
collision ruby and sapphire (collision gemstones) 10
indicators 5

3000 2700 2400 2100 1800 1500 1200 900 600 300 0
(Ma)

Figure 4. Comparisons of key petrotectonic indicators for plate tectonics and timings of major glacial episodes (vertical blue
regions) spanning the last 3 Ga, modified from Stern & Miller [61]. (a) Ophiolites provide evidence of seafloor spreading,
subduction initiation and horizontal motions consistent with plate tectonics (seafloor spreading and subduction initiation
proxy). (b) Indicators of subduction zone metamorphism—blueschists, glaucophane-bearing eclogites, lawsonite-bearing
metamorphic rocks and jadeitites—form only in the cool, fluid-rich environments in and above subduction zones (subduction
proxies). (c) UHP metamorphic rocks and the gemstone ruby proxy continental collision and deep subduction of continental
crust (continental collision proxies). Modified after Stern et al. [62]. (Online version in colour.)

uncovered by erosion and are difficult to obliterate. The fact that the multiple lines of evidence
shown in figure 4 agree indicates that preservation bias is not a serious concern for the Archaean
and Proterozoic record.
Palaeomagnetic evidence is potentially a key constraint but becomes less reliable with
age [64]. Critical evaluation of this line of evidence for the purposes of understanding when
palaeomagnetic constraints are robust and when they are not, and, on this basis, what the
palaeomagnetic evidence says about when cratons moved relative to each other (=plate tectonic
episodes) and when they did not (single lid), will require dedicated syntheses by members of the
practising palaeomagnetic community.
The absence of ophiolites, blueschist and UHP terranes for approximately 800 Ma—
approximately the time of the ‘boring billion’ or Earth’s middle age [40]—provides further
support that this was a single lid interval, as shown in figure 5. It is also significant that
there are some 1.8–2.1 Ga ophiolites, suggesting that this may have been a short ‘proto-plate
tectonic episode’ (figure 5). Nevertheless, it is increasingly acknowledged that Earth went through
major tectonic changes in Neoproterozoic time (e.g. [65–67]). The appearance of low-T, high-P
metamorphism indicative of ‘cold subduction’ begins approximately 800 Ma [68]. The evidence
is overwhelming that this was when modern plate tectonics as defined at the start of this paper
began.

(e) Other phenomena explained by a Neoproterozoic start of plate tectonics

Three additional Earth history mysteries are explained by recognizing that the modern episode
of plate tectonics began in Neoproterozoic: the kimberlite record, Neoproterozoic snowball Earth,
and the acceleration of biological evolution. These revelations are discussed briefly below.
Kimberlite eruptions—the water- and carbon dioxide-charged explosions from deep in the
lithosphere that carry diamonds to the surface—are concentrated in Neoproterozoic and younger
time (figure 6a). This has long been recognized but until recently unexplained. Stern et al. [62]
argued that kimberlites became increasingly abundant from the Neoproterozoic onwards because
subduction began to deliver increasingly large volumes of water and CO2 deep into the mantle
 Downloaded from http://rsta.royalsocietypublishing.org/ on October 2, 2018

plt tect single lid vigorous single lid T (°C) 13

stage 5 stage 4 3C 3B stage 3A stage 2 HP 1
very active 1700

rsta.royalsocietypublishing.org Phil. Trans. R. Soc. A 376: 20170406

........................................................
‘boring billion’

PPT
active LHB
least active 1600

true
re 1500
subduction
er atu
temp
ntle
ma
1400

1300
0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5
age (Ga)

Figure 5. Modified version from Hawkesworth et al. [65] to reflect the author’s preferred interpretation of Earth’s tectonic
history. HP, heat-pipe tectonics; LHB, Late Heavy Bombardment; PPT, proto-plate tectonics. (Online version in colour.)

(a) kimberlite ages (b)

800
790 >1.0 Ga: single lid kimberlites are
very rare
oceanic crust continental crust

600 graphite carbonate-rich lithosphere lithosphere

diamond
diamonds asthenosphere
frequency

carbonate-rich mantle
425
400 (c)
<1.0 Ga:
kimberlite kimberlite plate teconics

200
102
hydrous
38
9 18 5 9 19 2 7 carbonate-rich fluids
0 derived from the
0 500 1000 1500 2000 2500 3000 subducted slab
accumulate at the base of
age range (Ma)
the lithosphere and
explosively erupt as kimberlites

Figure 6. Kimberlites and the evolution of plate tectonics. (a) Histogram of kimberlite ages, binned each 500 Ma. Note the
great increase in kimberlite in Neoproterozoic and younger time (less than 750 Ma). (b,c) Simple explanation why abundance of
kimberlites increased approximately 750 Ma. (b) Before approximately 1 Ga, no plate tectonics and no deep subduction. Flux of
water to mantle is low so fluid pressure at the top of the asthenosphere is low. (c) After approximately 1 Ga, plate tectonics and
deep subduction delivers more water deeper into the mantle. Upward-infiltrating water interacts with carbonated peridotite
mantle, generating abundant H2 O–CO2 fluids and increasing fluid pressure at the top of the asthenosphere. Eventually build-up
of fluid pressure breaks to the surface as kimberlite. Modified after Stern et al. [62]. (Online version in colour.)

then. Large quantities of water pumped deep into the mantle slowly percolated up from deeply
subducted slabs to destabilize mantle carbonates. The combined H2 O–CO2 fluid accumulated
at the base of the lithosphere, resulting in especially high volatile pressures at the base of the
cratonic lithosphere. Volatile accumulation continued until it was released by explosive eruptions
from this depth to form kimberlites. Kimberlites did occur before Neoproterozoic time but they
 Downloaded from http://rsta.royalsocietypublishing.org/ on October 2, 2018

(a) (b) (c) (d)

14
hydrothermal circulation
Sl

rsta.royalsocietypublishing.org Phil. Trans. R. Soc. A 376: 20170406

T2

........................................................
V V V V V V
V V V T2 T2 V
V
V V
a=

V
a
V a V V
V T2
a T1 T3
amphibolite a a T1 a T2
T3

E2 E3
a T1 a E2 M2
E2 a a E3 E2 M2 E3
M1 a M1 a E3
M2 M2
E1 E2
MANTLE E1
M2
M1 E2
M2 M3

E1
E2
T1: TTG melt MANTLE
DIAPIR DEPLETED
T2: 2nd generation TTG 1 M2
MANTLE
T3: 3rd generation TTG
E1: eclogite and pyroxenite

a
MANTLE
E2: 2nd gen. eclog. and pyrox. DIAPIR
2
E3: 3rd gen. eclog. and pyrox. E1 MANTLE
DIAPIR
M1: mantle melt MANTLE 3
M2: 2nd generation mantle melt DIAPIR
2

Figure 7. Cartoon modified after Bédard [69] showing how water might be delivered to the upper mantle and how felsic
magmas (TTG suites) might be generated by melting crustal amphibolite. (a) A large mantle plume releases melt (M1) that
constructs a thick volcanic crust, the lower part of which is metamorphosed into amphibolite. Underplating magma, which
causes melting at the base of the crust, forms a first generation of tonalitic melt (T1) with complementary eclogitic to pyroxenitic
restites (E1). (b) Smaller delaminated bodies mix into the shallow upper mantle and trigger the formation of a second generation
of mantle melt (M2). (c) The first generation crustal restites are largely destroyed as M2 melts are generated, collect and
ascend. New melt from a second mantle diapir also contributes to M2. M2 underplates the crust to form a second generation of
tonalite melt (T2) by melting amphibolites, yielding a second generation of restites and cumulates (E2). Older tonalites (T1) are
extensively remobilized at this time, and also contribute to T2. The voluminous T2 tonalites are buoyant and trigger a second
cycle of partial crustal convective overturn. (d) As M2 magmatism wanes, the underplated layer cools and crystallizes. The
restites and cumulates (E2) delaminate into the mantle, triggering the formation of a third generation of mantle melts (M3),
and destroying the second generation restites. Melting of underplated M2 melt and relict lavas generate a third generation of
tonalitic to granodioritic melt (T3), also yielding a third generation of restites and cumulates (E3). Older tonalitic rocks (T1 and
T2) are extensively remobilized and represent the dominant part of T3. The voluminous T3 tonalites/granodiorites are buoyant
and trigger a third cycle of partial convective overturn in the crust.

were much rarer than after Neoproterozoic time (figure 6a). Single lid tectonics could deliver small
volumes of surface fluids and volatiles to the upper mantle by sagduction and drips (figures 7 and
6b) but the massively greater delivery of fluids and volatiles deep into the mantle needed to cause
frequent kimberlite eruptions could not occur until plate tectonics and sustained subduction got
underway in Neoproterozoic time (figure 6c). An alternate interpretation is offered by Tappe
et al. [70].
Neoproterozoic snowball Earth is another mystery in Earth history that can be explained
by a Neoproterozoic onset of the modern episode of plate tectonics. Stern & Miller [61] noted
that the transition from Mesoproterozoic single lid tectonics to plate tectonics should have
disturbed the climate equilibrium established by the previous billion-year-long stasis of silicate
weathering–greenhouse gas feedbacks by re-distributing continents, increasing explosive arc
volcanism, creating relief and stimulating mantle plumes. Formation of subduction zones could
have redistributed mass sufficiently to caused true polar wander. These disruptions could have
caused spectacular carbon isotope variations along with several episodes of Neoproterozoic
snowball Earth. Stern & Miller [61] argued that the transition to plate tectonics could have caused
nearly all of the proposed geodynamic and oceanographic triggers for Neoproterozoic snowball
Earth events, and could also have contributed to biological triggers. Only extraterrestrial triggers
cannot be reconciled with the hypothesis that the Neoproterozoic climate crisis was caused by a
transition from single lid to plate tectonics.
 Downloaded from http://rsta.royalsocietypublishing.org/ on October 2, 2018

Marinoan
Huronian global glaciations Sturtian Gaskiers 15
seawater d13C 10

d13Ccarb

rsta.royalsocietypublishing.org Phil. Trans. R. Soc. A 376: 20170406

5

........................................................
0
Neo- B –5
Archaean Palaeoproterozoic proter- T –10
Mesoproterozoic ozoic I Tr Phanerozoic

3000 2700 2400 2100 1800 1500 1200 900 600 300 0
(Ma)

Figure 8. Timings of major global glacial episodes (vertical blue regions) spanning the last 3 Ga superimposed on a seawater
δ13 C curve. Neoproterozoic C-isotope excursions B, I, T, Tr and S denote Bitter Springs, Islay, Tayshir, Trezona and Shuram
anomalies, respectively. Note that the Neoproterozoic was a remarkable episode of change on Earth’s surface, consistent with a
transition from single lid tectonics to plate tectonics during Neoproterozoic time. Modified from [61]. (Online version in colour.)

An intriguing result of the suggestion that the transition from single lid to plate tectonics
caused the Neoproterozoic climate crisis is that the stratigraphic record provides information
for how long it took for the transition from single lid to formation of the first subduction zone
and a two-plate planet to development of a global plate mosaic to occur. If we assume that
individual ‘snowball’ episodes (Sturtian, Marinoan and Gaskier; vertical blue lines in figure 8)
and major C isotope excursions (Bitter Springs, Islay, Trezona, and Shuram; figure 8) reflect the
formation of new plates and subduction zones, then it took about 230 Ma for a global plate mosaic
to form. There may not be a simple one-to-one correspondence between these climate and isotopic
excursions on the one hand and formation of new plates and subduction zones on the other,
but the approximately 230 Ma interval of climate and isotopic instability may approximate how
long it took to accomplish the transition from the climate and isotopic equilibrium associated
with the Mesoproterozoic single lid tectonic regime to that of the modern plate tectonic
episode.
Finally, the start of plate tectonics in Neoproterozoic time provides new insights into the
question of why biological evolution accelerated at that time. Eukaryotic life experienced a
major diversification approximately 800 Ma [71]. The Neoproterozoic also witnessed the first
appearance of marine planktonic single cellular nitrogen-fixing cyanobacteria and non-nitrogen-
fixing picocyanobacteria [72]. The oldest known animal body fossils are found in approximately
571–566 Ma sediments. Trace fossil evidence of bilateral locomotion also occurs in ca 565 Ma
sediments. Simple skeletonized metazoans occur only in rocks deposited during the last 8–9
million years of the Ediacaran Period [73]. Metazoans began to take over ecosystems during the
Ediacaran (635 to 541 Ma) but the developmental toolkits were established in the Cryogenian
(720 to 635 Ma) [74]. Biologists are very interested to know what stimulated this burst of
diversification.
The link between Neoproterozoic tectonics and biological evolution is acknowledged [75],
but a link between acceleration of biological evolution rates and the start of plate tectonics
has not been considered heretofore. Such a link makes sense because creation of new habitats,
isolation and interspecies competition is difficult and rare for single lid tectonic regimes but is
easy and common with continents and plate tectonics (figure 9b–g). It is perfectly consistent with
interpreting a single lid regime for the boring billion (ca 1800 to 800 Ma) that biological evolution
proceeded so slowly during this interval. In contrast, new habitats are constantly being created
and destroyed by plate tectonics. New continental margins form during continental break-up,
increasing habitats and isolation, which spur diversification and speciation. These habitats are
destroyed and species occupying similar niches are forced to compete when continents collide.
By rapidly creating and destroying new habitats, plate tectonics expedites biological evolution
[76]. Single lid tectonic regimes cannot stimulate biological evolution, although it is difficult to
quantify the effects of these different evolutionary determinants. Nonetheless, a transition from
 Downloaded from http://rsta.royalsocietypublishing.org/ on October 2, 2018

(a)
single lid tectonics plate tectonics 16

Demospongiae

Echinodemata
Hemichordata

Onychophora
Brachiopoda
(b) (e)

Arthropoda
Vertebrata

Priapulida
t1

Nemertea

Mollusca

Annelida

Cnidaria
Bryozoa

rsta.royalsocietypublishing.org Phil. Trans. R. Soc. A 376: 20170406

........................................................
land
P1 P1
Late
H1
Ordovician

460
Middle H1
C1
Palaeozoic

480 Early
C1
500
Cambrian

stage 5
Terrenuvian series 2

520
stage 4
stage 3 ocean
stage 2

540
Nama assm.
Fortunian
(c) t2 cold (f)
White Sea assm. P2a P2b

~500 Ma
560 Avalon assm.
H2a H2b
time in millions of years (Ma)

580
Ediacaran

Gaskiers glaciation
C2a C2b
600 P2 H2 C2
620
P2c
640 Marinoan glaciation
warm H2c C2c
Neoproterozoic

660
(d) (g)
Cryogenian

680 t3
phyla
700 classes P3a C3b
stem groups
Sturtian glaciation
720 Ediacaran genera
c
740
crown group estimates P3 H3 C3 H3
known stratigraphic range-macrofossils
known stratigraphic range-biomarkers
Tonian

760 potential range extension

780

800
0 20 40 60 80 100 120 140 160 180 exogenic pressure (bolide) plume and LIP
taxa

Figure 9. Animal evolution in the context of a Neoproterozoic transition to plate tectonics from Mesoproterozoic single lid
tectonics. (a) Animal fossil record compared with the molecular divergence estimates for 13 different animal lineages. The known
fossil record of animals is shown at the class and phylum levels (hatching indicates ‘stem’ lineages, i.e. lineages that belong to a
specific phylum but not to any of its living classes); green shows the record of macroscopic Ediacaran fossils (see scale at bottom).
Shown in thick black lines are known fossil records of each of these 13 lineages through the Cryogenian–Ordovician; most
lineages first appear in Cambrian time, consistent with the animal fossil record. The extent of these stratigraphic ranges mirrors
molecular estimates for each crown group (coloured circles), highlighting the general accuracy of the molecular clock [74]. (b–g)
Cartoons showing how natural selection and evolution vary on a simplified Earth-like planet with subequal areas of continents
and oceans and three interdependent life forms (plant ‘P’, herbivore ‘H’ and carnivore ‘C’) over a supercontinent cycle at three
different times (1, 2 and 3) at approximately 100 million year intervals. Top of each panel corresponds to high latitudes (cool,
arctic), bottom is equatorial (warm). Exogenic evolutionary pressures exist regardless of whether or not plate tectonics occurs,
as does mantle plume activity including Large Igneous Provinces (LIPs). (b–d) Planet without plate tectonics, little change in
continental configuration, only climatic isolation, few barriers, and slow climate change. Evolutionary pressures are dominantly
biological and exogenic. (e–g) The situation for the same planet with plate tectonics over the course of a supercontinent cycle.
This provides many opportunities for isolation, diversification under different conditions of natural selection and evolution; this
is ‘rift pump’. Evolutionary rift pumping continues until continents collide, when different species co-mingle and compete and
new ecological systems are established; this is ‘collision pump’. Endogenic evolutionary pressures are more important and are
dominated by plate tectonics effects.

Mesoproterozoic single lid to Ediacaran plate tectonics helps explain the accelerating pace of
biological evolution beginning in Neoproterozoic time.

5. Conclusion
Below are 10 observations resulting from considerations explored in this paper:

1. Plate tectonics is a very unusual convective style for a silicate planet. All other active
silicate bodies are encased in a single lithospheric lid.
2. What is common in space may also be common over Earth history, and a significant part
of Earth history likely involved single lid tectonic regimes. Plate tectonic and single lid
tectonic regimes may have alternated over this history.
 Downloaded from http://rsta.royalsocietypublishing.org/ on October 2, 2018

3. Geodynamic studies conclude that it is the sinking of dense oceanic lithosphere in

17
subduction zones that drives plate motions. Because Earth’s upper mantle has cooled
150–200°C since Archaean time, dense oceanic lithosphere has thickened and buoyant

rsta.royalsocietypublishing.org Phil. Trans. R. Soc. A 376: 20170406

........................................................
oceanic crust has thinned, making the lithosphere denser—and plate tectonics more
likely—with time.
4. There is a wide range of single lid tectonic regimes observed on Io (heat-pipe single
lid), Venus (vigorous single lid) and Mars (sluggish single lid). These variations in
single lid behaviour largely reflect the sub-lithospheric thermal structure of the silicate
body and lithospheric thickness, strength and composition. A variety of single lid
styles is also expected to have characterized Earth as its mantle cooled and lithospheric
mantle thickened and oceanic crust thinned. Possible variants include heat-pipe, plutonic
squishy lid, stagnant lid and sluggish lid, and these single lid modes are variously
affected by sagduction, eclogitic dripping, delamination and plume–lid interactions.
Earth’s tectonic regime during the boring billion/Earth’s Middle Age (approx. 1.7–0.8 Ga)
was probably a protracted episode of sluggish single lid.
5. The transition from single lid to plate tectonics requires a lithospheric weakness
approximately 1000 km long or longer. Early in Earth history this could have formed by
large bolide impacts. Later in Earth history progressively damaged lithosphere and/or
plumehead–lithosphere interactions would have provided the weak zones needed for
the first subduction zone to form. Subsequent subduction zones needed to make the
global plate mosaic may have formed similarly or exploited new weakness caused by
plate tectonics such as transform faults and fracture zones.
6. The formation of continental crust and plate tectonics are distinct processes and should
not be conflated because continental crust can and almost certainly did form without
plate tectonics. The zircon U-Pb age spectra need to be critically reconsidered in light
of the likelihood of single lid tectonic episodes in Earth history. The scarcity of zircons
greater than 3.0 Ga indicates that little felsic continental crust was produced before this
time and/or preserved, implying a thin, weak lithosphere dominated by basaltic and
komatiitic crust that easily collapsed back into the mantle. Plate tectonics produces and
destroys continental crust at a quasi-continuous rate, so large peaks at approximately
2.7 Ga and approximately 1.9 Ga are unlikely to have formed by plate tectonic processes.
The dominant peak approximately 2.7 Ga may mark a major episode of mantle overturn
leading to widespread granite melting, while the approximately 1.9 Ga peak may mark
a similar episode of mantle overturn associated with proto-plate tectonics, inferred from
the presence of ophiolites about this time. The smaller variations in U-Pb zircon ages
observed for the past approximately 1 Ga are more consistent with modern plate tectonic
processes than the larger variations observed for older rocks.
7. Plate tectonics produces unique igneous assemblages such as ophiolites as well as unique
metamorphic assemblages such as blueschists and UHP terranes. With the exception of
a brief episode approximately 2.0 Ga when some ophiolites formed, these assemblages
do not appear in the rock record until Neoproterozoic time, signifying that the modern
episode of plate tectonics began in the past 1 Ga.
8. The increase in the abundance of kimberlites since the Neoproterozoic is explainable as
being due to the onset of massive transport of surface water and CO2 deep into the Earth
(mantle ingassing) due to subduction, which began on a large scale in the Neoproterozoic.
These volatiles migrate upward from deeply subducted slabs and are trapped at the base
of cratonic lithosphere. Build-up of volatile pressure at the base of cratonic lithosphere
over many millions of years leads to explosive eruption of kimberlites.
9. The transition from the boring billion single lid tectonic regime to plate tectonics should
have disrupted Earth’s climate equilibrium. An otherwise inexplicable major episode of
dramatic climate and C isotope swings known as Neoproterozoic snowball Earth was
likely to have been caused by this transition.
 Downloaded from http://rsta.royalsocietypublishing.org/ on October 2, 2018

10. The duration of Neoproterozoic climate and isotopic variations suggests that the
18
transformation from single lid to the first subduction zone (two plates) to a complete
global, plate tectonic mosaic took approximately 230 Ma to accomplish.

rsta.royalsocietypublishing.org Phil. Trans. R. Soc. A 376: 20170406

........................................................
Data accessibility. This article has no data.
Competing interests. The author declares that he has no competing interests.
Funding. No funding sources.
Acknowledgements. Thanks to Taras Gerya for advice about geodynamic modelling. I greatly appreciate co-
authors who have helped me understand the multiple lines of evidence that are woven together in this essay:
Marzieh Baes, Taras Gerya, Chip Groat, George Harlow, Matt Leybourne, Nate Miller, Stephan Sobolev, Paul
Tackley, Tatsuki Tsujimori and Scott Whattam. This paper is dedicated to the memory of Kevin Burke, who
inspired me to think big thoughts; Kevin is likely to have vehemently disagreed with the main points of this
paper. This paper benefitted from critical comments by Peter Cawood, Michael Brown and Anders Scherstén.
This is UTD Geosciences contribution no. 1322.

References
1. McKenzie D, Parker RL. 1967 The North Pacific: an example of tectonics on a sphere. Nature
216, 1276–1280. (doi:10.1038/2161276a0)
2. Stern RJ, Gerya T. In press. Subduction initiation in nature and models: a review.
Tectonophysics. (doi:10.1016/j.tecto.2017.10.014)
3. Stern RJ, Gerya T, Tackley P. 2018 Planetoid tectonics: perspectives from silicate planets, dwarf
planets, large moons, large asteroids. Geosci. Front. 9, 103–119. (doi:10.1016/j.gsf.2017.06.004)
4. Barrell J. 1914 The strength of the Earth’s crust. J. Geol. 22, 289–314. (doi:10.1086/622155)
5. Anderson DL. 1995 Lithosphere, asthenosphere, and perisphere. Rev. Geophys. 33, 125–149.
(doi:10.1029/94RG02785)
6. Solomatov VS, Moresi L-N. 1996 Stagnant lid convection on Venus. J. Geophys. Res. 101,
4737–4753. (doi:10.1029/95JE03361)
7. Rozel AR, Golabek GJ, Jain C, Tackley PJ, Gerya T. 2017 Continental crust formation on early
Earth controlled by intrusive magmatism. Nature 545, 332–335. (doi:10.1038/nature22042)
8. Breuer D, Spohn T. 2003 Early plate tectonics versus single plate tectonics on Mars: evidence
from magnetic field history and crust evolution. J. Geophys. Res. Planets 108, 8-1–8-13.
(doi:10.1029/2002JE001999)
9. O’Neill C, Roberts NMW. 2018 Lid tectonics—Preface. Geosci. Front. 9, 1–2. (doi:10.1016/j.gsf.
2017.10.004)
10. Fischer R, Gerya T. 2016 Early Earth plume-lid tectonics: a high-resolution 3D numerical
modelling approach. J. Geodyn. 100, 198–214. (doi:10.1016/j.jog.2016.03.004)
11. Nédélec A, Monnereau M, Toplis, MJ. 2017 The Hadean–Archaean transition at 4 Ga:
from magma trapping in the mantle to volcanic resurfacing of the Earth. Terra Nova 29,
218–223. (doi:10.1111/ter.12266)
12. Wyman D. 2018 Do cratons preserve evidence of stagnant lid tectonics? Geosci. Front. 9, 3–17.
(doi:10.1016/j.gsf.2017.02.001)
13. Smithies RH, Ivanic TJ, Lowrey JR, Morris PA, Barnes SJ, Wyche S, Lu YJ. 2018 Two distinct
origins for Archean greenstone belts. Earth Planet. Sci. Lett. 487, 106–116. (doi:10.1016/
j.epsl.2018.01.034)
14. Herzberg C, Condie K, Korenaga J. 2010 Thermal history of the Earth and its petrological
expression. Earth Planet. Sci. Lett. 292, 79–88. (doi:10.1016/j.epsl.2010.01.022)
15. Condie KC, Aster RC, van Hunen J. 2016 A great thermal divergence in the mantle beginning
2.5 Ga: geochemical constraints from greenstone basalts and komatiites. Geosci. Front. 7,
543–553. (doi:10.1016/j.gsf.2016.01.006)
16. Forsyth D, Uyeda S. 1975 On the relative importance of the driving force of plate motion.
Geophys. J. R. Astr. Soc. 43, 163–200. (doi:10.1111/j.1365-246X.1975.tb00631.x)
17. Conrad CP, Lithgow-Bertelloni C. 2002 How mantle slabs drive plate tectonics. Science 298,
207–209. (doi:10.1126/science.1074161)
18. Korenaga T, Korenaga J. 2016 Evolution of young oceanic lithosphere and the meaning
of seafloor subsidence rate. J. Geophys. Res. Solid Earth 121, 6315–6332. (doi:10.1002/2016JB
013395)
 Downloaded from http://rsta.royalsocietypublishing.org/ on October 2, 2018

19. Gurnis M, Hall C, Lavier L. 2004 Evolving force balance during incipient subduction. Geochem.
19
Geophys. Geosyst. 5, Q07001. (doi:10.1029/2003GC000681)
20. Zhou X, Li Z-H, Gerya TV, Stern RJ, Xu Z, Zhang J. 2018 Subduction initiation dynamics

rsta.royalsocietypublishing.org Phil. Trans. R. Soc. A 376: 20170406

........................................................
along a transform fault control trench curvature and ophiolite ages. Geology 46, 607–610.
(doi:10.1130/G40154.1)
21. Sizova E, Gerya T, Stüwe K. 2015 Generation of felsic crust in the Archean: a geodynamic
modeling perspective. Precambrian Res. 271, 198–224. (doi:10.1016/j.precamres.2015.10.005)
22. Seton M et al. 2012 Global continental and ocean basin reconstructions since 200 Ma. Earth Sci.
Rev. 113, 212–270. (doi:10.1016/j.earscirev.2012.03.002)
23. Arculus RJ et al. 2015 A record of spontaneous subduction initiation in the Izu-Bonin-Mariana
arc. Nat. Geosci. 8, 728–733. (doi:10.1038/ngeo2515)
24. Torsvik TH, Doubrovine PV, Steinberger B, Gaina C, Spakman W, Domier M. 2017
Pacific plate motion change caused the Hawaiian-Emperor Bend. Nat. Commun. 8, 15660.
(doi:10.1038/ncomms15660)
25. O’Neill C, Marchi S, Zhang S, Bottke W. 2017 Impact-driven subduction on the Hadean Earth.
Nat. Geosci. 10, 793–797. (doi:10.1038/ngeo3029)
26. Korenaga J. 2013 Initiation and evolution of plate tectonics on Earth: theories and
observations. Annu. Rev. Earth Planet. Sci. 41, 117–151. (doi:10.1146/annurev-earth-050212-
124208)
27. Bercovici D, Ricard Y. 2014 Plate tectonics, damage and inheritance. Nature 508, 513–516.
(doi:10.1038/nature13072)
28. Gerya T, Stern RJ, Baes M, Sobolev S, Whattam S. 2015 Plume-induced subduction initiation
triggered plate tectonics on Earth. Nature 527, 221–225. (doi:10.1038/nature15752)
29. Ernst RE, Buchan KL. 2002 Maximum size and distribution in time and space of mantle
plumes: evidence from large igneous provinces. J. Geodyn. 34, 309–342. (doi:10.1016/S0264-
3707(02)00025-X)
30. Regenauer-Lieb K, Yuen DA, Branlund J. 2001 The initiation of subduction: criticality by
addition of water? Science 294, 578–580. (doi:10.1126/science.1063891)
31. Grevemeyer I, Ranero CR, Ivandic M. 2018 Structure of oceanic crust and serpentinization at
subduction trenches. Geosphere 14, 395–418. (doi:10.1130/GES01537.1)
32. Garfunkel Z, Anderson CA, Schubert G. 1986 Mantle circulation and the lateral migration of
subducted slabs. J. Geophys. Res. Solid Earth 91, 7205–7223. (doi:10.1029/JB091iB07p07205)
33. Gerya TV, Connolly JAD, Yuen DA. 2008 Why is terrestrial subduction one-sided? Geology 36,
43–46. (doi:10.1130/G24060A.1)
34. Shirey SB, Richardson SH. 2011 Start of the Wilson Cycle at 3 Ga shown by diamonds from
subcontinental mantle. Science 333, 434–436. (doi:10.1126/science.1206275)
35. Tang M, Chen K, Rudnick RL. 2016 Archean upper crust transition from mafic to felsic marks
the onset of plate tectonics. Science 351, 372–375. (doi:10.1126/science.aad5513)
36. Elkins-Tanton LT. 2008 Linked magma ocean solidification and atmospheric growth for Earth
and Mars. Earth Planet. Sci. Lett. 271, 181–191. (doi:10.1016/j.epsl.2008.03.062)
37. Moore WB, Webb AG. 2013 Heat-pipe Earth. Nature 501, 501–505. (doi:10.1038/nature12473)
38. Condie KC, O’Neill C, Aster RC. 2009 Evidence and implications for a widespread magmatic
shutdown for 250 my on Earth. Earth Planet. Sci. Lett. 282, 294–298. (doi:10.1016/j.epsl.2009.
03.033)
39. Holland HD. 2006 The oxygenation of the atmosphere and oceans. Phil. Trans. R. Soc. B 361,
903–915. (doi:10.1098/rstb.2006.1838)
40. Cawood PA, Hawkesworth CJ. 2014 Earth’s middle age. Geology 42, 503–506. (doi:10.1130/
G35402.1)
41. Davies GF. 1992 On the emergence of plate tectonics. Geology 20, 963–966. (doi:10.1130/
0091-7613(1992)020<0963:OTEOPT>2.3.CO;2)
42. Johnson TE, Brown M, Kaus BJ, VanTongeren JA. 2014 Delamination and recycling of
Archaean crust caused by gravitational instabilities. Nat. Geosci. 7, 47–52. (doi:10.1038/
ngeo2019)
43. Chowdhury P, Gerya T, Chakraborty S. 2017 Emergence of silicic continents as the lower crust
peels off on a hot plate-tectonic Earth. Nat. Geosci. 10, 698–703. (doi:10.1038/ngeo3010)
44. Behn MD, Grove TL. 2015 Melting systematics in mid-ocean ridge basalts: application of a
plagioclase-spinel melting model to global variations in major element chemistry and crustal
thickness. J. Geophys. Res. Solid Earth 120, 4863–4886. (doi:10.1002/2015JB011885)
 Downloaded from http://rsta.royalsocietypublishing.org/ on October 2, 2018

45. Cadio C, Korenaga J. 2016 Macroscopic strength of oceanic lithosphere revealed by ubiquitous
20
fracture-zone instabilities. Earth Planet. Sci. Lett. 449, 295–301. (doi:10.1016/j.epsl.2016.05.027)
46. Van Thienen P, van den Berg AP, Vlaar NJ. 2004 On the formation of continental silicic melts

rsta.royalsocietypublishing.org Phil. Trans. R. Soc. A 376: 20170406

........................................................
in thermochemical mantle convection models: implications for early Earth. Tectonophysics 394,
111–124. (doi:10.1016/j.tecto.2004.07.058)
47. Moyen J-F, van Hunen J. 2012 Short-term episodicity of Archean plate tectonics. Geology 40,
451–454. (doi:10.1130/G322894.1)
48. Thébaut N, Rey PF. 2013 Archean gravity-driven tectonics on hot and flooded continents:
controls on long-lived mineralised hydrothermal systems away from continental margins.
Precambrian Res. 229, 93–104. (doi:10.1016/j.precamres.2012.03.001)
49. Campbell IH, Taylor SR. 1983 No water, no granites—no oceans, no continents. Geophys. Res.
Lett. 10, 1061–1064. (doi:10.1029/GL010i011p01061)
50. Hirose K. 1997 Melting experiments on lherzolite KLB-1 under hydrous conditions and
generation of high-magnesium andesitic melts. Geology 25, 42–44. (doi:10.1130/0091-7613
(1997)025<0042:MEOLKU>2.3.CO;2)
51. Wolf MB, Wyllie PJ. 1994 Dehydration-melting of amphibolite at 10 kbar: the effects of
temperature and time. Contrib. Mineral. Petrol. 115, 369–383. (doi:10.1007/BF00320972)
52. Petford N, Gallagher K. 2001 Partial melting of mafic (amphibolitic) lower crust by
periodic influx of basaltic magma. Earth Planet. Sci. Lett. 193, 483–499. (doi:10.1016/S0012-
821X(01)00481-2)
53. Mojzsis SJ, Harrison TM, Pidgeon RT. 2001 Oxygen-isotope evidence from ancient zircons for
liquid water at the Earth’s surface 4,300 Myr ago. Nature 409, 178–181. (doi:10.1038/35051557)
54. Ito E, Clayton RN. 1983 Submarine metamorphism of gabbros from the Mid-Cayman
Rise: an oxygen isotopic study. Geochim. Cosmochim. Acta 47, 535–546. (doi:10.1016/0016-
7037(83)90276-4)
55. Moyen J-F, Laurent O. 2018 Archean tectonic systems: a view from igneous rocks. Lithos
302–303, 99–125. (doi:10.1016/j.lithos.2017.11.038)
56. Condie KC, Aster RC. 2010 Episodic zircon age spectra of orogenic granitoids:
the supercontinent connection and continental growth. Precambrian Res. 180, 227–236.
(doi:10.1016/j.precamres.2010.03.008)
57. Rollinson H. 2017 There were no large volumes of felsic continental crust in the early Earth.
Geosphere 13, 235–246. (doi:10.1130/GES01437.1)
58. Condie KC, Arndt N, Davaille A, Puetz SJ. 2017 Zircon age peaks: production or preservation
of continental crust? Geosphere 13, 227–234. (doi:10.1130/GES01361.1)
59. Dickinson WR. 2008 Impact of differential zircon fertility of granitoid basement rocks in North
America on age populations of detrital zircons and implications for granite petrogenesis. Earth
Planet. Sci. Lett. 275, 80–92. (doi:10.1016/j.epsl.2008.08.003)
60. Stern RJ, Scholl D. 2010 Yin and Yang of continental crust creation and destruction by plate
tectonics. Int. Geol. Rev. 52, 1–31. (doi:10.1080/00206810903332322)
61. Stern RJ, Miller NR. 2018 Did the transition to plate tectonics cause Neoproterozoic Snowball
Earth? Terra Nova 30, 87–94. (doi:10.1111/ ter.12321)
62. Stern RJ, Leybourne MI, Tsujimori T. 2016 Kimberlites and the start of plate tectonics. Geology
44, 799–802. (doi:10.1130/G38024.1)
63. Stern RJ. 2005 Evidence from ophiolites, blueschists, ultra-high pressure metamorphic
terranes that the modern episode of subduction tectonics began in Neoproterozoic time.
Geology 33, 557–560. (doi:10.1130/G21365.1)
64. Scotese C. 2004 A continental drift flipbook. J. Geol. 112, 729–741. (doi:10.1086/424867)
65. Hawkesworth CJ, Cawood PA, Dhuime B. 2016 Tectonics and crustal evolution. GSA Today 26,
4–11. (doi:10.1130/GSATG272A.1)
66. Brown M. 2010 Paired metamorphic belts revisited. Gondwana Res. 18, 46–59. (doi:10.1016/
j.gr.2009.11.004)
67. Ernst WG, Sleep NH, Tsujimori T. 2016 Plate-tectonic evolution of the Earth: bottom-up and
top-down mantle circulation. Can. J. Earth Sci. 53, 1103–1120. (doi:10.1139/cjes-2015-0126)
68. Brown M, Johnson T. 2018 Secular change in metamorphism and the onset of global plate
tectonics. Am. Mineral. 103, 181–196. (doi:10.2138/am-2018-6166)
69. Bédard JH. 2006 A catalytic delamination-driven model for coupled genesis of Archean
crust and subcontinental lithospheric mantle. Geochim. Cosmochim. Acta 70, 1188–1214.
(doi:10.1016/j.gca.2005.11.008)
 Downloaded from http://rsta.royalsocietypublishing.org/ on October 2, 2018

70. Tappe S, Smart K, Torsvik T, Massuyeau M, deWit M. 2018 Geodynamics of kimberlites on a

21
cooling Earth: clues to plate tectonic evolution and deep volatile cycles. Earth Planet. Sci. Lett.
484, 1–14. (doi:10.1016/j.epsl.2017.12.013)

rsta.royalsocietypublishing.org Phil. Trans. R. Soc. A 376: 20170406

........................................................
71. Knoll AH. 2014 Paleobiological perspectives on early eukaryotic evolution. Cold Spring Harb.
Perspect. Biol. 2014, 6a016121.
72. Sanchez-Baracaldo P, Ridgwell A, Raven JA. 2014 A Neoproterozoic in the marine nitrogen
cycle. Curr. Biol. 24, 652–657. (doi:10.1016/j.cub.2014.01.041)
73. Knoll AH, Nowak MA. 2017 The timetable of evolution. Sci. Adv. 3, e1603076. (doi:10.1126/
sciadv.1603076)
74. Erwin H, Laflamme M, Tweedt SM, Sperling EA, Pisani D, Peterson KJ. 2011 The Cambrian
conundrum: early divergence and later ecological success in the early history of animals.
Science 334, 1091–1097. (doi:10.1126/science.1206375)
75. McKenzie NR, Hughes NC, Gill BC, Myrow PM. 2014 Plate tectonic influences
on Neoproterozoic-early Paleozoic climate and animal evolution. Geology 42, 127–130.
(doi:10.1130/G34962.1)
76. Valentine JW, Moores EM. 1972 Global tectonics and the fossil record. J. Geol. 80, 167–184.
(doi:10.1086/627723)

View publication stats