When Did Plate Tectonics Begin on Earth?

UPDATE 26 April 2016: On Friday 22 April during the European Geosciences Union 2016 General Assembly in Vienna, a Great Debate was held.  The question “Plate tectonics started in the Paleo-Archean or Earlier?” was debated. Please take a look and enjoy!

by Bob Stern (University of Texas at Dallas)

Busy person’s summary:  This is a long blog (~5000 words).  If you are too busy to read all of it, here’s the story in a nutshell: there will be a meeting this summer in Switzerland to discuss four related questions about the evolution of the solid Earth: “When did Plate Tectonics begin on Earth?”, “How did Plate Tectonics begin?”, “What was Earth’s tectonic style before Plate Tectonics began?, and “Why is it important to understand when and how Plate Tectonics began on Earth, and What came before?”. In preparation for this meeting, we will post blogs each month on each of the four related topics. This month’s topic concerns “When did Plate Tectonics start?”.  This blog begins by observing that there is no consensus about when Plate Tectonics began and then discusses four reasons why the controversy persists. The first reason is that many geoscientists do not fully understand what drives plate motions today, which makes it impossible to understand when these forces became strong enough to start Plate Tectonics. The second reason is that each geoscientist determines what is convincing evidence, so there are bound to be disagreements. The third reason is that uncertainties about preservation persist; each geoscientist decides how much of the evidence we need to resolve this controversy has been removed and how much remains? The final reason that the controversy persists is that strict interpretation of the Principle of Uniformitarianism hinders the creative thinking needed to make progress towards answering the question “When did Plate Tectonics start?”.

Complete Blog: This summer there will be what promises to be a most interesting international geoscientific meeting in the Swiss Alps.  The meeting is sponsored by the Eidgenössische Technische Hochschule Zürich (ETH), a leading technical university in Switzerland. Geoscientists from all over the world will assemble for a week at the ETH conference facility in Ascona to discuss four related questions about the evolution of the solid Earth: “When did Plate Tectonics begin on Earth?”, “How did Plate Tectonics begin?”, “What was Earth’s tectonic style before Plate Tectonics began?, and “Why is it important to understand when and how Plate Tectonics began on Earth, and What came before?”  This is an open meeting and anyone can apply to join what is sure to be a wide-ranging discussion; more information can be found here.

In this and blogs to come, we will provide some background information on each of the four topics; one blog will be posted on each of these different topics each month through June.  In this month’s (March) posting, some aspects of the question “When did Plate Tectonics begin?”  are explored. In April, Taras Gerya (ETH, Switzerland) will blog about “What was Earth’s tectonic style before Plate Tectonics began?” In May, Dave Brecovici (Yale U., USA) will blog about “How did Plate Tectonics begin?” The final blog before the meeting will be Lindy Elkins-Tanton (Arizona State U., USA), who will discuss some aspects of “Why is it important to understand when and how Plate Tectonics began on Earth, and What came before?”. We are thinking about blogging during the meeting as well.

Back to this month’s question “When did Plate Tectonics begin on Earth?” If readers think they will receive a definitive answer in this blog, they will be disappointed.  Some readers may have their own answer, which they may hold with confidence. Other readers may not be sure, and if these folks grab the nearest geoscientist and asks their opinion they are likely to get a firm answer. However, while individual geoscientists may have their own convictions (I certainly do, see Stern 2005; also check out a previous Speaking of Geoscience blog), the geoscientific community as a whole does not.  Figure 1. nicely summarizes the wide range of opinions on this topic. From this figure it appears that most (70%) workers conclude that Plate Tectonics began in Archean time but there are also those who argue that our unique style of convection began much earlier (>4.2 Ga) as well as much later (<1 Ga). Even for the majority “Plate Tectonics began in the Archean” group, there is 1.1 Ga difference in estimated age of onset. Clearly, the question “when did Plate Tectonics begin on Earth?” is very controversial!

Why is it so difficult to come up with a definitive answer?  It isn’t because geoscientists haven’t been working on this problem.  This will be the third focused workshop on the topic.  The first was the GSA Penrose conference “Pre-Mesozoic Plate Tectonics: How far back in Earth history can the Wilson Cycle be extended?” held in Vail, CO USA in 1975. It is interesting to see what geoscientists thought about the problem 40 years ago: “While recognizing the present difficulties of erecting quantitative models, the consensus was that plate tectonic motion, and thus the Wilson Cycle, took place during the Paleozoic Era.  A similar consensus was that Archean global tectonics were radically different from Mesozoic-Cenozoic Plate Tectonics, although some argued for Archean microplate tectonics”  (Dewey and Spall, 1975).  The second workshop on the problem was also a GSA Penrose conference held in Lander, WY USA in 2006 “When Did Plate Tectonics Begin?”.  The consensus of the Lander Penrose was that “modern-style” Plate Tectonics evolved from an earlier form of proto–plate tectonics in Archean time. The Lander Penrose took a straw vote before and after the meeting as to when participants thought Plate Tectonics began. Although in both cases about 70% voted that Plate Tectonics began sometime between 2.5 and 4 Ga, 20% of the participants changed their vote between the two ballots. One participant favored the idea that “modern-style” Plate Tectonics began about 1 Ga (That dissenter was me).

Different geoscientists attended the two Penrose conference but it is still surprising that the geoscientific community seems to have changed its collective mind so completely between 1975 (Archean global tectonics were radically different from Mesozoic-Cenozoic Plate Tectonics) and 2006 (Plate Tectonics began sometime between 2.5 and 4 Ga,).  Why is there such volubility in opinion about when Plate Tectonics began on Earth?

I can imagine four reasons why we cannot place the question “When did Plate Tectonics begin?” into the “Settled Science” category: 1) Incomplete nature of the Plate Tectonic revolution; 2) The challenges of evidence; 3) Uncertainties about preservation; and 4) the two-edged sword of Uniformitarianism.   Each of these points is interesting, and some of them have multiple aspects, for example the challenges of evidence. The rest of this blog looks more carefully at these four challenges and how these uncertainties impact the discussion “When did Plate Tectonics begin?”.

1) Incomplete Nature of the Plate Tectonic Revolution: What is meant by the phrase “incomplete nature of the Plate Tectonic revolution” and why would this affect thinking about when Plate Tectonics began?  Let’s address the first half of this question first, then the second half. Plate Tectonic theory is a purely kinematic explanation of how the lithospheric plates move, it was articulated and embraced by the geoscientific community without a dynamic explanation of why the plates move.  Plate Tectonics explains that there are three types of plate boundaries: divergent, where new plate is created; convergent, where plate is destroyed; and transform, where two plates slide past each other without plate being created or destroyed.  There are clear rules for plate motion, specified by poles of rotation and angular plate velocities.  Plate Tectonics is an elegant description of how the outer part of the solid Earth behaves but it is only kinematic.   This was enough at the time it was articulated in the late 1960’s because it resolved the controversy about continental drift, which had been festering for half a century.  Vigorous opposition to Wegener’s idea climaxed at a 1926 AAPG conference (Holmes, 1928; Anonymous, 1929), and this opposition was largely based on a perceived lack of a mechanism powerful enough to move continents.  Plate Tectonics provided the mechanism and resolved the controversy. From this historical perspective maybe it is understandable that those involved in the Plate Tectonic revolution were satisfied with a purely kinematic Plate Tectonic theory.

Although it is not part of the formal definition of Plate Tectonics, we now know that convergent plate margins are the surficial expressions of deep Earth systems called subduction zones (Stern, 2002). Subduction zones are regions where oceanic lithosphere sinks deep into the mantle (here’s a 9 minute animation that shows how Plate Tectonics and subduction are related). Subducted lithospheric slabs may stagnate just above the 670 km discontinuity, such as beneath China, or may sink through this discontinuity into the deep mantle, as is seen beneath the Marianas.  As a result, the deep subduction of oceanic lithosphere is an implicit part of Plate Tectonics now.

To understand why the lack of a wide understanding of what makes the plates move – Plate Tectonic dynamics – creates problems for understanding when Plate Tectonics began, we have to first consider what are the possible explanations for plate motions.  Beginning geology students (let’s call this the “Geology 101 explanation”) are taught that mantle convection moves the plates (Fig. 2), but this is a tautology (definition: the saying of the same thing twice in different words). Plate Tectonics is the surface expression of Earth’s convection, so saying that the plates move because of mantle convection is the same as saying that the surface expression of convecting body moves because of convection – geo-gibberish!  We have to dig deeper into what we understand about how and why Earth’s mantle convects if we are to understand when the forces that drive Plate Tectonics today became large enough for Plate Tectonics begin. (Note: the question of “What was Earth’s tectonic style before Plate Tectonics began will be the topic of Taras Gerya’s blog in April).

Let’s now look at the second half of this question: why do we need to understand why Earth convects before we can seriously address how Earth’s convective style has changed through time (which is another way of asking “When did Plate Tectonics begin?”). Convection is driven by buoyancy differences and can be caused by less dense material rising or by denser material sinking, or by a combination.   Density differences can be due to composition (for example, dense eclogite sinking through peridotite) as well as temperature (because the coefficient of thermal expansion for silicate minerals is >0, warmer peridotite is less dense than cooler peridotite).  Phase changes also play a role in determining density.  We know from the existence of Plate Tectonics that Earth’s mantle convects, but which of these possible explanations is responsible? If we know what are the driving and resisting forces that control Plate Tectonics today, we can imagine how this “force balance” (Gurnis et al., 2004) changed through time.

It turns out that we do have a good understanding of the forces that drive Plate Tectonics, and have known this for quite some time.  Beginning with the seminal study of Forsythe and Uyeda (1975) geodynamicists have figured out that it is mostly the sinking of dense oceanic lithosphere in subduction zones that moves the plates.  The slab pull force (F_SP in Fig. 3) is by far the most important plate-driving force, but secondary forces of ridge push (F_RP in Fig. 3) and drag on the base of the continental lithosphere (F_CD in Fig. 3) also contribute.  This understanding continues to be refined and quantified, for example by geodynamic modeling that shows how the age and length of the subducted slab controls plate velocities (Conrad and Lithgow-Bertelloni, 2002).

In spite of what geodynamics has learned, the fundamental dynamic misconception – the “Geology 101 explanation”  – persists (Fig. 2).

How does this dynamic misconception impede clear thinking about when Plate Tectonics began? If the “Geology 101 explanation” is correct and mantle convection due to escaping internal heat drives plate motions, then it stands to reason that the hotter early Earth (and there is consensus that the Archean mantle was hotter than modern Earth’s by 100-300°C; Herzberg et al., 2010) would have convected more vigorously than it does today and the Plate Tectonics may have existed since the Earth solidified. On the other hand, if our present geodynamic understanding is correct and plate motions are mostly determined by the sinking of lithosphere in subduction zones, then the question of “When did Plate Tectonics begin” requires addressing the question: “When did lithosphere become dense enough to sink into the mantle?”  Because lithospheric density increases with thickness and the thickness of a planet’s lithosphere thickens with time as it cools, the current geodynamic understanding of what drives plate motions favors a more recent onset of Plate Tectonics than does the erroneous “Geology 101 explanation” that the plates move because the Earth is losing heat.

2) The Challenges of Evidence; This set of problems concerns two sets of issues: the broad breadth of evidence and the fact that some lines of evidence are more compellingly diagnostic of plate tectonics than others.  Below are some thoughts on these “evidence-related” problems.

Let’s talk first about challenges stemming from the broad breadth of evidence. There is a huge range of potential lines of evidence that needs to be considered if we are to resolve when Plate Tectonics began – stable and radiogenic isotopes, igneous petrology, metamorphic petrology, paleomagnetics, geodynamics (just to name a few) – and no single geoscientist is expert enough to evaluate all of them. We naturally are most impressed by the types of geoscientific evidence that we are familiar with. For example, the isotope geochemist is most impressed by isotopic evidence, whereas the paleomagnetic expert is most convinced by paleomagnetic evidence.  Correspondingly, the paleomag expert can be as baffled by the isotopic evidence as the isotope geochemist is by the paleomagnetic evidence.

There are a lot of observations generated by the many geoscientific subdisciplines that need to be considered as potential lines of evidence, including paleobiology, paleoclimate, and planetology.  The skills needed to evaluate the wide range of evidence varies from geoscientist to geoscientist, but all of us are stronger in some subdisciplines than in others.  Fortunately, the geoscientific community has all of the expertise needed to evaluate all the evidence, but individual geoscientists do not. That’s one of the most important reasons for meetings like we will have Ascona, to bring the geoscientists with the different expertises together for discussion, teaching, and learning.  That and to involve the young geoscientists needed for this discussion to continue, evolve, and advance.

Let’s talk now about the second “evidence challenge”.  This issue here is that some lines of evidence are more convincing than others for answering the question “When did Plate Tectonics begin?”.  The list of compelling indicators of Plate Tectonic (and pre-Plate Tectonic) diagnostic indicators is still “under construction”. At present, all participants in the debate are free to define for themselves what is convincing evidence, and everyone has different standards. Confusion due to the different qualities of evidence is often magnified by the fact that all scientists like to inflate the significance of their research.  It’s much easier to get your manuscript accepted for a top journal if you claim to have found evidence for when Plate Tectonics began instead of just an interesting result, the significance of which is unclear.

When and if the debate “When did Plate Tectonics begin?” is finally resolved, we are likely to look back and see that some lines of evidence used in the debate were “false positives”, that some evidence originally thought to require Plate Tectonics could be explained in other ways.  There are lots of possible candidates for “false positives” in the recent literature.  There is compelling isotopic evidence (e.g., C isotopic compositions of diamonds, S isotopes of basalts) that surface materials have long been recycled deep into the mantle.  A recent study of 3.5 Ga Barberton komatiites found large decoupling of Nd and Hf isotopic systems and lithophile trace element signatures that were interpreted as indicating that pelagic sediments in the lower-mantle source of these lavas (Blichert-Toft et al., 2015). Does this require Plate Tectonics and deep subduction or can deep recycling be accomplished in other ways, such as bolide impacts and delamination? Certainly isotopic evidence of deep recycling could be evidence for Plate Tectonics but is Plate Tectonics required?  To me, the answer is “No!” There are surely other ways to move surface materials to depth, as can be glimpsed from active stagnant lid tectonic activity on Venus, Mars, and Jupiter’s innermost moon, Io.   Bolide impacts, which must have been very common in Hadean and Early Archean time would have injected material into convecting asthenospheric mantle. Delamination of crust could also inject surface materials into circulating asthenosphere.  (Stay tuned for Taras Gerya’s blog next month which will explore some aspects of “stagnant lid” tectonics).

Other “false positives” could result come from overinterpreting geochemical and petrologic observations to require Plate Tectonics and deep subduction.  Hydrous mineral inclusions in >4.0 Ga zircons could be due to deep subduction (Hopkins et al., 2010) and Plate Tectonics but is this the only possible explanation?  Eclogite inclusions in diamonds first appear ~3.0 Ga; this first appearance of eclogite inclusion could be due to Plate Tectonics and deep subduction (Shirey and Richardson, 2011) but is this the only way these could form?  A rapid decrease in Ni/Co and Cr/Zn ratios in detrital sediments between 3.0 and 2.5 Ga could reflect the beginning of Plate Tectonics during this time interval but is there no other explanation possible?  Archean greenstone belts are superficially similar to ophiolites and some authors cannot resist the temptation to obscure the differences (Furnes et al., 2014).

Certainly the beginning of Plate Tectonics was an important tectonic transition, but there may have been other important tectonic transitions that occurred that are recorded in isotopes, geochemistry and petrology.   If so, these indicators could give false positives for when Plate Tectonics began.

3) Uncertainties about Preservation: A third issue concerns the fact that the rock record is incomplete and becomes more incomplete the further we go back in time. We have almost no rocks from the first 700 million years of Earth history, our evidence for this time comes mostly from the remarkable zircons of Jack Hills in western Australia.  The incompleteness of the early rock record reflects a wide range of processes that operated in the ancient Earth, which was undoubtedly vigorous in terms of tectonic and igneous activity, whether or not there was Plate Tectonics.  Meteorite impacts battered the surface.  Igneous activity remelted the crust. Metamorphism changed the original characteristics of rocks, sometimes making these unrecognizable. The key question here is whether or not the diagnostic rock assemblages that must have been produced when Plate Tectonics began have been completely removed and reworked by these processes or whether some remnants persist. Is it only inclusions in minerals and cryptic geochemical indicators that still preserve evidence of Plate Tectonic beginnings, or do some of the diagnostic rocks and crustal tracts still exist?

An interesting example of the preservation challenge concerns interpretation of the histogram of U-Pb ages for zircons shown in Figure 4.  Zircons have become the single most important mineral for geoscientists interested in Earth history, and this is especially true for those of us interested in continental crust formation. Because zircons mostly form when felsic melts crystallize to form granitic rocks, zircon U-Pb ages reliably track the growth of the continental crust.  Tremendous datasets now exist, 37,830 individual U-Pb zircon ages were used for the study of Condie and Aster (2009)(Fig. 4A), and 100,445 for the study of Voice et al. (2011)(Fig. 4B). These two datasets differ somewhat.  Condie and Aster (2009) report ages for three groups of rocks and sediments: ancient detrital sediments, modern river sands, and granitic igneous rocks intruded during mountain-building processes.  The larger dataset of Voice et al. (2011) is only for detrital sediments. Nevertheless, the two studies reveal very similar results, except for <500 Ma samples (Condie and Aster 2009 database focused on Precambrian rocks).  There are five major peaks: ~2.7 Ga, ~1.9 Ga, ~1.0 Ga, ~600 Ma, and ~200 Ma, with possible sixth peak ~3.2 Ga.

What do these age peaks mean?  The significance is in the eye of the beholder.  One perspective takes these data at face value, accepting these to reflect the most important episodes of continental crust formation (that’s how I see it). Another perspective infers that these are preservation peaks, not production peaks. These investigators (e.g., Hawkesworth et al., 2009) begin with the strict Uniformitarianist (see next section) assumption that Plate Tectonics has always operated and that Plate Tectonics has for the most part produced new continental crust at convergent plate margins, at broadly similar rates. They argue that the apparently episodic nature of continental crust production inferred from these peaks is very hard to explain by Plate Tectonic processes, which is likely to be characterized by continuous production (and destruction; see Stern and Scholl 2010 for more information about how Plate Tectonics destroys as much continental crust as it creates, if not more).  They infer that these peaks therefore must be times of enhanced preservation and argue further that the peaks reflect the greater preservation potential of rocks formed during the latter stages of ocean closure and collision. They conclude that the record is therefore biased by the construction of supercontinents. Personally, I am baffled by this interpretation, for three reasons.  First, erosion and subduction will remove all the zircons that have been produced, not just ones of a certain age. Second, there are lots of basins adjacent to but set well back from convergent plate margins today, marginal basins in the Western Pacific and foreland basins in the western Americas.  These are found hundreds of kilometers from the plate boundary and if similar basins existed during earlier times of Plate Tectonic activity, the zircon record these preserve would be very difficult to destroy. Third, the best proxy we have for the zircon record produced by supercontinent formation is provided by the ongoing India-Eurasia collision. The zircons shed by the Himalayas into the Indus and Ganges-Brahmaputra rivers shows little evidence for the collision; the zircons in these river sediments overwhelmingly show unrelated ages much older than the collision.

Another interesting “preservation problem” is captured by Figure 5, which shows the distribution through time of five different Plate Tectonic indicators.  These are not indirect and minute indicators like isotopes or mineral inclusions, these are rock assemblages that occur in the field on the scale of tens to hundreds of kilometers. Such assemblages include ophiolites, which demonstrate some sort of seafloor spreading (Fig. 5A) and four different subduction indicators: blueschists, UHPs, Lawsonite-bearing rocks, and jadeitites (Fig. 5B).  Blueschists are widely acknowledged to only form in subduction zones (Palin and White, 2016). Ultrahigh-pressure (UHP) metamorphic rocks contain coesite or diamond and require subduction of continental crust to depths of at least 100 km. Lawsonite formation also requires conditions that today are only found in subduction zones.  The gemstone Jadeitite today only forms in subduction zones. Take a look at Figure 5, note that all five of these indicators first appear (with a few exceptions) in the Neoproterozoic or Early Paleozoic. Is this an artifact of preservation or a reasonably reliable indication of what really happened?

4) The Two-edged Sword of Uniformitarianism

Uniformitarianism is the view of Earth history encapsulated by the saying: “the present is the key to the past”.   The immensity of geologic time required by Uniformitarianism was encapsulated by James Hutton’s phrase “no vestige of a beginning, no prospect of an end”. This phrase, emphasizing the enormity of deep time, was used by Hutton principally to refute the strict biblical interpretation that the Earth was only a few thousand years old that dominated the thinking of his time. It is still useful today in refuting creationists like those of my own city (Dallas, Texas), which hosts the Institute for Creation Research .  The reader is invited to take a look at this glitzy website to see what we geoscientists are up against. The ICR home page pulldown “Resources” includes a section “Age of the Earth” which presents a detailed bible-based chronology to argue that the Earth is less than 2000 years old!

Uniformitarianism is still very useful for explaining what science has revealed about the antiquity of the Earth and the timescales of evolution to the many gullible and well-meaning people who are vulnerable to such nonsense. Uniformitarianism is much less useful for discussions of Earth history among rational Earth scientists, including the question of when did Plate Tectonics begin.  Following the Huttonian dictum “no vestige of a beginning, no prospect of an end”, application of strict Uniformitarianism to the question at hand would have Plate Tectonics operating from the beginning of Earth history (e.g., Windley, 1993).

Given the fact that Uniformitarianism is a key part of how we present our science to the public and how we teach new students, how are we to reconcile our Uniformitarianism “face” with a critical investigation of Earth history?  I don’t know, but it is useful to look a bit more deeply into Uniformitarianism beyond the way we use it for public outreach and undergraduate teaching.  The remarkable paleobiologist and polymath  Stephen Jay Gould, in one of his first scientific papers, argued against application of strict Uniformitarianist principles to Earth science research. Gould’s (1965) 104 word abstract succinctly states his useful insights and is reproduced below:

“Uniformitarianism is a dual concept. Substantive uniformitarianism (a testable theory of geologic change postulating uniformity of rates or material conditions) is false and stifling to hypothesis formation. Methodological uniformitarians (a procedural principle asserting spatial and temporal invariance of natural laws) belongs to the definition of science and is not unique to geology. Methodological uniformitarianism enabled Lyell to exclude the miraculous from geologic explanation; its invocation today is anachronistic since the question of divine intervention is no longer an issue in science.  Substantive uniformitarianism, an incorrect theory, should be abandoned.  Methodological uniformitarianism, now a superfluous term, is best confined to the past history of geology.”

I disagree somewhat with Gould (1965) because I still find Uniformitarianism to be useful in talking with the public, especially because they are subjected to disinformation from groups like ICR and their ilk and in teaching beginning geology students.  I strongly agree with Gould (1965) that Uniformitarianism is either anachronistic or “hypothesis-stifling” in discussions with my peers.  While Uniformitarianism is still a useful “white lie” along the lines of Santa Claus and the Easter Bunny for engaging beginning students and the general public, we know that substantive uniformitarianism is not useful for reconstructing Earth history because so many things happened in Earth history that are not happening today.  Examples of non-uniformitarian events include the impact to form the moon, Earth’s magma ocean, the heavy bombardment, generation of komatiites in the Archean, Proterozoic banded Fe formations, Neoroterozoic Snowball Earth, the Cambrian biological explosion, and the great Phanerozoic  extinction events.  Substantitive uniformitarian – including the idea that Plate Tectonics has always been like we see today –  is nonsense and inhibits objective interpretation of the rock record..

Concluding remarks: I hope this blog has stimulated the reader’s mental juices to learn more and contribute to our exploration of “When did Plate Tectonics begin?”. We are going to have a lot of fun exploring this topic at Ascona.  Please join us if you can, and please follow the monthly blogs to come. Don’t forget: in April, Taras Gerya will blog about “What was Earth’s tectonic style before Plate Tectonics began?” In May, Dave Brecovici will blog about “How did Plate Tectonics begin?” The final blog before the meeting will be Lindy Elkins-Tanton, who will discuss some aspects of “Why is it important to understand when and how Plate Tectonics began on Earth, and What came before?”

P.S. The interested reader is also encouraged to look at the “on-demand” video lectures on these topics presented at 2015 Fall AGU meeting ( Go to http://bit.ly/1TLTHh7 to log in; once in look at Union session U41A: When and How Did Plate Tectonics Begin, What Came Before, and Why is This Controversy Important for Understanding the Earth? There are 5 interesting talks, by Jun Korenaga, Michael Brown, Patrice Rey, David Bercovicci, and Linda Elkins-Tanton.

References

Anonymous 1929. Theory of Continental Drift: A symposium on the origin and movement of land masses both inter-contiental and intra-continental, as proposed by Alfred Wegener. Quarterly Journal of the Royal Meteorological Society 55 (229), 101-103.

Blichert-Toft, J., Arndt, N.T., Wilson, A., and Coetzee, G. 2015.  Hf and Nd isotope systematics of early Archean komatiites from surface sampling and ICDP drilling in the Barberton Greenstone Belt, South Africa.  American Mineralogist 100 2396-2411

Condie, K. and Aster, R., 2009. Zircon Age Episodicity and Growth of Continental Crust, EOS transactions. AGU, 90, 13 October, 2009

Conrad, C. P., and Lithgow-Bertelloni, C., 2002. How Mantle Slabs Drive Plate Tectonics. Science 298, 207-209.

Dilek, Y., Ophiolite pulses, mantle plumes and orogeny. Geol. Soc. London, Spec. Publ.218, 9–19 (2003).

Dewey, J., and Spall, H., 1975. Pre-Mesozoic plate tectonics: How far back in Earth history can the Wilson Cycle be extended? Geology 3, 422-424.

Forsyth, D., and Uyeda, S., 1975.  On the Relative Importance of the Driving Forces of Plate Motion.  Geophysical Journal International 43, 163-200.

Furnes, H., de Witt, M., and Dilek, Y., 2014.  Four billion years of ophiolites reveal secular trends in oceanic crust formation. Geoscience Frontiers 5, 571-603.

Gould, S. J., 1965. Is Uniformitariansim Necessary? Am. J. Sci. 263, 223-228.

Gurnis, M., Hall, C., Lavier, L., 2004. Evolving force balance during incipient subduction. Geochem. Geophys. Geosyst. 5, Q07001.

Harlow, G. E., Tsujimori, T. & Sorensen S. S., 2015. Jadeitites and plate tectonics. Annu. Rev. Earth Sci. 43, 105–138.

Hawkesworth, C., Cawood, P., Kemp, T., Storey, C., and Dhuime, B., 2009, A matter of preservation: Science 323, 49–50.

Herzberg, C., Condie, K., and Korenaga, J., 2010.  Thermal history of the Earth and its petrological expression. Earth and Planetary Science Letters 292, 79-88.

Holmes, A., 1928. Continental Drift. Nature 3073, 431-433.

Hopkins, M. D., Harrison, T.M., and Manning, C. E., 2010.  Constraints on Hadean geodynamics from mineral inclusions in >4 Ga zircons. Earth Planet. Sci. Lett. 298 367-376.

Korenaga, J., 2013. Initiation and Evolution of Plate Tectonics on Earth: Theories and Observations. Annual Reviews of Earth and lanetary Sciences 41, 117-151.

Palin, R. M., and White, R. W., 2016. Emergence of blueschists on Earth linked to secular changes in oceanic crust composition. Nature Geosciences 9, 60-64

Peltonen, P. and Kontinen, A., 2004.The Jormua ophiolite: A mafic-ultramafic complex from an ancient ocean-continent transition zone, in Kusky, T., ed., Precambrian Ophiolites and Related Rocks: Amsterdam, Elsevier, p. 35–71.

Shirey S.B., and Richardson S.H., 2011. Start of the Wilson cycle at 3 Ga shown by diamonds from subcontinental mantle. Science 333, 434–436.

Stern, R.J., 2002. Subduction Zones.  Reviews of Geophysics 40, 10.1029/2001RG000108

Stern, R.J. 2005. Evidence from Ophiolites, Blueschists, and Ultra-High Pressure Metamorphic Terranes that the Modern Episode of Subduction Tectonics Began in Neoproterozoic Time. Geology 33,7, 557-560.

Stern, R.J., and Scholl, D. 2010. Yin and Yang of Continental Crust Creation and Destruction by Plate Tectonics.  International Geology Review 52, 1-31.

Stern R. J., Tsujimori, T., Harlow, G. E. and Groat, L. A., 2013. Plate tectonic gemstones. Geology 41, 723–726.

Ting, M., Chen, K., and Rudnick, R. L., 2016. Archean upper crust transition from mafic to felsic marks the onset of plate tectonics. Science 351, 372-375.

Tsujimori, T. and Ernst, W. G., 2014. Lawsonite blueschists and lawsonite eclogites as proxies for paleo-subduction zone processes: A review. J. Metamorphic Geol. 32, 437–454.

Voice, P.J., Kowalewski, M., and Eriksson, K.A., 2011, Quantifying the timing and rate of crustal evolution: Global compilation of radiometrically dated detrital zircon grains. The Journal of Geology 119, 109–126.

Windley, B. F., 1993. Uniformitarianism today: Plate tectonics is the key to the past. J. Geol. Soc. London 150, 7-19.