Peripatus Home Page pix1Black.gif (807 bytes) Paleontology >> StratigraphyUpdated: 17-Dec-2018 



This page outlines some of the important concepts to understanding stratigraphy.

Keywords: stratigraphy, paleontology, uniformitarianism, fossil record, evolution


The science of determining the correct sequencing and relative dating of rocks is called stratigraphy, and – at a stretch – its origins can be said to lie with the Persian philosopher and physician, Avicenna (AD 980-1037 or thereabouts), who argued that the morphology of landscapes is largely due to the action of running water, and that great intervals of time, “during which the mountains themselves might be somewhat diminished in size”, were involved.

Consistent with the chauvinism of the times, his views were largely ignored by medieval European scholars whose understanding, instead, was thwarted by a literal reading of Genesis. Throughout the fifteenth through seventeenth centuries, most Europeans believed that land features were either formed when the Earth was created, or else the product of sudden, violent catastrophes. Supernatural causes were attributed to some of these events, such as the biblical Flood. This philosophy came to be known as “catastrophism”, although force majeure might serve equally well.

Nevertheless, some scholars were able to see more clearly. Possibly one of the first, certainly the best known, was Leonardo, who understood that fossils were not relics of the Flood but the remnants of creatures that once lived in a sea that formerly covered the area. There were others, but this is not a history lesson so we will next mention James Hutton (1726-1797). Hutton proposed an model of cyclic erosion and sedimentation in a paper first read before the Fellows of the Royal Society of Edinburgh in 1785, and finally published in 1788. Although he did not have our current understanding of the causal mechanisms – they would not come for nearly two centuries – his contribution was remarkable for presaging an essentially modern view of sedimentation and landscape evolution, and, crucially, in relying upon natural rather than supernatural mechanisms to explain them.

Charles Lyell (1797-1875) took up the baton with his Principles of geology published in 1830-1833 in which he argued that the formation of Earth’s surface took place through countless small changes occurring over vast periods of time. Again, he appealed to known natural processes rather than divine intervention, a philosophy which came to be known as “uniformitarianism” and gradually displaced catastrophism.

It is common to read about rocks (or fossils) belonging to a certain “Period” – say the Jurassic Period or the Cretaceous Period – and in the very early days of the science, these may have represented our limits of understanding. Nowadays, however, very few stratigraphers use such grand “period-scale” divisions of the geological timescale in their day-to-day work. Nevertheless, these are the divisions discussed on this web site, in part because they are so well-known, but mainly because they are still instructive and demonstrate important stratigraphic principles in a manner which other divisions do not.

Stratigraphy is employed on this web site as a supporting framework for the paleontology essays, rather than as a science in its own right. (I only half apologise for that.) So, whereas subdisciplines such as lithostratigraphy and chronostratigraphy are fascinating in themselves, here they will get a once-over-lightly, at best, and the focus will be on biostratigraphy.

Sedimentary Rocks

To a first-order approximation – the level of detail that might be covered in secondary (high) school science – there are three kinds of rocks: igneous, sedimentary and metamorphic. Igneous rocks have cooled from a molten precursor either erupted onto the surface or intruded below ground, whereas sedimentary rocks have formed from the consolidation of erosion products in various environments, usually but not always involving water. Metamorphic rocks have formed from the alteration of pre-existing rocks by heating, deformation, or chemical processes.

Stratigraphy predominantly concerns itself with the sedimentary kind: Rocks which have formed when any pre-existing rocks have been eroded to (generally) small particles, transported some distance where they have accumulated and consolidated to form new rocks.


When rocks are being laid down by a typical sedimentary process – water-borne particles falling to the bottom of a lake bed, for example – it is obvious that the first particles to settle will fall to the bottom, whereas particles which settle later will settle on top of the first lot. The layers at the bottom will be older than those on top. Later, when the sediments have become consolidated to rock, that relationship will still generally hold true. Subsequent deformation, perhaps associated with uplift and exposure, may tilt originally horizontal beds or even, more rarely, completely overturn them, and it may be quite difficult to tell which way the original bedding lay.

Principles, Definitions and Units


A principal aim of stratigraphy is correlation: determining which rocks correspond in some sense, to which others. Where the detailed nature of the rock, as determined by grain-size, colour, and various other characteristics which can generally be determined in the field and mapped, we may establish a lithostratigraphic correlation. The units of lithostratigraphy are rock units; the “formation” is the fundamental unit, although they may be subdivided into members, or aggregated into groups and higher units.

Where rocks have the same fossil content, we may establish a biostratigraphic correlation, although there are many confounding issues we must be aware of. An important one is “facies control”. No fossil is found everywhere; to a greater or lesser extent, the distribution of almost every organism is restricted in some way, and so it will seldom or never be found fossilised in some environments. For example, fossils of marine organisms such as molluscs, trilobites, etc., will never be found in inland coal deposits. (It is theoretically possible that a seagull could carry a shellfish inland and drop it, that it could subsequently fossilise and, against all odds, be found by a paleontologist. It could happen, and sometimes weird and wonderful things like that do happen. But it is highly unlikely and for practical purposes, we can ignore such very rare circumstances.)

Where we have reason to believe the rocks are the same age, we may establish a chronostratigraphic correlation. It is usual to distinguish between rocks of a certain age and the age interval itself; there are even different units represented for the two concepts. For example, the Jurassic System is the grouping of all rocks which fall above the defined base (see here) but below the base defined for the Cretaceous System. The Jurassic Period is the interval of time during which the Jurassic rocks were deposited. Our present best estimate of this interval is from 201 to 145 million years ago, although both of these estimates are uncertain, expecially the latter. So here is one fundamental difference between the two: the rock units are a matter of definition; the ages are something we have to try and find out.



emphasis Walshes point that type section => geochron unit => chronostrat unit


bits from Walsh

McGowran 2005, p. 393, makes the interesting (though typically tortured) comment that “this traditional stratigraphy, focussed on the Hedberg Triad of rocks, fossils and time (lithostratigraphy, biostratigraphy, and chronostratigraphy), progressively disenfranchised stratigraphy from its birthright, which is nothing less than the study of Earth history. Disenfranchisement happened in two ways: by an overemphasis on coarse, arbitrary and non-genetic codification of the data that divorced the data from Earth history; and by emphasizing the distinctness of rock, fossil and time units, thereby obscuring their interrelationships and ceding the important scientific questions to sedimentology (which focuses at the small scale, so that the big questions went unasked altogether). ... [T]raditional stratigraphy got mired in the codes and rules of the framework category to the detriment of the real science in the phenomenon category.”

Personally, as a practising biostratigrapher, I have never found this to be true. On the contrary, I find it invaluable to explicitly think about the geological data in terms of rocks, fossils and time – not constantly, but very often. I believe it is essential for a good practitioner never to forget that these three concepts represent real (natural) phenomena, which are not the same, no matter how we might conflate them in our heads.

However, I also know good scientists who would agree with McGowran, at least to some extent, and if they could be bothered to untangle his prose.

So, here, we reach the point where the reader is left to decide for themselves.


Biostratigraphic correlations are often used as proxies for time correlations. For example, the oldest fossil occurrence of a particular fossil is presumed be related to its evolution, which only ever happens once. (Species do not go extinct and then evolve again.) Depending on the type of organism, it might then spread out around a region, or the whole world, slowly or quickly. Marine microplankton are believed to mostly spread passively as they are carried about by ocean currents. It may take decades for a newly evolved planktonic organism – a foraminifera, say – to spread around an entire hemisphere, but in the context of our ability to tell geological time, this is effectively instantaneous. The first appearance of the foraminifera Praeorbulina glomerosa is generally believed to have occurred about 15.97 million years ago and so there is reason to suppose that any rock sequence which exhibits that first appearance event (called a “first appearance datum” or FAD) might also be around that age. However, it is important to know that there are definitely also counter-examples, where first appearances are demonstrably time-transgressive (“diachronous”) or the fossil organism is so sensitive to the local environment that its absence might more plausibly be explained by other reasons.

It is common to read about rocks (or fossils) belonging to a certain “Period” – say the Jurassic Period or the Cretaceous Period, and so on, and in the very early days of the science, these may have represented our limits of understanding. Nowadays, however, very few stratigraphers use such grand “period-scale” divisions of the geological timescale in their day-to-day work. Instead, they typically use units which represent much shorter intervals of time. Precisely what these units are will depend upon the kind of data available, the extent to which the data are time-transgressive, and – to a considerable extent – the history of geological study in the local area.

Uniformitarianism (and the New Catastrophism)

“Thus, by the 1960s there was a powerful consensus among opinion leaders in stratigraphy, tectonics, palaeontology and evolutionary biology that the record in the rocks is episodic preservationally but gradualistic in its message and can be accommodated in Lyellian uniformitarianism” notes McGowran 2005 (p. 387), and quite correctly, too.

He is quite wrong, however, when he goes on to say that we perceive things differently, now.

Yes, continental geodynamics are episodic, the stratigraphic record is punctuated by numerous, globally synchronous depositional sequences that reflect rapid eustatic sea-level changes, and the Phanerozoic fossil record is indeed punctuated by mass extinctions.

But these “new” realisations – this belated sophistry – completely miss the point. The great triumph of uniformitarianism was to replace “deluges” and other unexplained (and unexplainable) catastrophes, for which no plausible causal mechanisms are available, with “natural” processes which can be explained using known physics, and which, in some form, can be observed.

McGowran is literally correct in his analysis, but he demolishes a straw-man which never existed. This attitude has become quite fashionable of late. Despite being nothing more than a fad, it is no less annoying for all that. Uniformitarianism is nothing less that the triumph of evidence-based explanation over deus ex machina, and the tiny minds who point to occasional meteor impacts and say, “Look; Lyell was wrong!” are unfit to tie Lyell’s boot laces.


Biostratigraphy is the study of geological history – in particular, the sequencing and correlation of events – by means of fossils.

Fossils are extremely useful for establishing the relative ages of rocks and they can even give us qualitative hints about the absolute ages of rocks. We can intuit that plants and animals which look quite unlike anything around today, and especially whole assemblages of such exotic life forms, are likely to be ancient, whereas assemblages of things which look very similar to the modern biota cannot be very old.

However, biostratigraphy really comes into its own when it comes to making correlations – establishing which rocks are approximately the same age – which cannot be accomplished so economically by any other means.


Just as modern organisms have limited geographical ranges – kangaroos from Australia, kiwis from New Zealand – so ancient organisms had limited ranges also.



Hutton, J. 1788: Theory of the earth; or an investigation of the laws observable in the composition, dissolution, and restoration of land upon the globe. Earth and Environmental Science Transactions of the Royal Society of Edinburgh 1: 209-304.

Lyell, C. 1830-1833: Principles of geology. John Murray, London (3 vols.).

McGowran, B. 2005: Biostratigraphy - Microfossils and geological time. Cambridge University Press: 1-459.

 Peripatus Home Page pix1Black.gif (807 bytes) Paleontology >> Stratigraphy

Hits counted from 11 Dec 2018: Counter for tumblr
My Traffic Estimate