(i) Preamble

So far we’ve discussed general aspects of species and community distributions and the role of climate in generating global distribution patterns. We’ve had some historical perspective (e.g. climate changes during the Pleistocene), but we’ve mostly dealt with current processes. We will now explore major historical processes with profound influences on animal distribution: continental drift and glaciation.

Much of the appreciation of historical processes that influence animal distributions owes its beginnings to people like Charles Lyell who used the composition and distribution of fossils to infer past changes in climate and sea levels and the Scot, James Hutton who championed the idea the "the present is the key to the past". In other words, we should be able to observe current processes that will yield insights into events in the past that have shaped biogeographic pattern. This idea of "uniformitarianism" eventually supplanted an earlier view that most patterns in biogeography were driven by episodic, catastrophic events (although the latter may still contribute).

In order to discuss and appreciation historical events in biogeography, we need to appreciate the time scale over which such processes have operated. The geological time scale provides the basic time scale for the history of the earth. You should become familiar with the general divisions of the time scale (i.e. Eons, Eras, Periods, and Epochs) and their durations in time, i.e. if the period "Permian" is mentioned you should know that it represents a time period of roughly 290- 250 million years ago.

The time scale was achieved by dating stratigraphic layers of rock and fossils of fairly uniform composition that had wide geographic distributions. By using half-lives of radioisotopes found within such deposits, estimates of the ages of these deposits could be made. For instance, isotopes of uranium have VERY long half-lives (> 500 million years) and decay to lead. The decay of these isotopes provided the current estimate of the earth’s age of about 4.6 billion years. By contrast, carbon 14 (C-14) decays to C-12 with a half-life of about 5800 years and, therefore, is useful for dating much more recent events.

(ii) Continental drift: history and basic tenets

Continental drift essentially encompasses the idea that the present day distribution of the earth’s continents is different from what is was in the past. The changes in distribution (and orientation) of the continents came about because they have drifted or "rafted" as they rode on crustal plates that floated on a semi-liquid upper mantle.

The most common observation that lead to various individuals postulating about changes in continental positions was the complementarity of coastlines of several continents, the most notable example being how closely South America and Africa appear to "fit". This idea was first put forth in the mid 1500’s by a Dutch cartographer (Ortelius) and was often reiterated by others such as Von Humboldt and Darwin in the 1800’s. The first succinct description of continental drift as a hypothesis was articulated by Alfred Wegener a German meteorologist in the early 1900’s. F.B. Taylor, an American geologist, independently (and slightly earlier) speculated that the continents drifted and that mountain ranges were formed at the leading edges of continents as they "ploughed" across the earth with oceans forming at the trailing edges.

Alfred Wegener (1880-1930)

Wegener published the basics of his hypothesis for continental drift in a series of books during the 1910-30 period with the definitive outline published in 1929. The basic tenents of continental drift according to Wegener were:

(1) The continents were composed of ligher, less dense rocks and the oceanic rocks which suggested that the continents could "float" atop a semi-liquid mantle (see below).

(2) The major landmasses of the earth were once united as a single "supercontinent" called Pangaea. The breakup of Pangea began in the Mesozoic with the splitting of Africa and South America (North America and Europe still united until much later).

(3) The breakup of Pangea began with the formation of a giant "rift valley" which as it widened became an ocean basin and the progenitor of the Atlantic Ocean. The mid-ocean ridges and trenches (becoming known to people at this time) marked the zones where the continents were once joined and formed as the continents split apart, respectively. Earthquakes and volcanoes are signals of such activity.

(4) The shapes of the continents are more or less as they have always been, allowing historical reconstruction of their margins. When the continents are matched-up by the fit of their margins, similarities in fossils, stratigraphy, and ancient climates between now separated continents strongly suggest they were once united.

(5) Rates of movement of the continents vary, Greenland moves at the slowest rate and has only recently (<100,000 yr) separated from Europe. Wegener suggested that continents moved at rates of up to 36 m/yr (he was off by quite a bit)!

(6) The above processes are gradual, and ongoing and likely have their ultimate causes in heating processes deep in the earth’s mantle (this was an extension of Lyell's ideas on "uniformitarianism").

Initially, these ideas were rejected. Wegener presented no information on a detailed mechanism for continental drift (only vague references to "mantle heating", celestial forces, etc), and many disjunct distributions could still be explained just as adequately by land bridges between continents. As late as the 1950’s many still believed that it was a crazy idea and that animals were much easier to move, even if by "rafting" across thousands of km of ocean, than continents. The lack of a detailed mechanism, in the form of plate tectonics, was the crucial limitation and would not be well accepted until over 30 yr after Wegener’s death in 1930 (tragically, he frozen to death on Greenland!).

(iii) Evidence for Continental Drift

Until the theory of (and evidence for) plate tectonics was established post 1950, four kinds of evidence for the possibility of continental drift had been presented:

1.Stratigraphic evidence:

the "lining-up" of distinct rock types and assemblages on now distant continents. Note the clustering of Precambrian rock between continents in the Northern Hemisphere (I) and basaltic deposits in the Southern Hemisphere (II) in the figure below.

2. Paleoclimatic and Paleontological evidence:

Late Paleozoic glacial deposits are common to all continents of the Southern Hemisphere. As the glaciers moved they made deep grooves (striations) in the underlying rock. The orientation of the grooves makes it possible to determine the direction of glacier movement. A map of inferred movement is shown on the upper panel below (I). The shaded areas represent areas of glacial deposits and most of the arrows suggest movement of glaciers from the ocean onto the land (note arrows on east coast of S. American and southern margin of Australia). Also, note the heavy black line which indicates the distributional limits of the so-called "Glossopteris" flora which is a fossil record of vegetation that grew at the marginal edges of glaciers. The distribution of this flora occurs on now widely separated continents. Both the direction of the glacier movements and distribution of the Glossopteris flora make more sense when viewed in the lower panel with the continents all together as Pangea (II, i.e. the landmass held a central massive glacier sheet that extended outwards in all directions accounting for the direction of the striations and the distributional limits of Glossopteris flora).

In addition to the distribution of plant fossils (see above), many animal fossils in the Paleozoic glacial deposits that were shared among now separate continents suggested that they were one landmass with a similar climate. A few examples are given in the figure below. The distribution of the Triassic land reptile, Lystrosaurus, is often cited as a classic example. It was "mammal-like" (actually ecologically like a modern day hippopotamus) which specialized on chomping down on aquatic plants that grew over widespread areas in Antarctica, India, and Africa. Other fossil deposits shared among continents include the alligator-like Mesosaurus and Cynognathus.

 3. Disjunct distributions:

Disjunct distributions of a wide diversity of living taxa (especially those not well equipped for long-distance dispersal through the sea (beetles, clawed frogs – see below) suggest that they were at one time part of a more continuous distribution. The panels below show (from the top) disjunctions in a tribe of carabid beetles, galaxid fishes, Proteaceae (a plant family), and clawed frogs (Pipidae) in the southern hemisphere and suggest that their ancestors radiated and dispersed across a much larger continuous landmass. What do the distributions say about the timing of these radiations relative to the splitting of Pangaea into northern "Laurasia" and southern "Gondwanaland" (i.e. note that there are very few instances of any dispersal into the Northern Hemisphere)??

Note: the present-day ranges of each taxon are in dark shading.

(iv) Mechanism of continental drift: Plate tectonics

An excellent web site maintained by the U.S. Geological Survey that describes much of the history and mechanics history and mechanics of continental drift and plate tectonics. Please visit for any more details on what follows below, for some cool graphics, and figures for aspects of what is discussed below.

 There were developments in four major areas of "geosciences" that spurred the development of the theory of plate tectonics as the mechanism driving continental drift:

1. Sea floor mapping:

Even fairly primitive bathymetric surveys began to dispel the notion that the ocean floor was uniform in structure and featureless. By the mid 1800’s, naval surveys had identified large mountainous areas in the mid-Atlantic ("Middle Ground"). Echo-sounding developments in the early 1900’s refined these early maps and identified the mountain chain in the middle Atlantic Basin as the "mid-Atlantic Ridge". Also discovered were "seamounts" (submerged volcanic islands that tended to be associated with ridge areas) and "guyots" (deeper volcanic islands that appeared to be eroded seamounts and tended to be found farther away from oceanic ridges than seamounts). Later still, the mid-Atlantic Ridge was discovered to be part of a 50,000 km long chain of ridges mid-ocean ridges that wove through all the worlds ocean basins. It’s the largest single geographical feature on the earth and is 800 km wide in places and rises to over 4,500 m in some areas.

2. Magnetic striping

The basaltic rocks of the ocean floor contain large amounts of the magnetic mineral, magnetite. Through the use of "magnetometers", geologists are able to measure the magnetic polarity of rock bearing magnetic material (including continental rock formations). "Normal" polarity means that the polarity of the magnetite is the same as the current day earth’s magnetic field (i.e. the magnetite crystals act like tiny magnets that whole poles line up in a North-South orientation). "Reversed" polarity means the poles of the magnetite are opposite to that of the earth today (i.e. South lines up with the earth’s magnetic North). Reversals in the polarity of the earth’s magnetic field are a natural, if not completely understood, phenomenon, that take place (on average every million years or so). These polar reversals are revealed because as molten rock (magma) cools, the orientation of the magnetite is "locked in" as the rock solidifies. Mapping of the ocean floor demonstrated that: (a) there were regular patterns of alternating polarity in the rock as one moved out from the oceanic ridges, (b) the alternating patterns were symmetrical on either side of the ridges, and (c) that the age of different strips increased as one moved away from the ridge crests.

The figure below illustrates the magnetic striping and the symmetrical pattern of polar reversals.

A magnetic map of the ocean floor for areas just offshore from B.C. and Washington State can be viewed at by clicking HERE.

3. Sea floor spreading

Putting the observations of ocean floor terrain (ridges, seamounts, guyots) and of magnetic striping together prompted various people to propose the idea of "sea floor spreading". This idea, in its barest form, proposes that the mid-ocean ridge areas are zones of structural weakness through which magma is extruded to the ocean floor from deep within the earth’s mantle (see below). At these ridge "suture zones", the earth’s crust is literally being "torn apart" as new magma rises to create new ocean crust. As it rises through the ridge suture zones it displaces older crust and causes the ocean floor to "spread out" from the ridges. These ideas neatly explained: (a) why older rocks (and guyots) are found as one moves away from ridges (young rocks and seamounts tend to be found near the ridge crests), (b) the phenomenon of magnetic striping, and (c) why the youngest rocks near the ocean ridge crests have the current day magnetic polarity.

Clearly, however, if the sea floor spreading ideas were correct then if new ocean crust is continually being generated, why has the earth not increased in size over geological time? Harry Hess, a Princeton geologist and ex-Navy man, broadened the idea of sea floor spreading in the early 1960’s to include the idea of oceanic crust "recycling". If new crust is being formed, and the earth has remained a constant size – then old crust must be being "consumed" somewhere else! The sites of oceanic crust consumption were suggested to be the very deep ocean trenches that tend to be found where oceanic plates meet each other or where ocean plates meet continental plates.

4. Earthquake activity concentration

Seismology, or the study of earthquakes, has contributed to the theory of plate tectonics in that it has documented that most earthquake activity tends to be associated with areas of ocean trenches and spreading ridges. Pinpointing the locations of the world’s earthquake zones helped to focus attention on these areas and, hence, on the dynamic movements of plates. Essentially, these earthquakes (and to some extent volcanic activity hotspots) were vivid examples of the dynamics of plate movements (and their consequences!). Earthquakes are caused by the slipping of plates as they move apart to create "faults" (fractures within rock formations): "extensional" faults at spreading centres, "compressional faults" where plates collide, and "transform faults" when plates slip laterally with respect to one another. The most severe earthquakes tend to be found at compressional faults.

Current model of Continental Drift

The current working model of continental drift involves large convective forces generated by the different degrees of heating and resultant solid, semi-solid, and liquid states of the three major zones of the earth: the crust, mantle, and core. Think of a large pot of boiling water; the differential heating of different layers of the water generate water flow through convection. The core has "inner" and "outer" components that are about 1,200 km and 2,200 km in radius, respectively. They tend to be made of solid (inner) and liquid (outer) combinations of iron and nickel. The mantle is about 2,900 km thick and is only semi-solid and tends to composed of more iron/magnesium. The plates are typically about 100 km in thickness with continental portions tending to be thicker than oceanic crust which averaged about 5 km thick.

The upper portion of the mantle and the crust, together constitute the "lithosphere". The lower portion of the mantle, which is in a more liquid state, is called the "asthenosphere". The consensus is that the lighter lithosphere floats atop the more liquid, but denser asthenosphere to generate continental drift.

The lithosphere has the aforementioned "structural weak spots" that form the ocean ridges and which form the "breaks" between any two adjacent continental and oceanic plates. For a nice colour representation of the different plates click HERE.

The drifting of the continents is driven by three basic kinds of forces: "ridge push", "mantle drag", and "slab pull". These are illustrated in the figure below. Basically, ridge push is the upward push of magma from the asthenosphere through the mid-ocean ridges and is the chief force generating new ocean floor. As the seafloor spreads out from the ridges owing to ridge push, the lithosphere interacts with the underlying semiliquid asthenosphere to generate mantle drag, a friction force. Finally, when one plate meets another, one has to give. The denser plate (tends to be oceanic plates when they meet continental plates) is "subducted" (i.e. it sinks) below the lighter plate. As gravity pulls the heavier plate downwards into the molten magma, this generates so-called "slab pull’’. The pull is generated by the sinking leading edge that pulls the trailing portion of the plate downwards into the "subduction zone". The relative strengths of the three forces varies from place to place, but slab pull may be the dominant forces in many areas.

These forces and the interactions between plates generate: (a) subduction zones (aka convergence zones where two plates meet), (b) spreading zones (where plates are separating), and (c) strike-slip faults (where adjacent plates "grind" or "slip" against one another, say of they are roughly equal in density). Classic results of such interactions between plates consist of ocean trenches that form at subduction zones. Mountain building can also occur as these convergence zones when the convergence is between equally dense continental plates. Areas of concentrated volcanism characterize many spreading zones (e.g. Iceland marks a spot where the North American and Eurasian plates are spreading apart, Great Rift Lakes of Africa have resulted from local spreading fissures within the African plate). Intense seismic activity (earthquakes) is associated with strike-slip zones (e.g. the San Andreas Fault is perhaps the most infamous).

Various subduction, spreading, and strike-slip zones are illustrated below. Very deep ocean trenches at subduction zones are known along the Aleutian Peninsula, near New Zealand, and the famous "Marianas Trench'' where the Pacific and Philippine plates meet (the trench has been measured at over 11,000 m deep!).

(v) Tectonic history of the earth

 We know most about the tectonic history of the earth over the last 300 million years. The basic starting point for discussions of continental drift is the existence of the supercontinent known as "Pangaea". In reality, Pangaea appears to have been around for a relatively "brief" time; only about 60 million years. Before that time, a supercontinent known as "Rodinia" existed about 750 mya. That continent fractured and the northern continents that made up the landmass known as "Laurasia" (see below) had a rather fractured history and more or less drifted around the globe independently from each other from 750 to 260 million years ago. "Gondwanaland", the landmass comprised of the southern continents (Africa, SA, Antarctica, Australia) appears to have had a longer history as a unit, persisting more or less intact for most of the Phanerozoic (although its position changed over that time). A more detailed "time-lapse" account of the continents is on reserve in Woodward Library and below is a time-lapse simulation of the drift of the continents:

A few points to note:

The breakup of Pangaea was initiated in the Northern Hemisphere in the early Jurassic with expansion of the "Turgai Sea" in the Arctic. By this time, Pangaea was more or less subdivided into Laurasia and Gondwanaland by the emergence of the Tethyan Seaway. The expansion of the Turgai Sea resulted in a split between North America + Europe and Asia.

There has been a long history of interconnectedness between North America and Eurasia, either through connections between eastern NA and Eurasia or via "Beringia" (between the extreme northwestern portion of NA and northeastern Russia). The faunal similarities between northern North America and Eurasia is reflected in the designation of so-called "Holarctic" faunas. For example, North America and Eurasia share about 20 % of their mammal fauna at the family level compared to NA sharing 8% and 6% of mammal faunas with Africa and Australia, respectively.

Gondwanaland broke up in three major stages (see figure below). The first separated the landmass into eastern and western Gondwanaland (SA and Africa split from all others). The second stage was marked by the initial separation of SA and Africa and the separation of India and Madagascar from Australia+Antartica. The final stage started some 100 mya and saw the separation of Australia/NZ from Antarctica and the separation of India and Madagascar. India apparently travelled north at a rate of some 15cm/yr and the collision of the India Plate with the Eurasian Plate resulted in the mountain building activities forming the Himalayas. The long period of isolation of Australia is reflected in its generally sharing a low percentage of its faunas with other areas. For modern mammals, Australia shares less than 20% of its fauna with any other major continent and has a very high incidence of endemism; about 91% of all Australian terrestrial mammals are found there and nowhere else. By contrast, North America has about only about 13% endemism in its terresrtrial mammals and Eurasia about 3% (the latter has been connected to many other areas for a long time period).

References:

Brown, J.H. and Lomolino, M.V. 1998. Biogeography. 2nd Ed. Chapter 6.

Cox, A. et al., 1964. Reversals of the earth's magnetic field. Science 144: 1537-1543.

Dalziel, I.W.D. 1995. Earth before Pangaea. Sci. Am. Jan. 1995: 58-63.

Fisher, A. 1988. What flips earth’s field? Popular Sci. Jan. 1988: 71-113.

Fooden, J., 1972. Breakup of Pangaea and isolation of relict mammals in Australia, South America, and Madagascar. Science, Wash. 175: 894-898.

Hays, J.D. & Pitman, W.C., 1973. Lithospheric plate motion, sea level changes and climatic and ecological consequences. Nature, Lond. 246: 18-22.

Jardine, N. & McKenzie, D., 1972. Continental drift and the dispersal and evolution of organisms. Nature, Lond. 235 (January 7): 20-24.

Thornton, I.W.B., 1980. Plate tectonics and the distribution of the insect family philotarsidae order Psocoptera in the Southwest Pacific. Palaeogeography 31(2-4): 251-266.