Dating Mammalian Evolution

By Claire Asher, on 28 March 2014

When the age of the dinosaurs ended around 65 million years ago, mammals stepped in to fill the gap, and the age of the placentals began. However, whether early placental mammals were already present on Earth before the demise of the dinosaurs has been the subject of a long standing debate. Recent research in GEE used genomic data, in combination with fossil evidence, to show that the earliest placental mammals were indeed scurrying between the feet of dinosaurs.

Shrew-like mammals scurry between the feet of dinosaursThe huge diversity of placental mammals on Earth today first appeared shortly after the mass extinction event that killed the dinosaurs. It is thought that the loss of the dinosaurs, along with much of life on Earth, freed up niches which placental mammals to evolved to fill. But were early placental mammals present, waiting in the wings, during the age of the dinosaurs, or did they appear rapidly after their demise? One recent study suggested that, based on fossil evidence, the placental mammals must have appeared after the cretaceous-tertiary boundary (KT) when dinosaurs and most life on Earth was wiped out. However, a recent paper by GEE’s Mario dos Reis and Ziheng Yang, in collaboration with Philip Donoghue from the University of Bristol, highlights flaws in the methods used in this study, and utilitsed a more thorough approach to show that early placental mammals likely predated the KT boundary.

Using genetic sequence data from over 14,000 genes, combined with fossil evidence, GEE researchers applied 3 alternative statistical methods to estimate the age of the earliest placental mammal; ancestor to all modern placental mammals. Although different statistical methods yielded slightly different estimates, and differed in their accuracy, they all agreed that placental mammals must have already been around before the dinosaurs went extinct. The adaptive radiation of mammals that occurred after the extinction of the dinosaurs was dramatic, but it was initiated by a few shrew-like species which had already evolved. This study highlights the importance of using statistical methods to estimate the true age of ancestral species; the age of the oldest fossils is not the same as the age of the ancestral species that gave rise to them, and statistical techniques must be employed to estimate this. Using both molecular and fossil evidence to inform estimates also provides more robust evidence for the true age of the first placental mammals, and the theory that the earliest ancestors of placentals predated the disappearance of the dinosaurs.

Original Article:

This research was made possible by funding from the Biotechnology and Biological Sciences Research Council (BBSRC).

The Dynamics of Population Declines

By Claire Asher, on 2 July 2013

Widespread declines in wildlife populations are a major concern, but global and regional targets to reduce the rate of biodiversity loss have so far not been met. Conservation practitioners are faced with an increasingly difficult task of balancing limited funding against increasing human pressures on wild populations. Methods for identifying detrimental human activities, identifying and classifying declines and prioritising species for conservation interventions are critical to ensure conservation efforts are invested wisely over the coming decades. Researchers at the Institute of Zoology (IOZ), Imperial College London and GEE have been working towards developing innovative methods for identifying pressures and determining conservation priorities in wild mammal populations.

Theoretical work suggests that different human pressures may result in different types of population decline, leaving a fingerprint of anthropogenic stress on the history of a population. Human activity might cause a constant pressure, or one that changes in proportion to the abundance of a species. Pressures may become more extreme as populations become smaller, for example when the value of a species to hunters increases as it becomes rarer. By contrast, a constant hunting effort would create a declining pressure over time, as individuals become increasingly difficult to find. Previous work by CBER’s Georgina Mace has suggested that the shape of the population decline curve observed may be characteristic of certain types of human pressure. A recent paper in Ecology and Evolution by IOZ’s Martina Di Fonzo, and Ben Collen and Georgina Mace from GEE, uses computer simulations and long-term monitoring data from nearly 60 species of mammal worldwide to test these theoretical predictions.

Different Population Decline Curve Shapes for Different Types of Human Pressure

Different Population Decline
Curve Shapes for Different Types
of Human Pressure

Taking a number of biological traits into account, such as life-history speed (e.g. generation time) and the carrying capacity of the environment, Di Fonzo, Collen and Mace (2013) simulated population size trends under a number of different types of human pressure. They looked at threats that are constant, increase or decrease in intensity as the population declines, and threats that act in proportion to the population size or that remain fixed as population size changes. They then statistically characterised the shape of the resulting curves using three different statistical models (linear, exponential, quadratic) and different curve shapes (concave or convex). This produced mixed results. Some types of pressure consistently produced the same type of decline curve regardless of the biological and environmental characteristics of the population, but others were more strongly influenced by generation length and habitat carrying capacity. This makes sense, since populations with a short generation time are likely to be better able to cope with and recover from human pressure. For most types of pressure, however, some tell-tale signs were usually identifiable. For example, pressures that are proportional to population size and increase over time tended to produce convex declines, whilst proportional pressure that decreases over time is more likely to produce a concave curve.

From Simulation to the Wild
That’s all very well and good, but how well does the simulated data match up to real-world wildlife populations? Using data from the Living Planet Index, the authors attempted to characterise the shape of real population declines and compared the model predictions with data about real human pressures impacting on these populations. Across 124 populations for 57 different species of mammal, they found that populations were most commonly experiencing concave declines, suggesting human pressures that are in proportion to population size, but decrease in intensity over time. There was some association between curve-shape and the actual sources of pressure, with exponential concave declines being associated with habitats suffering from exploitation, habitat loss, invasive species and pollution, and quadratic convex declines being more characteristic of disease-affected populations. However, most populations were subject to multiple human-pressures simultaneously, which may have partly obscured the relationship predicted by theory.

Population decline curves may not enable us to identify specifically which threats are affecting a population, however they can reveal important details of how those pressures are changing over time. The International Union for the Conservation of Nature (IUCN) Red list attempts to categorise species’ conservation status. Currently, the Red List includes population size decline when assessing conservation status, however this does not take into account whether those declines are accelerating or decelerating over time. Incorporating information about the shape of the decline curve can provide important insights for conservation – species experiencing accelerating declines may be prioritised when determining how to use limited funding. Population-level data can highlight the impact of human activities more rapidly than studying species-level information, as populations are likely to show more rapid responses to human change, and may act as an early warning of longer-term species decline.

The dynamics of real populations are complex, and species’ biological traits in combination with multiple human pressures acting on a population simultaneously can mean that reality does not always match perfectly with theory. However, studying the shape of population declines can reveal characteristics of the pressure being exerted on the population, may provide an early-warning system for species-level declines, and can be used to inform conservation priorities. This relatively simple method for assessing the type of decline a population is experiencing can provide valuable information to conservationists about which types of pressures are most influential, and how to act to prevent extinction.

Original Article:

() Ecology and Evolution

This project was made possible by funding from the Natural Environment Research Council (NERC)

Skulls and Teeth: Constraints on Mammalian Evolution

By Claire Asher, on 7 June 2013

When we think about evolution, we tend to imagine the appearance and behaviour of animals as infinitely malleable, changing and adapting to perfectly suit the whim of natural selection. However, in many cases, evolution is actually surprisingly constrained. The potential for a species to evolve into new environments, diets or lifestyles may be limited if the necessary variation isn’t available in the population. In this case, the raw materials for natural selection to act upon are missing and a species cannot evolve in a particular direction. Evolutionary constraints may be developmental or genetic in origin, and may in part explain why some groups of species are more diverse than others.

Early in mammalian evolution, around 160 million years ago, mammals diverged into two main groups; placental mammals (e.g. humans, rats, elephants) and marsupials (e.g. kangaroos, possums, koalas). Placental and marsupial mammals use a different strategy for early development; mammalian infants develop either internally (placental mammals) or in a pouch (marsupials). Despite a similar period of evolutionary history, there are far fewer species of marsupials, and as a group they show less diversity in terms of locomotion and diet. New research suggests their unusual developmental strategy may have constrained their evolution, limiting the variety of present-day marsupials.

A Koala (Phascolarctos cinereus)
A Marsupial Mammal


A Chipmunk (Neotamias umbrinus)
A Placental Mammal

Pouch-Living Limits Marsupial Diversity
Placental mammals give birth to reasonably well-developed young after a lengthy pregnancy. However, marsupials, such as kangaroos and koalas, do things a bit differently. The infant is ejected very early in development, and must crawl to a specialised pouch where it can complete the rest of its development, feeding by suckling. This developmental strategy is pretty demanding on the young marsupial, and requires that they are strong enough to crawl and suckle at a very young age. There is evidence that the shoulder has been constrained by the demands of crawling at a young age. Has suckling imposed similar constraints on the skull?

During development, embryonic connective tissue begins to form into bone through a process known as ossification. In marsupials, ossification in the skull and particularly in facial bones occurs much earlier in development compared to placental mammals. This pattern is probably necessary to allow marsupial young to suckle at such a young age. Early facial ossification has wide-ranging effects on brain, tooth and muscle development, and may have limited the diversity of facial shapes available to marsupials. Suckling may have been a major constraint on marsupial evolution.

Facial bones (viscerocranium)
and braincase (neurocranium)

In a recent study conducted in GEE, Bennett and Goswami (2013) measured a number of skull characteristics for 125 species of placental and marsupial mammals. Previous studies have failed to find evidence for evolutionary constraints affecting mammalian skulls, but these studies primarily focused on carnivores, and considered the whole skull at once. By looking at different regions of the skull separately, Bennett and Goswami found that marsupials exhibit less variation between species in the facial bones (viscerocranium) than placental mammals. This pattern is not present for the braincase (neurocranium), which is not involved in suckling behaviour, and where bones ossify much later. These results suggest that early suckling might have constrained the diversity of marsupial skull shapes, and may go some way to explaining why marsupials are less diverse.

Something to Get Your Teeth Into
Mammalian diversity has probably been constrained by a variety of factors, not just bone ossification. Another recent study from GEE and Earth Sciences by Halliday and Goswami (2013) showed that tooth growth patterns may have restricted the relative size and shape of mammalian teeth, potentially influencing diet. Mammalian tooth development is thought to proceed by an ‘inhibitory cascade’, whereby the growth of each developing tooth is controlled by a balance between stimulatory signals from the embryonic tissue, and inhibitory signals from the adjacent tooth. This results in the middle molar being intermediate in size between the first and last molar.

Coloured teeth

Lower right jaw of two early mammals, showing the first molar (yellow), second molar (green) and third molar (red).
Pentacodon occultus (left) – a very early insectivorous mammal
Hyracotherium (right) – the first horse, a browsing herbivore
Photograph by Thomas Halliday

Across 154 species of mammal, despite considerable variation in tooth shape and size, the relative sizes of the three molars conformed to the same basic pattern in about two thirds of species. The largest tooth was almost always either the first or last molar, in line with a common pattern of tooth development across all mammals. Diet is strongly related to tooth size and shape: mammals with a larger third molar (coloured red) tend to eat more plant matter, while mammals with a larger first molar (coloured yellow) tend to be meat-eaters. Across 180 million years of evolution, almost all mammals still conform to the same basic pattern of tooth development that their early ancestors used.

The Exception to the Rule
One extinct group of mammals, called condylarths, bucked the trend in molar shape. Their middle molar was largest, inconsistent with an inhibitory cascade. Perhaps this was an archaic adaptation to a new diet – condylarths were among the first omnivorous mammals, and a similar pattern is found in modern-day bears. The developmental constraint of the may not have been too strong for this particular deviation, though, and could have been achieved by a fairly simple modification to the inhibitory cascade.

Adaptation is not infinitely plastic. Natural selection can only work with the raw materials of variation in the population. Sometimes, one aspect of life constrains another, but if the pay-off is great enough, evolution will often find a way, despite the constraints.

Original Articles:

()Testing the Inhibitory Cascade Model in Mesozoic and Cenozoic Mammaliaforms BMC Evolutionary Biology 13: 79


This project was made possible by funding from the Natural Environment Research Council (NERC), as well as Abbey Research and Collaboration Awards and the University of London Central Research fund.