Transitions: The Evolution of Life

The Eocene saw the rise of the euprimates (a term coined by Elwyn Simons). An alternative term is “primates of modern aspect”. Going back to R. D. Martin’s definition of fossil primates the euprimates share claws replaced by nails, opposable hallux (big toe) and postorbital closure (among other things) in common with all modern primates.The euprimates are divided into two families, the adapidae and omomyidae. The adapidae are composed of two subfamilies: nothartinae (5 genera) and adapinae (14 genera). The omomyidae are divided into three subfamilies: anaptomorphinae (15 genera), omomyinae (12 genera) and microchoerinae (4 genera) Plus two species (Arapahovius and Loveina) of uncertain affinities (for those who know something of taxonomy they are Omomyidae incerta sedis).

In general, Omomyids are tarsierlike in their morphology. For example, the dental formula (which gives the number of each type of tooth) is often similar (in particular the lower dental formula is sometimes 1 incisor, 1 canine, 3 premolars and 3 molars - as seen in tarsiers). The crania have tapered snouts and the ectotympanic bone is tubular - as in tarsiers. The nasal and olfactory regions are diminished in size and the eye orbits are expanded.

The adapids, on the other hand, are generally considered to be related to lemurs. They have a divergent hallux, flattened nails on the digits. Some (such as notharctus)have a postorbital bar and a petrosal bulla. Encephalization quotients (a ration of brain and body weight) have been calculated ad are withing the range of other Eocene mammals and are slightly lower than say, the Oligocene anthropoid Aegyptopithecus. In the middle ear region they bear some resemblance to lemurs. The ectotympanic bone (which supports the tympanic membrane) is variable in adapids ranging from a free ringlike structure (as in lemurs) to one that is expanded to form part of the lateral bull wall (as in lorises and Aegyptopithecus).

The morphology of both leads to several interesting problems. First, what are the phylogenetic relationships? There are three competing theories.

1) Omomyids share some characteristics with plesiadapiformes and at one time the plesiadapiformes were thought to have given rise to omoyids. In this theory the adapids were considered ancestral to anthropoids and the prosimians (other than tarsiers).

2) Since no shared derived characters link tarsiers and anthropoids to the other prosimians it has been suggested that plesiadapiformes gave rise to the euprimates which split into two branches. One branch was composed of adapids, lemurs and lorises, the other was composed of omomyids, tarsiers and anthropoids. In this theory, tarsiers are more closely related to omomyids than to anthropoids.

3)A variant of number 2, except tarsiers are more closely related to anthropoids than they are to omomyids.

There is a further complication. In both 2 and 3 above tarsiers are grouped with anthropoids and adapids are grouped with lemurs and lorises. The problem is adapids share quite a few traits with anthropoids, tarsiers share some traits with anthropoids but not lemurs and lorises. Paleontological data supported a linking of adapids and anthropoids. Comparitive anatomy (hemochorial placenta, presence of a retinal fovea, for example) and biochemical data supported a relationship between tarsiers (and consequently omomyids) and anthropoids. This led to something of a stalemate. If tarsiers (and hence omomyidae) were more closely related to the anthropoids (as the anatomical and biochemical data suggested) then adapids (as the paleontological data suggested) couldn’t be. Which was right. A very intersting solution to this problem was presented by Gingerich and Schoeninger in 1977. The suggestion wasn’t paid much attention to until 1986, when Rasmussen (in his 1986 paper “Anthropoid Origins: A Possible Solution to the Adapidae-Omomyidae Paradox”) revived it. Grant the paleontological evidence that relates the omomyidae to tarsiers and adapidae to anthropoids. Lemurs and lorises would then form a sister group to both the omomyidae-tarsier group and the adapidae-anthropoid group. Consequently, tarsiers would be more closely related to anthropoids than to lemurs and lorises - which satifies the anatomical and biochemical evidence and the omomyid-tarsier and adapid-anthropoid groups could still be kept - satisfying the paleontological evidence. It’s a good theory, unfortunately, one small fact stands in the way. This is the traits which seem to relate adapids to lemurs and lorises. In this theory, the traits relating adapids to lemurs and lorises are due to parallel evolution. Which has raised some objections since parsimony requires little or no parallel evolution.
It’s been my experience that these kinds of situations come up a lot in primate - and human - evolution. No matter what phylogenies you create parallel evolution always comes into play. Personally, I consider a certain amount of parallel evolution to be a fact of life in primate evolution.

I examined several definitions of the order primates in a previous post and looked at how any definition of primates has to be constricted when applied to the fossil record. In particular, the closer one gets to the common ancestor between primates and other mammals (insectivores for example) the harder it becomes to tell what is a primate and what is not.

Currently, there are four theories as to which group primates arose from (although, there is general agreement that thier orgins lie in the Order Insectivora):

1) Erinoaceomorpha (hedgehogs)
2) The suborder Lipotyphla - which contains shrews, moles, tenrecs and selenodons.
3)Tree shrews
4) Archaic taxa such as Leptictidae or Apatemyidae - which may or may not be insectivores.
The earliest identifiable primates are the plesiadapiformes. The plesiadapiformes occur from the mid Paleocene to the Eocene (from about 65 mya to about 53 mya). They inhabited both North America and Europe. The plesiadapiformes are an infraorder composed of six families and almost forty genera. The families are:

1) Plesiadapidae (five genera)
2) Carpolostidae (three genera)
3) Saxonellidae (one genus)
4) Microsyopidae (24 genera)
5) Paromomyidae (three genera)
6) Picrodontidae (two genera)

The phylogeny of most of these groups has been worked out in greater or lesser detail. Consider the Plesiadapidae. The earliest genus was Pronothodectes. In North America the earliest member of this genus was Pro. matthewi. Pro. matthewi gave rise to Pro. jepi. From here it gets complicated. Pro. jepi gave rise to two different groups, Nanodectes and Plesiadapis. The Nanodectes lineage goes as follows: N. intermedius, N. gazini, N. simpsoni, and N. gidleyi. The Plesiadapid branch goes as follows: Ples. praecursor, Ples anceps. Ples anceps gave rise to two lineages. The first goes: Chiromyoides minor, C. caesor, C. potior, C. major. The second lineage of Ples. anceps goes as follows: Ples. res, Ples. churchilli. Ples. churchilli also split into two lineages. The first goes: Ples. fodinatus, Ples. dubius. The second lineage is Ples. simonsi, Ples. cookei. The phylogeny of the other families is equally complicated.

Comparitive work on modern primate dentitions has allowed us to come up with some general guidelines on determining things like diet. Based on this we say that the plesiadapiformes were primarily insectivorous with diets resembling modern prosimmians such as lemurs and galagos. However by the late Paleocene - Early Eocene some plesiadapiformes were adopting diets of fruit or leaves.

How are plesiadapifomes related to later primates? There are actually three different views on this question. One view is that plesiadapiformes are not primates because they are distinct from later primates. A second view places them in a suborder with tarsiiformes (because both groups share some traits in common - enlarged, protruding incisors, similar configurations of inner ear anatomy among others). A third view is that they are the earliest primate radiation (because they have primatelike teeth that make them important for understanding primate origins).

For Further Reading:

Primate-like mammals:
A stunning diversity in the tree tops

Archonta (primates, bats, tree shrews and flying lemurs)

The pictures below are all of primates.

The jury is still out as to whether the tree shrew (pictured below) is a primate, or is related to the primates.

Primates come in all shapes and sizes, so the question “how do anthropologists and paleontologists” is a natural one. The question necessarily involves a great deal of comparitive anatomy but I have tried to keep it to a minimum. Readers who need an overview can read this article on the anatomy of the skull. I have written this post in the form of a question and answer session.

What is a primate?
That is actually a good question and the answer is quite complicated.

What do you mean? Aren’t monkeys monkeys?
Well yes, but it’s more complicated than that. It is easier to define modern living primates. Primates as a group do share some unique, universal (among primates) features not shared by other mammals. Unfortunately, the also share features in common with other mammals.

Could you give an example of a unique (or diagnostic) feature that separates one mammal group from another?
Sure! The double pulley configuration of the astralgus (a bone in the hind limb) is diagnostic of artiodactyls.

But, what about primates?
That is a good question. Mivart first defined the order primates. His defination was:

Unguiculate, claviculate placental mammals, with orbits encircled by bone, three kinds of teeth, at least at one time of life; brain always with a posterior lobe and calcerine fissure; the innermost digit of at least one pair of extremities opposable; hallux with a flat nail or none; a well developed caecum; penis pendulous; testes scrotal; always two pectoral mammae.

Wow, that’s a lot!
Actually there is more. Mivart gave his definition in 1873. In 1959 Le Gros Clark added to it:

Preservation of generalised limb structure with primitive pentadactyly (five fingers). Enhancement of free mobility of the digits, especially of the pollex and hallux (both used for grasping). Replacement of sharp, compressed claws by flat nails; development of verysensitive tactile pads on the digits. Progressive shortening of the snout. Elaboration of the visual apparatus, with development of varying degrees of binocular vision. Reduction of the olfactory apparatus. Loss of certain elements of the primitive mammalian dentition. Preservation of a simple molar cusp pattern. Progressive expansion and elaboration of the brain especially of the cerebral cortex. Progressive and increasingly efficient development of gestational processes.

That seems pretty thorough. Is there more?
Yes, there is.

I was afraid of that.
Please don’t interupt. In 1967 Napier added two more:

Prolongation of postnatal life periods. Progressive development of truncal uprightness leading to a facultative bipedalism.

That’s a lot of information, where did you get it?
Mainly from R. D. Martin’s paper “Primates: A Definition”

So what’s the problem with the above definition of primates?
There are two problems. First, some of the above are actually trends, some of which are not features. Instead they refer to developments found only in some members of the group (remember, we are not trying to trace ancestor-descendent relationships at this point. We are trying to provide a definition of an order of mammals). Second, some of these are either traits that are probably primitive features of placental mammals or have arisen by convergence.

So, then how do we define primates?
Martin choose to examine living primates with an eye to creating a new definition, which I won’t bore you with since it is rather long.

You said the definition applies to living primates, what about fossils?
For fossils the definition has to be modified somewhat, but first we have to talk about tree-shrews.

?Tree-shrews?
Yes, you see Le Gros Clark argued that the Tupaiidae are more closely related to primates than to any other placental mammal and should be included in the order primates. This has been argued about ever since. Martin used tree-shrews as a test case for his definition of primates and decided (correctly, I think) that they were not primates.

That’s a pretty scientific approach!
Yes, it is. Paleoanthropology has a well developed scientific methodology and a rich body of theory to draw on.

So what about the fossils?
Since we have only skeletons to examine the definition has to be contracted somewhat. This is what is left:

Well developed, divergent hallux with flat terminal phalanx in the foot. Elongated distal segment of the calcaneus. Relatively large, convergent orbits with restricted interorbital distance. Postorbital bar present; ethmoid exposure in the orbit possible (depending in interorbital distance relative to skull size). Petrosal bulla. Relativly large braincase. Sylvian sulcus on endocast. Dental formula maximally 2.1.3.3/2.1.3.3. Premaxilla short; upper incisors arranged more trnsversly than longitudally. Molars with low, rounded cusps. Lower molars with raised, enlarged talonids.

So, does it identify fossil primates?
Yes, it does. According to this definition omomyids and adapids, for example, are primates.

What about plesiadapids?
The jury is still out on this issue. But see the next post.

For Further Reading:

Primate Adaptation and Evolution”

Primate Evolution: An Introduction to Mans Place in Nature

Primate Evolution

The Evolution of Primate Behavior

Major Topics in Primate and Human Evolution

Additionally, if you have access to a good (University) library:

The American Journal of Physical Anthropology
The Journal of Human Evolution
Folia Primatologica