Mesozoic
Eucynodonts

Trevor's
essays HOME

The Evolution of Mesozoic Mammals, a Rough Sketch

PLEASE NOTE: THIS PROJECT IS NOT SCIENTIFIC. IT IS A HOBBY.
"I was looking for information on an old mammal and found this lot. What is this project?"
It's got lots of information on old mammals. For a short bit of background information, see here.

"I was looking for the feature on the evolution of breasts and lactation."
see Putting the Mamms onto Mammals.

The Evolution of Mesozoic Mammals, a Rough Sketch
A Fifty word version

Mammals evolved from cynodont protomammals and made their first appearance during the Upper Triassic. Crown-group mammals are known from the Middle Jurassic. The earliest identified representatives of the three existing lines, (eutherians -including placentals, metatherians -including marsupials, and the egg-laying monotremes), are all from the Lower Cretaceous.

What follows are my own unqualified ramblings. Treat with caution.

I thought I'd try to produce a narrative overview of my on-line directories, and explain a bit about the reasoning behind the structure.

Mesozoic Eucynodonts, Genus Index:
http://home.arcor.de/ktdykes/meseucaz.htm

As the structure has much to do with the present understanding of evolutionary development, (none of which is in any way my work), this also provides an opportunity to draw up some sort of sketch of that too, as far as I'm able. There are many gaps in the fossil record. Nevertheless, the broad sweep of much anatomical development is surprisingly well evidenced. Of course, new fossils and methods of study will change and deepen understanding as they become available. As far as possible, I'll attempt to use reasonably accessible English, though there will be some paleo-tongue-twisters.

BEFORE THE MESOZOIC PLUS MUCH OF THE TRIASSIC

Crudely put, mammals evolved from reptiles, which evolved from amphibians. That's a bit too crude, and I prefer to think of the first 'reptiles' as pre-reptiles. They also get called stem-reptiles and many more complicated things besides. That seems to be the lineage from which 'mammal-like reptiles', known as synapsids, emerged. This development probably occurred approximately 300 million years ago during the Upper Carboniferous. These synapsids flourished during the following Permian period, and then seem to have been sliced off in their prime during a mass extinction. Only three lines are known to have crept into the Triassic.

The physically smallest of these animals, resplendent with the name of therocephalians, aren't known beyond the Middle Triassic. The plant eating 'two-dog-teethers', the dicynodonts, disappeared during the lower Upper Triassic. The third group, the cynodonts, seem to have mostly gone with them. Some of these were the dominant predators of their days; a role which was assumed by new fangled things called dinosaurs. There were also other representatives with herbivorial habits.

However, some of the more derived cynodonts, the eucynodonts, ('true dog tooth'), held on. Rather like Tom Jones in the late 1990s, they even enjoyed a further flourish to their career. And they're still singing Sex Bomb now.

EUCYNODONTIA (THE 'ADVANCED' CYNODONTS)

The first widely recognized member of this pack is Cynognathus, ('dog jaw'). It's known mainly from the Lower Triassic, though some finds may pre-date this. It was the top dog of its time; a wolf-sized predator with pretty much a worldwide range. As befitting the first acknowledged eucynodont, it's kennelled on my opening directory. It enjoys the company of various colleagues, which are mostly hard to place more accurately.

In the same subdirectory can be found the remaining, non-mammalian Triassic lineages. I'm not confident of the validity of the scheme I'm presently using, but it seems to be the best I can do for now. It involves a clear distinction between herbivore and carnivore lines, which is far neater than any such arrangement today. Then again, this has to do with anatomy rather than with the actual diet. The giant panda is clearly a carnivore, despite its love of bamboozling. Our ancestry lies amongst the meat-eaters, anchored somewhere within the strangely named superfamily, Chiniquodontoidea. In an effort to be more precise, the dromatheriids can't be ruled out as culprits. However, this is perhaps because they're so poorly known. Most genera are based on isolated, very mammal-like teeth.
The Triassic herbivores are presently corralled together under the term of gomphodonts.

Triassic Eucynodonts
http://home.arcor.de/ktdykes/trias.htm

INTO THE JURASSIC

At least three groups of small to tiny eucynodonts survived into the Jurassic. Enough remains have been recovered to show all these critters were closely related, although this cuts across the traditional classes of Reptilia and Mammalia, if both are defined widely.

I refer to one as 'Jurassic cynodonts'. If taken too literally, this title could be misleading. Not all the listed creatures were Jurassic, and all eucynodonts are cynodonts. However, a fair number of people, (eg. the viewers of the first episode of Walking with Dinosaurs), have at least heard the word cynodonts, although it might not spring readily to mind. They were those things in the burrow being annoyed by the predatory dinos. I'm cheating slightly and assuming people will accept this usage as indicatiing nearly-mammals. It's home to the plant-munching tritylodontids.

According to some views, tritys are descendants of traversodonts. However, the prevailing opinion now sees them as specialised probainognathians, and close relatives of mammals.

As of April 2005, a seperate directory was established for titheledontans; a loose grouping of insectivores which just reached the Jurassic. They could be even closer to mammals. As their lifestyle was so similar to the earliest mammals, trithes may have become obsolete. The tritys -the herbivores- enjoyed better luck. They've even been reported from the upper Lower Cretaceous of Siberia and Japan.

Jurassic Cynodonts; Tritylodontidae
http://home.arcor.de/ktdykes/jtherap.htm

Tritheledonta
http://home.arcor.de/ktdykes/tritheledonta.htm

The third lineage I mentioned is a family called Morganucodontidae. (Ultimately, the name derives from Mr Morgan and a tooth.) Sticking to a broad definition, these represent the first known mammals. A little anatomy:
- The lower jaw of a reptile, (here used to include non-mammalian synapsids), is constructed from a number of bones. The mammalian equivalent is a single bone.
- There's a different joint for attaching the jaw to the skull.
- Typical reptiles have one kind of tooth. Like most other mammals with choppers, you have four. They work as a set. That's why we only experience a single change of gnashers. If we enjoyed serial replacement, there'd constantly be gaps, and the dentition wouldn't work as a well drilled team.
- The development of some bones around the mouths of newborn mammals has to be well advanced, so as to allow suckling.
- Our hearing is exceptionally good for animals. Our inner ear, which is encased within a single bone, (four in reptiles), has a long channel; the cochlear canal. This enables us to process sound more effectively.
- Unlike reptiles, mammals reach a maximum size and stop growing. Coming to think of it, so do reps. In their case though, this is caused by death.

The best known member of this family is Morganucodon. It happens to be remarkably well preserved, and the following is clear:
- Whilst its lower jaw was actually multi-boned, the 'extra' (postdentary) bones were tiny.
- Whilst it possessed both types of jaw joint, (as was the case amongst some more mammal-like protomammals), the mammalian one was overwhelmingly dominant.
- It had strongly differentiated teeth and replaced them only once. There's no sign of serial replacement on any specimen, and the wear facets indicate that the individual teeth had been in use for more or less the same length of time.
- The bone development in babies allowed for suckling.
- Its ear was housed in a single bone and had an elongated cochlear canal.
- If allowed a full lifespan, Morganucodon stopped growing long before it dropped dead.

It will never be possible to say with certainty that this is the big mama of all mammals, (for goodness sakes, it was only about 10 cm long). However, it does an extremely good impersonation of being just that.

Both in terms of time and geography, Morganucodon was a wide ranging genus. Remains have been reported from the Upper Triassic to the Lower Jurassic, and from across the northern hemisphere. The best specimens have been recovered from the Jurassic of China. Before the excess rock was cleaned away, one Chinese fossil was tentatively referred to as a baby. This turned out to be tantalizingly incorrect. It's the skull of a tiny adult, (or near adult), which is now known as Hadrocodium. This creature weighed about the same as a non-Mesozoic paperclip. In terms of anatomy, it's also considerably further along the evolutionary road towards our selves. Due to a lack of much to compare it with, a better understanding of this midget must await further discoveries. I've placed it on this directory for want of anywhere more appropriate.

A family with the somewhat frightening name of Megazostrodontidae might be somewhere near to the docodonts, though most opinions see closer relationships with the morganucodontids.

Basal Mammaliaformes, Morganucodontidae, Megazostrodontidae and Hadrocodium
http://home.arcor.de/ktdykes/morganu.htm

EARLY 'SYMMETRODONTS' AND OTHER STRANGE WORDS

The next three directories are rather isolated affairs. There are clues about affinities with other groupings, though decisive evidence is lacking.

Kuehneotherium and friends
Briefly put, scrappy. Symmetrodonts is a word that sometimes crops up in articles and books concerning the Mesozoic. In a very rough summary: first mammal = Morganucodon, then came symmetrodonts, followed by things called pantotheres and then us, with various stations in between. And very loosely speaking, that's probably something like right. As a valid group, however, Symmetrodonta can be extremely dubious, depending upon what's included. Symmetrodonts are named for the vaguely triangular arrangement of points on their cheek teeth. There's a big middle one and another to each side. This is sort of symmetrical. It's also not that different to some of the teeth of Morganucodon. It's a condition that goes back into our early mammalian roots, which makes it pretty much useless for defining relationships.

As well as being difficult to spell, the kuehneotheriids are sometimes termed 'obtuse angle symmetrodonts', which has to do with the configuration of these cusps. They're known only from bits of jaw and teeth, which hale from the Upper Triassic of Western Europe and the Lower Jurassic of India and Wales. The jaw construction is of a 'primitive' grade. There were later 'obtuse angle symmetrodonts' too, but there's little evidence to suggest this is a sign of kinship. The "amphiodontids" fall into this category. They're poorly known and of uncertain affinities. The various genera probably aren't even related to each other.

Kuehneotheriidae and Co and "Amphiodontidae"
http://home.arcor.de/ktdykes/kuehneo.htm

Docodonts plus

The modest woutersiids are only known from Upper Triassic teeth from Western Europe. Docodont teeth are rather distinctive to those who appreciate such things, and the relationship could be near to an ancestral one, though this is increasingly seen as doubtful. With the exception of the wonderously preserved Haldonodon from Portugal, fully paid up docodonts are not well known. Typical fossils are teeth and jaw fragments from the Upper Jurassic of North America. An increasing range of fossils from Asia seem to be fleshing out a distinctive family of docodonts called Tegotheriidae.
According to a couple of recent and highly derived finds from South America, this lineage survived right up until near the end of the Cretaceous. A number of 'northern Jurassic' groups made an eccentric and late comeback in Patagonia, though they've not recorded since.

Woutersiidae and Docodonta
http://home.arcor.de/ktdykes/doctridon.htm

Haramiyidans
If the kuehneotheriids are scrappy, this lot must be mega-scrappy. Apart from one jaw with a few bits and pieces, this entire group is known solely from teeth. Nevertheless, they do have their attractions. They're largely confined to the Upper Triassic of Western Europe and Greenland. A couple of later genera, (from the Middle and Upper Jurassic of the UK and Tanzania respectively), are perhaps better described as haramiyidan-like.

The dental details show these critters were predominantly herbivores. The teeth have fine quantities of cusps on them, reminiscent of the multituberculate mammals. An ancestral relationship between the two lines has some support. One problem though is the lack of evidence. Furthermore, what evidence there is tends to say 'probably not'. These haramis chewed in a side-to-side way. Multis were backwards and forwards chompers, which suggests their jaws must have been hinged very differently. This could perhaps have been a dramatic bit of reengineering from the workshop of evolution. If so, there's no fossil evidence for it yet.

It's not even clear whether these critters were less, more or about as derived as the most basal mammals. They might be non-mammals according to even the stretchiest of definitions.

Haramiyida
http://home.arcor.de/ktdykes/harami.htm

MULTITUBERCULATES

This subdirectory covers the best known of Mesozoic mammals. The name might be difficult enough to remember, but it appears fairly often in books which include mammals at all. They were herbivores to omnivores, with lots of cusps on their teeth, thus the name. They also had a couple of tusk-like lower teeth, and upper incisors reminiscent of rodents. Multis probably led a similar kind of lifestyle and survived well beyond the Mesozoic. They finally seem to have been replaced by rodents about 40 million years ago, leaving no descendants. The first proto-mices turn up during the Paleocene, which is the time immediately following the extinction of Tyrannosaurus and all. This catastrophe had no noticeable effects on the multis. They were diverse both before and after it. As rodents became more common though, multis declined. They didn't go out with a bang, but more amidst a growing chorus of squeaks and eeks.

The oldest (unnamed) multituberculate fossils are from the Middle Jurassic of England. The earliest substantial material comes from Guimarota in Portugal, a site which is Upper Jurassic. A bit later, allodontid multis turn up in the rocks of North America. In the Mesozoic sites where multituberculates are found, they're generally the most common and diverse of the mammal fauna. This probably reflects their role in the northern hemisphere at the time. It also makes northern mammal sites where multis haven't been found intriguing. However, the explanation might be 'not yet found'.

According to the most recent scientific view, most multituberculates can be reasonably assigned into two suborders. The most basal members are within 'Plagiaulacida'. The apostrophies indicate that this doesn't represent a natural group; it doesn't include one ancestor and all of its descendants. It's the safest interpretation presently possible with the fossils available. These creatures inhabit the first two directories. The chronological range of the group extends from the Middle Jurassic until the Lower Cretaceous.

Basal Multituberculata
http://home.arcor.de/ktdykes/plagiau.htm

Plagiaulacidae, Albionbaataridae, Eobaataridae & Arginbaataridae
http://home.arcor.de/ktdykes/basmult.htm

One Lower Cretaceous genus from Mongolia, Arginbaartar, shares characteristics with these plagis. It seems to have its origins within or near to the allodontids. It also has characteristics in common with the more derived multis, the cimolodontans. These facts do not mean it must represent a link between the two groups. Cimolodontan genera such as Paracimexomys are under suspicion.

Cimolodonta seems to be a monophyletic suborder. That means it includes one ancestor and all of its descendants. Its most basal members are the just cited Paracimexomys and its kin. This suborder includes all the more derived multituberculates. This means that cimolodontans survived from the Lower Cretaceous until at least the Eocene, (35 - 55 million years ago).

'Basal' Cimolodonta, Cimolomyidae, Boffiidae and Kogaionidae
http://home.arcor.de/ktdykes/cimolod.htm

Whilst the relationships of multis to each other is quite well known, their placement within the broader scheme of Mammaldom is unclear. Their descendants are easily described; none, dead, gone. Not a sausage. The best available candidates as their ancestors are the haramiyidans, but this seems far from convincing. There are apparently some similarities in the braincase and the internal ear structure of multis and monotremes, (platypus and Co), but that's about as far as it goes. For reasons I don't pretend to understand, certain aspects of skull and limb anatomy suggest these things were more 'primitive' than some lineages confined to the Mesozoic. Nevertheless, they out-survived them - and why not? We and our egg-laying monotreme cousins are more closely related with each other, than either of us are to these multis.

And this points to an interesting sting in the tail; or in the hips. Where known, the pelvis of multituberculates is rather narrowly built. It seems too narrow to have enabled the females to lay eggs. This suggests, where such evidence has been preserved, that at least some multis reproduced a lá live birthing. Given our nearer relationship with monotremes, this implies to me that live birthing must have evolved at least twice within mammalian history. Either that, or they produced extraordinarily tiny eggs.

Djadochtatherioidea
http://home.arcor.de/ktdykes/djado.htm

Eucosmodontidae, Microcosmodontidae and Taeniolabidoidea
http://home.arcor.de/ktdykes/taenio.htm

Ptilodontoidea
http://home.arcor.de/ktdykes/ptilodon.htm

Multis are more or less restricted to the northern hemisphere, with the exception of some fossils from the Upper Cretaceous of Patagonia. These were described as ?Ferugliotherium, which is a genus within Gondwanatheria. This group has been assigned to the multis in the past, but that's not the present view. They may be related to the predominantly South American, tooth-poor edentates; anteaters and the like. However, some poorly preserved material is apparently multituberculate. Although the gondwanatherians don't belong in this subdirectory, I've left them there anyway, at least until better information is available.

Gondwanatheria
http://home.arcor.de/ktdykes/gondwanatheria.htm

In the case of multis, I utterly ignore the implied restriction of the term Mesozoic, and follow them to their demise. Given the numbers involved, it seemed pointless not to.

Multituberculata
http://home.arcor.de/ktdykes/multis.htm

TRICONODONTS

This is another rather slippery term based on dental characteristics. It's to do with three- coned molars, which is a typical enough mammalian condition. Some interpretations have included those early morganucodontids within this group. Parts of Triconodonta seem to be genuinely related, whilst others aren't.

Some of these animals were giants amongst the mammals of the time. They grew to cat, and even dog-size. Others were the more usual mouse-small and smaller. Good remains have been found, including one virtually complete representative, Jeholodens. The main habitats of triconodonts were in the northern hemisphere of the Upper Jurassic and Lower Cretaceous. Some North American representatives are known from the lower Upper Cretaceous.

I've arranged the entries into three sections:

- The first category is a hold-all for triconodonts of some kind or other. It contains bits and pieces of various groups. Most are 'amphilestids', which is a term based on some fairly bland characteristics. There is a proposed family, but it's no longer convincing.

- Some mainly Asiatic 'amphilestids' of the Lower Cretaceous are close relatives of each other. I've put them into an enclosure for gobiconodontids, (which sounds like something out of Star Wars), and Repenomamus.

- My third paddock is Triconodontidae. With a couple of possible exceptions, these animals appear to be members of the same family. The aforementioned Jeholodens is held to be at least close, if not even closer.

Some of these triconodont fossils help relate a postscript to the 'reptile-mammal' transition. When the mammalian jaw joint had become the only hinge, the 'reptilian' joint bones didn't simply disappear. They'd already been involved in the sense of hearing, and that's the career they pursued. They became the incus and the malleus; small bones inside the ear. The tiny contemporary of the Chinese morganucodontids, Hadrocodium, is the earliest known possessor of such sophisticated listening devices. However, this still left other pieces of reptilian paraphernalia lying around. There was a thing called Meckel's groove on the inside of the lower jaw. Generally, it contains a cartilage which is involved in mouth opening. Those small bones form from the end of it. With their new place of residence in mammalian ears, this link severs.

Although we no longer have Herr Meckel's most excellent groove, his cartilage boogies on. It formed our original embryonic jaw, before bone replaced it to do the job properly. We have small bits of it in our ears, but most of it was reabsorbed into the jaw and the groove never got groovy.

Repenomamus has a cool Meckel's groove. There are even remains of the fossilized cartilage. It presumably had some use for it. Those closely related Star Wars characters, the gobiconodontids, also grooved, as did other Jurassic and Lower Cretaceous triconodontids; eg. Arundelconodon. It's not known from later representatives though.

This ungrooviness seems to have taken hold in parallel among various mammal lineages, though at differing rates. Lower Cretaceous representatives of our own line, Eutheria, were also Meckelian groovers. Sooner or later, we all found it old hat and it faded away. That's fashion for you.

Most triconodonts, (including the 'amphilestids'), probably fit within Crown-group Mammalia, somewhere towards the start. Crown-group Mammalia is the most recent common ancestor of existing mammals, and all of its descendants. Amongst the more tightly defined Triconodontidae, there are a number of characteristics reminiscent of those egg-laying monotremes, though other factors don't support a relationship of any intimacy.

Triconodonta
http://home.arcor.de/ktdykes/triconodonta.htm

CROWN-GROUP MAMMALIA, THE ROAD TO THERIA

TINODONTIDS AND SPALACOTHERIIDS

These two families aren't closely related, but they were contemporaries. They're also both 'symmetrodonts'. The first mentioned are more 'obtuse angle' critters, whilst the latter are 'acute angle'-rs. They have reasonably 'advanced' lower jaws, in comparison to the old fashioned kuehneotheriids. They're also very possibly within Crown-group Mammalia.

The tinodontids were late Jurassic - early Cretaceous residents of North America, Europe and perhaps Mongolia. The jaw remains show they're more closely related to us than they were to basal mammals.

The exotic sounding spalocotheriids had much the same geographical and chronological range. It's clear that they do form a valid family. Fossils are generally limited to odd scraps of jaw and isolated teeth. However, these have enough peculiarities to show kinship. There's one superb find from China called Zhangheotherium. It's either a member of this family or closely related. In some respects, this fabulously preserved fossil's rather dull.

Up until 1997, all 'symmetrodonts' were based only on jaw fragments and teeth. Trying to reach any conclusions about the rest of the body was down to educated guesswork. A reasonable bash might have been: shrewish in size, limbs splayed out somewhat -(in line with other groups with broadly similar jaws, and unlike types with more 'modern' ones)-, insectivorous; something like that. When the virtually complete Zhangheotherium was described, the guesswork was broadly correct. About the only major surprise it offered was its excellent state of preservation. This is a bit like getting a birthday present, you desperately wanted thirty years ago. That's unfair. It's a remarkable fossil. It's been overshadowed somewhat by the even more fantastic finds from its home of Liaoning.

Comparison to a second specimen suggests that the original specimen very possibly represents a male. This is on account of a spur of bone on the ankle, which the more recently found fossil doesn't possess. In monotremes at any rate, this is a sex based difference. In general, this feature is restricted to Mr platypussies and echidnas.

Tinodontidae and Spalacotheriidae
http://home.arcor.de/ktdykes/symmetro.htm

MAMMALS AMONGST THE PALEO-LIZARDS OF AUS

The next two directories have strong Australian connections. The structure is inspired by a somewhat controversial viewpoint. As these tribosphenic teeth are so unusual, it was assumed that they could have evolved only once. A relatively recent theory says: hang on. What if they evolved twice?

A clever bit of interpretation, backed up with an immense amount of observation and testing, strongly suggests there are significant and characteristic differences between 'northern' tribosphenic teeth, (Boreosphenida), and 'southern' ones, (Australosphenida). Whilst broadly similar in function, the construction is not the same. This implies the dual evolution of tribosphenic molars. Unfortunately, I haven't got any idea whether the Dorset Tribactonodon is anatomically a 'northerner', a 'southerner' or somewhere in between. Based on common sense, I expect it'll be the former, but this is a question of anatomy. In my present ignorance, I dumped it here.

The inclusion of Shuotherium is also controversial. All lower tribosphenic molars have a grinding surface known as a talonid. Those of Shuotherium have no such thing. Instead, there's a pseudo-talonid. It's a similar sort of structure but in an unexpected position. This is situated in front of a feature known as the trigonid, which is the reverse of the tribosphenic pattern. It 'should' be behind. I'm following the notion that this Chinese (and British) creature represents a sister line of these southern australosphenidans. However, I'm increasingly impressed by another line of thinking, which links shuos with docodonts.

The third grouping on the first directory of Aus contains the down under connections, and the attractions of the dual origin theory. This section is home to the earliest known australosphenidans and presently houses four genera. Two come from the Middle Jurassic of Argentina and Madagascar. The other couple are from the Lower Cretaceous of Australia. This pair is also a matter of debate. An alternative view is that they're early placental mammals. This is partly based on the dental formula and some similarities of the jaw construction with hedgehogs. I'm presently more heavily influenced by the dual originist view. This interprets Australosphenida as including Monotremata; the duck-billed platypus and echidnas. However, subsequent research suggests that monotremes don't, and never did have tribosphenic molars. If correct, than they can't be australosphenids.

One reason why there's so little consensus on these issues is that there are so few fossils. Until the last few years, very little Mesozoic mammal material had been recovered from the southern hemisphere. This is beginning to change.

Tribactonodon, Shuotheriidae and Australosphenida
http://home.arcor.de/ktdykes/australo.htm

MESOZOIC MAMMALS? THE PLATYPUS AND FRIENDS

I had a lot of fun compiling this next directory. It starts off amongst the usual scraps and teeth of the Mesozoic. The fossil record of long dead relatives of the platypus isn't extensive, but it's slowly improving. One unexpected novelty is provided by a couple of slightly post Mesozoic teeth. Monotremes are confined to Australasia. There was no record of their ever having had a broader range. This changed in 1992 with the publication of Monotrematum, which turned up in Patagonia. Increasing numbers of paleontologists enjoy working there.

Whilst this find was surprising, the presence of platy-ancestors in South America is in line with the geography of the time. It was linked to Australia via the Antarctic. That's the route marsupials used to reach Aus. Although 98% covered by ice, Antarctica boasts a number of extinct marsupials. It's even yielded bits of an Eocene sloth. Apart from being larger, these Patagonian teeth are similar to those of the later Australian, Obdurodon, which was clearly a close relative of the modern platypus. The quantity of web material available on the platy is predictably enormous.

There are a couple of other egg-laying mammals, though they're not as famous. These are the echidnas, which look something like large hedgehogs with a touch of anteater. One of the first Westerners to meet an echidna was Captain Bligh, shortly after his enforced departure from HMS Bounty. He made a sketch before dinner. Echidnas are very easy to catch, tasty and have an entertaining sex life.

As with the multituberculates, I strayed way beyond the Mesozoic with this group.

Monotremata
http://home.arcor.de/ktdykes/monotrem.htm

DRYOLESTIDS AND COLLEAGUES

These creatures started out as a northern group of insect crunchers. There are also several African representatives from the Upper Jurassic (Tanzania) and the Lower Cretaceous (Morocco). However, the associated floral and faunal fossils are also broadly 'northern'. Probably the most peculiar feature of dryolestids was their enthusiasm for enormous numbers of teeth. Unless you've lost any, you've got two molars on either side of your upper and lower jaws. The wisdom tooth represents a third, and that's as many as any placental. Marsupials luxuriate with up to four. Dryolestids wouldn't have dreamt of leaving home with less than seven, and a few even went as far as nine. That's on each side.

Collectively, dryolestids, paurodontids and the like, (plus critters such as the prototribosphenidans, below), are also known as eupantotheres. Knowledge is usually restricted to jaws and teeth, which are generally Upper Jurassic - Lower Cretaceous. A spectacular exception is the near complete paurodontid, Henkelotherium, which is from Guimarota in Portugal. (Paurodontids were somewhat less enthusiastic in their taste for molars, though they were also pretty toothy.) The skeleton suggests a life in the trees. The construction of its legs is more 'advanced' than is the case for the living monotremes.

Freaky dryolestids took part in the 'Jurassic Revival Festival' of Upper Cretaceous Patagonia. Some of these ravers made it into the Paleocene. Quite where these lines were hanging out for much of the Cretaceous isn't known. Suitable southern hemisphere fossil sites are doubtless out there somewhere, so please go and find them.

Dryolestida
http://home.arcor.de/ktdykes/dryolest.htm

PROTOTRIBOSPHENIDANS AND THE LIKE

Prototribosphenic mammals were those with nearly tribosphenic molars. That's those grinding- cutting teeth mentioned above. The ancestors of the northern tribosphenidans would have been amongst them, and they probably looked quite like a genus called Peramus, from the Lower Cretaceous of Dorset, England. It's amongst the most basal looking members of a grouping called Zatheria, which begins with the most recent common ancestor of Peramus and myself. Of course, the ancestor would have lived earlier, and stem-zatherians have now been identified from Upper Jurassic Guimarota, Portugal.

As well as dental developments, there was some significant change underway in the inner ear, as revealed by the Lower Cretaceous South American, Vincelestes. Back at Morganucodon, I mentioned an elongated cochlear canal. That's the thing sound waves travel along. In ourselves and marsupials, this canal is significantly longer. It's coiled. In the case of Vincelestes, this already exceeded 270°, which is a more pronounced coiling than is known from those egg-laying monotremes. They've also further enhanced their hearing abilities, but they used a different trick. Rather than lengthening the bony canal still more, monotremes have worked on their aural membranes. Amongst other things, this might suggest that these prototribosphenidans were effectively proto-northern-tribosphenidans. The southerners had branched off long before.

Stem-zatherians, zatherians and Peramuridae
http://home.arcor.de/ktdykes/cladoth.htm

BOREOSPHENIDANS AND THE EMERGENCE OF THERIA

This is where the northerners finally catch up with their dentally sophisticated cousins. The southerners had already invested in high tech, dual purpose molars back in the Middle Jurassic. Our ancestors didn't get with it until the Lower Cretaceous, (or perhaps somewhat earlier). Maybe they'd been too busy working on their bony cochlear canals.

This is also an area of intermingling shades of grey. People like tidy stories; beginning- middle-end. With that in mind, from the boreosphenidans emerged the therians, ('beasts'). The therians branched. One line carried on for a while as therians. A second evolved into eutherians, (placentals plus a bit). A third led to metatherians, (marsupials plus a bit. Did I mention marsupials were originally northerners?).

The boreosphenidans are known from the lowest of the Lower Cretaceous. As is to be expected, the therians, basal eutherians and basal metatherians were all broadly similar. There are clear diagnostic differences between placentals and marsupials. However, these distinguishing characteristics crystallized over many generations. This can make recognizing early representatives extremely difficult, especially when most of the material available is fragmentary.

Boreosphenida
http://home.arcor.de/ktdykes/boreosphen.htm

Theria
http://home.arcor.de/ktdykes/tribo.htm

Repetition for the sake of clarity. Basal therians + the eutherians (= placentals plus) + the metatherians (= marsupials plus) = Theria.

MESOZOIC MARSUPIALS

Until December 2003, the known fossil record contained a eutherian which was 25 million years or so older than the earliest known metatherian. There's just been an equalizer. A beautiful specimen from Liaoning, China called Sinodelphys has dramatically filled a gap in the ancestory of marsupials. It's 125 million years old and also now shares the record for the earliest direct evidence of fossilized fur. If anyone would care to uncover an earlier, undoubted therian this would be most welcome.

The cast list of Mesozoic metatherians is composed of opossums and proto-possums. Cuddly kangaroos and sex-mad, male koalas arose much later. It also begins in their nurseries of Asia and North America. The only evidence from the southern hemisphere is an Upper Cretaceous tooth from Madagascar. (Update: I'm just printing out a new paper. This tooth may not be from a marsupial.) I've heard of South American sites, but they generally turn out to be slightly post-Mesozoic. However, given the diversity of Paleocene marsupials of that continent, there's a reasonable likelihood that some Cretaceous representatives were endemic.

The first directory provides sanctuary for 'basal' metatherians, deltatheroidans and asiadelphians. the 'basal' members the aforementioned Sinodelphys and the earliest North American representative, Kokopellia. This genus is based on some bits of 100 million year old jaw from Utah and Oklahoma. They're accompanied by isolated Upper Cretaceous, North American teeth.

Deltatheridium is a reasonably well preserved creature from the Upper Cretaceous of Mongolia. It has some marsupial characteristics; eg. part of the dental formula. However, it lacks other marsupial details. As a consequence, it's been interpreted as both a basal boreosphenidan, (albeit a late one), and a basal metatherian. However, recent research has confirmed its placement within Metatheria.

The best known asiadelphian is Asiatherium from the Upper Cretaceous of Mongolia. Unfortunately, I don't yet have much information available on these doubtlessly charming critters.

'Basal' metatherians, Deltatheroida and Asiadelphia
http://home.arcor.de/ktdykes/basmeta.htm

The second instalment concerns mainly North American metatherians of the Upper Cretaceous. The first group's called Stagodontidae. They seem to have been brought to an end by the K-T extinction events which accounted for the last dinosaurs, (non-birdy). One of their number, Didelphodon, is among the largest known of Mesozoic mammals. It was roughly badger-sized, and managed to negotiate a cameo role as an egg eater in the last episode of Walking with Dinosaurs.

They're partnered by their friends and relatives, the pediomyids. These are also unknown in the north beyond the Cretaceous. However, as one of the family, Monodelphopsis, has turned up in the Paleocene of Brazil, they seem to have been involved in spreading the marsupial message to the south.

Stagodontidae and Pediomyidae
http://home.arcor.de/ktdykes/stago.htm

The final Metatheria directory features 'basal' Didelphidae. These are full-blown opossums and their origins also seem to be in the Upper Cretaceous of North America. They became very cosmopolitan. There's little to indicate that they fared any better than the stagodontids in the north. However, like the pediomyids, they also spread south. Later, the Virginia possum returned to its roots. It's also in this family.

As is Peradectes. I'd be surprised if all the reports of this genus are accurate. It's perhaps known from the Cretaceous of both Americas, (though I have my suspicions concerning the correctness of this). It has most definitely been recovered from the Eocene of Europe and possibly Africa. Peradectes is one of several marsupials from the fine fossil sites of Germany, (eg. Messel near Darmstadt).

In view of the diversity of post-Mesozoic Marsupialdom, I felt compelled to more or less stop at the Mesozoic, which is in many ways a pity.

'Basal' Didelphidae & Co
http://home.arcor.de/ktdykes/boreometa.htm

MY ANCESTORS

THE DAWN MOTHERS

Eutheria contains all placental mammals and our immediate ancestors. As with Metatheria, the oldest and most basal known representative kicks this group off with a bang. It's another absolute stunner from Liaoning, China.

Eomaia was described in 2002. Its name means 'dawn mother' and it's about 125 million years old. If anything earlier ever turns up, I suppose it'll have to be called 'night-before granny' or some such. As it's exquisitely preserved, it also substantiates a number of details concerning early eutherids.

It was reasonable to assume that all mammals must have had at least some hair, even though such a feature almost never fossilizes. 'Almost never' is a synonym for 'occasionally does'. The oldest direct evidence was provided by a 60 million year old, Chinese multituberculate named Lambdopsalis. Eomaia doubled that figure. Extensive traces of fur and finer hairs are well preserved.

Where evidence exists and with the general exception of placentals, all mammals have or had epipubic bones. These stick out from the pelvis and are also known as marsupial bones. Mrs Kangaroos use them as pouch supports. They probably started out as leg muscle anchors in pre-mammalian days. (They're known from those strangely named tritylodontids. Most mammals today have no such things, but that's because they're placentals. Eomaia had them, as did various basal colleagues from Mongolia, (Lower and Upper Cretaceous).

Another detail, which had already been observed in several of these Mongolians, concerns the teeth. Those between the canine and the large cheek teeth are called premolars. The maximum count for all existing placentals is four per side. As some basal, but clearly eutherian Mongolian fossils had five, it was reasonable to assume, that the earliest representatives would have had at least that number. Eomaia has five.

The pelvis of Eomaia is relatively narrow. This suggests live birthing, but with poorly developed babies, a pattern known from marsupials. As the babies would have been small, they wouldn't have enjoyed a prolonged residence in the womb, which means they'd have had no use for a placenta. This suggests that Eomaia, though certainly a eutherian, was a kind of pre-placental placental mammal. As this method of reproduction was followed, or is generally followed, by Eomaia and marsupials, it's reasonable to say this must have been the original, ancestral pattern; the therian method. Placental assisted birthing developed later, (as well as independently and differently amongst a few marsupials).

As far as Herr Meckel might be concerned, Eomaia was certainly a groover.

This is not only one of the most astonishingly beautiful fossils ever found. It's also very obliging in the way it conforms to and confirms many expectations. That's why I've given it these few hundred words.

More may be enjoyed or jeered at on the first Eutheria directory. It's a collection of basal representatives, hard-to-placers and odd bits and pieces. It also contains some info on the earliest North American eutherian, Montanalestes, and a photo of a particularly cute European representative, (Self MY).

'Basal' Eutheria
http://home.arcor.de/ktdykes/baseuth.htm

THEIR CHILDREN

As far as I understand the concept, (which isn't necessarily any further than the distance to the nearest door), Epitheria is something akin to all eutherian mammals except for the most basal ones, and the line which led to the predominantly South American near-toothless wonders of Xenarthra, (the aforementioned edentates). However, I've decided to abandon attempting to use the category and have adopted a looser approach. All the following are reasonably described as small insectivores.

The first section is a basket of critters which might belong elsewhere. As is so often the case, interpretations vary. It's a small Asiatic collection, ranging from the Lower Cretaceous to the Upper. It's the sort of section I enjoy trying to close down, as and when possible. The second group is Asioryctidae, which comprises a couple of reasonably well-known genera. They're possibly related to the zalambdalestids on the final directory. Finally, we have the gypsonictopids; an exciting sounding group. Prokennalestes is another Lower Cretaceous eutherian, whilst its companions are from the Upper Cretaceous of Asia and North America. The structure of the page is fairly described as liable to subsidence.

Basal Eutheria Two
http://home.arcor.de/ktdykes/basepi.htm

OF HOOFS AND DOLPHINS, ELEPHANTS AND FINS

There are differing views on the scheme I'm following for this next collection, but it provides a very convenient structure, a fine narrative context and some great links. It's one interpretation and reasonable alternatives are certainly possible.

Ungulates are more generally understood to be animals with hoofs. There are odd-toed ungulates, (eg. horses), and even-toed representatives, (eg. sheep). Ungulatomorpha unites all imaginable ungulates together with some more surprising kith and kin; elephants, sea cows, aardvarks and whales count amongst the latter. The thinking is that all these animals have a common ancestor, and possible candidates have been dug up in the Upper Cretaceous rocks of Central Asia, Europe and North America. Whilst there's enough evidence to support a 'maybe', I wouldn't wish to put it more strongly than that. This centres on things often called zhelestids, which may, at least in part, be closely related with the zalambdalestids, (below).

Perhaps surprisingly, the controversy doesn't involve the connections between whales and ungulates. That's more a matter of which ungulates. The extinct, carnivorous and terrestrial mesonychids are favourites, whilst proto-hippos also have some support. There's also a further candidate for the mum-gulate; Protungulatum of the Upper Cretaceous- Paleocene of North America. It's the placement of the slightly earlier fossils that's less clear. They're perhaps best described as 'could be-s' in terms of proto-hoofers, but they might represent extinct lineages. Or something else.

'Basal' Ungulatomorpha
http://home.arcor.de/ktdykes/ungulato.htm

PROTO-MICE AND PROTO-CATS?

The inhabitants of this final directory are also more tidly arranged than they might deserve. I've followed a couple of possible scenarios.

There is some evidence that zalambdalestids, (further furry inhabitants of the Central Asiatic Upper Cretaceous), could be the forerunners of Glires, which is composed of rodents and bunny rabbits. All members of Glires share a suite of dental characteristics, some of which are also evident in the thrilling sounding zalambdalestids. This could be a matter of ancestry. I've gone along with this 'could', but don't intend it as meaning 'is'. They look like perhaps ancestors.

I've followed a similar line with cimolestids, though I doubt that many of the included 'members' ever paid their joining fees for this family. The version of Cimolestidae I've adopted looks suspicious. Nevertheless, Cimolestes has been accused of being the inventor of all cats; and even of all carnivores. Given that the forerunners of all existing mammals were small, squeaking insect-eaters, it might be guilty. If not, something broadly similar was.

Zalambdalestidae and Cimolestidae
http://home.arcor.de/ktdykes/zalambda.htm

FINAL CAUTION

As this is intended as a rough and not necessarily reliable sketch, I'm not including any references, but am indebted to many sources. These are listed in the bibliographies of the individual directories.

My reasoning is as follows: people with less knowledge of this subject than I have, might gain the impression that I'm some kind of expert. Those with more knowledge can assure them that I'm no such thing. My knowledge is superficial. If I listed references, it might make this essay look like some kind of academic endeavour. It's not. It's a hobby piece for my own interest. Mind you, I take hobbying seriously. This overview may also sometimes slip out of kilter with the directories, which are being constantly updated.

Trevor Dykes (not a paleontologist)
March 2003
Last update 25.4.2005
Ktdykes@arcor.de


"That article was interesting. Are there any more on-line?"
I'm pleased you asked. Have a look here.