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.

This directory is now one of three containing triconodont mammals and several genera with no clear affinities. Hopefully, the reconstruction hasn't involved inadvertent destruction. This particular page contains various odds and sods and 'amphilestids'. Those searching for gobiconodontids and allies (including Repenomamus the Magnificent) should file along to the gobicondontid enclosure. Should, however, your tastes desire triconodontids, then feel free to visit the triconodontid playground. There were two main reasons for this reconstruction. Firstly, there was more desire to cause mischief and confusion. Secondly, all groups were previously contained on a single directory, and it became vast. Tricondont mammals Triconodonta Osborn, 1888 is a group of mouse to cat-sized insectivores and carnivores. In contrast to more basal lineages, they had the typical mammalian arrangement of one lower jaw bone, (the dentary), and three bones in the middle ear, (incus, malleus and stapes). However, they retained various old-fashioned touches. For example, the pelvis and rear legs displayed a more 'reptilian' posture, than is the case with any existing mammals. The molar teeth, which give the group their name, were relatively simple constructions. A somewhat tangled past Over the years, Triconodonta came to include Morganucodontidae, Sinoconodontidae, Megazostrodontidae and etc. The dubious legitimacy of that was summed up quite nicely by this observation from David Marjanovic (pers. comm., 2001); "My canines are also triconodont, and more beautifully triconodont at that, that's the plesiomorphy." As arranged on this page, this taxon has some validity, which is none of my doing. A brief historical sketch Triconodonta was established as an order for Mesozoic mammals with molars featuring a straight line of three main cusps. What couldn't then be appreciated is that this basic characteristic dates back to at least the Lower Triassic. Thrinaxodon has teeth like that, and it's not quite a eucynodont. As a primitive feature can be maintained in disparate lineages, it's not a secure sign of particular affinities. A somewhat narrower concept of Eutriconodonta may be more useful. This presently provides a refuge for the families of Amphilestidae, Gobiconodontidae (including Repenomamidae) and Triconodontidae. However, whether all these groups are close relatives is far from sure, (Kemp 2005, p.150-152 provides a useful outline, which I consulted). Various other animals have been shoehorned in with the amphilestids, eg. Austrotriconodon. It shouldn't be included, but that's nevertheless a better place than anywhere else presently available. A guest star Fruitafossor also features on this page, but doesn't belong here. It's affinities are presently unresolved. Rather than giving the genus its own directory, I'll let it hang out with what might be some of its relatives. Wracking brains The following paragraph might have fallen into redundancy! One possible clue regarding the relationship of triconodonts to other mammalian groups is in the structure of the brain, where known from casts. It's classed as cryptomesencephalic: "characterised by an expanded vermis, no cerebellar hemispheres, and lack of the dorsal midbrain exposure." This is a feature shared only with multituberculates, and not with therians, 'proper' mammals. Their brains were apparently originally eumesencephalic: "characterized by a wide cerebellum with extensive cerebellar hemispheres and large dorsal midbrain exposure," (quotations from Kielan-Jaworowska & Hurum 2001, p.394). New thinking on brains A newer study, (Kielan-Jaworowska & Lancaster, 2004), may well have put the kybosh on that. It concerns endocranial casts (brain prints) of the multituberculate Kryptobaatar. While lying around dead, sand found its way into the skulls of several specimens. As it packed in around the brain, details were left reflected in the material. Walk in the sand and you'll leave footprints. Get sand in the skull and the result may well be an endocranial cast. Either case can provide information on anatomy. However, your footprint isn't a faithful representation. At the very least, it'll be bigger than your actual foot. And the tracks you leave in squidgy sand will differ to prints in dry sand. Walk in fine sand and the result may well be a relatively uninformative mini-crater. Should you be destined for the beach next summer, you can conduct your own experiments. Such distorting factors can be misleading. In previous studies of endocasts, it had been concluded that there were significant differences in brain construction of triconodonts and multis on the one side, and therians on the other: "Multituberculate and eutriconodontan endocasts differ from those of primitive therian mammals in their lack of visible midbrain exposure on the dorsal side and in having a vermis-like bulge (recognized herein as the cast of a large sinus - the superior cistern) inserted between the cerebal hemispheres", (p.177). A difficulty with interpreting brain prints of long dead animals is that they can't be compared directly with the brain responsible. In correspondence with the lead author, (p.178), HJ Jerrison pointed out that the apparently 'enlarged vermis' can also be observered in koalas, (don't ask how he did that), but this bulge is actually the result of the superior cistern. Consequently, the authors reconsidered the situation in multis and Triconodon. With the benefit of new specimens, they concluded that previous interpretations may well be incorrect. "In the light of the present interpretation of multituberculate and eutriconodontan endocasts the terms cryptomesencephalic and eumesencephalic brains should be abandoned." (I've decided to leave them here because: a. they are in the literature; b. interpretations might change again; c. they might still be of use in Scrabble.) In short: the brain structure doesn't seem as different from that of therians as previously thought. This may suggest a closer relationship, (p.179). Link: Kielan-Jaworowska & Lancaster, 2004, Acta Palaeontologica Polonica http://app.pan.pl/acta49/app49-177.pdf The full paper is freely accessible on-line. There are similarities in features of the foot structure between triconodonts and multis. If common dental details could be cited too, this would be most convenient. The difficulty is that, up to now, they can't be, which leaves things very inconclusive.

Links: Toby White, Palaeos, Triconodonta http://www.palaeos.com/Vertebrates/Units/430Mammalia/430.400.html "However, the most recent analysis gives some reason to believe that there is a solid core of truth to the Triconodonta. Luo et al. (2002) find surprisingly robust support for at least some kind of Triconodonta -- some clade of triconodonts more inclusive than just the Triconodontidae. Since the deconstruction of this taxon may have been premature, it will be treated as a clade for the present." Once broadly considered a decidedly dubious taxon, Triconodonta refuses to be so easily dismantled. However, Amphilestidae is no longer seen as a proper clade, (eg. Rougier et al 2001, p.5). They cite it as paraphyletic; ie. not containing all the descendants of a common ancestor. Mikko K. Haaramo, Triconodonta Mikko Haaramo's Triconodonta Mikko K. Haaramo, Amphilestidae Mikko Haaramo's Amphilestidae David Atkins, Origin of the Modern Therian Mammals http://gwu.edu/~atkins/newwebpages/Mammalia/primitive.primmammals.html A concise look at basal mammals, including Docodonta and Triconodonta.

Fruitafossor A. Triconodonts B. 'Amphilestids'

FRUITAFOSSOR & Co Remarks: This section is presently housing a mammal with no particular home in my directories. This here Fruitafossor was previously accompanied by the even more extraordinary Volaticotherium but, as that now seems to be some of 'amphilestid' triconodont, it's been moved to the appropriate enclosure. Time-Line: Upper Jurassic: Fruitafossor

Genus: Fruitafossor Luo ZX & Wible JR, 2005 'Fruita digger' Remarks: The first part of the generic name refers to Fruita, a town in Colorado. 'Fossorial' is a reference to digging abilities.

Species:	Fruitafossor windscheffelia Luo ZX & Wible JR, 2005
Place:	(Burro) Morrison Formation, Colorado
Country:	USA
Age:	Kimmeridgian, Upper Jurassic
Remarks:	The following is based upon my reading of Luo & Wible, 2005. Lots of mammals make part of their living by looking after ants and termites, but some are particularly adapted for this challenge. Echidnas and anteaters take a toothless approach to the subject. They use their long, sticky tongues to harvest prodigious quantities of prey. Armadillos meet the same objective by having simplified teeth in large amounts. (A large armadillo mouth can possess up to a hundred.) Being small, (the lower jaw's about 1.2cm in length), Fruitafossor wouldn't have needed to process as many victims, but it came supplied with appropriate specialisations. In form, the teeth are broadly similar to their equivalents in armadillos and aardvarks, (p.103). They're very different to the more generalised dentitions of their Jurassic contemporaries. The excellently preserved front legs are adapted for digging. The lumbar vertebrae have an unusual double articulation, similar to the scheme found in modern xenarthran spines, (armadillos, anteaters and sloths.) With the exception of Fruitafossor, such a characteristic is presently only known from the placental order of Xenarthra. So this is a Jurassic xenarthran? Categorically not. The similarities are a matter of convergence; similar characteristics arising separately to deal with the same kind of challenges. The unusual articulation system of the lumbar vertebrae is xenarthran-like. (The authors refer to it as xenarthrous and not xenarthran.) Fruitafossor was far more distantly related with an armadillo than I am. This was a Jurassic eater of ants and termites, not an extremely early, small anteater. The genus represents a previously unknown line of mammals, which implies that close relatives haven't yet been discovered. Despite the structure of the vertebrae, Fruitafossor has no particular bearing on the origins of Xenarthra. Significant differences far outweigh the similarities. Unusually good for the time of Earth Upper Jurassic mammals are generally known from teeth. Sometimes, these may be found with bits of jaw. This specimen is much more generous. The lower jaws are almost fully preserved, as are parts of the uppers and skull. About 40% of the skeleton is available. A complete front leg and paw is an extreme rarity, and parts of the rear legs are present as well. Substantial remains of the spine and shoulders are also represented. For lovers of teeth The known dental formula per side is: (lowers): three incisors, one canine, three premolars and three molars; (uppers): ?, one, three and three respectively. The number of teeth is relatively low for mammals of this age, but it was an unusual critter. Distinctive molars and ancient origins The molars are completely unlike anything ever seen before in a Mesozoic mammal. They are tubular and lack enamel. Each has a single, open-ended root. Open-ended roots are associated with teeth which continue to grow throughout a life time. The absence of enamel meant they were relatively soft, and erosion ensured the owver didn't become too long in the tooth. These teeth are elliptical in cross-section, and have strong similarities with those of extant armadillos and, (to a lesser degree), aardvarks. Some of the molars of Tertiary palaeanodonts were also similar, (p.104). This suggests the same sort of diet; a fondness for ants, termites and other creepy crawlies, perhaps supplemented by small vertebrates and bits of vegetation. Remains of nests attest to the presence of at least four species of termites in the Morrison Formation, and the same applies for ants, (Hasiotis, 1999). Many other characteristics of the animal show this genus is more basal than therians, trechnotherians or even theriiforms, (trechnotherians and multituberculates). It represents a lineage with an earlier origin than any of those mammals. Digging for victory Should you feel a hunger for ants, then digging would be a very useful ability. Rather than relying upon ants you happen to see wandering around, this skill allows you to go for the nest. The front legs and paws point to a talented digger. The construction of the shoulders, (which involves an ancient bone called a caracoid), indicates a capability for movements more in line with monotremes than moles. The scapula is broadly reminiscent of the equivalents in monotremes, Morganucodon and Haldanodon. The humerus is built to anchor a large teres muscle. Lower down the arm, the ulna could probably have boasted to Popeye about its impressive triceps and further muscles. (I read the Carnegie Museum press release the day after I wrote that sentence - see the link below. It turned out that the fossil is nicknamed Popeye.) Hands of a termite terminator When it comes to the number of fingers, Fruitafossor isn't as primitive as I am. I favour the antiquated count of five. This genus had simplified things down to four. With the exception of the innermost finger, (known as digit 1 or the thumb), the carpals, metacarpals and phalanges are short and stumpy. (Digit 1 is somewhat more gracefully built). The claw-bearing phalanx is the longest bone in each instance, and its end is widened and flattened for shifting soil. Such robust bones are typical for mammals specialised for digging. Distinctions of the paw Although some characteristics of the paw are familiar, others are distinctive. Metacarpal 1 has a similar width to that of the rest, and the relative ratios of width and length are contrasts between this taxon and monotremes, (p.105). The stockiness of the metacarpals and phalanges differ from the more slender build of eutriconodonts, multis and trechnotherians. An archaic upper arm "The humerus lacks a distinctive humeral head but has a broad intertubercular groove; the two distal humeral condyles for the radius and ulna are spherical, widely separated, and arranged obliquely." These characteristics are found in non-mammalian tritylodontids, tritheledontids and basal mammalian morganucodontids. Wrists This area of the body resembles the states found in eutriconodonts (this directory), multis and Zhangheotherium, as the proportions of bones called the hamate, lunate and scaphoid are similar. Metatherians have a hamate which is much enlarged, (hypertrophied). On the other hand, eutherians favour an enlarged trapezium, (p.106). Neither features in the arsenal of Fruitafossor. Rib tickling If you'd care to undress and examine your chest and stomach, (or those of a willing and obliging volunteer), you should discover your ribs end about two-thirds of the way down. The lower bones of your spine are the lumbar vertebrae, and I predict that they lack ribs. The corresponding bones of this genus have: "unfused and mobile lumbar ribs down to the penultimate lumbar vertebrae..., a feature of cynodonts... and monotremes", (references omitted). Xenarthran-style vertebrae Although Xenarthra may sound like something from a science fiction book, the truth is stranger and more interesting. It's an order of placental mammals, (armadillos, anteaters and sloths as mentioned above). The name points to 'strange joints', but that has nothing to do with certain cafés in Amsterdam. These back bones have an 'extra' articulation with one another, and can be found on both the lumbar and thoracic vertebrae. In the living world, this characteristic is only found in xenarthrans. Surprisingly enough, a similar condition turned up in Fruitafossor. As the genus isn't closely related, (it's not even a therian, let alone a placental), this must have evolved independently. Lower jaw The lower jaw of all extant mammals is exclusively formed by the dentary. This isn't the case for Fruitafossor and a variety of Mesozoic lineages, (especially the basal ones). A bit behind the tooth row is an upwards projection termed the coronoid process. You've got one on each side of your face. Among other functions, it's useful for providing the lower part of the joint with your skull. This genus has a roughly triangular dent on the inner side of the bone, and it's between the coronoid process and the final molar. This is a facet for an 'extra' bone called the coronoid, (which isn't the same as the process). Should you not be able to locate your coronoid, there's no need for concern. You never had one in the first place, and that's fine for us therians. Going down towards the bottom of the jaw brings us to a groove. This is another antiquated feature, and it's known as Meckel's groove. There's no mature mammal on the planet with anything like that today, but they're popular with other vertebrates. In this case, it's fairly broad at the back, (which is slightly behind that coronoid facet), but it tapers quickly and is little more than a slit when level with the second molar. The authors interpret the breadth of this groove as suggesting some middle earbones, (the incus and malleus), were still attached with the dentary. That wouldn't necessarily mean they must've formed part of the jaw. Affinities Poorly known. Analysis of a matrix of data on 422 morphological characters from 96 taxa, (including some non-mammals and extant genera), placed Fruitafossor between Monotremata and Eutriconodonta. Holotype LACM 150948 is a resident of the Natural History Museum, Los Angeles, (p.107). The specific names honours the discoverer, Wally Windscheffel. Acknowledgements Thanks are due to a number of posters to the Dinosaur Mailing List. We'll let Tim Williams represent them, seeing as he provided the first posting, (see the links below). As the publication appeared on the 1st of April, I was somewhat sceptical of the initial mention, but I couldn't find a joke. There wasn't one. Further thanks go to Serdar Mayda, Jeff Hecht and Micky Mortimer.
Reference:	Luo & Wible (2005), A Late Jurassic digging mammal and early mammalian diversification, Science, 308, p.103-107.

Links: Carnegie Museum of Natural History, Press Release 31.3.2005 http://www.carnegiemnh.org/news/05-jan-mar/032405popeye.htm "Researchers discover new species of mammal with unique features otherwise only known to appear 100 million years later..." Estimates of body length and weight are given as 15cm and 30g. It's also mentioned that the fossil was found in 1998. The 'Available Images' are well worth a look. Pittsburgh Post-Gazette, New find suggests more mammals lived among the dinosaurs by Byron Spice http://www.post-gazette.com/pg/05091/481073.stm "Wally Windscheffel doesn't recall whether he or his pal Charles Safris saw the gray, fist-sized rock first. It might have just been a rock, after all; only the black speck on one end suggested it was something more..." This is both an accessibly and nicely written report, which pays due attention to the importance of volunteers. You don't have to be a highly trained specialist to make significant discoveries. The required qualifications are interest, persistence, common sense and luck. CNN, Fossil found of termite-eating mammal http://www.cnn.com/2005/TECH/science/03/31/mammal.termites.reut/index.html An April Fools' Day story which wasn't a hoax. Dinosaur Mailing List, Tim Williams, 31.3.2005 http://www.cnn.com/2005/TECH/science/03/31/mammal.termites.reut/index.html Fruitafossor - new Mesozoic mammal. The resultant thread contains much interesting info and discussion.

Fruitafossor A. Triconodonts B. 'Amphilestids'

A. TRICONODONTS

Taxon: within Triconodonta Osborn, 1888 This section's somewhat messy in terms of systematics. It houses the triconodonts which don't fit into the subsequent sections. Some may be allied with 'amphilestids' on account of broad similarities, but these could also be shared primitive retentions. Genera: Acinacodus, Austrotriconodon, Ferganodon, Jeholodens, Klamelia, Kryptotherium, Yanoconodon other reports Time-Line: Upper Cretaceous: Austrotriconodon Lower Cretaceous: Acinacodus, Jeholodens, Kryptotherium Upper Jurassic: Comodon, Tendagurodon, Triconolestes, other reports Dorset Middle Jurassic: Ferganodon, Klamelia?

Genus: Acinacodus Lopatin AV, Maschenko EN & Averianov OA, 2010 Family: Amphidontidae Simpson, 1925 'Akinak tooth' Remarks: As will be obvious to everybody, the generic name is based upon the Latin word for an 'akinak' and the Greek for 'tooth'. What d'you mean you don't know what an akinak is? Then you're obviously not a member of the Tagar. This was a short form of Scythian sword favoured by them.

Species:	Acinacodus tagaricus Lopatin et al, 2010
Place:	Shestakovo 1 locality, western Siberia
Country:	Russia
Age:	Lower Cretaceous
Remarks:	I have the English language edition of the paper, which isn't quite the original citation. That's just as well seeing as Russian is all Greek to me! Anyway, the English version is Lopatin et Al (2010), A new genus of triconodont mammals from the Early Cretaceous of western Siberia, Doklady Biological Sciences, 433, p.282-285. Thanks are due to Alexander. Holotype PIN 4463/5 is part of a right dentary now at home in the Borissiak Paleontological Institute in Moscow. It was harvested by EN Maschenko in 2009. The specific name honours the Tagar, a human culture in that area of Siberia.
Reference:	Lopatin AV, Maschenko EN & Averianov AO (2010), Russian language (A new genus of triconodont mammals from the Early Cretaceous of Western Siberia), Doklady Akademii Nauk, 433(6), p.846-849.

Genus: Austrotriconodon Bonaparte JF, 1986 'southern Triconodon' Family: Austrotricondontidae Bonaparte JF, 1992 This genus is descriptively rather than taxonomically triconodont, (Cifelli et al 1999, p.202). Further thoughts come from Kemp, 2005 (p.152). The molars have a line of three main cusps and are triconodont in that sense. The central cusp is overwhelmingly dominanat. As far as it goes, (not very far), that's more in keeping with amphilestid rather than triconodontid habits. It's less than likely this genus has relevant affinities with any other known line of triconodonts. More informative comparisons would require comparable fossils.

Species:	Austrotriconodon mckennai Bonaparte JF, 1986
Place:	Los Alamitos Formation, Patagonia
Country:	Argentina
Age:	Campanian, Upper Cretaceous
Remarks:	A couple of premolars have since also been referred to this species. They live at the Museo Argentino de Ciencias Naturales Bernardino Rivadavia in Buenos Aires.
Reference:	Bonaparte (1986), Sobre Mesungulatum houssayi y nuevos mamiferes cretácious du Patagonia. Actas IV Congreso Argentino de Paleontologia y Biostratigrafia, 2, p.48-61.

Link: I Congresso "Osvaldo A Reig", 13-17.3.2002, Buenos Aries http://www.caece.edu.ar/fundacionhn/Investigacion/fa010.htm The congress was dedicated to the memory of Osvaldo Alfredo Reig (1929-1992). Abstract 28 of the Communiciones Libres is by Laura Chornogubsky. It contains some information on the newly referred premolars, (Spanish).

Species:	Austrotriconodon sepulvedai Bonaparte JF, 1992
Place:	Los Alamitos Formation, Patagonia
Country:	Argentina
Age:	Campanian, Upper Cretaceous
Remarks:	And, as with A. mckennai, a premolar housed in the same collection has subsequently been referred to this species.
Reference:	Bonaparte (1992), Una nueva especie de Triconodonta (Mammalia), de la Formación Los Alamitos, Provincia de Rio Negro y comentarios sobre su fauna de mamiferos. Ameghiniana 29(2), p.99-110.

Genus: Ferganodon Martin T & Averianov AO, 2007 'Fergana tooth' Family: Klameliidae Martin & Averianov, 2007 Remarks: The description has arrived (thanks go to Lisbon), and further information will be added at some stage (note from 4.5.2007). The time is nigh. The generic name includes a homage to the Fergana Valley, seeing as it was kind enough to house the relevant locality.

Species:	Ferganodon narynensis Martin T & Averianov AO, 2007
Place:	Balabansai Svita
Country:	Kyrgyzstan
Age:	Callovian, Middle Jurassic
Remarks:	I wrote the following paragraph based only upon the abstract of the paper. Looking at it now, some months later, it seems to be a reasonable enough quick summary, and I may as well let it remain. That's bound to result in a bit of repetition but, if the universe were only energy then I guess that'd be no matter (as in it doesn't matter). Klameliidae As well as establishing a new genus for one lower molar, the authors scoured the globe for traces of its relatives. The call of this quest was answered by the previously lonely Klamelia of Inner Mongolia. Indeed, such affection did the two genera display towards one another, that they decided (after taking advice) to found a new family together. Their wider affinities may well lie with the gobiconodontids, an accusation which is not new. According to the abstract, these klameliid lower molars (or molariforms in the terminology employed) are parallelogram-shaped things, the cusps e-d-f 'imbricate' rather than 'interlocking', there's an odd cusp to the rear of the labial end of a cingulum, and the enamel of the labial side of the tooth is vertically folded. The following is based upon my reading of Martin & Averianov, 2007, and thanks have winged their way to Lisbon. In the north of Kyrgyzstan's Fergana Valley lies the town of Tashkumyr, and my atlas appears to be completely indifferent to these facts. As it's a cheap German atlas, I would expect most residents of Tashkumyr to feel the same way about that book. It does manage to find room for the Naryn River, and that runs through the relevant area. Doubtlessly, this town and its surroundings can boast of many attractions for any visiting cartographers and humans, and three of those are provided by Middle Jurassic mammal localities (p.43). Given the global paucity of fossil-bearing rock from that time for land critters, each is a most welcome addition to the inventory. So far, they've come up with twenty or so mammalian teeth, although these are mostly fragmentary. Most common are the docodonts. There are at least two genera of these surviving scallywags from the fairly early days of mammalian evolution; Tashkumyrodon and an undescribed friend closer to Dsungarodon from Inner Mongolia. (A count of three or four species is mentioned on page 47.) Also present is at least one eutriconodont; Ferganodon. It's been given a berth in the newly established family of Klameliidae. Family familiarities Two klameliids have now been named, Ferganodon and the earlier described Klamelia from China. Some characteristics are share with the gobiconodontids, and they help explain why Klamelia has sometimes been referred to that family; for example, by its original authors. Both groups have lower molars with crown height and length close to being equal; cusp b (if present) and c are: "... diverging from rather than parallel to the central cusp a (in lateral view)..." (The partial quotation arises from my difficulties with understanding that point.); cusp c is taller than b and its base is positioned at a greater elevation. However, klameliid lower molars have an odd, parallelogram outline. It's vaguely reminiscent to me of a poorly drawn diamond shape on an appropriate playing card, but none of the sides share the same length, Apparently, this indicates an imbrication of molars rather than the interlocking favoured by gobiconodontids, with that being based on cups e-d-f. (That bit means, for gobis, there are three relevant cusps. The d is on the back of a tooth, and it fits between cusps e and f on the following crown. That's interlocking. Somebody might have to help me out with the pleasures of klameliid imbrication.) Anyway, klameliids also have an odd molar cusp towards the back of the buccal side; "a distolabial cingulid cusp". Additionally, the enamel on that side of the crown features vertical folds. While not yet appreciating all the ins and outs of those distinctions, I'm not paid to actually do so. The main point is we seem to have a new family of smallish, eutriconodont killers, and their known range is limited to Kyrgyzstan and China during the Middle Jurassic. Jaw Attempting to reconstruct the jaw of Ferganodon wouldn't be wise, seeing as only a single tooth is presently known, so it wasn't done. Nevertheless, part of Klamelia's dentary is known, and that also exhibits gobico-like touches. Not gobico-style is the widened final premolar and large mental foramina beneath the molars. A gobico prefers more modest ones beneath the premolars. At least, so it used to seem. The size and placement of these foramina (holes) is now known to be more variable among the gobicos (including Repenomamus). Lower molars The functional main cusps for klameliids had more or less been reduced to two instead of three. The foremost of the trio, b, is small for Ferganodon and not found on any teeth of Klamelia. A cusp with that designation was reported by the original authors of that genus, but this was apparently a case of misidentification. Some gobicos experienced a similar process of reduction on at least their front molars. That could be a sign of kinship (p.45). Distinctions One difference between Ferganodon and Klamelia has just been mentioned. Cusp b is still present for this new genus. Additionally, cusp d is higher, the cingulid cusp to the rear of the buccal side better formed, cusp e is reduced and the f cusp stronger. A cingulum on the lingual side is weak and incomplete in the case of Fergan. This tooth hosts a line of five cusps running along its middle, and their notations could be off-putting for some. From front to back they're termed f, b, a, c and d. This makes more sense to a mammalogist than an alphabet fetishist, should the latter insist that sentences should always begin with an a. Indexers might work in such a manner, but authors and mammalogists pursue other objectives. The major cusp receives the letter a, stands centrally and occupies around 45% of the available floor space. It's the first among unequals. Wear has truncated its crowning glory, and also left an impressive facet on its buccal side. Cusp b performs unconvincingly at its feet and, when complete, it wouldn't have managed even half the height. The rear member of the main trio, c, has its base positioned at a much greater elevation on the crown; roughly equivalent to the halfway line for a, if it had such a thing. Ignoring f for the moment, and the rear of d (at the back of the tooth), all the cusps in this line have front and rear slopes forming sharp, pointy blades for slicing misery from the life of some unfortunate victim. This tooth isn't designed for universal animal welfare. Strengthening the line The lingual cingulum gets interrupted at the base of cusp a and further forward at b. There's an embayment at that point. This presumably accommodated a bulge on the back of the preceding tooth kindly provided by its d cusp. That would allow the basis of an interlocking mechanism to stabilize the dental row. However, such a system would have benefited from some modifications of further cusps; for example, of e and f. Sneakily jumping forward t page 47, the authors aren't suggesting this Ferganodon-stage is anything like a forerunner of a Gobiconodon-stage. Rather, the imbrication system of klameliids is likely also derived from a basal e-d-f interlocking system in a lineage which then took a different approach. In those terms it's a refinement. Sadly, I've still not caught on as to what this imbrication entails. Holotype ZIN 94214 is a lower right molar presently imprisoned at the Zoological Institute, Saint Petersburg. The specific name sings praise for the waters of the Naryn River. This is richly merited as the fossil locality is found on the left bank. Some readers may be left wondering which of the two banks is the left one. It can be distinguished very easily, as it's the bank with this fossil locality on it.
Reference:	Martin & Averianov (2007), A previously unrecognized group of Jurassic triconodontan mammals from Central Asia, Naturwissenschaften, 94(1), p.43-48.

Genus: Jeholodens Ji Q, Luo Z & Ji S-A,1999 'Jehol tooth' Family: Jeholodentidae Luo Z-X, Chen P, Li G & Chen M, 2007 Remarks: For a brief, reader-friendly introduction: "Jeholodens (...) has relatively short fingers and toes, and claws that are too broad in cross-section to have provided a particularly good grip," (Weil, 2002). This was a small ground-dweller. Jehol is an ancient name for western Liaoning.

Species:	Jeholodens jenkinsi Ji Q, Luo Z & Ji S-A, 1999
Place:	Yixian Formation, Liaoning
Country:	China
Age:	Barremian, Lower Cretaceous
Remarks:	The following is based upon my reading of Ji et al, 1999. In terms of the Upper Jurassic-Lower Cretaceous, triconodonts provided the mammalian equivalent of giant predators; terrifying monsters of up to cat-sizes. Admittedly, 12 metre long allosaurs and 25 metre sauropods probably failed to notice this, but small vertebrates had reason to stay clear. The best preserved genus is Jeholodens. However, its dimensions were rather modest. Judging by the sketch, (p.327): skull -well over a full centimetre; neck and body -about 3.5cm; tail -4cm+! Be afraid. Be very afraid indeed. The title of the description contains the word 'mosaic'. While apt, this word is almost always appropriate for fossils. The bodies of animals are mixtures of archaic features, subtle refinements and occasional elements of novelty. That's how evolution works. A couple of reasons for employing 'mosaic' in this instance are: the pelvic girdle and rear legs are relatively primitive, (p.326); and the pectoral girdle is more reminiscent of further derived mammals. In praise of preservation As a small note of appreciation for the remarkable state of preservation; the tail consists of lots of very small vertebrae. There were and are thirty. Despite the passing of 125 million years, they're all present, correct and fully articulated. The neck has seven, there are fifteen thoracic vertebrae (with ribs), and seven in the lumbar region. While these anatomical details aren't surprising, their conservation is astonishing. Also in attendance are the epipubic bones. The text to Figure 1 also contains an accessible couple of comments on the interpretation of the limbs. The rear legs are 'sprawling', whereas modern mammals have legs going directly down from the hips. While the front legs are joint on in a more familiar style, the elbows are also 'sprawling'. Teeth The dental formula per side is: (uppers): four incisors, one canine, two premolars and four molars; (lowers): four, one, two and four respectively, (p.328). As is to be expected for triconodonts, (thus the name), the molars have three main cusps arranged in a central line along the crown. The dominant cusp is the middle one, (A or a for uppers or lowers). A fourth smaller cusp is present at the back of the first two uppers and all lowers, (D or d). The dominance of the main cusp is a basal characteristic in comparison to triconodontids, in which the cusp height is more uniform. The incisors are spoon-shaped, and this is presently only known from this genus for triconodonts. The second of the lowers is a milk tooth, as a replacement was erupting behind. The m4 was also in the early stages of development. As this insectivore must have been a subadult or younger, it presumably also had a bit more growing to do. Molar alignment (Update: Doubts on this point have been expressed by Rougier, Isah & Manabe, 2007. As stated below in the additional notes, they think a measure of distortion may have changed the relative positions of the uppers and lowers. If correct, that would have cusp a occluding between A and C.) "In J. jenkensi lower molar cusp a occludes into the valley-groove between cusps A and B of the opposite upper molar, thus differing from amphilestids and gobiconodontids..., in which lower cusp a occludes into the embrasure anterior to upper cusp B", (reference number omitted). And further: "The lower molars are interlocked, with a crescent-shaped distal cusp d of the preceding molar fitted into the concave mesial margin of cusp b of the succeeding molar, which is diagnositic of the Triconodontidae." Shoulder "The dorsal part of the scapula has a prominent triangular area..., similar to the large attachment for the terres major muscle in monotremes..., a condition also present in the archaic therian Zhangheotherium. The clavicle is curved. Its lateral end has a tapering point with a reduced contact to the acromion, whereas in living monotremes the two structures have a rigid and broad articulation", (reference numbers omitted). The effect is that the joint of the scapula and clavicle was probably to some degree mobile. This is similar to: Gobiconodon, multituberculates, Zhangheotherium and existing therians. It's a contrast with monotremes. Adding some idle speculation of my own, this could be a sign of common descent in some or all cases. Alternatively, similarities might have arisen by separate routes in two or more lineages. Upper arm On its distal end, (which is the one nearest the hand), the humerus has radial and ulnar condyles, and there's an incipient trochlea for the ulna. Along with some further details, this reflects the condition in Zhangheotherium, and it contrasts with non-mammalian cynodonts, morganucodontids, multis and monotremes, (no ulnar trochlea). Again, this may or mayn't be suggestive of common descent. With regard to these derived features at the front end of the animal, the authors note: "Our phylogenetic analysis indicates that many apomorphies of the pectoral girdle and forelimb in J. jenkinsi are independently derived and convergent with those of therians", (p.329). Waist and back legs These areas share basal features with morganucodontids, tritylodontids and other non-mammals, (p.329). However, just because a characteristic is relatively primitive, that doesn't mean it's necessarily antiquated. One of the plesiomorphies is the presence of epipubic bones. As it happens, I haven't heard of their presence in any non-mammals other than tritys. (Then again, that shouldn't be thought of as overly significant.) The epipubis is found in all living non-placental mammals, and it was also popular with basal eutherians. In Hollywood terminology, epipubic bones could be represented by Laurel and Hardy. They were once required viewing in every cinema. While primitive, they still have their loyal fans, (marsups, monotremes and the movie goers in Dipwytch.) However, there is a list of further details from the pelvis and legs, which are most reminiscent of non-mammals and morganucodontids. Some of these are really antiquated; more like the movies of Abbot and Costello. They may be slightly later than Laurel and Hardy, but their films were nowhere near as derived. For further details, refer to the relevant paper and films. Feet Cretaceous mammal feet don't grow on trees. They're extremely rare, but exceptionally well preserved in this instance. The construction is in line with a ground dweller, who could climb to some extent. This is a useful attribute for small critters, to whom flat ground is in short supply. What's clear is that Jeholodens wasn't an arboreal acrobat. Tricondont diversity and possible relationships While tricondonts have an abundant and varied fossil record, this genus was the first, fully articulated skeleton. In this study Jeholodens, (and Gobiconodon), were held to be outside of Crown-group Mammalia. However, subsequent and wider evaluations show this conclusion isn't necessarily correct. Triconodont relationships within Mammaldom aren't presently resolved. Within Eutriconodonta, this genus is more closely related with triconodontids than with amphilestids or gobiconodontids. A clearer picture is difficult to envisage without more information on the other relevant taxa. Holotype The holotype, BMN 2139 a and b, is a nearly complete skeleton and partial skull "preserved as two conterplates". It's an inmate of the National Geological Museum of China, Beijing. The specific name honours the pioneering work on mammalian postskeletal morphology conducted by FA Jenkins Jnr. (Thanks from me are due to David Marjanovic and Dr Zhexi Luo, who generously sent the paper and an additional drawing of the skeleton, which is on my wall.) Additional notes "Jeholodens was originally interpreted as the sister group of Triconodontidae", (Rougier et al 2001, p.18). It still seems likely that this genus is within or close to that family, (eg. Luo et al 2002, p.9). 'Molars' behaving badly Rougier, Isah & Manabe, 2007 share some gossip on this theme on page 88-89. The preserved skeleton shows an apparent one-to-one occlusion of molars, but note that word 'apparent'. The main cusp of an upper molar, M1 for example, is aligned neatly with the main cusp of the corresponding lower. However, this may not reflect the actual occlusal pattern. If it did, then that'd be odd. As with the other fossils from the upper Yixian fauna, this one's been squashed as flat as a fairly thickish pancake, and things could've got pushed somewhat out of position. Half a tooth-length would suffice. The reasoning behind this concerns the three main cusps of the upper molar. Rather than being arranged in a strict straight line, a small degree of triangulisation is present (p.89). That's achieved by the central cusp having shifted sidewards of its front and rear colleagues. As far as is known, all other mammals with any degree of molar cusp triangulation have alternate occlusion. Cusp A of uppers comes down between the rear of a lower tooth and the front of the following one. The same would be expected behaviour for Jeholodens.
Reference:	Ji, Luo & Ji (1999), A Chinese triconodont mammal and mosaic evolution of mammalian skeleton. Nature 398, p.326-330.

Links: Carnegie Museum of Natural History, Pittsburgh http://www.carnegiemuseums.org/cmnh/news/99-jan-mar/jeholodens/jeholodens_images.html A press release with some good pictures. http://www.carnegiemuseums.org/cmnh/news/99-jan-mar/990322jeholodens_comments.html A good, short read by Timothy Rowe Nature: A Mosaic of a Mammal http://www.nature.com/nsu/990325/990325-3.html

Genus: Klamelia Chow M & Rich THV, 1984 'from Klamelia' Family: Klameliidae Martin & Averianov, 2007 Remarks: Martin & Averianov (2007) accuse this genus of being related to their new friend, Ferganodon of Kyrgyzstan. The generic name honours the district of the locality, Klamelia. It also made an appearance in print prior to its official publication. This was in the role of a nomen nudum in a 1983 book by Savage and Russell.

Species:	Klamelia zhaopengi Chow M & Rich THV, 1984
Place:	Junggar Basin, Xinjiang
Country:	China
Age:	Middle Jurassic (late) or Upper Jurassic (early)
Remarks:	The first part of this entry is based upon Chow & Rich, 1984, and thanks are due to the supplier. This paper had its critics and, hopefully, I've managed to take some discussion by Martin & Averianov, 2007 into account. I made sure to read that paper before turning to this older one. Three particular points come to mind. # C & R interpreted the first preserved tooth as being the final premolar. An alternative view, cited in C & R's paper, was that it could be a double-rooted canine, and that'd naturally have implications for identifications of the teeth behind. M & A concur with C & R's conclusion, but not with all their reasoning. # C & R identified a particular molar cusp as cusp b. Comparisons with Ferganodon, described in 2007, suggest it's probably cusp e. In the following notes, should it occur, then I'll be calling it "b" and perhaps other rude names. # Thirdly, C & R referred this genus to Gobiconodontinae, a new subfamily. This concept has since been raised to the status of a family (perhaps including Repenomamidae), and Klamelia was ejected. Despite the subsequent increase in gobiconodontid diversity, this genus was seen as being too different to maintain its membership. Nevertheless, similarities there are. All these various factors have contributed to making my notes on the original publication fairly brief. A Shishugou jaw Events began in September 1980 (p.226). Two Chinese researchers came across part of a jaw from the Shishugou Formation of Inner Mongolia's Junggar Basin, deposits now identified as dating from the Middle Jurassic (p.227). Unlike the norms displayed by triconodontids, the main cusp of lower molars is much larger than the other cusps. The owner also can't have been a morganucodontid as cusp g, the kühneocone, isn't present. According to C & R, the jaw must've been relatively short at the front with room for a maximum of six teeth prior to the molars. I don't happen to know how secure this accusation of a foreshortened mandible is now held to be. In any case, it was deduced from recognising that the: "posterior border of the midline symphysis" occurs below the meeting point of m2 and m3. Generally, this exciting event takes place somewhere below the canine or premolars. Additionally, there's a steep slope for the rear of this midline symphysis, and that's reported as showing the jaw didn't go all too far forward from the first premolar. This shortage of room would indicate a low number of anterior teeth. Among triconodontans, such a condition applies for what are now three specimens of Gobiconodon (p.228); G. borissiaki, G. hoburensis and G. ostromi. (They didn't all go by those names in 1984.) Further grounds for linking these critters with Klamelia, pointed out by Martin & Averianov, 2007, is that the final premolar is unusually large. It's also transversely expanded for Klamelia, and such a condition isn't known for Mesozoic canines. (That's at least partly why these authors also identify the tooth as a premolar.) Unfortunately, most of the crown has gone. Molars It now seems clear, as stated by C & R, there were at least six molars. They increase in size from front to rear, but their characters are otherwise fairly consistent (p.228). The basis for drawing a body between types of postcanine was simply a matter of size. A lower molar crown has a line of four cusps running along its middle, and these were originally interpreted as being a-d. However, as mentioned, it now seems that "b" is actually e. Largest in all imaginable dimensions is cusp a, the central one. Cusp c is considerably smaller and the others, regardless of their actual identities, are even more diminutive. There's a well developed, continuous cingulum on the lingual side of molars, with buccal ones restricted to areas at the front and rear. Holotype IVPP V6447 entertains visitors to the Institute of Vertebrate Paleontology and Anthropology, Beijing. It's part of a left jaw preserving evidence of two premolars and six molars; the crowns of the final premolar and first four molars, with roots for the remainder. Its specific name honours the discoverers Zhao Xiquing and Peng Xiling. Additional notes Zhang, 1984 (p.4) reports the genus is based upon a damage lower jaw containing one premolar, four molars and evidence of further roots both in front and behind. The mandible is quite large and deep. There are three mental foramina on the labial surface, and a Meckelian groove on the inside. Doubts have been expressed about that latter point by others. At least six molars were present when alive. Affinities with morganucodontids have been suggested, but subsequent research hasn't supported that. The original authors placed this genus within Gobiconodontinae, (now Gobiconodontidae), with the reference for that taxon as for the genus. However, subsequent analysis by others suggests Klamelia is not closely related to the gobiconodontids. Rougier et al 2001 plump for " Mammliaformes incertae sedis," (p.16). Early in 2007, it was transferred to the newly established family of Klameliidae by Martin & Averianov. The citation is the same as for their new genus of Ferganodon. As seen by Martin & Averianov, 2007 The identity of the tooth positions displayed by this critter has been the subject of disagreement even such it described. An anonymous reviewer voiced objections in that very paper. Chow & Rich interpreted the first tooth as being the final premolar. Roots forward of it suggest the set would've contained at least two premolars and six molars. "Oh ho!" said somebody, "but that tooth's procumbent and could be a double-rooted canine. Such items are known. If so, then those first four 'molars' would actually be premolars." At least, a suggestion along those general lines was mooted. While accepting some of the original reasoning was insecure, and doing so in agreement with the just mentioned Rougier & Co (2001), M & A also accept the original tooth identification was correct. The disputed tooth is transversely expanded at the base (p.47), a condition not known for any other Mesozoic mammalian canines. It is recorded for some premolars. Another clarification concerns the identity of a lower molar cusp designated as b. Comparisons with Ferganodon show, while this cusp is present for that genus, it's been strongly reduced. Cusp b has actually gone entirely from the crowns of Klamelia, and the previously designated cusp 'b' is actually a different development occurring on the cingulum; possibly cusp e as known from amphilestid and gobiconodontid eutriconodonts.
Reference:	Chow & Rich (1984), A new triconodontan (Mammalia) from the Jurassic of China, Journal of Vertebrate Paleontology, 3(4), p.226-231.

Link: DINOSAURS FROM XINJIANG, CHINA, James Clark and Xu Xing, 2001 http://www.gwu.edu/~clade/faculty/clark/china.html An overview of the history and results of fossiling in Xinjiang, home of Klamelia.

Genus: Kryptotherium Sigogneau-Russell D, 2003 'Hidden beast' Remarks: Presently, all I've seen is the abstract. This genus is said to be of a new lineage of triconodonts not found in Laurasia. Thanks are due to George (you know who you are) for mentioning the critter.

Species:	Kryptotherium polysphenos Sigogneau-Russell D, 2003
Place:	Anoual
Country:	Morocco
Age:	Berriasian?, Lower Cretaceous
Remarks:
Reference:	Sigogneau-Russell (2003), Diversity of triconodont mammals from the early Cretaceous of north Africa: Affinities of the amphilestids, Palaeovertebrata, 32(1), p.27-55.

Link: Palaeovertebrata, 2003 http://cat.inist.fr/?aModele=afficheN&cpsidt=15795185 The abstract for Kryptotherium and Goniconodon palaios.

Genus: Yanoconodon Luo Z-X, Chen P, Li G & Chen M, 2007 'Yan cuspate tooth' Family: Jeholodentidae Luo Z-X, Chen P, Li G & Chen M, 2007 Remarks: The generic name honours the Yan Mountains in the north of Hebei Province, the site of the animal's grave.

Species:	Yanoconodon allini Luo Z-X, Chen P, Li G & Chen M, 2007
Place:	Daluozyou locality, Hebei Province, equivalent to the Yixian Formation, Liaoning
Country:	China
Age:	Barremian, Lower Cretaceous
Remarks:	The following is based upon my reading of Luo et al, 2007, and thanks are due to the supplier. This entry will commence with a prologue on recent developments among Mesozoic mammals in northeastern China. And then the curtain will open on the main stage. Sportingly, William Shakespeare has agreed to assist with the subheadings. I thank you verily, kind sir. Prologue 2006 saw an astonishing development in the world of Mesozoic furry animals. The miraculous treasures of the Yixian Formation (and now the equivalent from Hebei) found themselves challenged by at least equally miraculous marvels. The Yixian team's Akidolestes thrilled with its surprisingly present lumbar ribs, but the gloss was knocked off the show by a megastar that leapt out the Daohugou beds of Inner Mongolia. This was a semi-aquatic docodont named Castorocauda, what with its beaver-like tails, webbed feet and postdentary bones preserved delightfully postdentarily. As if that weren't astonishment enough (and it was for this critic), Daohugou finished the year with Volaticotherium, a gliding mammal complete with a flight membrane. Yixian -the cradle of so many wonders- wondered what to come up with next. It had to think about this until the middle of March, and then asked similarly aged rock in the neighbouring Province to help out. The postdentary bones gave it an idea. These things include elements involved in mammalian hearing; the forerunners of our middle ear bone known as the malleus, for example. At some stage, evolution shifted these bones from a storage cupboard on the back of the inside of the lower jaw, and stuffed them into the ears. Given the proximity of the rear dentary and your ear, the distance involved isn't that great. Nevertheless, it was a neat trick, and left some of the audience trying to figure out how it was done. Another possibility concerned those briefly mentioned lumbar ribs of Akidolestes. These are things my wife doesn't have. (I assure you, I've checked her body most thoroughly. There are no ribs anywhere on her lower spine.) Akidolestes did have them but not, oddly enough, some close relatives. These lumbar ribs seem to come and go in some districts of the Mammalian Anarcho-Syndicalist Association, or go and then come again, and that sounds weird behaviour for body parts. Hebei Province rock strata considered the options and their new effort, Yanoconodon, sank its triconodont teeth firmly into both at once. "Like a rich jewel in an Ethiope's ear." Once upon a time, there were several small bones attached to the back of mammalian jaws, and these somehow moved their place of employment and enjoyment into the middle ear. "How did they travel?", the audience asked. "That's easy," replied Yanoconodon. "Transport was provided by my ossified Meckelian cartilage." While a few of the children in the audience clapped knowingly, most spectators simply stared as if none the wiser. The new jeholodentid decided to elaborate. "Look, I've got an ossified Meckelian cartilage here to show you. This long bit at the front fits into the groove on the inside of my jaw, low down towards the base. And here, at the back, there's a branch that kinks in towards the inside of my head, and those tiny, tiny bones on the end of it get shoved into my ear. Simple," it continued. "Actually, this sort of thing happened to all you living mammals as well, but that was when you were embryos, be you a platypus or a placental. I expect you've forgotten. For some reason I can't fathom, all you lot then let your Meckelian cartilages fade away. I liked mine so much, that I ossified it -made it bony. Some might term that a retained paedomorphic feature. Personally, I think it's simply stylish and other mammals are jealous. There's nothing childish about me, and I'll scream very loudly if you say otherwise." "Oh, and what about these comings and goings of lumbar ribs?" "Lumbar ribs are cool, too! However, one thing at a time. I'll show you mine a bit later. All I'll say for now is that the answer's in Hox gene patterning. Now, I think it high time that I was introduced." "My ever-gentle cousin, welcome hither." The molars have a simple triconodont style (p.288); three main cusps in a straight line along the length of the crown. In contrast to 'amphilestids' and gobiconodontids there are no accessory cuspules on a lower lingual cingulum. The dental formula per side is: (uppers): 2 incisors, 1 canine, 2 premolars and 3 molars; (lowers): 2, 1, 2 and 3 respectively. In both numbers and form, these teeth are much like the set sported by Jeholodens. The diagnostic differences between these two genera are found in the skeleton. Despite the presence of a Meckelian groove, there's no trace of a postdentary trough for 'extra' lower jaw bone storage. Those bones had gone ear-wards in contrast to more basal mammals such as Morganucodon. Going by my informal and approximate measurements from the sketch in the paper, our new distant cousin has a skull length of about 3 centimetres, a fairly elongated body of around ten cm and a tail length that may have stretched to about seven, although the latter is only partly preserved. This critter is the size of a standard mouse. Apart from much of the tail little of the body is missing; spine, shoulders, pelvis, legs, partial paws... Admittedly, the right rear leg peters out at the ankle and, should it be somehow brought back to life, that could create difficulties for scurrying around. However, even that wouldn't matter much in the circumstances. The small break in the spine would make scurrying completely impossible anyway. Compared to such an injury, an absent paw isn't worth waving at. Such completeness brings two words to mind: 'fan' and 'tastic'. "I am made of that self metal as my sister, and prize me at her worth." Discussions on the closer affinities don't require much attention in this case. It strongly resembles Jeholodens, so the authors assigned both sisters to the same new family, Jeholodentidae. In terms of the morphology, the two sisters could consider teeth-swapping parties. As if to emphasize their natural affection for one another, both holotypes even have their final molars only partly erupted. For Yanoconodon there are no signs of replacement occurring for any of the teeth, and all are of the adult set. The animal was either almost an adult or an adult of a species with some delayed eruption. All middle ear bones that might be expected on an ossified Meckelian cartilage are preserved attached to the end of a branch of it. This juts off from the main line directed along the jaw, and proceeds to the ear. An ossified cartilage wasn't originally identified with Jeholodens. However, re-examination showed it to be present although detached. An ossified Meckelian cartilage isn't a novelty, but it's still an impressive achievement, as a similar structure had already been found on a number of Yixian mammals including Repenomamus. But what is new, is the preservation of those middle ear bones still in place. This isn't a smoking gun showing that evolution pulled the trigger from such or such an angle. It's more decisive than that. In this case, the pistol's poised at the side of the head, with its barrel shoved into the ear. You don't need Inspector Columbo to solve this crime. Last year, Castorocauda displayed its very delicate postdentary bones preserved in their trough. This year, Yanoconodon shows us the same bones in place, but in a different place. "Out, you rogue! you pluck my foot away: Take that, and mend the plucking off the other." Should Yanoconodon ever want to walk again, then its limbs (and spine) would require some repairs and renovations. For example, more toes would come in handy. Nevertheless, it's impressively fit for a 125 million year old, and redrawing the original blueprint needs little imagination. For such a challenge, a right foot isn't essential as long as the left paw's available. We vertebrates are a symmetrical lot. The shoulder girdle is termed "therian-like", and praised for its mobility skills. Turning to the hind quarters of the critter, characters of the humerus give it a more old-fashioned flavour in line with basal mammalians and non-mammalian cynodonts, and the femur concurs (p.289). Lots of apomorphies common for multituberculates and therians aren't found in the limbs and paws of Yanoconodon, those limbs sprawled somewhat sidewards, and the short, stubby toe bones are suggestive of a ground-based lifestyle or perhaps, to some degree, a subterranean one. The owner was no arboreal acrobat. "She bears a duke's revenues on her back, and in her heart she scorns our poverty." In the case of myself or her sister, Jeholodens, it's obvious which vertebrae belong to our thoracic and which to our lumbar regions. The upper lot have ribs whereas the lowers don't. Rather surprisingly, given the family relationship, the spine of Yanoconodon follows an apparently radically different drum beat in this regard. As is typical for non-mammalian vertebrates, ribs continue down to near the pelvis while gradually diminishing in size. That makes a clear distinction problematic. The authors have opted for 18 of the first vertebrae and eight of the latter. That adds up to a total of 26 in all, and it's a lot; the same number as known from Repenomamus. Sister Jeholodens totals 22 and 19 or 20 are popular for extant mammals. Given their general similarities, these differences between the two jeholodentids are interesting. On the face of things, this presence and absence of lumbar ribs is a puzzle. This matter is of wider interest than just a family affair. Oddly, lumbar ribs displayed a habit of coming, going and coming again among various mammalian lineages. Some experimental research suggests this apparent trick isn't actually complicated to replicate. It requires either a bit of genetic manipulation or relatively minor mutation. More on that will be found below. Eight caudal vertebrae are also preserved on the skeleton, but the rest of the tail has gone. The reconstruction sketch in the paper seems to be based on an assumption of about twenty of these bones in all. Had an assumption simply been based on the number known from Jeholodens, then that would've required thirty. Presumably, reasons exist for thinking this tail would've been shorter. "Friends, platypuses, embryos, lend me your ears..." The assembly of bones preserved on the Meckelian cartilage include the ectotympanic, malleus and incus, and their form and proportions are broadly similar to their equivalents in the platypus ( Ornithorhynchus). Edgar Allin, after whom the species is named, predicted that there must've been an intermediate evolutionary stage when this cartilage had been connected at the front to the lower jaw, but its end would be deflected inwards (p.290). Hypothetically, the stage was set and Yanoconodon has now provided a convincing performance upon it. A considerable body of well-tried evidence supported such a hypothesis, eg. the embryological development of mammals and reptiles, but such a performer has never previously been available. If this specimen had been published five years ago, then the applause in the auditorium would've been deafening. However, miraculous performances from even more gobsmackingly talented artistes -Castorocauda is my personal darling- lessen the impact to just astonishment and deep admiration. Using my personal judgement as an entirely objective gauge, Yanoconodon scores no more than eleven out of ten on the wow-ometer. Such fine preservation is simply impossible rather than complexly impossible. It's a matter of 'As you Like it' rather than a literal, springtime 'Midsummer Night's Dream'. I'm sorry t sound blasé, but eleven out of ten is the highest I can go here. OK, then make it twelve. The main branch line of this Meckelian cartilage runs along the groove in the dentary, as Yanoconodon has already told us, and a branch at the end aims to the ear. The migrationary group of middle ear bones of the marsupial, Monodelphys, follow the self-same path during their embryological development. Furthermore, as the cartilage is curved and due to the alignments involved, jaw movements happen to have little knock on effect on the middle ear bones attached to it. The cartilage and arrangement of bones in this, at least near-adult critter are also similar to the pattern known from embryonic monotremes, eg. the platypus, and embryonic placentals, eg. you. The main difference is that the cartilage was retained and ossified instead of being disposed of after transportation (p.291). "Age cannot wither her, nor custom stale her infinite variety..." And now it's time to get back to the back; the vertebrae. Living mammals are rather economical with vertebrae. Virtually all have seven of the things in the neck -there are a few exceptions, eg. sloths- and, as far as I'm aware, no fossil mammals took a different approach. (Seven is also the number in the Lower Permian cynodont Thrinaxodon.) However, the vertebrae of the torso are lower in number than for Yanoconodon; 13 to 14 thoracic, five to six lumbar vertebrae, the latter with no ribs. That's a maximum of 20 compared to this critter's 26. As our long dead ancestors among early eucynodonts had lumbar ribs, then this lack of the things is a refinement; a derived character. Their absence is steered by things called Hox10 paralogues applying a break on their development. Should you be an interfering so and so, and happen to know what you're doing, then you could apply a triple knock out on the correct Hox10 paralogues of, say, a mouse trying to mind its own business, and the border between the thoracic and lumbar vertebrae can shift. A similarly ill-mannered assault on Hox11 paralogues can confuse lumbar vertebrae into becoming sacral ones instead. Shutting out the functions of Hox10 genes can result in the development of lumbar ribs and a graduated transition from the thoracic to the lumbar series. You can actually encourage the mouse to grow an approximation of the style of ribs worn naturally by Yanoconodon. The processes involved aren't complicated. However, don't expect any thanks, particularly as the rodent wouldn't be feeling in the best of health by the time you were studying its skeleton. It'd no longer to be equipped to feel anything at all. In short, whether deliberately inflicted or from chance mutations, switching genetic breaks on or off can do odd things to your vertebrae, but none of them require magic. This could account for the loss and/or reappearance of lumbar ribs in various lineages of mammals. Naturally, that raises a further intriguing question. If redeveloping lumbar ribs is relatively straightforward, and as the things must've had their advantages in some circumstances, then why don't any placentals get to make use of them through natural means? I don't pretend to know the answer to that. However, advantageous characters for one set of circumstances can be disadvantageous elsewhere, and may also impose restrictions. It occurs to me that, if my wife were to have had lumbar ribs, then our reproductive strategy involving a prolonged period of gestation would presumably not have been viable due to the extra stability behind the abdomen. Mind you, my wife's far from a typical placental mammal, or else I wouldn't have married her. In any case, all advantages come with costs. "The web of life is of mingled yarn, good and ill together." Holotype NJU-P06001A and B are the positive and negative slabs containing the fossil, and both are scholarship students at Nanjing University. The specific name honours Edgar Allin for his contributions on the evolution of mammalian ears.
Reference:	Luo et al (2007), A new eutriconodont mammal and evolutionary development in early mammals, Nature, 446, p.288-293.

Other reports: Morocco Reportedly, Denise Sigogneau-Russell presented a provisional paper on triconodonts from Morocco at the 11th International Symposium on Dental Morphology, (September 1998), at Oulu, Finland. Reference: SVP News Bulletin No. 177, October 1999. Prasad & Manhas, (2002), suggests Gobiconodon may be among them, though I've found no confirmation of this. Cloverly Formation, Big Horn County, Montana, USA Indeterminate triconodontid remains have been reported. Stored at the Peabody Museum, Yale, USA. Mexico Indeterminate material has been reported from an ?Early-Middle Jurassic location. La Boca springs to mind, but that's a guess. This is included in Table 1 of: Reference: Rose KD, Cifelli R L & Lipka TR (2001), Second tricondont dentary from the Early Cretaceous of Maryland: J. of Vertebrate Paleontology, 21 (3), p.628-632. Update 2008 saw the publication of three tricondont taxa from Mexico's La Boca Formation; Bocaconodon, Huasteconodon and Victoriacondon. India, Kota Formation, Andrah Pradesh A further indeterminate molar is described by Prasad & Manhas, (2002). England, Dorset A relatively large, undescribed triconodont of some kind was detected wandering around near Durlston Bay in the earliest Cretaceous. This is apparently something like Austrotriconodon. It features in Sigogneau-Russell & Kielan-Jaworowska (2002), Mammals from the Purbeck Limestone Group of Dorset, southern England, p. 241-255, in Milner & Batten (eds), Life and Environment in Purbeck Times. Special Paper in Palaeontology 68. Thanks are due to Darren Naish for posting the information on the Dinosaur Mailing List, (2.1.2003). I still managed to miss it for a year. Should this paper turn up by some means or other, it would be very welcome.

Fruitafossor A. Triconodonts B. 'Amphilestids'

B. 'Amphildestids'

Taxa: "Amphilestidae" Osborn, 1888 Amphilestidae is presently a fairly dubious but useful family label, and soon I may give some indication as to why. Notes are being written. In the meantime, you may as well take my word for it. 'Amphilestids' are among the earliest known eutriconodonts, and a strict interpretation would probably restrict the family to Amphilestes and Phascolotherium of the Middle Jurassic. In contrast to triconodontids, the middle cusp of the main three on a molar is much the highest, a condition which applies to others such as Jeholodens. The triconodontid interlocking mechanism for molars isn't developed, and the premolars are symmetrical, (Kemp 2005, p.152). A word of warning I should also point out the inclusion of entries in this subdirectory has been carried out in line with my own judgement, and doesn't follow a single proposal from any formal study. I don't approve of me going ahead like that but, given the circumstances, it seems to be the best method despite the risk of getting things wrong. "Amphilestids" and a club for Irish Americans The following notes have been written while reading Rougier, Isah & Manabe, 2007. They didn't actually discuss things using quite the same terms. The problem, and usefulness, of essentially junk categories like "Amphilestidae" could be compared with an Irish American social club, especially one which shows no great interest as to whether its members are ethnically any more Irish than I am. As, to be sure, to be sure, one of my grannies was born and raised literally within kissing distance of the Blarney Stone, in a blarney metaphorical sense, there'd be a fair chance of my ethnic credentials being stronger than those of many of the O'Muellers, McPatels or Pat Wzyworzkys. Qualifications to join consist of: a. wanting to, b. being proposed by a couple of established members; and c. that'll do. It's not even essential to drink Guinness or own an LP or two by The Dubliners or The Chieftains. Should anybody be involved with an Irish American Social Club as enlightened as that, then please accept my congratulations. It's an admirable approach to take. Cheers! Requirements for membership of "Amphilestidae" have been stricter than that but, as they happen to reflect a fairly basal mammalian molar style, then they were as widely available as records by The Dubliners or Chieftains now are. Both groups kept churning out their wares for decades, and very similar LPs can be picked up in record shops all over the planet. The requirements are lower molars with a reasonably straight line of three main cusps (p.73), the middle of which (a) is dominant. The first (b) and the third (c) are close to each other in size, and a couple of small cingular cuspules also occur. Two roots are also part of the equipment. In the absence of any disqualifying accessories, eg. a heavily cuspulated cingulum, you can become a member of "Amphilestidae". As it stands, the category is nowhere near restricted enough for purists. However, it happens to be a useful meeting place for reasonably like-minded molars. And, given the shortage of information on further details such as skulls, bodies and so on, racially cleansing it into monophyletic exclusivity will probably remain no more than a unrealisable good idea. Time For what it's worth, animals referable to as amphis had a career ranging from at least the Middle Jurassic until the Lower Cretaceous; small mammalian killers. How and whether they were actually related is poorly understood. In case anybody should be wondering, the inorganic nature of typical fossils -all organic remains have been subjected to mineral replacement- means the answer is no, genetic comparisons wouldn't help at all. Literal petrifications are made of stone, and don't have any DNA to test. "Amphilestid" teeth are primitive things, and some amphis probably ultimately led to triconodontids, " "symmetrodonts", boreosphenidans (including me) and asustralosphenidans. The style is very deeply rooted within the soils of the mammalian flower garden. Genera: Amphilestes, Aploconodon, Argentoconodon, Comodon, Didelphys sp. (=Phascolotherium),"Phascolodon" (= Comodon), Hakusanodon, Ichthyoconodon. Juchilestes, Kemchugia, Liaotherium, Meiconodon, Phascolotheridium (= Comodon), Phascolotherium, Tendagurodon, Thylacotherium (=Phascolotherium), Triconolestes, Volaticotherium other reports Time-Line: Lower Cretaceous:Hakusanodon, Ichthyocondon, Juchilestes, Kemchugia, Meiconodon Upper Jurassic: Aploconodon, Comodon, Tendagurodon, Triconolestes Middle Jurassic: Amphilestes, Argentoconodon, Liaotherium, Phascolotherium, Volaticotherium

Genus: Amphilestes 'paired robber' Family: Amphilestidae Osborn, 1888 Remarks: 'Amphilestes' was also applied to a dragonfly genus in 1862. This was subsequently renamed, Rhinagrion.

Species:	Amphilestes broderipii (Owen, 1845)
Place:	Stonesfield slate, Oxfordshire
Country:	England
Age:	Bathonian (mid), Middle Jurassic
Remarks:	Several jaw specimens are in the collection of the Oxford University Museum. "Data are limited to characters from the lower dentition and mandible", (Luo, Kielan-Jaworowska & Cifelli 2002, p.9). Upper molars: In this genus, Aploconodon and Comodon (= Phascolodon), the middle cusp A is larger than B or C, which are about equal with each other. Those latter mentioned cusps have shifted to the external (buccal) side of the tooth, (Datta & Das, 2001). Simpson, 1935 (p.165) offers a lower dental formula of four incisors, one canine, four premolars and five molars per side. "The premolars are symmetrical with large central and small anterior and posterior accessory cusps. On the molars the central cusp is much larger than the anterior and posterior cusps." The species name was authored by Owen in 1845, but I don't know which generic name he used at the time. The later date of 1859 may have significance.
Reference:

Genus: Aploconodon Simpson GG, 1925 Family: Amphilestidae Osborn, 1888

Species:	Aploconodon comoensis Simpson GG, 1925
Place:	Morrison Formation, Wyoming
Country:	USA
Age:	Upper Jurassic
Remarks:	Knowledge is limited to the holotype; lower dentition, (Rougier et al 2003, p.11).
Reference:	Simpson (1925), Mesozoic Mammalia. American triconodonts: part 2. American Journal of Science, 5, p.334-358.

Genus: Argentoconodon Rougier WR, Garrido A, Gaetano L, Puerta PF, Corbitt C & Nowacek MJ, 2007 'Argentinean coned tooth' Remarks: Should anybody require further assistance with working out why this generic name was selected, then please accept my profound sympathy.

Species:	Argentoconodon fariasorum Rougier et al, 2007
Place:	Canadon Asfalto Formation, Chubut, Patagonia
Country:	Argentina
Age:	?Middle Jurassic
Remarks:	The following is based upon my reading of Gaetano LC & Rougier GW, 2011, and not the original description. That got damaged by a torpedo fired from the submarines of hindsight armed with lots of additional information. Odd as it certainly seems to me, I'd already written some words on the possibilities of booby traps into my notes before reaching the part on the 2007 misidentification of an upper molar as a lower one. (Whilst I had quickly read through the paper before writing anything, that point managed to float by unnoticed.) Rather than throwing it away, I've left my original entry tacked on down below. I'll repeat one sentence from it here: "The build of the tooth suggests it's neither the first or last of the right lower series, although the possibility of it being an upper left can't yet be discounted (p.6)." Argentoconodon take two, Prelude This study concerns fossils recovered from a Jurassic lake in Patagonia, and there are various challenges confronting people trying to interpret them. Firstly, the limitation and condition of the collected material can leave much hope for fuller and better specimens but, for the while, researchers can only work with that which they've presently got. Secondly, specimens can only be compared with other fossils that happen to be known, and not with unknown ones that might be wished for. And, as most available material has been assembled from localities far away from this ancient Patagonian lake (many not just far in crow-flying distance, but also in geological time), the best comparisons now possible may turn out to be not overly appropriate. Time will tell. Furthermore, as all research on the local mammal remains has taken place during the last decade, you don't have mistakes made by pioneers from fifty or a hundred years ago to benefit from. Other fossil mammal faunas from Patagonia, all of which are younger than the Mesozoic, have been known of since Charles Darwin passed through in the 1830s. Later, towards the end of the same century and into the start of the next, Carlos Ameghino scoured the region successfully for further sites and fossils while Florentino, his brother, concentrated his attentions more on trying to interpret things; frequently somewhat wrongly both in terms of age of deposits and the affinities of the animals. As GG Simpson says of the latter point (Simpson, 1980 p.13): "He was caught in what may in retrospect be seen as a sort of then almost unavoidable booby trap: the phenomenon of evolutionary convergence, which still is a source of disquiet and precaution to paleontologists and other biologists." And later in the same paragraph: "That is especially likely to occur when the groups involved are in different geographic regions separated by barriers to expansion and migration." The paleontologists now engaged with the isolated mammal remains from the Canadon Asfalto Formation may turn out to be caught by still unseeable booby traps. I don't doubt they know this themselves. Still, nobody can do more than the best they can with the information they have at hand. (To repeat, that was written before I noticed the reinterpretation of the molar, and wasn't some kind of short-term prediction! Between you and me, I was thinking in terms of the difficulties that can arise by attempting to place sparse Gondwanan remains into Laurasian-centred frameworks which, at present, is the only viable course available in this case.) Argentina's coned tooth The genus was established in 2007 on the basis of an isolated molar, all that was then available. Subsequently, more remains have turned up so as to substantially increase the evidence (p.829). One helpful specimen provided most the teeth and some scraps from the body. Similarities were recognized with two critters in particular; the very poorly represented Ichthyconodon from the Lower Cretaceous of Morocco, and the outrageous gliding show-off Volaticotherium, a fossil so complete that it can leave you feeling decidedly queasy in the face of such undeserved and overwhelming over-generosity. While that superstar may be of a similar age to Argentoconodon, it practiced its aerobactics in thoroughly northern Inner Mongolia. While it would have been wonderful to compare this genus with other South American triconodonts of the Jurassic, that option was impossible on the account of there being just this one. Anyway, the authors found support for placing all three mentioned genera, plus several more, into the triconodontid subfamily of Alticonodontinae. Additionally, what's known of 'Argentina's coned tooth' doesn't rule out the possibility of a similar lifestyle to Volaticotherium, but it's too sparse to make a strong case for assuming that was so. The extant squirrel family, Sciuridae, includes members who glide. Others, such as our local red squirrels, would require the help of a powerful catapult to enjoy anything approaching their way of getting from point a to tree. And a crash helmet and a helpful net at the intended landing spot as well. Newlings The helpful additional specimens (p.822) are a fragmentary jaw half with one molar, and part of a skeleton including the jaws (with some teeth in place) and associated isolated teeth. Something caused them to fall out. A further unassociated tooth and a few more fragments were also found. Lower jaw The left dentary is close to being complete and is equipped with two molars. The more damaged right partner has one molar and a probable premolar (p.833). The bone is slender but, as it reaches the level of the molars, the upper margin begins to ascend, and that results in an increase in depth. At the top of the rear of the jaw is the coronoid process. Although incompletely represented, this was comparatively wide for this critter, and its wall, just behind the third molar, goes steeply up; much of the way towards being, but not literally, vertical. On the internal side a Meckelian groove can be seen, although not too clearly. This would've housed Herr Meckel's most excellent cartilage, and it's recognisable low down from about the middle of the m1 tooth, and runs forward from there. Further back along the jaw there's no indication of the presence of any trough for postdentary bones. Dental formula The tooth numbers per side appear to be: Uppers: ?2 incisors, 1 canine, 4 premolars and 4 molars; Lowers: 2, 1, 4 and 3 respectively. Lower incisors and canine No incisors were found implanted, but two isolated ones with some distictions were identified as from two separate lower positions, and that seems to be the total number involved (p.834). One is a small tooth whilst the other is markedly even smaller with a circular root in cross-section. Based upon comparisons with other triconodonts, eg. Gobiconodon and Volaticotherium, the larger of the pair is likely i1. A further isolated tooth seems right for a lower canine. It's much taller than either incisor and would, indeed, have looked down upon all other teeth in the entire row (p.835). It was single-rooted. Lower premolars Table 1 of the paper contains measurements for length and height, but I'll restrict myself to the former for material associated with the partial skeleton: p2 1.26mm; p3 1.59. Two very similar teeth, a left and a right one, are probably p2s. Each has two clear cusps present with the one denoted a dominating much of the crown. A wide but shallow valley comes between that and cusp c to the rear. These teeth are single-rooted. A further specimen is perhaps a left p3. This has two roots and that's the number of alveoli for such a tooth on the jaw. The crown's a bit more complex than the aforementioned positions as there's a third cusp right at the back, and these decrease in size from a to d. Over half the length of the crown is hogged by cusp a, the small c sits on the other side of a notch and, still further back, tiny d juts out from the base of the crown to overhang the rear root. That root is implanted diagonally into the jaw whilst its front partner goes in vertically. Not much can be related of the p4 beyond its presence. A bit of breakage shows an unerupted p4 in front of the first molar. Lower molars A full set comprises three but lengths are only available for the first pair: m1 3.26mm, m2 3.83. These teeth are very narrow (p.838), with four cusps forming a line along the length (from front to rear, b, a, c and d). Cusp bases begin at similar levels on the crowns, and the tips of the first three recurve increasingly strongly from b to c. That trio of cusps are close in size although the central one, a, is a bit larger. There are distinctions between both the first two molars, with the most readily comprehensible being the greater length of m2. Finer points are given as well (p.839). The first molar is positioned further externally in the mouth of m2, while the latter is implanted at a higher position on the jaw. Two roots in a bit of jaw represent the remains of an m3. They serve to show this tooth was much smaller than the other molars. Upper incisors and canine The probable number of incisors is two, but that's not certain. In comparable critters, eg. Volaticotherium, the I1 is far larger than the I2 and, presumably, a similar pattern is appropriate here as well (p.833). Suitable representatives were among the isolated teeth, and they're single-rooted. The also single-rooted canine is higher than the I1, but less so then for its lower counterpart. It curves to the rear towards its tip. (For lovers of heights, the lower canine achieves 4.4mm, the upper 4.0 and the I1 3.55.) Upper premolars The teeth examined are a mixture of left and right. Lengths: P1 1.00mm, P2 1.75, P3 2.14, P4 (or DP4) 2.48. There are probably four premolars per side (p.835), and these become increasingly complex along the line. The fourth isn't all that unlike the molars, and it can't presently be ruled out that it is actually a molar or, indeed, a deciduous premolar. (Milk premolars are frequently more molar-like than the replacement teeth.) The first three identified upper premolars are deductions based on isolated teeth whereas the fourth, the more complex form, is still in position on the jaw. If correctly identified, then the first three premolars increase in size with Ps2 and 3 bearing one more cusp than P1. Descriptions of all are naturally in the paper but, as something like a close-ish representative for them all, I'll merely muse a bit upon P2. Oh, and mention that the P1 is single-rooted while its colleagues have two roots. P2 has a crown with three cusps although, should you look too casually, you could miss noticing the third. That's D; little more than a bump at the back. Cusp A dominates the crown and is four times the height of C. The tip of A is slightly recumbent. The possible P4 is more reminiscent of the molars than other premolars, in that it has well developed cusps. However, there are only three rather than four. These are accused of being B, A and C. B's at the front, an area of the crown with no cusp on the other premolars. The tooth length is less than for the first molar. Another contrast to all molars is that the front and rear cusps, that's B and C, don't enjoy full support from the roots, and that's more usually found with premolars. The level of complexity could indicate a deciduous tooth, although it is in line with the situation known for Volaticotherium for the position its authors referred to as a P4 (whether rightly or wrongly). Upper molars Lengths: M2 (2 specimens) 2.85 - 3.39mm, M3 (1 sp.) 3.38, M4 1 sp.) 1.98. There were probably four molar positions, but only three could be measured for length and, again, both left and right teeth are involved. The teeth from the first three positions have five cusps. From front to rear these are E, B, A, C and D. One of the two M2s is the type specimen of this genus which, originally, was thought likely to be a lower. The new information outs that as incorrect. A partly preserved M1 is still in the jaw. Cusp A was taller than the similarly heighted B and C, but both the state of preservation and still lingering matrix conspire to limit visibility. There were two roots. Two more cooperative specimens, including the holotype and another in position on a jaw, were identified as M2s. Like the first molar this is a double-rooted tooth, but it's larger. Cusp B is half the height of A while C is almost as tall as the latter. E and D, at the front and rear respectively, are on the border between the crown and root. The D cusp is set buccal of the E cusp of the following tooth. An isolated molar is likely a left M3. As the main cusp is a smidge more lingual than B and C, there's a touch of triangulation to the arrangement. (As it happens, I very recently read accusations of modest obtuse-angled triangulation for the 'amphilested' known as Juchilestes from the Lower Cretaceous of China.) Cusp B is about two-thirds the height of A which, again, is but little taller than C. Two roots are present. The identification of M4 is more tentative (p.838). This would be the shortest of the molars, should the identity be correct, only single-rooted, and lacking a cusp B. Due to the presence of matrix it can only be seen from one side. Affinities The authors revised previously established data matrices for triconodonts (p.839), and added nine further genera including this one and Volaticotherium. A switch was then pushed and a convenient computer was asked if it wouldn't mind sorting through the info. It opined upon a close relationship for that pair within Alticonodontinae along with Ichthyconodon, a number of North American Cretaceous taxa and Meiconodon. The computer fell about laughing when asked about close kinship between this crew and various other 'amphilestids' including Jeholodens and Amphilestes. That was interpreted as meaning no, and such derision wasn't unexpected. Although similarities were noted between the available limb bones of Argento and Volati, a strong basis isn't there for concluding both must have been arboreal gliders (p.840). The preservation is too poor in the case of Argento. Triconodontidae According to those findings, Argentoconodon, Volaticotherium and Ichthyoconodon are members of Triconodontidae (as previously found in the case of Ichthyo). Consequently, I could move the relevant entries to that directory. However, I'd like confirmation from elsewhere before doing that and, in the mean time, took the simpler route of moving Ichthyo here into this looser enclosure instead. Holotype MEF-PV 1877 was a single tooth but it's now enjoying the companionship of further specimens. It resides in the Museo Paleontologico Egidio Feruglio in Trelew, and the specific name honours the family of Ricardo Farias. They've provided much welcomed assistance to researchers interrogating parts of their land over the last few decades. Update from September 2011 What follows is the entry for this genus as first writen, but with warnings inserted. The original authors weren't sure whether they were looking at a lower or upper molar, but plumped on the grounds of probability for a lower. Further finds, described in 2011, have indicated that the type fossil is an upper. Consequently, much of the text below was rendered irrelevant as comparisons were made with the wrong teeth! The following is based upon my reading of Rougier et al, 2007. According to the news programme in my dream channels, festivities in Patagonia were in full swing through much of the summer of 2007. Almost the entire population were still celebrating the birth of a new member of the Jurassic royal family in the Republic of Argentina. Excitement had spread like wild fire with reports of Henosferus safe arrival into the scientific annals, only the second South American mammal genus of the age. Filled with joy as both came from the same Patagonian formation, people naturally took to the streets to dance with their neighbours. This partying continued for several months, and it wasn't until late August that tiredness began to take its toll, and a start was made on clearing up the empty bottles, plates now bereft of goodies, trestle tables, streamers and balloons. Something approaching mundane normality started to return. The dream channel news team haven't yet reported on local reactions to the publication of Argentoconodon, South America's third Jurassic mammal. This hit the news stands in September. Naturally, Patagonians won't be expected to begrudge such an honour being bestowed upon other parts of the continent but, as this one was also from the same locality, an ancient pond in the Canadon Asfalto Formation... Patagonian Pond 3 Rest of the Continent FC 0. Problem paragraph The molar is an upper, not a lower! (Added September 2011) Whereas Henosferus and Asfaltomylos are represented by lower jaws from closely related australosphenidan mammals, Argentoconodon (which I kept misspelling in my notes!) is restricted to a single tooth from a different mammalian stock; at least descriptively a triconodont of presently unclear affinities (p.1). This presumably lower molar ( molariform in the stricter language of the authors) shows similarities with some fossils from North America and Africa, and has "amphilestid" traits. However, the systematics of that group (or groups) are presently far from settled. None too surprisingly, this geographically remote Gondwanan fails to add clarity to such wider issues. "Amphilestid" is used n a loose sense in the paper as, presently, a more formal usage isn't viable (p.2). It refers to mammals which generally have a near symmetrical arrangement of five cusps on molars; three main ones and two cingular cusps. The central cusp (A on uppers or a for lowers) is always the largest. Basic toothology The molar features a straight line of three main cusps running along the centre of the crown, and a further small cuspule at either end (p.4). Another small cusp occurs lingual of the valley between a and c, a position known as g (similar to the so called kuehneocone of morganucodontids). Cusps b and c are close with each other in height while a is considerably taller. Both e and d (front and rear cuspules respectively) jut outwards. The crown is extremely narrow, and all main cusps are jauntily recurved. Problem paragraph The molar is an upper, not a lower! (Added September 2011) The build of the tooth suggests it's neither the first or last of the right lower series, although the possibility of it being an upper left can't yet be discounted (p.6). A length of three millimetres means it's comparatively large for a mammal of the age. A lack of clear cingula, and the near symmetry of the cusps in side view, make the orientations more difficult to determine. A possible area of wear on the side of cusp b is equivocal. It may simply have resulted from chipped enamel. Cusp g is more helpful, assuming that's its correct identity. Comparisons with northern triconodonts suggest that's indeed what it is, but Gondwanan ones might have been pursuing different developments. The positioning of the front and rear cuspules indicates the probability of there having been slight overlaps with the neighbouring teeth and, in line with mammalian norms, the front of this tooth presumably overlapped the buccal side of its predecessor. Such contact is sometimes termed echelon. Cusp a, the largest, is centrally positioned between both roots (p.7). As with other cusps, one side is somewhat more convex than the other, and that's likely the buccal one. Unless very peculiar indeed, the curvature of cusps gives them a tilt towards the rear; recurvature. if eucynodont postcanines have cusp tips behaving like this, then recurvature is the norm. The front root is somewhat larger than the rear one and, although preserved depths are similar, breakage at the tip suggests it may also have been longer. Descriptive triconodonts If you were to say that a triconodont is an animal with postcanines featuring a centrally situated line of there main cusps, then you'd end up with a vague group involving convergence. It would include various non-mammalian cynodonts, eg Pseudotriconodon. Limiting it to critters with (as far as is known) diphyodont dental replacement would still leave a vague collection containing convergence. You'd have morganucodontids, "amphilestids", gobiconodontids, triconodontids and more. While there's some monophyletic core in all those families, its extent isn't necessarily all that clear, especially for "amphilestids", and not all those groups are closely related. There does seem to be a natural taxon of Eutriconodonta involving at least some amphis along with gobis and trics but how, or whether, Argentoconodon features among them is unclear. Some similarities shared with Ichthyoconodon and Triconolestes could have come about due to common descent. Then again, the vast white spaces on the fossil map of Gondwana leave the present assessment possibilities in the realm of the tentative. Comparative anatomy requires comparisons, and little of possible relevance is available from Gondwana. Laurasian lineages may be of little assistance, but at least some are known. Morganucodontids These very basal mammals are known from the Gondwanan Lower Jurassic as well as from northern landmasses. That was back in the days of the mega-continent of Pangaea when terrestrial radiations could play themselves out on a global stage. Given morganucodontids fluttering flags in Europe, North America, Asia and Africa, they could well turn up in South America too, should deposits of the right age start yielding mammal fossils there. Problem paragraph The molar is an upper, not a lower! (Added September 2011) While all that's well and good, Argentoconodon isn't a morganucodontid. Cusp b is properly part of the crown, rather than being a development launched from the cingulum, and its size is similar to that of c. Also in contrast to Morganucodon and Co, there's a shortage of cingula for additional cuspules to populate. The cusp g is in a morganucodontid-like position, but the Argentinean tooth is otherwise too derived for such affiliations. It may well be a descendant of something very much like a morganucodontid, but that applies for me as well. Apart from the g cusp, it shares more similarities with eutriconodontans. Problem paragraph The molar is an upper, not a lower! (Added September 2011) Triconodontids As a rule, triconodontids share a suite of traits that Argentoconodon doesn't have. For example, cusps a to d are all similarly sized. That's not so in this case. Interlocking of lower molars is accomplished by means of a groove on the front root, and that accommodates cusp d of the neighbouring tooth (p.9). This interlocking system isn't developed for a basal member named Priacodon, and its cusp d is smaller than for the others. However, molars are also wider than for Argentino, cusp b is higher than c, and e doesn't develop. There's no room for the critter in that particular inn. Problem paragraph The molar is an upper, not a lower! (Added September 2011) "Amphilestids" It shares "amphilestid" characters, but there's something of a snag. As some of them are primitive traits, then if they resulted from common inheritance, they could have persisted in lineages of not necessarily closely related mammals. Its amphi qualifications include possession of a crown with a line of five cusps arranged in something like a symmetrical manner, the dominance of cusp a, a close height similarity between b and c, the narrowness of molars in comparison to morganucodontids, a large cusp e, no or only a reduced cingulum (the remnant being cusp g in this instance), and a relative shortage of accessory cuspules. Not terribly surprisingly, no established genera volunteered to accommodate the new South American. Problem paragraph The molar is an upper, not a lower! (Added September 2011) Amphilestes tried its best, but it has a smaller a cusp and there's a strong lingual cingulum. That latter 'defect' also afflicts Phascolotherium. Both those genera also have a bulge on the buccal base of cusp a, as do most amphis. That's not noticeable for Argentoconodon. Details also rule out Upper Jurassic North Americans, Jeholodens of China and Uncle Tom Cobley and all. One similarity, this echelon overlapping, may be shared with Klamelia. However, a different study indicates that could be problematic. This concerns the identity of a cusp towards the front of the crown, which was originally identified as being b. Rougier & Co write, with regards to a difference for Klamelia (p.10): "...the presence of a cingulum that includes a poorly developed cusp b." I don't wish to spread gossip, but Martin & Averianov, 2007 have reason to believe that's not its identity. They accused Klam of not having a cusp b. Rather, it's probably cusp e in a somewhat shifted position (M & A 2007, p.45). Still, I don't suppose anybody will come to blows on this matter. Problem paragraph The molar is an upper, not a lower! (Added September 2011) Affinities In short, Rougier & Co compared the specimen with a whole range of amphis and others, and came up with no satisfactory close matches (p.11). However, similarities seemed strongest with Triconolestes of North America and Ichthyoconodon from Lower Cretaceous Morocco. They may perhaps nod towards a common ancestor. The wide geographic range of Morganucodon demonstrates the opportunities that were available for migrants in earlier generations, and perhaps a randy amphi roamed as well. More specimens are required to help fill out the presently scant details known from Gondwana. As further triconodont material has recently been obtained from South America -it's Patagonia's favourite ancient pond again, a bit more detail will likely be added. It either represents the same taxon or a closely related one and (p.12), should the latter apply, the local residents may well be celebrating yet again.
Reference:	Rougier et al (2007), First Jurassic triconodont from South America, American Museum Novitates, 3580, p.1-17.

Link: American Museum Novitates 3580 http://digitallibrary.amnh.org/dspace/bitstream/2246/5873/1/N3580.pdf Rougier et al, 2007 is presently freely accessible in pdf format.

Genus: Comodon (Simpson, 1925) Kretzoi M & Kretzoi M, 2000 'Como tooth' Aka: "Phascolodon" Simpson GG, 1925; Phascolotheridium Cifelli RL & Dykes TD, 2001 Remarks: The original generic name was preoccupied by Phascolodon vorticella Stein, 1859, a ciliated protistan (a freshwater plankton). Co-authoring a replacement was very interesting, (update, ...though superfluous!). Thanks are due to Peter Eigner of Deutsch Goritz, Austria and Dr Genoveva Esteban of the Windermere Laboratory, England. Their advice on the literary citation from 1859 was generous and much appreciated. It has subsequently become apparent that a replacement name for "Phascolodon" had already been published in 2000. By seniority, the valid name is indeed Comodon, which refers to Como Bluff, Wyoming.

Link: Peter Eigner http://members.magnet.at/p.eigner/Diversity.html A team picture of various plankton. Phascolodon is in the top left corner, number 127. It's clearly not a mammal. A fine image which the Windermere Lab team assembled for an open day, (thanks for sending a copy, Genoveva!). Family: Amphilestidae Osborn, 1888, Amphilestinae (Osborn, 1888)

Species:	Comodon gidleyi (Simpson, 1925) Kretzoi M & Kretzoi M, 2000
Aka:	Phascolodon gidleyi Simpson, 1925; Phascolotheridium gidleyi Cifelli & Dykes, 2001
Place:	Morrison Formation, Wyoming
Country:	USA
Age:	Upper Jurassic
Remarks:	Although smaller (and later) than Phascolotherium, Simpson observed a broad anatomical similarity between the two genera. Our 2001 replacement name sought to maintain the spirit and intent of his description. So it goes. The original fossil is in the collection of the Smithsonian Institute, while Yale has a cast copy.
References:	Simpson (1925), Mesozoic Mammalia. I. American triconodonts: part 2.--American Journal of Science 10, p.334-358.
	Ketzoi & Ketzoi (2000), Fossilium Catalogus 1: Animalia. Pars 137--Index Generum et Subgenerum Mammalium. Backhuys Publishers, Leiden, 726 p.
	Cifelli & Dykes (2001), Phascolotheridium, a new name for the genus Phascolodon Simpson, 1925 (Vertebrata, Mammalia) preoccupied by Phascolodon Stein, 1859 (Ciliophora, Phyllopharyngea). Acta Palaeontologica Polonica 46(3), p.392.

Link: Acta Palaeontologica Polonica 46(3) http://www.paleo.pan.pl/acta/acta46-3.htm The contents page.

Genus: Hakusanodon Rougier GW, Isah S & Manabe M, 2007 'Hakusan tooth' Family?: Amphilestidae Osborn, 1888 Remarks: Hakusan is a volcano near Kuwajima, and it's held to be sacred by followers of Shinto and/or vertebrate paleontology. As well as honouring the mountain for those reasons, the authors may have been informed it would blow its top in the case of a different natural feature being given precedence. Hakusan literally translates as 'white mountain'. Pre-ramble I've read x hundred various papers and books on Mesozoic cynodonts of all possible hues, and many have contained astonishing stuff; 240 million year old burrow complexes, 125 million year old fur coasts, a semi-aquatic docodont with a wonderfully preserved tail and webbed feet... This is the golden age of vertebrate paleontology, and it's a huge stroke of luck to be able to enjoy it. However, none of those papers managed to astonish me in the way Rougier et al, 2007 did on page 80. It's due to a trifling matter from 2001 and, if it hadn't been for an unnoticed publication a year prior to that, its significance would've soared to minor. There was this Jurassic mammal with a preoccupied name, and I happened to notice it before anybody else on the planet! Excepting for a couple of Kretzois. Anyway, there I was innocently reading my Rougier, Isah & Manabe, 2007, listening to them speak of premolars and molars on page 80. Suddenly, they said: "Simpson (1925b) described a similar condition in Comodon Kretzoi and Kretzoi, 2000 (formerly Phascolodon Simpson, 1925b, = Phascolotheridium Cifelli and Dykes, 2001); Cifelli 2002a)." Not that I'm famously observant, but I couldn't help noticing that Dykes chap is me. An appearance in the reference section wasn't one of my anticipations.

Species:	Hakusanodon archaeus Rougier GW, Isaji S & Manabe M, 2007
Place:	Hakusan, Ishikawa Province
Country:	Japan
Age:	Lower Cretaceous
Remarks:	The following is based upon my reading of Rougier et al, 2007. This paper offers much more in its forty plus pages than just a description of a partial lower jaw and, although my notes won't reflect much of it, there's plenty of significant, informative stuff. Using a data matrix of 62 characters and 28 taxa, the authors attempt to tackle the family tree(s) of triconodonts. A considerable portion of the discussion centres on the nature and merits of the 62 characters employed. As nobody assumes this is anything like the concluding song sung in the exploration of that particular theme, their observations and considerations will undoubtedly be of use in future projects by both themselves and other researchers. This is a resource for wider issues in the continuing concert. Japanese triconodonts Several fairly complete triconodont mandibles have been recovered from the Kuwajima Formation in western Japan, although only one features in this paper (p.73). It's seen as being a close relative of Jurassic "amphilestids" from North America and Europe. That's why I've chosen to place it in this section. However, given the prevailing lack of clarity about amphi systematics, it doesn't receive an official referral to a family or even a particular order. A problem lies in the artificial nature of several traditionally used Mesozoic mammalian "orders" including "Symmetrodonta", "Triconodonta", and the family of "Amphilestidae". While they're presumably partly natural taxa for descendants of genuine common ancestors, separating the corn from the chaff of simply the occurrence of similar features, or their retention, in not closely related lineages isn't yet attainable with the available resources. Additional notes This generic name appeared in the cladogramme presented by Rougier et al, 2001 (p.20). Their results suggest a possible relationship to Comodon, (the triconodont originally known as Phascolodon). However, it appeared there as a nomen nudum. The planned publication failed to materialize in 1999. That was intended to be Rougier GW, Isaji S & Manabe M, An Early Cretaceous Japanese triconodont and a revision of triconodont phylogeny. Journ. of Vert. Paleont. 19, Supplement 3, 72A. There was a slight change of schedule for one reason or another, and it finally come out in a different journal in 2007. Terminology The authors don't use the words premolar or molar in their description of this fossil. Instead, they write of premolariforms and molariforms. This reflects a difficulty with some Mesozoic mammals. In several cases, it's now known that some molar-like teeth were actually replacements. Strictly speaking, molars have to be first generation teeth without replacements. The primitive habits of critters such as Repenomamus the Magnificent make the identification of genuine molars problematic. However, for reasons of language familiarity, I tend to stick with more generally understood words. Not all readers necessarily know quite what molars are, let alone trying to cope with an additional, odd looking -iform. Locality An outcrop of Cretaceous rock, which the local residents affectionately term the Kaseki-Kobe fossil bluff -should they happen to be geologists (p.76), can be found loitering along the Tedori River in the vicinity of the city of Hakusan. It's from the upper reaches of the Kuwajima Formation, itself part of the Tetori Group of three formations. Plant fossils have long been known from the area, but the discovery of vertebrate remains is a more recent development. The rock built up in river channels and on a floodplain. The flora and fauna are diverse and its age, probably upper Neocomian, makes it unusual; an otherwise poorly known chunk of geo-time for terrestrial wildlife. Typical vertebrate finds occur as isolated teeth and bits of bone. The fauna packed one heck of a eucynodont punch. It includes Lower Cretaceous, non-mammalian cynodonts; at least two genera of gloriously late tritylodontids. They've already graced the pages of a number of studies, but haven't yet been named. Size and tooth numbers According to my estimate, based on the sketch, the preserved fragment of mandible has a length of around 1.2cm. For lovers of such things the known tooth formula is: (lowers): incisors ?, 1 canine, 3 premolars and 5 molars (p.77). Alert readers may recall these are more correctly termed premolariforms and molariforms, and that's my final reminder. Postcanine cusps are blunt, poorly separated and somewhat recurved at their tips. The front cusp b is always stronger than the c one. A brush with the teeth The absence of the front of the jaw ensures information on the incisors means the term 'limited' would be an exaggeration. Nothing at all could be more fairly termed 'infinite'. The canine wandered off as well, but it did have the decency to leave its former domicile behind, the alveolus (aka hole). It was evidently a large tooth and may have been double-rooted. Behind this come a pair of alveoli for the first premolar; a similar length as for the p2. These holes have been plugged up with bone growth, a primitive sort of habit displayed by a variety of basal mammals; morganucodontids, Kuehneotherium, Hadrocodium, sometimes Dinnetherium and docodonts. As the jaw grew bigger, this position got converted into a diastema (p.80). The second premolar took a more dutiful approach. It's a simply built, double-rooted tooth with a main cusp at the middle. A small colleague occurs at the front of the crown. An even weedier effort at the rear appears to have shrunk somewhat from wear. The base on the buccal is a touch thickened, whereas a small cingulum occurs lingually. That becomes larger on teeth further back along the row. Premolar number three is a bit larger than p2. Its lingual cingulum is divided into two areas by a ridge. Molars The double-rooted m1 is long and five-cusped. The largest of those, known as a, stands proudly on the space between both roots, and is recurved. The base of the buccal slope of cusp a is prominently bulged, as for all molars, a similarity known from Comodon of North America's Upper Jurassic. The labial cusp base has a pair of concavities. Cusp b has been cut off in its prime, meaning it's broken, but the remains indicated it was larger than cusp c. Both are cone shaped and separated from the main cusp by shallow notches. The five-cusped style is achieved, in contrast to the premolars, by accessory cusps at the front (d) and rear (e). These are lingually situated; not aligned with the main cusps. Contact between molars in the same row is of a fashion known as echelon; d overlaps the tooth in front lingually, and e does so buccally with the following one. No sophisticated interlocking system is developed. With the exception of the fifth, all molars have a similar morphological build, with the main difference being size. Somewhat unusually, this decreases consistently along the line from front to rear. The m5, which was only starting to erupt, has a tri-cusped crown supported by a single root. That, in combination with the plugged p1 alveolus, suggests the tooth was delivered fairly late during a lifetime. Jaw Three mental foramina are found along the front part of the outer surface of the dentary and, behind them come three rather suspicious grooves. The owner presumably wasn't very happy about those. They seem to be bite marks. Affinities The molars have the qualifications required for membership of "Amphilestidae" (p.81); cusp a is dominant, b and c are close in size, and cingular cusps are small. However, the main characteristics of "amphilestids" arose early in the mammalian career. Dinnetherium, dating from the Lower Jurassic, has them. An analysis of the data matrix indicated the closest match lies with an Upper Jurassic amphi named Comodon. As stated, however, there's no formal referral of Hakusanodon to any family. Holotype SBEI 006 works as a licensed-to-kill secret agent for the Shiramine Board of Education, Ishikawa Prefecture. Its remit includes punishing inattention and truancy, so local students have now been warned to behave! The specific name derives from the Greek archaeus, 'old'. Additional notes With thanks to Dr Guillermo W. Rougier for some background information and a copy of the eventual paper.
Reference:	Rougier et al (2007), An Early Cretaceous mammal from the Kuwajima Formation (Tetori Group), Japan, and a reassessment of triconodont phylogeny, Annals of the Carnegie Museum, 76(2), p.73-115.

Genus: Ichthyoconodon Sigogneau-Russell D, 1995 'fish coned tooth' Remarks: Gaetano & Rougier, 2011 report their analysis places this poorly known genus into the 'amphilestid' subfamily of Alticonodontinae (p.839). That's why I've placed this entry into this directory.

Species:	Ichthyoconodon jaworowskorum Sigogneau-Russell D, 1995
Place:	Anoual
Country:	Morocco
Age:	Berriasian, Lower Cretaceous
Remarks:	According to Rose et al, (JVP 21(3), 2001), the affinities of this genus are also enigmatic. (With thanks to Vince Ward and David Marjanovic).
Reference:	Sigogneau-Russell (1995), Two possibly aquatic triconodont mammals from the Early Cretaceous of Morocco. Acta Palaeontologica Polonica, 40(2), p.149-162.

Genus: Juchilestes Gao CL. Wilson GP, Luo ZX, Meng O & Wang X, 2010 'Sawtooth robber' Remarks: The generic name derives from Ju chi, Chinese for the tooth of a saw, and this refers to the line of triconodont cusps on molars. I originally had the year of publication down as 2009, which is the date on my copy. Officially, it appeared in print in 2010.

Species:	Juchilestes liaoningensis Geo, Wilson, Luo, Meng & Wang, 2010
Place:	Yixian Formation, Liaoning
Country:	China
Age:	Lower Cretaceous
Remarks:	The following is based upon my reading of Gao et al, 2010. I happen to have a pre-publication copy from the end of 2009, and the pagination deviates from the official version. It simply runs from 1 to 10. Among the relatively few humans with an interest in such things, China's Yixian Formation is justly famed for, among other treasures, unusually large to vast killer mammals from the Lower Cretaceous. These include the respectably dog-sized thug Repenomamus giganticus, the biggest known Mesozoic mammal in the world. Such actors, however, are celebrities from the lower Yixian fauna. The upper Yixian cast found so far could all have their life-sized (or death-sized) potraits, including all scrumptious details of their delightful bodies, comfortably accommodated on a single sheet of A4 paper. And there'd be ample room left for an expressive signature, informative notes and a number of games of noughts and crosses. That is far more typical for Mesozoic mammals, although the completeness of some of the specimens is anything but normal. A welcome enrichment of the upper fauna, Juchilestes, is less than complete. It's presently known from a skull with its lower jaws. Like all the earlier lower Yixian fauna so far it's a eutriconodont killer. With a skull length of a touch over four centimetres though, it was a smaller mammal than any of those. The locality it's from, Lujiatun, is a deviant for the upper Yixian fauna in that it yields fossils preserved in 3D rather than flattened to pancakes. Amphi-up The molars of the new genus are of a style known as amphilestid, and that used to be seen as being a sufficient basis for a family, Amphilestidae (p.1). Increasingly, that "family" has been suspected of and then finally outed as being a paraphyletic hodge-podge of mammals sharing similarities based on a deeply primitive arrangement in disparate lineages. There was also a problem of rampant lower-ism. Upper molars couldn't be taken much into account as known 'amphilestids' had failed to preserve any. Juchilestes managed to retain evidence of its full allotment and, for good measure, much of the skull. Taking advantage of the newly available details, the authors tested some new and old issues. A new matter concerns the affinities of this 'saw tooth'. Of known suspects its closest relative is Hakusanodon from Japan, and that implies kinship with only some taxa referred to over time as amphilestids. Far, far more distant is a relationship with the Middle Jurassic 'amphilestids' of England; eg. Amphilestes. Consequently, the alleged family of Amphilestidae, as formulated, isn't a natural taxon, and (when practical) will require redefinition. Also in doubt is the integrity of Eutriconodonta. However, that's seen here as unresolved. Talking Eutriconodonta blues This category presently provides a home for at least four families; Amphilestidae, Triconodontidae, Gobiconodontidae and Jeholodontidae (p.2). The first of those is composed of mammals sharing a similar style of lower molar. The central cusp is dominant, two smaller cusps (one to the fore, one behind) accompany it, and there's no "tongue-in-groove" interlocking mechanism with neighbouring teeth. Gobiconodontid arrangements are similar, but their teeth are robuster. Testing consistently supposrts the monophyletic credentials of Triconodontidae and the Gobi family, and consistently fails to do the same for the amphis. Being so disgraced, that 'family' is now regularly slurred with inverted commas representing its hoggle-poggle nature; 'Amphilestidae'. The jehols, should you be wondering, have impeccable familial credentials, but it only contains two genera, both of which are from the upper Yixian fauna A brief word from WTF Please, please, please, somebody find a Yixian trity. The presence of a Yixian friend for Japan's Hakusanodon, and the occurrence in both faunas of eobaatarid multituberculates, are two candles of faunal similarity that keep a cherished dream alive. We, if within our capabilities, should be readily eternally grateful for the discovery of a non-mammalian tritylodontid from the Lower Cretaceous of China. The People's Republic has managed Upper Jurassic tritys. Siberia and Japan have provided teeth from the Lower Cretaceous. Although, as yet, we know of not the slightest hint of such a critter, we'd love to hear of a Yixian trity, preferably one that's complete and furry. This plea has been financed by the Web of Trity Fans. Back to 'saw tooth robber' The jaws were CT-scanned and no indication whatsoever was found of any unerupted or replacement teeth still lurking in its crypts. That shows that this animal was an adult, with its full and final set of choppers, when something rudely curtailed the beating of its passionate heart. Teeth The dental formular per side is: Uppers: 4 incisors, 1 canine, 3 premolars and 5 molars. Lowers: 4, 1, 3, and 6 respectively. With reference back to Japan, Juchi bears most resemblance to Hakusanodon. Common characters include the presence of only three premolars, a decrease in size of the latter molars, and the near identical shape of molar cusp outlines when seen from both the internal and external views. However, the front cusp on lowers (b) is taller than the rear cusp (c) on Juchi's final premolar whereas, for Haku, the reverse applies. The a cusp rises vertically for Juchi but slants to the rear for Haku. Furthermore, all Haku's postcanines are smaller by 16 to 20% (p.4). Head lines A more detailed description of the skull is planned for a later date, although some points are mentioned along with the fuller treatments for the lower jaw and teeth. Following 'saw tooth's death, nature contrived to narrow it's skull by squashing and adding some distortions (p.5). Nevertheless, its clear the original length was a touch over 4cm. If compared to a standard mouse or many Mesozoic mammals, that's quite large, but it would've qualified as a smalling in the known weird world of the lower Yixian. Lower jaw The dentary bone is relatively deep beneath the teeth. For example, there's room for about one-and-a-half crown heights beneath the m2 position, and the depth below the following two teeth is somewhat more. The internal side features a wide groove for a Meckelian cartilage running forewards until it reaches the level of the p3 premolar. There were at least four foramina (small holes for nerves and other such cables) in the wall of the jaw; two small ones beneath the canine and others below the p2 and p3. Back from the tooth row, the coronoid process rises diagonally to reach far above the height of the teeth. Lower front teeth Alveoli are present for four incisors with three of those on right side in place. All are implanted a bit procumbently. The first two are small with the third, as judged from its alveolus, presumably similar. The i4 is larger, but not by all that much. The canine is tall, easily the tallest tooth in the whole set, thick and set vertically in the jaw. It's single-rooted. This front part of the dentition is accused of being similar to that of Phaslascotherium. A diastema interrupts proceedings between the canine and premolars and, as revealed by CT-scan, this vacant area has from resulted one (or perhaps two) previous teeth having been dispensed with. The areas where roots once plied their trade have been plugged with bone. Two former alveoli are involved. Lower premolars There are three premolars per side, a count also known from Hakusanodon. The p2 is dominated by cusp a set slightly forwards of the middle of the crown. Cusps b and c are present but small. There are no cingulids. A short gap occurs between this tooth and p3, a premolar implanted higher up on the jaw at the same level as the molars. This p3 is taller, longer and wider than the tooth in front, and its b cusp is more distinct. Lower molars Each side features a set of six. The first differs from the final premolar by being larger, having roots that go vertically down rather than splaying, and having cusps b and c still more distinct. Of those, b is somewhat taller. There's also an interesting difference evident in the degree of wear. The first molar is more worn than either the p3 or m2, and that indicates it erupted prior to both of them as, indeed, did my m1. That's usual for us toothy mammals. It's also a good clue for the identity of the tooth in conjunction with its morphology. The first pair of molars are the longest in the series, with the size then decreasing along the line (p.7), gradually at first but then more radically for m5 and 6. Upper teeth, incisors and canine Each side featured four incisors, although only I3 and 4 have retained complete crowns. Remains of the first two show them both to have been smallings in comparison to the better preserved pair. A groove comes behind the incisors so as to accommodate the upper part of the lower canine. Its upper counterpart is a robust, single-rooted tooth with its root about 75% the length of its crown. Upper cheek teeth In case anybody's forgotten, and seeing as this could form the basis for a question on a tv quiz show soon, Juchilestes had three upper premolars and five molars. That's a total of eight postcanines, a relatively high number for known eutriconodonts. No diastema separates the double-rooted P1, a small tooth, from the canine. The second premolar is about twice as long as the first. It's thin with a low, centrally positioned A cusp, and little attempt at further cusps. There are bumps, however, in positions appropriate for B (front) and C (back). Similar modest prominences also grace the crown of P3 but, in that instance, their positions are buccal in comparison to A, thus resulting in a degree of triangulation. As for the molars, the second and third have better formed B and C cusps than the first displays, itself a wider tooth than the premolars. Both can also boast about a cusp E on the external cingulum. Beyond that pair crown morphology increasingly simplifies for the M4 and 5, with those last two teeth being set further inwards towards the tongue than the other molars. The M5 is merely half the length of its upper colleagues. For these molars as a whole, a bit more so for uppers than the lowers, there's a modicum of triangulation. As cusps B and C are slightly external of A, this is a minor case of obtuse-angled triangulation. The style of occlusion for upper and lower teeth is known as embrasure occlusion. That means the upper cusp A fits between cusp c at the rear of a lower molar and cusp b of its follower. This system is also known from other 'amphilestids', gobiconodontids and obtuse-angled 'symmetrodonts'. It doesn't occur for Jeholodens, triconodontids or dear old ancient Morganucodon. Wide obtuse-angled triangulation The earliest outbreak of this mammalian obtuse-angled trigginess is known from ancient critters such as Kuehneotherium. Kuehneos were clearly not members of crown-group Mammalia. Consequently, other crown-group obtuse-angled 'symmetrodonts', for example the spalacotheriids, can't have inherited that condition via descent from kuehneos (p.9). And this trait provides no substantial reason for pleading intimate kinship between spalacos and this Chinese 'saw tooth', especially as there are plenty of reasons for ruling that out. It's held independently as convergence in different lineages, and in combination with so called embrasure occlusion of upper and lower teeth. Both spalacos and Juchi have that in common too, as do other mammals. Family values? No thanks! Analysis of characters from a range of mostly triconodont mammals left the supposed family of Amphilestidae looking severely dysfunctional (p.8). 'Saw tooth' enthusiastically shacked up with Comodon and Hakusanodon, and that whole trio voiced close affection for Aploconodon and Amphidon. However, as they expressed the opinion that Amphilestes was so distantly related so as to be out of sight, they wanted no place in a family named after such a rum duck. Holotype The type fossil, D2607, performs before the admiring public at the Dalian Museum of Natural History in Liaoning Province. That final information leaves you equipped to figure the meaning of the specific name out for yourself.
Reference:	Gao et Al (2010), A new mammal skull from the Lower Cretaceous of China with implications for the evolution of obtuse-angled molars and 'amphilestid' eutriconodonts, Proceedings of the Royal Society B, 277, p.237-246.

Genus: Kemchugia Averianov AO, Skutschas PP, Lopatin, AV, Leshchinskiy SV, Rezvyi AS & Fayngerts AV, 2005 'from Kemchug' Amphilestidae Osborn, 1888 Remarks: The following is based upon my reading of Averianov et al, 2005. The original entry for this genus mentioned my lack of information, and a request for somebody to forward a copy of the paper. This was seen by Dr Averianov and his response was most generous. Many thanks. The generic name honours the Bol'shoi Kemchug River of Western Siberia, which kindly assists by providing canoe access to the shy, remote fossil locality in an otherwise stubborn forest. Known remains aren't exactly plentiful (p.3). There's certainly an upper tooth and most probably a fragment of a lower. So much for numbers.

Species:	Kemchugia magna Averianov et al, 2005
Place:	Bol'shoi Kemchug 3, Krasnoyarsk Territory, Western Siberia
Country:	Russia
Age:	Lower Cretaceous
Remarks:	Upper molar snippets Basal features include (p.3) the relatively straight alignment of the three main cusps along the central length of the crown, the lack of an ectoflexus bay, the convexity of the face on the lingual side, the presence of two cuspules at the front (termed E and F) separated by a small bay, and complete cingula along both the buccal and labial sides equipped with some cuspules. Derived features include the high main cusp (A) and cusp E being aligned with the main cusps. F is buccal from E. I'm actually slightly unclear on a couple of points, as the wording seems contradictory. While cusps B and C being: "about half as high as cusp A" is interpreted as basal, their close similarity in height is stated to be derived. Perhaps it would be clearer in Russia... but not to me. Anyway, if that some of that language seems off-puttingly esoteric, you're allowed to forget it. There won't be a quiz, and it's unnecessary to understand every single word. If you could read scientific descriptions using my brain, then you'd be use to relaxed yet engaged incomprehension. Comparisons Attempting to compare the upper molars of amphilestids presently poses a slight problem, and this lies in their severe shortage. So far, strictly interpreted amphilestids have tended to avoid providing uppers. The relatively pronounced height difference between cusp A compared to B and C is a similarity with Gobiconodon, but that genus favoured an ectoflexus bay and also had a concave area centrally located on the lingual side. That isn't the case for Kem. Repenomamus had been busy reducing cusps B and C and the cingulum of the buccal side. Furthermore, it had also been inflating its main cusp. Kem hadn't. Morganuc O'Dons? Originally, the upper molar from Bol'shoi Kemchug 3 had been tentatively referred to the ultra basal mammalian family of Morganucodontidae, albeit with a question mark. It's broadly similar, but cusp A is more dominant, and B and C are closer to one another in height. There are also indications that the pattern of occlusion differed, as is known to be the case for morganucodontids and amphilestids. In the case of the amphis, cusp C left wear on the front labial part of cusp a on the following lower lower tooth. Meanwhile, upper A interacted with lower cusp c. This is termed "embrasure occlusion. Morganucs settled for direct cooperation between the corresponding uppers and lowers; M1 with m1 and so forth. The wear present on Kemchugia points in the amphilestid direction, both for the holotype and referred lower. Roots and dimensions The upper molar was supported by a pair of long roots, both of which are preserved. With the width of the crown amounting to 65% of the length, it's relatively wide. Due to incompleteness, the length of the lower can't be fully measured, but what is preserved suggests it was about the same as for the upper, and that would make it proportionately narrower. Upper molar: length 3.2mm, width 2.1. Possible lower molar: width 1.3mm. Other local mammals A couple of fragments of dentary, and an isolated lower molar, point to the presence of a second, smaller amphilestid in the fauna. These fossils are described and discussed in the paper, and assigned to Amphilestidae indeterminate. Also present were at least two species of Gobiconodon. Holotype The type fossil, PM TGU 16/6-401, us an upper right molar attending Tomsk State University. The specific name is Latin for 'great', and this celebrates its enormous dimensions. A third of a centimetre might not be much of a length for a man to boast about, but it's relatively huge for an amphilestid. Additional notes The first time I heard of this genus, the information came from the Paleobiology Database, so thanks for the information. That resource is supported by some most illustrious names (according to its pages), so it should prove very useful. Still, there's no according to this critic, my jokes are generally better. You can compare them, should it appeal: Paleobiology Database.
Reference:	Averianov et al (2005), Early Cretaceous mammals from Bol'shoi Kemchug 3 locality in West Siberia, Russia, Russian Journal of Theriology, 4(1), p.1-12.

Lower Cretaceous eucynodonts of Western Siberia This article is also loosely based on my reading of Averianov et al, 2005. I grew up in Dorset in southern England, and was often underway on a bike. A proper one, not one of those noisy things with motors. It's almost impossible to get more than five miles away from at least a substantial village. When human legs served as Shank's pony for transporting tradables and saleables to and from market then, even in reasonably flat areas, living more than five miles from a market would have been inconvenient. In the hillier north of the county, crows resting midway between two erstwhile markets are more likely to have only three or four miles further to flap in either direction. That's what I was used to. Scotland, which begins something like 350 miles to the north, almost qualified as being on a separate continent. This background leaves me struggling to grasp the scales involved in Russian geography. It might be thought that Moscow is near the eastern edge of Europe. However, check a map and you'll find the Ural Mountains are some 1,500km further east. That's a serious tract of territory. To reach the city of Tomsk, in western Siberia, you'd still have to cycle on for around another 1,500km. That brings us somewhere nearish to the Bol'shoi Kemchug fossil locality. Kind of nearish. It's only about a further 400km east. However, you won't be able to cycle all the rest of the route. Trees beyond trees According to some scriptwriters, Siberia is a boundless tundra stretching beyond the horizons of even the imagination into yet more boundless tundra. Bits of it are. There also happen to be boundless forests stretching beyond the totally obscured horizon into more boundless forest, and Bol'shoi Kemchug 3 is stuck in the middle of one of them. This is the taiga counterpoint to the tundra permafrost. Should you get there, then you could strike up a conversation with a tree seeing as you're not likely to find a human to chat with, excepting for any that you might've brought along. The nearest road is impossibly far away, and the nearest settlement is presumably somewhere further along it. And yet, despite the remoteness, paleontological researchers from Tomsk State University found this fossil site in 2000 (p.1). How to find a fossil site in the middle of a deserted nowhere Reaching this location involves a great deal of paddling in canoes (p.2), and that's what a joint expedition from Tomsk and Saint Petersburg did in 2001 and 2003. This is the only practical means of transportation, and it places severe restrictions on the amount of supplies that could be taken along, and the quantity of material that could be brought back. The latter point helps explain why the number of specimens acquired was relatively low. Matrix had to be processed on site; 470 kilos in the first instance and 750 during the second. While bringing out tons of the stuff for later examination might sound like a good idea, as this would result in sub-aqua canoeing, you'd be left down in the creek with or without a paddle. The convenient river happens to be a small one. While that reveals how the locality was reached, it also leaves an interesting question open. How on Earth was the place discovered? I couldn't resist asking Saint Petersburg, and Sasha took a bit of time to explain. The fossil locality in the middle of the taiga was essentially discovered by studying maps. A geographical map showed cliff exposures in a remote though accessible place, and a geological map provided information on the extent of the Ilek Formation dating from the Lower Cretaceous. Each supplied complementary strokes for an 'x' marking the spot of a likely fossil site at Bol'shoi Kemchug 3. It was a case of using all available resources intelligently before attempting to stick your oar in. Subsequent prospecting showed the 'x' did indeed mark a good spot for hidden treasure. And so, later, paleontological pirates came a-plundering. A well behaved formation The Ilek Formation happens to behave itself in a helpful, consistent manner. It had already lifted its veils from two faunas with eucynodont remains from Kemerovo Province, at least one of which had been known since the discovery of two Psittacosaurus dinos in 1953. (I just had occasion to check something in Maschenko et al (2002), and that's stated on page 75.) These localities are Shestakovo 1 and 3. As the trio are all of much the same age, the vertebrate faunas are also broadly similar. However, the restricted sampling possibilities at Bol'shoi Kemchug have resulted in more limited returns; a couple of dozen vertebrate taxa: fish, amphibians, various reptiles including pterosaurs, dinos -with an indeterminate bird, Xenocretosuchus and at least five mammals (p.3). Eucynodonts of Bol'shoi Kemchug 3 and similarly aged Siberians There are now three known Lower Cretaceous eucynodont localities in the Ilek Formation; Bol'shoi Kemchug 3, and Shestakovo 1 and 3 (p.10). The fact that they're in the same formation provides a rather strong clue that they're geographically fairly close. Shestakovo is in Kemerovo Province and somewhat to the south. (To put it another way, east of Novosibirsk.) Being about the same age and in much the same area, faunal overlaps should be expected. Indeed, as far fewer specimens have been collected at Bol'shoi Kemchug 3, 'absentees' could simply reflect sampling bias. Such taxa could have been resident but not yet found. An overlap with at least one (perhaps both, I'm not sure) Shestakovo locality is a gloriously late non-mammal tritylodontid. This is Xenocretosuchus, one of the last non-mammalian cynodonts to have graced the face of the Earth. Even pretty good paleo books for the general reader may well relate that these critters died out during the Middle Jurassic. That conclusion was reasonable enough a decade ago, but then Siberia's Lower Cretaceous superstar eloquently objected. The chorus has since been joined by friends from Japan. Squeaks of support have also emerged from Upper Jurassic finds at the Junggar Basin of China and, in June 2007, southwest Mongolia. Another intriguing peculiarity of the faunas is that eutriconodonts are the most common mammals. Of the taxa known from Bol'shoi Kemchug 3 so far, the triconodont element comprises 100%. That surely can't reflect the ecological reality. Nevertheless, eutriconodonts are also in the majority at Shestakovo (p.11). Interestingly, to me at least, is that something similar can be said of the lower Yixian fauna in China. Perhaps this will turn out to be a coincidence of no particular significance, but it isn't something that generally applies to northern hemisphere faunas of this sort of age, or other ages. I'm still hoping the Yixian will one day deliver a trity, but there's been no sign of any thus far. Will somebody pretty please dig one up? As the Shestakovo sites have been more extensively sampled, it's hardly surprising that their mammal diversity is greater than Bol'shoi Kemchug. The second most common taxon at Shestakovo 1 is a tinodontid 'symmetrodont'. Rare remains of a docodont, Sibirotherium, also occur. A pretribosphenic mammal, Kiyatherium, has been arrested for limited loitering at both the Shestakovo localities. Amphilestid triconodonts haven't been found there with certainty, although one tooth may perhaps indicated a limited presence. Further Mesozoic site summaries can be found at Localities. Meet the Bol'shoi 3 eucynodonts Non-mammals Tritylodontidae: Xenocretosuchus sp. Mammals Triconodonta (Eutriconodonta): Kemchugia magna, amphilestinae indet., Gobiconodon sp. A, G. sp. B, Mammalia indet.

Genus: Liaotherium Zhou M, Cheng Z & Wang Y, 1991 'Liao beast' "Etymology: Liao [= 'far away'] is the abbreviation for Liaoning [= 'far away' and 'quiet'] where the fossil was collected," (Zhou et al, 1991). Family: Amphilestidae? Osborn, 1888

Species:	Liaotherium gracile Zhou M, Cheng Z & Wang Y, 1991
Place:	Jiulongshan Formation, Liaoning
Country:	China
Age:	Callovian?, Middle Jurassic
Remarks:	Whilst best known for fossils from the Lower Cretaceous, Liaoning is a geographical term, rather than a geological one. The Langi Formation, which overlies the Jiulongshan, has been K--Ar dated to about 158Ma (Wang DF, 1984). Studies based on conchostraca, aka clam shrimps, (Wang SE, 1984), and insects (Hong, 1984) support a Middle Jurassic diagnosis. This particular location seems to have had continuous deposits from at least the Mid J until the Lower Cret. With thanks to David Legg for alerting me to the genus and David Marjanovic, who supplied the English summary of the Chinese publication. Reportedly, In the Shadow of the Dinosaurs, 1994, this is regarded as a nomen dubium. (With thanks to Vince Ward). This phrase from the original publication might give a clue as to why: "as was observed in the field when it was first exposed, this molar has three cusps arranging in a line." This perhaps suggests that such a feature can no longer be seen, for whatever reason. According to Averianov 2002, (p.712), this could equally well be a "symmetrodont" of some kind or other. The main reason for its diagnosis as an amphilestid was its possession of a line of three cusps. A similar condition is also known from non-amphilestids, such as on the last molar of Gobitheriodon.
Reference:	Zhou M, Cheng Z & Wang Y, (1991), A note on a Jurassic mammalian lower jaw bone from the west of Liaoning Province. Vertebr. PalAsiatica 29 i.

Genus: Phascolotherium (Broderip, 1828) Owen, 1838 aka: Didelphys Broderip, 1828; Thylacotherium Valenciennes, 1838 Family: Amphilestidae Osborn, 1888

Species:	Phascolotherium bucklandi (Broderip, 1828), Owen, 1838
Place:	Stonesfield slate, Oxfordshire
Country:	England
Age:	Bathonian, Middle Jurassic
Remarks:	Rougier, Isah & Manabe, 2007 contains some information on page 80. The genus was described as having only two premolars per lower jaw half, but that turned out to be not quite correct. A more recently prepared juvenile has a tooth housed in what would have become a diastema later in life. It's only get itself to blame for that not happening. Sheer carelessness. So, originally, a premolar developed there, but it was dispensed with as the animal's thoughts turned towards having sex, a strict scientific definition of the word 'adult'. Additional comments One specimen, BMNH 112, resides in the Natural History Museum, London. Several jaws are studying at Oxford University.
Reference:	Owen (1838), On the jaws of the Thylacotherium prevostii (Valenciennes) from Stonesfield. Proceedings of the Geological Society of London 3, p.5-9.

Links: University of Oxford, UK http://www.earth.ox.ac.uk/Research/furtherinfo.htm I can't presently locate this link, which is a pity. It included a photo. However, this specimen catalogue has appeared: http://www.oum.ox.ac.uk/onlinedb/geotypes/geosearc.htm. Shikha Chanda, Dinodrastic! http://www.chanda.freeserve.co.uk/oxygen.htm I'm not sure I accept the theories, but there's a nice illustration.

Genus: Tendagurodon Heinrich WD, 1998 'Tendaguru tooth'

Species:	Tendagurodon janenschi Heinrich WD, 1998
Place:	Middle Saurian Bed, Tendaguru
Country:	Tanzania
Age:	Kimmeridgian-Tithonian, Upper Jurassic
Remarks:	The following is largely based upon my reading of Heinrich, 1998. The German Tendaguru excavations lasted from 1909 until 1913. This activity resulted in a mass of material, much of which found its way back to the Humboldt University Museum in Berlin. Although first described 85 years later, that's how Tendagurodon also came to light. About 500kg of matrix was dissolved in acid at the museum, in order to find out what fossils it contained, (p.271). The answer was all kinds of stuff. Meet the tooth Amongst the finds was a probable lower molar. It's double-rooted and from the right side of the jaw. The crown is long and narrow, and adorned with a nearly straight line of three main cusps, (a,b,c) and two accessory ones (d at the rear and e at the front). This procession is slightly ( lingually) positioned. The comparative size order is a > c > b > d > e, (p.272). The central cusp a is easily dominant, with its peak near the midpoint of the line. The rearmost of the trio, c, is closer to a than b is, and also in a more elevated position, (p.273). V-shaped vallies are formed between the descending slopes of the main cusps and the ascending ones of its assistants. The d accessory cusp is a bit larger than e, which juts out to overhang the front of the tooth, a bit reminiscent of guttering in profile extending beyond a roof. Wear facets aren't apparent on the crown, but the tip of cusp a has been blunted. Singularly underequipped and other details. From p.274 comes: "A striking feature of the Tendaguru triconodontid is the absence of well-developed cingula". A cingulum is a kind of girdle which may surround the base of a mammalian tooth. They can be handy for storing further accessory cusps on, among other uses. 'Tendaguru tooth' wasn't an enthusiast. There are faint cingula beneath cusps b and e. Otherwise, there's a cingula lack of interest. Also lacking is any hospitable hollow on the front of the crown. Many triconodonts had such devices to cater for protruberances on the back of the preceding teeth. These interlock with each other and provide for stability along the row. This mammal didn't bother with anything of that nature. The maximum length and width of the specimen are 2.5mm and 0.7mm. Affinities It's relatively straightforward to understand what this genus isn't. The three main cusps of triconodontids are broadly similar in height with one another, (p. 281), in contrast to Tendagurodon. The relative proportions also differ to gobiconodontids (a > c ~ b), austrotriconodontids (a > b > c) and amphilestids (a > b ~ c). The scheme here most closely resembles the morganucodontids. However, the relevant positionings of the cusps to each other are different, and they generally have a lingual cingulum with cuspules on their lower molars. There's also no sign of a cusp g (kuehneocone) on this tooth. It's not a late (and unusually large) morganucodontid. Tendagurodon is some kind of tricondont tooth or other, which seems to have maintained some basal characteristics. Holotype The holotype and only specimen known so far is called MB.Ma46911. It works at the Museum of Natural History, Humboldt-University. The species name is in honour of Werner Janensch, who was the leader of the German expedition. He had a good innings, (1878 -1969). Neighbours This location is far better known for its extensive dinosaur fauna. Remains of Brachiosaurus, Dicraeosaurus and Barosaurus are relatively common. Also recovered have been fossils from theropods, ornithopods, stegosaurs and non-dinos such as crocodiles and pterosaurs.
Reference:	Heinrich WD (1998), Late Jurassic mammals from Tendaguru, Tanzania, East Africa. J. of Mammalian Evolution, 5, p.269-290.

Genus: Triconolestes Engelmann GF & Callison G, 1998 'three coned thief' Family: Amphilestidae Osborn, 1888

Species:	Triconolestes curvicuspis Engelmann GF & Callison G, 1998
Place:	Dinosaur National Monument, Utah
Country:	USA
Age:	Kimmeridgian, Upper Jurassic
Remarks:	The following notes are derived from my reading of Engelmann & Callison, 1998. The holotype, (the only known specimen), is a fragmentary lower ?left molar known as DINO 10780. "Although we are loathe to make a single fragmentary tooth the holotype for a new species and to create a new genus for it, there is enough of the specimen to recognize its probable affinities as being within the amphilestid triconodonts and that it differs from other known taxa in ways typically distinguished at the level of genus", (p.346). The preserved portion shows three cusps but there's no indication of a cingulum. It probably had two roots. The cusps display a slight curvature, (probably on the lingual -'tongue'- side), which suggests this fragment is the front part of a tooth. Comparisons with other triconodont specimens indicate it's likely a left molar.
Reference:	Engelmann & Callison (1998), Mammalian faunas of the Morrison Formation. Modern Geology 23, p.343-379.

Genus: Volaticotherium Meng J, Hu Y, Wang Y, Wang X & Li C, 2006 'gliding beast' Order: ?Volaticotheria Meng et al, 2006 Remarks: The generic name refers to the critter having been a gliding beast. Update A 2011 study has outed this genus as an "amphilestid". Consequently, I've now moved this entry accordingly but haven't changed anything in the entry. Of Mice and Men This is an attempted sketch of the increase in knowledge concerning the diversity of lifestyles of Mesozoic mammals, and it's been provoked by the publication of the fabulous Volaticotherium. Information centred on that fossil is below. Thanks are due to Mr John Steinbeck, who has agreed to help out with the title and subheadings. At least, he hasn't objected. I must also thank David Marjanovic for spotting the unfortunate error. I forgot to check a detail, and that oversight turned out to be a spoiler and a half. In Dubious Battle In around 1820, those who knew their stuff generally agreed that mammals hadn't existed during the Mesozoic, 'The Age of the Reptiles'. This was a sensible enough belief given the lack of evidence for supposing the contrary. William Buckland had a couple of reasons to doubt its correctness; small slivers of Middle Jurassic mammal jaws given to him by a student in 1812. I don't know whether he quite realized their true nature at that time, but he certainly did in 1824, as Buckland mentioned them in his 'Notice on Megalosaurus'. Still, some people were determined to maintain a mammal-free Mesozoic, and it took much of the 1830s before the excessive scepticism fully dissolved. Mammals had lived alongside the giant saurians and they were obviously, judging by the then available evidence, small insectivores and presumably marsupials. So was it thought. Despite that marsup conclusion being gloriously wrong, the 1857 description of multituberculate fossils from Dorset by Hugh Falconer appeared to strengthen it. Multis are characterized by having numerous bumps on their molar teeth, and that also happens to apply for the extant Kangaroo rat (Hypsiprymnus ). Being aware of this, Falconer recommended that his new friend, Plagiaulax, enjoy a similarly plant-based diet. This was disputed at the time by Richard Owen who argued for meatier tastes, but a rodent-like approach to slack herbivory does seem the likely option. This was an enormous leap forward for Mesozoic mammals. It was now known they were dead European marsupials consisting of small bits of jaw. Some favoured grubs while others grubbed around for planty favours. And, amidst great excitement, another leap leapt in 1859. Owen described jaw remains which, rather than being tiny, were vast; comparable in size with rather small cats. Some Mesozoic mammals obviously messed up vertebrates. Travels with Charley Gradually, material started to accumulate from wider afield, especially in the Mid West of North America. As these collections came to include actual Cretaceous marsupials, they eventually helped to de-marsupialize the non-marsups. A clearer picture began to emerge but, as little more than jaws and isolated teeth were involved, it didn't add much depth to the information on the diversity of lifestyles. While the geographical range was on the way to becoming global, typical Mesozoic mammals largely remained bits of small jaw with occasional forays towards small cat dimensions. It was a bit like owning a Game Boy in a world without batteries. It wasn't really until the 1970s that more extensively preserved and informative skeletons began to be described, and most came from Mongolia. (There were some earlier exceptions to this, but not very many.) And, over the 1980s and the first half of the '90s, a few mammals began to lose some of their anonymity as regards behaviour. A Mongolian multi, Nemegtbaatar, was publicly accused of being a generally slow mover capable of short bursts of more vigorous hopping. You need legs to reach conclusions like that, and this animal had some. Furthermore, Mongolian multis were accused of having pelvises too narrow to have allowed platypus-style reproduction; ie. egg-laying. That suggests either uncommonly small eggs (not impossible) or live babies (not impossible). Europe also became reinvigorated with a sumptuous series of Iberian contributions, most of which came for a Jurassic coalmine in Portugal. Guimarota managed to provide a partial skeleton of what seems to be a small, arboreal insect hunter named Henkelotherium. It also produced much of a docodont worm miner, Haldanodon, an animal suspected of semi-aquatic tendencies. It somewhat resembles a modern European swimmer, the desman. In earlier times, sensibly attempting lifestyle assessments as specific as those could hardly have been done as convincingly, and the scrappy Mesozoic fossils could only have laid the blame upon themselves. The much better material from Mongolia and Iberia was like manna from heaven; a paradisiacal Garden of Eden for paleontology. Or so it appeared. East of Eden The Province of Liaoning is in northeastern China, and it had never been the kind of place to receive invitations to 'celebrity' television shows. History happened there with a small and quiet h, or sometimes not at all. Miners mined in small mines, farmers farmed small farms, and a group of local musicians musicated musingly, generating national levels of complete indifference. Their audience doubtlessly loved them, but it was a small and spacious crowd. One of their number, Mr Zhang He, made bonsai gardens as a side line in his spare time. For this purpose Mr Zhang required stones, and Liaoning has plenty of those. His search for attractive stones brought him into contact with fossils and, after a few years, Liaoning managed to send the global paleontological community babblingly and delightedly gaga. The fossil fields yielded ridiculous quantities of simply stunning superbness. Zhang He's fossil collecting became prodigious beyond prodigy itself, and several of his finds ignited the paleontological equivalent of an on-going gold rush. One of his first children to take to the wing was Confuciusornis, the most complete Cretaceous bird the world had ever known. Archaeopteryx von Solnhofen is understandably famous, and joy jumps with every new specimen. Around ten have hatched over the last 150 years. Hundreds of Confuciusornis have flocked from Liaoning in the last 15. And that was merely the prelude to a stampede of birdy and non-birdy dinosaurs, famously feathery and fluffy. By no means all of those could be termed the chicks of Zhang, but he loaded the gun for firing the opening shot, and also prepared the mammalian barrage. The first blast of that provided Zhangheotherium in 1997, the most complete Cretaceous 'symmetrodont' mammal the world had ever known; ditto for Jeholodens ( Triconodonta, 1999), ditto for Eomaia (Eutheria, 2002), ditto for Sinodelphys (Metatheria, 2003)... Ditto, ditto, ditto. On average, since 1997, at least one startlingly complete mammal specimen gets to establish a new taxon every year. We know the number of bones in the tail of Jeholodens (30), the last meal of a Repenomamus, the qualities of the fur of Sinodelphys. And it's also possible to reach well-grounded conclusions on the reproduction strategy of Eomaia and the sex of Mr Zhang's pet. I experience firm evidence that I'm a male, and it's not much more flaccid for that type fossil. The extraordinary place yielded 125 million year old mammal fur. What more could anybody possibly want? To a God Unknown Well, still older fur. In 2005, a Monopoly Commission enquiry would've concluded that the sole global supplier of Mesozoic hair needed competition, and other areas of northeastern China began to get jealous. However, rather than simply stamping its feet in anger and frustration, Inner Mongolia decided on action. "If Liaoning can come up with Mesozoic hair," it said to itself, "then I'll do so too, and raise them x million years." And, having made that New Year Resolution, it marched forward with Castorocauda in February, 2006. You may recall mention of a Portuguese, possibly semi-aquatic docodont named Haldanodon. Castorocauda ('beaver tail') was also a docodont and undoubtedly semi-aquatic. A sting was delivered with this tail. The preservation is so astounding, that its soft scales are still to be seen. As are traces of its webbed feet. Not even ten gallons of Oil of Ulay will help keep your body looking as young for such a stretch of time. Having opened the year with what is simply the most spectacular fossil of the 21st century, Inner Mongolia decided to try and overturn that decision with a further scorcher from the same formation; Volaticotherium. This individual also took greater care of its hair than any medicated shampoo could ever allow, although the age is presently unclear; presumably younger than 165 million but older than 125 million years. (The more recent part of the range is said to be most likely.) However, this one sent the fur flying by keeping much of it on a flight membrane. This critter was an early mammal glider, and could conceivably have been travelling through the air even before Archaeopteryx von Solnhofen was an egg. If not that, then it's certainly far earlier than the first bats flapped from the proto-belfry. Apart from these incredible fossils, Inner Mongolia has achieved something of tantalising and exciting promise. As was demonstrated in Edinburgh in the days of Burke and Hare during the 1820s, the best chances of finding a corpse go to those with spades in cemeteries. Where there's one, then there'll probably be more. And for that prospect, I offer sincere thanks to the Manufacturer(s) of the planet. From: The Grapes of Wrath (Chapter 11) "And the mice moved in and stored weed seeds in corners, in boxes, in the backs of drawers in the kitchens. And weasels came to hunt the mice, and the brown owls flew shrieking in and out again." The Patron Saint of mammal paleontologists is George Gaylord Simpson, an American researcher, whose tireless quest for knowledge on the roots of his family tree was only curtailed by death in 1984. Back in 1925, he was working with several reasonably complete, but not very well preserved skulls of Cretaceous mammals from Mongolia, and he received the challenge of being the sole author of the multituberculate, Djadochtatherium. Here's how he addressed that in the first page of his study: "The writer is deeply indebted to Professor Osborn and Doctor Matthew for the great privilege of naming and describing this important form." However, while the owner of the skull certainly received a name, the knowledge to be gleaned on its lifestyle was still so sketchy, that its owner remained largely anonymous regardless. A small, probably herbivorous mammal was the richness of its biography. It's not much of an obituary. Nevertheless, it was many miles along Route 66 from scraps of jaw and isolated teeth, and that's why the task was seen as a "great privilege". About a human lifetime later, even more complete specimens from Mongolia and Iberia were reactivating some Mesozoic mammals into dynamic animals, with plausible personalities, and the convoy of appreciation was travelling west at a further lick of speed. And then northeastern China moved things up by a considerable number of gears and, year after year, gives rise to more stunners, with lives you can relate to. Zhangheotheirum has the body (all of it) of a small, terrestrial generalist insectivore, and that's how he earned his living. Why 'he'? How can fossils be sexed? Occasionally, it can be very easy. There's a non-Mesozoic locality called Messel, and it's famous for its exquisite, nigh on 50 million year old mammals. Several of its charming mini-horses and at least one of the bats were female. They were pregnant when they died, and men don't behave like that. And at least one of the local lemurs is male. Like many primate chaps, its equipment enjoyed the support of a penis bone. Perhaps some females enjoyed it too, but somewhat differently. Zhangheotherium is probably a he, and the reason is a spur of bone on the back of its ankle. These occur on monotremes such as the platypus, but only adult males retain and develop them. As they nevertheless frequent the ankles of juvenile platymisses, they presumably were once useful for both sexes. For this reason, it would be unwise to assume it's a sexual characteristic on a Cretaceous mammal known form only one fossil. However, subsequently, a second specimen was found. There's no sign of this ankle spur, and that's likely the mark of a Zhang-she-otherium. Once there was a War The upper Yixian fauna of Liaoning has provided a number of other terrestrial insectivore competitors for Zhangheotherium, but there's presently something of a puzzling shortage of herbivores. Generally, when a fauna is graced by a multituberculate, they're numerous. In this case, there's only Sinobaatar. Perhaps more await. The Yixian actually offers two mammal faunas, and the older of them is also a curiosity. As known so far, it's populated by unusually large, terrestrial, mammalian terrorisers; modest pussy-cat sized giants. The only traces of smaller mammals were in the stomach of one of them, Repenomamus robustus, an abnormally large mammal for the time of Earth. She'd also just eaten a dinosaur baby. Even more abnormal is her sister. Repenomamus giganticus astonished paleontologists with its respectable dog-size. Its head and body length exceed 70cm. Trees appear to have been fashionable for some upper Yixian mammals. Limb proportions and claws of the therians, Eomaia and Sinodelphys, compare closely with some tree-loving mammals of our time, but not with the more radical specialists. The trees probably formed at least part of their realm, but there is a slight snag. While, for example, American opossums are built for trees, not all of them appreciate this fact. You get possums in places with not a trunk to be seen, and they still have a typically possomy build. The new Inner Mongolian, Volaticotherium, is a much more extreme case. Gliding mammals make no sense without trees. While bats can fly above water, plains and deserts, the largest extant gliding mammal would hit an insurmountable problem after a couple of hundred metres. It's a thing known as the ground. These northeastern Chinese localities have already drawn back the curtains of time, and populated the distant past with living, breathing, moving, eating, breeding mammals, and no view like this was available to anybody until just ten years ago. This is a great privilege GG Simpson was unable to enjoy, and it's probably well beyond his wildest dreams, let alone his expectations. The even more incredible aspect is this. This carnival isn't likely to end soon. With any luck at all, there's a heck of a lot more to come. While the Yixian faunas are only challenged in quality by that of Inner Mongolia, they're not yet outrageously diverse. A few Cretaceous faunas provide two dozen species. (Post Cretaceous ones can double and triple that number.) The Yixian Formation is close to half this total and, as said, there are actually two Yixian communities. A doubling of the known population is far from out of the question. And as Inner Mongolia, that fauna is presently home to only a couple of mammalian hermits with outrageous personalities; a swimming megastar and an aerial acrobat. Keep digging, as there must be more bodies in those cemeteries. "And Teresina discovered, by a method she had found to be infallible, that she was going to have a baby. As she poured a quart of the new beans into the kettle, she wondered idly which one of Danny's friends was responsible." - Tortilla Flat, chapter 13.

Species:	Volaticotherium antiquum (Meng, Hu, Wang, Wang & Li, 2006) Meng, Hu, Wang, Wang & Li, 2007
Aka:	Volaticotherium antiquus Meng et al, 2006
Place:	Daohugou fossi beds, Inner Mongolia
Country:	China
Age:	?Middle Jurassic
Remarks:	Update - March 2007 The authors found they'd made a mistake with their Latin grammar. According to the rules of nomenclature, the specific name has to be in agreement with the gender of the generic name. As therium is neuter an adjustment was required. The following is based upon my reading of Meng et al, 2006. As with the article above, thanks are again due to Mr John Steinbeck for contributing most the subheadings and several quotations. It's been a privilege and a pleasure to have collaborated with this Nobel Prize winning author, and I'm confident he'd feel much the same way. Sweet Thursday People may wonder how I can pay the bills while writing lots of stuff about ancient mammals. I certainly do. The trick seems to involve selling secondhand books and odds and ends and, on Thursday 14th December, 2006, it worked exceptionally well. Two original 1952 German Mickey Maus comics helped manfully with establishing a new record. Late in the day, with a pleasing pile of orders from EBay, Amazon, Booklooker and ZVAB toppling jubilantly off the table so as to carpet the floor, something even better washed in with the e-mail tide. It was a long cherished copy of Meng et al, 2006, hot off the virtual press. (I'd been after one for hours, and Jeff kindly put me out of my satisfied misery.) "Thursday was that magic kind of day. Miss Winch, who took pride in her foul disposition before noon, said good morning to the postman", (Chapter 10). While it's true it was the evening here in Germany, it was doubtlessly morning elsewhere. On such days, you can feel as if you could fly. Volaticotherium generally felt something of that nature, although it was more a matter of gliding (p.889). The evidence for this is close at both hand and foot. It comes from a freaky fossil site of the Daohugou fossil beds of Inner Mongolia where, for some absurd constellation of happystances, conditions were just right for allowing the preservation of soft body tissue. In this case, that includes fur impressions and an impressive flight membrane. Modern flying squirrels have similar flaps of skin, and they're useful for gliding respectable distances between forest trees. Larger ones can manage stretches of a couple of hundred metres with little loss of height. This sort of feature is generally a no no for fossilization and, as they require the support of light, elongated bones, those easily breakable things are also poor candidates. These considerations leave the fossil cupboard largely bare, when it comes to flighty and gliding mammals. For example, the exceptional conditions of Messel, an Eocene locality in Germany, have provided over a hundred specimens of bats from six separate species, but there's little available of their obviously less batty ancestors. A few older fossils are known, but none of them are much older. Cup of Gold Volaticotherium was a modest squirrel kind of size estimated at something like 17cm (combined head and body length), and weighing in with its fighting weight set at 70 grammes. So far, it's the only mammalian pilot known from the entire Mesozoic (which contained more than two-thirds of mammalian history), but its flight membrane and long limbs are far from the entire capacity of its comparative weirdness. It enjoys no close affinities with any established order of mammals. Thus the need for Volaticotheria. Many eccentricities of its beautiful body are connected with its gliding and ultra-arboreal life style. It was a passionate lover of trees, and its broken heart at each departure from one was compensated for by its joy in arriving on another. At least, I hope so. It was an extravagant character, which cared nothing for keeping its feet securely planted on the ground. Its teeth say that getting high on plant products wasn't its only interest, when it came to matters of natural history, as it clearly also enjoyed radically reengineering the bodies of insects. The incisors are small cones and not overly destructive, but the long, sharp fangs had no hospitable intentions towards beetles, and the severe, inwardly curved cusps of the molars were not for expressing and exhibiting tender loving care towards bugs. It was mean and nasty, and the whole dentition was definitely best avoided by invertebrates interested in issues of personal health. Despite the title of the old song, suicide isn't necessarily painless. The dental formula per side is as follows: (lowers): 2 incisors, 1 canine, 4 premolars, 2 molars; (uppers): 3, 1, 4 and 3 respectively. The teeth provide some of the weirdnesses, but not because of their intentions towards insects. The molars are essentially triconodont-like, but they have derivations in distinction from any known triconodont mammal. A triconodont arrangement refers to a simple straight line of three main cusps parading along the central line of a molar, and that's the situation for Volaticotherium. However, uppers and lowers differ more strongly from each other than is generally the case, cusp tips curve back towards the throat, an m1 cusp known as d is "labially overlapping" with the b of m2 (and don't ask me for clarification), and small extra cuspules are absent from the fronts of lower premolars. Also missing is any interlocking mechanism along the postcanine row. Usually, triconodont dentitions were interlocked in some way or other for enhanced stability when biting. The Pearl Lower teeth are of limited value without uppers, and the fossil is blessed with both sets. While the cusps of the lower molars sweep back quite dramatically, like waves about to break down onto the beach, those on uppers point back less informatively, perhaps like trios of fingers of well intentioned but less than certain directors: "The throat? I think it's somewhere back there." These uppers bear most resemblance to two molars of Ichthyoconodon; isolated teeth thought to be lowers. However, those cusps tilt less to the rear and aren't as tall. Even should they be uppers, they differ. The lower molars, meanwhile, only resemble those on the corresponding jaw half of Volaticotherium, the glider of Inner Mongolia. In other words, they're like nothing else ever seen. In terms of the number of bones making up the lower jaw, namely one, this critter is typically mammalian. Its lower jaw is all dentary. And, as the plentiful impressions attest, it was undeniably furry. However, the upper jaw continued with a basal feature, in possessing a branch of bone on the premaxilla known as the internasal process. But, and in contrast to earlier models such as Morganucodon, Sinoconodon and non-mammalian cynodonts, this doesn't run as far back and it avoids meeting up with the nasal bone. The Wayward Bus Obviously, Volaticotherium had intimate relatives, was on excellent terms with them, but who they were isn't clear. The triconodont-like dentition is inherited from the bearer of a primitive mammalian tooth fashion, and its peculiarities rule out direct affinities with the main stream of the triconodont river of Eutriconodonta. In an attempt to untangle what's what, the researchers ran an analysis of 58 genera and 435 anatomical traits. The results place this critter within Crown-group Mammalia, and suggest it's a relative of a clade containing eutriconodonts, multituberculates and trechnotherians such as the spalacotheriid 'symmetrodonts' and myself. However, it's related to that wide group but not within it. Close relatives may be sitting in a waiting room of a bus station somewhere or other but, as yet, none have arrived at a convenient destination. The Moon is Down The sexy feature of this mammal is likely to be utterly uninformative as to the interrelationships of Crown-group Mammalia, but that doesn't matter a damn. It's yearn-provocatively, gut-stirringly beautiful. Strip off, sing out, jump up and down and feel great to be alive. If you find somebody irresistibly sexy, then the identity of their parents, siblings, cousins, grandparents and Uncle Tom Cobley and all, just ain't relevant. And even if they happen to be in the area, they shouldn't expect invitations into the bedroom, and they've got no bloody business to knock on the door. Preserved in the rock is an extensive, hooptedoodle of a flight membrane, a patagium (p.890). It's thickly furred and relatively large. Just how large is impossible to say for sure, as it's not fully exposed and seems to be folded up in places. It runs between the extremities of the front and back legs, and makes it absolutely obvious that this animal was a glider (p.891). It's similar to the arrangement known from flying squirrels; tree-top dwellers which, odd as it may sound to some, are largely nocturnal. Gliding from tree to tree in the dark may sound like a great way of breaking your neck, but the squirrels manage to do this in safety, and the canopy of thick forests tends to be fairly dark at high noon. Unlike those rodents, however, Volaticotherium was as full-blooded an insectivore as typical bats. That's not to say it hunted in such a manner. Catching insects in mid air requires: a. a way of pin-pointing their position; and b. extreme aerial manouevrability. Volaticotherium was a glider, not an astonishingly sophisticated jet fighter fitted with radar. Lots of anatomical adaptations are connected with gliding flight, and this includes the impressively long limbs. A flight membrane requires a lot of surface area, and elongated legs contribute much to this objective, and they're typically worn by gliding mammals. They're far from typical for Mesozoic mammals, which generally had rather stubby limbs. No others come anywhere near this critter in terms of proportional length. The caudal vertebrae form the tail, and these bones are long, wide and flat. The foremost are surrounded by the remains of the flight membrane, and they presumably provided further support. Other details of these vertebrae demonstrate the tail was relatively rigid. This all combines to produce a stiff and flat tail for extra stability during flight (p.892). Viva Zapata! The rear legs are both close to being preserved in a fully articulated condition, and the made an even better job of completeness when their owner was alive. An oddity of the femur bone is the absence of a neck between its shaft and head, and this lack hasn't been previously encountered on a mammal. The femoral head fitted into a socket on the pelvis called the acetabulum. As the head is small, then this acetabulum must've have been appropriately shallow, but the hip region isn't preserved. This combination would have placed restrictions upon the rotational abilities of the joint, and could have thus further enhanced the stability of the limb when jumping horizontally for joy betwixt trees. Lots of limb structure shows adaptation for arboreal adventurousness, and these include climbing skills as well as gliding. The construction of the paws, for example, results in proportionately long and flexible fingers and toes, and would've allowed their owner a strong grip for moving through and along the branches in search of snacks. The fingers and toes don't differ all that much from each other. Only one final phalanx is available, and it happens to be from a toe. Its build indicates it was crowned by a splendid claw. Body weights of modern mammal gliders range from 10g to 1500, and this coincides with bats. It provides a sensible basis for estimating the weight of our deadling friend. The skull's incomplete, but a 28.3mm jaw suggests 3.5cm is a likely figure. This was atop a body of 12 to 14cm. That makes it comparable in size to an American flying squirrel named Cyril or, more officially, Glaucomys volans. As luck would have it, Cyril has popped round for tea and bickies, and has agreed to climb on the scales; 70g. As most flying squirrels are nocturnal, this habit may also have been followed by Volaticotherium, but this is a decision best made by the available prey. Holotype IVPP V14739 is much of a squashed corpse in the employment of the Institute of Vertebrate Paleontology and Paleoanthropology, Beijing. The specific name is Latin for 'ancient', and that word's English for 'antiquus'. It could have something to do with the age. Postscript, hooptedoodle "No," said Mack. "Sometimes I want a book to break loose with a bunch of hooptedoodle. The guy's writing it, give him a chance to do a little hooptedoodle. Spin up some pretty words maybe, or sing a little song with language. That's nice. But I wish it was set aside so I don't have to read it. I don't want hooptedoodle to get mixed up in the story. So if the guy that's writing it wants hooptedoodle, he ought to put it right at first. Then I can skip it if I want to, or maybe go back to it after I known how the story come out." - Sweet Thursday, the Prologue
References:	Meng J, Hu Y, Wang Y, Wang X & Li C (2006), A Mesozoic gliding mammal from northeastern China, Nature, 444, p.889-893.
	Meng J, Hu Y, Wang Y, Wang X & Li C (2007), Corrigendum, A Mesozoic gliding mammal from northeastern China, Nature, 446, p.102.

Links: Hairy Museum of Natural History, December 16, 2006 http://news.bbc.co.uk/2/hi/science/nature/6176061.stm Short and sweet by Matt Celeskey. BBC, 13.12.2006, Earliest flying mammal discovered http://news.bbc.co.uk/2/hi/science/nature/6176061.stm Aunty Beeb takes a peep. As accessible summaries go, it's accessible and goes. About Sweet Thusday by John Steinbeck http://www.ac.wwu.edu/~stephan/Steinbeck/sweet.html John insisted on an advert.

Other reports: Dorset, England The Middle Jurassic site of Watton Cliff may have provided an amphilested upper molar (Averianov et al, 2007, p.4). Although described as being a lower tooth from Amphilestes, it has a cingulum running around all the crown, and this is more usual with upper molars. It's similar to the front molars of Gobiconodon and, in contrast to the recently described Kemchugia, it's proportionately narrower and lacks cuingula cusps. Should it happen to be an upper, then it has a distinct ectoflexus, and that's also unlike Kem. Cañadon Asfalto Formation, Argentina The 2006 Society of Vertebrate Paleontology Abstracts include Rougier G, Garrido A, Gaetano L, Puerta P & Novacek M on p.116A. They report on mammal fossils from the Queso Ralloado Quarry, and this is presently the only source of remains for the Middle Jurassic of South America. It yielded Asfaltomylos and a related critter (undescribed). The place has also now joined in with triconodonts or, at least, one molar. (The above mentioned australosphenidans managed bits of jaw as well, but you can't have everything.) The crown has a line of five cusps arrayed along the middle of the tooth, running from front to back. The middle one, a, is tallest. Its neighbours, b and c, and close together in height, while d and e are better termed accessory cusps (on account of being rather small). A pathetic cusp g is also present, but no cingulum worthy of the name bothered to develop. It's most similar to the later North American Triconolestes curvicuspis from Utah. However, peculiarities are also in evidence, so its closer affinities are presently awaiting resolution.

Help: Should anybody have any further information, I'd be pleased to hear of it. Regarding references and Bibliography: I haven't and can't verify the references, so beware. Traditional papers used in constructing this page are in the bibliography. If you feel these are too few, then send some more. With thanks to all the featured sources. back to top Trevor Dykes, November 2002 (Relaunched November 2007) Most recent update: 27.9.2011 (Many of the entries were originally on a different page and date back to May 2001). Ktdykes@arcor.de

With further thanks for assistance due to: Mr David Marjanovic, for corrections and suggestions. Dr Zhexi Lou for the encouragement, kind words and various papers. HitBox Central for the thrilling animations. Dr John Alroy, for information concerning Jugulator, via his North American Fossil Mammal Systematics Database, Serdar Mayda. The Society of Vertebrate Paleontology's Bibliography of Fossil Vertebrates (John Damuth) http://www.bfvol.org/ BIOSIS: The Index to Organism Names http://www.biosis.org.uk/triton/indexfm.htm Polyglot Paleonotologist http://www.uhmc.sunysb.edu/anatomicalsci/paleo/terms.html This homepage is packed with goodies, especially for the keen anatomist.

Bibliography: Averianov AO, (2002), Early Cretaceous "symmetrodont" mammal Gobitheriodon from Mongolia and the classification of "Symmetrodonta". Acta Palaeontologica Polonica 47 (4), p.705-716. Averianov AO, Skutschas PP, Lopatin AV, Leshchinskiy SV, Rezvyi AS & Fayngerts AV (2005), Early Cretaceous mammals from Bol'shoi Kemchug 3 locality in West Siberia, Russia, Russian Journal of Theriology, 4(1), p.1-12. Chow M & Rich THV (1984), A new triconodontan (Mammalia) from the Jurassic of China, Journal of Vertebrate Paleontology, 3(4), p.226-231. Cifelli RL (1999), Therian teeth of unusual design from the Mid-Cretaceous (Albian-Cenomanian) Cedar Mountain Formation of Utah. Journal of Mammalian Evolution, 6 (3), p.247-270. Cifelli RL, Lipka TR, Schaff CR & Rowe TB (1999), First Early Cretaceous mammal from the eastern seaboard of the United States. Journal of Paleontology, 19(2), p.199-203. Datta PM & Das DP (2001), Indozostrodon simpsoni, gen. et sp. nov., an Early Jurassic megozostrodontid mammal from India. Journal of Vertebrate Paleontology, Vol. 21(3), p.528-534. Engelmann GF & Callison G (1998), Mammalian Faunas of the Morrison Formation, Modern Geology, Vo 34 (4), p.343-379. Gaetano LC & Rougier GW (2011), New materials of Argentocondon fariassorium (Mammaliaformes, Tricondondonidae) from the Jurassic of Argentina and its bearing on triconodont phylogeny, Journal of Vertebrate Paleontology, 31(4), p.829-843. Gao C-L, Wilson GP, Luo Z-X, Maga AM, Meng Q & Wang X (2010), A new mammal skull from the Lower Cretaceous of China with implications for the evolution of obtuse-angled molars and 'amphilestid' eutriconodonts, Proceedings of the Royal Society B, 277, p.237-246. Hasiotis ST (1999), Continental ichnofossils from the Upper Jurassic Morrison Formation, Western Interior, USA: what organism behaviour tells us about Jurassic environments and climates. In (eds. Santucci VL & McClelland L) National Park Service Paleontological Research, vol 4, p.121-125. Heinrich W-D (1998), Late Jurassic Mammals from Tendaguru, Tanzania, East Africa. Journal of Mammalian Evolution, Vol 5 (4), p.269-290. Ji Q, Luo Z & Ji S (1999), A Chinese triconodont mammal and mosaic evolution of mammalian skeleton. Nature 398, p.326-330. Kemp TS (2005), The Origin and Evolution of Mammals, Oxford University Press, pp.331. Kielan-Jaworowska Z & Hurum JH (2001), Phylogeny and systematics of multituberculate mammals, Palaeontology, Vol 44 (3), p.389-429. Kielan-Jaworowska Z & Lancaster TE (2004), A new reconstruction of multituberculate endocranial casts and encephalization quotient of Kryptobaatar, Acta Palaeontologica Polonica 49(2), p.177-188. Lopatin AV, Maschenko EN & Averianov AO (2010), A new genus of triconodont mammals from the Early Cretaceous of western Siberia, Doklady Biological Sciences, 433, p.282-285. Luo Z-X, Chen P, Li G & Chen M (2007), A new eutriconodont mammal and evolutionary development in early mammals, Nature, 446, p.288-293. Luo Z-X, Kielan-Jaworowska Z & Cifelli RL (2002), In quest for a phylogeny of Mesozoic mammals. Acta Palaeontologica Polonica 47 (1), p.1-78. Luo Z-X & Wible JR (2005), A Late Jurassic digging mammal and early mammalian diversification, Science, 308, p.103-107. Martin T & Averianov AO (2007), A previously unrecognized group of Middle Jurassic triconodontan mammals from Central Asia, Naturwissenschaften, 94, p.43-48. McKenna MC & Bell SK, (1997), Classification of Mammals Above the Species Level. Columbia University Press. Meng J, Hu Y, Wang Yuanqing & Li C, (2003), The ossified Meckel's cartilage and internal groove in Mesozoic mammaliaforms: implications to origin of the definitive mammalian middle ear, Zoological Journal of the Linnean Society, 138, p.431-448. Meng J, Hu Y, Wang Y, Wang X & Li C (2007), Corrigendum, A Mesozoic gliding mammal from northeastern China, Nature, 446, p.102. Prasad GVR & Manhas BK, (2002), Triconodont mammals from the Jurassic Kota Formation of India. Geodiversitas 24 (2), p.445-464. Rose KD, Cifelli RL & Lipka TR, (2001), Second triconodont dentary from the Early Cretaceous of Maryland, Journal of Vertebrate Paleontology, 21(3), p.628-632. Rougier GW, Garrido A, Gaetano L, Puerta PF, Corbitt C & Novacek MJ (2007), First Jurassic triconodont from South America, American Museum Novitates, 3580, p.1-17. Rougier GW, Isah S & Manabe M (2007), An Early Cretaceous mammal from the Kuwajima Formation (Tetori Group), Japan, and a reassessment of triconodont phylogeny, Annals of Carnegie Museum, 76(2), p.73-115. Rougier GW, Novacek MJ, McKenna MC & Wible JR (2001), Gobiconodonts from the Early Cretaceous of Oshih (Ashile), Mongolia. American Museum Novitates, 3348, 30pp. Rougier GW, Spurlin BK & Kik PK (2003), A New Specimen of Eurylambda aequicrurius and Considerations on "Symmetrodont" Dentition and Relationships. American Museum Novitates 3398, p.1-15. Simpson GG (1980), Splendid Isolation, The Curious History of South American Mammals, Yale University Press, pp266. Simpson GG (1980), Splendid Isolation, The Curious History of South American Mammals, Yale University Press, pp266. Simpson GG; Olson EC (1935/1976), The First Mammals, The Quarterly Review of Biology, 51, 50th Anniversary Special Issue, 1926-1976, p.45-73. (Note: this is a reissue with an introduction by Olson. The remainder of the paper is by Simpson, and has the page numbers 154-180.) Weil A (2002), Mammalian evolution: Upwards and onwards, Nature 416, p.798-799. Zhang Fakui (1984): The Fossil Record of Mesozoic Mammals in China. Vertebrata PalAsiatica XXII, No. 1, Jan 1984, p. 29-38. The translation of this article from the Chinese was made by Will Downs and obtained courtesy of the Polyglot Paleontologist website.