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.

'HARAMIYIDA'

Taxon: 'Haramiyida' Hahn, Sigogneau-Russell & Wouters, 1989 Reference: Hahn et al (1989), New data on Theroteinidae -their relations with Paulchoffatiidae and Haramiyidae, Geol. Palaeont., 23, p.205-215. I previously credited this taxon to Hahn, 1973. That was incorrect and should have read Haramiyoidea, "proposed as suborder of Multituberculata," (McKenna & Bell, 1997). Note: There will be emendments to the family affiliations presently noted on this directory. They'll take into account the scheme proposed in a study by Hahn & Hahn in 2006. 'Haramiyidans' seem to be the earliest known herbivores amongst basal mammals, assuming they are mammals. Their teeth, which are by far the most common remains, resemble those of the multituberculates. The jaw however, based on Haramiyavia, is less derived; "at the level of evolution of Morganucodon and Kuehneotherium, with a groove for ear ossicles on the dentary" (Butler PM, 2000). It’s tempting to infer that this order represents the ancestors of the multis, a group which appears to have survived until about 40 million years ago. Whilst this is not impossible, the evidence available is insufficient to be conclusive. As an illustration and with reference to material from Greenland: "Jenkins' group identified haramiyid jaws with the teeth in place, as well as additional parts of the skeleton. Among other revelations from the fossils, the teeth in the upper jaw do not fit the classic multituberculate arrangement, with the second molar offset towards the centre of the mouth. "That shows rather well that haramiyids are not closely related to multituberculates," says William A. Clemens of the University of California, Berkeley, one of the leaders of the multituberculate symposium," (Monastersky 1996, p.379). A confession in advance You might have noticed the inverted commas. If not, have another look; 'Haramiyida'. These aren't used for fun. This taxon is in some disgrace for being paraphyletic, (Butler & Hooker 2005, p.185). It doesn't consist of one ancestor and all of its descendants. However, it's not the fault of the 'haramiyidans' that their fossil record is presently fairly lousy. They had more important business to attend to when alive. The general picture Kemp, 2005 (p.140-141) provides a concise overview of the postcanines. Harami molars are larger than many of their equivalents from contemporary mammals, but not by much. These are blessed with lots of cusps and are generally double-rooted. The crowns are wide and have a line of three large cusps on one edge, with five smaller ones on the opposing side. As there's a connecting ridge at one end and a basin in between, the general idea is a crater surrounded by a number of hills. Originally, it was thought that upper and lower molars were pretty much mirror images of each other, and minor details lead to the establishment of two genera: Thomasia and Haramiya. The suggestion was subsequently made that these could also represent lowers and uppers of only one genus, and the discovery of Haramiyavia provided confirmation for that. Relatives? The teeth of Theroteinida and Eleutherodontida, (nothing more is yet known), are perhaps more accurately described as 'haramiyidan'-like. "But haramiyids are known from beds as old as Norian: hence, if they are related to multituberculates, an astonishingly early divergence of crown mammals -not to mention a series of putative sister taxa to crown mammals- is implied," (Cifelli, 2001). More precise affinities will probably remain unclear until better evidence is forthcoming. Butler & Hooker, 2005 maintain that 'haramiyids' are still stronger candidates for having given rise to the multis than morganucodontids are: "As long as we only have teeth to of the critical taxa, we feel it necessary to adopt the Allotheria concept as a working hypothesis; no doubt the discovery of mammalian skeletal material in the Jurassic will throw new light on the problem", (p.206). The concept of Allotheria unites haramis and multis as the sister line of other mammals. It's not a scheme I presently follow. Geographical range Most fossils have been reported from Europe, but some are known from Africa and Greenland. A description in January 2005 extended the published range to the Junggar Basin of Inner Mongolia. For those of a technical persuasion, it may be safer to refer to this group as haramyioidens rather than 'haramiyidans'. It's a wider (and uglier) term. Time 'Haramiyan' fossils first appear in the Upper Triassic. Until 1999, the last traces were Middle Jurassic. However, Allostaffia then turned up in the Upper Jurassic of Tanzania. Nobody had informed it of any extinction. If Butler and Hooker are correct with their interpretaion of hahnodontidids, members of the group even made it as far as the Lower Cretaceous: Hahnodon and Denisodon. I'm presently treating that pair as multituberculates, and waiting for further opinions on the matter.

Links: Palaeos, Mammaliaformes: Allotheria http://www.palaeos.com/Vertebrates/Units/Unit420/420.100.html See Haramiyidia. For non-anatomists, such as myself, these notes are hard work. They are nevertheless highly informative. I find my struggles with the technical language are well worth the effort, although I wouldn’t claim to understand every word. Haramis according to Hahn & Hahn, a scheme I'm not presently following The following is based upon my reading of Hahn & Hahn, 2006 and thanks are due to the supplier. Some innocent visitors are likely to find some of the language used thoroughly offensive. There is bound to be some overlap with the notes above. Being known almost exclusively from isolated teeth, haramis are very good at being mysterious. Their placement within Mammalia isn't even secure, possible affinities with multituberculates are proposed by some but disputed by others, and how -or if- the various harami strands should be woven into a unified design, one reflecting the actual systematics, isn't clear either. The shortage of evidence is fine fertilizer for a rich crop of question marks. Taking into account the meagre crumbs that are available, the dynamic duo of Hahns from Marburg, Germany have contributed a further review of these extraordinarily shy extinct eucynodonts. They've also endeavoured to unravel any evolutionary tendencies observable in the known record. However, with the fossil cupboard so bare, the chances of coming up with the correct picture must be very low, regardless of how careful and experienced the researchers might be. Such a task isn't for those afraid that subsequent events will show up errors. I'd imagine that will happen here. Knowing this is likely shouldn't provide sufficient grounds for taking the easiest option; ie. not bothering. Despite shortages, some evidence is available, the supply has increased a bit, and getting towards an accurate understanding will never progress without attempts to make sense of what's known. Hahn & Hahn's haramis, the broad scheme Following previous authors the Hahns retain Haramiyida as an order, and divide it into two subfamilies. This had already been down by Butler in 2ooo, but they found the names he proposed to be inappropriate; Theroteinida and Haramiyoidea. Rather dully, suborders shouldn't have the same ending as their order and -oidea pertains to the rank of superfamily. Therefore, they emended them (p.189). The theroteininans, a word that's horrible to spell, are regarded as being more basal than the haramiyinans. Theroteinina is accused of containing a single known family housing two genera, Theroteinus and the later Millsodon Haramiyina (p.190) is associated with greater diversity, and subdivides into perhaps five families within two superfamilies (new ranks). The first superfamily proposed is a lonely looking Haramiyavioidea. Presently, its sole occupant is Greenland's Haramiyavia, and that genus is also allowed to be in its own family. To put things into plain English, Haramiyavia is a haramiyaviid haramiyavioid haramiyinan haramiyid. That should be easy enough to remember. The second superfamily is the more complex Haramiyoidea. On the face of it, one detail of that lot could cause some harami lovers to shake their heads in astonishment. It's a matter of time. Haramis are typically associated with the faunas of Rhaetian-Liassic Europe, with their sparse traces fading drastically beyond the Lower Jurassic. A "probable" member of one haramiyoid family happens to date from the end days of the Upper Cretaceous; Avashishta from India. When I first heard of that critter, it struck me as astonishing. It extended the known range of haramis by something like 85 million years, so perhaps those wanting a pet hadrosaur needn't yet give up their hopes. However, the possibility isn't beyond my belief for two reasons. Firstly, a fossil happens to have been found from Maastrichtian rocks, and it matches best with haramis. Using a similar logic, if somebody could show me part of a hadrosaur that was alive in the very recent past, I'd be suitably astounded, have to readjust my opinion as best I could, and then I'd want to know more. By the way, a live multituberculate mammal would be even more welcome. The second reason concerns the known map of the Gondwanan mammalian fossil record. It contains vast blank spaces. Whatever eventually comes to light from the southern hemisphere, astonishment will very probably accompany it. Haramiyoidea According to Hahn & Hahn's 2006 scheme, this superfamily contains Haramiyidae, Eleutherodontidae, Allostaffiidae and perhaps Hahnodontidae (the somewhat disputed Hahnodon and relatives, possibly multis). Although not of direct concern to their study, they see this superfamily as possibly being related to the multituberculates. This directory I haven't taken the opportunity to re-order the entries on this directory in accordance with these proposals, which doesn't mean I'm dismissive of them. Rather, I'm awaiting further discussion and other opinions. Presently, I'm sticking to a straight A-Z arrangement of haramis. I also don't intend to give some account of the argumentation supporting the Hahns' conclusions. As the last word is far from falling on the matter of harami systematics, various bits could go out of date sooner or later. Nevertheless, their allocations to families have been noted in the entries below. The earliest harami? In the Addenda section of page 191 comes news of a possible new harami. The word "probably" is used. "The configuration of at least one isolated tooth is similar to that of Thomasia, but the crown has only six cusps (three in each row), and only one root is present." I've no intention of offering quotations from an unpublished paper I've seen, but I happen to know this interpretation will be robustly challenged. If the teeth in question were from a harami, then they'd extend the range of the group back by something like 25 million years. The fact that there's only one root is a characteristic not in keeping with haramis, and it'll be argued that similarities of crown morphology are matters of convergence rather than common descent. Although the chronological extension involved would be considerably less than that leapt by Avashishta, I'd actually find it far more bewildering in its implications, given what's known of the Triassic fossil record. I should also mention details the Hahns have certainly got incorrect. They rightly state these finds come from the Cynognathus Assemblage Zone of Karoo, South Africa. However, they say "uppermost" and refer to "Dr. T. Abdala". This led me to contact Fernando Abdala, whose first name doesn't begin with a 'T'. The specimens come from subzone A of the Cynognathus AZ, and that's the lowermost one. Rather than dating from the Middle Triassic, they're Lower Triassic (personal communication, 2007). Expanding diversity Prior to the 1980s, haramis comprised only two 'genera' from the Rhaetian-Liassic of Europe, and it was thought probable (correctly so) that these were based upon upper and lower teeth of but one genus. Unfortunately, there was no way of resolving that matter for sure. Such was the state of play after a mere 130 years of relevant research. The subsequent three decades could seem giddy with riches in comparison. Firm confirmation allowed Haramiya to be buried within Thomasia, and that naturally reduced the count of valid genera. However, new localities have rushed in to more than fill that loss. Rudely ignoring such exotics as the alleged 'Middle' Triassic South African and the actually tentative Upper Cretaceous Indian, the counter now stands shows at least eight genera ranging from Greenland, through Europe, and now into Asia and Africa where, in the latter case, the harami clock was still ticking during the Upper Jurassic. Additionally, and not according to the opinion of the Hahns, Morocco may house Lower Cretaceous haramis, should Hahnodon & Co so qualify (p.173). To return to one point mentioned above, the Hahns argue that haramis: "are indeed members of the Allotheria Marsh, 1880 and related to the Multituberculata Cope, 1884" (p.174). If correct in all their opinions, then a "Middle" Triassic harami, which would actually be a Lower Triassic one, would have staggering implications for the concept of Crown-group Mammalia which, according to large amounts of recent research, the multis are part of. Actually, even ignoring that possibility, I can't for the life of me reconcile multis as being both allotherians and crown-group mammals. It's not a matter addressed in the paper, so it would be wrong to assume the Hahns even implied any such thing. They don't argue multis belong within the crown-group. John H Burkitt, Mammals, A World Listing of Living and Extinct Species http://cougarhillweb.org/mammals.pdf This project is a definition of the word monumental. T Mike Keesey, The Dinosauricon, Ages of the Mesozoic http://www.dinosauricon.com/times/index.html A good guide to the chronology of long ago. Mikko K Haaramo, Haramiyidae Mikko Haaramo's Haramiyidae By habit, my starting point.

Genera: Allostaffia, Avashishta, Eleutherodon, Eleutherodon (partly = Millsodon), Haramiya (= Thomasia), Haramiyavia, Hypnoprymnopsis (?=Hypsiprymnopsis), Hypsiprymnopsis (dubious), Kirtlingtonia, Microcleptes (= Haramiya), Microlestes (= Haramiya), Millsodon, Mojo, Plieningeria (=Thomasia), Staffia (= Allostaffia), Sineleutherus, Theroteinus, Thomasia, other reports Time-Line: Upper Cretaceous: Avashishta Lower Cretaceous: Upper Jurassic: Allostaffia, Sineleutherus Middle Jurassic: Eleutherodon, Kirtlingtonia, Millsodon Lower Jurassic: Thomasia Upper Triassic: Haramiyavia, Hypsiprymnopsis, Mojo, Thomasia

Genus: Allostaffia (Heinrich WD, 1999) Heinrich WD, 2004 Aka: Staffia Heinrich WD, 1999 Family: Allostaffiidae Hahn & Hahn, 2006: Suborder Haramiyina G Hahn, 1973 Remarks: The original generic name turned out to be preoccupied. Thanks are due to David Marjanovic for the tip off.

Species:	Allostaffia aenigmatica (Heinrich WD, 1999) Heinrich WD, 2004
Aka:	Staffia aenigmatica Heinrich WD, 1999
Place:	Tendaguru
Country:	Tanzania
Age:	Kimmeridgian-Tithonian, Upper Jurassic
Remarks:	This is a surprise. It's late but is a 'haramiyidan'. G & R Hahn reportedly say that eleutherodontid affinities cannot be excluded. Further material was described in 2001.
References:	Heinrich W-D (2004), Allostaffia, a new genus name for Staffia Heinrich, 1999 (Allotheria, Haramiyida) preoccupied by Staffia Schubert, 1911 (Protista, Foraminifera). Mitt Mus Natkd Berl Geowiss Reihe 7, 153, Zoological Record Volume 141.
	Heinrich W-D (1999), First haramiyid (Mammalia, Allotheria) from the Mesozoic of Gondwana. Mittlg. aus dem Mus. f. Naturkunde in Berlin, Geowiss. Reihe, vol 2, p.159-170.

Link: Darren Naish http://www.soft.net.uk/richardforrest/svpca/personalaccounts/dn2000/10.synapsids.html An early mention of the fossil.

Tendaguru, Tanzania -Upper Jurassic The following is loosely based upon my reading of Heinrich, 1998, p.270-271. Tendaguru is in the southeast of Tanzania and, until the First World War, it formed part of the German Empire. As the local residents had never been asked for their views concerning imperial affinities, the British kindly removed the Teutonic yoke and replaced it with one fashioned from sturdy English oak instead. Obviously, the local people weren't consulted on this development either. Tendaguru was also the location for one of the largest dino digs ever undertaken. The German Tendaguru expedition lasted from 1909 until 1913, and involved the construction of a shanty town for several thousand locals. They were allowed to do the actual digging and carrying, of which there was plenty. And some of the fossils recovered were both staggeringly heavy and complete. Most famous is a specimen of Brachiosaurus, which may be admired in the Humboldt Museum; all 22 metres of it. As this locality is forty miles inland, the labourers who manhandled these remains to the coast deserve a damn-sight more recognition for their work than they'll ever receive. As well as the dominant genera of Brachiosaurus, Dicraeosaurus and Barosaurus, the varied dino-fauna also included theropods, ornithopods and stegosaurs. Of the latter, Kentrosaurus might be said to be a fairly insubstantial representative with a length of only 2.5 metres, but this view would be revised by anybody who'd helped carry the thing. Also living in the area were crocodiles and pterosaurs. Doubtlessly impressed by the fossil on display in Berlin, a British expedition worked in Tendaguru from 1924 until 1931. However, the remains 'liberated' to London were more fragmentary. There are three dinosaur yielding beds present and these were cleverly designated as the Lower, Middle and Upper Saurian Beds. Mammal remains have been recovered from the higher two. Biostratigraphic studies indicate the age of these deposits is Kimmeridgian-Tithonian. The German expedition took home a further souvenir from what was then considered to be German East Africa; 500kg of matrix. This was examined at a later date which, as things turned out, was eighty years or so later. Surprising amounts of material hang around for decades in museums and universities. When eventually processed, this matrix was found to contain remains of various vertebrates: lizards, crocs, pterosaurs, mammals and dinosaur teeth. As Brachiosaurus hogs the limelight, spare a thought for the marginally less well-known Allotaffia. This is the youngest known haramiyid, (as presently constituted). This was then an ancient and venerable mammalian (or near-mammalian) lineage, which more typically turns up in rocks from the Upper Triassic and Lower Jurassic. Very little is presently known about haramiyids beyond their teeth but, nevertheless, they had a history spreading across at least sixty million years. The dino fauna has strong affinities with Upper Jurassic contemporaries from North America (The Morrison Formation), and Europe (see Guimarota, although the emphasis there is on microvertebrates). The absence of any known multituberculates at Tendaguru is an intriguing contrast. However, given that the mammal material is so limited, it would seem premature to attribute significance to this. Absence of evidence is not necessarily evidence of absence, as a well-used saying goes. Further Mesozoic site summaries can be found at Localities. Meet the Mammals of Tendaguru (and the haramiyid, if it's non-mammalian) Haramiyida Allostaffia aenigmatica Triconodonta Tendagurodon janenschi Dryodlestoidea Brancatherulum tendugurense ?Peramura Tendagurutherium dietrichi

Genus: Avashishta Anantharaman S, Wilson GP, Das Sarma DS & Clemens WA, 2006 'Survivor' Family: ?Allostaffiidae Hahn & Hahn, 2006: Suborder Haramiyina G Hahn, 1973 Aka: Avashistha (a typo -H & H, 2006 p.179) Remarks: The referral of this genus to the family is tentative, and the generic name comes from the Sanskrit word avashist, 'that which survives'. Thanks are due to Dino Hunter for uploading the citation. Prelude, The spread of haramis 'Haramiyidans' used to be poorly understood, possibly mammalian teeth known from a couple of Rhaetian-Liassic deposits (Upper Triassic - Lower Jurassic) from a few localities in Western Europe. A small ripple of excitement spread across the surface of this generally still pool in 1983. A report from Arizona of a possible harami tooth from the Kayenta Formation registered a potential -not unlikely- presence in North America as well. European diversity gradually began to increase thanks to further locations and, in 1997, some long-term dilemmas were resolved by teeth still in situ on jaw remains from Greenland. At last there was clarity regarding the identification of upper and lower postcanines, and their correct orientations. The geographical spread was well underway; much of the northern hemisphere rather than just Europe. Then came a couple of chronological expansions for haramis. Fossils from Oxfordshire demonstrated some, eg Eleutherodon, managed to scamper around towards the latter parts of the Middle Jurassic (1998). Much more recently, their territory was pushed far east by the same genus being found in Inner Mongolia (2005). In terms of time, that expansion was fairly modest. It was put into the shade by a further development in 1999. Research into Upper Jurassic matrix obtained from Tendaguru, Tanzania revealed much later haramis had lived in Africa; deep in the southern landmass of Gondwana. According to a disputed reinterpretation, more may have been active during the Lower Cretaceous of Morocco, but this is still contested by Hahn & Hahn. In a 2006 paper, they maintain their view that Hahnodon and Denisodon are multituberculates Middle Triassic harami? Before turning to the astonishing possibility of Avashishta actually being a harami from towards the end of the Cretaceous, Hahn & Hahn, 2006 mention some apparently remarkable news. The Abstract states: "The stratigraphic range of the order is extended by new finds from the Middle Triassic of South Africa (Addendum 4) to the Upper Cretaceous of Asia (Addendum 1)." Addendum 4 reports that a certain Dr T Abdala has come across some: "teeth probably belonging to the Haramiyida from the uppermost Cynognathus Assemblage Zone (~Late Olenekian) of the Karroo Basin in South Africa" (p.191). Unfortunately, much of that's incorrect. Dr F (not T) Abdala and colleagues have collected haramiyidan-like teeth from subzone A of that unit, and that dates from the Lower Triassic (not Middle). In this case, 'A' isn't "uppermost" but lowermost. While the crown morphology is haramiyidan-like, their single-rootedness (correctly reported by the Hahns) is utterly un-haramiyidan-like. I'm reliably informed these teeth will be referred to an otherwise unknown, non-mammalian cynodont, and not a haramiyidan. A publication to that effect is in press. Naturally, also in planning is the addition of information to this directory from the Hahn's most welcome review of the group. A harami presence in the Lower Triassic would've involved a chronological extension of perhaps 25 million years. That's dramatic stuff, and the long hiatus in the reasonably good succession of Karoo faunas (and those from elsewhere) would've been puzzling to come to terms with. Avashishta, an Upper Cretaceous perhaps harami On the face of it, should Anantharaman & Co really have caught an Upper K harami (a claim they don't actually make), then the leap of time is even more dramatic; around 70 million years of so later than Allostaffia. But, even prior to having read the Avashish paper, stuck me as being more plausible than the "Middle" Triassic announcement. While certainly surprising, the possible presence of haramis in India doesn't seem as outlandish in terms of the relevant known fossil record; the Gondwanan one. I can even fall for the temptations of a parallel. At about the same time that Allostaffia was doing its stuff in the Upper Jurassic of Tanzania, the continents of Europe and North America were enriched by a variety of mammals called dryolestoids; then fashionable insectivores. This line became rare as the Lower Cretaceous progressed and then vanished from view in the north. However, the fossil record of Gondwana is far more poorly known and, as very few traces have been found, it's a perilous game to say whether traces actually faded or are yet be found. In any event, Upper Cretaceous Patagonian faunas turned out to be dominated by dryolestoids, and they even seemed to launch a mini Campanian invasion into Wyoming. (The sole tooth found there so far suggests it wasn't all that overwhelming.) If that lineage persisted in Gondwana, then the same possibility has to permissable for others including haramis. A lack of fossils beyond the Middle Jurassic in Laurasia, despite a considerable number of localities, provides strong evidence of a northern extinction at about that time. An absence of fossils in Gondwana is a different matter, given the paucity of fossil sites. That particular map is chock-a-block with massive vacant spaces.

Species:	Avashishta bacharamensis Anantharaman S, Wilson GP, Das Sarma DS & Clemens WA, 2006
Place:	near Bacharam, Andhra Pradesh, Deccan Traps
Country:	India
Age:	Maastrichtian, Upper Cretaceous
Remarks:	The following is based upon my reading of Anantharaman et al, 2006 and thanks are due to the supplier. It seems this tooth must be from a fourth Deccan Traps' mammal locality, as it's the first to be described from the Infratrappean beds (p.488). Three other sites have yielded fossils from the Intertrappean beds. While also Maastrichtian in age, Avashishta isn't precisely contemporaneous with the other Upper Cretaceous Indian faunas. The reader is also assured that a collection of specimens from both dormitories (beds) includes several new eutherians and a gondwanatherian mammal. That would presumably be the since published Dakshina (almost certainly aka Bharattherium). Only a single specimen pertained to Avashishta, and the authors sensibly term it a molariform. I'm going to call it a molar as, although strictly not necessarily correct, the word looks less frightening to many people. They also presently favour a hypothesis linking haramis with multituberculates, a position for which support seems to be strengthening again. However, they term it: "a paraphyletic stem group, consisting of all allotherians not belonging to Multituberculata". This is something akin to expanding your concept of close relatives, in order to include all descendants of a personal ancestor alive some time prior to the development of agriculture; say ten thousand years ago. If I were you, I'd make a start on writing this year's Christmas cards about a decade ago. Actually, that's not a very apt analogy, seeing as that's massively understating things. Suffice it to say that some haramis (as presently used) may well have shared only very ancient relatives; 'cousins' many times removed. In any case, as this grouping isn't based upon direct common descent, its name receives quotation marks as a warning; "Haramiyida". The authors also point out that their use of dental terminology need not be appropriate, and their interpretation of orientations is tentative. Their guidance is drawn from research into haramiyid and eleurtherodontid critters, unknown beyond the Middle Jurassic, and the characteristics don't necessarily directly apply for animals born nigh on a hundred million years later during the end days of the Cretaceous. However, it's a matter of using the best means available despite their limitations. A survivor It's thought, but not considered certain, that the tooth is an upper molar from the right side of the mouth. If the orientations are correct, then it differs from other haramis by having its main valley confined to the rear part of the crown. In contrast to haramiyids (apart from Allostaffia if so seen), both cusp rows have five clearly separated cusps, and a third row comprised of a single large cusp on the lingual side. That gives a cusp formula of (buccal to lingual) of 5, 5, 1. The oval shaped outline contrasts to eleutherodontids as does the lack of a row of lingual cusps (plural). Running from the buccal side, the rows are termed A, B and BB. A vagary of history means that numbers for uppers run from back to front. For example, the front cusp of row A is A5, and the rear one of row B is B1. This is partly a matter of convention, but it's mainly done so as to confuse unwary students. Some damage mars the front of the postcanine, enamel is missing from the rear and sides, and the roots have got to meet their maker. It's described as saddle-shaped when seen from the side, and the roof ascends both towards the front and rear. Seen from above, the crown becomes narrower towards the back. Five cusps make up the buccal row, and the slope of their internal sides is steep. The first four of them are similarly sized whereas the last, A1, is damaged. The two front cusps of the lingual row are the largest members of that gang, and the next two form from a low ridge. Lingual of B5 is a fairly large cusp termed BB (p.489). The length of the crown was at least 2.43mm, and the maximum width attained 1.33. Affinities? The obvious option with multicusped molars of the Cretaceous is to compare them with the teeth of multituberculates. It always made sense before. Modes of wear reflected a style of chewing known as palnial. However, that didn't apply here. The longitudinal valley between the cusps is a groove for a haramiyidan-like orthal movement, and its features compare more closely with haramis. As mentioned, however, there are also distinctions from both haramiyids and eleutherodontids. These debar Avashishta from both families. A gremlin appears to have been at work on a brief discussion of Eleutherodon. Correctly, it's accused of being known from England and China. It's also said to be known from Middle Jurassic Wales. Presently, no eucynodonts of that age have turned up in Wales. The relevant locality is presumably Watton Cliff and, according to traditional reports, that happens to be in Dorset. I come from Dorset and nowhere in the county gets nearer to the Welsh border than about sixty miles, not even on very windy days. Returning to matchmaking for Avashishta, the closest likeness is provided by Allostaffia. In the view of these authors, the sides were misidentified in the original description of that Tanzanian genus. They see "lingual" as being buccal and vice versa. That changes the "A" row to B and a short "mesiobuccal" cingulum becomes cusp BB on the front of the lingual side. So seen, similarities include the saddle-like shape in profile view; low and clearly distinct cusps (but their bluntness could reflect similar levels of wear); a groove between cusp rows that is deepest at the rear; the buccal wall of this grove being the steepest; BB cusps at the front of crowns. Differences lie in wider, deeper and longer grooves for Allostaffia; ridges linking the front and rear of cusp rows being stronger for Avashishta; and this later genus has a shallow, wide valley between the B row and the lingual edge of the crown. Obviously not discussed in this study is a proposal made later by Hahn & Hahn, 2006. Those authors established a family named Allostaffiidae, and voiced the possibility of Avashishta perhaps being a member. However, this referral is no more than tentative. Even so, it's interesting to note they in no way challenged its possible membership of the "Haramiyida". Holotype GSI/SR/PAL - B215 is employed in Hyderabad by the Geological Survey of India, Southern Division. The specific name refers to the village of Bacharam, which is near to the locality.
References:	Anantharaman et Al (2006), A possible Late Cretaceous "Haramiyidan" from India, Journal of Vertebrate Palaeontology, 26(2), p.488-490.

Genus: Eleutherodon Kermack KA, Kermack DM, Lees PM & Mills JRE, 1998 'freeman tooth' Family: Eleutherodontidae Kermack et al, 1998: Suborder Haramiyina G Hahn, 1973 Remarks: Thanks are due to Frans Keijsper for attempting to decipher a Dutch translation of the generic name as 'free tooth'. As I've now got, but haven't yet read the paper, I can report it actually means 'freeman tooth'. The non-serf in question is Eric Freeman, the first to publish on mammal fossils from the Forest Marble.

Reassigned species: E. oxfordensis (partly) see Millsodon superstes and Hahnotherium antiquum; E. species (China) = Sineleutherus uryguricus

Species:	Eleutherodon oxfordensis Kermack et al, 1998
Place:	Forest Marble, Oxfordshire and Watton Cliff, Dorset
Country:	England
Age:	upper Bathonian, Middle Jurassic
Remarks:	The following is largely based upon my reading of Butler & Hooker, 2005. Originally, thirteen isolated teeth from Oxfordshire were referred to this species. There were four different types as could occur with four tooth positions. Butler & Hooker (p.186) reassigned one to Millsodon and two to Hahnotherium. However, they replaced them with four further E. specimens from Oxfordshire and two more from Dorset. Their descriptions centred on these new fossils. Upper molars These have a narrow end and, as known for sure from Haramiyavia and Thomasia, that's presumably the distal end; ie. the bit nearest the back of the mouth. A characteristic of the genus is that the largest cusp is located there. In front are three rows of lots of little cusps, and they run forwards towards the mesial end. The external (buccal) row is on a ridge which forms the wall of the main lengthways groove. The opposite wall is provided by the middle row. A shallower groove is found between that and the internal (lingual) cusp row. That lingual file also connects the aforementioned large cusp, (B1 in this terminology), with an enlarged cusp on the internal corner at the front of the crown. (Short version: three rows of cusps run along the molar, and there are grooves between them.) The grooves are decorated by small ridges running across their floors. These are also termed flutings. Sizes There's quite a degree of variation in size and proportions between upper molars, (p.187). Twelve vary in length from 2.1 to 3 millimetres. The width/length ratios for eleven of them range between 0.67 and 0.93. Some of this probably reflects different molar positions. Seven fall between 2.4 to 2.9mm in length. Four are shorter (2.1-2.25) and relatively broader. These include the type fossil. Roots At least five specimens preserve roots. All have an undivided root-trunk, which is deeper than the crown is high. This can be subdivided into various elements lower down. For example, BMNH 46832 has four such branches. There's one at the back for the large cusp B1, another at the middle on the buccal side, and two at the front (one per side). In some cases no additional roots are present, (p.188). In all instances this is un-multituberculate-like. With some Upper Cretaceous exceptions, multis were content with two distinct and similarly sized roots on upper molars. Lower molars A pair of the new specimens are incomplete lowers. A Dorset one, BMNH 46851, spent some time rolling around after the death of the owner. This playful behaviour was presumably at the behest of moving water, and it cost most of its enamel. However, the decorative fluting in the basin is preserved and wear was light during life. Seen from the occlusal perspective, the crown is a longish oval with maximum measurements of 2.2 and 1.55mm, (length and width respectively, p.189). The highest cusp is on a prominence at the front. This is termed b2. A line of (probably) six cusps runs along a ridge on the buccal side, and there are traces of a broadly similar number on the lingual aspect. Both lines carried on until the back of the tooth. The central basin was surrounded. Holotype The holotype is a right upper molar named BMNH M46460, and it resides in the collection of the Natural History Museum, London. At a guess, I'd imagine the specific name refers to Oxford. Additionally Kemp, 2005 (p.141) states these teeth are similar to those of Theroteinus. Differences include an extra line of cusps on the uppers, and more cusps on all molars. Furthermore, wear patterns are consistent with propalinal jaw action in this genus.
Reference:	Kermack et al (1998), New multituberculate-like teeth from the Middle Jurassic of England. Acta Palaeontologica Polonica 43(4), p581-606.

Link: Acta Palaeontologica Polonica 43(4) http://www.paleo.pan.pl/acta/acta43-4.htm The Abstract.

Species:	Eleutherodon sp. Maisch MW, Matzke AT, Grossmann F, Stöhr H, Pfretzschner H-U & Sun G, 2005
Place:	Junggar Basin, Xinjiang
Country:	China
Age:	Jurassic
Remarks:	Update, 2010 This material was subsequently reassigned to a new genus, Sineleutherus. Rather than the locality corresponding to the Toutunhe Formation, as thought in 2005, it's now considered to be part of the Qigu Formation dating from the Oxfordian stage of the Upper Jurassic. I've moved my original text down the page.
Reference:	Maisch et al (2005), The first haramiyoid mammal from Asia, Naturwissenschaften, 92(1), p.40-44.

Links: Naturwissenschaften, 92(1). The abstract The description is available to subscribers. Naturwissenschaften, 92(1). Maisch et Al, 2005 The whole paper is presently available in pdf format.

Genus: Haramiyavia Jenkins FA, Gatesy SM, Shubin NH & Amarai WW, 1997 'for Haramiya' Family: Haramiyaviidae Butler, 2000: Suborder Haramiyina G Hahn, 1973

Species:	Haramiyavia clemmenseni Jenkins et al, 1997
Place:	eastern Greenland
Country:	Greenland
Age:	Upper Triassic or perhaps Lower Jurassic.
Remarks:	This is much the best known 'haramiyidan'. Remains consist of a jaw with teeth in place, and some further pieces of the postcranial skeleton. It provides strong evidence that Haramayida is a distinct order, rather than a sub-group within the Multituberculata. Kemp, 2005 (p.140-141) reports a lower jaw length of slightly over three centimetres. The animal was a touch larger than mini. It probably compares with the lower area of the tritylodontid size range. What's left of the rear of the dentary appears mammalian. Whether the back housed a dentary condyle is unclear, but it wouldn't have looked out of place. This condyle is a feature of the jaw joint. Teeth There were four incisors per jaw half. Three of the lowers pointed more or less forwards, and that provided the uppers with a work bench. There's a diastema followed by a pathetic canine. Four premolars have progressively more complex architecture working backwards along the jaw, and the set is completed by three molars Rest of the animal I was going to guess at a skull length of four centimetres. As that's the estimate mentioned by Kemp, I don't need to. Parts of the skeleton are available, but not many of them. These point to a light and agile animal.
Reference:	Jenkins et al (1997), Haramiyids and Triassic mammalian evolution. Nature Vol. 385, p.715-718.

Links: SVPCA 2000, Percy M Butler http://www.soft.net.uk/richardforrest/svpca/2000/abstracts/abstractsae.html Another look at 'haramiyidans'. A concise, informative abstract. Dr Steven M Gatesy, Brown University, RI. http://www.brown.edu/Departments/EEB/images/gatesy_pictures/11.JPEG A sketch reconstruction of the dentary. The scale bar suggests that the complete bone was about 3cm in length.

Genus: Hypsiprymnopsis Aka: ?Hypnoprymnopsis Remarks: My best available information is that Hypno. is probably a misspelling of Hypsi.

Species:	Hypsiprymnopsis rhaeticus Dawkins, 1864
Place:	Somerset
Continent:	England
Age:	Upper Triassic
Remarks:	This is a nomen dubum, (dubious). It's based on an isolated tooth from somewhere near the Bristol Channel. The fossil got lost and the locality is now unknown. (With thanks to Vince Ward for some info).
Reference:

Link: Prehistoric Data Files, Angellis http://www.angellis.net/Web/PDfiles/rept-go.htm This is a listing of a broad range of genera. Mesozoic mammals are hidden away amongst 'Marsupials'.

Genus: Kirtlingtonia Butler PM & Hooker JJ, 2005 'from Kirtlington' Family: Eleutherodontidae Kermack, Kermack, Lees & Mills, 1998: Suborder Haramiyina G Hahn, 1973 Remarks: Presently, the genus is only known from one site. The name is geographical.

Species:	Kirtlingtonia catenata Butler PM & Hooker JJ, 2005
Place:	Forest Marble, Oxfordshire
Country:	England
Age:	Bathonian, Middle Jurassic
Remarks:	The following is based upon my reading of Butler & Hooker, 2005. Fossils are presently restricted to two upper molars. An upper premolar may also belong. Upper molars The upper molar has a central basin banked by a crest on the buccal side, (p.192). This crest contains a cusp at the front and another in the middle, with minor bumps trailing along behind. The basin is shallow and contains no ornamental fluting in contrast to Eleutherodon. On the lingual side is the central cusp row. The foremost cusp is moderately high and its followed by around half-a-dozen successively smaller colleagues; a mother duck with six ducklings, with the chain swaying externally as we reach the back swimmers of the file. At the end this chain approaches the cusp in the middle of the buccal crest. Other features include a large cusp at the rear of the crown and a shrinking chain of three or more cusps. Lingual of the central row is found an irregular congregation of small cusps. Cut off before its prime The type fossil is a tooth that never erupted. It's 2.2 millimetres long and 1.7 wide. A second broadly similar specimen did get to do some work. It preserves an undivided root, which is a characteristic associated with the final molars of 'haramiyans'. That paratype shows some wear, but only as scratches going along the basin, (p.193). Breakages may be responsible for there being no evidence of further wear in several potentially strategic places. Holotype The holotype is right upper molar crown. It goes by the name of BMNH M46497 and lives at the Natural History Museum, London. The specific name is Latin for chain, and this refers to the chains of minor cusps.
Reference:	Butler & Hooker (2005), New teeth of allotherian mammals from the English Bathonian, including the earliest multituberculates, Acta Palaeontologica Polonica, 50(2), p.185-207.

Genus: Millsodon Butler PM & Hooker JJ, 2005 'Mills' tooth' Aka: Eleutherodon (partly) Family: Theroteinidae Sigogneau-Russell, Frank & Hemmerlé, 1986: Suborder Theroteinina Hahn, Sigogneau-Russell & Wouters, 1989 Remarks: The generic name is in memory of Professor JRE Mills, and honurs his orthodontic expertise, which helped greatly towards understanding how Mesozoic mammals used their teeth.

Species:	Millsodon superstes Butler PM & Hooker JJ, 2005
Aka:	Eleutherodon oxfordensis (partly)
Place:	Forest Marble, Oxfordshire and Watton Cliff, Dorset
Country:	England
Age:	Bathonian, Middle Jurassic
Remarks:	The following is based upon my reading of Butler & Hooker, 2005. The genus was established on the basis of two lower molars which were originally assigned to Eleutherodon. They're an m1 and an m2. An upper molar may also belong, although its location is now unknown. It was supposed to arrive in London from Oxford but ran off. It's presently represented by photos and sketches. A quick tour of a tooth The type fossil is the first molar from the right of the series, (p.189). The outline from the occlusal perspective is a longish oval (length 2.4mm, width 1.4). The back (distal) is rounded and the front of the crown looks as if somebody cut an oblique slice off. (Nobody did.) At the front is a cusp (a1) positioned somewhat lingually. It's tall, blunt and dominates the landscape, taking up 40% of the available length. Fine details have been eroded away, but two ridges run from front to back around a central basin. The lingual ridge is a continuation of the large cusp, and it falls in height along the way. There are traces of two former cusps on it. The ridge on the buccal side is lower, and it also meets the dominant front cusp. This ridge features a lengthy exposure of dentine, which probably resulted from wear. There's a worn, longitudinal groove in the central basin, and it's deepest near to the midpoint; more wear. The single root is broken at 0.7mm. Grooves suggest it may have separated further down. Comparisons The size is similar to the larger molars of Eleutherodon, which also has the basin, ridges and a dominant front cusp. However, that's on the buccal side rather than the lingual one. Nor does the lingual ridge make contact with the cusp, as in Millsodon. The major cusp of 'Mill's tooth' is both relatively larger and more upright. Eleu has numerous cusps on the lingual ridge rather than just a couple, and the decorative fluting of the basin isn't evident in this genus. Comparisons with Thomasia and Allostaffia also failed to satisfy. Holotype The holotype is right lower molar from Oxfordshire, (m1). It's known to its fans as BMNH M46645 and works at the Natural History Museum, London. The specific name is Latin for a survivor. Given the animal's complete lack of health for the past 165 million years, it shouldn't be taken too literally.
Reference:	Butler & Hooker (2005), New teeth of allotherian mammals from the English Bathonian, including the earliest multituberculates, Acta Palaeontologica Polonica, 50(2), p.185-207.

Link: Acta Palaeontologica Polonica, 50(2) http://app.pan.pl/acta50/app50-185.pdf Butler & Hooker, 2005 is presently freely accessible in pdf format.

Genus: Mojo Hahn G, Lepage J-C & Wouters G, 1987 'Mojo' Family: ?Theroteinidae Sigogneau-Russell D, Frank RM & Hemmerlé J, 1986: suborder ? Theroteinina Hahn, Sigogneau-Russell & Wouters, 1989 Remarks: The original authors tentatively referred this somewhat bizarrely named genus to ?Paulchoffatiidae: Multituberculata. Presently, I'm assuming that's incorrect. I can't think of a concise translation for the name, as English often doesn't provide alternatives for the specialized vocabulary associated with Louisianan Voodoo. The explanation as to why this name was selected is of potential interest to gossip columnists, but it isn't actually stated in the paper. A mojo is a charm fashioned from cat bones, and these feline features are supposed to attract female attentions to lucky fellows. Purely out of scientific curiosity, I conducted some tests to find out whether this approach brings the desired results and, surprisingly enough, my experiences suggest it works. The first experiments involved such bones being carried by a tom cat, and so many human females found his advances so compelling, that I took to calling him Irresis-Tibbles. He received plenty of pets, pats, coos and a saucer of milk during a half-hour stroll. Impressed, I undertook efforts to extract several of his lucky bones for a mojo of my own. If anything, the effects were even more impressive. Within seconds I was leapt upon by the cat's owner and a policewoman. As a consequence, I've got plenty of time available for writing these notes; six months assuming parole for good behaviour. None of that explains why this genus is called Mojo.

Species:	Mojo usuratus Hahn, Lepage & Wouters, 1987
Place:	Habay-la-Vielle
Country:	Belgium
Age:	Rhaetian, Upper Triassic
Remarks:	The following is largely based upon my reading of Hahn et al, 1987, and thanks are due to the supplier. As may be guessed from its title, that paper's in German, and this is a language I speak like a native of Westsachsen. That area, which is more generally called Wessex, happens to be in southern England but my German's gar nicht so schlecht. Non-German speakers might also wish to thank the authors for providing reasonably lengthy summaries in both English and French. My own brief introduction This genus is extremely poorly known even in comparison to other haramis (assuming that's what it is). Despite the paucity of remains, if the authors are correct about its affinities, then it'd be remarkably tantalizing; rather like a very brief glimpse of a naked body part of a loved one through a heavily steamed up window. You're not quite sure which part you can see, but that quick flash stirs your blood and makes you sick with yearning for more. The entire animal was undoubtedly attractive and fascinating, but it could be more significant than that; lump-in-the-throat, weight-in-the-heart, yearningly, word-chokingly downright sexy. However, it could also be a misleading tease; a girl deliberately on purpose allowing a quick peep of bra-strapped shoulder to a hopeful, still thoroughly experienced love-sick virgin, only for her then to instantly climb into the number 25 bus and disappear from your life. As you watch the bus depart, your stomach churns with feelings of inadequacy, it begins to rain and you haven't even got enough money left for a bag of chips to sustain you during the four mile walk home. PS: Memory tells me an experienced virgin isn't a contradiction in terms. We're talking of a single, heavily worn tooth; perhaps an upper premolar. The thing is though, that this solitary tooth couldn't have come from the mouth of just any old animal, and quite whose mouth it was is most definitely worth knowing. It could perhaps have been some kind of "haramiyid" mammal, albeit an apparently weird one. Then again, it could stem from a multituberculate, albeit a weird and very early one or, and potentially of even more interest, perhaps it was something else. More fossils of Mojo are urgently required. Getting to know Jo The tooth is reasonably described as multituberculate-like in some striking ways (p.39), and dates from the Upper Triassic. If the bonebed in question were some tens of millions of years younger, then it's difficult to imagine there'd have been qualms about accepting its multi credentials and, given its heavily worn state, it probably wouldn't have prompted all that much consideration. However, should it actually be from a multi, then it happens to be around 50 million years older than any other known specimens. The authors were confident enough to offer a tentative referral to the family of Paulchoffatiidae, rather than simply to Multidom. Mojo was found hiding in a bonebed in Luxembourg, Belgian. This is a southern area neighbouring the Grand Duchy of Luxembourg. The locality came to light due to a road construction project, and was first announced in 1983. It yielded lots of fish teeth, some reptile choppers and a few from cynodonts. These include remains of 'haramiyidans'. And then there was this oddity. According to the diagnosis, the cusps of the postcanines of this genus are fully eroded by wear, although I somehow doubt that applies when they've just been delivered. Nevertheless, the only specimen happens to be extremely worn. Furthermore, that level of wear is given high significance. Naturally enough, it also means various structural details are presently unknowable. What remains shows the crown had two rows of cusps running along its length, and there's a cingulum-like feature at, presumably, the front. One cusp row was wider than the other, and that's most likely the buccal one. The tooth, only about two-thirds of which is actually preserved, was held in place by a pair of roots. The original length would've been about 2.25mm and the width 1.8. Those sparse details reveal quite a bit. As there was a pair of roots and a great deal of wear, this is a tooth from an animal with, at most, very restricted dental replacement. Plenty of wear means the tooth saw much service. Excepting for mammals, toothy critters are serial replacers, and such extreme coronal eradication simply doesn't come about from use. There isn't enough time. Adding a sense of direction The presumed orientations of the tooth are based upon comparisons with paulchoffatiid multituberculates. Paulchoffie upper premolars have buccal cusp rows that are wider than the lingual ones and, if present, additional cusps are more frequently found at the front. If this is appropriate, then it's the rear third of the crown that's got lost. While ruins show the positions of the former cusp rows, the cusps themselves have been worn down to their bases and, at the front, they've melted into each other. Towards the rear of the preserved tooth stand the remains of a narrow ridge between the rows, and that demonstrates wear was less pronounced than at the front. The foremost halves of the rows appear to have been occupied by similarly sized cusps, but how many followers each had can't be seen. In front of row A (buccal) is another worn down cusp (termed C). Lingual from that runs a narrow, cingulum-like ridge. This ridge stands below the level of the cusp bases of both A and B, and has a connection with cusp C. Despite its deep position, it's also been strongly effected by wear. The side walls of the crown run almost vertically down whereas the front wall is less steep. On the front, below the level of both C and the cingulum-like feature, a large facet caused by pressure can be seen. This probably resulted from contact with the preceding tooth. Wear It's plentiful (p.42). This shows the tooth was well used, but also provides clues concerning diet. Teeth in this state can't have been used for puncturing insect armour or chewing yummy portions of meat. They're grinders for dealing with plants. That's not to say the owner can't possibly have been anything but a strict vegetarian, but botany was the primary interest, and the preferred plant stuff can't have been soft. (Should you wish to doubt the security of that assumption, then fine. However, then try to find a carnivore with horizontally worn down teeth.) It's also interesting that the level of wear is clearly stronger at the front of the crown. This pattern must reflect the actions of a complexly constructed upper tooth. Affinities Comparative anatomy has an inevitable limitation. Comparisons can only be made with other stuff that to be available, and ancient faunas are at best only partially known. In this instance, four cynodont groups offered their services and, despite it seeming surprising, paulchoffie multis provided the closest match. It's conceivable that a presently unknown possibility could manage an even better one, but such an option requires a new discovery. No further candidates have since emerged. The degree of wear rules out all insectivores and carnivores. That leaves traversodontids, tritylodontids, haramis and multis. Travie teeth can show lots of wear. However, their upper postcanines are generally wider than long, and they don't possess rows of cusps running along them (p.43). This isn't from a travie. Tritylodontids are also discounted. Only lower postcanines have two cusp rows, but these are of similar widths. No known trity has anything comparable with the cingulum-like structure, and wear is never as extreme as that seen for Mojo. The count of two cusp rows for upper postcanines does occur among haramis, and a transverse structure can be present at the front of crowns. However, it isn't as large as the cingulum-like one here, and there's no record of anything corresponding to cusp C. Furthermore, the valley between the cusp rows is wide for haramis rather than narrow and, again, such extreme levels of wear aren't known. Cusps are left distinct. The authors found most similarities with paulchoffie multituberculates, thus the tentative referral. Wear can be great upon upper premolars. In contrast to most multis, rear paulchoffie premolars were grinders rather than shearers. Cusp rows do differ in width, and are separated by relatively narrow valleys. Small accessory cusps can also be found to the front. However, these are conical bumps rather than akin to this cingulum-like structure. The overall character of this tooth is a bit of a mongrel. The cusp constellation at the front is more reminiscent of haramis (p.45), and such an affiliation is more in line with expectations based upon the time of Earth; something like 50 million years prior to the first uncontested multi fossils. Given the known range of variability in occlusal outlines and cusp configurations among paulchoffies, it appears to be the time factor -not the morphology- which leads to the greatest doubt. Common sense wants to suggest this is too early for a multi to have scampered around, but common sense doesn't know everything. More evidence, please I've got a 2006 review of haramis by Hahn and Hahn. As yet, I've only read it briefly. Mojo receives some discussion, but its placement is still unclear. More closely comparable fossils haven't yet come to light. Some information from that paper will appear later. As well as having an oddly attractive name, this lack of certainty makes Mojo an oddly attractive genus. Whatever it might turn out to be, weird harami relative, remarkably early multi or something else, it's a small relic of a largely unheard story. While already familiar tales can retain their interest, unfolding ones tend to provide more oomph for the curious. This is a small yet significant find, but what it signifies must await further discoveries. Holotype No catalogue number is mentioned in the description. The specimen, presumably an upper premolariform, resides in the collection of the Institut royal des Sciences naturelles de Belgique, Brussels. The specific name is Latin for 'versehen mit Gebrauch / Abnutzung'; something like: signs of use / wear. It was recovered from bonebed HLV-2 of the "Sables de Mortinsart-Formation". Additional notes Some interpretations place this possibly near to Paulchoffatiidae, (Multituberculata), rather than 'Haramiyida'. This view doesn't seem to be currently widely held. "However, one of us (G.W.R.) has studied the specimen and believes it to be a theroteinid", (Wible JR & Rougier GW 2000, p.5).
Reference:	Hahn et al (1987), Ein Multituberculaten-Zahn aus der Ober-Trias von Gaume (S. Belgien). Bulletin de la Sociéte belge de Géologie, 96 p.39-47.

Link: Espace Paléo 3j: Habay-la-Vielle, Dominique Delsate http://home.pi.be/~nerodj2/Npagehtm/txtvtp2_hb1.htm Microvertebrates from the Ardenne of Belgium, (French).

Genus: Sineleutherus Martin T, Averianov AO & Pfretzschner H-U, 2010 'Chinese Eleutherodon' Family: Eleutherodontidae Kermack et al, 1998: Suborder Haramiyina G Hahn, 1973 Aka: Eleutherodon (partly) Remarks: The generic name refers to a country named in honour of the Emperor-God Qin Shi Huangdi, and an English harami genus deviously named after Mr Eric Freeman. The authors noted similarities between the teeth of this Chinese critter and the earlier described and earlier demised Eleutherodon. Indeed, a tooth previously assigned to that genus is now included in this new taxon

Species:	Sineleutherus uryguricus Martin et al, 2010
Aka:	Eleutherodon sp. (partly)
Place:	Qigu Formation, Xinjiang
Country:	China
Age:	Oxfordian, Upper Jurassic
Remarks:	I've got the paper, but haven't yet added much information from it. Thanks are due to Kees. Holotype The type fossil, SGP 2001/33, is a lower left molar presently chewing over its future at the Universität Bonn. At some future point, possibly after celebrating graduation, it and other Qigu fossils will be returning home to work at the Research Center of Paleontology and Stratigraphy, Jilin University, Changchun. The specific name celebrates the Uygur people who live in the area. Additional irrelevancy I happened to sell one of these authors a book on pocket watches via Amazon in 2009! The fossil previously known as Eleutherodon sp. The authors of the 2010 paper found fault with some of the following information published in 2005. For a start, they could distinguish the fossil from the English genus, but they had the advantage of further specimens being avaiable for interview. As I haven't yet made any changes to the text below, it's still 'infected' with bits that are somewhat misleading. In addition, the 2005 interpretation of this locality was that it corresponded to the Toutunhe Formation, and was tentatively dated as being from the Bathonian part of the Middle Jurassic. New research indicates that was wrong. The following is based upon my reading of Maisch et Al, 2005. Thanks are also due to Dr Michael Maisch for advance notification on this find, as he let me know in 2001. The Junggar Basin of Inner Mongolia is becoming increasingly generous with fossils from a number of localities and ages. This fossil, a lower molar, is the first evidence of Mesozoic mammals from the southern area of the Basin (assuming haramis were mammals, p.40). It's also the first 'haramiyid' from the whole of Asia, and the first vertebrate from the Formation belonging to a taxon known from elsewhere (England). That suggests the possibility that the Toutunhe Formation is similar in age to the Bathonian Forest Marble, and that some measure of faunal interchange occurred. Shuotherium concurs on this, as it's known from elsewhere in China and Oxfordshire. Setting This fossil came from the Liuhuanggou Gorge, which is located about 40km southwest of Urumgi. It was in a bonebed about 15 metres below the junction of the Toutunhe and the overlying Qigu Formation, and was accompanied by isolated remains of various fish, temnospondyl amphibians and reptiles. Awaiting description are a few teeth from a new docodont mammal. A further horizon with fossils is found a bit higher up, so that's presumably somewhat younger. It's termed the turtle-archosaur-amphibian assemblage as it contains... I think I'll leave you to work that out for yourselves. The truth of the tooth The crown has only two rows of cusps rather than three, and that's characteristic of lower harami molars (p.41). "The position of the main mesiobuccal cusp allows identification as a left lower molar." It's somewhat wider at the front than the back, and the cusp rows don't run in parallel. They converge to the rear (p.42). There are five cusps in the buccal row, with the front one being much the biggest. Following the conventions of harami terminology this is b2. The next in line is over 80% smaller and a third less the size of its immediate follower. The final three decrease in size along the line, although the last still outdoes the second cusp. Wear was strongest at the rear of the crown. If there were a cusp positioned between both rows no sign of it has been preserved. Four cusps constitute the lingual row and they're fairly similar in size. The first, presumably a1, is much inferior to b2 while a2 is a bit higher then a1. The final lingual cusp is the lowest in the line. The two rows are separated by a wide basin extending over half the width of the crown, and this area is marked with ridges caused by wear. They run from the cusps and are most pronounced for the rear three cusps in each row. What remains of the root suggests it had three sub-divisions. Affinities Among the harami credentials of the tooth are two rows of cusps with the foremost buccal cusp being the largest. The basin between the rows extends for most of the crown's length, and that rules out close affinities with Theroteinus. It's a character of haramiyoids. As the authors could find no strong distinctions from the Oxfordian Eleutherodon, they referred the fossil to that genus. There are fewer cusps than found on any of the British equivalents, but there's a considerable degree of variation in that regard, and this could reflect different tooth positions. Otherwise the overall morphology matches well. If the difference in cusp numbers were maintained across a larger sample, then it could have significance. However, as the Chinese sample is presently restricted to a single specimen, there's no useful information as to variation available. Middle Jurassic poverty Until this description appeared, the inventory of Middle Jurassic mammals from Asia (the largest continent on the planet consisted of (perhaps) Shuotherium and Liaotherium, a possible docodont upper molar from Kyrgyzstan and a lower molar known as Tashkumyrodon (also Kyrgyzstan). That isn't exactly a lot (p.43) and any increase is welcome. Generally, Middle and Upper Jurassic vertebrates from Central and East Asia don't belong to taxa which occur elsewhere. Much of the continent was isolated. However, both Eleutherodon and Shuotherium were able to buck this trend. Similar principles prevail with regard to contemporary isolated bits of land. The ocean provides an effective barrier for larger tetrapods, but the small fry can disperse over considerable stretches of the stuff. For example, most Pacific islands have been populated by rodents. Despite the isolation, some measure of migration was clearly possible for mini mammals. Maisch MW, Matzke AT, Grossmann F, Stöhr H, Pfretzschner H-U & Sun G (2005), The first haramiyoid mammal from Asia, Naturwissenschaften, 92(1), p.40-44.
Reference:	Martin, Averianov & Pfretzschner (2010), Mammals from the Late Jurassic Qigu Formation in the southern Junggar Basin, Xinjiang, northwest China, Palaeobio Palaeonv, 90, p.295-319.

Genus: Theroteinus Sigogneau-Russell D, Frank RM & Hemmerlé J, 1986 Aka: Theriotenus Family: Theroteinidae Sigogneau-Russell D, Frank RM & Hemmerlé J, 1986: suborder: Theroteinina Hahn, Sigogoneau-Russell & Wouters, 1989

Species:	Theroteinus nikolai Sigogneau-Russell D, Frank RM & Hemmerlé J, 1986
Place:	Saint-Nicolas-de-Point
Country:	France
Age:	Rhaetian, Upper Triassic
Remarks:	According to Kemp, 2005 (p.141)... This genus is based on wide, isolated postcanines with two or three rows of low, round cusps. "The root was probably divided into two or three, and the histology of the enamel resembles that of other mammals." These could apparently be milk teeth of other 'haramiyidans'. Another suggestion is the speciailisations may indicate a different diet; hard and brittle food. Wear reveals no element of propalinal movement in the jaw action. That leaves me wondering how they could possibly be harami milk teeth, given that the jaws worked in different ways. Affinities and inclinations The following is based upon my reading of Butler & MacIntyre, 1994, pages 455-456. Theroteinus is an eccentric candidate for a harami, but there do seem to be some signs of kinship. However, there are also some highly significant differences in the teeth. (Nothing other is known for a theroteinid.) The cusps of molars are low and rounded in this case, and wear indicates a predominantly orthal approach to tackling food. That contrasts to the backward drag (palinal) chewing stroke of haramiyids. The jaws must've worked differently. This could have something to do with a specialisation on particular foodstuffs. The hard and brittle option is mentioned above, but an earlier suggestion plumped for soft 'fruits'. Postcanines As with haramiyids, the lower postcanines have two cusp rows either side of a central basin, and this is closed at the rear by a u-shaped ridge. A cusp technically termed c occurs on a cingulum at the front, and this could be homologous with the harami cusp then known as b. It's the uppers which complicated matters, as they have three cusp rows rather than a pair. In language emended by myself, there is a 'C' row on the lingual margin beyond the 'B' row. (No inverted commas are used in the 1994 paper, and I'm not sure what the current terminology is.) If related with haramis, which seems likely, then the therodontinid middle 'B' row is presumably a new development, and it's the 'C' row that would correspond with the harami 'B'. This barricading of the central basin would've been fatal for the chewing action of critters such as Thomasia, but this family had different table manners.
Reference:	Sigogneau-Russell et al (1986), A new family of mammals from the lower part of the French Rhaetic. p. 99-108 in Kevin Padian (ed.), The beginning of the Age of Dinosaurs. Faunal changes across the Triassic-Jurassic boundary. Cambridge University Press.

Link: Acta Palaeontologica Polonica 45 (4) http://app.pan.pl/acta45-4.htm#Butler Butler PM, 2000. Review of the early allotherian mammals. The Abstract. This is my source for the sub-orders Theroteinida and Haramiyoidea.

Genus: Thomasia (Plieninger, 1847) Poche, 1908 'for Thomas' Family: Haramiyidae Simpson, 1947: Suborder Haramiyina G Hahn, 1973 Aka: Microlestes ('small thief') Plieninger, 1847; Plieningeria ('for Plieninger') Krausse, 1919;Microleptes ('small thief') Simpson, 1928; Haramiya ('petty thief' feminine) Simpson, 1947 Remarks: According to the German language version of Wikipedia, the generic name honours Oldfield Thomas. References: Poche F (1908), Einige notwendige Änderungen in der mammalogischen Nomenclatur. Zoologische Annalen, 2 pp. 269-272. Simpson (1947), Haramiya, new name, replacing Microcleptes Simpson, 1928. J. Paleontology 21(5), p.497. Remarks: This genus turned out to represent the lower teeth of Haramiya. Thomasia takes precedence on the basis of seniority. This entry has been constructed in line with the proposals of Butler & MacIntyre, 1994. At least, that was the intention. As that study was too early to benefit from the information supplied by Haramiyavia (the only harami known from anything other than isolated teeth), some of the thinking has been overtaken; eg. speculation on possible dental formulae (correct to some extent), and a tentative (very tentative) hypothesis of possible affinities with non-mammalian traversodontids. However, the following sentence from the Abstract is presumably still valid: "The chewing movement resembled in principle that of the traversodontid Scalenodon." Before beginning my notes, I've also referred to Kemp, 2005 (p.140-141) for a convenient and more contemporary summary of subsequent developments. I've also updated or emended some terminology. At least, I hope that's what I've done. I did intend attempting to keep the entry for the genus fairly concise, and I've failed most happily. The earlier paragraphs are something like a summary, and then things get a bit more detailed. Those searching for a brief snack are free to move on without sampling all the dishes. The direct subject of the study was provided by teeth obtained from Holwell Quarry in Somerset. These had been collected and described earlier (p.433). However, as a considerably larger number of similar fossils had then recently been recovered from Saint-Nicolas-de-Point in northwest France, that new source of information justified a review. Material included both upper and lower molars but, unfortunately, nothing other than isolated teeth was known. While the molars appeared complementary, that alone didn't necessarily mean they had to be from the same animal. However, the large French sample pointed strongly to such a conclusion for statistical and morphological reasons, and the later discovery of Haramiyavia closed the issue with its long awaited jaws. But that was after this study. It was correctly deduced that the uppers (assigned to Haramiya) and the lowers (Thomasia) belonged together. Differences used to construct two 'genera' were matters of up or down. (Gremlins seem to have gone to work on page 140 of Kemp, 2005, and they've playfully given the lowers to Haramiya. They shouldn't have done that.) All teeth aren't equal Some readers may now be able to remove their own cheek teeth from their mouths (the premolars and molars), and take a quick look at them. Others would be best advised to refrain from that. Those teeth differ. That sort of variation had earlier led to an outbreak of dental speciesitis; the establishment of new 'species' based on teeth from different positions. Recognition of such matters can be simplified by larger collections, and one had just become available. The authors drew out their knives and cut the number of Holwell species down to two or, less likely, three. And this conclusion was rooted in the gums of size. The larger remains were assigned to Thomasia moorei and the smaller to T. antiqua. Euthanasia consigned T. anglica, H. butleri and (probably) H. fissurae to the latter. Upper molars from this quarry were found to be more derived than their earlier French colleagues, but evolution's busy elves hadn't felt moved to tinker noticeably with the lowers. Nearview mirrors Although broadly similar in this case, upper and lower postcanines aren't identical. The basic idea is a crown with two rows of cusps (A and B) separated by a basin. One end (termed + is the paper) is nearest to the largest cusp of the B row, and the other (-) features a curved ridge closing one end of that central basin and connecting the cusp rows. For lower molars the + end is to the front. This is reversed for uppers. Wear on the teeth provides clues as to how they worked, as it's inflicted by collisions while eating things and foodstuff. The primary method of processing yummies involved the uppers being dragged backwards along the lowers (a palinal action), with little sidewards movement. As well as a mirroring of the + and - ends, features on the flanks of the crown surface are also reflected; the A row is buccal on uppers while being lingual on lowers. (Note: In this paper, a capital letter is used in both cases. I know this practice has changed, and lower case letters will be used for the lowers further down the page. I'll briefly mention why later.) It could also be useful to mention that, in some studies, the + end was consistently termed anterior, and that would be wrong for the uppers. Nevertheless, that's the end cusp numbers run from: A or B 1, 2, 3 and so on. (That may have since changed but I'm unsure.) Rounding up Thomasia and changing names The first specimen captured was named Microlestes antiquus by Plieninger in 1847 (p.434), and it was arrested at Degerloch in Germany. Being Upper Triassic (Rhaetian), it's older than the Holwell specimens. As things turned out, that generic name had already been used for a beetle. (I'm not sure whether Schmidt-Goebel was referring to Paul, John, George or Ringo in 1846.) However, six decades elapsed before a new name was proposed. With no obvious regard for the intentions of the original author, Poche strode in with Thomasia. While there's no compulsion to respect the first intentions with nomenclatural changes, it would strike some as being polite and parsimonious to do so. Perhaps views differed in 1908 and, in any event, Thomasia is valid. At some stage the specific name was also emended. Similar fossil had been collected from Holwell in Somerset, and these were used to establish a second species, M. moorei, with the name honouring the discoverer. This meant the original species at least enjoyed some companionship at the re-naming ceremony. However, Simpson then viewed the English material as meriting a separate genus and, being of impeccable manners, he selected Microcleptus in 1928. This had the same meaning as Pleininger's choice; 'small robber'. However, either Gravenhorst (1829) or Newman (1840) (sources vary) had already used that for another insect. Undaunted by roving beyond Latin and Greek, Simpson turned to Arabic for Haramiya ('petty thief'). What he'd also done (in 1928) was to establish two further harami species; M. fussurae and T. anglia. Subsequent researchers added more harami specimens. Kühne visited Holwell (1946), Germany provided some for von Huene (1923) and Hahn (1973), Peyer obtained a number from Hallau in Switzerland (1956), and then Belgium donated a solitary one (1984) for Wouters and colleagues. It is possible (although extremely unlikely) that the President of France felt it essential to discover French harami teeth, as the Republic was falling behind its neighbours. In any event, the 200 plus specimens from Saint-Nicholas-de-Port would've been a fine response to any such prayers. This single site (which apparently now sports a rubbish tip) set a new world record for harami population density. They were first discussed by Sigogneau-Russell in 1989, and she also established a new species the following year (H. butleri). This Gallic numerical superiority, however, still provided nothing other than small, isolated teeth. While they were very welcome, Greenland finally decided to step in and trump them for novelty. It did so by releasing the jaws and some skeletal remains of Haramiyavia. And harami lovers applauded the feat in approval. Dental formula (of a relative) The authors then outline various interpretations regarding how the teeth were positioned, and how they operated as parts of a set (or perhaps two sets). Which way round they go and so on. As Haramiyavia was kind enough to supply the answers, much of this discussion has lost its bite. It does seem worth mentioning that Parrington got the alignments right in 1947, and Sigogneau-Russell thought both 'genera' belonged together in 1989, with Haramiya being the uppers. She also deduced (purely from isolated teeth) that two or three molars were present on each jaw half. Kemp, 2005 offers the formula for Haramiyavia (p.140), which is presumably similar; (Uppers): 4 incisors, 1 canine, 4 premolars, 3 molars; (Lowers): 4, 1, 4 and 3 respectively. He also adds that the upper incisors are short and vertical, whereas the first three lowers are forward pointing things (procumbent), and they provide a working surface for their abovelings. The presence of canines was something as a surprise, seeing as none had previously been noticed, but they are small. (Repetition for clarity: That is the formula for a haramiyid, and not strictly for Thomasia). Hang on! If Haramiya is actually Thomasia, then why call it a haramiyid? Thanks for the interest, but there was no need to shout. The reason is that the family is called Haramiyidae. Should a genus name become invalid because it's found to belong to an earlier established one, or because the name is preoccupied, or for any other esoteric cause, then this is no reason to alter established names of wider taxa; eg. families or orders. Thus, while Haramiya has fallen out of formal use, haramiyid, haramiyidan and so on, have not. Thomasia remains a harami. This is anything but unparalleled. One of the most famous of non-birdie dinosaurs is Triceratops, and it's a ceratopsian. This is regardless of the detail that the genus Ceratops has got six legs. At least, its healthy members have, seeing as they're beetles. I mean to say, if I turn out to be in a lawful sense 'illegitimate' (If a person can be 'illegitimate', then there's something severely wrong with the law.), then why on Earth should my kids be expected to change their names? Mammals? Of course my children are mammals! Are you trying to insinuate... Oh, apologies. I've just realised I posed that question, and it relates to haramis. Butler & MacIntyre give a brief summary of observable mammalian characteristics. The 1994 list included the molars being multicusped and double-rooted but, as stated, that also applies for the non-mammalian tritylodontids. Another point is awarded for the enamel, termed 'preprismatic', the sort of level of organisation sported by Morganucodon. Their small size has also been accused of being suggestive of: "a high level of homeostasis". (I would imagine that's in conjunction with all available information. Smallness alone signifies smallness.) As mentioned earlier, the jaw of Haramiyavia provided some further support for membership of Mammalia, but not enough to be conclusive. What's known of the anatomy is consistent with haramis either being mammals or merely extremely close relatives. Dimensions The entire determinable size range for the upper and lower postcanines from various positions fell between 1.35-2.55 millimetres. Various other lengths, widths and heights of features are also given, should the thirst be unquenchable. Wear (p.436) Teeth got worn down in several ways, and this provides information. Wear facets form at points of regular contact between uppers and lowers, and these can help demonstrate the forces and directions of movement involved. More wear results from abrasive food, and that can eradicate cusps. Further effects may be added by remains being played about with while being washed along in mischievous rivers or tides, and also from being collected and processed. However, if wear occurs in the corresponding places on several specimens, then this is unlikely to be random, and there will often being confirming features showing wear was caused from dental activity, (eg. scratches). Haramiya - a quick memory prompt place It may be recalled that this generic name was used for the upper teeth. If not, then I've just reminded you. For orientation, the + end of these postcanines is at the back, and the A row of cusps is on the buccal side. I've emended the text used in the 1994, and here's why. The numbering of cusps began from the + end, which is actually the rear. Both A1 and B1 were the final cusps, and that seems rather perverse. I'm simply going to insert inverted commas around the position; ie. where, in the paper, A1 is written, this'll appear here as 'A1'. I'd imagine the numbering should run from front to back, and it's conceivable the letters ought to be swapped round but, being uninformed on the matter, it's not something I'm going to do here. That would be guesswork. (Perhaps current practice actually features in one of the other entries on this page, and I've forgotten it.) Upper postcanines The uppers from Holwell fall into two broad groups previously distinguished by Sigogneau-Russell. These are based on proportional relationships and morphological characters, rather than absolute size, and appear to indicate molars and premolars. Group I has 'A' and 'B' cusp rows of equal lengths, and the - end (on the front of uppers) is relatively wide. With group II, the 'B' row stops further from the - end, and the crown is proportionately narrower there. Subgroups were also recognised in the wider sample of Saint-Nicholas-de-Port, and this could reflect different molar positions. H. group I (molars Fourteen were found in the Holwell haul; eleven larger ones for T. (H.) moorei and three from T. (H.) fissurae. I'm inclined to confine myself to mentions of the type fossils, so as to just provide a sniff of the much wider feast in the paper. Some information on lower teeth can be found below in the entry for T. anglica. The following only concerns uppers. T. moorei M211 is from T. moorei. The widest part of the crown is near the + margin (the rear), where the 'A1' and 'B1' cusps are found. That margin extends somewhat further on the 'B' row than for the 'A' row, and the occlusal outline is something like a rounded, oblongish sort of shape that's been pinched and pulled about in several areas. The 'A' row (buccal) had three cusps and 'A1' is broken. Its base is a bit shorter than that of 'A2', which is higher than 'A3'. This file of three runs more or less in parallel with their five friends on the other shore of the basin; the 'B' row. Of those, 'B1' is the highest, and 'B4' and 'B5' are similar to each other in that quality. A cingulum features on the 'A' row side of the tooth, but there's also a bit of a cingular ridge on the opposing side running from 'B2' to 'B3'. The - end (front) houses a flattened u-shaped ridge. This connects both cusp rows and closes off the central basin. This particular specimen hadn't been subjected to much wear, but there was some on the tips of the cusps, and on a small cusp und cingulum at the + end. The roots aren't preserved. Bases, however, are present for two and these suggest roots of similar sizes. T. anglica M2401A, the type of H. fissurae (p.441), is a molar with a roughly elliptical occlusal outline, and the half of the crown bearing the 'B' cusp row extends further at the + end (back) than does the 'A' side. The crown is consequently "very oblique" in that region. It has a length of 1.95mm and a width of 1.55. The 'A' row boasts three cusps, with 'A1' being broken and relatively small. The 'B' row curves somewhat along its course, 'B1' is set further towards the midline of the crown, and there are only four cusps in this parade. There are no cingula and wear has been modest. Two of the H. fissurae specimens are complete, and they are close in size to smaller teeth from T. moorei. However, they're proportionately wider and the + end is more oblique. Of the cusps, 'A1' is comparatively smaller, and 'B1' further towards the crown midline, thus the curve of the 'B' row. There's no distinct 'B5' but can be a bit of a bump on the u-shaped ridge at the - end. The 'B2' is also not as distant from that end. (The distance of 'B2' from the - end is used for diagnostic purposes with harami species, but it's not my intention to provide possibly poor summaries of diagnoses. The only worthwhile source for a diagnosis is the paper itself.) H. Group II These teeth could be premolars (p.442) and, as the size range is variable (1.4 - 2.26mm) at Saint-Nicholas-de-Port) presumably they come from different positions. In comparison to Group I uppers, the 'B' row of cusps is shorter and gives way to a small ridge beyond the level of 'A3'. The - end (front) is also narrower. A further difference is the greater ratio of length compared to width. Holwell contributed two specimens, one larger (1.8mm long) and the other smaller (1.35). In terms of proportions they conform to H. butleri, and could be from distinct positions. The larger one resides in Bath and answers to the name of M217. Its - end (front) narrows to a curved margin, whereas the + end is wider and fairly straight, with a bulged extension beyond the 'B' cusp row (lingual). Not much wear occurred but there's damage caused by water transport (rolling). The 'A' row has three clear cusps, with 'A3' positioned slightly more to the medial than its colleagues. 'B1' is about as high as 'A1', and the rest of that row has been erased. It transforms itself into a ridge when level with 'A3'. Remains of roots suggest a subdivided one lurked at the + end, and a narrower root was present at the front. Lower postcanines At this juncture, the terminology in the study flips around. The + end is the front rather than the rear, and the a cusps are on the lingual side. All this means that cusp numbers run from the anterior margin, as is generally the case, and I've got no reason for using inverted commas. Nevertheless, I have emended the notation for cusps. Capital letters are used in the paper. The practice followed in Butler & Hooker, 2005 is to employ lower case ones, and that's what I intend to do here. To give a flavour by example of this highly complex translation, I'm converting A1 to a1 and so on. Intentional exceptions may happen in quotations. Unintentional cock ups could also slip in. The distinctions between upper postcanines (formerly known as Haramiya) and lowers are rather subtle, and some lie in the cusps: "... A1 is higher than B1, and A3 is very small or absent" (p.443). There are again Groups I and II, but only examples of the former were found in the material from Holwell Quarry. In contrast to the corresponding upper teeth, the lowers of Group I are proportionately narrow, and this is more pronounce to the rear (- end). The front of the crown doesn't widen (as it usually does for H.), and the sides are almost parallel with, sometimes, convergence towards the anterior. Furthermore, the basin separating the cusp rows isn't as wide, cusp b1 isn't positioned further towards the midline than its colleagues, and the b row fails to curve or, at least, curves with very modest enthusiasm. Of the roots, only the front one is wide whereas the rear one expresses itself with more elongated tendencies. Group II lower postcanines, presumably premolars, are narrow to the fore as there's no distinct cusp b1. It could possibly have amalgamated with a1. Across the Channel A large variation in lengths of Group I lower postcanines is consistent with different positions. The range spreads from 1.09 - 2.49mm, and the long ones are cuspier. No a3 is found on smaller models, and b3 only occurs on the u-shaped ridge at the rear of the basin. It's further forward on larger examples as part of a straight line of four cusps, with a b5 set somewhat lingually. There are several other details which vary in accordance with the molar length, and one concerns the roots. They don't divide on the smallings. It was thought these also simpler teeth were possibly from the rear of sets. Whether confirmation was provided by Haramiyavia is something I don't know. Back to Somerset If these considerations on positions are right, then the available Holwell specimens included nine front molars (larger), one posterior one but no premolars (Group II). The smaller tooth is named H7. It was collected by Kühne and now studies in the museum of Zoology, Cambridge University. It's 1.75mm long and 1.25 wide. The front margin (+) is broader than the oblique rear one, so it's sensibly accused of being 'trapezoidal' in occlusal outline. That obliqueness results from the b cusp row being considerably longer than the a one. Excepting for a1 and b1, all cusps have been eroded away on this specimen. What remains is consistent with two cusps in the lingual row and three for b. The ridge at the rear of the central basin has heavy wear directed down into the basin. More wear scars the external slopes of the b cusp row, and has left facets on the fronts of both remaining cusps. The single root was probably not divided, but preservation doesn't permit certainty. Some information on one of the larger specimens is located in the enclosure for T. antiqua, and that may be visited below. Genetic synonymity Discussion concerning the then evidence for Haramiya being the upper teeth of Thomasia is summarized on page 448. As the discovery of Haramiyavia seems to have shoved the matter beyond any reasonable doubt, that limits its interest. One detail could be of use. The number of postcanines referred to both 'genera' from Saint-Nicholas-de-Port were near equal; T. narrowly won 85:81. If the ref had allowed more injury time, then it's possible H. could've caught up and perhaps taken the lead. The reason for the closeness of the match is that the match (in another sense) was valid. The opinion of species Several species have been established over the years (p.449), but the distinctions seem to have reflected different tooth positions and some degree of variation. As well as questioning the usefulness of T. anglica, these authors also challenge H. butleri, which was based on upper front molars. They note, simply on the basis of size, that a few such specimens might be from a larger species, and they do differ significantly from Holwell's T. (H.) moorei. In contrast to the conservatives inclinations of lower postcanines, there are more adventurous traits in the Holwell uppers compared to those from Saint-Nicholas-de-Port. The 'A1' cusp is considerably smaller, and the 'B' row is more curved. These could be developments caused by the passage of time. Dating remains from fissures can be especially difficult, and the age of Holwell fossils has been much debated. There are now usually thought to be Liassic (Lower Jurassic). Firmer information shows the Saint-Nicholas-de-Port location is no more recent than Rhaetian (Upper Triassic). Several more species do seem to be among the whole harvest of European fossils. A couple of upper postcanines indicate a species of T. (H.) moorei dimensions, while one partial tooth from France is probably from a still larger animal. An upper molar from la Gaume (Belgium) and a lower from Halberstadt (Germany) belong to two further species. The authors concluded all then known haramiyidans belonged to only one genus, and divided that into four named species, with two being new. I show dutifully attempt to follow that here unless there are reasons to do otherwise. Chewing like Thomasia While I would now like to go into graphic and rousing detail about the sexual performance of our long dead relatives (and mention how this area of behaviour has been continually taking to ever greater heights of achievement in subsequent lines of mammals, culminating in the achievements of odd, derived primates known as, oh... It's those bonobo show-offs at it yet again!), the fossil record of Mesozoic eucynodont sex organs isn't terribly good. We'd better have a bite to eat instead. Pages 450-451 are largely dedicated to discussion on tooth occlusion, operation and chewing action as inferred from the teeth and their patterns of wear. It's splendidly detailed stuff and can be admired in the paper. As I seem to recall including bits on this subject in several entries on this directory, an approach with some brevity and perhaps a pinch of levity appears a good idea, and that could excuse the paragraph immediately above. Then again, perhaps it was blatant exploitation of sex. Suffice it to say, as a reminder, the upper postcanines are much like mirrored reflections of the lowers. This is unusual but not without chronological antecedent, as the same can be sensibly said for lots of rodents, and there is known to be an association with eating methods. In the case of those rodents, the main action involved is termed propalinal chewing; ie. a longitudinal action with little sidewards activity. That also happens to nicely match the wear paterns for Thomasia. This was handily researched by Parrington in 1947. As may be recalled, the crowns have a central basin between two rows of cusps; 'A' and 'B' or a and b. The a is lingual for lowers while 'A' is buccal on uppers, but these points apply for them both. The word 'internal' here refers to the basin and 'external' to the opposing face of the cusps. (I've selected that pair of terms.) OK, the 'A'(a) row wears internally but never externally, but 'B'(b) gets worn on both sides. This is because 'A'(a) occluded with the external side of the opposing 'B'(b) row, whereas 'B'(b) rows moved through the central basin. Pardon? An upper 'B' row travelled outwards over the tooth below, and was then drawn downwards and backwards. It landed towards the middle of the front of the lower tooth at an angle of about 40°, filed along the basin, and then departed out over the ridge to the back. This kind of behaviour wears the front of the lower tooth, scratches the basin floor lengthwise, can batter the rear ridge, pummels the internal sides of both banks of the basin and both flanks of the 'B' row being dragged back. But it never vandalises the external face of the a row. And the situation is very similar for wear on the upper teeth. Going straight, but differently While this here longitudinal chewing action is superficially common to both dead haramis and many living rodents, it results from opposite processes (p.451). The effective stroke for Thomasia was palinal, and that refers to the backwards drag. Rodents favour the forward push method (proal). Either was works well enough for grinding dinner. Affinities This theme is considered on pages 451-456. Three hypotheses for relationship of haramis were offered. 1. Affinities with the non-mammalian traversodonts; 2. Haramis were mammals with links to multituberculates; 3. They were part of a non-mammalian line of very mammal-like tendencies along with travies, tritylodontids and multis. At that time, as stated on page 456 with regard to those hypotheses: "On the basis of parsimony, hypothesis 1 is to be preferred, but the problem will be solved only when more information is available, particularly from the early Jurassic. (A small 'e' is used for 'early' in the text, so it must've been an informal usage.) Of course, the more extensive remains provided by Haramiyavia enriched the picture, links with the travies seem untenable, and close links with tritys unconvincing. Multis? Butler has since come to favour the possibility of multi affinities for at least some haramis, especially the theroteinids such as Theroteinus. (Butler & Hooker, 2005 contains some discussion on this.) As his co-author, MacIntyre, died prior to publication, his interest in the matter faded. Most researchers see no particular basis for linking haramis with multis. The nature of these beasts remains delightfully obscure. Final caution I do not know whether some of the terminology used in the 1994 paper is still current, and am sure in several cases it isn't. There may have been further refinements. Botanical department Thomasia is also an Australian flowering plant genus; eg. T. stelligera. As botanical and zoological nomenclature are not unified, plants and animals can share a common name.

Links: Photographer: Greig, D. http://155.187.10.12/images/photo_cd/809931012633/081.html An attractive photo of the plant in bloom. The Dinosaur Mailing List, Peter Depuydt http://www.cmnh.org/fun/dinosaur-archive/1997Apr/msg00643.html There's a lot of good information lying around in this archive.

Species:	Thomasia antiqua (Plieninger WH von, 1847) Simpson, 1928
Aka:	Haramiya butleri Sigogneau-Russell, 1990; H. fissurae (Simpson GG, 1928); Microlestes antiquus Plieninger, 1847; ?Microcleptes fissurae Simpson, 1928; Thomasia anglica Simpson, 1928.
Place:	Degerloch & Olgahain, Germany; Hallau, Switzerland; Boisset, Saint-Nicolas-de-Point & Varangéville, France (all Norian late - Rhaetian early); Holwell quarries, England (?Lower Jurassic).
Locations:	Western Europe
Age:	Rhaetian-Liassic, Upper Triassic - Lower Jurassic
Remarks:	Upper teeth For editorial reasons, some discussion is included in the above entry for the genus. Rather than writing it all out again, let's be brief. The essential difference between the uppers of this species and those of T. moorei is size. This is the smaller version. It can be sensible assumed that this trait was reflected in other body areas, but only teeth are known. Lower molar The following is based upon my reading of Butler & MacIntyre, 1994; a study of specimens from Holwell Quarry, Somerset. Among the material was the type fossil for T. anglica (p.434), was the authors transferred to this species. It's one of the larger postcanines they had available, and that appears to indicate it's a molar. The length is 1.95mm and its width 1.3. The lingual and buccal sides of the crown run sociably parallel with one another, the front margin (+) is relatively straight and the rear one more has a more curved attitude. The highest tooth cusp is a1 in the lingual row, and that's typical for lowers. It's kept company by an unimpressive a3. (There's no mention of an a2). The b row has five cusps, and the rearmost is a bit more towards the central plane than its co-workers, and inhabits the ridge closing the back of the basin between both rows. A small cusp termed b (a lower letter is used for this in the paper, although other cusps were capitalised) is also part of the furnishings. As this tooth shows no signs of wear, its former owner dropped dead before deriving much benefit from growing the thing. However, at least the continuing sharpness of its cusps can still provide aesthetic pleasure for subsequent eucynodonts, so gratitude is merited. Besides, information on wear has been delivered by specimens from better fed critters. These also inform us they were double-rooted. If it's T. anglica, then why mention it under T. antiqua? I did say it had been transferred but, admittedly, saying something about why may be in order. The lower postcanines appear to have been evolutionarily conservative things. When a wider sample became available from Saint-Nicholas-de-Point, there was no obvious way of diagnostically keeping the species already proposed apart (p.447). Of course, that was half-a-century too late for Simpson, who established T. anglica in 1929. He could only refer to what was then available. T. antiqua was thought to differ on grounds of a stronger a3 cusp, and the presence of a cusp to the front called b. However, these were found to be variable factors in the wider sample, and that made them diagnostically worthless. The English 'species' was rendered a junior synonym in the process. Although it's conceivable (and even likely, given the chronological gap between the localities) the English haramis differed very obviously from their older relatives, should they have been stood in an identity line up, that doesn't show up in the known teeth. There's presently no need for separate taxa. Holotypes The holotype, a lower molariform, can be admired at the Staatliches Museum für Naturkunde in Stuttgart, Germany. This came from Degerloch, Baden-Württemberg and, presuming it has such a thing, I don't presently know its catalogue number. T. anglica was anchored on a tooth in the collection of the Bath Geological Museum, and it luxuriates under the name BATGM M219. This hailed from Holwell in Somerset, with the specific name presumably referring to England. It had been collected by Charles Moore. Mr Moore found yet mor... H. fissurae, discovered hiding in a fissure, was another of his children. It now lives at the Natural History Museum, London, and answers to the call of BMNH M2401A.
References:	von Plieninger (1847), Microlestes antiquus und Sargodon tomicus in der Grenzbrccia von Degerloch. Jahresh. Ver. Natk. Württemberg III 164-167, pl.I, figs. 3-10. (Sargodon is a semionotid fish).
	Poche F (1908), Einige notwendige Änderungen in der mammalogischen Nomenclatur, Zool. Annln, 2, p.269-272.
	Simpson (1928), A catalogue of the Mesozoic Mammalia in the Geological Department of the British Museum, London, British Museum.
	Sigogneau-Russell (1990), Reconnaissance formelle d'une novelle espèce d'Haramiya dans l'hypodigme francaise des Haramiyidae (Mammalia, Allotheria). Bull.Mus.Natl.His.Nat., Sect.C: Sci.Terre: Paléontol. Géol. Minéral (Sér.4) 12p 85-88.

Links: Syren CRL http://www.multimania.com/crcrl/articles/syrencrl.htm A detailed report on fossil finds from Luxembourg, including T. antiqua, (in French). Butler PM & MacIntyre GT, 1994 General Science Abstract Service The abstract of: Review of the British Haramiyidae (? Mammalia, Allotheria), their Molar Occlusion and Relationships, Philosophical Transactions: Biological Sciences, 345, p.433-458.

Species:	Thomasia moorei (Owen, 1871)
Aka:	Microlestes moorei Owen, 1871; Haramiya moorei
Place:	Holwell quarries & Saint-Nicolas-de-Point & Hallau
Locations:	England & France & Switzerland
Age:	Rhaetian-Liassic, Upper Triassic - Lower Jurassic
Remarks:	Upper teeth For editorial reasons, some discussion is included in the above entry for the genus. Rather than writing it all out again, let's be brief. The essential difference between the uppers of this species and those of T. antiqua is size. This is the larger version. It can be sensible assumed that this trait was reflected in other body areas, but only teeth are known. (I know you may have just read a remarkably similar paragraph for the other species a minute ago, but the editorial reasons are the same.) Hallau Butler & MacIntyre, 1994 includes (p.441): Clemens reported this species to be present in Switzerland. Unless subsequent finds have turned up, this refers to a single upper specimen of the right sort of stage. However, the tooth is somewhat eccentric in that it lacks a 'B' cusp. Type fossil The holotype comes from Holwell, Somerset. It is (or perhaps was) affectionately known as BATGM M214 and is (or was) part of the Charles Moore collection in the Bath Geological Museum. In any event, a lectoholotype, BATGM M211, has a similar pedigree. The specific name honours the collector.
Reference:	Owen, (1871), Monograph on the fossil Mammalia of the Mesozoic formations. Palaeontological Society Monograph, 24, p.1-115.

Saint-Nicolas-de-Port, France, Upper Triassic The first part of this article is based upon my reading of Godefroit & Battail, 1997. Saint-Nicolas-de-Pont is in northeastern France, (p.567). The fossils recovered present evidence of a diverse fauna, remains of which were deposited just off-shore. Fossil bones were found in the area as long ago as 1851, (p.568). A decade later, a stratum was reported as rich with fish and reptiles. More recent excavations commenced in a quarry in the 1970s, and several tons of sediment have been worked through by various researchers. Of particular interest were mammal and mammal-like teeth, but studies have also been undertaken on amphibians, reptiles and fish. Such a vivid insight into paleoecological conditions is rare. The exposed section of rock contains a number of bonebeds which, (p.569), become progressively thinner in ascending sequence. Most vertebrate fossils found since 1975 have come from the same bed, with a thickness varying between 20 and 100cm. Remains are generally restricted to teeth. The age of this location has been a matter of some dispute, (p.570), though there's at least a consensus for Norian-Rhaetian. Most the rest of the paper is devoted to descriptions and a useful summary of Upper Triassic and Lower Jurassic non-mammalian eucynodonts. As noted in the Conclusions, (p.627), the diversity of cynodonts at Saint-Nicolas is unprecedented. It encompasses nine genera of small non-mammals and seven of basal mammals. These are listed in the enclosure below. Most of the non-mammals were insectivores / carnivores but three were herbivores. The dental morphology of the latter is relatively basal for traversodonts, despite these being the most recently known representatives. The next part is based upon my reading of Godefroit & Cuny, 1997. The cynodonts had plenty of company. Many isolasted teeth of archosauriformes have been recovered, (p.2). There were phytosaurs, which filled a crocodile-like niche until their replacement by professional crocodiles. Eudimorphodon, an early pterosaur, was in the sky and Plateosaurus (Dinosauria) was attempting to keep the vegetation cropped. Three other herbivorous dinos were assisting with this, (early ornithischians), and 13 meateating archiosaurformes were doing their best to hinder their efforts. Most abundant of these were those phytosaurs, (p.6), despite the relative imminence of their extinction. The number of specimens amounts to several hundred. Two distinct types of phytosaur teeth have been identified. While this may indicate two separate species, it's also possibly the result of the presence of mature and juvenile representatives, (p.8). This study provides more than descriptions of teeth. Late Triassic ornithischian dinosaurs are discussed on page 21, and this is followed by considerations on the age of the assemblage. Pages 23-26 are dedicated to an overview of the vertebrate community: "a group of organisms living together within a definite locality", (p.23). This is about intermeshing food-chains with primary producers, end-consumers and middlers, (like most of us), and ecological niches. These animals were flesh-blood beings, and not fossils. "The mixture of terrestrial, marine and freshwater vertebrates within the Saint-Nicolas-de-Port community suggests a deltaic or a coastal palaeoenvironment." Dominant in the bonebed are the marine critters, but they're not diverse. There were lots of similar fish around, and that's what the pterosaurs ate, (though a differing dentition suggests the juveniles were perhaps insectivores). Several species of lungfish are also present. Stepping partly out of the water brings us to a number of amphibians; capitosauroids, metoposaurids and plagiosaurids: "The latter were undoubtedly suction-gulpers, living permanently on the bottom". The first mentioned were large and somewhat like crocodiles, reaching lengths of up to three metres, (p.24). Terrestrial vertebrates are uncommon, (none too surprising considering this is a marine sediment). Nevertheless, they're diverse. No plant remains from the land have been identified, but the herbivores must've eaten something. Small planteaters included the traversodonts and haramiyid mammals of this directory, (if they are mammals). The largest resident, Plateosaurus, reached about seven metres in length, (p.25). An array of insectivores implies a plentiful supply of insects. The mammalian and non-mammalian eucynodonts were complemented by small, adult and juvenile reptiles; eg. sphenodontians. In turn, this meant food was available for those carnivorous archosauriformes. Further Mesozoic site summaries can be found at Localities. An Upper Triassic Ramble. This is a 6,000 word romp concerning Saint-Nicolas-de-Port. It's written in an accessible style for general readers. Should you ever have wanted to go on a day-trip to the deep past, the tourist guide awaits. It's hosted by Mike K-H at France for Freebooters. Eucynodonts of Saint-Nicolas-de-Port (Genera 16, Species 21) Non-mammalian eucynodonts (Genera 9, Species 11) Carnivores and insectivores Gaumia longiradiata; "Hahnia" obliqua; Lepagia gaumensis; Meurthodon gallicus; Pseudotriconodon wildi; Tricuspes sigogneauae; T. tapeinodon; T. tuebingensis Herbivores (traversodonts) Maubeugia lotharingica; aff. Microscalenodon; Rosieria delsatei Mammals (Genera 7, Species 10) 'Haramiyida' (this directory) Theroteinus nikolai; Thomasia antiqua; T. moorei Morganucodontidae Morganucodon sp.; Brachyzostrodon coupatezi; B. maior Woutersiidae Woutersia mirabilis; W. butleri Kuehneotheria Delsatia rhupotopi; Kuehneotherium sp.

Species:	Thomasia hahni Butler P & MacIntyre GT, 1994
Place:	Halberstadt
Locations:	Germany
Age:	Norian late - Rhaetian early, Upper Triassic
Remarks:	The following is based upon my reading of Butler & MacIntyre, 1994. This mighty lower molar has a length of 1.6mm (p.449), and the crown is proportionately narrow. As it's been entirely stripped of enamel, it's possible some details may have disappeared, eg. small cusps. In any case, there's no sign of a b4 cusp, b2 is about 33% of the crown length away from the - end (posterior), and cusp a3 is relatively large. Holotype The paper doesn't directly say. At the time, it was the only specimen to have been found, so that's the one. I'd presume the specific name honours Gerhard Hahn, who provided the original description, but that's not specified either.
Reference:	Butler & MacIntyre (1994), Review of the British Haramiyidae (? Mammalia, Allotheria), their molar occlusion and relationships, Philosophical Transactions of the Royal Society B, 345, p.433-458.

Species:	Thomasia woutersi Butler P & MacIntyre GT, 1994
Aka:	Haramiya fissurae (in part)
Place:	Habay-la-Vielle & Attert
Locations:	Belgium
Age:	Rhaetian, Upper Triassic
Remarks:	The following is based upon my reading of Butler & MacIntyre, 1994. Originally (p.441), the then sole known specimen (an upper molar) was referred to H. fissurae. While being similar, however, it's about one-third the size and this is seen as a rather extreme difference. The length is 1.3mm (p.449), and there's no 'B5' cusp. The + end (rear) is oblique, and the 'B' row of cusps (lingual) curves in the manner favoured by H. moorei (see T. moorei). Holotype Refreshingly, I've got no idea. The authors helpfully refer me to the study by Wouters G, Sigogneau-Russell D & Lepage JC (1984), Découverte d'une dent d`Haramiyidé (Mammalia) dans des niveaux rhetiens de la Gaume (en Lorraine belge), Géologie 93, p.351-353, and do so on a number of occasions. It's doubtlessly an excellent idea. Unfortunately, the French is rather beyond me* and I haven't got a copy*. Unusually for a paper proposing a new species, the type fossil isn't specified beyond it being the one in the 1984 study, and there are no illustrations (see the 1984 study). It also doesn't comment on the reason for the specific name, but I'm prepared to guess at a connection with M. Wouters. * Of course, both those difficulties are of my own making.
Reference:	Butler & MacIntyre (1994), Review of the British Haramiyidae (?Mammalia, Allotheria), their molar occlusion and relationships, Philosophical Transactions of the Royal Society B, 345, p.433-458.

Other reports: Kayenta Formation, Arizona, USA Butler & MacIntyre (1994) make brief mention of a single tooth from this Lower Jurassic locality. They state is's "of aberrant morphology" (p.434). It apparently had a walk-on roll in a 1983 study by Jenkins, Crompton & Downs in 1983: Mesozoic mammals from Arizon: new evidence on mammalian occlusion, Science, 222, p.1233-1235. Talking of walking, I can now stride across to my own copy of Jenkins et al, 1983, and thanks are due to the supplier. They report (p.1233) that an Arizonan tooth may represent a haramiyid. It has two cusp rows separated by a valley, and a crest cuts them off at one end by linking the cusps. One of these rows has three members while the other numbers only two. The larger of those is positioned by the open end of the valley. A trio of roots supported the crown. The number of cusps is rather low, and the size is unusually small for a harami; 0.66mm in length as opposed to the usual range of between 1 to 3mm in Europe. Another possibility could be an abnormally worn tritylodontid tooth. Company from more specimens would be most welcome.

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 all 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, September 2001. Last update: 22.11.2010 Ktdykes@arcor.de

With further thanks due to: 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 Prof. Pascal Godefroit, for supplying the informative papers. Mr David Marjanovic, for his helpful corrections and suggestions. Shirley Sparks, for kindly supplying the paper by Dr Savage. Lots of people for kindly finding fossils and describing them. The harmiyids and forces of nature for supplying them.

Bibliography: Anantharaman S, Wilson GP, Das Sarma DC & Clemens WA (2006), A possible Late Cretaceous "Haramiyidan" from India, Journal of Vertebrate Palaeontology, 26(2), p.488-490. Butler PM & Hooker JJ (2005), New teeth of allotherian mammals from the English Bathonian, including the earliest multituberculates, Acta Palaeontologica Polonica, 50(2), p.185-207. Butler PM & MacIntyre GT (1994), Review of the British Haramiyidae (? Mammalia, Allotheria), their molar occlusion and relationships, Philosophical Transactions of the Royal Society B, 345, p.433-458. Cifelli RL (2001), Early Mammal Radiations, Journal of Paleontology, vol 75 (6), p.1214-1226. Cox B, Dixon D, Gardiner B & Savage RJG (1989), Dinosaurier und andere Tiere der Vorzeit, Mosaik Verlag (Sonderausgabe für Gondrom Verlag, 1994), ISBN 3 8112 1138 2 Godefroit P (1997), Reptilian, therapsid and mammalian teeth from the Upper Triassic of Varangéville (northeastern France). Bulletin de L’Institut des Sciences Naturelles de Belgique: Sciences de la Terre 67, p83-102. Godefroit P (1999), New traversodontid (Therapsida: Cynodontia) teeth from the Upper Triassic of Habay-la-Vielle (southern Belgium). Paläontologische Zeitschrift 73 (3/4), p.385-394, 6 Abb., 1 Tab., Stuttgart Godefroit P & Battail B (1997), Late Triassic cynodonts from Saint-Nicolas-de- port (north-eastern France). Geodiversitas, 19(3), p.567-631. Hahn G Hahn R (2006), Evolutionary tendencies and systematic arrangement in the Haramiyida (Mammalia), Geologica et Palaeontologica, 40, p.173-193. Hahn G, Lepage J-C & Wouters G (1987), Ein Multituberculaten-Zahn aus der Ober-Trias von Gaume (S-Belgien), Bulletin de la Societe belge de Geologie, 96, p.39-47. Heinrich W-D (1998), Late Jurassic Mammals from Tendaguru, Tanzania, East Africa. Journal of Mammalian Evolution, Vol5 (4), p.269-290. Jenkins FA, Crompton AW & Downs WR (1983), Mesozoic mammals from Arizona: New evidence of mammalian evolution. Science 222, p.1233-1235. 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. Luo Z-X, Kielan-Jaworowska Z & Cifelli RL (2002), In quest for a phylogeny of Mesozoic mammals. APP 47 (1), p.1-78. Maisch MW, Matzke TM, Grossmann F, Stöhr H, Pfretzschner H-U & Sun G (2005), The first haramiyoid mammal from Asia, Naturwissenschaften, 92(1), p.40-44. Martin T, Averianov AO & Pfretzschner H-U (2010), Mammals from the Late Jurassic Qigu Formation in the southern Junggar Basin, Xinjiang, northwest China, Palaeobio Palaeonv, 90, p.295-319. McKenna MC & Bell SK (1997), Classification of Mammals Above the Species Level. Columbia University Press. Monastersky R (1996), The lost tribe of the mammals, Science News, 150 (24), p.378-379. Savage RJG (1989), British mammals of the Mesozoic Era, Biological Journal of the Linnean Society, 38, p.3-7. Storrs GW (1994), Fossil vertebrate faunas of the British Rhaetian (latest Triassic) in Vertebrate Palaeobiology (eds.) Benton MJ & Norman DB., Zoological Journal of the Linnean Society, 112, p.217-259. Wible JR & Rougier GR (2000), Cranal anatomy of Kryptobaatar dashzevegi Mammalia, Multituberculata), and its bearing on the evolution of mammalian characters. Bulletin of the American Museum of Natural History, 247, pp. 124.