A. MIDDLE JURASSIC MULTITUBERCULATES

Taxa: Hahnotheriidae Butler PM & Hooker JJ, 2005 Kermackodontidae Butler PM & Hooker JJ, 2005 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. Multituberculates have a number of key contrasts to 'haramiyidans', and it's generally thought that some of these preclude close affinities. However, Butler and Hooker 2005 aren't prepared to rule out a special relationship despite contrasts they raise on page 194. Multis have a horizontal chewing action which produces characteristic wear. This reflects differences in how the jaws worked. A multi speciality is that the upper molar, M2, is positioned lingually relative to the corresponding lower m2. Those teeth occluded differently. Furthermore, lower multi premolars have blade-like specialisations. The equivalents in harami mouths don't. The first enclosure of this directory presently provides a home for the earliest known multis, and they're represented by a few isolated teeth from the upper Middle Jurassic of southern England. Assuming the owners were herbivores, they would have been part of a surprisingly diverse assemblage of small, eucynodont botanists. The 'haramiyidans' seem to have been the most common but, as the size of the available sample isn't large, biases may be in play. Whether haramis were mammals is unclear. Also about were tritylodontids, and they don't qualify as mammals in the present opinion of anybody. Multis are mammals and, up until 2005, they were all divided into two suborders. The paraphyletic 'Plagiaulacida' contained the most basal forms. More advanced multis form the monophyletic Cimolodonta. The placement of the two established genera from the Middle Jurassic in this broad scheme is clear. They don't fit. If plagis can be termed primitive, then pre-primitive seems appropriate. Genera: Eleutherodon (partly = Hahnotherium), Hahnotherium, Kermackodon, other reports Time-Line: Middle Jurassic: Hahnotherium, Kermackodon

Genus: Hahnotherium Butler PM & Hooker JJ, 2005 ‘Hahn’s beast' Aka: Eleutherodon (partly) Family: Hahnotheriidae Butler & Hooker, 2005 Remarks: 'Hahn's beast' is named in honour of Professor Gerhard Hahn in recognition of his work on early multis.

Species:	Hahnotherium antiquum Butler PM & Hooker JJ, 2005
Aka:	Eleutherodon oxfordensis (partly)
Place:	Forest Marble, Oxfordshire
Country:	England
Age:	upper Bathonian, Middle Jurassic
Remarks:	The following is based upon my reading of Butler & Hooker, 2005. The specimens described are an upper molar, a lower molar and an upper premolar, (p.200). The first two were originally misinterpreted as a lower and upper respectively, and referred to Eleutherodon. The premolar has since been lost, although photos and sketches are known. Upper molar This tooth is an M2 and the type fossil of the genus. The wear groove between the buccal and lingual rows of cusps isn't basined. Horizontal longitudinal grooves are characteristic of multitubercualtes and not 'haramiyids'. It has similarities with the equivalent teeth of paulchoffatiid multis. The crown is broadly oval seen from the occlusal perspective but, (as is usual for paulchoffies), the front edge looks as if it's been cut diagonally, as the lingual side is longer. The length is 1.9mm and the width 1.55. The proportions are consistent with paulchoffie M2s. This identification is strengthened by: no wear on the lingual sides of the lingual cusps; wear on the buccal faces of the buccal cusps; a contact wear facet on the front but not the back of the crown. Cusps The buccal side has a row of five cusps separated from one another by grooves. In contrast to Kermackodon no longitudinal crest is present. The front two cusps are tallest. The third, B3, is smallest and situated comparatively slightly lingually. At the rear of the crown, B5 is even further towards the middle. Excepting for Meketichoffatia, paulchoffies have either no cusps on this side beyond B2, or much reduced ones. There are seven cusps on a ridge on the lingual side of the tooth, and use has worn them down. The largest are in the middle of the file, (L3~L5). The remainder become sequentially smaller in both directions. As with the buccal row, grooves separate the cusps, but the longitudinal crest nevertheless crosses them. At the front of the tooth, an oblique crest connects L1 and B1. At the back, as no cusp is located between L7 and B5, the central valley is open, (in contrast to Kermackodon). Ridges of enamel run down into the valley from the cusps on either side, and the effect looks similar to the fluting of Eleutherodon. Down below Wear on the upper molar has been caused by the partner downstairs. The concentrations of damage are consistent with this tooth having been situated lingually of the m2. That's characteristic of multis. Roots Two are present. One's at the front and the other behind. However, a large part of the base is missing from the buccal side, and that probably housed a third root. Lower molar The right m2 has a lingual row of three clear cusps, while the buccal contains a ridge with small elevations along it. Length and width both reach 1.6mm. The lingual cusps grow progressively smaller along the line from front to back, and their lingual faces have been worn flat. The buccal side of the crown has also been subject to erosion. Remains of four cusps are visible running from the front, and more may have once been situated behind. The referral of this specimen to the species is provisional. Holotype The holotype is known to its friends as BMNH M46797. It's a left upper molar (M2) in the collection of the Natural History Museum, London. The specific name is Latin for ancient, and this is a reference to the relatively early age. I wonder what the Latin for 'even more ancient' is. It might become relevant should any ancestors be found.
Reference:	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.

Genus: Kermackodon Butler PM & Hooker JJ, 2005 ‘Kermack's tooth' Family: Kermackodontidae Butler & Hooker, 2005 Remarks: The generic name is in memory of Professor Kenneth A Kermack, and honours his work on Mesozoic mammals.

Species:	Kermackodon multicuspis Butler PM & Hooker JJ, 2005
Place:	Forest Marble, Oxfordshire
Country:	England
Age:	upper Bathonian, Middle Jurassic
Remarks:	The following is based upon my reading of Butler & Hooker, 2005. The one known molar is an upper one, M2. It's heart shaped from the occlusal perspective, and measures 2.85mm long and 2.6 wide, (p.195). The back is pointed, (probably exaggeratedly so due to breakage). Nevertheless, this characteristic is shared with most paulchoffatiids but not other 'plagiaulacidans'. A valley runs along the top of the crown between two somewhat curved rows of cusps, which converge towards each other at the back. Each row has pointed cusps joined by a longitudinal crest. Cusp rows The buccal row has five cusps, and the rear one (B5) is the tallest. In front of that is B4. As it's either damaged or worn, its original height is uncertain. The other three are further forwards still and smaller, being placed on a ridge running from the B4. Seven cusps form the lingual row. Highest is the sixth (L6), but it's not as tall as B5. The preceding five decrease in size progressively towards the front. L7 is smaller than L6 and found on a prominence near to the midline of the crown, where it blocks off much of the central valley. This is another feature shared with paulchoffies but not other plagis. At the front of the central valley are found an irregular congregation of small cuspules on both sides. Three small cingulum cusps occur on the lingual margin of the crown below L3-L5. There may have been more as the margin is broken behind them. Not a plagi 'Plagiaulacidan' M2s have blunter cusps than this tooth, and they're separated by grooves running across the crown. With 'Kermack's tooth' the cusps are more numerous and sharper. They're connected by crests running along the crown. The construction follows a different general plan. Wear and tear A groove caused by wear runs along the central valley. It's horizontal; not basined as in 'haramiyidans'. It's also narrow suggesting the excavation was performed by a sharp cusp on the lower molar. Wear is also evident on the buccal ridges of some of the lingual cusps, but not on their internal faces. The corresponding lower teeth was somewhat offset; a feature of multis. (Further clues suggest the lower tooth had a row of high cusps on its lingual side, and probably a poorly developed buccal row. Size and relations 2.85mm might sound rather small when compared to Mesozoic dinosaurs, but it's modestly ginormous in terms of Jurassic and Lower Cretaceous plagi multituberculates. It would have been a healthy size for the 'haramiyidan' Eleutherodon. However, the lack of a basined wear groove and no third row of cusps, and the presence of various multi characteristics prohibit harami affinities. It has more similarities with paulchoffies, although it's too cuspy among other qualities. One multi molar is known that shares the pointiness of these cusps. It's an M1 of Proalbionbaatar, (p.197). That tooth also has an unusual number of cusps, (six or seven in the lingual row), and they're connected by a longitudinal crest. A direct comparison with an albionbaatarid M2 would require the discovery of such a tooth, and none are yet known. Lower premolar The specimen gives a convincing performance of being the most primitive multi p4 yet seen. It differs to most later versions, (eg. its triangular when viewed from the side), but this is the sort of contrast to be expected. The length and width are 3.5 and 1.5mm, similar to Plagiaulax becklesii. It's narrower at the front than the back, and the dominant feature is the long crest shaped roughly like the roof of a house. That's what produces the triangular profile. The highest point is reached just in front of the mid point. This crest is a blade with a maximum height of 2.2mm. Three serrations are located near the highest point. Low ridges on the buccal surface are associated with the first two, and they reach about halfway down the wall. The third serration has a ridge running straight down to cingulum level. Getting to the back The rear slope is worn, but a fourth serration is indicated about half the way down, and there's a corresponding ridge on the lingual side. This kind of ribbing is known from later multis, which generally favour lots more serrations. However, the paulchoffies had only four serrations, although they are more evenly spaced than for Kermackodon. Until this description, the paulchoffies were about the most basal of known multis. An absent groove and the roots Paulchoffies also have a groove on the front of their p4s, and this provides some support for the back of the p3. I would imagine, (but don't know), this had to do with additional stability of the tooth row. Whether I'm correct or not, no such groove is on this more ancient tooth, (p.198). The Kermackodon p4 has two impressive roots. These are curved to some extent, which suggests the tooth may have sat in the jaw at a somewhat jaunty tilt. Holotype The holotype, BMNH M46822, is a left upper molar in the collection of the Natural History Museum, London. The specific name is Latin and means 'many sharp points'. A lower left premolar (p4) and an a right upper have also been referred.
Reference:	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.

Other reports: England "Reported but not yet described from the Middle Jurassic (Bathonian) of Europe (England)." McKenna & Bell, (1997), p.36. If accurate, this would be the earliest known uncontested material. Those details are consistent with a subsequent publication, Kermack KA, Kermack DM, Lees PM & Mills JRE (1998), New multituberculate-like teeth from the Middle Jurassic of England. Acta Palaeontologica Polonica 43 (4), p.581-606. This is Eleutherodon oxfordensis, which I presently list as a member of 'Haramayida'. However, (Kielan-Jaworowska & Hurum, 2001, p.411): "Until recently, the oldest uncontested multituberculates were from the Kimmeridgian of Portugal. Freeman (1976) described from the Middle Jurassic Forest Marble of Dorset, England, a strongly worn, three-rooted tooth, and from the Middle Jurassic of Oxfordshire (1979) an incisor, which might be records of multituberculates. PM Butler (pers. comm. 1999) informed us that in the collection from the Bathonian Forest Marble at Kirtlington, Oxfordshire, housed in The Natural History Museum, London, there are isolated multituberculate upper and lower premolars..." Some of this material was described in 2005 (see above).

A. Middle Jurassic Multitubercuates B. ‘Basal’ Multituberculata C. Paulchoffatiidae & Hahnodontidae D. Pinheirodontidae

B. ‘BASAL’ MULTITUBERCULATA

Taxon: None. A collection of mostly some of the least derived genera plus others, whose exact position in the scheme of Multidom is not yet clear. They’re 'plagiaulacidans' of one kind or another, with the possible exception of Ameribaatar. However, the taxon of Plagiaulacida has fallen into some disrepute, though it's still a reasonably convenient label for the more basal multituberculates. Careful Some of the genera in this section probably aren't "basal" multis in any meaningful way. They're simply included here because I can't presently think of anywhere in particular to house them. Basal multis "It is difficult to decide whether the allodontid or paulchoffatiid line is more plesiomorphic; both lines are characterised by different mixtures of plesiomorphic and derived characters. We tentatively accept that members of the allodontid line might be the most plesiomorphic multituberculates, as they retain the plesiomorphic structure of the lower molars with two rows of well-separated cusps, smooth enamel (lack of grooves and ribbing on the upper and lower molars) and a small I3... They all retain fiver upper and four lower premolars," (Kielan-Jaworowska & Hurum, 2001). I3 is an upper incisor. Glirodon, Allodontidae and Zofiabaataridae are members of the allodontid line. Genera: Allodon (= Ctenacodon, Psalodon), Ameribaatar, Corriebaatar, Ctenacodon, Glirodon, Janumys, Morrisonodon Psalodon, Zofiabaatar, other reports Time-Line: Upper Cretaceous: Ameribaatar, Janumys Lower Cretaceous: Corriebaatar Upper Jurassic: Ctenacodon, Glirodon, Morrisonodon, Psalodon, Zofiabaatar

Genus: Ameribaatar Eaton JG & Cifelli RL, 2001 'American hero'

Species:	Ameribaatar zofiae Eaton JG & Cifelli RL, 2001
Place:	Cedar Mountain Formation, Utah
Country:	USA
Age:	Albian (late) - Cenomanian (early), Upper Cretaceous
Remarks:	This genus "is of uncertain subordinal affinities," (Eaton & Cifelli, 2001). It’s got to go somewhere though, so why not here? The species name honours Zofia Kielan-Jaworowska.
Reference:	Eaton & Cifelli (2001), Multituberculate mammals from near the Early-Late Cretaceous boundary, Cedar Mountain Formation, Utah. Acta Palaeontologica Palonica 46(4), p.453-518.

Link: Acta Palaentologica Polonica 46(4) http://www.paleo.pan.pl/acta/acta46-4.htm#Eaton The abstract, featuring a number of the then residents of Utah. Also involves Janumys (below).

Genus: Corriebaatar Rich TH, Vickers-Rich P, Flannery TH, Kear BP, Cantrill DJ, Komarower P, Kool L, Pickering D, Trusler P, Morton S, Klaveren N van & Fitzgerald EMG, 2009 'Corrie's hero' Remarks: Should Australia feel the need of a new cricket team, then the authors of this paper could be worth contacting. There are no problems as far as the numbers go, and it wouldn't even be necessary to nominate a substitute fielder. As for the necessary commitment and talent, even putting such questions would be superfluous. Admittedly, several might be more familiar with baseball bats, but that could provide the crucial element of surprise. The Corrie to whom this hero is dedicated is Corrie Williams in recognition of: "... her discovery of a Gondwana multituberculate." Which one isn't explicitely revealed.

Species:	Corriebaatar marywaltersae Rich et Al, 2009
Place:	Flat Rocks, Wonthaggi Formation, Victoria
Country:	Australia
Age:	Aptian, Lower Cretaceous
Remarks:	The following is based upon my reading of Rich et al, 2009. The Lower Cretaceous locality of Flat Rocks, a couple of hundred km from Melbourne, has deemed it fit to dish up what appears to be a p4 lower premolar of a multi mammal (p.1). This 3.5mm small tooth is a big addition to the small glimpses of Gondwanan multituberculates presently on offer. Tentatively, it's been questionably referred to Cimolodonta. I've deliberately written both 'tentatively' and 'questionably' into the same sentence, as that strikes me as being an apt summary of the level of certainly that's justified in this case. Gondwanan multis While multis are widespread in the Jurassic-Paleocene terrestrial fossil faunas of the northern hemisphere, and still register a presence into the Eocene, they're much scarcer in the south. However, the numbers are increasing, and their low count belongs in a context of all Gondwanan mammals being poorly represented between the Middle Jurassic until the Upper Cretaceous. Disputed multis (or haramiyidans) frequent the Lower Cretaceous of Morocco. Given some overlaps with European faunas, eg. Gobiconodon, a degree of faunal exchange with Laurasia was clearly possible in that case, and some authors see those Moroccan hahnodontids as being relatives of Euro- paulchoffatiids. The other possible Gondwanan multis stand in an even dimmer light. A fragment of a tooth in the Upper Cretaceous of Madagascar, and a riddle awaiting resolution in Patagonia. A p4 premolar from there seems to be multi and, to some, it goes by the name of Argentodites. According to others, it's more likely a ferugliotheriid gondwanatherian, and implies that such mammals may also have been multis or, at least, that the feruglios were. The Australian newling, Corriebaatar, spreads no obvious light on that matter. It seems to be a fragment of a multi lineage song of some sort, but either the bizarre accent or unknown language leaves its meaning awaiting explanation. For now, it's best called an aberration. When it came to a classification within Multituberculata as presently arranged, there were effectively four options available; no comment, 'Plagiaulacida', Cimolodonta or something else that defies any worthwhile definition. The first of those would be safe but boring, and the fourth unhelpful although seemingly quite likely. As 'Plagiaulacida' could be ruled out under the current diagnosis, the result was a tentative assignment to Cimolodonta "... as a working hypothesis" (quote from page 3). The public goes wild! In anticipation of this publication, much of the human population of Australia took to the streets in joyous and spontaneous celebration. That was on the evening of December 31st and, admittedly, there could've been some minor connection with New Year's Eve. I prefer to interpret it as a slightly premature party in honour and acknowledgement of Corriebaatar. So far only a single specimen has been found; a vast 5mm long slab of fossil jaw bone containing a premolar and the front root of an m1 molar. The premolar is both well preserved and peculiar. One oddity involves a word termed exodaenodont, a term that, until now, puzzled the hell out of me as well. It refers to an enamel bulge that covers part of the front root on its buccal side, and extends on down. This exodaenodont condition is present for Corrie, but with a touch not previously reported for any multi. There's also a similar lobe of enamel on the buccal surface of the rear root, a state of affairs the authors call "double exodaenodont" (p. 2, and the authors use quotation marks). A look at 'Corrie's hero' Several non-multis have superficially multi-like p4s (of a style known as plagiaulacoid, which isn't the same word as plagiaulacidan!), but they don't possess the exodaenodont characteristic just mentioned (p.2). With one possible exception, namely Pinheirodon pygmaeus, "plagiaulacidan" multis have a row of cusps on the buccal side of the tooth or, at least, remnants of an ancestral row (pits or a single cusp). Corriebaatar shows no such condition and, as far as P. pygmaeus goes, the lack of a cusp on the only known specimen could've resulted from abrasion. There are at least two currently unique features for 'Corrie's hero' debarring it from any established family; the "double exodaenodont" business and a hollow in the rear root on its lingual side. Such excavations are found on both sides of each root, and these presumably had some function for the preceding (should there have been a p3) and the following tooth. However, the rear lingual hollow hasn't been reported before. A further peculiarity, in comparison to other cimolodontans, concerns the ribbing running down the sides of the tooth from the serrations of the blade. There are only two such ridges on the buccal side and a lonely lingual singleton. Cimolods generally had a fondness for such ribbing although, in the case of Cimexomys gracilis, the numbers are low; three or four (buccal), two or three (lingual). Corrie's baatar is an even less enthusiastic follower of that particular fashion. Another touch not typical for cimolods is the relatively flat roof of the tooth. Usually, in side view, there's a serrated arch of a blade; a shape designated arcuate or, in less pronounced forms, triangular. The profile in this instance is more typical for plagis; rectangular. As the authors point out, it's not entirely unknown for cimolods as Microcosmodon conus is no more arcuate than Connie. The nature of the beast Personally, the p4 looks to me as if at least three options are open; a plagi descendant with some parallel enhancements in common with cimolods; a peculiar cimolod perhaps retaining traits from plagi ancestors; something else, eg. a member of a "pre"-plagi lineage we know sod all about. The vast tracts of unknown white space on the hardly known map of Gondwanan Multidom are like a large dot-to-dot picture with virtually all the dots missing; like a massive white advertizing hoarding with "watch this space" written somewhere in the middle in microscopic text. It doesn't even manage an obvious resemblance to the South American p4 known as Argentodites (and/or a ferugliotheriid). Corrie is hypothetically referred to Cimolodonta because it doesn't seem to be a plagi (p.3). Presumably, as this is more a provisional bookeeping arrangement than a formal declaration of affinities, enquiries concerning its cimolod credentials were less than passport control rigorous. A frontier Comparing the relatively thin Flat Rocks mammal inventory (four genera) with the later and larger Los Alamitos community of Patagonia is clearly problematic. Further discoveries could change the possible conclusions. Presently, the only apparent ordinal overlap could be multis. Flat Rocks otherwise provides ausktribosphenidans ( Ausktribosphenos and Bishops) and a monotreme (Teinolophos). Nothing of the sort has been discovered in the Patagonian Upper Cretaceous. There may be a further Australian overlap from Lightning Ridge in New South Wales, but the emphasis is strongly upon that 'may'. A possible dryolestoid was reported in 2003: an extremely tentative possible dryolestoid. Western Antarctica was geographically closer to Patagonia than Australia, but its Lower Cretaceous plant fossils nevertheless share more similarities with Australian floras (p.4). Matches with South America are rare. This applies both for larger plant remains, pollen and spores. That suggests a large scale interchange with South America was, at that time, not possible. This may at least partly be a result of the climatic conditions (p.5). Both geographic Antarctica and at least much of Aus were within the southern polar circle. Holotype The type fossil, NMV P216655, resides in the paleontological collection of Melbourne's Museum Victoria. The specific name is for Mary Walters who, in 2004, found the fossil. Whether that means she found it whilst Dr Williams discovered it, or the good doctor discovered a different Gondwanan multi, is unknown to me.
Reference:	Rich et al (2009), An Australian multituberculate and its paleobiogeographic implications, Acta Polonica Palaeontologica, 54(1), p.1-6.

Link: Acta Palaeontologlca Polonica, 54(1) http://www.app.pan.pl/archive/published/app54/app54-001.pdf Rich et al, 2009 is presently freely accessible on-line in pdf format.

Genus: Ctenacodon Marsh OC, 1879 'comb tooth' Aka: Allodon Marsh, 1881 Family: Allodontidae Marsh, 1889

Link: Ctenacodon http://www.angellis.net/Web/DFG-mam/Ctenacodon.htm A sketch from the gallery of VRW.

Reassigned species: C. brentbaatar see 'Ctenacodon' brentbaatar; C. portens Marsh, 1887 see Psalodon portens

Species:	Ctenacodon falconeri (Owen R), 1871
Place:	Durlston Bay
Country:	England
Age:	Lower Cretaceous
Remarks:	It's listed in the table in Savage, 1989, (p.5). However, the correctness of the identification has been called into doubt, (Engelman & Callison 1998, p.578).
Reference:	Owen, (1871), Monograph on the fossil Mammalia of the Mesozoic formations. Palaeontological Society Monograph, 24, p.1-115.

Species:	Ctenacodon laticeps (Marsh OC, 1881) Simpson GG, 1927 or 1929
Aka:	Allodon laticeps Marsh OC, 1881
Place:	Morrison Formation, Wyoming and Utah
Country:	USA
Age:	Upper Jurassic
Remarks:	The following is largely based upon my reading of Engelmann, 2004. The 1984 discovery of a microvertebrate site at the Dinosaur National Monument, Utah, has resulted in the recovery of mammal fossils, (p.54). Both hand-quarrying and screen-washing methods have been used. The first yielded a few more complete specimens, while the second technique produced mainly isolated teeth. The best represented mammalian group in the fauna is Multituberculata. This study was primarily concerned with a damaged and partial upper palate, (which is the top of the mouth). As the morphology conforms to the holotype, it's been referred to C. laticeps, (which was established for upper jaw fossils). Excavation caused some damage and the loss of teeth, (p.53). However, relevant information was preserved in the surrounding matrix. The palate broke into two but parts of both dental rows are present. Of jaws and teeth The left maxilla contains several premolars in situ, (P3 and P5), with P5 being the last tooth position preserved. Aveoli attest to the original existence of three more double-rooted premolars. In front of P1 is a circular hole for a single-rooted canine. This was rather surprising as it had been thought Ctenacodon lacked an upper canine, which is the case for further derived multis. The right portion contains parts of both premaxilla and maxilla. The former has the I3 and an alveolus for a single-rooted I2. Two premolars (P1 and P3) are relatively well preserved. Three others and a small canine were present. What's available of the premolars conforms to this species. P1-P3 are tri-cusped and something like triangular in occlusal perspective. P4 and P5 are longer and four-cusped. The canine seems to have been small and probably uni-cusped. Both the I3 and frontmost premolar were larger than it, (p.57). Screenwashing turned up several isolated teeth which seem to be I3s of this species. Two of these are about 40% smaller, which suggests the possibility they're milk teeth, (p.59). The number of upper incisors isn't known, but its relative, Psalodon, had three per side. The morphology of the I3 in Ctenacodon is more complex than in either that genus or in Glirodon, (simple and nearly peg-like). The lack of a mesial cusp is a feature shared with paulchoffatiids and the shape of a wear facet resembles that on the holotype of Henkelodon naias. Such factors could be clues for close affinities or shared plesiomorphies. Replacement patterns In a study by Szalay in 1965 (which I now have), a Ctenacodon specimen was described which had both the permanent and deciduous P2 still in place. The former had three cusps and its precursor five, (p.60). In existing mammals deciduous premolars are generally more complicated or molariform than their successors. As this trend's so widespread, (triconodonts, dryolestids, therians and apparently multis), this may well be plesiomorphic for mammals. (This wouldn't be terribly surprising, seeing as some premolars in juveniles have to work as molars, until those teeth are sufficiently developed.) Holotype The holotype, collected by Reed WH in 1880, is in the Peabody Museum, Yale. This species may be the same as C. serratus, (Martin & Foster 1998, p.385).
Reference:	Marsh (1881), New Jurassic mammals. Am. J. Sci. (3) xxi: p.511-513.

Link: The Peabody VP On-line Catalogue http://george.peabody.yale.edu/vp/ Meet the Ctenacodon specimens of Yale.

Species:	Ctenacodon nanus Marsh OC, 1881
Place:	Morrison Formation, Wyoming
Country:	USA
Age:	Upper Jurassic
Remarks:	This type fossil is likewise at Yale. Reportedly, Simpson synonymized this species with C. serratus in 1929, (Martin & Foster 1998, p.385).
Reference:

Species:	Ctenacodon scindens Simpson GG, 1928 or 1929
Place:	Morrison Formation, Wyoming
Country:	USA
Age:	Upper Jurassic
Remarks:	Also at Yale and originally assigned to C. serratus. This species is based upon lower teeth. Carpenter 1998 (p.398) contains the following lengths for postcanine teeth: premolars: p2 1.1mm, p3 2.0mm, p4 1.5mm.
Reference:

Link: American Museum of Natural History http://sabertooth.amnh.org/fm/3004-01 Ctenacodon scindens has also been sighted in the collection of the AMNH, New York.

Species:	Ctenacodon serratus Marsh OC, 1879
Aka:?	C. nanus
Place:	Morrison Formation, Wyoming
Country:	USA
Age:	Upper Jurassic
Remarks:	This is also a lower dentition species, and specimens are at Yale. The holotype is called YPM 11833. It may represent the same species as C. laticeps. Carpenter 1998 (p.398) contains the following lengths for postcanine teeth: premolars: p2 0.8-1.0mm, p3 1.1-1.2mm, p4 1.6-1.7mm; molars: m1 1.1-1.3mm.
Reference:	Marsh (1879), Notice of new Jurassic mammals. Amer. J. of Sci., 3pp., xviii, p.296-398.

Genus: Glirodon Engelmann GF & Callison G, 1999 'rodent tooth' Remarks: The name's derived from Greek. As rodents have incisors with similarly restricted bands of enamel, Glirodon is rodent-like in that respect.

Species:	Glirodon grandis Engelmann GF & Callison G, 1999
Place:	Dinosaur National Monument, Utah & Fruita, Colorado
Country:	USA
Age:	Upper Jurassic
Remarks:	The following is based upon my reading of Engelmann & Callison, 1999. The genus is based upon partial skulls and other remains found at two localities of the Morrison Formation; Dinosaur National Monument, Utah and the Fruita Paleontological Area, Colorado (p.161). The front of the skull is best represented including all teeth, and many characters are relatively basal for multituberculate mammals. This relative primitiveness includes the dental formula per side: (uppers): 3 incisors, 1 canine, 5 premolars and 2 molars; (lowers): 1, 0, 4 and 2 respectively. Most multis had no canines, and the number of lower premolars became reduced to a single functional tooth generally accompanied by a pathetic, vestigial partner to the front. (That disappeared completely for many late multi models.) In terms of construction, the postcanines mostly resemble those of its contemporary, Ctenacodon. However, there are several surprising derivations. Enamel on the second upper and solitary lower incisors is restricted to a band, rather than covering all surfaces, and second upper molars have a third row of cusps positioned on the buccal side of the crown. Such characteristics are known from some later multis, but their presence here could sensibly be termed unexpected. To dream up a playful analogy, I suppose it could be a bit like finding Beethoven had included a score for an electric guitar in his Ode to Joy. Tchaikovsky's 1812 Symphony perhaps, but Ludwig von making provision for Jimmy Hendrix? Actually, I can hear it playing now in my head, and it's good enough to cause the death and dead composer to roll over in his grave. As the monophyletic cimolodontans don't possess such an archaic crowd of teeth, that's a conveniently comprehensible reason as to why Glirodon can't be referred to as even an early member of that orchestra. It was told to find a place within a loosely defined Plagiaulacoidea ( I'd prefer 'Plagiaulacida'), but which place remains unclear. For one thing, no other plagis with restricted bands of enamel on incisors are known. Bits and pieces A total of six specimens provided information for this study, and they displayed an impressive understanding of team work. The captain from Utah opted to be the type fossil by proffering much of the front of its head in a fairly well preserved condition, along with a decidedly trashy scrap of mandible. A partial skull from Colorado is less handsome, but it does provide confirmation on a number of points, and adds some details not shown by its friend. Despite the poor nature of the Utah mandible, it nevertheless made a solid contribution by confirming the identities of a trio of jaw remnants from Fruita, and that locality donated a further upper jaw. Skull and size The features of the skull are described in loving and extensive detail; more detail than my brain can cope with. A view from above, the perspective from the side and discussion on the palate... I'm not going to attempt summarising it. Apart from anything else, it'd be daft as it's been done expertly in the paper, and that's always the best source to drink from, if possible. The snout is comparatively short (p.162) and, as it becomes strongly wider along its course towards the cheeks, it has something of a triangular shape. Measurements aren't stated in the study, so I've done a couple of rough estimates based upon the scale bars accompanying the graphics. For example, the preserved length of the partial skull from Fruita is something like 14mm, and around eleven of those run from the tip of the snout to the rear of the tooth row. Although the complete skull would've been longer, the specific name of grandis should be viewed with an appropriate sense of proportion. In terms of lower postcanine length, Glirodon actually falls about midway between two other Morrison multis, Ctenacodon and the larger Psalodon. There are some mice which would make this 'giant' look a bit on the small side. Upper teeth to the front of them... As far as is known, and this still applies in 2007, the tooth numbers in the upper jaw reflect the original multi tradition; three incisors, a canine, five premolars and two molars. For a while, in the 1960s-70s, some research suggested the possible presence of six premolars, and that number can be found in corners of the literature. It involved a tentative misinterpretation of the upper canine worn by a paulchoffatiid, an odd one with several cusps. Although present in spirit on parade (p.167), the first incisor is a pathetically sized thing known only from roots. As the I2s of both sides are reasonably large and close together, there wasn't much space between them for accommodating anything impressive. I2 actually manages to be the largest member of the upper teeth. It tip is well worn and has enamel restricted to a band at the front. That's the rodent resemblance which inspired the name. Despite its root being large (p.168), it doesn't go far back into the jaw and, in contrast to the rodent counterpart, it doesn't appear to be open-rooted. Open-rootedness is a trick used by real rodents and some other mammals, and it enables teeth to be ever-growing through the owner's lifetime. This is particularly useful if loads of gnawing wears the material at the tip away, as the loss gets compensated for by reinforcements. Glirodon received no such renewal services. While considerably smaller, I3 was a tooth with a job of work to do. It's simply constructed but shows plenty of wear, although breakage may have exaggerated the picture (p.169). The canine in the set is a small cone with a single cusp. That contrasts with the more elaborate models known from some nigh contemporary paulchoffatiid multis from Europe. The cusp shows a bit of wear, and it's supported by a single root. More typically, this tooth wasn't part of the multi approach to living. As has already occurred for the lower jaw of Glirodon, the canine was dispensed with. There is actually a measure of individual variation with canines. A specimen from Fruita has a more daring design featuring two closely positioned cusps. Regularly sized diastemata provide free space between I2, I3, C and P1 whereas, further back, the remaining postcanines are crowded together. Upper teeth to the rear... The first trio of the premolars are much the same, although some eccentricity occurred during the career of the holotype. Ignoring that for the moment, these crowns have a larger buccal cusp and two smaller lingual ones. The buccal cusp is also taller. The two lingual ones are arranged in a straightish line, with the rear members being a shade larger than their partners; larger in terms of width rather than height. Crests connect the pair. Further crests run from both to connect them with the buccal cusp (p.170). These teeth are double-rooted, have a degree of wear on the cusps, and have somewhat asymmetrical crowns. The eccentric left P2 of the type fossil manages only one lingual cusp, seemingly the rear one. Its intended partner failed to develop. The structure of P4 points to a more specialised function; slicing. Its length is greater than the width, and the crown contains a crest connecting a straight line of four small cusps. These seem to align with the lingual cusps of the first three premolars (p.171). As applies for all premolars the tooth is double-rooted. P5 manages to provide a reasonable impersonation of the molars although, thanks to heavy wear and damage inflicted by crushing of the fossil, some details are obscured. Each side has a cusp row containing at least three members, and there may have been a fourth in the lingual team. If so, then it was on a heavily worn cingulum that continues as a well developed cingulum. The advanced degree of wear demonsrtates this tooth formed part of the grinding mechanism, and this wear is heaviest on the lingual side, most particularly towards the front. In this case, when it comes to proportions, the length doesn't enjoy a great advantage over the width. The M1 molar is similar to the P5 but larger, its cusps are stronger and it lacks any cingula. The buccal cusp row contains a threesome of uniformly sized cusps. Its opposing row contains four and is more reminiscent of its premolar counterpart in that the first cusp is the smallest. M2 has an unexpected novelty; the origins of a third cusp row on the buccal side. This is like a cingulum but it divides into two recognisable cusps. Normally, plagis of the age have nothing of this kind upon their upper molars. A pair of cusps contribute a middle row, and these are larger than all other cusps on the crown. Meanwhile, as I'm sure you're anxious to hear, the lingual row has a trio of members, and they become progressively larger along the line from front to rear. Lower jaw Although a car insurance company would regard the mandible supplied with the type fossil as a write off, the wreckage happens to include teeth, and they correspond with those of a much better specimen (p.173). That one's somewhat battered but reasonably complete and, based on the graphics, I'm guestimating its length at around 13mm. That doesn't include the length of the impressive incisor, which juts forward for a bit before executing a steeply inclined, getting-close-to-vertical ascent. At least, its direction is closer to the vertical than horizontal plane, and its maximum height rises above that of any point of the jaw. A diastema of a millimetre or so isolates it from the rest of the teeth. The coronoid process beyond the tooth row is broad but comparatively low. It fails to rise above the maximum height of the postcanines. Its full profile is uncertain due to breakage. The tooth to the fore Overall, the teeth are similar to the choppers of Ctenacodon, another Morrison multi (p.175). As stated later by the authors, the similarities were pronounced enough to tempt them towards a referral to the same family. However, as well as an eccentric upper molar with three cusp rows, the dentition also packed a surprise among the lower teeth; an apparently un-plagi-like incisor. At least, nothing of the like is known from any other plagi. This tooth is strong, tapers along its course and has a chisel-sharp edge. Enamel only frequents the front and foremost parts of the sides. It ceases above the level of the alveolus, and that's taken as suggesting it wasn't an open-rooted, ever-growing tooth. Aside from that detail, it's reminiscent of rodents. Lower postcanines If it weren't for their larger size, then distinguishing the postcanines from Ctenacodon would be far from easy. They consist of four premolars and two molars. The first premolar is small, simple and rather boring. It attempts to add a bit of interest with a jaunty tilt backwards at its tip, but its owner was so unimpressed that it couldn't be bothered growing any more than a single root. What remains of p2 is a better effort, although a large chunk is missing. Both its roots are in place and the tip happens to be stuck on the front of the following premolar with matrix acting as glue. This is a case of the roof staying up despite the absence of the walls, and indicates the former owner was horribly ignorant about laws of gravity or essential aesthetics. A probable p2 is included with the wreckage of the type fossil's lower jaw. It's much like p1 but twice as high and long. Enamel goes down further over the front root than the rear one, and this is apparently typical for multis and also applies for the other pair of lower premolars. The crown does manage to bear a single serration. Number three of this quartet is more characterful. The crown's high at the front; double the height of the rear wall. Nevertheless, the build, implantation and tilt of the tooth all conspire to ensure the highest point s, when compared to the jaw, at the back. This tilting produced an occluding working surface running near straight along the crown, and it possesses a crest with three serrations near by. Any attractive details on the buccal side have fallen victim to a large wear facet. The p4 premolar has a length in excess of its height, and five serrations can be seen. There was also a row of cuspules towards the rear on the buccal are of the crown, and these ran forward starting from the above the rear root (p.176). It seems to have had at least three distinct members, but wear has accounted for them. This wear facet matches the orientation of the one on its predecessor. The alignment of all the premolars ensured they effectively acted as a single blade mechanism for shear delight. With later multis, the p4 expanded at the expense of the anterior premolars, and assumed all shearing responsibilities. Lower molars Teeth from both positions happen to be much alike, and that would make assigning isolated lower molars to particular positions difficult. Fortunately, if on a jaw, it's a doddle. The front one's m1. Even I could manage that. They're slightly rectangular in outline with two cusp rows; 3 buccal members and 3 lingual ones. The foremost buccal cusps of m1 have nearly been wiped away by wear, and the plane of that facet again aligns with that of the preceding tooth. Only the lonely While Glirodon must've had a mummy and daddy, cousins, uncles, grandmother and so on, clear affinities with other taxa are securely blanketed by opaque fog. Its surprisingly 'advanced' seeming characters sit puzzlingly alongside ultra conservative multi traits. Even the strong similarities of most teeth shared with allodontids (eg. Ctenacodon) offer only little assistance. They could be bequests retained from ancient ancestors in two only distantly related lineages; not necessarily inheritances from a relatively recent one. Glirodon's rodent-like and (more aptly) cimolodontan-like touches on some teeth presently render it too eccentric for an interpretations beyond: "a weird "plagi". It's a lonely heart yearning for some company. Holotype The specific name refers to the relatively large size of the multi; large, that is, among smallings. The type fossil, DINO 10822, is one of the stars in the collection of the Dinosaur National Monument, Utah. Fans flock to view it in great numbers, and yawn at local dullards such as Diplodo-thingy and Allosaur-what-not. Additional notes Other than being a large, early American multituberculate, the nature of this beast is somewhat unclear. It's an "allodontid (two families and the genus Glirodon)", (Kielan-Jaworowska & Hurum, 2001). It had gliriform incisors; incisors with the "enamel reduced to a stripe on the front side." Apparently, this condition evolved several times among multis. (With thanks for the info to David Marjanovic). It's based on a portion of snout. "Glirodon retains the plesiomorphic 'plagiaulacidan' dental formula and shares with Allodontidae the structure of the upper premolars (Pl.1 figs 2-4). It differs from the Paulchoffatiidae and Plagiaulacidae in having a single-cusped I3," (Kielan-Jaworowska & Hurum, 2001, p.401-402). I3 refers to an upper incisor.
Reference:	Engelmann & Callison, (1999), Glirodon grandis, a new multituberculate mammal from the Upper Jurassic Morrison Formation. Vertebrate Paleontology of Utah, pp.161-178, in Utah Geological Survey, (ed. Gillette DD), 8/99.

Links: The Mapstore, Utah http://mapstore.utah.gov/ Buy your copy on-line and send me one. An Inventory of paleontological resources from the National Parks and monuments in Colorado http://www.aqd.nps.gov/grd/geology/paleo/pub/fossil_conference_6/scott.htm A detailed survey of Colorado fossils, including this one from Utah!

Genus: Janumys Eaton & Cifelli, 2001 Remarks: Kusuhashi, 2008 contains some information on ?Janumys Eaton & Cifelli, 2001. I don't happen to know why that question mark is there. It could mean this doesn't apply to all fossils assigned (or provisionally assigned) to the genus. In any case, it's relevant for some reason. Apparently (p.379), in a paper I haven't seen, Hahn & Hahn assigned ?Janumys to Eobaartaridae. There's no discussion on the merits of this, and I'm not certain what precisely is meant. I'm not going to move this entry until I'm sure that's what should be done, and I'm not going to simply assume the entire genus is implicated.

Species:	Janumys erebos Eaton & Cifelli, 2001
Place:	Cedar Mountain Formation, Utah
Country:	USA
Age:	Albian-Cenomanian, Lower-Upper Cretaceous boundary
Remarks:	"Two of the multituberculates (Janumys erebos gen. et sp. n. and an unidentified taxon) are provisionally placed among 'Plagiaulacida'," (Eaton & Cifelli, 2001).
Reference:	Eaton & Cifelli (2001), Multituberculate mammals from near the Early-Late Cretaceous boundary, Cedar Mountain Formation, Utah. Acta Palaeontologica Polonica 46 (4), p.453-518.

Genus: Morrisonodon Hahn & Hahn 2004 Family: Allodontidae Remarks: Further information would be welcome.

Species:	Morrisonodon brentbaatar
Place:	Morrison Formation
Country:	USA
Age:	Upper Jurassic
Remarks:	The information I presently have is modest. However, the specific name needs me to 99.99% suspect that this fossil was formally known as "Ctenacodon" brentbaatar, Bakker RT, 1998. It's perhaps difficult to discern without a microscope, but I think I recognize a measure of similartiy between the two, and "C." b was known to have been misassigned to that other genus. Nevertheless, as I'd rather be 100% right, I'm not yet prepared to list the authorship citation for the species. It probably differs to the one for the genus. Affinities Hahn, G & Hahn, R (2006), Fossilium Catalogus Pars, 140, Backhuys Publishers, Leiden is another study that I haven't seen, but I do have the contents list. It's a contribution to the internal affinities of 'plagiaulacidans'. This genus is listed as a member of the Allodontidae. Possible citations: Bakker (1998), Dinosaur mid-life crisis: the Jurassic-Cretaceous transition in Wyoming and Colorado. in Lucas, Kirkland & Estep (eds.), Lower and Middle Cretaceous terrestrial ecosystems, New Mexico Museum of Nat Hist and Sci, Bulletin 14, p.67-77. Hahn & Hahn, 2004 The dentition of the Plagiaulacida (Multituberculata, Late Jurassic to Early Cretaceous), Geol Palaeontol 38. If somebody could remove my margin for doubt by forwarding a copy -or information from it- then please feel free to do so. Thanks are due to George Georgalis.
Reference:

Genus: Psalodon Simpson GG, 1926 Aka: Allodon ('different tooth'), Ctenacodon ('comb tooth') Family: Allodontidae Marsh, 1889 Remarks: A link to the Peabody Museum Specimen Catalogue is attached to Ctenacodon, (above). Although I haven't yet been able to track down the species involved, this genus certainly dates to 1926. The main difference between the two genera is apparently a matter of size. Psalodon is larger than C.; (eg. p4 tooth is about 3 mm long, as opposed to less than 2). Reference: Simpson (1926), Mesozoic Mammalia. IV. The multituberculates as living animals. Am. J. of Sci., 11, p.228-250.

Species:	Psalodon fortis (Marsh OC, 1887) Simpson GG, 1927 or 1929
Aka:	Allodon fortis Marsh, 1887
Place:	Morrison Formation, Wyoming
Country:	USA
Age:	Upper Jurassic
Remarks:	The holotype is at the Peabody Museum, Yale. This is possibly the same species as P. potens. This species is based upon a premaxilla.
Reference:	Marsh (1887), American Jurassic mammals. Am. J. Sci. (3) xxxiii: p.326-348.

Species:	?Psalodon marshi Simpson GG, 1929
Place:	Morrison Formation, Wyoming
Country:	USA
Age:	Upper Jurassic
Remarks:	Several Peabody specimens of a possible P. species. It may represent the lower teeth of P. potens, (Martin & Foster 1998, p.387). A further specimen resides at the South Dakota School of Mines and Technology. This is a fragment of right jaw with a couple of teeth, (p4-m1). It was recovered from the Little Houston Quarry in the Black Hills. Other finds from this quarry include much plant material, fish, a possible lizard and various dinosaur fossils: Allosaurus, Camarasaurus, Stegosaurus, Barosaurus or Diplodocus, Apatosaurus, Camarasaurus, Dryosaurus and Othnielia. The holotype works in the Smithsonian Institute, Washington, (USNM 2684). Carpenter 1998 (p.398) contains the following lengths for postcanine teeth: premolars: p2 1.7mm, p3 1.7mm, p4 3.0mm; molars: m1 2.1mm.
Reference:	Simpson (1929), American Mesozoic Mammalia, Mem. of the Peabody Museum 3, p.1-235.

Species:	Psalodon potens (Marsh OC, 1887) Simpson GG, 1927 or 1929
Aka:	Ctenacodon potens Marsh OC, 1887
Place:	Morrison Formation, Wyoming
Country:	USA
Age:	Upper Jurassic
Remarks:	A further Yale holotype. This is based on a maxilla.
References:	Marsh (1887), American Jurassic mammals. Am. J. Sci. (3) xxxiii: p.326-348.

Genus: Zofiabaatar Bakker RT & Carpenter K, 1990 'Zofia’s hero' (Zofia Kielan-Jaworowska) Family?: Zofiabaataridae Bakker RT, 1992

Species:	Zofiabaatar pulcher Bakker RT & Carpenter K, 1990
Place:	Pine Tree Ridge, Morrison Formation, Wyoming
Country:	USA
Age:	Upper Jurassic
Remarks:	Much of the following is based upon my reading of Carpenter, 1998. Although this genus and the paurodontid Foxraptor were established in 1990, the original description included some unfortunate problems. I don't want to delve into or dwell upon them. However, a paragraph from page 394 provides some context. "Numerous errors regarding Zofiabaatar and Foxraptor by Bakker and Carpenter (1990) necessitates redescription of the specimens. The turtle, Uluops, is now under restudy by Gene Gaffney who will correct errors in Carpenter and Bakker (1990). In both contributions I appear as an author in name only and without my knowledge." Both genera come from a site called Breakfast Beach, Pine Tree Ridge. This is near the uppermost part of the Morrison Formation, (p.393). Consequently, these fossils must be somewhat more recent than other Morrison mammals, (eg from Como Bluff to the west, p.394). Robert Bakker found the locality in 1977, and bulk sampling began two years later. Zofiabaatar is known from a left lower jaw minus the incisor, the first premolar and the second molar. The fragility of the fossil required particularly careful preparation. Carpenter places the genus within Plagiaulacidae, although other researchers recognise a separate, presently monotypic family. I'm following Kielan-Jaworowska & Hurum, 2001, who refer it to Zofiabaatar. (Carpenter's usage of Plagiaulacidae also includes the allodontids Ctenacodon and Psalodon.) Mandible The mandible differs from ('other' according to Carpenter) plagiaulacids in being relatively short and deep, (p.395). A feature at the back termed the pterygoid fossa is particularly large. "The tooth row is oblique to the long axis of the jaw as it is in other plagiaulacids." At its longest point, the fossil is a touch over two centimetres in length, (p.396). The coronoid process is also enlarged and robust, when compared to other plagis. Teeth As far as its known, the tooth formula is: (lowers): four premolars and two molars. Incisors None are known for sure, although one was featured in the original description. However, it doesn't resemble any plagiaulacid incisor, and looks more like a tooth from the premaxilla of a small theropod dinosaur. It was allegedly found a foot from the jaw, but this seems to have been a groundless claim. Premolars The premolars increase in size along the series from p1 to p4, and they're blade-like. The rear edge of each is overlain by the front edge of its following colleague. The entire ensemble operates as one, near-continuous blade with a complete length of just over half a centimetre. In comparison to Ctenacodon and Plagiaulax, the serrations and cusps are less distinct. The p1 was small, bulbous and went walkabout when the rock was opened; ie it got lost. The p4 is about twice as long as p2 and p3 combined. The crown's damaged, but it had at least seven serrations. The three premolars now present were double-rooted. Excepting for the p4, the roots were of similar sizes. In the case of the exception, the rear root is larger, (p.397). The following lengths are from Table 1: p2 0.88mm; p3 1.24mm; p4 2.6mm. Molars When viewed from above, (occlusal), the m1 is roughly square in outline. The crown is 1.31mm in length. It has only two rows of cusps, labial and lingual, each of which contains but a pair. The ones on the labial side show much wear, and this is a feature typical of plagiaulacids. The second molar mentioned in the dental formula is neither present nor discussed. There's precious little room between the m1 and the coronoid process for another tooth, so the m2 can't have been impressive, (which isn't unusual for multis). A misleading perspective In the 1990 paper, the mandible was said to be of a "Cretaceous grade". Carpenter finds this unjustified. The jaw happens to be exposed in the lingual perspective so that the inner side is visible. More usually, it's the external surface which can be seen. Right at the back is found the condyle, and this was the basis of the claim. However, the shape and orientation of the condyle can vary on the labial and internal sides, (eg. Chulsanbaatar), and the original reason for establishing the family were rejected, (p.398) Family matters In a later study, Kielan-Jaworowska & Hurum (2001) offered a diagnosis of Zofiabataaridae which involved some further points. I don't know how Carpenter views their opinions, but he could hardly be expected to take them into account in advance. "We regard the Zofiabaataridae as belonging to the allodontid line", (Kielan-Jaworowska & Hurum, 2001, p.401). They observe that the family: "Differ from Allodontidae and Plagiaulacidae in having very short m1 with two cusps in two rows, and from Plagiaulacidae in having lower molar cusps separated. Zofiabaatar differs from all 'plagiaulacidans' in having the condyle facing upwards, rather than posteriorly, and shares this character with some advanced Djadochtatherioidea and Taeniolabidoidea. It differs also from other 'plagiaulacidan' genera in having strongly enlarged pterygoid fossa." McKenna & Bell, (1997), located the genus within Plagiaulacidae, though the term was used for a paraphyletic group, rather than a complete clade. Holotype The holotype, UCM 42329, is in the collection of the University of Colorado Museum. It's also been incorrectly referred to as UCM 42239.
Reference:	Bakker & Carpenter (1990), A new latest Jurassic vertebrate fauna, from the highest levels of the Morrison Formation at Como Bluffs, Wyoming. Part III. The mammals; a new multituberculate and a new paurodont. Hunteria, 2, p.4-8.

Link: Prehistoric Data Files http://www.angellis.net/Web/DFG-mam/Zofiabataar.htm A sketch of Zofia’s hero.

Other reports: Japan It's a multi. I've put it here because it had to go somewhere. I've no idea about its affinities. Reference: Takada T, Matsuoka H & Setoguchi T (2001), The first multituberculate from Japan: In: Proceedings of the Eighth Annual Meeting of the Chinese Society of Vertebrate Paleontology, edited by Deng T & Wang Y, China Ocean Press, p. 55-58. This presumably involved fossils of subsequently described eobaatarids. See either probably Hakusanobaatar or perhaps Tedoribaatar. Both were described in 2008. USA "Reported but not yet described from the Early Cretaceous (Albian) of North America (Texas)." McKenna & Bell, (1997), p.36. Multi material was collected around fifty years ago at Trinity, Texas. As of 2001 it was still undescribed, (Kielan-Jaworowska & Hurum 2001, p.410). A tantalising clue is present in Simpson 1959, p.406: "The multituberculates, not yet fully described, are said to be nearer those of the late Jurassic (Plagiaulacidae) than those of the late Cretaceous." Dinosaur National Monument, Utah "There are a smaller number of teeth, including both upper and lower premolars and molars, that are very similar morphologically to C. laticeps and/or C. serratus but are about 50% larger. They are substantially larger than any Ctenacodon from Como Bluff and only slightly smaller than species (probably only a single species) of Psalodon from Como Bluff... and we believe this represents a new species that is best referred to Ctenacodon", (Engelmann & Callison 1998, p.362-363). They also report an upper P4 tooth which may represent a new species of Psalodon, (p.363). Several other teeth from the same location are reported on page 364. These are upper, front premolars and they correspond to the size of Ctenacodon laticeps, but the morphology's different. These have four main cusps, which is a feature known from paulchoffatiids, (four or more, next section). The relative lack of wear suggests the possibility that these may be milk teeth, and not necessarily paulchoffies. Australia, Flat Rocks, Victoria "The tooth has a blade-like form such as is known in only four groups of mammals. Only one of these groups is known from the Mesozoic Era when the dinosaurs lived. This group is called the multituberculates... What we seem to have is either a highly unusual multituberculate or a heretofore unknown group of mammals." Whatever this turns out to be, it won't qualify as a basal multi. It shouldn't be on this page, but a more suitable placement would require more information. Updat: Described as Corriebaatar in 2009. Provisionally, it's been referred to as a possible cimolodontan. However, I've added the relevant entry to the above section for now. Link: Monash University, Dinosaur Dreaming, 2004 Annual Report on excavations http://www.sci.monash.edu/msc/dinodream/ The report includes a photo. Excavations will continue, and more informative fossils may turn up. These people are very determined. (Thanks are due to Dann Pigdon, who originally posted the report on the Dinosaur Mailing List.)

A. Middle Jurassic Multitubercuates B. ‘Basal’ Multituberculata C. Paulchoffatiidae & Hahnodontidae D. Pinheirodontidae

C. PAULCHOFFATIIDAE & HAHNODONTIDAE

Taxon: Paulchoffatiidae Hahn G, 1969 Taxon: Hahnodontidae Sigogneau-Russell, 1991 Paulchoffatiidae is perhaps the most basal known family within Multituberculata, at least in terms of the anatomy of the lower jaw and premolars. It's certainly the oldest named family for which remains are known, although there are now some multi teeth from the Middle Jurassic of England. Doubts have been expressed as to whether the taxon is monophyletic, (if it contains all the descendents of a common ancestor). Many of the species names have indirectly come from the book, Guimarota - A Jurassic Ecosystem, (see two links down). A listing was kindly supplied by David Marjanovic. Of paulchoffies, hahnodontids and friends Hahnodon and its newly described sibling, Denisodon, belong to a separate but related family. Hahn & Hahn, 2003 proposes the new superfamily of Paulchoffatioidea, presently composed of three families: Hahnodontidae, Paulchoffatiidae and Pinheirodontidae. Various common peculiarities of the dentition (technically known as autapomorphies) form the core of their reasoning (p.350). "These autapomorphies definitely exclude the Paulchoffatioidea from the ancestry of younger multituberculates ("Cimolodonta"). The Paulchoffatioidea are thus the oldest known side branch on the multituberculate evolutionary tree." Furthermore, the authors mention basal structures of the skull, (known specifically from the paulchoffatiids), but exclude them from the discussion in this particular paper. They also give a list of further plesiomorphic (basal) characteristics, (p.350-351): - The upper jaw has relatively many teeth, including a canine and five premolars. - The lower premolar 4 is pretty much the same length as the p3, and has only four serrations. - There's a row of fairly large cusps on the p3 and the p4. - The premolars are grinders rather than cutters. - "The angle between the longitudinal axis of the dentary and the tooth row is only 7 to 20°." - The tooth enamel, (in Paulchoffatiidae at least), isn't prismatic All these features are primitive for multis. Family name The name is naturally enough derived from the genus Paulchoffatia, and that was chosen to honour Paul Choffat, a pioneer of Portuguese geology. Should you not be familiar with the word, it can look like a bit of a mouthful. However, they can be rendered much more friendly by terming them paulchoffies. The animals won't mind at all. The angle of longitudinal axis... That potentially baffling character is mentioned in the concise list of plesiomorphic characteristics above. It's been found useful for sorting one lot of paulchoffies from another, but what the hell does it mean? I left it in quotation marks because I wasn't sure. Reading Hahn, 1971 has helped make things clearer, and thanks are due to the donor. Typically, the row of lower postcanine in multis takes a kind of diagonal approach to sitting on in the jaw (p.10). This concerns the alignment of the entire row, rather than the attitude of each individual. The rear molar is set near to the internal edge of the jaw bone ( lingual). the rest of the postcanines are placed progressively ever further externally towards the buccal edge with the first premolar closest to it. That allows a paleontologist -should they feel so inclined- to plot a couple of lines, as there's an angle of divergence between the direction of the postcanine row as compared to the inner line of the jaw. Grab a piece of paper and a ruler if needs be. Draw a line going straight ahead. Keeping the ruler secured to the start of that line, tilt the thing to some degree in either direction, and now draw a relatively diagonal line. The space between them both is this here 'angle of longitudinal axis' comparison. Generally for multis, the exercise would do little more than help you pass some time. That's not so for paulchoffies. A low scorer in this contest is Paulchoffatia with an angle of merely 10°. Kuehneodon, on the other (more derived) paw, could boast of around 20°, and strut itself in this respect more in line with the general tramp of multi-orthodoxy. Further back for your roots A consequence of this sideward shift in the tooth row was the opening up of jaw space for an expanded root of the single lower incisor. That allowed for an enlargement and realignment of this tooth. Rather than standing up for itself, the tooth took to pointing in a forwards slant while being securely anchored into place. The incisor in Paulchoffatia is nearer to being vertical than its more procumbent counterpart in Kuehneodon. Such procumbent teeth are popular for several purposes among various mammals, and this isn't the only route natural selection could've taken to produce such a thing. However, it was the direction allocated to multis. Some tooth points Staying with our new pair of paulchoffie pals (Paulchoffatia and Kuehneodon), Hahn offers some discussion of the lower premolar situation. For p2-p4 (p.17), the lengths aren't all that different, and this is a primitive trait for multis. Furthermore, shearing enhancements are very limited with four or -occasionally- five projections from the cusps on the lingual side of the crown. Cusps located on the buccal side attain only around a quarter of their height. What transpired in somewhat more derived multis is that the p4 expanded to around double the length of the p3, and the count of those projections increased to at least six (p.18). The buccal cusps were also subject to further reduction. Although paulchoffies had some kind of potentially usable blades for cutting available on lower premolars, this actually eroded away with increasing usage and age (p.20), and the teeth could then only be used as grinders. However, nature went to work on those potential shearing abilities for other multis, and enhanced them yet further. In terms of known multi premolars, paulchoffies are archaic eccentrics. Upper teeth Hahn gave details on teeth then assigned to Paulchoffatia, but that can't be their present taxon of residence. Only lower jaws have been identified for that genus. Uppers are probably known, but the problem is which they are. As only a single paulchoffatiin had been found in the fauna, matching uppers and lowers seemed straightforward. However, a second and more successful bout of excavations began at Guimarota in 1973, and paulchoffatiin multis decided to multiply prolifically. Quite which genus Hahn was describing is a mystery to me. Nevertheless, it's at least a very close relative of Paul Choffat's personal pet, as it's in the same subfamily. It was thought to be unique among multis for having six upper premolars a side (p.20), but that was actually incorrect. Generally, multis didn't have canines. However, this critter had one, and it happens to be abnormally premolariform. The canine and first four premolars are all roughly circular in outline seen from the occlusal perspective and each is blessed with four cusps (p.21). However, the first tooth is nevertheless a canine. Apart from its position in the dental row, it also differs a bit from those following canines as the buccal cusp to the front is poorly formed. Furthermore, this tooth is single-rooted and required a larger alveolus. That feature on a toothless fragment led Kühne to conclude it was for a canine in 1961 (p.22), and he turned out to be correct. As the third incisor possessed an abnormally complex crown for such a tooth in a mammal, the canine must've felt moved to also develop eccentrically. Upper premolars The P5 of a paulchoffiin possessed three cusp rows whereas the multi norm is two. In this case, an extra line of five had developed on the buccal side (p.24). The middle member of this gang was by far the largest. This refinement enhanced the chewing talents of the tooth (p.25) whereas, in general, multi fashions favoured rear premolars built for shearing. Additional notes on paulchoffies Kielan-Jaworowska & Ensom, 1992 offer some observations specifically on family traits (p.102). Paulchoffie characteristics include: an upper incisor (I3) with three or four cusps; the first lower molar (m1) has a cingulum at the front with cuspules, and two cusp rows in which one buccal cusp is particularly big; a basin-like crown for the m2 with just one cusp at the front on the lingual side. For the upper molars (and as for multis in general), the M2 is set lingually of the M1. The basin-like m2 is very untypical of multis. In more advanced critters, the crown of that tooth developed into being a multi-cusped grinder. Dead end "The paulchoffatiid and plagiaulacid lines are more closely related to one another... than either of them is to the allodontid line," (Kielan-Jaworowska & Hurum, 2001, p.402). They go on to observe, (same page): "We follow this opinion (see also Hahn 1993) and regard the Paulchoffatiidae as a monophyletic dead-end of the 'Plagiaulacida'." Guimarotan multis Directly from Guimarota - A Jurassic Ecosystem, (Hahn G & Hahn R 2000, p.97 -with thanks to Father Christmas): "Multituberculates are small animals, usually of the size of a mouse or a rat. They are the longest lived group amongst the known mammals, with a lifetime as a group from the Late Jurassic to the Early Tertiary (Oligocene) [Granger & Simpson 1929], a time span of more than 100 million years. Within Mesozoic mammal communities, they occupied the ecological niche that is filled by the rodents today: they are small, omnivorous to herbivorous, exhibit a significant variety, and occur in high numbers of individuals. If multituberculates are found in a fossil locality, they are usually the most common group of mammals (Hahn 1978a)." The authors go on to report that most finds from Guimarota represent paulchoffatiids. So far only remains of skulls, lower jaws and isolated teeth have been recovered. The state of preservation is generally not good, with the bones being often broken and distorted. Another difficulty is that, with the exception of Kuehneodon, no correlation has so far been possible between lower and upper jaws. As some genera are based on the lower, with others based upon the uppers, there's almost certainly some overlap involved. Rather than representing a dozen actual genera, seven or eight is more likely.

Links: Toby White, Palaeos: The Vertebrates, Allotheria http://www.palaeos.com/Vertebrates/Units/Unit420/420.100.html These anatomy crib notes restrict Paulchoffatiidae to the genus of Paulchoffatia, (sensu Simmons, 1993). It seems this isn’t in line with the more recent work of Kielan-Jaworowska and Hurum (2001). Guimarota - A Jurassic Ecosystem, edited by Martin T & Krebs B, 2000 http://www.pfeil-verlag.de/07pala/e2_80.html A sneak preview of what seems to be a well worthwhile read. Several sample pages can also be viewed, (pdf). Amongst the non-mammal fossils are Archaeopteryx-like teeth, which is the first evidence I’ve come across for Jurassic birds outside of Bavaria. They also predate the famous fossils of the Solnhofen Formation. This was the most intensive paleontological excavation in European history. The Fossils of Guimarota, Portugal ?????????????.??? This is a virtual link. For some incomprehensible reason, such a homepage doesn’t exist. The location sounds fascinating. Would some kind person please construct a suitable site, so I can have a real link to it? Genera: Bathmochoffatia, Calveodon (= Galveodon), Denisodon, Galveodon, Guimarota (?!?), Guimarotodon, Hahnodon, Henkelodon, Kielanodon, Kuehneodon, Meketibolodon, Meketichoffatia, Parachoffatia (preoccupied = Plesiochoffatia), Paulchoffatia, Plesiochoffatia, Pseudobolodon, Renatodon, Sunnyodon, Xenachoffatia, other reports Time-Line: Lower Cretaceous: Denisodon, Galveodon, Hahnodon, Kuehneodon (Galve), Paulchoffatia (Galve) Upper Jurassic: Bathmochoffatia, Guimarotodon, Henkelodon, Kielanodon, Kuehneodon (Guimarota), Meketibolodon, Meketichoffatia, Paulchoffatia (Guimarota), Plesiochoffatia, Pseudobolodon, Renatodon, Sunnyodon, Xenachoffatia

Genus: Bathmochoffatia Hahn G & Hahn R, 1998

Species:	Bathmochoffatia hapax Hahn G & Hahn R, 1998
Place:	Guimarota
Country:	Portugal
Age:	Kimmeridgian, Upper Jurassic
Remarks:	Known from one upper molar. Merci beaucoups David Marjanovic.
Reference:	Hahn & Hahn 1998), Neue Beobachtungen an Plagiaulacoidea (Multituberculata) des Ober-Juras. 3. Der Bau der Molaren bei den Paulchoffatiidae. Berliner Geowissenschaftliche Abhandlungen, E, 28, p.39-84.

If you're wondering what 'Kimmeridgian' means, it's a an epoch of geological time, running from approximately 154 -151 million years ago. It's named after Kimmeridge Bay, Dorset, England, (pictured). This cliff is the type-locality of the epoch. It's fossil rich, though the remains found aren't in the best state of preservation. What the photo shows reasonably well are the strata, which were deposited one upon the other. You might also be able to make out the fairly regular heaps of scree at the base of the cliffs, which are constantly collapsing. Kimmeridge is part of one of the largest on-land oilfields in Europe.

Genus: Denisodon Hahn G & Hahn R, 2003 'Denis's tooth' Family: Hahnodontidae Sigogneau-Russell D, 1991 Remarks: Named after Dr Denise Sigogneau-Russell in honour or her contribution to the knowledge of Mesozoic mammals.

Species:	Denisodon moroccensis Hahn G & Hahn R, 2003
Place:	Anoual
Country:	Morocco
Age:	?Berriasian, Lower Cretaceous
Remarks:	The following is based upon my reading of Hahn & Hahn, 2003. This is described as being represented by an isolated lower molar, (m2), which is a full millimetre in length, (p.352). (Note, as stated below, doubts have been expressed about this.) The crown is well preserved, though the two roots are broken. Seen from above, the shape is something like that of an hour-glass. The type fossil is known as MNHN SA 97 and it's a resident of the Muséum national d`Histoire naturelle in Paris. The specific name is after Morocco, (but you probably worked that out for yourself). Room for doubt Butler & Hooker, 2005 are of the opinion that this might be the m1 of Hahnodon, (p.202). Furthermore, there's no evidence on either tooth consistent with a horizontal chewing action. Wear suggests crushing rather than grinding. There's also no wear suggesting the upper M2 had a more lingual position than its lower equivalent, as is the case for other multituberculates. As a consequence, they refer hahnodontids to 'Haramiya' rather than Multitubercualta. I'll leave the pair on this directory for now, and await further opinions.
Reference:	Hahn & Hahn (1992), New multituberculate teeth from the Early Cretaceous of Morocco, Acta Palaeontologica Polonica 48 (3), p.349-356.

Link: APP 48 (3), Hahn & Hahn, 2003 http://app.pan.pl/acta48/app48-349.pdf If you don't believe my account, have a look in the paper. It's on-line.

The Anoual Syncline, Morocco - Lower Cretaceous The following is based upon my reading of Hahn & Hahn, 2003. Until the 1980s the list of African Mesozoic mammals consisted of two genera from the Elliot Formation of southern Africa, one genus (Brancatherulum) from Tanzania, and a misidentified piece of quartz. A few more have since been described from elsewhere, but this Lower Cretaceous locality in the eastern High Atlas Mountains has produced a vast increase in the register. The age is probably Berriasian. The mammalian fauna from Anoual is impressively diverse, and still expanding, though remains are restricted to teeth, (as far as I'm aware). Before this publication added a further genus, twelve genera had been described and six more were thought to be present. Represented were, (according to the original interpretations): Triconodonta, 'Symmetrodonta', Pantotheria, Peramura, Tribosphenida and Multituberculata. Subsequent doubts have arisen concerning the affinities of the supposed 'symmetrodonts'. Further afield It's hardly possible to compare this haul with other Lower Cretaceous faunas from Africa, seeing as they're rather lacking. Abelodon from Cameroon is one exception but, as yet, it's very lonely. (If anyone's planning to visit Niger*, perhaps it might be worth checking the dinosaur sites. Then again, maybe the geological situation is unsuitable.) However, there are certainly similarities between the mammals of Amoual and their nearish contemporaries of Laurasia as known from Europe. One taxon, (the probably-not 'symmetrodont' named Thereuodon), has also been reported from the older rocks of Dorset, England. Normally, when multituberculates are present at Upper Jurassic and Cretaceous northern locations, they're relatively well-represented and often form the majority of among the mammals. However, for unknown reasons, they're a relatively small contingent in the Anoual fauna. Both of the first published triconodonts are fairly eccentric and fish has been proposed as a suitable diet. A couple more were susequently published. * Talking of Niger... Guess where Greg Wilson of the University of Washington (Seattle) went in 1990 along with Paul Sereno and crew, and guess what they were doing? Yes, looking for Lower Cretaceous mammals and other less interesting beasts. He told me after reading the bit above. The word on the paleo-street has it that around 6,000 pounds of sediment from various horizons were arrested on suspicion of being potentially interesting. That's the amount he admits to having dragged in personally, and it's still being processed. Others were more dino-inclined. Dinosaur expedition, Mesozoic Mammal Man http://www.projectexploration.org/niger2000/10_30_00_interview_wilson.htm An Interview with Greg Wilson by Gabrielle Lyon, photos by Mike Hettwer. Further Mesozoic site summaries can be found at Localities. Meet the Mammals of Anoual (15 genera, 16 species) Triconodonta (4 genera, 4 species) Dyskritodon amazighi; Ichthyoconodon jaworowskorum; Kryptotherium polysphenos; Gobiconodon palaios Multituberculata (2 genera, 2 species) Denisodon moroccensis; Hahnodon taqueti Dryolestoidea (3 genera, 3 species) Atlasodon monbaroni; Donodon perscriptoris; Thereuodon dahmani Peramura (1 genus, 1 species) Peramus sp. Zatheria and stem-Zatheria (3 genera, 3 species) Afriquiamus nessovi; Microderson laaroussii; Minimus richardfoxi Tribotheria (1 genus, 1 species) Tribotherium africanum Boreosphenida (1 genus, 2 species) Hypomylos phezlizoni; H. micros

Genus: Galveodon Hahn G & Hahn R, 1992 Aka: Calveodon

Species:	Galveodon nannothus Hahn G & Hahn R, 1992
Place:	Castellar Formation - Camarillas Formation, Galve
Country:	Spain
Age:	lower Barremian, Lower Cretaceous
Remarks:	I've spent months searching for information on a spelling mistake, (Calveodon)! There is a tooth in the collection of the museum in Galve. The genus is based upon an upper incisor (I2) and a lower premolar, either p3 or 4. These attest to a relatively small multi, (Canudo & Cuenca-Bescós 1996, p.221). Hahn & Hahn, 2002 The authors who established the species subsequently had an upper premolar from Galve to describe, and this is the taxon they referred it to (p.258). The tooth concerned is roughly oblong in occlusal view and blessed with two pairs of cusps. The rear is somewhat narrower than the front. Such tetracusp crowns for front upper premolars (P1-P3) are limited, as far as is known, to two Plagiau families; Paulchoffatiidae and Pinheirodontidae. Both happen to be represented in the Galve fauna. However, the closest similarity noted was with a P3 of Henkelodon naias, a paulchoffi. The crown of pinhei's premolars differs in detail. Therefore, a paulchoffie is more likely, and that suggests Galveodon rather than Lavocatia (Galve's pinhei). Of course, a presently unknown candidate can't be completely ruled out as the actual owner. A second factor is the size of this complete and well-preserved fossil. Its owner can't be fairly accused of having been a big mouth, not with a front premolar with a maximum length of 0.7mm. For those somewhat unfamiliar with the metric system, that translates as ever so, oh, ever so so tiny. Galveodon nannothus qualifies as a plausible owner on those grounds as well.
Reference:	Hahn & Hahn (1992), Neue Multituberculates-Zähne aus der Unter-Kreide (Barremian) von Spanien (Galve und Una). Geologica et Parlaeontologica, 28, p.143-162.

Link: Museo http://adigital.pntic.mec.es/~tronchon/dinos/museo.htm The homepage of the museum in Galve, (Spanish). Dinosaurs, pterosaurs, teeth…

Alleged Genus: Guimarota Hahn G, 1969 'Guimarota' (geographical)

Alleged species:	Guimarota freyi
Place:	Guimarota
Country:	Portugal
Alleged Age:	Kimmeridgian, Upper Jurassic
Remarks:	I thought this might have been the a misinterpretation of the name of the coalmine, but apparently not! I don't know what it's supposed to be based on. It may be a synonym for Guimarotodon. Another possibility is a mangling of Guirogatherium freyi.... The mysts of tery are clearing 'Myststery' is a joke. If you remove the middle 'st' and... Oh, never mind. I liked it! Information has come to hand that links this 'genus' and several others to Kühne in 1968. He published some names in a manuscript but didn't provide full descriptions. I'll add appropriate details as and when I get around to them. In any case, this name's not valid.
Reference:

Genus: Guimarotodon Hahn G, 1969 'Guimarota tooth' Remarks: Guimarotodon Hahn 1969 exhibits a more slender Corpus mandibulae than either Paulchoffatia or Meketibolodon. The most conspicuous character of this genus is the morphology of the P3-4 and the M1.", (which are upper teeth). "The incisor is relatively little curved and its root is of similar length as that of Meketibolodon and extends to underneath the posterior premolars." (Both quotations from Hahn & Hahn 2000, p.105-106).

Species:	Guimarotodon leiriensis Hahn G, 1969
Place:	Guimarota
Country:	Portugal
Age:	Kimmeridgian, Upper Jurassic
Remarks:	Three specimens; lower jaws.
Reference:	Hahn (1969), Beiträge zur Fauna der Grube Guimarota Nr. 3. Die Multituberculata. Paleontographica, A, 133, p.1-100.

Genus: Hahnodon Sigogneau-Russell D, 1991 'Hahn's tooth' Family: Hahnodontidae Sigogneau-Russell D, 1991

Not a species!:	Hahnodon purbeckensis ascribed to Sigogneau- Russell D, 1991
Remarks:	I have seen a reference to such a species, but this doesn't exist, (see Kielan-Jaworowska & Hurum, 2001, p.413). It could be the result of confusion with Gerhardodon purbeckensis Kielan-Jaworowska Z & Ensom PC, 1992, (compare with the named Homo sapiens, Gerhard Hahn).

Species:	Hahnodon taqueti Sigogneau-Russell D, 1991
Place:	Anoual
Country:	Morocco
Age:	?Berriasian, Lower Cretaceous
Remarks:	Caution: as mentioned below, doubts have been expressed about the interpretation of this fossil. Known from a single lower molar. "The holotype of H. taqueti (Sigogneau-Russell 1991a: figs. 1a,2a,b) is a left m2, as correctly stated by the author. The tooth resembles, besides the presence of the anterobuccal cusp, the m2 of Kuehneodon (compare Fig. 1C). Both teeth show a relatively large anterior and a smaller posterior lingual cusp, separated one from the other by the indentation of the lingual wall", (Hahn & Hahn 2003, p.351). A couple of further teeth from the same locality have been referred to the family, but not to a genus. The first is a rear upper molar, (P3 or P4). The other is a lower incisor, (the front teeth), (Hahn & Hahn, p.353-355). In contrast to both the paulchoffatiids and pinheirodontids, the enamel of hahnodontid lower molars is smooth, and there are further distinctions in details of the cusps. In terms of numbers of taxa, this location is unusually productive. There are now around a dozen published genera, with six more undescribed, (p.349). So far identified are members of Triconodonta, 'Symmetrodonta', Pantotheria, Peramura, Tribosphenida and Multituberculata. This is in stark contrast to the rest of Africa. The relative paucity of multis, (2 genera), is also in contrast to broadly contemporaneous fauna elsewhere. Where present, multituberculates tend to be amongst the most numerous groups. A differing view The interpretation of this tooth as a left lower molar wasn't certain, (Butler & Hooker 2005, p.201). Three cusps are present, with two on one side and a singleton on the other. The singleton was hesitatingly referred to as buccal. It's higher than the opposing pair. With other multis the lingual cusps are higher. If that precedent were followed, then this would be a right molar. Furthermore, the two authors are of the opinion that hahnodontids aren't multis. They reassigned them to 'Haramiya'. Should that position find support, then they will be relocated to the appropriate virtual enclosure.
Reference:	Sigogneau-Russell (1991), First evidence of Multituberculata (Mammalia) in the Mesozoic of Africa. Neues Jahrbuch für Geologie und Paläontologie, Monatshefte 2, p.119-125.

Link: The Prehistoric Data Files http://www.angellis.net/Web/PDfiles/MarsupS-em.htm An impressively broad index of genera and species information, which has enabled me to fill in a far few gaps; eg. Hahnodon.

Genus: Henkelodon Hahn G, 1977 'Henkel’s tooth'

Species:	Henkelodon guimarotensis
Place:	Guimarota
Country:	Portugal
Age:	Kimmeridgian, Upper Jurassic
Remarks:
Reference:

Species:	Henkelodon naias Hahn G, 1977
Place:	Guimarota
Country:	Portugal
Age:	Kimmeridgian, Upper Jurassic
Remarks:	The author and year were kindly supplied by Vince Ward. According to Kielan-Jaworowska & Hurum, 2001, (p.413), this genus was named in 1987. However, Hahn and Hahn 2000, (p.105), supports 1977. A specimen known as VJ 401-155 works in the Museum of the Geological Service, Lisbon. The species is known from one upper jaw.
Reference:	Hahn G (1977), Neue Schädel-Reste von Multituberculaten (Mamm.) aus dem Malm Portugals. Geologica et Palaeontologica, 11, p.161-186.

Genus: Kielanodon Hahn G, 1987 'Kielan’s tooth' (Zofia Kielan-Jaworowska?)

Species:	Kielanodon hopsoni Hahn G, 1987
Place:	Guimarota
Country:	Portugal
Age:	Kimmeridgian, Upper Jurassic
Remarks:	Three upper jaws.
Reference:	Hahn (1987), Neue Beobachtungen zum Schädel- und Gebiss-Bau der Paulchoffatiidae (Multituberculata, Ober-Jura). Paleovertebrata 17, p.155-196.

Genus: Kuehneodon Hahn G, 1969 'Kuehne’s tooth' Aka: Kuehnodon Remarks: Some details have come from Hahn, 1971. This author compares characters of Kuehneodon and Paulchoffatia. At that time, they were the only two members of the family known, although several later genera had been incorrectly designated as paulchoffies. Be that as it may, Kuehneodon was found to be the more derived of this pair, and the contrasts were large enough to win it a subfamily as a prize; Kuehneodontinae Hahn, 1971. Paulchoffatiinae was also established at the same time. Why? Two twigs of one family I'm ignoring information presented on upper jaws. Paulchoffatia has since been restricted only for some lower specimens, and it's now not clear which the relevant uppers are. That doesn't affect the differences between the fossils, but it makes it difficult to use the correct label. With regards to Kuehneodon, the lower teeth are implanted diagonally along the jaw, with the second (the rearmost) molar set near to the lingual margin of the bone and the first premolar closer to the to buccal margin. With Paulchoffatia this characteristic is much less extreme, as the tooth row is closer to running parallel with the inner edge of the jaw. Additionally, the lower incisor is significantly more procumbent for Kuehneodon. It shares both of these traits with the typical multi mainstream. However, aspects of the lower premolars align the genus up with the paulchoffies and not with the further derived 'Plagiaulacidans'. These teeth are of similar lengths, rather than the back one being considerably longer. Furthermore, they were built for grinding up food rather than slicing it. Additional notes Named in honour of paleontologist, Walther Kühne, pioneer of the Guimarota site in the late '50s and early '60s. He began working there as the mine ceased its economic life. Just as things started to get really interesting though, he moved on to Galve in Spain. A great deal of resources were invested in excavating Guimarota between 1973 and 1982, under the auspicious of the Freie Universität of Berlin. Kühne, also of that institution, seems to have felt the money could have been better used elsewhere. Luckily, the investment paid off; 1,000 mammal jaws, a number of skulls, two skeletons and 10,000 or so teeth. It's 20 years later and some of this material has yet to be prepared. The last I heard, they think they’ve got the mammal material sorted out. This is the one genus from the site whereby the lower and upper jaws have been found united. These: "exhibit the lowest number of derived characters ( apomorphies), and are thus closest to the main evolutionary lineage of the multituberculates", (Hahn & Hahn 2000, p.106).

Link: Acta Palaeontologica Polonica 46 (1), p.98 http://www.pfeil-verlag.de/07pala/e2_80r1.html Rich Cifelli's review of Guimarota: A Jurassic Ecosystem.

Species:	Kuehneodon dietrichi Hahn G, 1969
Place:	Guimarota
Country:	Portugal
Age:	Kimmeridgian, Upper Jurassic
Remarks:	The following is based largely upon my reading of Hahn, 1971. This study predates the start of the second and more extensive campaign of excavations at Guimarota, and was written when the supply of fossils was much more limited. Parallel developments This species is derived for a paulchoffie, as is the genus. It echoes some aspects of still more advanced multis (p.14) such as Plagiaulax. For example, both are descendants of animals which had replaced their p1 premolar with a toothless gap. The lower incisor is strongly curved in the two lineages. However, that doesn't make them descendants of a common ancestor with such features. Other details rule out Kuehneodon from any parental responsibilities towards Plagiaulax and Co. The similarities are matters of coincidence and not shared descent. Holotype The type fossil is known to its friends as VJ 4-155. This handsome right lower jaw works at the Museo Servicos Geologicos, Lisbon. Additional notes A lower jaw species based on twenty specimens.
Reference:	Hahn (1969), Beiträge zur Fauna der Grube Guimarota Nr. 3. Die Multituberculata. Paleontographica, A, 133, p.1-100.

Species:	Kuehneodon guimarotensis Hahn G, 1969
Place:	Guimarota
Country:	Portugal
Age:	Kimmeridgian, Upper Jurassic
Remarks:	Four lower jaws. Antunes, 1998 contains some information. One distinguishment of this species concerns the first lower premolar (p.136). In contrast to the other species within the genus, it's gone. The full set only retains p2-p4. It's also been a bit of a systematical rambler. Hahn established the species within this genus, Clemens & Jaworowska transferred it to Henkelodon in 1979 but, as it began squeaking out protests, Hahn brought it back to its original home in 1993. Holotype I'm recklessly guessing, but the specific name might just have some connection with the fossil locality. The type fossil, VJ 5-155, is a lower right mandible in the collection of the Museo dos Servicos Geologicas, Lisbon.
Reference:	Hahn (1969), Beiträge zur Fauna der Grube Guimarota Nr. 3. Die Multituberculata. Paleontographica, A, 133, p.1-100.

Species:	Kuehneodon simpsoni Hahn, 1969
Place:	Guimarota
Country:	Portugal
Age:	Kimmeridgian, Upper Jurassic
Remarks:	The following is based upon my reading of Hahn, 1971. An upper I3 incisor from this species is accused of displaying an interesting feature on page 17. There's "very deep wear". The reason that's unusual is because incisors don't generally get involved in heavier dental work. This level of wear suggests it was doing more food processing duties than simply nipping and holding, and this contrasts with normal multi behaviour. Further derived multis developed specialisations for shearing on their premolars, and this meant any mincing ambitions for front teeth would've been surplus to requirements. Upper molars The M1 has two cusp rows; two cusps make up the buccal row and four form its lingual colleague (p.29). The latter row is curved, and its remost member provides a blockage to the valley otherwise running between both rows. Holotype Information is limited to a single upper jaw. This is VJ 112-155 and also lives in the Museum of the Geological Service, Lisbon.
Reference:	Hahn (1969), Beiträge zur Fauna der Grube Guimarota Nr. 3. Die Multituberculata. Paleontographica, A, 133, p.1-100.

Species:	Kuehneodon dryas Hahn G, 1977
Place:	Guimarota
Country:	Portugal
Age:	Kimmeridgian, Upper Jurassic
Remarks:	Known from one upper jaw, (VJ 454-155), which is also housed in Lisbon.
Reference:	Hahn G (1977), Neue Schädel-Reste von Multituberculaten (Mamm.) aus dem Malm Portugals. Geologica et Palaeontologica, 11, p.161-186.

Species:	Kuehneodon uniradiculatus Hahn, 1978
Place:	Guimarota
Country:	Portugal
Age:	Kimmeridgian, Upper Jurassic
Remarks:	Known from four lower jaws. As the first lower premolar of this species is unusual for the genus in being single-rooted (as opposed to double-rooted or absent -Antunes 1998, p.139), I'd imagine that's the inspiration behind the specific name.
Reference:	Neue Unterkiefer von Multituberculaten aus dem Malm Portugals, Geologica et Palaeontologica 12, p.177-212.

Species:	Kuehneodon hahni Antunes MT, 1998
Place:	Lourinha Formation, Porto Novo Subunit
Country:	Portugal
Age:	upper Kimmeridgian-Tithonian, Upper Jurassic
Remarks:	Thanks are due to Kees Hazevoet for the notification. This one evaded my radar. Hopefully fuller information will follow. Actually, the entire paper arrived. The following is based upon my reading of Antunes, 1998. Despite it being a Portuguese journal, the author bravely attempted to wrestle with the notoriously tricky and agile English language. He put up a grand fight but, on occasions, the battle got a bit too hot. Still, the result is far beyond my Portuguese capabilities, and the meaning is usually clear despite some eccentricities. One slip provides a fine addition to the English language. I wish I'd thought of it myself. "I met again W. Kühne in Paris in 1967 and 1973. Our last meeting took place at the very Cuvier's flat on the Jardin des Plantes, whose tenant was then my much regretted friend, Professor Jean-Pierre Lehman." "Much regretted friend" can't have been intended but the phrase is magnificently useful. I've had much regretted friends myself. The main subject of the study's the description of the new multi material but, as a bonus, it also provides some personal memories of Walter Kühne, a somewhat colourful (deep red) researcher of Mesozoic eucynodonts. It contains some interesting background. However, this entry will concentrate on the new species of 'Kühne's tooth'. Fuller background on Kühne can be found in German sources. Hunting Jurassic mammals in Portugal Kühne scoured Iberia (and elsewhere) for Mesozoic mammals, and his greatest material legacy was the fabulously rich locality of Guimarota. However, a site at Pai Mogo near Lourinha somehow managed to evade his dogged detection systems. They evaded everybody's until I. Mateus stuck dino and croc eggshell fragments in 1993 (p.129). Investigation revealed a nest site for theropods. Such remains happen to be rare, and a thorough bout of sediment sieving and washing revealed lots of bones from embryos, some adult teeth and various other bits and pieces. These included a partial left lower jaw of a multituberculate mammal; the first paulchoffie in the country that wasn't from Guimarota. Kuehneodon hahni This is based upon a very small fragment of jaw. Preserved is the broken root of an incisor, evidence for four premolars (only p2-p3 are present), and what's probably a small part of an alveolus for the first molar (p.130). The preserved jaw isn't exactly immense. Going by the sketch, the maximum length is hardly more than four millimetres and the depth beneath the premolars struggles to attain 2.5 (p.132). This latter measurement is nevertheless enough to earn a reward of "distinctly robust". The bone on the labial side has a large mental foramen situated to the fore of the first premolar. Remains of crests on the bone provide information on the positions and proportions of muscles, seeing as that's where some were anchored. As is usual for multis, the masseter and pterygoid were strong and the temporalis weaker. Lower incisor The hole for this tooth attests that it was large but, as only part of the root is available, other details remain unresolved. It may have been procumbent and, looking at the sketch, that appears likely to me to some degree. Antunes doesn't rule out the possibility of it having been 'vertically' aligned as known from Paulchoffatia. However, I think this is a matter of degrees of procumbency according to my understanding of the word. While the incisor of that multi is less towards being horizontal, I wouldn't term it 'vertical'. In any case, x-rays failed to reveal how far into the jaw the root goes, although several factors suggest it gets no further than below the p3 premolar. This is of significance for useful comparative features. A shorter root and more upright stance are basal paulchoffie traits. Longer roots supporting more horizontal inclinations were fashionable for the trendier members, and are strongest expressed by at least some species in this genus. Lower premolars, roots and drawing lines There were four premolars in the lower set but the first and fourth weren't available for interview. At least the alveoli had the decency to turn up, and they showed the teeth were double-rooted. Naturally, they also reveal the positions of each tooth, and that's of some diagnostic significance. As is the case for multis in general, the lower postcanines didn't simply follow on behind one another in an orderly file along the middle of the jaw. The rearmost tooth is positioned close to the lingual side of the bone, whereas the foremost one is closer to the labial edge. The pair of premolars in between are inbetweenies in this sense too. This arrangement allows you to plot a couple of lines and compare them. One's the inner margin of the jaw and the second, the alignment of the postcanine teeth, runs somewhat diagonally apart from that. In comparison to the jaw bone, the teeth are set diagonally across it. As evolutionary developments led to an increase in this relative angle of diagonal divergence, it's diagnostically informative for sorting out your paulchoffies. With this species, the angle is estimated as being between 18.5-20°, and this happens to be a high value. Paulchoffatia delgadoi, for example, manages only 10°. The more derived Kuehneodon dietrichi attains about 20°. The Guimarota fauna provided plenty of lower jaws to measure, and that means this perhaps quirky sounding condition has been extensively tested. Lower premolar details The two examples still present are described as 'birradiculate', a word I'm not familiar with. It seems to mean double-rooted, and the first of the pair is somewhat the stronger. Both crowns are worn and not milk premolars. Although small, this critter was old enough for its adult p2 and p3, and had put them to work. It can be seen that the latter had a shearing blade with at least four serrations, and perhaps a fifth. The p2 measures 0.86mm (length) and 0.78 (width) (p.133). Its larger disciple achieves 1.09 and more then 0.78. The relative sizes of those teeth can also be used for comparisons. Affinities According to the author, this multi is so obviously a paulchoffie that the point requires no further justification. That relatively high angle for with regards to the diagonal arrangement of the premolars across the jaw is only similar to Kuehneodon, and the overall shape of the tooth row is also agreeable with such a conclusion. Taking further factors into consideration: "All in all, the accounted characters are largely enough to justify ascribing the Pai Mogo mandible to the Kuehneodontines and to Kuehneodon, since there are no reasons to do otherwise" (p.136). That genus has several species for which the lower teeth are known. K. guimarotensis had discarded the first of the premolars, and further factors also debar the new fossil from being assigned there. The p1 of K. uniradiculatus was still present but only single-rooted (p.139). The alveoli show it was double-rooted for this critter. That only left H. dietrichi or a new species. Only one point of general morphology was cited to rule out the first option. The gap (diastema) between the incisor and the first premolar is considerable shorter for the new specimen, and that brought about the new species. Size This factor provides support for ruling it out from the Guimarota species. The state of wear on the teeth indicates an adult age for the animal and, in comparison to any of the other known species, it was a bit of a smalling. The length of the lower premolar row will illustrate the point well enough, although it's also reflected in the lengths of the available teeth. The figures for p1 (or its vacant position for K. guimarotensis) to p4 are: K. hahni ca. 3.5mm (the estimate for this species); K. dietrichi 4.75-5.63; K. uniradiculatus 4.82-5.69; K. guimarotensis 4.17-5.0. Holotype Unless I've overlooked something, the sole specimen didn't seem to have a catalogue number when published or, at least, none is mentioned. It lived at the GEAL-Museum in Lourinha, and the specific name honours the contributions of Gerhard Hahn on Jurassic mammals, and particularly Portuguese paulchoffies.
Reference:	Antunes MT (1998), A new Upper Jurassic paulchoffatiid multituberculate (Mammalia) from Pai Mogo, Portugal and a few words on Walter Georg Kühne, Memórias da Academia das Ciências de Lisboa, Classe de Ciências, Colóquio - Colloquium, 37, p.125-153.

Species:	Kuehneodon barcasensis Hahn G & Hahn R, 2001
Place:	Porto das Barcas
Country:	Portugal
Age:	Upper Jurassic
Remarks:
Reference:	Hahn & Hahn (2001), Multituberculaten-zahne aus dem Ober- Jura von Porto das Barcas (Portugal). Paläontologische Zeitschrift, 74 (4), p. 593-586.

Species:	Kuehneodon sp.
Place:	Galve
Country:	Spain
Age:	Barremian, Lower Cretaceous
Remarks:
Reference:

Genus: Meketibolodon Hahn G, 1993 ?'no longer Bolodon' Remarks: "Meketibolodon Hahn 1993 is distinguished from the other genera by two characters: the tooth row is significantly convexly curved upwards, and the Corpus manibulae has angled margins ventrally (Hahn & Hahn 1998b). The incisor is more strongly curved than it is the case in Pauchoffatia, and its root is longer. The Corpus mandibulae is similarly more massive than in Paulchoffatia", (Hahn & Hahn 2000, p.105). This Corpus mandibulae thing is the bit of the lower jaw beneath the tooth row. A handy Greek dictionary offers a translation of meketi as 'no longer'.

Species:	Meketibolodon robustus (Hahn G, 1978)
Place:	Guimarota
Country:	Portugal
Age:	Kimmeridgian, Upper Jurassic
Remarks:	Based on lower jaws; nine specimens.
References:	Hahn G (1978), Neue Unterkiefer von Multituberculaten aus dem Malm Portugals. Geologica et Palaeontologica 12, p.177-212.
	Hahn (1993), The systematic arrangement of the Paulchoffatiidae (Multituberculata) revisited. Geol. Paleontol. 27, p.201-214.

Genus: Meketichoffatia Hahn G, 1993 ?'no longer for Choffat'

Species:	Meketichoffatia krausei Hahn G, 1993
Place:	Guimarota
Country:	Portugal
Age:	Kimmeridgian, Upper Jurassic
Remarks:	Based on two upper jaws. At least one of these is in the Lisbon collection, (VJ 446-155). One specimen, (or perhaps both) was / were originally identified as Pseudobolodon oreas, (see eg. Wible & Rougier 2000, p.96).
Reference:	Hahn (1993), The systematic arrangement of the Paulchoffatiidae (Multituberculata) revisited. Geol. Paleontol. 27, p.201-214.

Genus: Paulchoffatia Kühne WB, 1961 'for Paul Choffat' Aka: Paulchoffia Remarks: "Paulchoffatia Kühne 9161 is characterized by a massive Corpus mandibulae (the part of the jaw below the tooth row), a rounded lower margin of the jaw and a massive, only slightly curved and steeply inclined incisor with a short root", (Hahn & Hahn 2000, p.105). The generic name presumably honours the geologist, Léon Paul Choffat, (1849-1919).

Link: Choffat, Léon Paul, 1849-1919 http://www.snl.ch/dhs/externe/protect/textes/F42653.html A mini-biography, (French).

Species:	Paulchoffatia delgadoi Kühne WB, 1961
Place:	Guimarota
Country:	Portugal
Age:	Kimmeridgian, Upper Jurassic
Remarks:	The following is based largely upon my reading of Hahn, 1971. This study predates the start of the second and more extensive campaign of excavations at Guimarota, and was written when the supply of fossils was much more limited. Lower incisor All known multis have but one lower incisor per jaw half and, typically, it's impressively large and procumbent (p.11). It's large for this species too, but less inclined to procumbency. The crown angles forward with an inclination of around 40°. It also lacks pronounced curvature. The root is short in comparison to the counterparts of other multis, as it only manages to go back in the jaw to the level of the second premolar. Also eccentric is the detail that the root's stored directly beneath the premolars, rather than it being housed more to their lingual side. Holotype The type fossil, V.J. 1-155, is a lower left jaw in the collection of the Meseu dos Servicos Geologicos, Lisbon. Additional notes The skull probably had a length of 2,5cm. This is based on five lower jaw fossils.
Reference:	Kühne (1961), Eine Mammaliafauna aus dem Kimmeridge Portugals. Neues Jahrbuch für Geologie und Paläontologie, Monatshefte, 1961, p.374-381.

Species:	Paulchoffatia sp.
Place:	Galve
Country:	Spain
Age:	Barremian, Lower Cretaceous
Remarks:	Four lower jaws, Paulchoffatia sp., have also been recovered from Guimarota.
Reference:

The Guimarota multituberculates The following is based upon my reading of Hahn & Hahn, 2000. The place Guimarota is on the edge of the city of Leiria, which is near the middle of the Portuguese coast. A former coalmine has ensured the name of Guimarota is known across the world. It might not be familiar to all that many people, but its fan base is nevertheless global. What's so special about a disused coalmine? The answer was found in the coal. This was laid down fairly late during the Jurassic, (Kimmeridgian, between 154 and 151 million years ago.) As it happened to be deposited on top of acid-neutralising limestone, the produce of an ancient swamp turned out to be fossil rich. Even more unusual was that many of the remains were from small, land-living animals including mammals. These treasures are great rarities. Multituberculates The most diverse Guimarota mammals were the multituberculates; a lineage which was then relatively young and finally died out something like 35 million years ago. Apart from a few Middle Jurassic teeth from England, these are the oldest multis yet found. At this juncture, it'd be polite to offer a brief introduction to multis. As this order of mammals has been extinct for a very long time, the large majority of humanity has never heard of them. In ecological terms, they were the equivalents of rodents; mouse to rat-sized animals, which made their living as herbivores-omnivores. Multis have so far been found predominantly in the northern hemisphere, although a few Cretaceous fossils have come from Morocco and Argentina. As is the case at Guimarota, if multis are represented at a location, they tend to be among the most numerous mammals, (p.97), although there are exceptions to this. Portuguese pioneers Most of the multis of Guimarota are members of a family called Paulchoffatiidae. This rather challenging name commemorates Paul Choffat, an early Portuguese geologist. I trust he won't feel offended if I write about paulchoffies instead. It makes the animals sound much more friendly. In contrast to other mammals from this location, the multis have provided no remains beyond some skull material, jaws and isolated teeth. The state of preservation is generally not good. Nevertheless, the fossils suffice to show that paulchoffies were primitive. They lacked some characteristics typically associated with multituberculates. Paulchoffie skulls and jaws Seen from the side, the skull was long and low-roofed. As is common in basal mammals, the vault for the brain wasn't much higher then the level of the snout. There's little sign of anything which could be termed a forehead. The orbits for the eyes were large, and the joint with the skull was set far back. The dentary had a strong upwards pointing coronoid process beginning behind the molars. This was an attachment area for muscles. In contrast to further derived mammals, there was no backwards directed angular process. The lower mandible possessed another primitive touch. As there's a vestigial coronoid bone still in place on the internal surface, (p.98), the dentary formed marginally less than 100% of the lower jaw. In one case, a small septomaxilla may be involved in the upper jaw too, but other researchers interpret this as an artefact caused by damage to the snout. As well as being flat heads, the brain box was relatively short. Paulchoffies were doubtlessly charming mammals, but they were probably not very good at solving riddles. Teeth The teeth of paulchoffies were multicusped as with all multis. That's what the word multituberculate refers to, (p.99). The maximum arsenal per side consisted of: (uppers): three incisors, one canine, five premolars and two molars; (lowers): one, none, four and two respectively. Uppers With regards to the incisors, I1 has only a single cusp and it's a fairly small tooth. I2 has a large cusp at the front and one or more behind it. I3 has a ridge with a pair of cusps, and between nought and three others. Mammalian incisors are generally simple constructions. Complicating the crowns to this extent seems to have been an approach pursued only by multis. The canine and front three premolars mostly have three or four cusps each, but counts of two or five are also known. About the only substantial difference in these teeth is that the canine is single- rather than double-rooted. Such a cuspy canine is an oddity even for multis. These teeth were entirely dispensed with. The rear two premolars are longer teeth. (Length refers to the distance from front to back, and not gum to tip -that would be height.) The P4 and P5 exhibit two to three longitudinal rows of cusps, whereas the molars make do with a pair of them. The presence of three rows on premolars is unusual for multis. Lowers Although there was only a single incisor on the lower jaw, it was impressive. It was long and cone-shaped tooth, which pointed diagonally (or sometimes nearly horizontally) forward, and it was easily the highest in the set. The root was deep to deeper, and stretched back in the jaw until beneath the premolars or even further, (p.100). All multis were proud bearers of incisors like this, and they might well have been invaluable for preening. The lack of any other lower incisors or a canine ensured there was a diastema in front of the first premolar While the lower premolars have two cusp rows, the inner cusps are much enlarged and the external ones are reduced in size. One cutting edge has been favoured. These teeth increase in length from front to back. As multi evolution progressed, the first two fell quickly out of fashion, and the third declined to a pathetic relic. A massively enlarged p4 became a trademark of this order. In later models, it was the only premolar left. The m1 is the longer of the two molars, and it's equipped with a couple of rows of three cusps. The m2 was more or less unicusped on the inside at the front. Remains of further cusps can be found in a kind of rimmed pit on the rest of the tooth. The construction of the m2 is a further paulchoffie eccentricity. Sniffing around in the dark As the orbits were large, it's reasonable to assume that the eyes were. Where known, endocasts show that multis had brains with well-developed features at the front called olfactory bulbs, and these are associated with the sense of smell. Both of these characteristics suggest nocturnal animals. Multi feeding With most multis, there were three basic stages to oral food processing. Goodies were gathered with the incisors, sliced with the premolars and ground up by the millstones of the molars. The muscles provided a backwards to forwards chewing motion, which is the opposite to rodents. However, this wasn't properly developed in paulchoffies. The upper premolars were too wide to have functioned as shears. Although the lowers were organized with one main cutting edge, this was quickly ground down by the colleagues upstairs. Paulchoffies hadn't mastered the shearing abilities of more derived multis. They could gather and grind their food, but an effective shearing process required further refinement. Dental replacement There's no evidence for replacement molars in multis, so it's likely there were permanent teeth as in me, (p.101). (Molar replacement is known from some more basal mammals.) The premolars had an alternating replacement sequence, which is reminiscent of non-mammalian cynodonts. The uppers were substituted in the order P1-P3-P5-P2-P4, and this may help to explain some peculiarities with patterns of wear. There was probably only one episode of replacement for any of the teeth, but this isn't absolutely certain. Diet As the molars exhibit strong wear, they must've been used on hard foodstuff. Armoured insects or parts of plants are candidates. Unlike rodents and some multis, paulchoffie choppers didn't allow for gnawing. The incisors are closed-rooted and coated with enamel on all surfaces. Teeth suitable for gnawing are ever-growing and shelf-sharpening. The enamel is often restricted to the front and, as the softer dentine is eroded behind, a chisel-like upper edge is the inevitable result. The paulchoffie menu must've been centred on foods of a convenient size, as the dentition can't have been used for chopping stuff up. The animals were probably omnivores with a preference towards plants. The missing body As no postcranial remains are presently known, the paulchoffie skeleton can only be guessed at, although later multi fossils provide some assistance. Multis weren't built for endurance running at speed, but short dashes were probably possible, (p.102). If the build of paulchoffies was light, that could explain why none of the body is known. Jaws and teeth are hard parts and relatively abundant at Guimarota. Other local mammals were kind enough to provide a couple of partial skeletons. The complete lack of postcranial material from multis may not simply be the result of blind chance. Hopefully, future finds will bring more information, but none are expected from this location. Excavations stopped in 1982, and the disused mine has been left to the mercy of flood water ever since. A ripe old age? The chances of paulchoffie pensioners weren't high. The molars had low crowns, the enamel wasn't organised into prisms and the preferred diet was hard. Without multiple replacement, (or any replacement for the molars), the teeth wore down quickly, (p.103). Their owners probably couldn't have survived beyond a few years at best. The good news is that paulchoffies may well have led fairly peaceful lives, assuming they reached adulthood and stayed at home during the day. None of the meat-eating mammals were large enough to pose a serious threat, and nightbirds have no counterparts in the fauna. Unless any of the local dinos were partly or fully nocturnal, (which isn't impossible), suitable predators haven't been found. Sex and reproduction As a mammal myself, personal experience suggests that paulchoffies must've been fantastic lovers. Other than for the basics, I've no idea how they reproduced. Some later multis are known to have had a relatively narrow pelvis, (narrower than in living mammals). If they laid eggs, these must've been incredibly small. There's a good possibility they produced poorly developed live young. However, this isn't certain and those fossils come from the Upper Cretaceous. They aren't necessarily informative for more primitive multis from the Jurassic. Problems for paleontologists A dozen paulchoffie genera have been recognised in the fauna, but this figure is probably higher than the actual diversity. With only one exception, it hasn't been possible to match upper and lower jaws or teeth, (p.104). Five genera are based on skull material, four on lower jaws and three are represented by isolated molars. There's a strong chance that four or five of the genera would be unnecessary, if skulls and dentaries could be reliably united, (p.105). As the remains were transported to this last resting place, they probably weren't all residents of the same neighbourhood. Streams and rivers rounded them up post mortem, and unceremoniously dumped the animals into a mass grave. Intimate family business Kuehneodon is the only genus assuredly known from upper and lower jaws, (p.106). As this taxon possesses the least number of family peculiarities, it's somewhat more closely related with the main lineage of multis. For example, the upper premolars have the more usual number of two cusp rows, rather than the paulchoffie eccentricity for three. Consequently, it's been placed in its own subfamily, Kuehneodontinae. All the other Guimarota paulchoffies have been assigned to Paulchoffatiinae. A non-paulchoffie A further Guimarota genus, Proalbionbaatar, is represented by a couple of upper molars. These teeth are smaller and cuspier, and show the characteristics of a different family, Albionbaataridae. They may have been derived from paulchoffies. On a road to nowhere As paulchoffies had developed specialities not found in later multis, (for example, the unusual architecture of the second lower molar), they don't seem to have been the direct ancestors of at least most subsequent relatives. This was a fairly early side line. However, they generally conform to expectations concerning which characteristics basal multis should have possessed.

Genus: Plesiochoffatia Hahn G & Hahn R, 1999 'near Choffatia' Aka: Parachoffatia 'beside Choffatia' (preoccupied) Remarks: With thanks to David Marjanovic. Reference: Hahn & Hahn (1999), Nomenklatorische Notiz: Namens-Änderung bei Multituberculata (Mammalia). Geologica et Palaeontologica, 33, p.156.

Species:	Plesiochoffatia peparethos (Hahn & Hahn, 1998)
Aka:	Parachoffatia peparethos Hahn & Hahn, 1998
Place:	Guimarota
Country:	Portugal
Age:	Kimmeridgian, Upper Jurassic
Remarks:	One upper molar.
Reference:	Hahn & Hahn (1998), Neue Beobachtungen an Plagiaulacoidea (Multituberculata) des Ober Juras. -3. Der Bau der Molaren bei den Paulchoffatiidae. Berliner geowissenschaftliche Abhandlungen E 28, p39-84.

Species:	Plesiochoffatia staphylos (Hahn & Hahn, 1998)
Aka:	Parachoffatia staphylos Hahn & Hahn, 1998
Place:	Guimarota
Country:	Portugal
Age:	Kimmeridgian, Upper Jurassic
Remarks:	Three isolated upper molars.
Reference:	Hahn & Hahn (1998), Neue Beobachtungen an Plagiaulacoidea (Multituberculata) des Ober Juras. -3. Der Bau der Molaren bei den Paulchoffatiidae. Berliner geowissenschaftliche Abhandlungen E 28, p39-84.

Species:	Plesiochoffatia thoas (Hahn & Hahn, 1998)
Aka:	Parachoffatia thoas Hahn & Hahn, 1998
Place:	Guimarota
Country:	Portugal
Age:	Kimmeridgian, Upper Jurassic
Remarks:	More isolated upper molars.
Reference:	Hahn & Hahn (1998), Neue Beobachtungen an Plagiaulacoidea (Multituberculata) des Ober Juras. -3. Der Bau der Molaren bei den Paulchoffatiidae. Berliner geowissenschaftliche Abhandlungen E 28, p39-84.

Genus: Pseudobolodon Hahn G, 1977 'false Bolodon'

Species:	Pseudobolodon krebsi Hahn G & Hahn R, 1994
Place:	Guimarota
Country:	Portugal
Age:	Kimmeridgian, Upper Jurassic
Remarks:	This is based on two upper jaws. As for the septomaxilla mentioned in the title of the reference: "A small, triangular bone is placed between the premaxilla and the maxilla; this bone is interpreted as a septomaxilla. This is a bone that is typical for lower tetrapods (amphibians, reptiles), but is usually absent in mammals", (Hahn & Hahn 2000, p.98). The premaxilla is the front bit of your upper jaw, whilst the maxilla is the side of it. Like most mammals, you don't have this septomaxilla. However, according to other authors, nor did Pseudobolodon. "Our observations of the specimen of Pseudobolodon krebsi (V.J. 451-155) described by Hahn and Hahn (1994) suggest that the bone in question is likely part of the premaxilla and that what is illustrated by these authors (1994: tables 1,2) as a suture between the purported septomaxilla and premaxilla is merely a crack within the premaxilla. Therefore, pending new information, we consider the septomaxilla to be absent in multituberculates", Wible and Rougier 2000, p.76) The specimen they refer to is in the collection of the Museum of the Geological Service, Lisbon.
Reference:	Hahn & Hahn (1994), Nachweis des Septomaxillare bei Pseudobolodon krebsi n.sp. (Multituberculata) aus dem Malm Portugals, p.9-29, in Miscellanea Palaeontologica 3. - 531 p., 67 pl., Berl. Geowiss. Abh. Vol 13, editors Kohring R & Martin T.

Species:	Pseudobolodon oreas Hahn G, 1977
Place:	Guimarota
Country:	Portugal
Age:	Kimmeridgian, Upper Jurassic
Remarks:	Seven upper jaws.
Reference:	Hahn G (1977), Neue Schädel-Reste von Multituberculaten (Mamm.) aus dem Malm Portugals. Geologica et Palaeontologica, 11, p.161-186.

Species:	Pseudobolodon robustus Hahn, 1978
Remarks:	I’m assuming this is now Meketibolodon robustus (Hahn G, 1978). It's not listed in Hahn and Hahn 2000.

Genus: Renatodon 'Renata's tooth' Reference:

Species:	Renatodon amalthea
Place:	Guimarota
Country:	Portugal
Age:	Kimmeridgian, Upper Jurassic
Remarks:	My limited information on this genus has been gleaned from reading Kielan-Jaworowska et al, 2005, p.509-510, a study about a completely different multi. It's amazing what you can find when you're not looking for it. Prior to that study I'd never heard of this animal. The reference given wasn't mentioned specifically as a citation, but I assume that's most likely what it nevertheless is. Should anybody have a copy then please feel invited to provide clarity. You're welcome to send a copy as well. The paper is Hahn G (2001), Neue Beobachtungen an Schädel-Resten von Paulchoffatiidae (Multituberculata; Ober-Jura), Geologica et Paleontologica, 35, p.121-143. K-J & Co report that the 'anterior zygomatic ridge' of Renatodon has been photographed in plate 3.2. This tells us that part must be known and, therefore, material includes at least some of the skull. Page 510 mentions: "other paulchoffatiid genera discussed above". One was Kielanodon and the only other available candidate happens to be Renatodon. Logic suggests this must be a paulchoffie. Further info would be welcome.
Reference:

Genus: Sunnyodon Kielan-Jaworowska Z & Ensom PC, 1992 'Sunny tooth' Remarks: The name honours Sunnydown Farm, Langton Matravers, which is where the holotype was found.

Species:	Sunnyodon notleyi Kielan-Jaworowska Z & Ensom PC, 1992
Place:	Purbeck Limestone Group, Dorset
Country:	England
Age:	Berriasian, Lower Cretaceous
Remarks:	The following is based upon my reading of Kielan-Joworowska & Ensom, 1992. The sole specimen definitely referred to the genus is an upper premolar, which is probably a P5. A couple of other premolars may also belong. The holotype has similarities with both the P4 and P5 of Paulchoffatia, Kuehneodon and Kielanodon (p.107). As with those taxa there are two cusp rows. In the case of Sunny the formula is 2:4 (buccal-lingual). Both buccal cusps are situated towards the middle of the crown, and that's a contrast to the aforementioned genera. Also different from all other paulchoffies is a discernable incipient ridge lingual of both cusp rows. The constellation of the cusps is most like that known from Kielanodon, but this tooth is narrower and more oval in occlusal outline. The tooth is also comparatively small (length 0.94mm, width 0.67-0.87). Somewhat lingually of the buccal cusps are small cuspules, with one at the front and another at the rear. The cusps of the lingual row increase in size from front to back. A novel ridge The most obvious novelty is the lingual ridge towards the rear, and a similar feature adorns the M1 molar of plagiaulacids and relatives. Nevertheless, the cusps are more in keeping with premolars due to their conical shape and associated ridges. While the lingual ridge hadn't previously been seen on paulchoffie upper premolars, there is a precautionary proviso to bear in mind. Typically, these teeth are worn in that area, and this habit could have obscured the picture. The overall shape and size aren't suitable for Gerhardodon (which was subsequently referred to Pinheirodontidae), a somewhat larger Purbeckian multi. The possibility that this tooth could be a P4 also couldn't be eliminated. Holotype The type fossil, DORCM GS 18, is an upper premolar being held by kidnappers in the Dorset County Museum, Dorchester. The specific name honours Mr and Mrs RF Notley, who own the quarry at Sunnydown Farm. Additional notes - Age According to current understanding, this formation is usually considered early Lower Cretaceous. Paul Ensom has more recently written a highly readable booklet on the geology of Dorset, aimed at interested non-specialists, (see bibliography). It’s a short and informative read for non-specialists.
Reference:	Kielan-Jaworowska & Ensom (1992), Multituberculate Mammals from the Upper Jurassic Purbeck Limestone Formation of southern England. Palaeontology, 35, p.95-126.

Link: Palaeontology, 35 http://palaeontology.palass-pubs.org/pdf/Vol%2035/Pages%2095-126.pdf Kielan-Jaworowska & Ensom, 1992 is presently freely accessible in pdf format.

Bornholm, Denmark A multi upper premolar has been found in similarly aged rock on the island of Bornholm, (Skyttegård Member, Rabekke Formation, Nyker Group). It's been tentatively referred to Sunnyodon, and is Scandanavia's first contribution to Mesozoic mammal fossils. Lindgren J, Rees J, Siverson M & Cuny G (2004), The first Mesozoic mammal from Scandanavia, GFF, 126(4), p.325-330. Lund University paleontologists from Sweden have been invading a neighbouring island. They've kidnapped a 145 million year old multi tooth. Thanks are due to Jan Van Dijk for posting the notification on the Dinosaur Mailing List, (15.10.2004). Link: Lunds Universitet, Uråldrigt däggdjursfossil http://www.lu.se/o.o.i.s/1383?visa=pm&pm_id=186 I'm afraid I failed to pay due attention during my Swedish classes. Fortunately, Friend Lothar of Stuttgart has also been to Bornholm in 2004. Footprints suggest that a sauropod and an ankylosaur got there many years before him. A dromaeosaur tooth has also been found. Apparently, these finds have raised discussions concerning tourism. Presently, Bornholm highlights its charms by being referred to as the 'Sun Island'. Some people feel 'Dinosaur Island' would be more alluring. This would be silly. 'Multituberculate Island' would clearly be superior. Link: Bornholm Info http://www.bornholm.info/?langid=2 Welcome to the Multituberculate Island.

Genus: Xenachoffatia Hahn G & Hahn R, 1998 'for Xena Choffat'

Species:	Xenachoffatia oinopion Hahn & Hahn, 1998
Place:	Guimarota
Country:	Portugal
Age:	Kimmeridgian, Upper Jurassic
Remarks:	Based on three upper molars. With thanks to David Marjanovic.
Reference:	Hahn & Hahn (1998), Neue Beobachtungen an Plagiaulacoidea (Multituberculata) des Ober-Juras. 3. Der Bau der Molaren bei den Paulchoffatiidae. Berliner Geowissenschaftliche Abhandlungen, E, 28, p.39-84.

Other reports: Xxxxxxxxxxxxxx Xxxxxxxxxxxxxx

A. Middle Jurassic Multitubercuates B. ‘Basal’ Multituberculata C. Paulchoffatiidae & Hahnodontidae D. Pinheirodontidae