Comments by Mickey Mortimer

Etymology

(?)"gigantic Achilles hero" from achillis, for the calcaneal tendon and Mongolian bator, which means hero.

holotype (FR.MNUFR-15) (4.6-6.6 m) maxilla (290.8 mm), nine teeth (to 38 mm, FABL 17 mm), sixth cervical vertebra (34.6 mm), tenth cervical vertebra (51.3 mm), fourth?dorsal vertebra (53.8 mm), posterior dorsal vertebra (49.5 mm), anterior dorsal rib, posterior dorsal rib, mid-caudal vertebra, six distal caudal vertebrae, three chevrons, scapula, coracoid, radius (260 mm), metacarpal III (71 mm), phalanx I-1 (78 mm), manual ungual (90.5, 112 curve), ilium (531 mm), pubes (548 mm), ischium (378 mm), femur (505 mm), tibia (490.4 mm), phalanx II-2 (56.4 mm), pedal ungual II, metatarsal III (234.4 mm), phalanx III-2 (55 mm), metatarsal IV (209.6 mm)

Diagnosis

Hypapophyses absent in cervicodorsal vertebrae; very short anterior dorsal centra; two pairs of pleurocoels in posterior dorsal centra; pleurocoel-like foramina on caudal vertebrae; mid-caudal chevrons with sinuous ventral margin; manual elements very robust; sinuous ridges present on the lateral surface of the ilium, above the peduncles and acetabulum; anterior pubic foot slightly longer than posterior foot; metatarsal IV with distal lateral condyle strongly reduced; very stout pedal phalanx II-2.

Description

This dinosaur was described in an obscure Mongolian journal before it's time. Norell and Clark originally intended to publish it in American Museum Novitates with a comparative analysis. Unfortunately, it was released in an extremely preliminary form in Contributions of the Mongolian-American Paleontological Project without their knowledge. Indeed, the section "Habits and affinities of dromaeosaurian dinosaurs" was neither written nor seen by Norell and Clark prior to publication. The holotype was found as an associated but disarticulated specimen. No other dinosaurs were found in the area, so it is unlikely this is a chimaera as some (Burnham et al. 2000) have supposed.

Achillobator was about 4.6-6.6 meters long, based on comparison of several elements with Deinonychus. I estimate 5 meters is a believable length based on the shorter vertebral centra. For those of you wondering how this compares to other giant dromaeosaurids, Utahraptor was about the same size based on tibial length, while Megaraptor and "Kitadanisaurus" were about 28% larger based on lower arm length and metatarsal length respectively. Vertebral sutures are completely fused, so this is probably an adult.

For those wishing to illustrate Achillobator, the skull was probably very similar to Dromaeosaurus, while the neck, back and tail were shorter than Deinonychus, as were the arms, lower legs and pes. The propubic pelvis would have made the torso differently shaped and the sickle claw was smaller. The entire animal was much more robust.

The skull is represented by a maxilla and several teeth. The maxilla is deeper than Deinonychus and is missing most of the nasal process. Compared to Deinonychus, the nasal process projects more vertically, the premaxillary suture is more vertical and the jugal process is more ventrally projected. The antorbital fossa appears more recessed and the teeth are more widely spaced. The internal structure of the maxilla is partially known, as there is a medial excavation of the nasal process by a two-chambered sinus that opens laterally to form the maxillary fenestra. There are eleven alveoli and the interdental plates appear to be fused. The teeth are recurved and serrated, with anterior serrations being slightly smaller than posterior serrations (17-20 per 5 mm vs. 15-18 per 5 mm).

The sixth or seventh cervical is preserved. It's centrum is concave in front and taller than wide anteriorly. In addition, there are prominent epipophyses and the ribs are unfused. Compared to Deinonychus, the neural spine is transversely narrower and directed posteriorly, the neural canal is larger, the parapophyses are more pronounced, the diapophyses are smaller, the posterior articular surface is concave dorsally and the anterior articular surface is taller. The interspinal ligament scars extend futher dorsally than in Deinonychus as well.

The tenth cervical vertebra has an amphiplatyan centrum with deep pleurocoels. It is slightly wider than tall anteriorly and slightly taller than wide posteriorly. There is a broad keel ventrally and hyposphene-hypantra articulations are present.

A few dorsal vertebrae are known including a mid-dorsal (possibly fourth) and a posterior dorsal. The centra are amphicoelous with deep pleurocoels, double in posterior dorsals. Anterior pleurocoels are larger when there are two pairs. The articular surfaces are slightly higher than wide in the mid-dorsal. The ventral surface is slightly keeled, but there is no hypapophysis. Hyposphene-hypantrum articulations are present and the vertebral foramen is about 23% of centrum height. Other differences from Deinonychus include shorter centra and almost vertically directed prezygopophyses.

The text states only one rib was recovered, but two are figured. The rib the text mentions is supposed to be either the last cervical or first dorsal, but the two pictured come from the mid-anterior and posterior portion of the vertebral column. They are similar to Deinonychus, but exact comparison is difficult without having ribs from the same vertebrae to compare to.

There are seven caudal vertebrae preserved, one from the middle of the series, the rest from the distal tip. They are amphicoelous and the distal ones lack neural spines and transverse processes. There is a "pleurocoel-like foramen" on the basal surface of the transverse process that probably connects to the neural canal. Another foramen is also present, exiting from the anteroventral base of transverse process next to the pleurocoel-like foramen. Elongate prezygopophyses are present, although the postzygopophyses lack elongate rods. The prezygopophyses bifurcate more distally than Deinonychus, if at all. The neural spine is more prominent than Deinonychus, extending past the postzygopophyseal articular surfaces, and the centra are shorter.

Three mid-caudal and distal chevrons are present in the holotype. They are very short dorsoventrally, but have elongate anterior and posterior processes. The anterior processes are dromaeosaurid-like in their elongation, but must bifurcate more distally than Deinonychus. Other differences from Deinonychus are the sigmoideal ventral outline, more widely spaced articular facets and trifurcate anterior process.

A scapulocoracoid is preserved missing only the distal end. It is very similar to Deinonychus, with a very shallow acromial expansion and slender shaft. It narrows distally a bit, unlike Deinonychus. The coracoid is broadly similar to Deinonychus, being elongate with a prominent coracoid tubercle and foramen. The anteroventral surface is not as projecting as Deinonychus and the posterior process is much larger and triangular. The glenoid faces posteroventrally.

A radius is illustrated, but not described. It has a more expanded distal end than Deinonychus and the proximal end is expanded more gradually. It resembles dromaeosaurids in being longer than the metatarsus.

There are four manual elements preserved. The third metacarpal is stouter than in Deinonychus and bowed dorsally, but not laterally. Both manual phalanx I-I and manual ungual I are present, although the latter is not figured. The phalanx differs from Deinonychus in it's stoutness and small details of articulations, but is very similar in shape dorsally. Manual ungual I is reported to be laterally compressed, recurved and have a large flexor tubercle.

The ilium is very unique in structure. It is very tall with a short preacetabular process and longer postacetabular process. The preacetabular process is like a dorsoventrally expanded version of Deinonychus', with a posteroventrally slanted concave anterior edge. The dorsal margin is slightly convex and slopes ventrally over the postacetabular process. The postacetabular process is quite tall, extends ventrally past the ischial peduncle and has a vertical posterior margin, with a posterodorsal and a posteroventral tubercle. The pubic peduncle is nearly vertical and has a concave ventral edge facing slightly posteriorly. The ischial peduncle is reduced, perhaps with a prominent antitrochantor, and the acetabulum is partially closed off medially. Several sinuous ridges are present on the lateral surface above the peduncles and acetabulum. In addition, there are many other ridges and striations for various muscles on different areas of the medial and lateral surfaces.

The pubes are well-preserved and distict from other dromaeosaurids. I estimate they were projected ventrally or slightly posteriorly. They are straight with the proximal end expanded a bit anteriorly and a small obturator notch. There is no pectineal process and the shafts are circular in cross section. The distal foot is slightly longer anteriorly than posteriorly with a pointed anterior foot and a convex ventral edge. The two pubes are joined for 69% of their length and have straight lateral margins the whole way down, unlike the narrow foot of Deinonychus.

The ischium has a more proximally placed obturator process than Deinonychus (18% down shaft) and is longer compared to the pubis (69%). The pubic peduncle is narrower, the obturator process longer and the distal end is blunt. There is a small proximodorsal process. The two ischia were not fused, but may have had a mobile articulation.

The femur is moderately bowed anteriorly and round in cross section. The head is slightly declined and separated from the greater trochantor by a moderately depressed surface, while the lesser trochantor is small and barely separated from the greater trochantor. The posterior trochantor and a distally placed (~40% down shaft) fourth trochantor are also present. The anterodistal fossa is absent and unlike Deinonychus, the tibial condyle is much longer than the fibular condyle.

The tibia is 97% of femoral length and similar to Deinonychus in most respects, although stouter. The cnemial crest is much larger and directed more dorsally and the fibular crest is very proximally placed. What may be the facet for the astragalar ascending process is 29% of tibial length and expanded lateromedially.

Metatarsals III and IV are known. Metatarsal III is 46% of femoral length and clearly not artometatarsalian. The proximal end of metatarsal III is compressed transversely, but not nearly as much as in Deinonychus . Unlike Deinonychus, the distal end of metatarsal IV is ginglymoideal, with a very small lateral condyle.

Several pedal phalanges are mentioned, but only phalanx II-2 and ungual II are figured. Phalanx II-2 is similar to dromaeosaurids in that it has a large proximoventral heel, but it is much stouter than Deinonychus and even Adasaurus. The second pedal ungual is small compared to Deinonychus (151% of II-2) and differs in it's much larger flexor tubercle and proximally straight lateral groove. It approaches the manual unguals of Deinonychus more in these regards and I would not be surprised if it were a manual ungual.

Relationships

This theropod has many dromaeosaur-like characters, but is plesiomorphic in other regards. Because of this, it has been involved in discussions of dromaeosaurid paraphyly and seen as a chimaera. Burnham et al. (2000) state that the maxilla, ilium, ischium and caudal vertebrae share no unique characters with dromaeosaurids. This is obviously not true, as can be seen from the above description. It is obviously dromaeosaurid based on- fused interdental plates; elongate caudal prezygopophyses; elongate anterior chevron processes; concave anterior ilial edge that slants posteroventrally; elongate proximoventral heel on pedal phalanx II-2 that articulates with phalanx II-1. The only comment regarding it's placement within the Dromaeosauridae in the description is in the abstract, where the authors state it is most closely related to Dromaeosaurus. It has already been compared extensively to Deinonychus. It will now be compared to other dromaeosaurids.

Comparison to Dromaeosaurus is limited, but important. The maxilla is nearly identical in shape and the teeth share low DSDI ratios, unlike velociraptorines (although the anterior serrations in Dromaeosaurus are slightly larger than posterior serrations, the reverse of Achillobator). The number of maxillary teeth in Achillobator (11) is closer to Dromaeosaurus (9) than Deinonychus (16). Both Achillobator and Dromaeosaurus have stout pedal phalanx II-2, but it is much more stout in Achillobator.

The pelvis of Adasaurus is much more similar to Deinonychus than Achillobator. The few ways in which it is more similar to Achillobator include the more concave anterior edge, posteroventral tubercle and postacetabular process extending below the ischial peduncle. The pubis and ischium differ in the ways they do from Deinonychus. Both Achillobator and Adasaurus have fourth trochantors, but Achillobator's is more distally placed. The second pedal phalanges of digit II are the stoutest among dromaeosaurids in these two genera and the second pedal ungual is smallest (if it's not a manual ungual in Achillobator). Achillobator also has a large flexor tubercle on this ungual, unlike Adasaurus.

Utahraptor has few comparable elements. It's teeth have low DSDI ratios, although the anterior denticles are larger, like Dromaeosaurus. The caudals are similar to Deinonychus in their elongation, postzygopophyseal rods and reduced neural spine, but don't bifurcate until further distally like Achillobator. The tibia of Utahraptor is much more robust, with a more distally placed fibular crest and smaller cnemial crest like Deinonychus. The second pedal ungual has a smaller flexor tubercle like Deinonychus.

Even Megaraptor has a more gracile manual phalanx I-1 than Achillobator. Metatarsal III is also much more slender in this genus, with more extensive distal articular surfaces. Pedal ungual II has a smaller flexor tubercle and is longer and less recurved.

Velociraptor is quite different from Achillobator. The maxillary differences between Deinonychus and Achillobator are carried to an extreme in Velociraptor. Velociraptor differs from Achillobator in dorsal vertebral morphology based on the tall neural canal, longer centra, prominent hypapophyses and lack of posterior dorsal pleurocoels. The scapulocoracoid is different from both Achillobator and Deinonychus based on the laterally directed glenoid, prominent acromion process and more elongate twisted coracoid, and has a small posterior coracoid process like Deinonychus. Metacarpal III is slender, like Deinonychus, but straight in dorsal view, like Achillobator. Manual phalanx I-1 is more similar to Deinonychus. Velociraptor and Deinonychus are far more similar to each other in pelvic morphology than either is to Achillobator. Both Achillobator and Velociraptor have fourth trochantors, unlike some Deinonychus specimens. Velociraptor lacks a ginglymus on metatarsal IV, like Deinonychus , and the other pedal elements are similar to Deinonychus in the respects that they differ from Achillobator.

Saurornitholestes differs in the high DSDI ratio of it's teeth. The dorsal centra are longer and although only one pleurocoel is present per side, two openings are present inside each on posterior dorsals. Like Achillobator, the anterior is larger. The partial manual elements seem closer to Deinonychus than Achillobator. Referred materal includes a maxilla, which is similar in shape to Achillobator, but with a larger antorbital fossa. Pedal elements are also referred, which exhibit slender phalanx II-2 and enlarged second unguals with small flexor tubercles.

Bambiraptor has a much larger antorbital fossa with lateral sculpturing, but the maxilla is similar in shape. The teeth have high DSDI ratios, but are similar in number to Achillobator (9). The scapulocoracoid is very different with a laterally facing glenoid, prominent acromion process, twisted coracoid with a narrow neck and small posterior process. The manual elements are slender, like Deinonychus. Bambiraptor's pelvis is more similar to Deinonychus than Achillobator, additional differences from Achillobator including the hooked preacetabular process with convex anterior edge, absent anterior pubic foot and even more distally placed obturator process. Two similarities to Achillobator that Deinonychus lacks are a proximodorsal ischial process and medially reduced acetabulum. Unlike Achillobator, the femur lacks a fourth trochantor and the tibia has a smaller cnemial crest. The pedal elements are closer to Deinonychus based on the gracile metatarsi, elongate phalanx II-2 and small flexor tubercle on ungual II.

Sinornithosaurus has an extremely large antorbital fossa with extensive sculpturing, small premaxillo-maxillary suture, large DSDI ratio and much more elongate maxilla than Achillobator, but the number of teeth is identical (11). The pectoral girdle differs in the same aspects that Velociraptor does, but has a large triangular posterior coracoid process like Achillobator. The manual elements are slender. The pelvis resembles Achillobator more than Deinonychus based on the medially reduced acetabulum, proximodorsal ischial process and postacetabular process that extends ventral to the ischial peduncle. It is more similar to Deinonychus in other features and additionally differs from Achillobator in it's lack of a well-differentiated pubic foot, even shorter ischium with a more distally placed obturator process and mid-dorsal ischial process. The pes is more similar to Deinonychus, with the proximal end of metatarsal III reduced even further and elongate phalanx II-2.

Unenlagia differs because of it's single posterior dorsal pleurocoels, longer dorsal centra, laterally facing glenoid and prominent acromion process. The ilium shares some important features such as being very deep, having posterodorsal and posteroventral tubercles and a medially reduced acetabulum. Differences include a more anteriorly projecting preacetabular process, shorter shallower postacetabular process and more posteriorly projecting pubic peduncle. Other pelvic similarities are the small proximodorsal ischial process and vertical pubis. Unenlagia lacks an anterior pubic foot and obturator notch, has a much shorter ischium with a more distally placed obturator process and has pubes that taper distally in anterior view. Also, the femur lacks a fourth trochantor.

Variraptor has large hypapophyses on anterior dorsals and double pleurocoels on anterior centra, but single ones on posterior dorsals, the opposite of Achillobator. It is interesting to note the double pleurocoels of Variraptor are lined up vertically, unlike those of Achillobator. Variraptor also has more dorsally projecting prezygopophyses like Achillobator. Although the anterior dorsal centra of Achillobator are shorter than Variraptor, the posterior centra are longer.

Pyroraptor differs in it's large DSDI ratio, elongate pedal phalanx II-2 and pedal ungual II with small flexor tubercle.

As can be seen, Achillobator shares several characters with various dromaeosauids. There is a very complex character distribution which makes inferring phylogenetic relationships difficult. The eumaniraptoran section of my cladogram is currently in flux, with alvarezsaurids, Avimimus + Yandangornis + Pygostylia, troodontids, Rahonavis, Archaeopteryx and all dromaeosaurids in a polytomy. I have yet to add many of the characters mentioned here and in my Adasaurus post, so I expect resolution to occur once my number of characters is increased. As a preliminary analysis, I made a matrix with all dromaeosaurids mentioned here except Megaraptor, Variraptor and Pyroraptor (plus Troodontidae, Alvarezsauridae, Archaeopteryx and Rahonavis as outgroups) and 28 characters mentioned in this post. To fully resolve the cladogram, I had to remove Dromaeosaurus, Saurornitholestes and Utahraptor. What's important for this post is that the position of Achillobator was fully resolved in all analyses. It comes out as the sister group to Unenlagia, then to Adasaurus. Dromaeosaurus is also in this latter clade. Synapomorphies shared by at least some of these four taxa are: postacetabular process extends below ischial peduncle (also in Sinornithosaurus); posteroventral ilial tubercle; stout pedal phalanx II-2; second pedal ungual not enlarged. Synapomorphies shared by Achillobator and Unenlagia are: pelvis not strongly opisthopubic; very deep ilium; posterodorsal ilial tubercle; medially reduced acetabulum (also in Bambiraptor); proximodorsal ischial process (also in Bambiraptor and Sinornithosaurus). When lesser known dromaeosaurids were included in the fully resolved analysis one at a time, neither Utahraptor nor Saurornitholestes came out in the "dromaeosaurine" clade, although Variraptor grouped with Achillobator based on the short posterior dorsal centra. This is of course only preliminary, but it suggests that Achillobator may have secondarily lost several features (low DSDI ratio, opisthopubic pelvis, lack of anterior pubic foot, etc.) as opposed to being more basal than other dromaeosaurids. I conclude that Achillobator is a dromaeosaurid possibly most closely related to Unenlagia, and less closely related to Adasaurus and Dromaeosaurus.

Reference

Perle, A., Norell, M., and Clark, J. 1999. A new maniraptoran Theropod- Achillobator giganticus (Dromaeosauridae)- from the Upper Cretaceous of Burkhant, Mongolia. Contribution no. 101 of the Mongolian-American Paleontological Project. pp 1-105