Dromaeosaurs are Terrestrial Hawks

Denver Fowler and colleagues have just published a series of papers dealing with the reconstruction of predatory behavior as indicated by the proportions, curvature, and anatomy of the pes in theropod dinosaurs. They began this study investigating birds, and the range of ecology and behavior exhibited by a variety of birds. Then they expanded this to that ever-curious group, dromaeosaurs. That is the topic of the current paper, by Denver Fowler, Elizabeth Freedman, John Scannella and Robert Kambic, who describe the pedal anatomy of Deinonychus antirrhopus in relation to its possible predatory capabilities, including the premises of previous authors who’ve inferred the foot was used in climbing (including up the sides of very, very large prey).

I know I’ve had my moments of hyper argument when it comes to talking to Denver, but I have to say, this is a great paper. One of the interesting things about it is that it is written in a much more personal manner, including phrases like “vigorous retaliations” which, I would admit, I wouldn’t write into a paper in the first place and which doesn’t really appear commonplace in technical articles.

Despite this, though, the meat of the paper is hefty and well done.

To Have and Hold

They’ve taken aspects of the pedal anatomy and pushed the model of the grasping foot as accipitrid-like predatory foot to a strong degree. With physical manipulation and digital modelling, they’ve drawn from the work of Senter (2009) on attempting to find a biomechanical function that agrees with previous hypotheses, including a “digging” model used to help explain the incredibly large discrepancy in size between the unguals of the second and those of the third or fourth digits. The studies show that virtually all toes (excluding possibly the hallux) could flex tight enough for the unguals, especially including the keratinous sheath, to touch the palmar surface of the foot. This grants an extensive range of flexion in the foot. Senter (2006) also assessed the range of manus flexion and extension in dromaeosaurids, and Senter (2009) the range of possible extension of the pedal digits, finding that Deinonychus antirrhopus could curl its manual digits around in greater than 180° arc, but was unable to extend the digits by much at all, while in the pes the digits were extremely extensible and to similar degrees of motion, making it possible for all unguals to extend to the point the outer curve of the ungual could almost contact the metatarsus.

Deinonychosaur pes ranges of motion. Modified from Senter (2009), where I oriented the metatarsals to roughly the same degree. Ungual phalanges are missing in UA 5686, some nonungual phalanges are restroed from plaster in YPM 5205, and the pdII-3 ungual in the latter is considered deformed somewhat by Fowler et al. (2011), noting that Ostrom (1976) described a less-recurved pdII-3u which can substantially alter the range of "curling" or the taxon's position on the cursorial-arboreal gradient (Feduccia, 1993; see criticisms by Glen & Bennett, 2007). Specimens are not to scale.

As should be noted above, the curling of the second pedal digit (pd2) is shown by Senter (2009) to allow the pedal unguals to potentially contact the pes in two deinonychosaurs: Rahonavis ostromi (Forster et al., 1998) and Bambiraptor feinbergi (Burnham et al., 2000) [n1]; manipulation by Fowler et al. (2011) argues that the pes of MOR 747 (referred of Deinonychus antirrhopus Ostrom, 1969) should be able to curl so that virtually all of the toes will be able to contact the flexor surface of the metatarsus (or plantar surface of the pes). Senter (2009) argue that the pedal digit 2, while capable of high “curling”, was not so capable as that of the smaller Rahonavis ostromi (a basal deinonychosaur) or Bambiraptor feinbergi (represented by a juvenile specimen). A portion of this argument is centered not around pd2, but pd1, the hallux. Fowler et al. (2009) argue that the hallux is oriented somewhat medially in virtually all tetanurans, even in taxa with clearly cursorial components to their limbs, while this orientation merely becomes more medial in more avian-like taxa. They cite Norell & Makovicky (1997, 1998), who demonstrated with articulated pes of Velociraptor mongoliensis that the hallux is distally placed, has a distal offset to the articulation for pd1, and that the ungual is oriented medially when the articulations are considered. However, while I do not disagree with Fowler et al. (2011) that MTI and therefore pd1 are mediallt offset from the main metatarsus, there are problems associated with their articulation that renders the degree of offset problematic:

Pes of Velociraptor mongoliensis (Osborn, 1924), and Deinonychus antirrhopus (Ostrom, 1969). A (GIN 100/985) and B (GIN 100/986) modified after Norell & Makovicky (1997, 1998) respectively, and C (MOR 747) after Fowler et al. (2011). Scales bars in the original, representing 2cm, 1cm, and 5cm, respectively. White arrows in A and B point to MTI and its articulation with MTII (in A, it is closely associated, in B it is separated). Note the gap between the distal end of MTI in C, while the metatarsal is closely appressed otherwise. Yellow arrows indicate the angle (θ) by which MTI and MTII diverge; it is lowest in A, and higher in B and C.

The differing morphologies between the metatarsals I of all specimens above may make the similarities problematic; it implies, rather, that the relative angle of diverge is taxonomically signification, or individually variable, which may impact the utility of arguing the orientation of the hallux to MTII problematic. Nonetheless, a substantial amount of medial orientation is present in all taxa, and this supports the view that underestimation of the divergence has resulted in problems facing correctly evaluating hallucial relationships of these larger-bodied near-birds.

Fowler et al. (2011), in the new study, extend Senter’s premise to prey acquisition, applying work began by Fowler et al. (2009) on interdigital variation in raptorial birds (owls, hawks, falcons, ospreys, kites, eagles and vultures). They use a direct model of raptorial prey-handling behavior called mantling, which occurs after acquisition and is used during the rendering process when a raptor spreads its wings over the prey. Rendering involves the use of the feet gripping the prey while the beak pulls portions off the corpse. The purpose of mantling appears multiple, where Fowler et al. (2011) indicate it can be protective of the prey from possible invaders or to corral the prey; the actual function is variable, and may not be fully understood, and mantling can occur during threat displays rather than during feeding, and captive birds can exhibit it when stressed or malnourished (Fox, 1995).

Little is given to the discussion of the purpose of functional concerns for the manus in this prey acquisition style. The authors write:

Under the RPR model, if the same strategy was employed by paravians subduing prey, then the large manual unguals may have been used to pull escaping prey back under the feet of the predator in a raking action. This reconstruction lowers the hands to be used near the feet, consistent with the orientation of the palms while in this posture.
[Fowler et al., 2011:pg.9]

This model should be true regardless of the size of the manus, that the strength of the digits to restrain the movements of prey is only relative to the size or strength of the prey. So if the prey is tiny, smaller claws would be necessary, and indeed the authors consider the effect when dealing with basal deinonychosaurs like Tianyuraptor ostromi (Zheng et al., 2009), which has particularly short forelimbs limbs which would prevent equivalent “mantling” when grasping prey with the feet. If the prey were larger, the manus might be useful. But the problem here lies in the fact that no bird exhibits such behavior with the wings, as no bird has large manus claws with which to help keep prey restrained. I wouldn’t say this is a big problem, but it tacks the presence of manus claws into the equation of foot-based prey holding without leading to the question of retaining large manus claws at all. Part of the thesis of the paper is an emphasis of a transition in prey-handling from the manus to the foot. Despite this, smaller deinonychosaurs (e.g., Rahonavis ostromi [Forster et al., 1998] Senter, 2009) have similar ranges of motion in the pes as larger, and are typically considered more basal taxa than the larger and thus more derived Deinonychus antirrhopus. This is including the fact that the wings are clearly much larger in Rahonavis ostromi than in any other known deinonychosaur, while other basal deinonychosaurs (including the aforementioned Tianyuraptor ostromi) have shorter arms (this is also true of some unenlagiine dromaeosaurs, where Rahonavis ostromi is the only black sheep, especially if it doesn’t belong to this group).

Holotype of Tianyuraptor ostromi, STM1-3. From Zheng et al. (2010). Borrowed from Brian Switek's Dinosaur Tracking Blog.

Bite Me (Redux)

Fowler et al. (2011) discuss the elements of prey acquisition, but also of processing, especially in the manner in which raptors consume their prey. Because dromaeosaurs lack a beak (a subject I discussed here) their processing must be done with teeth (a thing I’ve been told that they have). Dromaeosaur teeth are characterized by particularly large denticles on the distal carina as well as occasionally having apical hooking of the denticles, a feature that varies in the degree of “hookedness”. They can be either extremely sloped in some teeth, essentially angled towards the crown apex, or outward (Currie et al., 1990). Denticles are useful for many purposes: As denticles, they produce both a point along the crown where pressure from the crown as a whole is imparted, focused toward small separate portions rather than as a whole if the carina were not a single ridge; they also provide slots (or kerfs) between which tissues such as ligament or muscle fibers may be caught, increasing the tearing and thus rendering power of the tooth (Abler, 1992, 2001). Shape of the denticles varies, with some denticles having a largely rounded aspect, others with a sort of tilted-peak, and still others more triangular than squared as well as the inverse. Most denticles are typically hooked, having the aspect of a square with one corner extended, and this corner is always the apical edge of the denticle; thus, the denticles always have a peak that points apically, even when the angle varies (Currie et al., 1990).

Teeth and denticles of Saurornitholestes langstoni (Sues, 1978), SEM closeup of distal denticles of RTMP 80.16.996 and drawing of the holotype, RTMP 74.10.5. Modified after Currie et al., 1990.

Fowler et al. argue that the hooking of the denticles (which as can be seen above point outward in a general 45° angle from the angle of the crown of this tooth) is related to the posture of the head, which when in mantling behavior would be nearly or sub-vertical, nose pointed down and between the legs. Thus the angle of the teeth would be horizontal and create pressure using gravity to assist in tearing of the flesh; in this manner, the denticles would be pointed not posteriorly but close to vertically, as they are generally oriented toward the back of the mouth, now upward. Such a model proposes that gravity would then pull engaged flesh downward against the denticles. Fowler et al. support this model by noting that the jaws are “not particularly robust,” citing work by Therrien et al. (2005) and Sakamoto (2010). These studies show that dromaeosaurid jaws act as simple levers, indicated by a largely level force profile along the jaw, and more suited to quick nip and pull motions than prolonged periods where the teeth are embedded in flesh, which is indicated by high force profiles in the back of the jaw as is seen in stouter jaws such as in Allosaurus fragilis, or Gorgosaurus libratus. The subjectivity of “robustness” aside, this suggests that the thin jaws, thin teeth, where slashing teeth. As such, the denticles are merely suited for quick pulls, nipping bits of flesh apart, rather than prolonged prey engagement, and the same is true for the broader-jawed, broader toothed Dromaeosaurus albertensis (see Therrein et al., 2005, for discussion).

The morphology of some dromaeosaur teeth differ strongly from the hooked denticles of taxa noted by Fowler et al. (2011). Note, above and below, the presence in Saurornitholestes langstoni (after Currie et al., 1990), that while the morphology of the “main” body of the denticle (or corpus) is hooked, each denticle is paired with a apical and a basal “keel,” which converge into the diaphyses between denticles (in some teeth, this diaphysis, the space between denticles, terminates in a narrow slot with a rounded bottom, called a cella. These keels afford the denticles less of a elongated, slender profile and more of a rounded, hump-like one, and can substantially alter the effect the hooking provides when the teeth are engaged in a substrate. It is interesting enough that one can treat the morphology of denticles as miniature teeth — so that each crown has it’s own “jaw” of teeth along the edge — and thus infer a general model that they might act like saw blades or a method of distributing compression force along the carina. Such a model may seem useful, but is certainly untested when it comes to predatory theropod dinosaurs, just as such a model might come under fire when adapted to ornithischians, or sauropods, or a range of denticulated, ziphodont or non-ziphodont dentition in Archosauria, mammals, etc.

Denticles in various taxa. A, Goyocephale lattimorei (after Perle et al., 1983); B, troodontid (after Currie et al., 1990); C, Richardoestesia gilmorei (after Currie et al., 1990); D, generic therizinosauroid, adapted after Erlikosaurus andrewsi and Nothronychys mckinleyi; E, Alxasaurus elesitaiensis (after Russell & Dong, 1994); F, Saurornitholestes langstoni (after Currie et al., 1990).

I hesitate to warn authors of adapting teeth as analogies to denticle function, especially in light of work by Abler (1992) that the function of denticles in some taxa may be appropriate to operating in a manner very different from the crowns (they work parallel to, not perpendicular to, penetration of muscle fibers, or can be used to pull through the flesh along the surface instead of pulling or hooking into it). This is certainly interesting if, for example, similar denticle morphology is known in disparate clades of dinosaur, including tyrannosaurids (Abler, 1992, where they are low, rounded, and somewhat quadrangular), Richardoestesia gilmorei (Currie et al., 1990, where they are narrow, rounded, and strongly quadrangular with a distal “apex” parallel to the carina) and Goyocephale lattimorei (Perle et al., 1983, where they are rounded, and even slightly hooked). We can, however, use knowledge of biomechanics (as Fowler et al., 2011, employ) to make inferences: a hooked tooth, for example, is suited to applying pressure at the tip, and its curvature would help reduce compression in the opposite direction; a rounded or blunt denticle is suited to withstanding pressure along the edge, and here richardoestesine and tyrannosaurid denticles are very similar, even while their teeth are so very different. Of course, deducing denticle function is not that simple, and something a good deal of work should go into; I would LOVE to do this, if I had the resources.

Thus, apical hooking and close spacing of tooth denticles is not a strictly theropod thing, as it occurs in the ornithischian Goyocephale lattimorei. It would be curious to apply the principle of denticle morphology to variation along the row, as this hasn’t been precisely quantified (something else I’d LOVE to do). It varies in troodontid jaws, for example (Currie, 1987; Currie et al., 1990), where the mesial-most teeth are inclined somewhat mesially and laterally from the middle and distal teeth, a feature only exaggerated by the mesial curvature of the mandible. This provides a U-shaped aspect to the jaw tip when viewed from above or below, a feature absent in dromaeosaurs, such that mesial crowns could have denticles facing not apically, but outwardly, and also that depending on the crown, parallel to the alveolar margin or perpendicular to it.

A selection of troodontid jaws in lingual and alveolar views. H indicates a cross-section of the jaw at midramus for D, taken at the breakage of the posterior end. Note that tooth inclination in D is mesial, while distal crowns are more vertical; this is regardless of the curvature of the mandible in lingual view. Scale bars all represents 1cm. (This is taken after a portion of an unpublished study on troodontid jaw variation, and attempted to quanitify published jaw morphologies.)

Neither Therrien et al. (2005) nor Sakamoto (2010) address troodontid jaw morphology and bite force estimates; estimating performance based on jaw profile and the brevity of the tooth row to jaw length (e.g., see my reconstruction of a troodontid skull and jaw here) I would predict a profile in the weak range, similar to ornithomimids rather than dromaeosaurids, regardless of the presence of teeth. Generalized size of the teeth suggest tooth size should play no role in modifying the profile attained from the jaw alone, and thus it could be estimated relatively easily for toothless jaws (I assume, and possibly incorrectly). With the rostrally tapering jaws (see above), rostral biting performance is extremely weak, and while it improves posteriorly, it should not be any stronger than the profile attained for a generalized dromaeosaurid. This suggests that, regardless of the argument by Fowler et al., that the teeth and denticle size difference offer no departure in the nip and pull method proposed for dromaeosaurids, troodontids should have a rostrally weaker bite than posteriorly in comparison to dromaeosaurids. Moreover, the different orientation of mesial to posterior teeth in the jaw, and the orientation and shape variance in such teeth can imply that mesial teeth certainly functioned differently than other teeth, and while variance is not necessarily quanitified yet, the difference from dromaeosaurids (Currie, 1987; Currie et al., 1990) necessitates a distinct perspective when dealing with jaw functional analogues, rather than implying they are merely scaled versions of dromaeosaur jaws.

With Arms Wide Open (and Apologies to Pearl Jam)

I am very curious where this study will go next. Fowler et al. have produced a wealth of quanitified data regarding toe function and proportions (Fowler et al., 2009, 2011). These analyses attempt to place into a behavioral context the function of the dromaeosaur pes and thus feeding ecology on the whole. They took the “raptors” off the sides of the tenontosaurs and engaged them with smaller prey. Cautiously, then, one must note that this is all itself speculative. Indeed, the presence of large, flexible hands and apparent wings are themselves bizarrely-linked to the question of pedal function, as Fowler et al., 2011, themselves note:

It remains paradoxical that the manual digits of paravians seem well-suited for flexion and grasping, yet would have borne flight feathers so as to make such an action difficult or clumsy (Ostrom, 1969; Senter, 2006; Carpenter, 2002). Further, paravian manual unguals are enlarged and strong expression of flexor tubercles suggests that the claws were capable of exerting considerable force, yet the limited range of motion of the forelimb (Senter, 2006) seems strongly adapted for flapping, rather than the flexibility required for prey manipulation (or indeed, climbing (Dececchi & Larsson, 2011)).
[Fowler et al., 2011:pg.8-9; I’ve modified the internal reference numbers to citations, a practice of PLoS ONE‘s that makes internal text more legible, but reference checking more cumbersome. All cited references are included below.]

Dromaeosaurids as mantling hawks, demonstrating Raptor Prey Restraint (RPR, or the "ripper" hypothesis). From Fowler et al. (2011). Illustrations by N. Carroll and L. E. Hall, and well done at that (although I admit the first looks much like Luis Rey's style, as I first thought it to be when seeing it).

Fowler et al., 2011, primarily concern themselves with the “mid-sized” dromaeosaurids, Velociraptor mongoliensis and Deinonychus antirrhopus, but it should be noted that in the remarks above, most of the work has been to compare broadly the small nonavian theropods Microraptor zhaoianus (Dececchi & Larsson, 2011 [n2]) and Archaeopteryx lithographica (Ostrom, 1969). These analysis draw parallels based solely on the apparent articulation of the shoulder girdle, or later the apparent presence of a series of quill knobs on the particularly short ulna of Velociraptor mongoliensis (Turner et al., 2007b). While the articulation of the various elements of the shoulder are questionable (for example, check out Jason Brougham’s deconstruction of the topic at his blog, Soft Dinosaurs (and here are links to part one, two, three, four and five of this series, and an outlier discussion on dromaeosaur girdles in general, prompted by Burnham et al.’s (2000) improbably shoulder articulation for Bambiraptor feinbergi), absolute limb length implies that any mechanical leverage of the limbs and its range of motion was less than that of any extant volant-capable bird (Ostrom, 1969; Chiappe & Padian, 1998; Dial, 2003), although it was clearly greater than antecedents. It is also generally considered that Velociraptor mongoliensis and Deinonychus antirrhopus were terrestrial and non-volant in almost any fashion, yet Fowler et al., 2011, supposed they might have employed the flapping wingstroke in a form of stability flapping while engaged with prey. Thus regardless of their archaic (relative to extant birds) limb design and small pectoral and brachial muscle systems compared to, say Confuciusornis sanctus (Dececchi & Larsson, 2011), the “wings” of larger dromaeosaurids (as supposed by Fowler et al., 2011) are modeled directly after the smaller, older, and phylogenentically more basal deinonychosaurs, such as Archaeopteryx lithographica.

So what to make of the “flapping first” model versus the competing ideas? It is a compelling idea, and it takes the potential of integrating actual feeding ecology rather than circumstantial prey acquisition (such as a lizard, mammal or bird in the stomach or gut region) based on biomechanics into account. I much prefer this method of reasoning, and do not see any distinct problems with this especially in connection with WAIR (Wing-Assisted Incline Running, as proposed by Dial, 2003). It should be noted, however, that speculations on this stem directly from the largest of dromaeosaurs, not the smallest. It seems unreasonable to paint the short armed, robust and relatively shorter legged, probable ambush predators of the Djadokhta and Cloverly Formations in the same manner as the smaller “near-avians” of the Solnhofen, Yixian and Jiufotang Formations. Taking phylogenetics into account, dromaeosaurids became larger, less arboreal, as they evolved, rather than smaller and more arboreal, or more flight-capable. This implies that models based on the largest are ill-suited to help explain the smallest, or that the model can explain the unusual pedal integument of Microraptor zhaoianus, Anchiornis huxleyi or Archaeopteryx lithographica (Xu et al., 2003; Hu et al., 2009; Longrich, 2006).

These issues are actually complicated by the problems of incorporated a particularly grasping suited manus with a grasping suited pes, but relegate the actual grasping duty only to the pes, leaving the manus free. This should, on the path toward flapping or corralling behavior, lead to a stiffer manus (one of those “Just So Stories” we’ve been taught), yet the larger dromaeosaurs seem to have very flexible digits, especially the third manus digit (Senter, 2009). Moreover, it seems improbable that the arm was as winged as that of much smaller dromaeosaurs, where the wings are not broad and rounded but slender and tapering sharply (in contrast to accipiters). The tail, acting as a counter-balancing rod, seems odd in that in Microraptor gui, unlike Archaeopteryx lithographica (see Xu et al., 2003; Longrich, 2006) only the distal tip of the tail bore a “retricial” array. Absence of a distinct tail fan in raptors (any of them) implies that the distal end only structure may have had less of an effect during mantling when used to “corral” prey, than to cover it during invasive aggression of possible thieves, or threat displays. That this behavior (as Fowler et al., 2011, note) occurs when the prey is dead, the effect of maintaining purchase during the struggle may be less of an explanation, rather than a passive tendency to cover the prey, enlarge the birds’ aspect for defense or for threat displays.

Oddly, unlike accipiters or in fact any raptor, only the second pedal ungual is elongated, whereas the other digits are shorter-clawed. This has associated effects: The pes appears to be functionally didactyl (Xing et al., 2010; Mudroch et al., 2011), the third and fourth digits accommodating this by being parallel rather than divergent, and the fourth toe longer relative to the third than in tridactyl theropods such as Coelophysis bauri or such. It is therefore tempting to say that the functional effect of the pes (using primarily the second pedal ungual to engage the prey as an effective gripping — as in, penetrating — “talon”) differs from that of extant birds of prey; Senter (2006) even implied a defensive or agonistic function for this reason.

Deinonychosaurs in relation to the RPR ("ripper") hypothesis presented by Fowler et al. (2011). The illustrations are mine. Characteristics of the changing body proportions and relative size of the limbs and unguals of the manus and pes, as well as of the skull and teeth, indicate a shift from primarily foot-based prey-restraint toward coupled forelimb/hindlimb prey restraint, as indicated by mantling hawks, but augmented with actual holding of the prey by larger-bodied dromaeosaurids (here, Velociraptor mongoliensis). Scale bar equals 20 cm.

The hypothesis presented by Fowler et al., 2011, is thus very well presented and reasoned, and I hope to see it firmly tested with respect to smaller dromaeosaurids, especially how the pedal adaptation is affected by the large “hindwings” in Microraptor gui and Anchiornis huxleyi, which appear largely absent or unpreserved closer to the body wall and are strongly developed below the tarsus. Because large feathers approximate to the engaging area could be easily damage, it suggests either a mechanism to move these feathers out of the way, or perhaps that the foot was not actually used to engage prey in just this manner, but perhaps in a more delicate manner. It is also notable that, as preserved, juvenile Sinornithosaurus sp. (NGMC 91, or “Dave“; Ji et al., 2001), at just above 50% body size of a putative adult Sinornithosaurus millennii (Xu et al., 1999), lacks any trace of fully pennaceous feathers on its arms, while they are present perpendicular to the extended femora; adult specimens preserve no pennaceous feathers at all, only filamentous integument, as present elsewhere on the juvenile body. These may suggest that these dromaeosaurids, putative microraptorines like Microraptor gui with fully fledged feathers, may not indicate the true extent of feathered limbs, and their effective “umbra” when mantling, possible for larger-bodied dromaeosaurs. There is to date a lack of positive evidence demonstrating that large pennaceous feathers covered the arm in such a manner to enable an mantling aspect, and this is further made problematic when the tail in raptors are used to both mantle and for balance while struggling with prey (or long after the prey is dead, suggesting as I said above a possible lack of connection).

Fowler et al. (2011) propose that basal deinonychosaurs were cursors, likely pursuit hunters. I would bring attention to Hartman’s discussion of the argument that early pre-birds could run around with their arms and tail as stabilizing aerofoils, where the exaptation is towards an elevated forelimb that could tangentially lead into WAIR (Dial, 2003), and a broader, shorter tail can develop more maneuverability. Instead, the trajectory was toward an apparently hypercursorial pes in troodontids (due to proportions and the arctometatarsalian pes), while the dromaeosaur lineage shortened this feature after the split with Microraptoria. Dromaeosaurids are short-legged, larger, and likely ambush hunters. Yet it is not these animals have have developed shoulder motility, broad sterna, or proportionally longer arms to body size. They have shortened the arms, shortened the leg, and reduced shoulder mobility. Thus, it might seem that instead of developing a should-arm “flapping” mechanism on the lineage to birds, they did so completely isolated and parallel to the evolution of the modern avian flight stroke. This is important when you consider that some dromaeosaurs (esp. the unenlagiines) had particularly short forelimbs, and form the sister group to the dromaeosaurines (sensu lato, Eudromaeosauria), while at the same time retained these features at small size (e.g., Buitreraraptor gonzalezorum, an unenlagiine/unenlagiid [argh, stupid Linnaean lack of sense!]). Depending on the order of divergence (microraptorians (unenlagiines, dromaeosaurines)) or (unenlagiines (microraptorians, dromaeosaurines)), thus could upset the trajectory favored in the paper that the arms and behavior of large-bodied dromaeosaurids had anything to do with the evolution of the avian flight stroke.

Holotype of Achillobator giganticus (Perle et al., 1999), FR.MUNFR.15, a particularly robust dromaeosaurid.

I suggest, rather, that the flight stroke mechanism was present basally, and exapted into the larger-bodied forms and coupled with prey acquisition and restraint. Smaller-bodied animals are either broadly feathered and likely pursuit runners who used foot-based restraint alone, or moderate-sized pursuit runners with short arms likely unable to use anything but foot-based restraint; larger-bodied dromaeosaurs, from the “turkey sized” Velociraptor mongoliensis on up to the stocky Achillobator giganticus. It is easy to imagine these animals as grappling predators, ambush hunters, rather than pursuit runners, and that this gives great explanatory power to the disparity in limb construction, skull structure, the proportions of the limb, etc. Here‘s a prescient argument on the “stockiness” of large-bodied dromaeosaurs focusing on the short-legged, robust Achillobator giganticus by fellow “armchair” paleo nut, Matt Marynuik. He presumes, as did Fowler et al., that the stockiness is linked to the large ungual on pd2.

Pes of EME PV.313, holotype of Balaur bondac, left pes in laft lateral view; arrows indicate relatively small second pedal ungual and relatively large first pedal digit. From Csiki et al., 2009.

One can deduce, from the fantastically large ilium, robust femur and short distal limb that the muscle mass controlled a strong leg, but that the animal was not a swift one, and was probably subcursorial. Here, I show it running. I will emend this to show a more plodding [plotting?] animal, far more likely to engage in foot-to-chest combat with animals its size or larger (or smaller, at need). It is absolutely amazing to think of this in context to the amazingly short and robust limb bones of Balaur bondoc, the even stockier dromaeosaur from Romania (Csiki et al., 2009). Such an animal might indiacte even more robust, heavily clawed forelimbs than Velociraptor mongoliensis (see above). While Andrea Cau takes Balaur bondoc to task in regards to Fowler et al.’s new study here, calling Balaur “Dodoraptor,” he finds it disconforms to the general model of larger robust forms at the terminous of dromaeosaurid evolution, along with Deinonychus antirrhopus, but rather at the base of a radiation of “dromaeosaurines,” i.e., Eudromaeosauria. But I think they have yet to go far enough. It amazes me that no one has yet placed this hypothesis in comparison to those theropods with the absolutely most fabulous, amazing and bizarre feet of all, derived therizinosaurids (or “segnosaurs,” in the ancient way). Take a look:

Pes of GIN 100/111, holotype of Erlikosaurus andrewsi (Barsbold & Perle, 1980). Left, pes in flexor view, after Barsbold & Perle (1980); right, pes in lateral view with digits splayed for viewing, after Perle (1981).

The unguals of the pes are all mediolaterally compressed, deep, and high high dorsal curvatures. The flexor tubercles of the unguals are more moderate than in most dromaeosaurids, but not substantially more than others (see Balaur bondoc, above). The first metatarsal likely oriented medially somewhat, as indicated at least by the medial orientation of the distal articular facet to the long axis of the shaft, which when articulated is oriented medially from MTII itself. Flexibility of the digits is unknown, although strong dorsoventral compression may place the digits in the low-flexibility range of ornithomimosaurs (Senter, 2009). Despite this, the raptorial appearance of the unguals is particularly fascinating in light of both the presumed herbivorous habits of the taxa (Barsbold & Perle, 1980), but also the amazingly mediportal or graviportal posture of the body, something somewhat approached by the largest dromaeosaurids. It would be interesting to see where this plots in future analyses. If the pedal anatomy indicates a grasping structure, and the dentition is clearly herbivorous, the utility of the foot as a grasping device may divorce the function from predation, where predatory grappling may simply be a use for such a foot, but not primarily indicated by it.

In closing, I think that the proposal of Fowler et al. (2011), mantling behavior in dromaeosaurids, neatly explains the functional anatomy of the leg and arm, although I would extend this and indicate that the arms would engage the prey regularly. Thus rather than simply borrowing the “mantling hawk” motif, the arms would likely be less feathered and more engaged in prey restraint.

[n1] Bambiraptor feinbergi was coined by Burnham et al. (2000) with the intention of honoring the entire Feinberg family, whom had donated resources to enable the acquisition by the AMNH of the holotype and paratype specimens. The formation of the name feinbergi is in keeping with treatments of names honoring single individuals, but according to the International Code of Zoological Nomenclature [ICZN] (1985, 3rd edition), names honoring multiple individuals must take the form –arum (when all honored persons are female) or –orum (when at least one honored person is male), but not –i (male) or –ae (female). In keeping with this tradition, Olshevsky (2000) emended the nomenclature to feinbergorum, which has since been followed (e.g., Turner et al., 2007a and Norell & Makovicky, 2004). However, as of the release of the 4th edition of the Code (which took effect on January 1st, 2000, a date antecedent to Olshevsky, 2000 [which was released towards the middle of that year), this requirement is no longer in effect. Currently, the only spellings that must be corrected are “clear evidence of an inadverdent error, such as a lapsus calami or a copyist’s or printer’s error” (Art.32.5.1) or “a name published with a diacritic or other mark” (Art.32.5.2). As stressed by the Code, “[i]ncorrect transliteration or latinization, or use of an inappropriate connecting vowel, are not to be considered inadvertent errors.” Thus, use and affirmation of “feinbergorum” as the correct spelling of the name is in error, and should be discontinued.
[n2] Dececchi & Larsson (2011) reference Mannion et al. (2009) who supported a climbing “crampon” function for the large pedal ungual, but based their analysis on a manual ungual, and Bin-Jeffrey & Rayfield (2009) [n3] refuted this. Both analyses draw particularly on comparison to extant scansorial birds, as in Fowler et al. (2009).
[n3] This is an abstract for the 2009 SVP meeting, and is thus technically “unpublished.” The citation (below) is provided solely under the basis of discussing Mannion et al. (2009) on the subject of claw function.

Abler, W. L. 1992. The serrated teeth of tyrannosaurid dinosaurs, and biting structures in other animals. Paleobiology 18(2):161-183.
Abler, W. L. 1999. The teeth of the tyrannosaurs. Scientific American 1999(9):50-51.
Abler, W. L. 2001. A kerf-and-drill model of tyrannosaur tooth serrations. pg.84-89 in Tanke, Carpenter & Skrepnick (eds.) Mesozoic Vertebrate Life: New Research Inspired by the Paleontology of Philip J. Currie. (Indiana University Press, Bloomington.)
Barsbold R. & Perle A. 1980. Segnosauria, a new infraorder of carnivorous dinosaurs. Acta Palaeontologica Polonica 25(2):187-195.
Bin-Jeffery, A. & Rayfield, E 2009. Finite element analysis of pedal claws to determine mode of life in birds, lizards and maniraptoran theropods. Journal of Vertebrate Paleontology 29(Supplement to 3):64A. [n2, see above]
Burnham, D. A., Derstler, K. L., Currie, P. J., Bakker, R. T., Zhou Z.-h. & Ostrom, J. H. 2000. Remarkable new birdlike dinosaur (Theropoda: Maniraptora) from the Upper Cretaceous of Montana. University of Kansas Paleontological Contributions, New Series 13:1–14.
Carpenter, K. 2002. Forelimb biomechanics of nonavian theropod dinosaurs in predation. Senckenbergiana Lethaea 82:59-76.
Chiappe, L. M. & Padian, K. 1998. The origin and early evolution of birds. Biological Review 73:1-42.
Currie, P. J. 1987. Bird-like characteristics of the jaws and teeth of troodontid theropods (Dinosauria, Saurischia). Journal of Vertebrate Paleontology 7(1):72-81.
Currie, P. J., Rigby, J. K., Jr. & Sloan, R. E.. 1990. Theropod teeth from the Judith River Formation of southern Alberta, Canada. pg.107-125 in Carpenter & Currie (eds.) Dinosaur Systematics: Approaches and Perspectives. (Cambridge University Press, New York City.)
Csiki, Z., Vremir, M., Brusatte, S. L. & Norell, M. A. 2010. An aberrant island-dwelling theropod dinosaur from the Late Cretaceous of Romania. Proceedings of the National Academy of Sciences of the United States of America 107 (35):15357–15361.
Dial, K. P. 2003. Wing-assisted incline running and the evolution of flight. Science 299:402–404.
Dececchi, T. A. & Larsson, H. C. E. 2011. Assessing arboreal adaptations of bird antecedents: Testing the ecological setting of the origin of the avian flight stroke. PLoS ONE 6(8):e22292.
Feduccia, A. 1993. Evidence from claw geometry indicating arboreal habits of Archaeopteryx. Science 259:790-793.
Forster, C. A., Sampson, S. D., Chiappe, L. M. & Krause, D. W. 1998. The theropod ancestry of birds: New evidence from the Late Cretaceous of Madagascar． Science 279:1915-1919.
Fowler, D. W., Freedman, E. A. & Scannella, J. B. 2009. Predatory functional morphology in raptors: Interdigital variation in talon size is related to prey restraint and immobilisation technique. PLoS ONE 4(11):e7999.
Fowler, D. W., Freedman, E. A., Scannella, J. B. & Kambic, R. E. 2011. The predatory ecology of Deinonychus and the origin of flapping in birds. PLoS ONE 6(12):e28964.
Fox, N. 1995. Development and behavior. pg.176-195 in Blaine (ed.) Understanding the Bird of Prey. (Hancock House.)
Glen, C. L. & Bennett, M. B.. 2007. Foraging modes of Mesozoic birds and non-avian theropods. Current Biology 17:R911-R912.
Hu D.-y., Hou L.-h., Zhang L.-j. & Xu X. 2009. A pre-Archaeopteryx troodontid theropod from China with long feathers on the metatarsus. Nature 461:640–643.
Ji Q., Norell, M. A., Gao, K.-q., Ji S.-a. & Ren D. 2001. The distribution of integumentary structures in a feathered dinosaur. Nature 410:1084-1087.
Longrich, N. 2006. Structure and function of hindlimb feathers in Archaeopteryx lithographica. Paleobiology 32(3):417-431.
Manning, P. L., Margetts, L., Johnson, M. R., Withers, P. J., Sellers, W. I., Falkingham, P. L., Mummery, P. M., Barrett, P. M. & Raymont, D. R. 2009. Biomechanics of dromaeosaurid dinosaur claws: Application of X-ray microtomography, nanoindentation, and finite element analysis. The Anatomical Record 292:1397–405.
Mudroch, A., Richter, U., Joger, U., Kosma, R., Idé, O. & Maga, a. 2011. Didactyl tracks of paravian theropods (Maniraptora) from the ?Middle Jurassic of Africa. PLoS ONE 6(2):e14642.
Norell, M. A. & Makovicky, P. J. 1997. Important features of the dromaeosaur skeleton: Information from a new specimen. American Museum Novitates 3215:1-28.
Norell, M. A. & Makovicky, P. J. 1999. Important features of the dromaeosaur skeleton. II. Information from newly collected specimens of Velociraptor mongoliensis. American Museum Novitates 3282:1-45.
Norell, M. A. & Makovicky, P. J. 2004. Dromaeosauridae. pg.196-209 in Weishampel, Dodson & Osmólska (eds.) The Dinosauria (2nd Edition). (University of California Press, Berkeley.)
Olshevsky, G. 2000. An annotated checklist of dinosaur species by continent. Mesozoic Meanderings 3:1-157.
Ostrom, J. H. 1969. Osteology of Deinonychus antirrhopus, an unusual theropod from the Lower Cretaceous of Montana. Bulletin of the Peabody Museum of Natural History 30:1–165.
Perle A. 1981. Новый сегнозаврид из верхнего мела Монголии [New segnosaurid from the Upper Cretaceous of Mongolia]. Трудй – Совместная Совестко-Монгольской Палеотологыческая Зкспедитсия — Joint Soviet-Mongolian Paleontological Expedition, Transactions 15:50–59.
Perle A., Maryańska, T. & Osmólska, H. 1983. Goyocephale lattimorei gen. et sp. n., a new flat-headed pachycephalosaur (Ornithischia, dinosauria) from the Upper Cretaceous of Mongolia. Acta Palaeontologica Polonica 27(1-4):115-127.
Perle, A., Norell, M.A. & Clark, J. 1999. A new maniraptoran theropod – Achillobator giganticus (Dromaeosauridae) – from the Upper Cretaceous of Burkhant, Mongolia. Contributions of the Mongolian-American Paleontological Project 101: 1–105.
Russell, D. A. & Dong Z.-m. 1994. The affinities of a new theropod from the Alxa Desert, Inner Mongolia, People’s Republic of China. Canadian Journal of Earth Sciences — Revue canadienne de sciences de la Terre 30(10):2107-2127.
Sakamoto, M. 2010. Jaw biomechanics and the evolution of biting performance in theropod dinosaurs. Proceedings of the Royal Society B 277:3327–3333.
Senter, P. 2006. Comparison of forelimb function between Deinonychus and Bambiraptor (Theropoda: Dromaeosauridae). Journal of Vertebrate Paleontology 26(4):897-906.
Senter, P. 2009. Pedal function in deinonychosaurs (Dinosauria: Theropoda): A comparative study. Bulletin of the Gunma Museum of Natural History 13:1-14.
Sues, H.-D. 1978. A new small theropod dinosaur from the Judith River Formation (Campanian) of Alberta, Canada. Zoological Journal of the Linnaean Society 62:381-400.
Therrien, F., Henderson, D. M. & Ruff, C. B. 2005. Bite me: Biomechanical models of theropod mandibles and implications for feeding behavior. pg.179-237 in Carpenter (ed.) The Carnivorous Dinosaurs. (Indiana University Press, Bloomington.)
Turner, A. H., Hwang, S. H. & Norell, M. A. 2007a. A small derived theropod from Öösh, Early Cretaceous, Baykhangor Mongolia. American Museum Novitates 3557:1-27.
Turner, A. H., Makovicky, P. J. & Norell, M. A. 2007b. Feather quill knobs in the dinosaur Velociraptor. Science 317:1721.
Xing L.-d., Harris, J. D., Sun D.-h. & Zhao H.-q. 2009. The earliest known deinonychosaur tracks from the Jurassic-Cretaceous boundary in Hebei Province, China. Acta Palaeontologica Sinica 48(4):662-671.
Xu X., Wang X.-l. & Wu X.-c. 1999. A dromaeosaurid dinosaur with a filamentous integument from the Yixian Formation of China. Nature 401:262-266.
Xu X., Zhou Z.-h., Wang X.-l, Kuang X.-w., Zhang F.-c. & Du X.-k. 2003. Four-winged dinosaurs from China. Nature 421:335–340.
Zheng X.-t., Xu X., You H.-l., Zhao Q. & Dong Z.-m. 2010. A short-armed dromaeosaurid from the Jehol Group of China with implications for early dromaeosaurid evolution. Proceedings of the Royal Society B 277:211–217.