One of the more interesting non-dental features of theropod dinosaur evolution is the orientation of the pubis. Historically, it was used to help affirm the transition of the modern bird form from that of particular theropod dinosaurs. Back then, it was Gauthier  and Ostrom  (and lesser known but nonetheless correct, Barsbold ) who tied retroversion of the pubis to birds and dromaeosaurid theropods, but this obscures that some historical perspectives, going back to Thomas Huxley of all people, correctly identified some features of the dinosaurian hip to that of birds. Of course, this was based on the occassional premise that Ornithischia (“bird-hipped ones”) was correctly named in more ways than one.
Opisthopuby is another subject I am fond of, for many reasons, and will not get into a large discussion on the topic as it relates to archosaurs in general (and it’s better to be broad on this topic given the prevalence of the condition, rather than confine ourselves to Dinosauria). In fact, I’d rather broaden this perspective to Sauropsida, or even Amniota (but that’s relatively easy: virtually all cynodont synaposids [including mammals] have a vertically to posteriorly-inclined, albeit short, pubis), because some taxa, many of whom are terrestrial, have particularly “unorthodox” pelvic organization, and it infers that pelvic organization relative to locomotion is particularly important (massive understatement).
The condition of a “reversed” or “retroverted” pubis is termed opisthopuby. The typical theropod orientation is propuby. These terms are not well-defined: They correspond only to the general seeming of the pubis to be pointed forward or backward, and it is not clear how this occurs. For example, some theropods have the distal portions of their pubis oriented caudally, but the proximal ends are straight vertically (G above); others have the reverse, a caudally-directed proximal portion, but a cranial projection of the distal end (not shown, as these refer to some theropods outside of the group commonly called Eumaniraptora [defined as the most recent common ancestor of Deinonychus antirrhopus and Vultur gryphus, and all of their descendants, 4] and Paraves [defined as all taxa closer to Vultur gryphus than to Oviraptor philoceratops, 4] — currently, without that many taxa closer to birds than oviraptorosaurs, but outside the dromaeosaur/bird node, these correspond to the same content). Some are in the middle, being almost completely vertical (D above), although most of these taxa have S-shaped pubes (H above).
It’s almost clear that the orientation of the pubis influences a large number of things, and the role of retroversion of portions of the shape will influence the attachment and position of various tendons and muscles. This makes the relative orientation of the pubis interesting in a finite way:
- Orientation of the pubis on the whole influences some, but not all features of the pelvic-hindlimb system.
- Segregating portions of the pubis based on the allocation of muscles and tendons will segregate taxa, and is thus of interest in a phylogenetic sense.
- Segregating portions of the pubis based on the allocation of muscles and tendons can alter the gait and posture of the limb, and this has interesting implications (duh) for the biomechanics of walking, running, the ability for bird-like theropods to launch, and so forth.
So where you draw the line, and how, relates to some pretty interesting questions and broad answers (if we knew them enough). Fortunately, while these questions have been answered for some taxa, few of them are Eumaniraptora/Paraves. Very few. This is fortunate only in that it allows a broad range of prospective studies. Go get ’em!
Oh, and if you want to guess which taxa produce the pelvises above, go ahead. I’ve excluded labels to permit this. This one will be easier than tooth denticles, which some people are just not familiar with (in part due to a lack of emphasis on their morphology [most analyses count number versus an arbitrary length, which has an undefined value relative to the tooth]).
 Gauthier, J. A. 1986. Saurischian monophyly and the origin of birds. In Padian (ed.) The Origin of Birds and the Evolution of Flight. Memoirs of the California Academy of Sciences 8:1-55.
 Ostrom, J. H. 1976. Archaeopteryx and the origin of birds. Biological Journal of the Linnean Society 8:91-182.
 Barsbold R. 1983. O “ptich’ikh” chertakh v stroyenii khishchnykh dinosavrov [“Avian” features in the morphology of predatory dinosaurs]. Trudy — Sovmyestnaya Sovyetsko-Mongol’skoy Palyeontologicheskiya Ekspeditsiya 24:96-103.
 Padian, K., Hutchinson, R. M. & Holtz, Jr., T. R. 1999. Phylogenetic definitions and nomenclature of the major taxonomic categories of the carnivorous Dinosauria (Theropoda). Journal of Vertebrate Paleontology 19(1):69-80.