The problems with The Origin and Evolution of Birds

Jeff Poling

Dr. Alan Feduccia is a well known ornithologist who is often sought out by the press whenever paleontologists announce an important find that further strengthens the dinosaurian ancestry of birds. The reason for this is that Dr. Feduccia is a well known critic of the dinosaur-bird theory.

Late last year Dr. Feduccia published a book called The Origin and Evolution of Birds. In it he explores his theories on the evolutionary history of birds, a history that lies outside of the Dinosauria. Some of the arguments he uses in arguing a separate ancestry for birds were recently debated on the Dinosaur Mailing List. -- ed.

On Dr. Feduccia's statements regarding the reversed first toe in birds and other theropods:

From: Stan Friesen <swf@ElSegundoCA.NCR.COM>

From: Dinogeorge@aol.com

Retroverted hallux (articulating in a shallow groove on the second metatarsal, distally on the palmar side) occurs in all theropods, and only in theropods, at and above the common ancestor of ceratosaurians and tetanurans. In almost all the larger cursorial forms, however, the retroverted hallux is secondarily reduced and points, in varying degrees, posteromedially.

Feduccia is off base again. This clearly indicates that the anisodactyl condition is ancestral for birds, and all other forms found in living birds are derived. Such radical changes in foot structure can, in my opinion, only occur if they are adaptive, contrary to what Feduccia says.

Feduccia also states he does not believe that theropods had anisodactyl feet!. (He associates the retroverted hallux with perching - which he may be correct about, but that just means that basal theropods were arboreal).

On Dr. Feduccia's claims that birds do not belong on the dinosaur family tree because the most birdlike non-avian dinosaurs, such as Velociraptor and its kin, appear in the fossil record long after the first known bird fossil:

From: "Thomas R. Holtz, Jr." <th81@umail.umd.edu>

At 12:23 PM 1/10/97 -0500, Caitlin Kiernan wrote:

He [Feduccia] seems to be ignoring the possibility of a long-forelimbed, arboreal dinosaur.

Long-forelimbed we have, and in forms apparently close to the ancestry of birds. We call them deinonychids. Other, slightly more distant, examples include troodontids and ornithomimids.

All of these groups are Cretaceous.

This says nothing about the two fundamental problems with Feduccia's studies:

1. Lack of a fossil in a stratum does not mean lack of a member of that taxon in that age. A couple of years ago, there were no known ornithomimids before the Aptian/Albian, and no known tyrannosaurids from before the Senonian. Now both are documented in the Barremian.

2. If known presence of an ancestral taxon older than the group in question is required, than Feduccia must accept that chimps and gorillas evolved from humans, that scyphozoan jellyfish evolved from anthozoan corals, that gnathostome vertebrates must have evolved from conodonts, and other unlikely events. Because of taphonomic biases of environment, preservablity, and rarity, some taxa within a clade have a better fossil record than their relatives. Just because one member of the group has a better chance of being preserved in the fossil record than the other does not mean it was ancestral to the other.

At 12:23 PM 1/10/97 -0500, Caitlin Kiernan wrote:

I think that Feduccia is concerned with finding a suitable dinosaur ancestor predating (or at least of roughly the same age) as the earliest known birds (and he seems to lean toward the possibility that this may be Protoavis, and since the Jurassic birds are so advanced, that means he's calling for an ancestor in the Carnian or earlier). He judges neither Eoraptor nor Herrerasaurus viable candidates (short-limbed, cursorials).

I'm going to go back over Feduccia's discussion of the morphologic problems he perceives with a theropod ancestry for birds this weekend, and maybe I'll wade into this a little more deeply.

From: "Darren Naish" <dwn194@soton.ac.uk>

Hi all. I made several confusing points a while back in a post on Jurassic maniraptors. Here's an attempt to clear them up and provide a bit more info. This kind of stuff is news to a lot of people (like, those who argue that bird-like theropods only post-date their bird descendants;-)).

On a recent TV show ("The Great Dinosaur Trail".. same series where the name Neovenator was flashed up <shudder..>), Bakker was talking all about "raptors" in the Morrison. I assumed he was talking about theropods in general - apparently he's got a site where large and small theropods together chewed up a bunch of Apatosaurus leg bones. But he seemed to make it clear enough that he really was talking about Morrison dromaeosaurs. What the..?

Little teeth provisionally referred to the Maniraptora are described and figured in Metcalf et al. (1992) - these come from the Gloucestershire Bathonian (Mid Jurassic) site here in England. I think they're also discussed by Metcalf and Walker (1993) in In the Shadow of the Dinosaurs, but I don't have that to hand. An amazing assortment of other small Bathonian taxa are reported from the same site, as are a few bigger things like a cetiosaur and a big megalosaur. There are also confusing rumors of a small, relatively complete theropod with big hooked claws. This one did the rumor circuit early last year, but still nothing on it. Expect more on April 1st I guess..

Anyhow, at least one of these Bathonian teeth is labeled tentatively as "dromaeosaurid-like (juvenile?)". Feduccia and colleagues might like to note that dromaeosaurid-like animals were, therefore, around prior to the Lower Cretaceous (as Tom Holtz has pointed out previously). Michael Benton (1993) provided an excellent overview of the same Bathonian site[. Of] the maniraptoran tooth he says: "New findings include some of the world's oldest frogs, salamanders and lizards, as well as some of the oldest members of the Maniraptora, the group that includes birds and their nearest dinosaurian relatives." The ceratosaur present in the deposit (restored, basically, as a Coelophysis in the dioramas) should not cause confusion with any Welsh sites (new coelophysid footprints), or anything from.. say, the Isle of Skye for example. Of course, it has. Me, I'm waiting for that new Sarcosaurus skull.

And a possible maniraptoran tail vert[ebrae] has been reported from the Isle of Skye too, leading Neil Clark to adorn the cover of Scottish Journal of Geology (and a mug, and a bunch of t-shirts apparently) with some cetiosaur-attacking feathery little dromaeosaur-look-alikes. Last I heard, Neil had not verified the identity of the vert[ebrae]. Neil? I think the Isle of Skye stuff is Bajocian-Bathonian, but having said that it's probably not. Anyway, it's definitely mid-Jurassic.

On other perceived problems with Dr. Feduccia's book:

From: GSP1954@aol.com

A number of people have sent notes to this board with favorable observations on [Dr.] Alan Feduccia's views on nondinosaurian bird origins in his new The Origin And Evolution Of Birds. Having read the chapters on the subject, I find them full of basic errors and highly questionable assertions. [There are] too many to detail in full here. Here are a few.

[Dr. Feduccia] asserts that the skull of Archaeopteryx is very avian, largely on reconstructions done a decade ago. He ignores the work - some based on the newest skull - by P. Wellnhofer, A. Elzanowski, L. Witmer and myself that shows that the old restorations are in error, and that the skull was much more theropodian than avian. Nor does he cite P. Currie's observation that the occipital wing of the braincase of Archaeopteryx is very similar to that of dromaeosaurs. [Dr. Feduccia] even says (based on the impossibly blurry results of an old CT scan) that Archaeopteryx had a double headed quadrate, [even] after the new skull has proven that this is absolutely false!

[Dr. Feduccia] says theropod shoulder girdles are not bird-like. The fully articulated skeleton of the new troodont Sinornithoides shows that the coracoids were large, their outer surfaces faced forwards, were angled sharply on the scapula blade, and articulated with long anterior grooves of a large sternal plate (not ossified in this juvenile) just like birds (especially the flightless sort). The shoulder girdles of dromaeosaurs and oviraptors were constructed in the same manner. Also, many of these advanced theropods could fold their arms in a near avian manner.

The similarity of the pubes of Archaeopteryx and theropods are dismissed. In fact the pubic shafts of the former and dromaeosaurs share a plate-like, slightly angled transverse cross-section not found in any other archosaurs.

[Dr. Feduccia] states that the "thecodont" Postosuchus is close to the origin of carnosaurs. This view is held by few if any today. All crurotarsal ankled archosaurs such as Postosuchus belong to the group that includes crocodilians, so Postosuchus is a near crocodilian, not a near dinosaur.

Little Triassic Megalancosaurus is proposed as an "avimorph" close to birds. The only photograph included is the one of the front end that looks vaguely avian. Not shown are photos of the rest of the skeleton, or a complete skeletal restoration, which shows that this deeply bizarre diapsid reptile mimics chameleons in design (without opposable toes and a prehensile tail!), and is an extremely nonavimorph.

I happen to agree with [Dr. Feduccia] that birds almost certainly evolved flight from high places (and I will shoot the next person whom illustrates a running Archaeopteryx chasing a dragonfly, which [is] too fast and agile to be hunted even by modern insect eating birds for heaven's sake!). Where we diverge is in his unsubstantiated insistence that small theropods - despite many having long clawed fingers and bird-like toes - were not good climbers.

I also agree with [Dr. Feduccia] that although cladistics is very important, it is also not phylogenetic nirvana. What [Dr. Feduccia] does not know is how overwhelming is the skull, skeletal, eggshell and nesting behavior evidence that advanced theropods are the ancestors of birds. [Dr.] Feduccia and other paleoornithologists sometimes say that we dinoologists do not understand bird anatomy well enough. Actually, we know birds quite well because they are the living dinosaurs we look at all the time. The real problem is that some paleoornithologists do not understand the anatomy of nonavian archosaurs well enough.

Of course, [according to Dr. Feduccia] I have been guilty of committing a "fantasy" by illustrating my small theropods with feathers, for which "there is no evidence," and would hinder the running of a biped by producing drag (is that why ostriches and cheetahs are naked?). Guess I should start removing all those feathers from my little dinosaur drawings. [Note: Greg's sarcastic comment is due to the fact that, provisionally upon full publication, the non-avian dinosaurs Pelicanimimus and Sinosauropteryx have been found preserved with feather or protofeather impressions. -- ed.]

[Dr. Feduccia] concludes that body insulating feathers are not necessarily correlated with endothermy despite the absence of a single living insulated e[ct]otherm in the modern world. He says Stegosaurus had a "walnut"-sized brain, and asserts that there is correlation between brain size and metabolism even though endothermic tuna have small simple brains while some ectothermic fish and sharks have really big complex brains. This is typical of his section on physiology, in which he embraces every argument against dinosaur endothermy and airily rejects all for.

This one in particular boggles my mind: arguing that dinos were big because they were ectotherms living in the warm Mesozoic, [Dr. Feduccia] says that "One need only to look at the highly active, large tropical reptiles, such as Nile crocodiles and Amazonian anacondas, to see how ectotherms have surpassed endotherms in the warm climates of the tropical zone" (its right there on p. 125). Surpassed? Surpassed elephants that reach 10 tons and nearly 14 ft and giraffes that tower 18 ft in the Namib Desert? What about hippos and rhinos?

In a recent interview [Dr. Feduccia] accused those who support dinosaurian ancestry of birds of being "arm wavers" without evidence to back up their new dogma. In truth it is he who is committing unsupported arm waving, [as] he cannot cite a single character that is found in birds and nondinosaurian archosaurs that is not also found in theropods. As for theropod-bird relationships being dogma, it is no more or less so than is the theory of cynodont-mammal relationships, the theory of evolution, or the theory of relativity.

An Ornithologist weighs in:

From: Ronald Orenstein <ornstn@inforamp.net>

It seems to me that [Dr.] Feduccia's biggest, overriding error is that he uses much of the book to proselytize for the "birds-did-not-evolve-from-dinosaurs view" with such messianic fervor that he appears to have lost the objectivity a book like this should have had. This is not to say that he is wrong (though I think he is), but that his views appear to have colored his ability to present the evidence. He pours scorn on anyone of the contrary opinion, stating more than once that they do not understand biology (though his own views on convergence seem peculiar at best -- he goes into great length explaining why all the similarities between birds and theropods must be the result of convergence, though they lived very different lives, but it does not seem to occur to him that the similarities he notes between birds and such probably arboreal forms as Longisquama might also be the result of convergence). His book, in short, is a polemic rather than the dispassionate text on fossil birds that we still very much need, and this tone makes it very difficult to judge whether any of the evidence is fairly, or completely, presented.

That said, I think the book is worth having, if only because there is nothing else like it, but I suspect (I am not a paleontologist myself so I cannot say this with any certainty) that it should, unfortunately, be taken with a grain of salt.

The man who placed tyrannosaurs close to birds in the maniraptoriformes has his say:

From: "Thomas R. Holtz, Jr." <th81@umail.umd.edu>

But after reading Alan Feduccia's recently published The Origin and Evolution of Birds" my impression is that the matter of bird origins has not been settled yet, and that according to some, birds may have evolved from some basal non-dinosaurian archosaur. No?

No (although I can see how [Dr.] Feduccia's book would mislead people).

[Dr.] Feduccia is the most prominent member of a relatively small group of ornithologists who adamantly object to the dinosaurian origin of birds (Larry Martin being another). They have gone so far as to say that the methodology of cladistics must be flawed, since time and again (and with different data sets, by independent researchers...) cladistic analyses result in birds being deeply nested within theropod dinosaurs.

An alternative, of course, is that the method is not flawed, but instead the initial assumption of Feduccia, Martin, etc. is.

You can read recent reviews of [Dr.] Feduccia's book in the American Journal of Science (review by Alan Brush), American Scientist (review by Kevin Padian), and most recently in Science (review by Larry Witmer). I'm sorry that I don't have the issue numbers hand, but they are all very recent (I literally just came from the library, where I read Witmer's review in the most recent issue of Science).

Although [Dr.] Feduccia has given some important questions to examine with regards to some characters, his knowledge of the current understanding of non-avian archosaurs is very out of date, as is his method of tracing phylogeny (i.e., assume you know what the ancestral conditions and behaviors were first, find a form that fits your notion second, and explain away any similarities with other forms as convergence).

What is more, some of his alleged proto-birds are demonstrably less bird-like than theropods in every conceivable anatomical way, but only isolated photographs of a few parts of the body are shown to emphasize the "birdiness" of the creatures. The worst case of this is on p. 89, concerning Megalancosaurus. Although the skull and forelimb are both shown, only a (demonstrably inaccurate) life reconstruction is shown of the body. For a complete look at the skeleton of this Triassic arboreal reptile (amazingly convergent with chameleons in some ways), see Renesto, S. 1994. Megalancosaurus, a possibly arboeral archosauromorph (Reptilia) from the Upper Triassic of northern Italy. Journal of Vertebrate Paleontology 14: 38-52.

More from Dr. Holtz in a back-and-forth discussion with "Cal King."

Cal King wrote:

I am still at a lost as to why you think [Dr. Feduccia's] book lacks scholarship. As to other book reviews, that is only other people's opinion. Is it because Feduccia recognizes the Class Aves but not the Avialae that he is getting all the bad rap?

From: "Thomas R. Holtz, Jr." <th81@umail.umd.edu>

No, it is because he consistently misrepresents the current state of paleontological knowledge. For example, he claims that only four species of non-avian theropod possess the semilunate carpal block. While this statement was true (in the 1974 and 1975 papers by Ostrom he is referencing), it is not true for 1995-6 (when he was writing the book): this structure has been reported in the peer-reviewed literature for at least fifteen different species in five or six different "families".

Cal King:

References please.

Dr. Holtz:

I'll give you the taxa, and a few references. You can hunt up the rest:

The dromaeosaurids Deinonychus antirrhopus (Ostrom, 1969, Peabody Mus. Bulletin 30), Velociraptor mongoliensis (Barsbold, 1983, Sovmetsnaia Sovestsko-Mongol'skaia Paleontologischeskaia Ekspeditsiia Trudy 19), and Sauronitholestes langstoni; the oviraptorids Oviraptor philoceratops (Barsbold 1983, above) and Ingenia yanshini (Barsbold 1981, Sovmetsnaia Sovestsko-Mongol'skaia Paleontologischeskaia Ekspeditsiia Trudy 15); the caengnathids Elmisaurus rarus and Chirostenotes pergracilis; the therizinosauroids Alxasaurus elesitaiensis (Dong & Russell 1993, Canadian Journal of Earth Sciences 30) and Therizinosaurus cheloniformis (Barsbold 1983); the troodontids Troodon formosus, Saurornithoides mongoliensis, Saurornithoides junior (Barsbold 1983), and Sinornithoides youngi (Dong & Russell 1993, Canadian Journal of Earth Sciences 30); and Coelurus fragilis and Ornitholestes hermanni (the Ostrom papers Feduccia cited in his own book).

Dr. Holtz:

He represents Postosuchus as a tyrannosaurid ancestor, but this animal has long been recognized as a) a chimera, composed of at least two different taxa and b) has clearly documented features (in both component taxa) linking it to different pseudosuchian (rather than dinosaurian) groups.

Cal King:

That seems to be a difference of taxonomic opinion.

Dr. Holtz:

Um, no. Feduccia simply cites the original 1985 Chattejee paper. He does not mention any of the more recent papers (by Parrish or by Murry & Long) on that taxon.

Besides, how would you know if it is simply a matter of taxonomic opinion or not. Have you read the Chatterjee paper, or Feduccia's book?

Dr. Holtz:

He states that no member of the dinosaur lineages closest to birds are present in the Jurassic, when one has been known to be in the Late Jurassic since the 1880s, another since 1899, and others found in the 1980s and teeth found in the Middle Jurassic in the 1990s. He states that there is a trend toward arm length reduction in the theropods through time: it is true that two lineages (tyrannosaurids and abelisaurids) show marked arm reductions, but other lines retained a relatively constant arm-length:torso-length or arm-length:hindlimb-length over time, and still others (which also share many derived characters with birds!) show an increase in time. He continues to refer to the Confusciusornis as Late Jurassic, even though by his own radiometric dates it is Early Cretaceous!

Cal King:

He (1996) did say, as a junior author in the Science article, that dating of the fossils remain controversial and that more field work is needed for more precise dating.

Dr. Holtz:

Yes. Nevertheless, their own published radiometric date placed the formation within the Early Cretaceous by the accepted geochronology of the time (they cite, instead, a wall poster (!) as support for a later Jurassic-Cretaceous transition).

Dr. Holtz:

He states that because paleontologists recognize that some coelurosaurs have features (primarily toothless beaks) similar to birds, that we know that convergence occurs (ooh, big surprise!). Although true furculae on allosaurids and tyrannosaurids had been reported earlier than some of the other references in the book, he still states that oviraptorids are the only non-avian theropods showing this structure.

All this in just the first two chapters!

Quite frankly, I don't care about the taxonomy he chooses to use. However, it is necessary for someone writing about the subject of bird origins to actually do the scholarly research on the recent work on the subject, even when it doesn't support your own ideas.

Hence, it is a work of poor scholarship. Any questions? By the way, "Cal", have you read the book, or are you just going by what you read in "Living Bird" or whichever non-technical magazine you came across?

Cal King:

Maybe he thinks that the other recent works, based presumably on cladistic methodology, are "unscholarly."

Dr. Holtz:

It is nevertheless important in a scholarly work to fairly present the opposing viewpoints, even if you disagree with them. Presumably, one would then present one's own evidence which refutes that of the other side. To do less is to hide contrary data, and hence "unscholarly".

Cal King:

If I am not mistaken, theropod fossils are not very abundant.

Dr. Holtz:

Well, relative to some taxa, this is true. Relative to others (for example, the subset of theropods called "birds"), however, non-avian theropod remains are pretty abundant. It depends upon the species, the local geology, the amount of time the region has been explored, etc.

Cal King:

Anyhow, the late Cretaceous bird-like dinosaurs may have a single common ancestor in the mid-late Cretaceous.

Dr. Holtz:

Well, no. To give you fair warning, you are heading directly into territory where (quite frankly) I am one of the top authorities worldwide, so I don't expect you to have the background that I do.

Although the best specimens of typical Late K theropod lineages tend to come from the spectacular assemblages of western North America and eastern/central Asia, every group has now had unquestionable remains found in at latest the late Early Cretaceous, and several of them unquestionable skeletal remains from the early Early Cretaceous. Among these:

• Dromaeosauridae: Utahraptor in the early EK (Barremian), teeth in the Middle Jurassic.
• Oviraptorosauria: Microventor in the late EK.
• Therizinosauroidea: Alxasaurus in the late EK, unnamed skeletal taxa in the early EK of Asia.
• Troodontidae: Sinornithoides in the late EK, possible teeth in the Late Jurassic of North America and Middle Jurassic of England.
• Ornithomimosauria: Pelecanimimus in the early EK of Spain, unnamed taxon in the early EK of Thailand.
• Tyrannosauridae: unquestioned remains in the late EK of Japan, Siamotyrannus (possibly a tyrannosaurid) from the early EK of Thailand.

Would you like to know more?

Cal King:

The important question is: how bird-like are these early Cretaceous "theropods?"

Dr. Holtz:

They demonstrate the same suite of bird-like characters found in the Late Cretaceous representatives of the same groups, with the following exceptions:

The prefrontal of Deinonychus is less reduced than in later forms

and

Pelecanimimus has many teeth (actually, a trait it shares with most Mesozoic birds), whereas Late K ornithomimosaurs were toothless (and thus like Cenozoic birds).

However, the derived characters used to link these various coelurosaurs with birds are present (where that part of the anatomy is known) in all these Early Cretaceous forms.

Cal King:

This ancestor may have acquired the bird-like traits independently of the avian lineage. Feduccia et al. argue that there is a "long avian history in the Jurassic," long before any known Dromaeosaurid has acquired their bird-like characters.

Dr. Holtz:

And, since all known dromaeosaurids HAVE these avian characteristics, how does Feduccia demonstrate that the Jurassic dromaeosaurids do not have these features?

Cal King:

Dr. Holtz:

Deinonychus is late Early Cretaceous (Aptian-Albian stages), and Utahraptor (now known by much more complete specimens: looks quite cool!) from the early Early Cretaceous (Barremian stage).

These fossils still post-date the late Jurassic Archaeopteryx.

Dr. Holtz:

Yep. But the Morrison ?dromaeosaurid and the English Middle Jurassic teeth predate Archaeopteryx. And besides, the two taxa I mentioned above were in response to you comment that dromaeosaurids were only known from the Late Cretaceous. Those two taxa are very definitely dromaeosaurids, and very definitely Early Cretaceous.

Cal King:

I meant that the bird-like Dromaeosaurids were known only from the late Cretaceous.

Dr. Holtz:

Oh, bloody hell, Cal: learn to read! Deinonychus and Utahraptor are as bird-like as any Late Cretaceous dromaeosaurid, and they are both Early Cretaceous.

Cal King:

The available evidence shows that as one goes back in time to Archaeopteryx and beyond, the bird-like dinosaurs are decidedly much less bird-like. (Zhou 1995. The Auk 112:962)

Dr. Holtz:

Wrongo! Zhou has demonstrated no such thing, and by relaying (as so many do) on Feduccia's work, he is only adding to the error.

Cal King:

Where do I find evidence to contradict Zhou's claim?

Dr. Holtz:

Examine the skeletons (or their descriptions) of the various Jurassic forms I mentioned in the previous postings.

Cal King:

The name is Paul and the reference his 1988 book "Predatory Dinosaurs of the World." If birds are not ancestral to theropods, then their bird-like traits must be acquired independently, hence the probability of convergence or parallelism remains high.

Dr. Holtz:

Or, birds are descendants of other dinosaurs, and their similarities stem from common evolutionary events. So, we have established the different possibilites. How do we choose from amongst these possibilities?

Cal King:

Find at least one *unique* synapomorph that could demonstrate a dinosaurian origin for birds.

Dr. Holtz:

How about: furcula (allosaurids, tyrannosaurids, troodontids, oviraptorosaurs, dromaeosaurids, etc.), semilunate carpal block (troodontids, oviraptorosaurs, dromaeosaurids, etc.) and others listed on p. 312 of Carroll (1997).

Cal King:

Hence Archosaurian ancestry for the birds is the more tenable alternative than dinosaurian origin.

Dr. Holtz:

"Cal", do you know what an 'archosaur' is? Are you aware that dinosaurs are a type of archosaur?

Cal King:

Besides, the Ornithischians have shown some remarkable convergences in pelvic structure to the birds.

Dr. Holtz:

Actually, they have very, very few. It is difficult to confuse the pelves of these two groups. Despite the name "bird-hipped", the only bird-like aspect of ornithischian pelves is the backwards turned pubis. However, the pubes of dromaeosaurids are much more similar to those of basal birds (Archaeopteryx and Sinornis, for example) than are the pubes of ornithischians.

Note: Dr. Feduccia apparently made the comment recently that dinosaur paleontologists do not know bird anatomy well enough to comment on the ancestry of birds, especially as it relates to their dinosaurian ancestry, a theory that Dr. Feduccia disputes. Most dinosaur paleontologists thoroughly deny this. According to Dr. Thom Holtz (and Gregory Paul, above), because the object of their studies are long dead, dinosaur paleontologists have to study living animals in order to learn and test animal biology. Most become very familiar with the anatomy of modern birds because they are the living animals most like theropod dinosaurs ... indeed, they are theropod dinosaurs. Dr. Holtz points out that ornithologists usually have not rigorously and thoroughly studied dinosaurs and it is therefore they that lack the knowledge to comment on the dinosaurian ancestry of birds. -- ed.