“Feathers” on the big, “feathers” on the small, but “feathers” for dinosaurs one and all?

Facebooktwittergoogle_plusredditpinterestlinkedintumblrmailby feather


Yutyrannus artwork by Brian Choo. Sciurumimus artwork by Arkady Rose

This year has seen the discovery of two big deal dinosaur specimens. At least they are a big deal in regards to dinosaur integument and, possibly, metabolism.

First off from a few months ago we had the announcement the theropod Yutyrannus hauli, the “beautiful feathered tyrant.”

Xu, X., Kebai, W., Ke, Z., Qingyu, M., Lida, X., Sullivan, C., Dongyu, H., Shuqing, C., Shuo, W. 2012. A Gigantic Feathered Dinosaur from the Lower Cretaceous of China. Nature. Vol.484:92-95

This was not just a single fossil, but a collection of three fossils (one might be tempted to call it a family group, but that would only be speculation). As with all other dinosaur fossils that have been found to have filamentous integument, these guys come from Liaoning, China. They are suspected to have come from the Jehol Group in the Yixian formation. I say suspected because the complete three specimen set was a purchase from a fossil dealer, an all too common occurrence for Chinese fossils. As such the provenance information is unknown. A lot of Chinese fossil dealers don’t like to give away the location of their find due to the potential loss of other profitable specimens. This current trend in China is a good example of what happens when capitalism comes into play with fossil collecting (something that the U.S. has been mostly, but not entirely, able to avoid). So it is currently uncertain whether these fossils are from the Yixian. However given that all the others guys are too it is probably a good bet. Given the sketchy nature in which many Yixian fossils are collected, coupled with the possibly large consequences of the find, one should naturally be skeptical of the fossil. Had it been one individual on multiple slabs I would question its validity as a real thing. However since Y.huali is known from three individuals, and the filaments seem to follow a consistent pattern around the body (compare that to the helter-skelter nature of Tianyulong‘s preservation), forgery seems unlikely. These guys are probably the real deal. This has some potentially far reaching consequences to interpretations of Late Cretaceous coelurosaurs and the Jehol Biota itself (more on this in a bit).

The second announcement came just a few weeks ago. This was the discovery of a potentially new, miniscule theropod from Bavaria Germany.

Rauhut, O.W.M., Foth, C., Tischlinger, H., Norell, M.A. 2012. Exceptionally Preserved Juvenile Megalosauroid Theropod Dinosaur with Filamentous Integument from the Late Jurassic of Germany. PNAS Early Edition:1203238109v1-201203238.

The specimen is exceptionally well preserved. So well preserved in fact that it actually looks like a plastic toy. While this degree of preservation warrants importance all its own, the main interest behind this new guy—dubbed: Sciurumimus albersdoerferi (Albersdörfer’s squirrel mimic)—is the apparent presence of filamentous integument on the body coupled with its apparent placement among much more basal theropods. This discovery has far reaching consequences for theropod integument interpretations. Note: As with Y.hauli, Sciurumimus albersdoerferi was also purchased from a private collector. I don’t suspect forgery here either as this was in Germany, where fossil dealing is neither a big problem nor a lucrative business. The exceptional detail on the specimen would also require a substantial amount of theropod knowledge to pull off. Anyone having that amount of knowledge is more likely to be a real paleontologist than a get rich quick forger.

The filamentous brouhaha

Wild turkey showing off its “beard” (the long strand of bristles hanging off the chest). Image from The Smoking Beard blog

As I mentioned above, the big appeal of both specimens was the presence (or apparent presence) of filamentous integument. I am purposely avoiding the use of both “feathers” and “protofeathers” in this post as both of these are loaded terms. They assume that these structures had something, if anything, to do with the formation of real feathers. This has been a current trend that I have seen both online and in the literature. It is a trend that I think has skewed many interpretations of features found on or near fossil specimens (e.g., see Ortega et al. 2010) . The fact is that we don’t know for certain that what we are seeing has anything to do with feathers. Both of Y.hauli and S.albersdoerferi preserve what Prum (1999) referred to as stage I feathers, or basically long shafts with no branches. This most basic hypothetical precursor is not all that descriptive. Basically any long filament falls under stage I feather territory, including mammalian, insect, and arachnid hairs. Even the dorsal spikes on many iguanian and gekkotan lizards have the potential to fall under this category, especially when all one has to look at is a 2D smashed fossil.

That theropods were on the line that lead to birds does suggest that these filaments had some role in the evolution of feathers, but even that train of thought may be misleading. Sawyer et al. (2003) were the first to point out that not every filament on a bird is a feather. In particular, Sawyer looked at the filaments that compose the “beard” of turkeys (Meleagris gallapovo). Despite being filamentous, hollow, and containing feather beta-keratins, these structures were not feathers, or even feather offshoots. They were outgrowths of the epidermis that were more akin to fingernails than to feathers. One of the take-home messages of the Sawyer et al. paper was to be very cautious about any stage-I feather, as true identification of these feathers requires a level of detail that is currently missing in the fossils.

Yutyrannus huali: 1.4 tonnes and fluffy

The presence of extensive filamentous integument on the body of the Y.huali specimens has some potentially large consequences for interpretations of large theropod body coverings. As discussed previously, since the discovery of filamentous coverings in maniraptoran coelurosaurs back in 98, there has been a slow, but steady push from many in the paleo community to add fuzz to dinosaurs whenever possible. All this despite a preponderance of evidence for most dinosaurs having scales. A common argument put forth to handle this apparent paradox has been that large size would be a deterrent to feather formation since larger animals would have surface area/volume ratios that would naturally result in heat retention anyway. Under this scenario larger dinosaurs would be scaly so as to promote heat loss, while the smaller guys could sport filaments.

The discovery of Yutyrannus hauli changes all of this. At an estimated 1.4 tonnes in size (based on ZCDM V5000, the largest specimen), this animal falls in the size range of medium to large size theropods like Allosaurus fragilis. At that size the heat retained in the body by simply being big would be enough to warrant a switch to heat-loss “promoting” scales. However the fossils tell a completely different story. Even at 1.4 tonnes, a large filament covered theropod shows no signs of feather loss. As the authors mentioned:

Although feather preservation is patchy in these specimens…the distribution of the preserved filamentous feathers in the three specimens of Y. huali implies that this taxon had an extensively feathered integument in life.

There may be another reason why Y.hauli was sporting filaments. We’ll cover that in a bit, but first…

Sciurumimus albersdoerferi: baby megalosauroid with a fluffy tail?

Sciurumimus (prior to description) shown under UV back in October 2011 at the Mineral Days in Munich Show. Apparent soft tissue can be seen near the base of the tail and around the tibia. The rest of the soft tissue requires higher magnification than this photo provides. Photo and UV setup by Helmut Tischlinger

Despite the better preservation and farther reaching connotations of the discovery of S.albersdoerferi, it seems to have made a much smaller splash in the paleo community than expected. The biggest reason behind this is probably due to the way in which it was originally announced. Rauhut et al. originally “leaked” images online of S.albersdoerferi some nine months ago before the description came out. At the time it was also rumoured that the phylogenetic placement was going to be more basal than other filamentous theropods.

So what of these filaments? Looking at the slab that contains the specimen one would be hard pressed to see anything integumentary at all. In visible light there are very little signs of integument on the specimen. The filament discovery came from UV analysis of the slab. Many fossils will fluoresce under UV light. By mixing and matching UV wavelengths it is possible to make certain features “pop out.” Rauhut et al. accomplished this for S.albersdoerferiby using the refined techniques of their co-author Helmut Tischlinger. Tischlinger’s work has appeared in multiple paleo papers over the past few years. His talent for UV is unmatched and has allowed researchers to view structures that would otherwise be obscured by the matrix that housed them. That said, just because something appears under UV does not necessarily mean it is real (note how the specimen also shows a few vertical black stripes under UV. This was likely the glue used to hold the specimen together).

This brings us to the observations Rauhut et al. made on the S.albersdoerferi specimen. The authors used a variety of lighting techniques to get the structures they wanted to highlight, to pop out. In some cases one can tell that there are two different things in place. For instance the bones of S.albersdoerferi fluoresce a nice greenish colour under the UV lighting scheme. Along the tibia, vertebrae, ilium, ischium, and part of the scapula, one can see different colours popping out from the underlying bone. The trouble is determining exactly what those colours represent. The authors argued that the yellow-coloured blobs seen along the base of the tail, were indicative of skin, while similar yellow blobs on the tibia represented muscle. Perhaps there is  some feature of soft tissue that would make it fluoresce this particular colour. However, when looking at the alleged filaments, they fluoresced a more greenish-blue colour. If these filaments were epidermal in nature then I don’t see why they would fluoresce a different colour from the rest of the skin. No other fossil examined under UV seems to show this type of distinction; at least none that I have read about. It seems more likely that the skin Rauhut et al. refer to is more muscle tissue on the base of the tail. This would appear to make sense as it basically looks just like the hypothesized muscle tissue by the tibia. The same could be argued for the apparent filaments above the scapular region (which fluoresce the same yellow colour as the muscle, collagen and “skin”). In that case they could be torn muscle fibres, or collagen fibres. As for the filaments themselves, it is strange that they appear to jut straight out from the bone, with no room for muscle or other soft tissue in between. Once again, if one looks at other dinosaur specimens that are known to preserve filamentous integument (e.g., Sinosauropteryx prima, Caudipteryx zoui, Microraptor gui), there tends to be a halo a few millimeters-centimeters from the bones, indicating where the body wall ended and the epidermal covering began. The only times that the epidermis runs close to the bone are towards the tip of the tail, or at the ends of the limbs, where soft tissue thickness is usually minimal.  We don’t see this in S.albersdoerferi. Instead the filaments seem to come right off the bone. Perhaps this S.albersdoerferi specimen had dried out prior to burial, though given the environment it was in this would seem unlikely.

Sinosauropteryx prima. Note the apparent empty space between the filaments and the bones. This would be where the internal tissues had lain. Photo from the Sam / Olai Ose / Skjaervoy Flickr stream via Wikipedia

Another possibility could be that these apparent filaments are actually preparation artifacts. If so that would make the discovery and description of Sciurumimus albersdoerferi very similar to the recent rediscription of Juravenator starki by Chiappe & Gohlich (2010). The biggest difference is that the alleged filaments for J.starki were much finer and could only be seen under UV using high magnification. In both cases neither specimen preserves unambiguous filaments. As it stands I would be hesitant to use either of these guys as examples of filamentous integument in theropods.

Yutyrannus the tyrannosaur?

As the title of the paper hinted at, Yutyrannus huali popped up as a tyrannosauroid on the dinosaur family tree. This places Y.huali in that nebulous region of Coelurosauria reserved for animals that seem to be closer to Tyrannosaurus rex and Daspletosaurus torosus than to other coelurosaurs. This position is somewhat problematic given what is currently known about dinosaur integument distribution. While Y.huali and the other putative tyrannosauroid Dilong paradoxus both present evidence of a filamentous covering on the body, the later evolved tyrannosaurids T.rex, Tarbosaurus bataar, Albertosaurus sarcophagus and D.torosus all preserve scaly integument.

This suggests one of three things are occurring.

1) Filamentous integument evolved somewhere around the base of Coelurosauria and was later lost at or near the base of Tyrannosauridae

2) Dilong and Yutyrannus are not actually tyrannosauroids

3) Tyrannosaurids are not actually tyrannosauroids

The first scenario is certainly possible. Evolutionary Development (Evo-Devo) tests have shown that making feathers from scales is “easier” than going the other way around (Dhouailly 2009), which would suggest that something about tyrannosaurid lifestyle selected for scaly integument over filaments or bare skin.

The second scenario seems a bit more likely (at least to me). The placement of Yutyrannus hauli in tyrannosauroidea has already been greeted with some initial skepticism. Part of the skepticism lay in the matrix choices the authors used for their placement of Yutyrannus (Choniere et al. 2010). This matrix was missing a lot of data and produced phylogenies that had weak statistical support, as evidenced by Bremer support values of 1 and 2 for many of the major branching points, including Maniraptora. Bremer support values act almost like statistics. They let researchers know how many extra steps it takes in a phylogeny for a relationship to change. The higher the number of steps, the more stable that relationship is (given the characters and taxa used). Typically a Bremer support value of 3 is considered okay, while 5 and up are considered highly supported. The tree used also contained substantial polytomies for groups like Tyrannosauroidea, Carcharodontosauria, and Tetanurae. As a test for the relationship of Alvarezsauroids among Maniraptors the tree did the job, but it was probably not the best choice for establishing tyrannosauroid relationships (especially given that it had weak support for tyrannosauroids in general). So there was that. There was also some question of affinity based on a certain aspect of the skull (a pneumatic midline headcrest). Crests such as this appear similar to carcharodontosaurs like Concavenator corcovatus. The authors noted this in the paper, but also pointed out that a similar crest is also seen in the tyrannosauroid Guanlong wucaii. Nonetheless it was enough to make some researchers consider Y.huali as a potential carcharodontosaur. If true then this would push the origin of filamentous integument closer to the base of tetanurae, a more inclusive group that contains theropods like Allosaurus fragilis. This would make for an even more messy phylogeny, as known skin impressions for Allosaurus show that scales were typical for this animal. Instead of one loss of filaments there would now be two or more losses. While not out of the realm of possibility, this is not a very parsimonious option (of course this is nothing compared to what Sciurumimus suggested, but we’ll get to that).

There is also the chance that Y.huali and D.paradoxus were closer to maniraptors than to tyrannosauroids? This scenario has some merit to it. As mentioned above, the tree choice for fitting Yutyrannus was already suspect, but even if one moves to better phylogenies that are more focused on tyrannosaur relationships (such as Brusatte et al 2010) one still finds weak support for the members of tyrannosauroidea, in particular Dilong, Guanlong and a handful of other guys. Coupled that with at least two relatively recent phylogenies that have moved Dilong closer to Maniraptora (Turner et al. 2007, Lee & Worthy 2011) and it would seem that membership of Tyrannosauroidea is a nebulous one with the potential for one or multiple taxa getting drawn into this portion of the tree based on vague/homoplasious (i.e., convergent) characters.

The third option could happen, but in a roundabout way. Should higher resolution analyses of Tyrannosauroidea wind up popping a lot of its current members into different clades, Tyrannosauridae could be left nested within a clade that is within an (essentially empty) clade, making the name Tyrannosauroidea useless. However given its stem-based definition (Brusatte et al. 2010), the potential usefulness of Tyrannosauroidea suggests it won’t be leaving anytime soon.

This is an area that only future finds and better phylogenetic analyses will help resolve.

Sciurumimus the megalosaur?

Continuing on the theme of questionable phylogenetic affinities, we have the placement of S.albersdoerferi as a basal megalosaurid. If this is correct, and the alleged filaments prove to be real, that would bring filamentous integument down almost to the base of Theropoda, creating a nasty mess for the evolution and loss of filaments in dinosaurs; far worse than if Yutyrannus hauli were a carcharodontosaur. Of course “if” is the operative word here.

While Yutyrannus hauli’s placement was greeted with some skepticism, Sciurumimus albersdoerferi‘s placement as a basal megalosaur has seen substantial more skepticism. There is good reason for this, as S.albersdoerferi is known from a single (albeit well preserved) juvenile specimen. When it comes to juveniles, ontogeny discombobulates phylogeny. Young individuals tend to show plesiomorphic traits that wind up placing them further down a phylogenetic tree than they would otherwise be. This makes dealing with juvenile animals in phylogenies very difficult. Typically the only time one would incorporate juveniles would be if one had a good growth series to work with. That way one could incorporate character changes that were known to happen during ontogeny.  There is no such growth series for S.alberdoerferi. To authors’ credit, they do address this a bit in the supplementary material, stating that they coded all characters that they considered to be ontogenetically variable, as “?” in the matrices.

Speaking of matrices, Rauhut et al. used three of them to test the phylogenetic placement of S.alberdoerferi. The first was from Smith et al. 2008, the second from Choniere et al. 2010, and the last from Benson et al. 2010. I mentioned the problem with Choniere et al. 2010 above, but Smith et al. 2008 suffered from a similar lack of coding completion. Benson et al. 2010 was better, but it did not look at Coelurosaurian relationships. Rauhut et al. used this last paper for their official analysis of S.alberdoerferi after “confirming” its placement as a basal megalosauroid in the previous analyses. The trouble with using matrices that have such large amounts of missing data, is that it can be very easy to move taxa around a tree with just one or two characters. The nebulous nature of S.albersdoerferi is made evident in the supplementary information of the paper as the authors repeatedly mention that this taxon’s placement had exceedingly low statistical support (bootstrap values below 50%. For reference a bootstrap value of 80% is considered minimal for good support). I find it strange that Rauhut et al. argued that because they were able to have Sciurumimus albersdoerferi place around megalosauroids in three separate analyses, it should count as strong support for its placement lower in tetanuran phylogeny. That all three analyses showed weak statistical support for most nodes suggests that these matrices are not that robust. Further let’s not forget that a bootstrap analysis takes that phylogeny, mixes things up and runs it again. That S.albersdoerferi and other theropods in these matrices were in the positions Rauhut found them in less than 50% of the time suggests that they either probably don’t belong there, or that the matrices used are too inadequate to resolve these relationships. Andrea Cau over at the Theropoda blog  input S.albersdoerferi in his “megamatrix” for theropods and did not have any trouble getting this little guy to nest comfortably in Coelurosauria (though I’m not sure with how much support). While this is not a published analysis, I would not be surprised to see future papers finding S.albersdoerferi to be a coelurosaur in the near future.

The feather-scale dichotomy

The location of preserved integument on the three specimens of Y.hauli provides insight into filament locations in dinosaurs. As previously mentioned, we have extensive data on preserved integument in most representative dinosaur species. Among species with preserved integument the vast majority exhibit scales. It is really only in coelurosaurs (and even then it is typically maniraptors) that we see the move towards filaments instead. Among this large dataset are the theropods Allosaurus (Pinegar et al 2003), Carnotaurus (Czerkas & Czerkas 1997), Gorgosaurus, Daspletosaurus, Albertosaurus, Tarbosaurus (Currie et al. 2003) and Tyrannosaurus. The last one, T.rex, comes from the the Black Hills Institute and their “Wyrex” specimen. More info (including skin picture) can be found here. Given such a large assortment of data showing the presence of scales, why is there such a push to put filaments if not full-on feathers on T.rex and other scaly dinosaurs?

The relationship between feathers and scales seems deceptively simple on the outset (left, from here). When viewed in more detail, such as in this beat-up Vietnamese chicken (right, from here), the relationship is more complicated. Note the lack of scales anywhere but on the tarsometatarsus.

One likely reason has to do with the scale impressions themselves. All of these scale impressions have come from isolated patches on the body, typically near the pelvis or in the thoracic region. These isolated pockets have lead some folks to question how extensive this scaly covering was. That is to say, some have argued that just because scales are seen on one portion of the body (e.g., the belly), that doesn’t mean that it was representative of the entire body, as scales could have been present on patches of the body that showed feather loss due to large size. This very argument has been made by both Xu et al. for Y.hauli and Rauhut et al. for S.albersdoerferi. The problem I have with this argument—much like the arguments from ontogeny—is that it routinely assumes that scales and naked skin are the same thing. Nothing could be further from the truth. Scales are a unique form of integument akin to hair and feathers, nails, and claws. Evo-Devo studies suggest that feathers evolved from scales by “hijacking” the scale developmental pathway early in the process (Somes 1990, Sawyer & Knapp 2003, Sawyer et al. 2005, Dhouailly 2009). Further, their relationship to feathers is such that the two integumentary types appear to be mutually exclusive. Integument forms through a series of cascading events, with integument in extant archosaurs occurring in a series of waves starting with the main body (trunk, tail, proximal limbs), then the neck, head and terminal limbs (Widelitz et al. 2000, Alibardi & Thompson 2001). Given this knowledge one may able to reasonably infer extent of integument based on where a skin impression was found. If it was found on the foot, or tarsometatarsal region one should be able to infer that integument was present along the foot and ankle, leaving the rest of the body as an unknown. Should the skin impression be found on the tail, pelvis, or thoracic region though, one should be able to reasonably infer that it was representative of the integument found across the body up to the head, ankle, and wrist regions. Evo-Devo data on scale and feather formation in birds (Sawyer and Knapp 2003, Dhouailly 2009) suggests that feathers covered the entire body when they first appeared. This prediction has so far been borne out by early feathered dinosaurs like Microraptor gui (Xu et al. 2003) and Anchiornis huxleyi (Hu et al. 2009) as well as the “first bird*:” Archaeopteryx lithographica (Christiansen & Bonde 2004).

This awkward image of an ostrich highlights the lack of scales on normally unfeathered regions of the body. Image from: Beautiful African Wild Animals Pets blog.

This all-or-none principle and antagonistic relationship of feathers and scales has been doubted in the paleo literature, with some authors arguing that Evo-Devo studies actually support a hodge-podge scale-feather relationship (Chiappe & Gohlich 2010, Rauhut et al. 2012). However, the references that are often cited for this do not actually support this view. Some such as Chang et al 2009, Widelitz et al. (2000, 2003) and Dhouailly et al. (1980) point to evidence of latent feather forming ability in avian scales. This ability was discovered through manipulation of scale epidermis by incorporating different compounds (e.g., B-Catenin, Retinoic Acid) during development. Results of these studies have shown that feathers can be induced from scale epidermis at certain times during development. However just because a feather was developing does not indicate that scales and feathers made good bedfellows. Quite the opposite in fact. As Dhouailly (1980) states:

It is clear, however, in view of the existence of domestic breeds of fowl with feathered feet (genetic ptilopody), that RA [Retinoic Acid] somehow interferes with scale morphogenesis and thereby reveals  a latent ability of avian foot integument to produce feathers. Apparently the formation of scales requires additional and possibly inhibitory region-specific information on top of the trivial and ‘ubiquitous feather message’ (Dhouailly,1978). Retinoic acid,by weakening the scale message, would leave the feather message free to be expressed.

To put it another way: scale formation in birds requires active feather suppression.

Some rather recent work on avian scale histology and development has uncovered yet another twist in the feather-scale dynamic. In general, avian scales are composed of three morphological types:

  1. Scuta – Large overlapping scales that cover the dorsal side of the tarsometatarsus and the foot digits
  2. Scutella – Somewhat smaller overlapping scales that cover the ventral side of the tarsometatarsus
  3. Reticula – Small non-overlapping scales that form on the underside of the foot

Whereas scuta and scutella contain the standard scale histological makeup (largely β-keratin with an α-keratin hinge), the reticula have been found to be composed of only α-keratin. Developmentally, reticula neither form from dermal condensations (a requirement for most other epidermal appendages) nor from a placode (as avian feathers and scales form from).  These data suggest that reticula are not true scales. So if reticula are not scales then what are they?

Feathers. Yep, feathers.

From Dhouailly 2009:

…reticula are not true cutaneous appendages, and appear to be feathers arrested in the initiation step of their morphogenesis: formation of a slight bump, without a placode.

…reticulate scales, which cover the plantar surface of all living birds, cannot be the remnants of the ancestral granulated beta-keratinized skin of first sauropsids, and correspond to a secondary, almost complete, inhibition of feather formation.

Reticula—these little guys—may actually be primordial feathers!

Bringing this all together it would seem that the most parsimonious integumentary covering for a dinosaur with known skin impressions,  like T.rex, would still be scales. The discovery of Yutyrannus huali further supports this view as filamentous impressions are found along the pelvis and tail region; areas that were previously argued to be scaly in large theropods when based on skin impressions.

So once again it would seem that the concept of the dinosaur with a feathered mohawk gets relegated to the fiction bin.

Yeah this is probably also fiction. Image by: Sammy Hall

* I put first bird in quotes to reflect the recent fuss over the relationship of Archaeopteryx to extant birds.

“Wooly” theropods of the “chilly” Mesozoic

One of the appealing aspects of a Xu et al. article is that the authors tend to cover their bases as best they can while trying to avoid speculating beyond the data. This article is no different. The authors entertain alternate reasons behind the presence/function of the filaments. One of those aspects has receive a bit more attention than it probably should have.

From the article:

Alternatively, if scales were indeed the dominant integumentary structures in most Late Cretaceous tyrannosauroids, the presence of long feathers in the gigantic Y. huali could represent an adaptation to an unusually cold environment.

As one can see from the image at the beginning of this post, this aspect of the paper has been blown a bit out of proportion. There is some indirect evidence to indicate that the Yixian was a time of unusual (for the Mesozoic) cold temperatures (Amiot et al. 2011). However—and this is the part the news organizations and blogs don’t mention—those temperatures aren’t all that cold.  Amiot et al. determined that the mean annual air temperature for the environment of the Jehol Biota was around 10°C (50°F). For perspective that is the same mean annual temperature of present day Ohio. These are hardly arctic conditions that we are talking about here (in fact Ohio gets bloody hot in the late Spring and Summer).

So it seems a bit weird that in what was likely a balmy to slightly cold environment (during part of the year) a 1.4 tonne theropod would be sporting a “wooly coat.” Also, while the preserved filaments were fairly long and shaggy (15-20cm [~6-8 inches]) that is well within the range of modern emu (Dromaius novaehollandiae) which lives in hotter environments than those predicted for the Jehol Biota. They are also a far cry from the actual wooly coats of mammoths and wooly rhinos (up to 0.91 meters [3 ft] long in the former). This brings us to another important thing to consider. Is the trend we see in mammals (thicker coats in winter, wooly coats for arctic animals) a physiological response to temperature, or a mammalian physiological response to temperature?

Yeah, these guys have thicker feathers than Y.huali, and they currently live in a hotter climate. Image by Benjamint444 under GFDL license from Wikipedia.

Put another way: is there such a thing as a wooly bird? Surprisingly a survey of the literature seems to indicate that birds don’t really change their feather thickness with latitude, or climate. While all birds molt, molting frequency and timing are taxon specific, with only some species having winter molts. It is generally thought that birds that molt prior to winter do so in preparation for the cold. Thus their “winter coat” is thicker than their summer coat. However there is very little analytical data to back this up. Irving et al. (1955) looked at a variety of birds from various latitudes and found no real difference in contour feather thickness.

Perhaps this is to be expected though. Unlike mammals, bird feathers grow on specialized regions of the body (tracts). These tracts limit the locations for new feathers to form. Feather tracts appear to make it difficult for a bird to put on more feathers at any one time of the year. This apparent limit to feather placement is borne out by the only observation I could find for a “winter coat” in birds.

The willow ptarmigan and Northern grouses do seem to put on thicker coats of feathers in the winter. They bypass the feather tract problem by making unique feathery structures referred to as “afterfeathers.”

After feathers are a secondary structure produced from the main axis, or rachis of the contour feather, having a plumulaceous barb and barbule structure greatly enhancing insulation…. The afterfeathers of ptarmigan in winter are three-fourths as long as the feathers from which they are derived, adding considerably to the thickness of the body plumage. – Marchand 1996

So it would appear that Yutyrannus hauli was unlikely to have been a “wooly theropod.” It was probably just a filament-covered coelurosaur that just so happened to get big.

Filaments and metabolism

Despite being the hardest and least likely thing to ever find evidence for, dinosaur metabolism continues to feature prominently in dinosaur studies. Finds such as this, that involve potentially insulative coverings, probably draw the most attention to the ultimate question of what “metabolic camp” dinosaurs were in. This is due in most part to the oft held assumption that insulation only benefits animals that produced heat internally. After all insulation by itself does not confer warmth but simply retards the flow of heat to or from a body (this is the reason why a thermos works in the summer and in the winter). If an ectothermic animal were to have insulation, so the argument goes, it would be unable to effectively raise its body temperature in the sun. There is surprisingly little empirical backing for this logic, most of it stems from mathematical models that predict heat flow in stationary (typically geometric) objects. There are extremely few physiological tests out there. This may help explain why one of the most popular physiological tests of insulation is that of Raymond Cowles (1958) and his study on dermal temperature regulation in amphibians and reptiles.

Part of Cowles’ paper made reference to a class project that Cowles had performed (or rather, had one of his students [Richard Grossman] perform) in which two heliothermic lizards—a desert iguana (Dipsosaurus dorsalis) and a chuckwalla (Sauromalus obesus)—were wrapped in a mink coat. Yep, you heard right, lizards wearing mink. The animals were then warmed up for 18 and 12 minutes respectively. Afterwards the animals were chilled to 15°C (59°F) before being reheated for 69 minutes and then chilled again. The results Cowles’ student obtained were a noticeable lag in heating and cooling. These heliothermic lizards were taking longer to warm up, but they were also taking longer to cool down. From this Cowles concluded that insulation would be a detriment to an ectothermic animal.

…animals might tend to profit from insulation by their retention of heat when once warmed,but this advantage would in turn be cancelled by the retardation in warming rates. For diurnal heliothermic animals there might be some extension of activity during the latter part of each day ,but a delayed or late start each morning would incur a penalty. For nocturnal animals there would seem to be similar cancellations of advantages and disadvantages, the disadvantages being aggravated by evaporative cooling if the insulated covering had absorbed either dew or rain. – Cowles 1958

And so the party line has stuck. However looking at Cowles’ original paper there are some noticeable flaws in the experiment that could greatly change the outcome. For starters, this experiment was performed by a student for a class project. This is not mean to knock Mr. Grossman’s work, but the fact that it was a class experiment meant that Cowles never bothered to write up the materials and methods for the experiment. Cowles only ever mentions that the animals were encased in “a crude ‘wrap-around’ of mink!” No mention is given to how tightly wrapped the lizard were, how extensive the covering was, or even how thick the mink covering was. All of these would be essential to an accurate assessment of the effect of insulation on an ectothermic animal. Cowles does not go into detail about the heating and cooling experiment either; only giving the reader two figures which mention a heating source, but not what that source was, or how extensive it was (are we talking about a spotlight?, a heat rock?, some form of heat tape?, multiple spotlights?, etc.). We are also not told how long the animals were allowed to acclimate, or what conditions they were allowed to thermoregulate in (thus letting us know how stressed the lizards may have been). All of these factors could greatly affect the outcome of the experiment. That none of this was really followed through makes this experiment really more of an anecdote more than anything else (indeed it isn’t even the main thrust of the paper).

A greater roadrunner (Geococcyx californianus) warming itself in the morning using ectothermy. Image by Greg Page

Methodological problems aside, Cowles’ assumption that insulative coverings are inert objects is also untrue. Anyone who has seen a scared cat, watched a dog dry off, or has had goosebumps knows that filamentous integument can be manipulated muscularly. This makes filamentous integument a dynamic addition to the body, allowing for increased, or decreased insulation at a moments notice. This swapping of insulative abilities has been demonstrated in the wild with roadrunners (Geococcyx californianus) and turkey vultures (Cathartes aura), both species which are known to employ ectothermic “sunning” postures to warm up in the morning (Ohmart & Lasiewski 1971, Clark & Ohmart 1985, Ruben & Jones 2000).

Then there is the other side of the coin, bradymetabolic animals that do have insulation. Moths, bees, caterpillars, tarantulas and leatherback sea turtles (Dermochelys coriacea) to name but a few. In some animals like the moths and leatherback, insulation serves to retain metabolically produced heat (but via muscular action rather than visceral as seen in mammals and birds), in other animals like the thousands of hairy spider and caterpillar species, the integument seems to act as a sense organ, or a deterrent to predators. Then there is the crazy “yeti crab” (Kiwa hirsuta) which uses its integument for…something we have yet to identify.

The importance of integument to bradymetabolic thermoregulators was also demonstrated by Seebacher (2003). The author used a mathematical model based on thermoregulating saltwater (Crocodylus porosus) and freshwater (C.johnstoni) crocodiles to estimate the average daily body temperatures of 701 dinosaurs operating under the assumption that they were bradymetabolic, ectothermic animals that thermoregulated. Under this model Seebacher found that body temperature fluctuations were greatest only in small (<100kg [220lbs]) dinosaurs living at mid-high latitudes. Yet, by simply incorporating insulation into animals like Sinosauropteryx prima (estimated body mass of 3.8kg [8.4 lbs]), body temperatures practically stabilized (variations of ≤3.1°C [~6°F]) while core body temperature increased (from 23.8°C [75°F] without insulation to 32.4°C [90°F] with insulation). Seebacher’s model was fairly conservative and incorporated some aspects from modern times (e.g., daily temperature flux) that may not have been true back in different epochs of the Mesozoic. Yet with these caveats in mind, his model was able to show the effectiveness of insulation for even a bradymetabolic animal.

Hypothetical core temperature flux in a 3.8kg Sinosauropteryx prima living at 50° Latitude in the Early Cretaceous. Adapted from Figure 8(C) in Seebacher 2003.

So with all this in mind, what does the presence of extensive filaments on Y.hauli tell us? It tells us that the animal had a layer of insulation around it that would have aided in maintaining core body temperature. How that temperature was obtained and maintained remains unknown. A key thing to remember is that given the location and currently estimated paleoclimate, an insulative covering would have been advantageous regardless of thermophysiology.



Alibardi, L., Thompson, M. 2001. Fine Structure of the Developing Epidermis in the Embryo of the American Alligator (Alligator mississippiensis, Crocodilia, Reptilia). J.Anat. Vol.198:265-282
Amiot, R., Wang, X., Zhou, Z., Wang, X., Buffetaut, E., Lecuyer, C., Ding, Z., Fluteau, F., Hibino, T., Kusuhashi, N., Mo, J., Suteethorn, V., Wang, Y., Xu, X., Zhang, F. 2011. Oxygen Isotopes of East Asian Dinosaurs Reveal Exceptionally cold Early Cretaceous Climates. PNAS. Vol.108(13):5179-5183
Benson, R.B.J., Carrano, M.T., Brusatte, S.L. 2010. A New Clade of Archaic Large-Bodied Predatory Dinosaurs (Theropoda: Allosauroidea) That Survived to the Latest Mesozoic. Naturwissenschaften Vol.97:71–78
Brusatte, S.L., Norell, M.A., Carr, T.D., Erickson, G.M., Hutchinson, J.R., Balanoff, A.M., Bever, G.S., Choiniere, J.N., Makovicky, P.J., Xu, X. 2010. Tyrannosaur Paleobiology: New Research on Ancient Exemplar Organisms. Science. Vol.329:1481-1485
Chang, C., Wu, P., Baker, R.E., Maini, P.K., Alibardi, L., Chuong, C-M. 2009. Reptile Scale Paradigm: Evo-Devo, Pattern Formation and Regeneration. Int.J.Dev.Biol. Vol.53:813-826
Chiappe, L.M., Gohlich, U.B. 2010. Anatomy of Juravenator starki (Theropoda: Coelurosauria) from the Late Jurassic of Germany. N.Jb.Geol.Palaont.Abh. Vol.258(3):257-296
Choiniere, J. N. Xu, X, Clark, J.M., Forster, C.A., Yu G., Fenglu H. 2010.  A Basal Alvarezsauroid Theropod from the Early Late Jurassic of Xinjiang, China. Science. Vol.327:571-574
Christiansen, P., Bonde, N. 2004. Body Plumage in Archaeopteryx: A Review and New Evidence from the Berlin Specimen. C.R.Pale. Vol.3:99-118
Clark, R.G., Ohmart, R.D. 1985. Spread-Winged Posture of Turkey Vultures: Single or Multiple Function? Condor. Vol.87:350-355
Cowles, R.B. 1958. Possible Origin of Dermal Temperature Regulation. Evolution Vol.12(3):347-357
Currie, P.J., Badamgarav, D., Koppelhus, E.B. 2003. The First Late Cretaceous Footprints from the Nemegt Locality in the Gobi of Mongolia. Ichnos. Vol.10:1-12
Czerkas, S. A., Czerkas, S.J.. 1997. The Integument and Life Restoration of Carnotaurus. in Wolberg, D.L., Rosenberg, G.D. (eds.), Dinofest International, Proceedings of the Symposium at Arizona State University, pp. 155–158. Philadelphia Academy of Natural Sciences, Philadelphia.
Dhouailly, D. 2009. A New Scenario for the Evolutionary Origin of Hair, Feather, and Avian Scales. J.Anat. Vol.214:587-606
Dhouailly, D., Hardy, M., Sengel, P. 1980. Formation of Feathers on Chick Foot Scales: A Stage-Dependent Morphogenetic Response to Retinoic Acid. J.Embryol.Exp.Morph. Vol. 58:63-78
Hu, D., Hou, L., Zhang, L., Xu, X. 2009. A Pre-Archaeopteryx Troodontid Theropod from China with Long Feathers on the Metatarsus. Nature. Vol.461:640-643
Irving, L., Krog, H., Monson, M. 1955. The Metabolism of Some Alaskan Animals in Winter and Summer. Phys.Zool. Vol.28(3):173-185
Lee, M. S. Y., Worthy, T.H.. 2011. Likelihood Reinstates Archaeopteryx as a Primitive Bird. Biology Letters doi: 10.1098/rsbl.2011.0884
Marchand, P.J. 1996. Life in the Cold. 3rd edition. University Press of New England. p.234
Ohmart, R.D., Lasiewski, R.C. 1971. Roadrunners: Energy Conservation by Hypothermia and Absorption of Sunlight. Science. Vol.172:67-69.
Ortega, F., Escasco, F., Sanz, J.L. 2010. A bizarre, Humped Carcharodontosauria (Theropoda) from the Lower Cretaceous of Spain. Nature. Vol. 467:203-206
Pinegar, R.T., Loewen, M.A., Cloward, K.C., Hunter, R.J., Weege, C.J. 2003. A Juvenile Allosaur with Preserved Integument from the Basal Morrison Formation of Central Wyoming. JVP. vol.23(3):87A-88A
Prum, R.O. 1999. Development and Evolutionary Origin of Feathers. J.Exp.Zool.(Mol.Dev.Evol) Vol.285:291-306
Rauhut, O.W.M., Foth, C., Tischlinger, H., Norell, M.A. 2012. Exceptionally Preserved Juvenile Megalosauroid Theropod Dinosaur with Filamentous Integument from the Late Jurassic of Germany. PNAS Early Edition:1203238109v1-201203238.
Ruben, J.A., Jones, T.D. 2000. Selective Factors Associated with the Origin of Fur and Feathers. Amer.Zool. Vol.40:585-596
Sawyer, R.H., Knapp, L.W. 2003. Avian Skin Development and the Evolutionary Origin of Feathers. J.Exp.Zool.(Mol.Dev.Evol) Vol.298B:57-72
Sawyer, R.H., Rogers, L., Washington, L., Glenn, T.C., Knapp, L.W. 2005. Evolutionary Origin of the Feather Epidermis. Dev.Dyn. Vol.232:256-267
Sawyer, R.H., Washington, L.D., Salvatore, B.A., Glenn, T.C., Knapp, L.W. 2003. Origin of Archosaurian Integumentary Appendages: The Bristles of the Wild Turkey Beard Express Feather-Type B Keratins. J.Exp.Zool.(Mol.Dev.Evol) Vol.297B:27-34
Seebacher, F. 2003. Dinosaur Body Temperatures: The Occurrence of Endothermy and Ectothermy. Paleobiology. Vol.29(1):105-122
Smith, N.D., Makovicky, P.J., Agnolin, F.L., Ezcurra, M.D., Pais, D.F., Salisbury, S.W. 2008. A Megaraptor-Like Theropod (Dinosauria: Tetanurae) in Australia: Support for Faunal Exchange Across Eastern and Western Gondwana in the Mid-Cretaceous. Proc.Biol.Sci Vol.275:2085–2093
Somes, R. G. 1990. Mutations and Major Variants of Plumage and Skin in Chickens. in “Poultry Breeding and Genetics” Crawford, D.R. (ed). Elsevier, New York
Turner, A.H., Pol, D., Clarke, J.A., Erickson, G.M., Norell, M.A. 2007. A Basal Dromaeosaurid and Size Evolution Preceding Avian Flight. Science. Vol.317:1378-1381
Widelitz, R.B., Jiang, T-X., Lu, J., Chuong, C-M. 2000. B-Catenin in Epithelial Morphogenesis: Conversion of Part of Avian Foot Scales into Feather Buds with a Mutated B-Catenin. Devl.Biol. Vol.210:98-114
Widelitz, R.B., Jiang, T.X., Yu, M., Shen, T., Shen, J-Y, Wu, P., Yu, Z., Chuong, C-M. 2003. Molecular Biology of Feather Morphogenesis: A Testable Model for Evo-Devo Research. J.Exp.Zool.(Mol.Dev.Evol). Vol.298B:109-122
Xu, X., Kebai, W., Ke, Z., Qingyu, M., Lida, X., Sullivan, C., Dongyu, H., Shuqing, C., Shuo, W. 2012. A Gigantic Feathered Dinosaur from the Lower Cretaceous of China. Nature. Vol.484:92-95
Xu, X., Zhou, Z, Wang, X., Kuang, X., Zhang, F., Du, X. 2003. Four-Winged Dinosaurs from China. Nature Vol.421:335-340
Facebooktwittergoogle_plusredditpinterestlinkedintumblrmailby feather

65 Responses to “Feathers” on the big, “feathers” on the small, but “feathers” for dinosaurs one and all?

  1. Avatar pilsator
    pilsator says:

    Very interesting post, Jura!

    I have been under the impression that evo-devo results suggest that the presence of feathers or scales on the body of any given theropod is mutually exclusive, but that pretty much all (proto)feathered taxa do show both feathers and scales (or rather, scutes, scutella and reticula).

    So if reticula are basically feathers, who says scales in some non-avialan dinosaurs might not be reticula in disguise? Are there morphological features telling the two apart, or is it a completely molecular thing? If we assume for a moment this were true, and that there are no morphological telltale signs to keep the two apart (correct me if I’m wrong), it might be that – playing devil’s advocate – scales in any ornithodiran might rather represent reticula.

    Sorry for knowing next to nothing about evo-devo stuff – is there any literature testing/refuting the homology of any kind of avian integument with crocodilian scales/scutes?

    • This is certainly possible. Bird scuta and croc tarsal scutes look awfully similar in the extant animals. They only differ in their embryology. I don’t think the resolution we get on skin impressions is enough to let us know one way or the other. At best we might be able to say whether scales were most likely primary or secondarily derived based on nearest common relatives. So in the case of tyrannosaurids, if Yutyrannus and Dilong are sitting on the right spot in the phylogeny, then tyrannosaurid scales were probably secondarily derived. This leads to interesting follow-up questions such as: what are the selection forces behind re-evolution of scales, or conversely, why don’t birds ever show selection for scale formation in secondarily featherless regions? A thorough comparison of skin impressions across Dinosauria might shed light on potential morphological differences between primary and secondary scales. I seem to remember something like this being published relatively recently, but it was only for hadrosaurs.

      As it currently stands, scales in dinosaurs, and possibly some non-crocodile-line pseudosuchians and other archosaurs, have the potential to be secondarily derived.

  2. Avatar Mathieu
    Mathieu says:

    Fantastic and critical examination of scales and feathers in dinosaurs, Jura! But I worry that this message is not reaching enough people, as memes such as all dinosaurs being feathered gather momentum and content at a frankly alarming rate, while the scientific skepticism that informed them is forgotten. I think that the harsh resistance you were met with when you commented on Brian Switek’s article regarding feathered tyrannosaurs spoke for itself. I think this meme of “most dinosaurs had feathers” in the paleontological community could actually be reaching a crisis point, where some of the more zealous feather lovers are starting to seriously consider fuzzy sauropods (with VERY scant, implicit evidence) more scientifically valid than a scaly Tyrannosaurus. Don’t get me wrong, I think it’s incredible that we share a world with living dinosaurs in the form of birds. They are an evolutionary dynasty that is totally global in scope, and is literally older than the continents on which we stand. Birds being dinosaurs means that falconry is a sport of cooperation between man and dinosaur, and this, to me, is absolutely wonderful. But like you said on Brian Switek’s article, let the evidence speak for itself, and paleo-artists saying that their Baryonyx with plumage is anything other than speculation, and somehow more scientifically accurate than the scaled renditions is far from ideal. But to the point of this long winded message, I worry that your well written articles regarding dinosaur integument are not reaching a broad enough audience. I urge you to consider starting a tumblr, where posts will be quickly reblogged and discussed amongst the paleo-communtiy there (met with resistance, sure, but at least discussed and seen by a wider audience of content producers) or even a narrated video posted to youtube, where eventual circulation is all but guaranteed.
    Once again, thank you very much for your intelligent and measured examination and discussion of dinosaur integument.
    Best Wishes, Mathieu.

    • Thanks for the support. I do think that we are in the middle of an over-feathering trend. Paul Barrett had a good poster at SVP that did critically examine the case for filament homology in dinosaurs. There was a news article on it a while back, but no paper as of yet. He did not find support for homology of filaments between pterosaurs, coelurosaurs, and marginocephalians. He also did not find support for just coelurosaurs and marginocephalian filaments. So there is a technical article (eventually) on the way that should help apply the brakes to some of this. This pendulum-like movement of dinosaurs between “classic reptiles” and birds has happened before. Dino art in the early to mid-nineties were very birdy looking. Things got less bird-like in the early 00’s, and now it’s swinging back towards birds again. On the bright side, the pendulum swings are always a little less each time.

      Thanks for the Tumbler suggestion. While I am aware of Tumblr and occasionally follow people on it, I haven’t thought much about moving my articles to it. I like the reblogging idea, and more exposure would be nice. I might have to look into what all that would take.

  3. As an artist who loves to draw dinosaurs, I once enjoyed depicting Tyrannosaurus rex and kin with feathers. Having recently re-examined the evidence you’ve cited, I’m no longer convinced of this “feathered tyrannosaurid” trope which has almost become the new orthodoxy in the paleo-fandom. That all the skin impressions recovered from *four* tyrannosaurid genera show featherless skin is really inconvenient for the feather-nazis’ case among other things. Thank you so much!

    Brandon S. Pilcher

  4. Sorry for the double post, but another thought occurred to me: if derived tyrannosaurids could evolve a secondary loss of feathers (or filamentous integument) as you’ve shown, couldn’t the same have happened in other theropod lineages now commonly portrayed as feathered? Take certain dromaesaurids (“raptors”) for instance. Unless the “advanced” feathers of maniraptorans would have been less expendable than the proto-feathers of basal tyrannosauroids, I don’t see why secondary feather absence couldn’t have happened in a minority of dromaeosaurids too. It may not seem likely at this stage and positive evidence in favor of it may not exist yet, but the fossil record always yields surprises.

    I wonder if the scalier skin in tyrannosaurids had something to do with body protection against dangerous prey? Do you know if reptilian scales would confer more protection than filamentous integument?

    • Sorry for my delayed response. Yes, it is certainly possible that dromaeosaurs could have re-evolved scales in place of filaments in a similar vein to that proposed for tyrannosaurids. When it comes to reversals it is generally thought that it is easier to go back to a previous path if one has not gone too far down the new path. So I tend to think animals that were just starting to experiment with filamentous integument might have had it easier reverting back to a scaly integument. However, despite this “rule of thumb,” the dromaeosaurid known as NGMC-91, more colloquially known as “Dave,” does show evidence that such a reversal did occur in the tarsometatarsus and feet of dromaeosaurs (since “Dave” appears later in dromaeosaur evolution than the feathered-hind-limbed Microraptor and Anchiornis). Evo-Devo studies show that at some point in bird evolution a similar reversal happened to their distal hindlimbs too (possibly as late as Neornithes). So it is possible that there was a species or family of dromaeosaurs (or other maniraptors) that dropped the filaments in favour of scales.

      As for protective properties of scales vs. filaments, I don’t know. Extant birds have almost paper-thin skin, and so rely heavily on their feathers for protection (and it seems to work just fine). The extensive beta-keratin covering on reptile scales seems to work as a good protective barrier for extant reptiles (in contrast, scaleless, alpha-keratin heavy Leatherback turtles tend to get badly scuffed up when crawling up beaches to lay eggs). Scales might be easier to keep clean and dry, which might explain why birds re-evolved them on their feet. Short of that I think this is a currently unanswered question that is ripe for study.

  5. Thank you for your answer. Given how adamant so many paleo-enthusiasts are about feathering up every dromaeosaurid that ever lived, it is refreshing to hear that at least in theory it would be possible for a few dromaeosaurs to revert to scales, for whatever reason.

    That said, when bringing up those tyrannosaurid scale impressions in a discourse on DeviantArt, someone mentioned that taphonomic factors might come into play. They claimed that the sediments which contain most tyrannosaurid fossil aren’t conducive to preserving feathers and so would only show scaly or naked skin impressions. Honestly I am not sure how much I buy that line of argument.

    • That argument pops up from time to time. The logic is sound. The only reason we know about filaments in some dinosaurs is due to the unique settings of their burials. The taphonomy of soft-tissue preservation is still in its infancy. Up until two years ago there was very little interest in what sedimentological settings were conducive to soft-tissue preservation. It was just generally assumed that the conditions had to be extremely unique and hence extremely rare. However, the recent discovery of filaments (I think the authors might have been jumping the gun a bit about complete feathers) in Ornithomimus specimens preserved in sandstone (Zelenitsky et al. 2012) has increased the potential rock types that could preserve soft tissues quite a bit. Prior to this discovery it was generally thought that requirements of sandstone formation (usually flooding or other high-energy conditions) were not compatible with soft-tissue preservation. Similarly, the discovery of a potential fleshy crest on Edmontosaurus (also preserved in sandstone) has also called into question what we originally thought about soft-tissue burial requirements (Bell et al. 2013). This is a burgeoning area of research that will likely change the way we collect fossils in the future.

      The other thing I usually notice about this argument is that it tends to make the mistake of associating scales with skin. That is proponents tend to think that because a fossil doesn’t preserve filaments then somehow that means it is preserving scales. However the conditions that preserve scales are pretty much identical to those that preserve filaments. A classic example that I’ve seen bandied about is the presence of feathers on Archaeopteryx but nothing on Compsognathus, yet both were preserved in the same sediment. Therefore we would assume that Compsognathus was scaly because no feathers were preserved. It’s an argument from negative evidence, which is always dangerous. It also ignores the fact that Compsognathus preserves no soft tissue at all. No scales, no skin, no internal organ stains, nothing. In contrast, the argument for scales on tyrannosaurids comes from the soft-tissue preservation of scales in a few fossils. It’s not that tyrannosaurids were scaly because filaments are not preserved, it’s because we have actual scales impressions to back up the claim. There isn’t really any example I can think of that shows the preservation of scales over feathers or vice versa. Where the soft-tissue gets preserved will vary, and that can effect our interpretations, but I haven’t really seen a case where one form of integument was favoured over another. You either preserve both or you preserve neither.


      Zelenitsky, D.K., Therrien, F., Erickson, G.M., Debuhr, C.L., Kobayashi, Y., Eberth, D.A., Hadfield, F. 2012. Feathered Non-Avian Dinosaurs from North America Provide Insight into Wing Origins. Science Vol. 338(6106):510–514.

      Bell, P.R., Fanti, F., Currie, P.J., Arbour, V.M. 2013. A Mummified Duck-Billed Dinosaur with a Soft-Tissue Cock’s Comb. Current Biology.

  6. By the way, do you think it possible that Nanuqsaurus (the recently discovered polar tyrannosaurid) might have re-evolved filaments as protection against the Alaskan wintry cold? I recently reconstructed it as scaly but possessing a lot of body fat like polar bears, but a commenter told me that dinosaurs and other diapsids lost the UCP1 gene that makes brown fat. If Nanuqsaurus lacked filaments and brown fat, how would it have kept warm?

    • UCP1 is pretty much a mammal-only protein. Birds have a similar protein known as UCP3, which seems to function in much the same way. It was long thought that reptiles lacked uncoupling proteins (since ectotherms would, theoretically, not need them). However work by Schwartz et al. (2008) and Rey et al. (2008) found that reptiles do have uncoupling proteins. Crocs have homologues of UCP2 and UCP3, whereas an analogue to UCP1 has been found in the lizard Zootoca vivipara (one that seems to have evolved from the same uncoupling protein family). All of these uncoupling proteins function during cold acclimation in all of these taxa, though UCP1 is the only one known to affect brown fat deposits (note: it doesn’t make brown fat, it burns it). Brown fat is typically thought of as a mammalian thing. It allows for non-shivering thermogenesis, which is a cheaper means of warming up than plain old shivering. No bird has been found to have brown fat (so far), and the means by which they create heat without shivering is still debated. However, swordfish and leatherback sea turtles do have brown fat. In fact leatherbacks have a tonne of it. So putting brown fat on Nanuqsaurus is not entirely out of the question.

      Nanuqsaurus could have also just used muscle power to keep itself warm too. Alternatively it could have simply operated at a lower temperature in the winter than in the summer. Work by Mark Goodwin and others (2011) found that regular old T.rex also seemed to operate at different core body temperatures throughout the year. As long as the proteins and enzymes are geared towards that operating temperature the animals shouldn’t suffer any performance drops (e.g., saltwater crocs acclimated to winter temperatures are able to move just as fast as they do in the summer [Glanville and Seebacher 2006]).

      It’s also pertinent to keep in mind that Cretaceous Alaska may have been just as dark as modern day Alaska, but it was a heck of a lot warmer. In fact it was warmer than areas today that house native reptiles and amphibians (e.g., Sweden). Edmontosaurus and Centrosaurus are known from these polar latitudes and their skin impressions also show a complement of scales. So, regardless, living up North back then did not require a dinosaur to be fuzzy.


      Glanville, E.J., Seebacher, F. 2006. Compensation for Environmental Change by Complementary Shifts of Thermal Sensitivity and Thermoregulatory Behaviour in an Ectotherm. J. Exp. Biol. Vol. 209:4869–4877.

      Goodwin, M., Stanton, K., Horner, J., Carlson, S. 2011. Oxygen Isotopic Variability and Preservation in Tyrannosaurus rex, Modern Ratites and Crocodylians: Revisiting the Thermophysiology of T.rex Using δ18O. JVP. Vol. 31(Sup 2):117A.

      Rey, B., Sibille, B., Romestaing, C., Belouze, M., Letexier, D., Servais, S., Barre, H., Duchamp, C., Voituron, Y. 2008. Reptilian Uncoupling Protein: Functionality and Expression in Sub-Zero Temperatures. J. Exp. Biol. Vol. 211:1456–1462.

      Schwartz, T.S., Murray, S., Seebacher, F. 2008. Novel Reptilian Uncoupling Proteins: Molecular Evolution and Gene Expression During Cold Acclimation. Proc. R. Soc. B. Vol. 275:979–985.

  7. Oh, another by-the-way, I’ve just seen another feather-fanboy on DA claim that the filaments on those tyrannosaurids could have rotted off before fossilization, citing the so-called “Montauk Monster” as his example (this was a dead raccoon whose fur had apparently rotted off). I don’t think this creature has yielded actual scales yet, but then feather-fanboys argue that those tyrannosaurid skin impressions show naked rather than real scaly skin.

    I apologize for bombarding your blog with these comments, but I’m in the middle of a fight with the feathered-dinosaur fandom on DA.

    • No worries about the bombardment. It’s nice to see the blog coming back to life again. I’ve been meaning to write more stuff for a while now, so this serves to inspire me to get the ball rolling once more.

      It’s unfortunate that Nanuqsaurus didn’t preserve more (it’s basically known only from a chin). Hopefully we will get more specimens in the future. As for the scales are really plucked skin argument, I’ve heard it before and I admit it could be possible. One of the things that came about from the recent integument papers from last year is that we need to be more rigorous with our analysis of skin impressions, and not just say: “scales found.” Sometimes it doesn’t even get that far. Having looked through many museum collections now I can tell you that there are way more dinosaur skin impressions that have been collected than have been written about.

      All that said, the tyrannosaurid skin impressions I’ve seen images of (e.g., Wyrex) sure look a lot like the scales one sees on geckos, and less like the helter-skelter cracks one sees in rhinos.

  8. The Wikipedia article on Gorgosaurus claims that it has yielded rounded/hexagonal scales from the tail and smooth skin without scales from elsewhere. However, I have also seen an excerpt from the Currie 2003 paper you cite in your blog post, and it generalizes tyrannosaurid skin impressions as pebbly like you describe.

    What irritates me about the current over-feathering trend more than anything else is how pushy its advocates are. You may recall that they were whining about the feather-less Gorgosaurus in the “Walking with Dinosaurs” movie last winter, portraying themselves as the champions of scientific accuracy despite what the science actually says. In my opinion their pretentiousness makes them even more insufferable than the stereotypical Jurassic Park fans who reject feathered dromaeosaurids.

    Ironically, feather-fetishism appears rooted in the same “reptiles are abhorrent” trope that makes scaly raptors seem scarier to certain laypeople. These guys want to use feathers and other filaments to divorce dinosaurs from the larger reptilian clade whom they stereotype as lethargic and cold-blooded. Even as our understanding of dinosaurs evolves, the underlying perception of reptiles as lower lifeforms has yet to evaporate.

  9. Avatar Mathieu
    Mathieu says:

    A thought just occured to me. I see reconstructions of permian synapsids with integument ranging from full scales to full fur and everything inbetween. Based on what we know, what do you think their integument would have been and at what evolutionary stages?

    • I haven’t spent as much time on synapsids as I have sauropsids. However what I can tell from my perusals of the literature is that there is far less contention about physiology or integument in pelycosaurs and therapsids as opposed to dinosaurs. This is rather disheartening as I think the evidence in favour of these interpretations for synapsids is just as ambiguous as the data for other extinct animals. I expect this general acceptance has a lot to do with an interest in these taxa as precursors to mammals, rather than an interest in the animals themselves.

      As to your question, pelycosaurs have generally been assumed to be scaly. This is largely a holdover from the days when they were still considered to be reptiles (albeit mammal-like reptiles). The only fossil evidence I have seen in favour of the scaly hypothesis comes from Reisz 1975, along with the recent paper by Niedźwiedzki and Bojanowski 2012. Reisz described the preservation of complete “belly scales” from the pelycosaur Archaeothyris florensis. The described scales were oblong and angled towards the mid-line. Each scale formed one arm of a ‘V’ that met in the mid-line. The images provided in the paper coupled with the illustrations makes these “scales” very different looking from the plate-like polygons that we normally see in reptiles. Regarding the scales Reisz wrote:

      Among the scales that have scattered some are turned over. These show that the sides of the scales have curled inwards, making the scales look narrower than
      they actually are. This may be due to an extended period of post-mortem exposure.

      The specimen is also extremely crushed, which further distorts the shape. I’m left leery of the interpretations of these tissues as scales, rather than skin folds or maybe some type of pseudoscale like those seen on rats (note: pseudoscale, in this case, means that it has a similar shape to epidermal scales but it lacks the necessary beta-keratin coating that we see on reptile scales). Even after taking into account postmortem crushing, the oblong shape of these belly scales suggests that if pelycosaurs were scaly looking, the texture would have quite different.

      Niedźwiedzki and Bojanowski’s paper looked at a potential body impression. They detailed the presence of individual rectangular scales from the belly and thigh of the animal. However their images do not support this interpretation (or much of anything, unfortunately). I suspect the impression they describe might be from water ripples and mud sliding rather than actual scales.

      Back to your point, pelycosaurs may have had a pseudoscaly appearance, though I’m suspicious of it. I wouldn’t be surprised if it turned out they had smooth, but keratinized skin, looking more like a dry amphibian than a lizard. As to when hair first appeared. It seems to be generally thought that hair started off as a mechanosensory structure (i.e., whiskers would be an example of the “first hair follicle”) that was later exapted for body-heat maintenance (Maderson 1972, Long 1972). Presence of foramina on the maxillae and premaxillae of therapsid jaws has been used as a proxy for these whiskers, and thus might count as the first occurrence of “hair” in synapsids (Brink 1956). This would suggest that fur first showed up in cynodonts. However it is important to keep in mind that this argument assumes that whiskers can be used to indicate the presence of fur throughout the entire body. There is not any real empirical backing for this claim, and even the proxies used for whiskers (foramina around the jaw) may not be that reliable (e.g., lizards also have similar foramina). Somewhat more recently Jaap Hillenius (2000) put forth a different proxy for hair presence in the ancestors of mammals. He looked at the evolutionary loss of the septomaxilla in synapsids and found that its loss was intimately associated with the movement of the nasolacrimal duct to the front of the nose (near the nostril). This seems to be associated with changes in the secretions of the Harderian gland (located around the orbit). Mammals are unique among vertebrates in that the Harderian gland secretes primarily lipids. These lipids are used to coat the fur on extant rats. This coating increases water resistance, which aids in thermoregulation and overall fur quality maintenance. So if this proxy holds true (Hillenius mentions that further work needs to be done to ensure that this behaviour is not unique to rats) then the first synapsid to have been covered in fur was probably a Mammaliaform such as Morganucodon. This is a much more recent origin of fur than is typically assumed. Given that a fur cleaning system probably evolved after fur, it’s probably safe to say that the origin would be slightly older than the taxa which showed this nasolacrimal duct rearrangement, but it probably wouldn’t be too much older. So then pelycosaurs, gorgonopsians, therocephalians, Lystrosaurus, and cynodonts up until ~Mammaliaformes would all be naked skinned.

      As for how fur evolved (gradual or all at once) I’m not sure. Maderson has done a lot of Evo-Devo work on this and has proposed a model that starts with sensory bristles starting off as little protrusions from the skin (in his model it is between the scales) that eventually get multiplied over the body (Maderson 1972, 2003). He started as far back as Pelycosauria. So one could imagine something along the lines of a smooth skinned Dimetrodon transitioning to a sparsely bristled Varanops to a more bristly Glanosuchus and Tritylodon before becoming full blown fur in critters like Morganucodon. Alternatively we could have had more smooth/leathery skinned guys up until some basal Mamaliaforme that, through a change in some regulatory genes, popped up a set of follicles all over its body, following the developmental “program” for epidermal development.

      I’m not sure how much sense all that made, but hopefully it gives some food for thought.

      ~ Jura


      Brink, A.S. 1956. Speculations on Some Advanced Mammalian Characteristics in the Higher Mammal-like Reptiles. Palaeontol. Afr. Vol. 4:77–96.

      Hillenius, W.J. 2000. Septomaxilla of Nonmammalian Synapsids: Soft-Tissue Correlates and a New Functional Interpretation. J. Morph. Vol. 245:29–50.

      Long, C.A. 1972. Two Hypotheses on the Origin of Lactation. Am. Nat. Vol. 106(947):141–144.

      Maderson, P.F.A. 1972. When? Why? How?: Some Speculations on the Evolution of the Vertebrate Integument. Am. Zool. Vol. 12(1):159–171.

      Maderson, P.F.A. 2003. Mammalian Skin Evolution: A Reevaluation. Exp. Derm. Vol. 12:233–236.

      Niedźwiedzki, G., Bojanowski, M. 2012. A Supposed Eupelycosaur Body Impression from the Early Permian of the Intra-Sudetic Basin, Poland. Ichnos. Vol. 19(3):150–155.

      Reisz, R. 1975. Pennsylvanian Pelycosaurs from Linton, Ohio and Nýřany, Czechoslovakia. J. Paleo. Vol. 49(3):522–527.

  10. Avatar Mathieu
    Mathieu says:

    Fascinating stuff! Thank you so much for that information. It sounds like it really could go either way so far for those animals. I look forward to any new information that could clinch our knowledge of what they were like. Also, in 2012, Dr. Mark Norrell said, and I quote: “We have as much evidence that T. rex was feathered, at least during some stage of its life, as we do that australopithecines like Lucy had hair.” Do you know of any evidence that clinches it this way?

    • I believe Norell’s argument was that, based on the phylogenetic bracketing of T.rex between dromaeosaurids and Sinosauropteryx prima, one could argue that it was fuzzy at some point in its life. This was also after the discovery of the putative tyrannosauroid Dilong paradoxus, which initially pushed fuzz in the direction of T.rex Similarly, we have no soft-tissue information for Australopithecus (as far as I know). It has just been assumed to have been hairy because of its bracketing deep within mammals. This is the same argument used to put hair on most Cenozoic mammals. That is all fine and good but Norell’s argument, much like many before, still associates scales with naked skin and does not really address the apparent incompatibility of those structures.

      • Avatar Mamenchi
        Mamenchi says:

        Hi, I have a question about dinosaur’s metabolism, I hope this forum is still active. One of the last articles says that thanks to oxygen isotope it has been possible to determine the temperature of the fossilized eggs of some small predators like oviraptor ( which as far as we know, might have been feathered). The temperature of the shells was about 32*C. Isn’t that a little low for a standard omeotherm (37*C), especially for an animal that many people consider as a warm-blooded because of its feathers?
        Another thing, I’ve read somewhere that some dinosaurs still show signs of growth rings in their bones(what species exactly?): isn’t that proof enough of ectothermy?
        One more, I know that tuatara is capable to survive at temperatures as low as 5*C, and has generally a low body temperature, but still moves actively in search of food. And yet it is a scaly creature, it doesn’t need feathers to keep warm. So, since they might have not been that necessary, why feathers evolved on some dinosaurs in the first place? There could have been many other ways to keep warm… Thanks, this forum is great, I hope you can solve my doubts!

        • Hey, sorry for the delayed reply. I’m in the midst of a hectic transition between jobs (and states) so the site was put on the back burner (ugh, again). All of that should be changing pretty soon.

          The work from Eagle and colleagues has been very interesting, largely because they found a neat way to take the temperature of extinct animals. I would be leery of taking the numbers they report at face value just yet, though. Fossils that are tens of millions of years old are bound to suffer some diagenetic problems. I appreciate that the authors did their due diligence in checking for diagenesis and taking multiple approaches to answering their question (indeed, they even had to throw out eggshell data from three different locales based on extensive diagenesis). I would like to see a larger data set of paleo-temperatures from a variety of dinosaurs, and sympatric species. That would give us a better idea of how consistent the data may be.

          The low body temperature on their oviraptorosaurs was an interesting finding. It is certainly lower than what we see in most birds, but it is not that far away from the temperatures that many mammals maintain (e.g., monotreme and marsupial core temperatures seem to range in the 30–33°C range [Feldhamer et al. 2014]). I think folks have been making the mistake of comparing these results to eutherian mammals, which are weirdos among mammals and make for misleading comparisons. However, all of that said, we need to keep in mind that what Eagle et al. did was find a way to take an extinct animal’s body temperature. They did not find a way to tell how the animal achieved that body temperature. Both tachymetabolic endotherms and bradymetabolic ectotherms operate within the same temperature range (it’s just that the tachymetabolic endotherms have a more restricted lower bound). The authors of the study stated that they did find that their oviraptorosaur data indicated body temperatures that were 6°C above the ground temperature, and they attributed this to internal heating, but it could just easily occurred via basking, and heat retention by the protofeathers.

          Regarding the growth rings, yes many dinosaurs show growth rings (theropods, ceratopsians, hadrosaurs, and even some sauropods), but they are not necessarily indicative of a bradymetabolic, ectothermic lifestyle. Köhler and colleagues (2012) showed that lines of arrested growth (LAGs) are commonplace among extant artiodactyls. Castanet et al. (2004) similarly found LAGs in primate bones, and penguin bones (Castanet 2006). So LAGs can tell us when bone growth stopped, but not the thermophysiology behind that bone growth (incidentally, the same is true for the infamous fibrolamellar bone).

          The selective pressure behind the evolution of feathers is currently a rife area of interest. It’s possible that feathers (filaments) evolved initially for display or camouflage. They may have evolved as an anti-predator strategy by starting off as toughened quills like what has been proposed for the filaments associated with Tianyulong confiusci and that singular Psittacosaurus specimen (Mayr et al. 2002). There are certainly multiple ways to keep warm, including just having a lower body temperature. There was an SVP talk years ago by Mark Goodwin, in which he did a detailed look at the long bone geochemistry of Tyrannosaurus rex and concluded that the specimen he looked at had two different optimal temperatures. A “summer” temperature and a “winter” temperature. So lowering one’s thermal set point is another viable alternative to staying active in lower temperatures.

          Hope some of this helped. Once again, sorry or the delay in responding.


          Castanet, J. 2006. Time Recording in Bone Microstructures of Endothermic Animals; Functional Relationships. C. R. Palevol Vol. 5:629–636.

          Castanet, J., Croci, S., Aujard, F., Perret, M., Cubo, J., de Margerie, E. 2004. Lines of Arrested Growth in Bone and Age Estimation in a Small Primate: Microcebus murinus. J. Zool. Lond. Vol. 263:31–39.

          Feldhamer, G.A., Drickhamer, L.C., Vessey, S.H., Merrit, J.F., Krajewski, C. 2014. Mammalolgy: Adaptation, Diversity, Ecology. Johns Hopkins U Press. p:177.

          Köhler, M., Marín-Moratalla, N., Jordana, X., Aanes, R. 2012. Seasonal Bone Growth and Physiology in Endotherms Shed Light on Dinosaur Physiology. Nature. Vol. 487:358–361.

          Mayr, G., Peters, D.S., Plodowski, G. 2002. Bristle-Like Integumentary Structures at the Tail of the Horned Dinosaur Psittacosaurus. Naturwissenschaften. Vol. 89:361–365.

  11. Avatar Brandon Pilcher
    Brandon Pilcher says:

    What are your thoughts on this infographic on feathered tyrannosaurids made by the people designing the “Saurian” computer game? The view I’ve taken on the whole issue is that feathers on derived tyrannosaurids has equivocal evidence supporting it—that is, it’s not impossible that feathers existed in heretofore patches on tyrannosaurids, surrounded perhaps by what you call “reticula”, but conclusions that feathered tyrannosaurids are NECESSARILY more accurate are premature. I’m concerned that enthusiasts for feathered tyrannosaurids are misrepresenting what is simply a possibility as confirmed fact on par with feathers on dromaeosaurids.

    The graphic in question:


    • I’ve seen this image making the rounds on the internet along with the general buzz that Saurian has been getting from the paleo community and I admit I don’t see the appeal. I get that seeing accurately reconstructed dinosaurs is something that every dinosaur paleontologist (myself included) wants to see, but there is a difference between “accurate based on our current knowledge” and “wishful thinking.” The latter is definitely the case for the integumentary covering in T. rex as shown in the infographic. That they use the Kulinda specimen as an example of feather-scale relationships despite the lack of data supporting any type of homology, is frustrating. Similar things can be said for Juravenator starki and Concavenator corcovatus, both of which have controversial integumentary inferences. The most frustrating thing about the integumentary choices given for the T. rex model is that they rely heavily on phylogenetic bracketing while completely ignoring the actual, hard data (i.e., skin impressions). Phylogenetic bracketing is useful for cases where we don’t have data, but when we do have data, bracketing needs to take a backseat. They also state that evo-devo studies supports their conclusion on how integument forms in crocodylians, but I have no idea where they are getting that from. Alibardi and Thompson (2000) showed pretty conclusively that integument forms in waves across the body of alligators, starting with the tail, and ending with the hands and feet. Similar patterns are seen in birds (Hamburger and Hamilton 1951; Alibardi 2005) and maybe lizards too (Alibardi 1996), so I have no idea how they are getting their “scaly tail, but feathered body” idea from.

      Similar artistic choices can be seen in their Triceratops model, where it has the long quills coming off the dorsum despite the fact that this is a controversial integumentary inference based on a specimen that has yet to be described. Thescelosaurus neglectus is even worse, brandishing feathers and quills despite being an ornithopod (animals that have only ever shown scaly integument). It doesn’t help that many of their scientific advisers are extremely liberal with their placement of integument in dinosaurs. This game feels more like an answer to dinosaur fans that insist on feathers on every dinosaur. I remember getting a similar feeling from the Indie film, Dino Hunt.

      As for the game itself, I still don’t know what one is supposed to do in it. From their description it sounds like they are hoping for something like Minecraft, or Black & White, in which there are no set goals, just a giant sandbox with which one can create one’s own story. I suspect that when this game eventually comes out it will be used mostly as a high-end digital terrarium. We’ll see, though. Everything is still pre-alpha so it still has a long way to go.


      Alibardi, L. 1996. Scale Morphogenesis during Embryonic Development in the Lizard Anolis lineatopus. J. Anat. 188:713–725.

      Alibardi, L. 2005. Keratinization and Lipogenesis in Epidermal Derivatives of the Zebrafinch, Taeniopygia guttata castanotis (Aves, Passeriformes, Ploecidae) during Embryonic Development. J. Morph. 251:294–308.

      Alibardi, L., Thompson, M.B. 2000. Scale Morphogenesis and Ulstrastructure of Dermis during Embryonic Development in the Alligator (Alligator mississippiensis, Crocodilia, Reptilia). Act. Zool. 81:325–338.

      Hamburger, V., Hamilton, H.L. 1951. A Series of Normal Stages in the Development of the Chick Embryo. J. Morph. 88:49–92.
      • I agree with you about Saurian. I became convinced a while ago that many of the people claiming to advocate “accurate dinosaurs” are pretentious and arrogant hypocrites, and that’s exactly the vibe I’ve gotten from the developers. The most infuriating part is that they’re passing themselves off as an authority that demands respect. They are full of themselves!

        • Avatar Arsene Kazaryan
          Arsene Kazaryan says:

          I got a lot of things I want to call them out for. Especially their arrogant attitude and lack of respect for fellow artists and other game developers out there.

          It seems to me that their only real backup for their choices is Trey the Explainer, whos videos I have had thrown at my face so many times, that I can pretty much predict every feather nazi doing so. And whenI show this article and give a brief explanation on how things work, he makes a silly comment about something, beign nutthurt about maybe his name being tagged in every comment, or saying that he don’t owe me no explanation for his views and etc.

          Saurian devs are well known feather nazis, there’s no doubt about that. I’m actually curious who is advising them on their “interesting” choices, considering how many flaws can be pointed out from them.

          I can name but a few that I find problematic in their choices:

          1. T.Rex feathers
          2. T.Rex lips
          3. Trikes changing shape of horns in a weird way
          4. The whole growth idea of pachycephalosaurus following Horner’s theory, which in their instance here I do not buy or support. Bones dont just disappear.
          5. Pachys have no cheeks, how are they grinding their food without it escaping?
          6. Feathered thescelosaurus, as known from the family that so far shows only scales, no hint of feathers
          7. Abuse of phylogenetic bracketing
          8. Comparing tyrannosaurid small scale patches to plucked chicken.. I mean, look at that image and then google a plucked bird, and then tell me to my face that they “look identical”. And afterwards, try googling chameleon or gecko skin and compare it to the prints, and THEN tell me what you really think. Know what im sayin?
          9. Dakotaraptor may not even be a real thing (see Jura’s reply to my questions down below for details)
          10. Triceratops has quills.

          That was me saying it from top of my head XD

  12. There is one thing I remember about Dilong and Yutyrannus that everyone seems to have overlooked: BOTH these taxa are from the Yixian Formation, which appears to have experienced colder winters than normal for the Mesozoic world. Noticeably this formation lacks “thermophiliac” (warm-climate) species like crocodilians that you find in places like Hell Creek. For me, this brings into question whether the filaments on either of them represent something all tyrannosauroids would have inherited, as opposed to unique adaptations for their cooler-than-average environment:


    • That is possible. Amiot et al. (2001) did find the the mean annual temp to hover around 10°C (though with a +/- 4° variance in their data) so it would have been fairly cool, at least by Mesozoic standards. If coelurosaurs all shared the potential to grow a filamentous covering, then there could have been some parallel evolution of fuzz during this time and place.

      All that said I would be wary of using any animal as a paleothermometer. Crocodylians in particular seem to be bad proxies for temperature. Nestler and Aiello-Lammens gave a talk about crocs as paleothermometers at the 2013 SVP in Las Angeles. They found that annual precipitation was the major driver of crocodylian expansion (e.g., modern-day alligators should be present as far north as Kansas or Nebraska, if temperature was the main driver). I think isotope data make better proxies for paleotemperatures than animals do, since animals can always evolve to live in a particular temperature regime. It is interesting that there are no crocodyliformes currently known from the Yixian. However, there are plenty of croc-like choristoderes like Monjurosuchus and Ikechosaurus, which seem to have filled similar niches. There may have been some competitive exclusion taking place, or the Jehol Biota may not have been productive enough to support a large crocodyliforme. Wang et al. (2010) suggested that the environment of the Jehol Biota may have been too harsh for terrestrial life (based on prominence of paedomorphic salamanders) whereas Jiang et al. (2014) suggested that the Biota is almost artifactual, having been created by frequent pyroclastic flows that engulfed terrestrial fauna from further away and deposited them into the lacustrine environment for preservation. Interestingly, this is a similar pattern to what we see in the fauna of the Prince Creek Formation in the North Slope of Alaska, except that is assumed to be based on seasonal flooding events rather than volcanoes. Either way it results in an extremely selective form of burial that may not be an accurate indicator of what was in the ecosystem at that time.


      Jiang, B., Harlow, G.E., Wohletz, K., Zhou, Z., Meng, J. 2014. New Evidence Suggests Pyroclastic Flows are Responsible for the Remarkable Preservation of the Jehol Biota. Nature Comm. 5:3151.

      Nestler, J., Aiello-Lammens, M. 2013. Modeling the Historical Range of Alligator and its Implications for Crocodylians as Paleoclimate Proxies. SVP abstract.

      Wang, Y., Dong, L., Evans, S.E. 2010. Jurassic-Cretaceous Herpetofaunas from the Jehol Associated Strata in NE China: Evolutionary and Ecological Implications. Bull. Chin. Ac. Sci. 24(2):76–79.
  13. Avatar Mamenchi
    Mamenchi says:

    I found a very interesting article about reptilian metabolism, that is based on the very very recent discovery of Tegu’s partial ability to control their own body temperature in the mating season:


    Do you think we could extend a similar method to dinosaurs as well?

    • Ah yes, the new Tattersall et al. article has been a very interesting discovery. I gave it a quick write up a few days ago and Tattersall himself gave a really nice rundown of how the discovery came about here. I think the biggest thing to come from this discovery is that we are still not sure how tachymetabolic, “automatic” endothermy arose in mammals and birds. That these two disparate groups share this unique metabolic suite of traits has lead many to look for a similar underlying pressure. One of those hypotheses is that automatic endothermy arose as a means of aiding in reproduction. This discovery of warming tegus adds support to that hypothesis.

      As for extending it to dinosaurs, we certainly could. We still don’t know if birds inherited their metabolic traits from theropods, or if automatic endothermy arose somewhere in the bird lineage. Either way I would say that the hypothesis that it arose because birds, or earlier dinosaurs were devoting a lot of time to head starting their kids, is possible. I certainly find it more plausible than the rather outdated hypothesis that automatic endothermy evolved to increase activity (which is still the most popular hypothesis despite weak support from the data).

  14. In this post, you state “just because a feather was developing does not indicate that scales and feathers made good bedfellows,” and also infer that the vast majority of dinosaurs were scaly. Many ornithischians have been discovered with some sort of “feathery” or at the very least filamentous integument – from Psittacosaurus to Tianyulong. With the discovery of Kulindadromaeus, an ornithischian which sported both feathery bits _and_ scaly bits, it seems to call into question these statements, as the Kulinda specimen shows no less than three types of scales, as well as feather-like structures. Will we be seeing an updated post with these new discoveries taken into account? I also want to point out the lack of discussion about other large, feathered theropods such as Therizinosaurs and ornithimimusaurs. I know the focus is on Tyrannosaurs in general, but you do comment on the majority of other dinosaurs being scaly – however, we’re seeing more and more that’s not necessarily the case. As messy as phylogenetic bracketing can be (and as much as you’ve inferred your distain for these methods in your post here), given all these new specimens and evidence, isn’t it fair to make the conjecture that it would be highly irregular for Tyrannosaurs to not have at least some sort of feather-type integument?

    • Regarding an updated post, I have been sort of doing that over the years as the occasional new discovery pops up.

      Getting more specific, in regards to ornithischians, I’ve expressed my skepticism over both SMF R 4970 (the Psittacosaurus) and Tianyulong before, but even assuming that the structures are genuinely part of the animal, that doesn’t make them homologous to feathers or protofeathers. As I discussed in this post, and my more recent take on the Kulinda specimen, there has been a tendency for the researchers describing these specimens, to just assume that the structures they are looking at are at all related to feathers. There has been very little testing (pretty much none) to validate that the structures have any association to actual feathers. That feathers and scales repeatedly show an adversarial relationship (borne out by a now extensive list of Evo-Devo studies) strongly suggests that none of these ornithischians were wearing structures related to feathers. We know from Evo-Devo data on modern turkeys that not every feather-like structure is actually a feather (Sawyer et al. 2003), so it’s not like archosaurs lacked the evolutionary toolkit to come up with something different, but similar. If these filaments don’t develop like feathers do, then they won’t be subjected to the same limitations that feathers do (i.e., replacing scales). That these ornithischian filaments differ so much from what we see in coelurosaurs adds further weight to the argument that they are convergently evolved structures. Further, phylogenetic testing for congruence (one of the three tests for homology) found that ornithischian filaments do not line up with coelurosaur filaments on the dinosaur family tree (Barrett et al. 2015). So the structures on the Kulinda specimen (and Tianyulong and SMF R 4970, if they are genuine) are likely separate evolutionary events from what we see in coelurosaurs.

      I also want to point out the lack of discussion about other large, feathered theropods such as Therizinosaurs and ornithimimusaurs. I know the focus is on Tyrannosaurs in general, but you do comment on the majority of other dinosaurs being scaly – however, we’re seeing more and more that’s not necessarily the case.

      I would argue that we are not. The first discovery of filaments in dinosaurs came in 1996. Now, 19 years later, we have over 40 “feathered dinosaur” specimens. On the outset, that does sound like dinosaurs are getting fluffier, but if we looked at where those specimens fall on the dinosaur family tree, 93% of those finds are within Theropoda, 88% are within Coelurosauria, and 79% are within Maniraptora. So at the end of the day, we can say with a fair amount of certainty that most, if not all, Maniraptors were fuzzy. Outside of Maniraptora we still see some strong evidence of filaments, including the recent find of protofeathers in ornithomimids (something that genuinely came as a surprise to me). Ornithomimids are just outside of Maniraptora, in the more inclusive Maniraptoriformes. All of these animals show evidence of genuine protofeathers (backed by ultrastructural analysis in the Ornithomimus specimen, though I think that is still just a SVP abstract). Once we get out of Maniraptoriformes things get dicey. Yes there is Dilong paradoxus and Yutyrannus huali, the “tyrannosauroids”, but there is also Juravenator starki (which I talked about in the original post), and every single tyrannosaurid that has skin impressions, all showing scales. These mixed data at the base of Coelurosauria suggests that filaments originated somewhere within this group. Once we go beyond Ceolurosauria, practically all the data comes up scaly. Our few exceptions are either questionable interpretations of osteological features (Concavenator corvatus) or questionable placement of a taxon within the phylogeny (i.e., the likely baby coelurosaur that is Sciurumimus albersdoerferi). The next time we see any kind of filamentous covering in a dinosaur is way over in derived ornithischians, which as I mentioned before, likely convergently evolved filamentous structures.

      As messy as phylogenetic bracketing can be (and as much as you’ve inferred your distain for these methods in your post here), given all these new specimens and evidence, isn’t it fair to make the conjecture that it would be highly irregular for Tyrannosaurs to not have at least some sort of feather-type integument?

      I can certainly understand how my writing can lead one to think that I despise phylogenetic bracketing. However, that is not the case. My work relies heavily on the limits imposed by phylogenetic bracketing, so I certainly understand its power. My problem is the rampant misuse of phylogenetic bracketing in both the paleophile field and in the professional field. The point of phylogenetic bracketing is that it limits our inferences of soft-tissue structures. Far too many people these days have been using phylogenetic bracketing as validation for putting structures on dinosaurs or inferring the presence of certain behaviours. The worst is when folks infer the presence of structure despite primary evidence to the contrary. This is what we have been seeing with the arguments for putting filaments on Tyrannosaurus rex. Witmer (1995) laid out the hierarchy of phylogenetic inference, and at the top of that hierarchy is the presence of a structure in the fossil itself. A classic example would be arguing that dinosaurs didn’t have horns because neither birds nor crocodylians have them. Extant phylogenetic bracketing supports this statement, but this inference is immediately squashed under the primary evidence of horns in ceratopsians. Returning to Tyrannosaurus, we have primary evidence of scales in Albertosaurus, Gorgosaurus, Tarbosaurus, and most importantly, Tyrannosaurus rex. The phylogenetic data suggests that they could be filamented, but it is trumped by the primary evidence, which shows that they were scaly.

      That said, it does seem weird that no tyrannosaurid shows evidence of filaments, but two tyrannosauroids do. I strongly suspect that this is due to the weak placement of taxa in Tyrannosauroidea (Bremer support is always 1, with bootstraps typically less than 50%), and that more rigorous phylogenetic analyses are going to pull most tyrannosauroids out and into other parts of the Coelurosaur tree (note that Dilong has come out as a basal maniraptor at least three times now, so there is already some precedent for this).


      Barrett, P.M., Evans, D.C., Campione, N.E. 2015. Evolution of Dinosaur Epidermal Structures. Biol Let. 11:20150229.

      Sawyer, R.H., Washington, L.D., Salvatore, B.A., Glenn, T.C., Knapp, L.W. 2003. Origin of Archosaurian Integumentary Appendages: The Bristles of the Wild Turkey Beard Express Feather-Type β-Keratins. J. Exp. Zool. (Mol. Dev. Evol.) Vol. 297B:27–34.

      Witmer, L. 1995. The Extant Phylogenetic Bracket and the Importance of Reconstructing Soft Tissues in Fossils. in Thomason, J. (ed). Functional Morphology in Vertebrate Paleontology. Cambridge Univ. Press. Cambridge, UK. pps:19-33.

      • Avatar Brandon S. Pilcher
        Brandon S. Pilcher says:

        I think an updated post, or maybe a whole new “master post” would be in order. Especially since the feather-nazi side put this out recently:


        They’re getting very aggressive and someone needs to call them out on their bullshit pronto.

        • Yeah, this isn’t the first time I’ve seen Trey’s stuff. I have considered making a response to that video series, but I think a more general post on what we know of integument development, might be a better choice.

          • Looking forward to that post. Another question I feel needs to be asked in these discussions is why tyrannosaurids would need a partial coat of filaments to begin with. If it was for insulation as commonly supposed, you’d think it would cover more of the body than just the upper part. It seems the only reason we’re told to presume tyrannosaurids had such half-assed feathering is, as Trey claims, the principle of phylogenetic bracketing. But doesn’t evolution also tend to phase out “vestigial” body parts that have lost their use over time? Surely heavy-yet-partial feathering on tyrannosaurids would present some utility to the animals if it had to be retained.

  15. Avatar Mamenchi
    Mamenchi says:

    Hello Jura, may I ask some more questions? Those are really important to me. I hope you can clear my doubts…

    Okay, according to this post, you said that reticula is practically an integument similar in appearance to scales, but in reality it is an integument made by feathers which have become scale-like structures throughout an evolutionary inverse process, right? And reticula is found at the base of birds’ feet, right? Okay, the first question is: is it possible for reticula to appear on other parts of the body of a bird or a reptile ( especially a dinosaur)?

    I was happy to read that now Dilong and Yutyrannus belong to the clade “Proceratosauria”. At lest for the time being… . So from now on we should not use these two dinosaurs as an excuse to put feathers on Tyrannosaurids ( and basically every other big theropod!!). Now from this originates my second question: LET’S SUPPOSE that Dilong and Yutyrannus still belong to the Tyrannosaurids. People assume that from “feathered” ancestors, tyrannosaurids retained feathers on their bodies ( or at least on some parts of their bodies ). However we have evidence of scales for the tyrannosaurids. Since it might be possible that feathers can “become scales again” by transforming into reticula, what if those scales impression are actually reticula??? I know we have pictures of the skin impressions on tyrannosaurids, but I am not good at distinguishing the difference between real scales and reticula. Could you look at the skin impressions and clear this doubt of mine?
    And if the skin impressions are actually reticula, what does this mean for tyrannisaurids? That they were covered in an integument that looks like scales but that in reality is something else?

    One last question: I’ve read that we have evidence of skin impression in Allosaurus. I’ve looked everywhere on the internet, but I couldn’t find any picture of those skin impressions. Do you know where I can find them?

    Thank you really much for your time and sorry if I asked a stupid question ( especially the one about reticula on tyrannosaurids).

    • Don’t worry about the questions. I’m happy to answer them as best I can.

      The story of sauropsid integument is a very interesting one, and I suspect one that I will need to devote a dedicated post to soon, as there seems to be lots of misinformation in the paleo community about it.

      Reticula are a very interesting integument. They are currently thought to be feather buds that were aborted very early in their development. They have very little to do with scales, other than looking superficially like them. I feel that this superficial look has gotten some folks to suggest that the scales we see in dinosaurs are just reticula. However, reticula don’t look much like scales at all. I regret that the image above is the one I used. There was a dearth of bird foot images at the time of that writing (at least ones that were not locked by copyright), so I used what I could. It’s not a very good image, because it does make it look like reticula are scale-like. In reality, reticula look more like a bunch of tiny, round spikes. If you ever brush up against a bird’s foot, you can feel how spiky they are. Dhouailly (2009) in regards to reticula wrote this piece (which I’m going to snip a lot, since it is all over the paper):

      The tuberculate scales, or reticula, which cover the plantar surface are made only of alpha-keratins…

      ….The overlapping of the scuta is sustained by a discrete dermal condensation…whereas the non-overlapping reticula do not display such a specialized structure of the dermis…

      The reticula, which cover the avian foot plantar surface, are very different from the scutate scales.

      Reticula-feather metaplasia never occurs in nature, which is easily understood, as it would prevent walking, climbing and perching.

      Chuong’s, as well as Niswander’s, group…did not obtain reticula-feather metaplasia, only scuta-feather metaplasia.

      ….The reticula-feather metaplasia has been obtained experimentally only once, by retinoic acid treatment during the appearance of reticula buds.

      ….Finally, reticula are not true cutaneous appendages, and appear to be feathers arrested in the initiation step of their morphogenesis: formation of a slight bump, without a placode.

      …the reticulate scales, which cover the plantar surface of all living birds, cannot be the remnants of the ancestral granulated beta-keratinized skin of first sauropsids, and correspond to a secondary, almost complete, inhibition of feather formation.

      To summarize, reticula are an innovations of the bottom of bird feet. Their formation strongly indicates that when feathers evolved, they covered everything on the body. That reticula form by so heavily downregulating feather development, suggest a strong evolutionary drive to keep this area feather free, and that inhibiting the feather-building program was evolutionarily “easier” than turning off feather-development entirely in this region.

      With all that said, I would say no, reticula probably never formed anywhere else but on the feet.

      This feeds into the tyrannosaurid question. If tyrannosaurids re-evolved scales from feathers, then they would be more in line with scutes and scutellae, not reticula as some folks online like to mention. This would make them true scales, although not homologous with reptile scales. Unfortunately all of our current data on scales in tyrannosaurids come from small skin impressions that no one has bothered to write much about. However, I am happy to say that there should be some news coming down the pike soon that might change some of that.

      Regarding the Allosaurus skin impressions, I have seen them, and I have an image of them that I can e-mail you. They currently remain unpublished. Pinegar et al. 2003, in an SVP abstract (which can be read here), mentioned them, but as is so often the case with non-feathery integument, it doesn’t seem to have warranted a full journal write up.

  16. Avatar Mamenchi
    Mamenchi says:

    Thank you really much Jura!! You’re right, more studies on dinosaur integument should be done to solve some doubts!
    A new abelisaurid about the same size as carnotaurus from the kem kem beds has been discovered. However it has yet to receive a name. We already possess integument on carnotaurus to assume that this family of theropods at least is covered in scales. Some authors ( Cau among them ) have suggested that those who have discovered the carnotaurus skin never made use of the word “scales” in their study. And because the skin impressions cover only the underside of the body of the animal, we could easily speculate that the upper part of carnotaurus might have had some sort of feathery cover.
    Maybe it’s me, but speculation recently seems to be the only thing that counts, instead of focusing on the direct evidences we possess. Of course they also admit that they could be wrong, it’s just that it’s another of those representations of dinosaurs that have been made because of this so called “feather revolution”. I’m not against feathers, I just wish we could fantasize less and study more those evidences we have some doubts about, instead of rushing into representations that might be just the result of our imagination.

    By the way Jura, if you still have that picture of the juvenile Allosaurus skin, you can freely e-mail it to me, here’s my address: kama_891@hotmail.it .

    • The trend I’m seeing is that this “speculation” more often than not takes the form of misapplying the principle of phylogenetic bracketing, as Jura mentioned. Quills on ceratopsians is an even better example than feathered tyrannosaurids because that fad is based on a singular specimen of the basal ceratopsian Psittacosaurus, yet it’s “inspired” the current generation of artists to draw every single species in Ceratopsia (even—no, especially Triceratops) as having quills. There seems to be no allowance for specializations unique to certain species or genera; if a given structure is found on a single specimen of one species, it must have been ubiquitous throughout the entire clade. And once that fallacious inference sets off an entire artistic trend and even affect people’s perception of accuracy, it becomes more irritating that a swarm of Cretaceous mosquitoes.

      Of course, the more that “speculation” makes dinosaurs resemble modern birds, the more popular it becomes in paleophile circles. I’m convinced more than anything it’s a hipster-like backlash to pop culture images of dinosaurs; their disproportionately negative attitude to Jurassic World says everything to me. It’s less rooted in science than a weird sort of artistic or academic snobbery/

  17. BTW, I thought I read somewhere that scutes in archosaurs might have developed from primitive feather-like filaments as shown by developmental research. If so, do you think dinosaurs like tyrannosaurids have evolved scutes in place of their ancestral fuzz? It makes more sense to me than a receding, vestigial coat of fuzz since it would actually have a selective advantage (body armor).

    • So that’s the argument that has been put forth about “scales evolving from feathers.” It is based on developmental data that shows that in order for scales to develop on a bird’s tarsometatarsus, the feather formation pathway need to be suppressed. From an evolutionary developmental perspective, this tells us that feather formation is running on the same basic scale development pathway that regular scales follow, and that the underlying pathway is still intact; it is just suppressed. It is a strong piece of evidence in favour of scales and feathers being homologous.

      It has also been largely misrepresented online and even in the paleo-literature. It does not mean that scales and feathers can exist together (in fact it suggests quite the opposite) and it also doesn’t mean that scales developed from feathers. It does tell us that the scales on bird feet are not truly homologous to reptilian scales, since their development is slightly different (bird scales develop from a condensation of cells called a placode. Reptilian scales do not have placodes).

      All that said, I have considered the thought that Tyrannosaurids may have re-evolved scales from a filamented ancestor. It remains possible. The one caveat I have with this scenario is that birds only exhibit scutes and scuttellae. The hexagonal, tuberculate scales seen in most dinosaur skin impressions, do not develop in birds. Danielle Dhouailly performed some really cool integument grafting experiments in the 70s. Her results indicated that birds only have the ability to develop scales on their tarsometatarsi. The scale building pathway has been completely overwritten everywhere else on the body. So it looks like that at least on the lineage that directly lead to birds, feathers were all over the body and were only later pushed back to the tarsometatarsus. Tyrannosaurs are close to birds, but not that close (especially compared to maniraptors), so it’s possible that the scale formation pathway was able to get switched back on over the whole body rather than just the tarsmetatarsi. It’s also possible (though less likely) that the filaments seen in tyrannosauroids, were unrelated to protofeathers. Unfortunately, without access to developmental data, the only way to really tell if scales re-evolved in Tyrannosaurids is to find the much coveted tyrannosaur with the feathered mohawk that seems to be all the rage in paleo-art these days.

      • Actually the scenario I meant to propose was SCUTES (or scuttellae) supplanting feathers in tyrannosaurids, not scales which you know are distinct. But then, the Wyrex skin impressions from the tail don’t look like proper scutes either, so they might not represent that process if scutes in theropods could only develop on the feet.

        One scenario I was going to suggest was that scales would have been basal to most dinosaur lineages, but they still had a genetic *potential* for filaments that could be turned on in certain environmental conditions. That could account for why most ornithischians have yielded scales but a select few have integument resembling filaments (even if they are not true feathers). I could see the same happening in tyrannosauroids; they could have been ancestrally scaly, but those in the Yixian formation like Dilong and the proceratosaurid Yutyrannus could have turned on filaments due to a shared environmental pressure (whatever that would be). I don’t think it is as simple as filaments in one lineage indicating an entire lineage had them, except maybe in a subgrouping like Maniraptora where they appear to have been ubiquitous.

  18. Sorry to irritate you yet again, but I’ve just seen tyrannosaurid skin impressions in a video advertising the “Dueling Dinosaurs” specimen. They’re described as finer-textured than those of its ceratopsian adversary, but they still look scaly or at least featherless to me. The one shown in the video comes from the arm (starting at the 3:53 mark).


    • Here’s a picture from the video:

      Kinda funny in light of the artistic trend to give T. rex et al little feathery wings.

      • No worries regarding comments. I’m always welcome to discussions, even if it takes me a while to write folks back. The Dueling Dinos fossils are very spectacular and have a lot of implications for tyrannosaurs. It’s just a shame that they are still in private hands. Hopefully they will eventually make there way to a museum.

        I remember hearing that the tyrannosaur fossil had apparently preserved some skin impressions, but I had not seen them until you posted the image. It’s very interesting. I wish it were at a higher resolution so we could really make out the detail. On the outset they appear to look very similar to previously published (and talked about) skin impressions for tyrannosaurs (i.e., small, tuberculate scales). There does appear to be a texture change as one moves further to the right of the image, but it’s hard to say for certain what exactly is going on there.

        I agree that it is funny (and frustrating) that we have yet more evidence for scales in tyrannosaurs, but that it will probably be ignored in favour of a more fuzzy appearance. I think that it is going to take something along the lines of a tyrannosaur mummy for some paleo folks to change there tune.

        All that said, it reminds me that I need to get back to cranking on my integument development post. It’s been a busy month, but I’ll try make some time for it.

        • I talked to the developers of Saurian about this. At first they dismissed it out of hand because it wasn’t “published” (which to me sounded like shooting the messenger). But then one of them stepped in to claim that, according to a secretive “inside source”, the impression actually came from the foot region and that their current T. rex design had already incorporated it. Very suspicious behavior on their part if you ask me, but I guess we’ll have to wait and see until these remains receive further study.

  19. Avatar Deltadromeus
    Deltadromeus says:

    Hi Jura,

    First I want to say that your posts on integument really made me think twice about how feathery Dinosaurs are in general – it seems very few share a similar opinion and go with what’s generally the popular trend in the paleocircles. Even the general “dinosaur fanbase” seems to be pushing for feathers on everything – no scales allowed (which in turn, seems to also be fueled in part by some revulsion to reptiles, or at least it seems that way to me). I am eagerly awaiting your post on dinosaur integument – you’ve piqued my curiosity on what we actually know, but sadly trying to dig up info on other sites come across as very biased towards feathers – even dinosaurs which have shown evidence of definitely being scaly aren’t safe from getting a feathered mohawk stuck on them these days!

    That brings me to a question I have – when you say “feathered all over”, how much of the face do you think would be covered in feathers? Would they go right up to the tip of the snout? Or would there be a little something left uncovered, like skin or scales? I wonder this in particular concerning the Dromaeosaur family, but also Yutyrannus – ‘ol Yu was show to be very much feathered, but I’m left wondering about its face, as well as the faces of Maniraptors in general (Oviraptorids seem to show beaks, so I can more or less guess what their snouts would look like). It seems popular to depict them with scaly faces, or even snouts that resemble toothy beaks, but I’m not entirely sure how correct that would actually be.

    The fossil of Microtaptor seemed to show a ton of fuzz even up to the snout, but there’s also claims drifting about the internet about the internet of a duck-like dromaeosaur which upon further google searches turns up absolutely nothing more – not even a name for the thing. I remember the fossil seemed to show some kind of bare face, more or less, but since I can’t dig up any other info on it on the internet, I’m wondering what happened to it. What was written about it stems from 2001, if I remember correctly.

    • Sorry for the super late reply. I was in the field for most of August. Regarding feathering, according to the Evo-Devo data it seems that when feathers formed, they took over every place that scales were present, all the way to the soles of the feet. Dhouailly (2009) notes that there is strong selective pressure to block the feathers on the bottom of the feet, which makes sense given how ungainly it would be to walk on fully formed feathers (though I doubt it ever got to that point, think about how hair growth gets stunted on the ankles of people who constantly wear socks).

      In contrast, our data on feather formation in birds and fossils of feathered dinosaurs indicates that the face was not under these same selective pressures. This is my long-winded way of saying that feathered dinosaurs likely had fully feathered (to the tip of the snout) faces. This would later get replaced as beaks formed (which use a similar developmental pathway to feathers and scales). Some species could certainly have had feather loss in their face, or even their face and neck (going down the turkey vulture path), but the exposed area was likely just skin rather than scales. For example, take a look at wood storks (Mycteria americana). They lost the feathers on their heads and necks and replaced them with skin that looks superficially like scales, but is formed entirely of alpha keratin and doesn’t show any of the serial arrangement of scalation (it looks more like plaque psoriasis).

      To date, the only evidence we currently have of scales coming in to replace feathers, is in the tarsometatarsal region of most birds and some dromaeosaurs. It seems to be an extremely rare event, with full-on accessory integument loss being more common.

      I’ve not heard about the duck-like dromaeosaur before. If true, it certainly sounds interesting.


      Dhouailly, D. 2009. A New Scenario for the Evolutionary Origin of Hair, Feather, and Avian Scales. J. Anat. Vol.214:587-606.

      • Avatar Deltadromeus
        Deltadromeus says:

        Thanks for answering. This clears up a lot of confusion I had in regards to how to depict dinosaurs. Would be great if the feathered from tip-to-toe depictions caught on over the beak-with-teeth restorations that I see so much, as I feel it would be incredibly interesting to play around with.

        I’ve tried to track down what I found about that duck-like dromaeosaur, but…nothing. If I remember what searches I used to find it, I’d link it. I just found it very weird that there was info around 2001, then…nothing. Not even a name, so I don’t even know if was a real thing or a hoax.

  20. Dear Jura,

    I have read your article with great pleasure and only regret that I havent had a chance to find it earlier. I am with you on pretty much every single account here, but I do have a few questions that I wanted to clear out. I hope you won’t mind answering them for me and I will try to make it a bit easier and make it more structured for you. Soem of these questions I am posting on behalf of my friends from Russia, who unfortunately dont speak English very well, so I was mostly translating it. Hence why the list may seem pretty big, but I hope it will be allright for you, so here we go:

    1. Do we have any evidence of feathers on Dakotaraptor? If so, what is it that we currently know that represents it? And what stage of feathers do you think they may have? Can bumpy skin prints (if any found) indicate the presence of feathers, or is it still debatable?

    2. What do you make of the comments, when people say that the skin impression from the specimen of Tyrannosaurus “Wyrex” supposively shows reticula of emu? Is this true and can it contradict the presence of the actual scales? Many of those who favour feathers in them seem to frequently use Juravenator as example, what is the best and the right way to resolve this?

    3. Is there really any find that can hint on feathers in tyrannosaurids or not?

    4. Many claim that feathers arent always possible to preserve, and thus try to use this as an argument to justify the possibility of feahters. I dont personally agree with it, since if scales are preserved, then so should be feathers. Could you please resolve this, if you can of course?

    I came across this post, and was wondering what you can make of that, really curious if this guy is onto something or not (will keep it anonymous, if you dont mind, since i dont know how he would feel about his name being mentioned):

    ” If people only understood just how rare it is to preserve feathers in fossil material , they would then understand why so many fossil finds dont show them . Its all about Study on a molecular level to infer the reason behind the feathering proposal. Just looking at what is in front of you isn’t necesarily the whole story. I do my homework and have been privy to information from the tops in the field and now understand why they had them . so many factors that just arent being understood by the general public is what is stopping some from getting the real picture. its ok, time will tell and it will come to light for the mainstream. then when everyone is part of this revalation i will sit back and smile with the few that already knew. the Evidence is there , not in the marks on the fossilized soil around the preserved bone but actually IN the bone through study of cellular structure and genetic fatures of avian code. You cant look at something face value and say see there is no evidence that shows it was there so it doesnt exist. if that were the case Camels for example would not have fleshy humps if we just looked at the skeleton. that is just a external feature that only in life or preservation of fossilized skin would help shed the light. funny thing is with feathering there is actually something there in the study of the cell structure that tells the story as opposed to relying on bold visable evidence with the naked eye. People also need to stop comparing Reptilain scales to Avian scales ,they are two completely different structures. we have Birds today that have avian scales with feathers from between the scales. The area between the scales has a underlayer of flesh which lay’s beneath the scale work, this is where the feathering support folicals form. there is also areas on Dinosaur skin samples that show Scales patterns stop and just fleshing material form just like in the legs of all Avians . some areas may not even be true avian scales but rather rough bumpy skin like in the neck of a turkey vulture.”

    5. Is it actually possible for tyrannosaurids to not be coelurosaurs? I heard somebody propose that idea a while back, but it was dismissed. No idea whetehr it was dismissed outright or after thorough testing and analysis, so wanted to know what you think on that.

    6. Do you perosnally think, from a logical point of view, that tyrannosaurids needed feathers? I personally see them more as a nuisance, since i cant imagine what kind of a hell they would go through when they have to molt and also the body temperature for the mass of something that weighs minimum of 2.5 tons (speaking of known large individuals, not the fragmented skeletons)? I find it pretty hard to imagine them being of any use and see it as nuisance more than anything. While all known feathered dinosaurs like ornithomimids and maniraptorae as examples, they at least are more flexible and are able to assist the molting process with their more useful arms, than those of tyrannosaurids. And for a parallel comparison, abelisaurids in essence are kinda sharing some basic ideas with tyrannosaurids in the way of having bigger heads, stronger legs and small useless arms. And are known to be scaly. Just speculating here. What do you make of this?

    7. Are you by any chance registered on Facebook and do you ahve a blog there as well? If not, I strongly recommend you do so, as this needs to be exposed and discussed, so that people see it and start actually researching these ideas more thoroughly, instead of trying to justify whats popular and the “fashion”, instead of talking science.

    8. You mentioned that you are a paeleontologist? May I ask for your actual name, if its ok and which country are you based in at the moment? If you dont want to answer in public, please feel free to email me ak150884@gmail.com (if you so wish). And also, could you please email me the pic of the allosaurus skin print? Would be very much appreciated, thank you tons!

    I hope you will get around to answering my questions and help me clear out some of these issues. I am personally a supporter of scaly tyrannosaurids and always have been, but untill i saw your article, i kinda allowed that 50/50 chance of them being feathered, but now im more inclined for scaly.

    • Hi Arsene,

      I told you I wouldn’t forget. 🙂

      1) There are no feathers known for Dakotaraptor. In fact, as it currently stands right now, Dakotaraptor steini may not be entirely real. The original description came from a bonebed that included multiple animals. At least one of those animals was a turtle that had part of its shell mistakenly associated with the holotype of D. steini (Arbour et al. 2016). Regardless, the placement of feathers on this species comes from inferences based on phylogenetic bracketing and potential proxies for feathers (quill knobs) on the ulnae. As a deinonychosaur, D. steini is well within the Penneraptora group, which would suggest true feathers (albeit symmetrical) were present in these species. I think that bumpy skin can mostly tell us that an animal didn’t have scales. The only skin evidence I can think of for feathers would be evidence of feather tracts. As far as I know neither of these have really been found preserved in fossils yet.

      2) I think it’s a reach to compare preserved scales in Tyrannosaurus, to the reticulae in emus. I don’t think they look alike at all and more resemble the turberculate scales on geckos (like this, for instance). As I mentioned in this post, reticulae differ substantially from true scales. Theoretically, some of these differences should preserve in fossil skin impressions, allowing an ultrastructural view of the imprints to determine if a structure is a reticulum or a scale. To date, no one has tried this yet.

      As for Juravenator being used as an example of scale-feather coexistence, I would need to see evidence that Juravenator actually preserves feathers (or filaments). Chiappe & Gohlich (2010)’s description leaves me wanting. Their images are terrible and they stayed cautious of their interpretations in their conclusions. That many in the online paleophile community still cite their paper as “proof” of this feather-scale coexistence is both frustrating and a little troubling.

      3) To date there is no find that shows or even hints at feathers or filaments in tyrannosaurids. Even the dueling dinosaurs fossils found a few years back only show scales on the preserved integument. Some folks have referred to tyrannosaurid skin impressions as being “naked” like a “plucked chicken” but closer examination always reveals small, tuberculate scales that are on par with the small scales we see in geckos. Ultimately, it would take a Yutyrannus like find to support a filamented tyrannosaurid position. We know that type of preservation is possible, thanks to the discovery of filamented Ornithomimus in Canada (Zelenitsky et al. 2012).

      4) Yeah, I’ve heard the argument about preservation bias too. While I agree that it takes a certain set of conditions to preserve feathers, I have yet to see any evidence that those conditions are any different from those that preserve scales. People like to cite Archaeopteryx as an example of feathers preserving where scales don’t, but all those Archaeopteryx specimens preserve are imprints of feathers, primaries and secondaries at that. There is nothing else there. No scale imprints, and no skin imprints. Sinosauropteryx preserves filaments around most of the body, but there is no preservation around the lower limbs. So areas where we might expect to see scales, were never preserved. In contrast, NGMC 01, a potential specimen of Sinornithosaurus, does preserve body filaments, as well as scales on the tarsometatarsal region. Similarly, Anchiornis fossils show feather preservation all over the body. There are no preserved scales, but largely because the areas we would expect to see scales, had feathers instead. In fact, close examinations of the feet of Archaeopteryx showed similar “feathery legs” results (Foth et al. 2014).

      So as far as I can tell, if conditions preserve scales they will also preserve feathers. Cases with one and not the other are always instances where only part of the skeleton shows preservation at all.

      As for the anonymous poster, I’m not sure what finds he is talking about, nor what he means about cellular material in the bone. Mary Schweitzer is the leader in the field of molecular paleontology and even she has stated how rare it is to get anything resembling original organic material from fossils. Still, the poster is right in that we should stop viewing the scales in birds as homologous to the scales in reptiles. We now have a wealth of evo-devo data that shows they are not truly homologous structures.

      5) It is certainly possible for tyrannosaurs to not be coelurosaurs. They were originally considered carnosaurs before more detailed systematics work revealed that they were a separate case of gigantism in theropods. It would take a compelling amount of evidence to pull them out of Coelurosauria, but it always remains possible. As it currently stands, tyrannosaurs are one of the most basally branching groups of Coelurosauria. It always seems that more and more finds keep pushing them closer to that base. So it could still happen, however unlikely that may be.

      6) I can’t see much use for filaments in tyrannosaurs. Of course that alone isn’t enough of a reason to dismiss the idea. Filaments in tyrannosaurs could have just been vestigial remnants from animals that did need them (much like the hair on elephants). The way I see it, if tyrannosaurids did evolve from filamented ancestors, they seem to have subsequently lost those filaments and re-evolved scales in their place. I have yet to see any specimens that support the whole filamented mohawk nonsense that seems to be all the rage these days.

      Although the Reptipage doesn’t exist on other media, I do provide easy sharing options for Facebook, Pinterest, Twitter, etc. I’m happy keeping the site here as a blog since it has a higher degree of permanence compared to the ephemeral nature of all the social media platforms. I always welcome folks to share my posts where relevant, though.


      Arbour, V.M., Zanno, L.E., Larson, D.W., Evans, D.C., Sues, H-D. 2016. The Furcuale of the Dromaeosaurid Dinosaur Dakotaraptor steini are Trionychid Turtle Entoplastra. PeerJ. 4:e1691

      DePalma, R.A., Burnham, D.A., Martin, L.D., Larson, P.L., Bakker, R.T. 2015. The First Giant Raptor (Theropoda: Dromaeosauridae) from the Hell Creek Formation. Paleo. Contrib. 14:1–16.

      Foth, C., Tischlinger, H., Rauhut, O.W.M. 2014. New Specimen of Archaeopteryx Provides Insights into the Evolution of Pennaceous Feathers. Nature. Vol. 511 Supplement:1–83.

      Zelenitsky, D., Therrien, F., Erickson, G.M., Kobayashi, Y., Eberth, D.A., Hadfield, F. 2012. Feathered Non-Avian Dinosaurs from North America Provide Insight into Wings Origins. Science. 338(6106):510–514.

  21. Sorry, forgot the most important part (got so carried away with thoughts on questions) – thank you for your article and for answering, hope to hear from you very soon and will await your response! 🙂

  22. Sorry once again for spamming this comment section with multiple posts, but I forgot to ask you if you have any related articles on the lips on theropods? Cause i see a lot of people going pretty far with that recently, and was wondering what your thoughts on this are? Cheers!

    • I missed Reisz’s talk on it this year at SVP, but I’m excited to read the paper. I think it has some merit to it, though as you said, we need to be careful not to go too far with things. It might be worth writing a post about it once the paper comes out.

  23. Avatar Brian Buck
    Brian Buck says:

    This section of the blog has a lot of great info.
    So, what do you make of this:

    • Thanks. The news of the protofeathered tail found in amber is pretty cool. I have a similar opinion to Steve Brusatte in that this could still be from an early bird. The only reason why I am more inclined to go with filamented dinosaur is because the authors’ analysis of the filaments indicates that they are most similar to type III protofeathers (although with a twist all their own), thus suggesting that this belonged to an animal that had yet to evolve true feathers. So that makes it anything from a Therizinosaur to a dromaeosaur. I think one of the most interesting things to take from this find is how well it agrees with developmental studies that also showed barbs and barb ridges to develop before follicles. That we found anything from a dinosaur in amber is amazing. Most animals the size of even hatchling dinosaurs, are usually thought to be strong enough to free themselves from runaway sap. This changes our search image and will hopefully result in more cool amber finding in the future.

  24. I suppose a small dino carcass (or portion thereof) could’ve been deposited in sap by whatever made a meal of it. Semi-arboreal maniraptors might be especially prone to such outcomes. Fantastic discovery at any rate.

  25. Avatar Brian Buck
    Brian Buck says:

    It is indeed a fantastic discovery. I hope it will be able to tell us just how well a preservative amber really is in regards to the soft tissues.

  26. Avatar Brian Buck
    Brian Buck says:

    I just saw yesterday on Facebook a link to a new Troodon fossil that supposedly shows both feather impressions and scale impressions near the tail. Have you seen this specimen?

    • Avatar Brandon S. Pilcher
      Brandon S. Pilcher says:

      I think those probably show naked skin, if not reticula. The guy who found the specimen says the skin impressions aren’t pebbly like that of most dinosaurs, but instead appear rough and irregular.

  27. Avatar Brandon S. Pilcher
    Brandon S. Pilcher says:

    New post on the subject of dinosaur integument, which cites this blog and you might find interesting:


  28. Avatar Thiago Chagas
    Thiago Chagas says:

    “Whereas scuta and scutella contain the standard scale histological makeup (largely β-keratin with an α-keratin hinge), the reticula have been found to be composed of only α-keratin” This part and where you talk about the relationship between body parts and integument caugh my attention. Well, have you ever considered the possibility of the scuta and scutella being formed by the same gene that makes the feathers (or something like this)? For example: if AA or Aa resulted on scaly feet while aa resulted on fluffy feet? A great example is the snowy owl http://918f9ce2f933e26d324a-a30cf25863fadb0ebfac784c1629c0d0.r84.cf2.rackcdn.com/40773d2871a804691d3853d981f50d6a.jpg

    And I see your point when you said “Given this knowledge one may able to reasonably infer extent of integument based on where a skin impression was found. If it was found on the foot, or tarsometatarsal region one should be able to infer that integument was present along the foot and ankle, leaving the rest of the body as an unknown. Should the skin impression be found on the tail, pelvis, or thoracic region though, one should be able to reasonably infer that it was representative of the integument found across the body up to the head, ankle, and wrist regions.” but the fact is that not every archosaur is exactly like T. rex, nor T. rex is exactly like every other archosaur in this aspect. Crocodiles are a great example http://i.dailymail.co.uk/i/pix/2016/06/24/12/359E9E2E00000578-3656376-Canadian_born_filmmaker_Franny_Plumrdige_has_been_pictured_divin-a-14_1466766300523.jpg http://i.purseblog.co/wp-content/uploads/Ph-siamese-crocodile-belly-3.jpg as you can see the scales that are found in the tigh are very different from those that are found on the belly, for example, all of this without mentioning this https://www.theguardian.com/science/2016/sep/14/scientists-reveal-most-accurate-depiction-of-a-dinosaur-ever-created the scales seen on psittacosaurus’ feet were equal to those seen on the rest of the leg, wich is completely different in other archosaurs.

    The fact is that evolution works trough mutations that we can not predict and may change a “rule” very rapidly. So just like crocodiles have different scales on their back and on their belly, T. rex could have had feathers on it’s back while showing scales on part of the tail

  29. Avatar Mamenchi
    Mamenchi says:

    Hi Jura, have you heard of the new “Daspletosaurus horneri” that presents a new facial integument never suspected for these types of theropods?