July 2011-July 2018: Marking 7 years of paleo-heresies.

On July 12, 2011
a new blogpost entitled, “Welcome to The Pterosaur Heresies” first appeared online. It was (and is) meant to be the newsletter for taxon additions to the large reptile tree (LRT, 1255 taxa) at ReptileEvolution.com. More complete explanations and documentation can be provided here than at ReptileEvolution.com.

Starting two days later (July 14, 2011) and for the next three days,
the several hypotheses of pterosaur origins were compared one with another.

About a week later (July 22, 2011)
a completely resolved family of pterosaurs was presented. This was the first one to include several specimens from all well-known genera and the first to include tiny Solnhofen pterosaurs, first listed by Peters 2007. Previously tiny pterosaurs had been ignored based on the false premise that they were juveniles of larger specimens. That is a disproved hypothesis that continues to make the rounds. And we said goodbye to the clade, “Pterodactyloidea” because now 4 clades are recovered that share all of the pterodactyloid-grade traits, while two others share some, but not all of those traits. Have other workers started to include tiny Solnhofen pterosaurs in their analyses? No.

On the last day of that first month (July 31, 2011)
a  phylogenetic analysis of just 235 taxa was presented that recovered a completely resolved and diphyletic Reptilia (= Amniota), with one branch, the new Lepidosauromorpha, containing turtles, pterosaurs and lepidosaurs and their many relatives. The other branch, the new Archosauromorpha, contained mammals, enaliosaurs, archosaurs and their many relatives. An amphibian-like reptile, Gephyrostegus was their last common ancestor.  Today, with more than 1000 additional taxa, the original topology from seven years ago remains unchanged. Have other workers started to include basal amphibian-like reptiles in their analyses? No.

In the seven years since July 2011
hundreds of exciting and heretical discoveries have been recovered. Some of these resolve long-standing problems by simply adding taxa. Others shed new light on topics that were not thought to be problems at all by simply adding taxa. Ironically, several other workers gained worldwide acclaim for ‘discovering’ relationships that were recovered in the LRT and promoted here years earlier. Still other workers continue to criticize the LRT, claiming it should have failed some time ago, but the LRT continues to grow.

a propagandistic pall was cast on the LRT, so most workers have ignored the taxon inclusion/exclusion suggestions offered here, leaving their work open to criticism from the ever-growing authority of the LRT.

Whatever the faults of the LRT,
the specimens included here need only be included in more focused analyses using independent character lists to test them. In other words, the faults don’t have to be employed, only the suggested taxa. When that happens, confirmation of the LRT has been the typical result. Why? Because the wide gamut and sheer number of taxa minimize the possibility of taxon exclusion, the number one problem in prior, less inclusive analyses. If you have a tetrapod of unknown affinity, test it here at the LRT.

One unexpected and disappointing discovery:
DNA analysis, the standard for crime-fighting and paternity questions, has not been able to replicate the results of wider trait studies. Rather, DNA studies lose their efficacy over large phylogenetic distances when compared to the trait-oriented LRT. Worse yet for paleontology, DNA cannot be used with most fossils. Unfortunately, many paleontologists still believe in the validity of DNA studies.

Figure 2. Dr. Sean Carroll and Dr. Antonis Rokas

Figure 1. Dr. Sean Carroll and Dr. Antonis Rokas

On that note…
Quoted from EvolutionNews.org, “Finally, a study published in Science in 2005 (Rokas and Carroll 2006) tried to use genes to reconstruct the relationships of the animal phyla, but concluded that “despite the amount of data and breadth of taxa analyzed, relationships among most [animal] phyla remained unresolved.” The following year, the same authors published a scientific paper titled, “Bushes in the Tree of Life,” which offered striking conclusions. The authors acknowledge that “a large fraction of single genes produce phylogenies of poor quality,” observing that one study “omitted 35% of single genes from their data matrix, because those genes produced phylogenies at odds with conventional wisdom.” The paper suggests that “certain critical parts of the [tree of life] may be difficult to resolve, regardless of the quantity of conventional data available.” The paper even contends that “the recurring discovery of persistently unresolved clades (bushes) should force a re-evaluation of several widely held assumptions of molecular systematics.”

I was not aware of that 2005 paper
before a few days ago. It needs to be more widely considered.

While other blogs journalistically report on the works of others,
the Pterosaur Heresies scientifically tests the work of others. That’s what sets it apart. That’s what makes it fun, interesting and rewarding. That’s what makes it controversial. Hopefully, that’s why you’re a subscriber. If, instead, you keep waiting for the LRT to crash and burn, well, that should have happened by now, don’t you think?

This July 2018,
seven years after it was started in 2011 with 235 taxa, there are 1000+ more taxa, all gradually blended in a tree topology that has been growing organically and with virtual complete resolution (some taxa known only from mandibles and other scraps are less resolved). Still, critics keep harping on the same perceived shortcomings (too many taxa, too few traits, not enough firsthand observation, lack of expertise)—while not harping on the shortcomings of traditional studies (principally taxon exclusion) that fail to produce gradually blended (= similar) sister taxa. There has always been a double standard at play, not only here, but for new hypotheses in geology, astronomy, physics, and paleontology. It’s universal and has been at work for centuries. It used to be that religious leaders led the charge against new ideas. Now we have PhDs trying to do the same.

Even scientists are not immune from this thing we call ‘human nature.’
Dr. J Ostrom complained about it, too. It’s human nature to follow authority, to go with the majority, and to suppress contra-indicators. Facts sometimes take decades to be widely accepted, and that’s just the way it is. It’s not acceptable, but that’s the way it is.

The beauty of science is
you, yes you can perform your own analysis to confirm or refute any analysis you read about here or anywhere. If I can do it… you can do it.

Thank you for your readership.
If there are subjects/taxa you want me to cover, or issues that need resolution, let me know. I look forward each day to corresponding with each and every one of you.

Peters D 2007. The origin and radiation of the Pterosauria. Flugsaurier. The Wellnhofer Pterosaur Meeting, Munich 27
Rokas A and Carroll SB 2006. Bushes in the Tree of Life. PLoS Biology, 4(11): 1899-1904.




Think of aardvarks and sloths as naked and hairy glyptodonts respectively

that’s what they really are… aardvarks are naked and sloths are hairy glyptodonts. And, yes, that comes as a surprise, it breaks a paradigm, it spins your head around, it’s heretical… and it’s exactly where the data takes us.

The Edentata is an odd clade
in which the basalmost taxa, like Barylambda, Glyptodon and Holmesina are very large. On the other hand, terminal extant and derived taxa, like Peltephilus and Cyclopesare much smaller, just the opposite of most mammal clades (in which smaller usually lead to larger, following Cope’s Rule.)

Figure 1. Subset of the LRT focusing on edentates and their outgroup, Barylambda. Here two glypotodonts nest at the bases of the two major clades.

Figure 1. Subset of the LRT focusing on edentates and their outgroup, Barylambda. Here two glypotodonts nest at the bases of the two major clades.

According to Wikipedia,
“Glyptodontinae (glyptodonts or glyptodontines) are an extinct subfamily of large, heavily armored armadillos which developed in South America and spread to North America.”

In the large reptile tree (LRT, 1252, edentate subset Fig. 1) the glyptodont, Glyptodon, nests between the massive Barylambda and giant sloths, followed by smaller tree sloths and small extinct horned armadillos, like Peltephilus and Fruitafossor. On another branch (Fig. 1) another large glyptodont, Holmesina, nests between the massive Barylambda and the much smaller aardvark, Orycteropus, the armadillo, Dasypus, and the anteaters, Tamandua and Cyclopes.

Such a big-to-small phylogenetic pattern,
is known as phylogenetic miniaturization or the Lilliput Effect and is often the product of neotony (adults retaining juvenile traits, including juvenile size).

Figure 2. Holmesina, the glyptodont ancestor to aardvarks, anteaters and armadillos.

Figure 2. Holmesina, the glyptodont ancestor to aardvarks, anteaters and armadillos. Those are aardvark hands (Fig. 3), glyptodont feet.

Holmesina (Fig. 2) is added to the LRT today.
Basically it is a longer-snouted glyptodont, basal to the longer snouted above-mentioned aardvarks, armadillos and anteaters.

Following a reader comment,
(suggesting ‘taxon exclusion’ was the issue that did not unite glyptodonts with armadillos) I was looking for a transitional taxon to more closely nest glyptodonts with armadillos, rather than sloths. I did so and the tree topology did not change when Holmesina was added. Armadillos are still one taxon removed from glyptodonts, but at least now we have a glyptodont on the long-nosed clade of aardvarks, etc.. As before, aardvarks nest between glyptodonts and armadillos. Looking at all the edentate taxa in detail and overall. I think this nesting and this tree topology seem very reasonable (= it produces a gradual accumulation of derived traits at all nodes and between all taxa).

Figure 3. Orycterpus, the extant aardvark, is a living sister to Barylambda from the Paleocene.

Figure 3. Orycterpus, the extant aardvark, is a living sister to Barylambda from the Paleocene. Aardvarks traditionally nest alone, but in the LRT they are edentates without armor… or hair.

Other workers, like Fernicola, Vizcaíno and Fariña 2008,
described the phylogeny of glyptodonts by putting taxa like Holmesina at the base while omitting Barylambda. Thus such studies do not present the full picture due to taxon exclusion. Everyone seems to omit Barylambda and all the other edentate outgroups back to Devonian tetrapods… but not the LRT.

Goodbye ‘Xenarthra’. Goodbye ‘Pilosa’. Goodbye ‘Cingulata’.
According to Wikipedia, “The order Pilosa is a group of placental mammals, extant today only in the Americas. It includes the anteaters and sloths, including the extinct ground sloths, which became extinct about 10,000 years ago.” According to Wikipedia, Cingulata, part of the superorder Xenarthra, is an order of armored New World placental mammals.” In the LRT ‘Xenarthra’ (Cope 1889) is a junior synonym for ‘Pilosa’ (Flower 1883) and that is a junior synonym for Edentata (Darwin 1859).

Darwin C 1859. On the origin of species.
Fernicola JC, Vizcaíno SF and Fariña RA 2008.
The evolution of armored xenarthrans and a phylogeny of the glyptodonts. Chapter 7 in: The Biology of the Xenarthra, Eds: Vizcaíno SF and Loughry WJ. University Press of Florida.
Gaudin TJ and Croft DA 2015. Paleogene Xenarthra and the evolution of South American mammals. Journal of Mammalogy 96 (4): 622–634. https://doi.org/10.1093/jmammal/gyv073






What would pterosaurs be, if tritosaurs were not known?

This is lesson 4 in taxon exclusion…
to see where select clades would nest in the absence of their proximal taxa.

Now that
the large reptile tree has grown more than three fold in the last seven years, it’s time to ask (or ask again) some phylogenetic questions.

Figure 1. Bergamodactylus compared to Cosesaurus. Hypothetical hatchling also shown.

Figure 1. Bergamodactylus compared to Cosesaurus. Hypothetical hatchling also shown.

pterosaurs are nested with archosauriformes, like Scleromochlus, close to dinosaurs, but only in the absence of fenestrasaurs and tritosaurs. In the large reptile tree (LRT, 1242 taxa), which includes representatives from all tetrapod clades, pterosaurs nest with fenestrasaurs (Peters 2000) and tritosaur, lepidosaurs (not prolacertiformes (contra Peters 2000, who did not test Huehuecuetzpalli, which came out in 1998).

In the absence of tritosaurs (and Archosauromorpha)
pterosaurs nest with drepanosaurs, both derived from Jesairosaurus.

In the absence of tritosaurs (and Lepidosauromorpha)
pterosaurs nest between Mei and Yi among the scansoriopterygid birds (Fig. 2) which are derived from Late Jurassic Solnhofen bird taxa, too late for the Late Triassic appearance of pterosaurs like Bergamodactylus (Fig. 1).

Figure 1. Two Mei long specimens, one in vivo, one in situ.  Click to enlarge.

Figure 2. Two Mei long specimens, one in vivo, one in situ.  Click to enlarge.

Taxon exclusion
has been the number one problem in traditional paleontology. That’s why the LRT includes such a wide gamut of taxa. The result is a minimizing of taxon exclusion and the problems that attend it.

Peters D 2000b. A Redescription of Four Prolacertiform Genera and Implications for Pterosaur Phylogenesis. Rivista Italiana di Paleontologia e Stratigrafia 106 (3): 293–336.

Dr. Baron tip-toes around the radiation of dinosaurs

Last year, Dr. Matthew Baron,
not even a year out from his PhD thesis, placed himself in the middle of controversy when Baron, Norman and Barrett 2017 resurrected the clade Ornithoscelida, wrongly uniting plant-eating Ornithischia with meat-eating Theropoda to the exclusion of plant-eating Sauropodomorpha, an invalid (due to taxon exclusion) hypothesis of relationships, we discussed earlier here.

Dr. Baron guessed,Maybe Ornithischia is actually so far removed from the base of the dinosaur tree that no studies, including my own, have been able to properly place them… Its an intriguing thought and one that needs examining properly.” By his own words, Dr. Baron is not yet an authority on the subject. That authority can only come from a wide gamut analysis that minimizes taxon exclusion, like the large reptile tree (LRT, 1236 taxa), which is something that anyone can produce. As noted last year (see citations below), Dr. Baron’s team excluded several relevant taxa.

Figure 2. Look familiar? Here are the pelves of Jeholosaurus and Chilesaurus compared. As discussed earlier, this is how the ornithischian pelvis evolved from that of Eoraptor and basal saurorpodomorpha.

Figure 1. Look familiar? Here are the pelves of Jeholosaurus and Chilesaurus compared. As discussed earlier, this is how the ornithischian pelvis evolved from that of Eoraptor and basal phytodinosauria.

Later Langer et al. 2017 argued against the Baron, Norman and Barrett interpretation. Baron, Norman and Barrett agued back, stating in Baron’s summary, “Langer et al.’s response showed that the alternative arrangement, that preserved the traditional model, was not statistically significantly different to our own hypothesis, and that was with much of our data having been altered, in ways that we perhaps disagree with strongly.”

Baron is correct is noting that Seeley’s original division, uniting sauropodomorphs with theropods based on pelvis orientation “just because a subgroup have gone on to lose a feature that was the ancestral condition for the wider group, it does not mean that we can then say that the other subgroups who have ‘hung on’ to the feature should be grouped together to the exclusion of the experimental group, at least based on that feature’s absence/presence, without other evidence.” Plus it would be one more example of pulling a Larry Martin.

Dr. Baron pulls out a bad example as his example of the above. He states, “In fact, Cetacea is more closely related to Carnivora than either group are to the Primates.” In counterpoint, in the large reptile tree (LRT, 1236 taxa) there is no clade “Cetacea.” Odontoceti arise from tenrecs and elephant shrews. Mysticeti arise from hippos and desmostylians. Carnivora split apart in the first dichotomy in Eutheria. Thus all other eutherians, including primates, odontocetes and mysticetes have a last common ancestor that is not a member of the Carnivora.

Dr. Baron bases the above quote on a phylogenetic error when he states, “Like I said before, you need to look at TOTAL EVIDENCE to come to this quite obvious conclusion, which means focusing on more anatomical evidence.” While this may sound reasonable and correct, a focus on anatomical evidence may lead to confusion due to convergence. Bottom line, it is more important to look at the phylogenetic placement of a taxon in order to determine what it is. This has to be done in the context of a wide gamut analysis that minimizes taxon exclusion using at least 150 (sometimes multi-state) characters (the LRT uses 238). Otherwise you’re cherry-picking taxa, something Baron, Norman and Barrett were guilty of by excluding bipedal crocs and several basal dinosaurs from their study (and we know this since the LRT includes them). Baron also cherry-picks traits in part 3 of his argument, pulling a Larry Martin several times in doing so. In a good phylogenetic analysis, like the LRT, you’ll see a gradual accumulation of traits. That means you’ll get a pubis with a transitional phase, a tiny predentary and other traits in gradual accumulation among the outgroups to Ornithischians.

Figure 1. Chilesaurus and kin, including Damonosaurus and basal phytodinosauria.

Figure 2. Chilesaurus and kin, including Damonosaurus and basal phytodinosauria.

Baron promises
“I will eat my shoes!” if Seeley’s dichotomy is correct. That’s an easy promise to make knowing there is a third hypothesis out there: the Theropod/Phytodinosaur dichotomy presented by Bakker (1986) and confirmed by the LRT in 2011.

Pertinent to this discussion
sometimes what a paleontologist does not say about a particular subject can be more important that what a paleontologist does say. I lump taxon exclusion and citation exclusion in the category of ‘what is not said.’

Bakker RT 1986. The Dinosaur Heresies.New Theories Unlocking the Mystery of the Dinosuars and Their Extinction. Illustrated. 481 pages. William Morrow & Company.
Baron MG and Barrett PM 2017. A dinosaur missing-link? Chilesaurus and the early evolution of ornithischian dinosaurs. Biology Letters 13, 20170220.
Baron MG, Norman DB and Barrett PM 2017.
A new hypothesis of dinosaur relationships and early dinosaur evolution. Nature 543: 501–506;  doi:10.1038/nature21700
Baron MG, Norman DB and Barrett PM 2017. Baron et al. reply. Nature 551: doi:10.1038/nature24012
Langer et al. (8 co-authors) 2017. Untangling the dinosaur family tree. Nature 551: doi:10.1038/nature24011
Novas FE et al. 2015. An enigmatic plant-eating theropod from the Late Jurassic period of Chile. Nature 522(7556), 331.

Relevant blogposts and theses from Dr. Baron:


What I think about Ornithischia

Thoughts on Ornithoscelida … over one year on … (part 1)

Thoughts on Ornithoscelida … over one year on … (part 2)

Chilesaurus – what is it?






The origin and evolution of bats part 4: distance vs. accuracy

we looked at bat origins here, here and here from several perspectives. Some of these are now invalid given the following scenario.

Today we’ll take a fresh look at
the behavior and traits of the closest bat relatives in the large reptile tree (LRT, 1233 taxa, subset Fig. 1) and see what they can tell us about bat origins. This is called ‘phylogenetic bracketing‘. In such a thought experiment we can put forth an educated guess regarding an unknown behavior or trait for a unknown taxa (e.g. pre-bats) if all related specimens share similar behaviors and traits inherited from a known or unknown last common ancestor.

We start off with a cladogram
focusing on bat relationships (Fig. 1) and take things one logical step at a time.

Figure 1. Subset of the LRT focusing on basal placentals, including bats.

Figure 1. Subset of the LRT focusing on basal placentals, including bats.

One. Living sister taxa.
The closest tested sister taxa to bats here (Fig. 1) are pangolins and colugos (flying lemurs) in order of increasing distance. The origin of bats and pangolins has remained a traditional enigma. Like the origin of pterosaurs and Longisquama, the surprise is, they are most closely related to each other, despite their current differences.

Two. Ancestral taxa
Th bat/colugo/pangolin clade had its genesis near the original dichotomy of placental mammals, when Carnivora split off from all others. At the next dichotomy the bat/colugo/pangolin clade split off from all others. So this clade is not far from an ancestral clades with living genera. Monodelphis, the short-tailed opossum today restricted to South America, nests just outside of all mammals with a placenta. Nandinia, the African palm civet, is a basal member of the Carnivora, somewhat larger than its Mesozoic forebearers.

Three. Timing for clade origins
The bat/colugo/pangolin clade had its origin in the Early Jurassic based on the more primitive egg-layers, Megazostrodon, Brasilitherium and Kuehneotherium in the Late Triassic and the much more derived arboreal multituberculate/rodent, Megaconus, in the Middle Jurassic. As you can see, Jurassic mammals remain extremely rare, currently represented only by the likes of Megaconus. Others will, no doubt, be discovered in time.

Four. Arboreality (tree niche)
Some bats, colugos and pangolins live in trees, and so do their last common ancestors, short-tailed opossums and African palm civets.

Five. Climbing trees
Bats no longer have to climb trees because they can fly. Colugos and pangolins both climb trees in a series of symmetrical short hops/extended reaches (colugo video, pangolin video), distinct from palm civets and short-tailed opossums, which put forth one hand at a time, like primates do.

Six. Descending trees.
Bats fly between trees. Colugos glide between trees. Pangolins use their prehensile tail to ease themselves down. The African palm civet drops out of trees in play. It also descends tree trunks like a squirrel, head first.

Seven. Nocturnal
Most bats, colugos, pangolins, palm civets and short-tailed opossums prefer to be active at night.

Eight. Omnivorous diet
Some bats eat insects, others prefer nectar or hanging fruit. Colugos prefer leaves, shoots, flowers, sap, and fruit. Pangolins eat ants. Palm civets and short-tailed opossums are omnivorous. African palm civets feed by holding their prey in their hand-like front paws, biting it repeatedly and then once dead, swallowing it whole.

Nine. Extradermal membranes
Colugos and bats both have extradermal membranes to their unguals that extend their glides in the former and enable flapping flying in the latter. Such membranes are lost in living pangolins, but the Early Cretaceous pangolin, Zhangheotherium appears to have scale-lined membranes between the elbows and knees. These were overlooked in the original description. The gliding membrane in colugos is fur-covered and camouflaged dorsally, naked underneath. In bats the flying membrane is naked, translucent and never camouflaged.

Ten. Mobile clavicle, interclavicle and scapula
The basal pangolin, Zhangheotherium, has a mobile clavicle-interclavicle and the large scapula rises above the  dorsal vertebrae, as in bats, but not colugos.

11. Sprawling femora
Zhangeotherium and bats share sprawling hind limbs, distinct from the more erect hind limbs of most limbed mammals.

12. Silent vs. noisy
African palm civets are noisy. Colugos and pangolins are largely silent. Bats are constantly chirping to one another and (micro-bats only) as part of their sonar attack system.

13. Enemies
All current enemies of bats (e.g. birds, snakes) evolved during or after the Late Cretaceous. Jurassic trees might have been a refuge for small early climbing mammals, like colugo, bat and pangolin ancestors. However…the minimally feathered, small theropod dinosaur, Sinosauropteryx, contained the jaws of Zhangheotherium, perhaps caught after descending from the trees or plucked out of lower branches. Certain pterosaurs (e.g. giant anurognathids) might have preyed on arboreal  mammals in the Jurassic, but no evidence of this is yet known.

FIgure x. Calcaneal spur in Zhangheotherium. Not venomous, but perhaps to anchor a uropatagium.

FIgure 2. Calcaneal spur in Zhangheotherium. Not venomous, but perhaps to anchor a uropatagium as in bats.

14. Calcaneal spurs
Hurum et al. 2006 originally considered the small spurs found on the calcaneum of Zhangheotherium (Fig. 2) similar to venom spurs found on the platypus, Ornithorhynchus. Phylogenetic bracketing indicates the closer homolog is with the basal bat, Onychonycteris, which has longer calcaneal spurs framing a trailing uropatagium.

Figure x. Monodelphis babies in an open pouch. This is how placentals began, slowly evolving from the less open pouch.

Figure 3 Monodelphis babies in an open pouch. This is how placentals began, slowly evolving from the less open pouch.

15. Newborns and mothers
All basal placental mammals give birth to helpless newborns that ride with the mother until mature enough to go out on its own. Monodelphis demonstrates a primitive version of this, protecting its ten young with lateral flaps of skin (Fig. 3). Carnivore mothers make nests for newborns (2-4 for African palm civets), but colugo, bat and pangolin mothers take their one or two babies everywhere they go, like marsupial mothers do. Zhangheotherium might have been fossilized with several newborns. (Fig. 4) and extradermal membranes between elbows and knees, as in bats and colugos. As we know from colugos, these extradermal membranes in basal pangolins (and Chriacus?) likely formed a playpen or nursery for developing young riding beneath their mother during the earliest stages of development.

Figure x. Zhangheotherium showing possible extradermal membranes (green) with keratinous scales (red) and several newborns scattered in the abdominal area, similar to Monodelphis in figure x.

Figure 4. Zhangheotherium showing possible extradermal membranes (light blue and green) with keratinous scales (red) and several newborns scattered in the abdominal area, similar to Monodelphis in figure x. These amorphous blobs with tiny tail bones need further inspection. Some may just be stains and shapes.

16. Curling (flexing the spine)
Mother opossums, palm civets, colugos, bats and pangolins are able to curl their spines so much that the mother’s mouth is able to assist wiggling newborns climb to the abdominal nipples. This curling ability is co-opted by pangolins as they defend themselves by rolling into a tight ball and by bats that catch prey in their tail before curling up to bite the victim as it is brought close to the jaws. Higher mammals lose the ability to curl ventrally in this manner. Humans and other primates have a limited ability to do this. Instead they use their hands. More derived mammals with stiffer backs have more developed newborns.

17. Upside-down vs. right-side up nursery for the young
Colugos may rest right-side up (preferring to hang from below a slightly leaning tree trunk) or upside down hanging by all fours beneath a horizontal branch. When doing so the mother’s extradermal membranes form walls making a protective nursery for the young ones.

By contrast, bats rest up-side down, sometimes hanging by only one locked foot. To fly bats simply release this foot lock, then plummet and start flapping. Bat membranes also provide a protective nursery for their young as they cling to their mothers’ chest and her wings fold over them.

Nowadays pangolins roll into a ball while nursing their young. Later in life, babies ride on the mother’s back and tail when able to do so. Zhangheotherium (Fig. 4) appears to have provided a colugo-like, but scale-lined membrane nursery for several growing babies. The late-surviving pre-bat, Chriacus (Fig. 5), likely did the same, based on phylogenetic bracketing.

18. Claws
Short-tailed opossums and African palm civets use their claws to climb trees and grab prey and fruit, bringing it to the mouth. So do basal primates. Colugos, bats and pangolins use their larger, curved claws principally to hang from trees, though living pangolins have co-opted their large claws to dig out ant and termite nests from trees and underground.

19. Distance vs. accuracy
Colugos leap and turn away from their tree trunk base in order to launch themselves into a glide. Can they do this while hanging beneath a branch? I don’t know. With their long limbs, colugos can just leap (without gliding) across gaps of 5m or more. With limbs extended, they can glide for 136m at 10m/second. Gliding is good for a quick escape from predators, and access to patches of food that are otherwise inaccessible. It does not save them energy to glide, let along climb back to a gliding height.

Bats drop from trees, then fly wherever they please, typically landing upside down on another high branch or cavern roof. The origin of bat flight enabled by flapping hyper-elongated webbed fingers is the key question here, and it is answered by combining all of the above numbered traits.

Before bats could fly Jurassic pre-bats had to climb trees, probably like colugos and pangolins do (see #5 above), before standing bipedally, but upside-down, on a horizontal branch. Why would they do that? To prepare to dive bomb insects on and in the leaf litter below. Here is where sonar became valuable, detecting insects in the leaf litter at night. Here is where the leaf litter became valuable, cushioning the early awkward landings of small dive-bombing pre-bats. Here is where flapping, even with small hands around colugo-like dermal membranes became valuable, at first in panic, then in gradually learning how to better direct the fall to cover the prey below.  (By analogy birds flap their wings vigorously while dropping to slow their descent.)

Upon landing the extended pre-bat nursery membranes ‘put a lid’ on the prey. Then, curling the tooth-line jaws toward the tail and the tail toward the jaws (see #16 above) spelled doom for the captured food item. Over time, larger fingers made better flapping parachutes. Ultimately flapping bats  learned how to hover before diving bombing their prey, like owls do. Later, after further development, bats gained the power and morphology to enable flight, slowly at first, then better and better to escape ground-dwelling predators and avoid having to climb a tree for the next attack. Only later did bats learn to use their sonar and flying skills on flying insects.

So what began as a small pouch, then a larger nursery membrane for bat and colugo infants became a killing zone for bat prey on the ground, another example of co-opting an old trait for a new behavior in derived taxa. Distinct from birds and pterosaurs, which used their nascent flapping behavior to ascend tree trunks to escape predators, create threat displays and slow their descents from branches, bats used their nascent flapping ability only to slow and direct their descent from branches. Distinct from colugos, which glided for distance, bats dropped for accuracy. Distance came later, after flight developed.

Remember the fall need not be far at first. Conifers can have very low branches and leaf litter can be a soft cushion for a mouse-sized mammal. Graduating slowly to higher branches provides bats a wider ‘field-of-view’ for their slowly developing sonar, and more time to develop flapping. Bat hind limbs are not long or heavily muscled. They are not good at leaping, like colugos.

Fruit eating bats could not have developed until flowering and fruit-bearing trees developed, later in the Cretaceous. The LRT and the fossil record indicates that fruit-eating bats are derived relative to smaller insect-eating bats. So sonar-emitting apparently was lost in fruit-eating bats, rather than never a part of their lineage. The great variation now seen in sonar-emitting bat morphology was likely developed during and after the Cretaceous, based on the current fossil record. I think we’ll find fully volant fossil bats in the Cretaceous someday.

I happened upon this idea while watching a pigeon descend from a roofline to a balcony beneath it and wondered if accuracy was more important for bats, while distance was more important for colugos. That distinction seems to be the key driver in both clades. In any case, it is important that any proposed scenario be viable at every point during the gradual evolution of new traits and behaviors. In this case, developing flapping forelimbs had to originate with a bipedal configuration, even it inverted. Developing sonar had to originate from simply listening to nocturnal insects and other small prey rustling in the leaf litter, not far below, gradually getting better in those families that randomly had slightly better skills once dive-bombing and trapping became the method for predation.

20. Bat ontogeny
Recapitulates this phylogenetic scenario. The fingers elongate last. 

21. Solitary vs. communal
Colugos and pangolins are solitary. So are African palm civets except when food is plentiful. Bats are communal, whether nesting in trees or caves. According to Kerth 2008, “Variable dispersal patterns, complex olfactory and acoustic communication, flexible context-related interactions, striking cooperative behaviors, and cryptic colony structures in the form of fission-fusion systems have been documented. tropical bats often form groups year-round, whereas sociality in temperate-zone species is sometimes restricted to certain times of the year. In most species, females form so-called maternity colonies to rear their young communally, whereas males are solitary, form groups of their own, or join female groups. In only a few species are both sexes solitary, meeting only to mate.”

Kerth concludes, “None of the three factors that I identify as important for the evolution of sociality in bats (ecological constraints, physiological demands, and demographic traits) can fully explain the frequency and diversity of group living in bats.”

Figure 1. Basal placentals at two scales, all arising from a Middle Jurassic sister to Monodelphis, based on the Earliest Cretaceous appearance of Zhangheotherium, in the lineage of pangolins.

Figure 5. Basal placentals at two scales, all arising from a Middle Jurassic sister to Monodelphis, based on the Earliest Cretaceous appearance of Zhangheotherium, in the lineage of pangolins..

22. Soles of the feet oriented opposite to those of most mammals
Distinct from most mammals, the knees of bats are splayed laterally, which should extend the toes laterally. However, the ankle is rotated another 90º producing a foot in which the soles are ventral during flight and while hanging. In the case of long-legged fish-eating bats, the feet help bring captured fish back to the mouth.

FIgure 1. Wondering if Chriacus had an inverted stance and dermopteran membranes? Comparisons to Onychonycteris and Pteropus.

FIgure 6. Wondering if Chriacus had an inverted stance and dermopteran membranes? Comparisons to Onychonycteris and Pteropus are shown. Yes, the knees are straight in derived fruit bats, bent in Onychonycteris and micro bats. The uropatagia are spread while inverted and while flying. Chriacus appears to be a much larger and much later-surviving version of much smaller Jurassic pre-bats. The membranes are conjectural and may have been lost in this large specimen, but it illustrates the possibility of a dive bombing taxon that covered prey like a casserole lid.

Why do bats hang upside down?
Without a phylogenetic or deep-time perspective, the following video is the best answer current bat workers can provide:

Bats are not using their wings to cool off.
A recent heat wave killed many fruit bats. They fell dead out of the trees (see below). None were creating a cooling breeze with their wings or extending their wings in a cooling fashion, like elephants sometimes do. Microbats that live in caves never have this problem.

Bat wings notes:

  1. Finger flexibility during flight varies greatly in bats.
  2. The flight stroke is otherwise bird-like with elbows raised above the back, nearly meeting at the midline, for maximum power at low airspeed, or less so for cruising at higher airspeeds.
  3. The large fingers do nothing else but push air for thrust and lift. They are not extended to cool the bat, nor do they extend or flash during courtship.
  4. Bat fingers hyper flex at the wrist to tuck away the flight membrane and reduce its surface area when not in use, as in pterosaurs and birds. When flexed they do little but envelope the bat and its clinging young.

Miscellaneous notes:

  1. Zhangheotherium was originally considered a symmetrodont mammal, but its teeth seems to converge with archaeocete whales in this regard. The reappearance of a more primitive symmetrodont molar shape is here considered an atavism in the evolution of toothlessness in both certain odontocetes and pangolins by convergence.
  2. The uncoiled cochlea of highly derived Zhangheotherium and multituberculates, has been traditionally considered a trait that nests these taxa in more basal branches of the mammal family tree. Here, in the LRT, these traits appear to be neotonous or atavistic developments that, taken alone, tend to confuse systematics. No traits should ever be taken alone to determine systematics. That would be ‘pulling a Larry Martin.’
  3. The initial splitting up of Pangaea in the Early Jurassic gave the previously dry climate a more lush, subtropical parade of cycads, conifers, ginkgoes and tree ferns. So there were plenty of standing and fallen trees for early mammals to gambol upon, learning how to climb and leap. The forest floor was likely cushioned with a carpet of leaves and fronds to absorb accidental falls and hunger-driven dive bombs mediated by fluttering pre-wings and large membranes co-opted for eventual flight.

Video showing a bat descending on a mouse in leaf litter appears here.

Byrnes, Libby, Lim & Spence. 2011. Gliding saves time but not energy in Malayan colugos. Journal of Experimental Biology http://dx.doi.org/10.1242/jeb.052993
Hurum JH, Luo Z-X and Kielan-Jaworowska Z 2006. Were mammals originally venomous? Acta Palaeontologica Polonica 51(1): 1–11.
Kerth G 2008. Causes and Consequences of Sociality in Bats. BioScience, Volume 58, Issue 8, 1 September 2008, Pages 737–746, https://doi.org/10.1641/B580810
Online here.

Mystery solved: Thylacoleo is a giant sugar glider…

no doubt, a little too big to glide…
and Thylacoleo (Fig. 2) is looking even less carnivorous in phylogenetic bracketing.

Sugar gliders
(Fig. 1) are phalangers (Fig. 6), a marsupial clade nesting between kangaroos and wombats (Fig. 5).

Figure 1. Petaurus breviceps skeleton in two views, plus a skull with mandible, lacking in the skeleton.

Figure 1. Sugar glider, Petaurus breviceps, skeleton in two views, plus a skull with mandible, lacking in the skeleton.

Adding the marsupial sugar glider,
Petaurus (Figs. 1, 3), and the cuscus, Phalanger (Fig. 6), to the large reptile tree (LRT, 1231 taxa) resolves a decades-old phylogenetic problem because Petaurus, the sugar glider, nests as a sister to Thylacoleo, the marsupial lion (Figs. 2, 4). Phalanger, the cuscus, nests as their last common ancestor, which has been suggested earlier.

According to the AustraliaMuseum website
“Most palaeontologists think that the ancestors of thylacoleonids were herbivores, an unusual occurrence since most carnivores evolved from other carnivorous lineages. One proposal suggests that thylacoleonids evolved from a possum ancestor (Phalangeroidea) based on dental formula, the skull of the cuscus Phalanger, and on a phalangerid-like musculature. Alternatively, evidence from certain skull features may show that thylacoleonids branched off the vombatiform line, the lineage that includes wombats and koalas.”

In the LRT,
wombats and koalas are now sister taxa to the cuscus clade. Without the sugar glider and the cuscus, the marsupial lion earlier nested with the wombat, Vombatus.

Just to be clear,
Phalanger is not an ancestor to Didelphis, the Virginia opossum, in the LRT, even though the Australian Museum called it a ‘possum ancestor.’

Figure 2. Thylacoleo skeleton compared to Petaurus skeleton to scale.

Figure 2. Thylacoleo skeleton compared to Petaurus skeleton to scale.

Long thought to be a super predator, 
in the midst of a clade of gentle wombat-like herbivores, Thylacoleo had, for its size, the strongest bite of any mammal, living or extinct, despite having tiny upper canines. This linking with sugar gliders further erodes the carnivorous hypothesis. 

Figure 3. Skulls of the genus Petaurus with many more teeth than in Thylacoleo, but in the same general pattern. Note the lower third premolar and its similarity to the same tooth in Thylacoleo.

Figure 3. Skulls of the genus Petaurus with many more teeth than in Thylacoleo, but in the same general pattern. Note the lower third premolar and its similarity to the same tooth in Thylacoleo. The big organe tooth at the tip of the dentary is the canine. The lower incisors are absent.

Arboreal or not?
Wikipedia reports, “The claws [of Thylacoleo] were well-suited to securing prey and for climbing trees.” And now we know how that came to be. Petaurus, despite its arboreal abilities, does not have a divergent thumb, like the one found in Thylacoleo.

Dentary canines
traditionally considered large, rodent-like incisors due to their placement, the anterior-most (medial-most) dentary teeth are actually canines. The incisors and their alveoli have disappeared. This can only be traced via phylogeny (see Arctocyon and Didelphis). The ancestrally small lower incisors are gone, replaced with ancestrally large large lower canines that meet medially like typical incisors. Notably, the lower canines maintain their traditional placement relationship to the upper canines (Fig. 6).

Even more interesting,
some marsupial taxa that experience a phylogenetic miniaturization, like Eurygenium (basal to Toxodon) the incisors reappear and the canines are not much larger than the incisors. That’s called a reversal or an atavism.

Figure 4. Thylacoleo skull. Many times larger than Petaurus, with fewer larger teeth, this is a giant sugar glider.

Figure 4. Thylacoleo skull. Many times larger than Petaurus, with fewer larger teeth, this is a giant sugar glider. The large orange tooth is the lower canine. The upper canine is a vestige. 

Thylacoleo was 71 cm tall at the shoulder, about 114-150cm long from head to tail tip, about the size of a jaguar.

Petaurus is 40cm long to the tail tip, about the size of a ‘flying’ squirrel. Loose folds of skin spanning the fore and hind limbs to the wrists and ankles are used to extend glides from tree to tree, or up to 140m. The diet includes sweet fruits and vegetables.

The sugar glider in vivo.

Figure 5. The sugar glider, Petaurus, in vivo. Note the wrinkled fur between the fore and hind limb. That’s the gliding membrane.

Petaurus species
According to Wikipedia, “There are six species, sugar glidersquirrel glidermahogany glidernorthern glideryellow-bellied glider and Biak glider, and are native to Australia or New Guinea.” Whichever one is closest to Thylacoleo has not been tested or determined.

Figure 2. Thylacoleo skeleton compared to Petaurus skeleton to scale.

Figure 5. Subset of the LRT focusing on Marsupialia, Metatheria and then nesting of Thylacoleo.

Petaurus breviceps (Waterhouse 1839; Early Miocene to present; up to 30cm) is the extant sugar glider, a nocturnal squirrel-like marsupial able to climb trees and glide with furry membranes between the fore and hind limbs. An opposable toe is present on each hind foot. Sharp claws tip every digit.

Phalanger orientalis (Pallas 1766; 34 cm in length) is a nocturnal arboreal folivore marsupial known as thte Northern common cuscus. Commonly considered a ‘possum’ the cuscus nests between wombats and kangaroos, basal to sugar gliders and marsupial lions.

Figure 6. The cuscus (genus: Phalanger orientalis) nests with Petaurus and Thylacoleo in the LRT.

Figure 6. The cuscus (genus: Phalanger orientalis) nests with Petaurus and Thylacoleo in the LRT. Those anterior dentary teeth look like incisors, but phylogenetically are actually canines.

Thylacoleo carnifex (Owen 1859; Pliocene-Pleistocene; 1.14 m long) was a giant sugar glider like Petaurus. Thylacoleo had the strongest bite of any mammal with the largest, sharpest molars of any mammal. It had fewer but larger teeth than Petaurus. The manus included retractable claws. The pes had a very large heel bone (calcaneum). This supposedly carnivorous ‘marsupial lion’ nests with herbivores. Pedal digit 1 likely had a phalanx and claw, but it has not been shown.

Goldingay RL 1989. The behavioral ecology of the gliding marsupial, Petaurus australis. Research Online. University of Wollongong Thesis Collection. PDF
Owen R 1859. On the fossil mammals of Australia. Part II. Description of a mutilated skull of the large marsupial carnivore (Thylacoleo carnifex Owen), from a calcareous conglomerate stratum, eighty miles S. W. of Melbourne, Victoria. Philosophical Transactions of the Royal Society 149, 309-322. 
Waterhouse GR 1838. Observations on certain modifications observed in the dentition of the Flying Opossums (the genus Petaurus of authors). Proceedings of the Zoological Society of London. 4: 149–153.


Time to trash the widely embraced ‘Azhdarchoidea’

According to Wikipedia
Unwin (2003) defined the group Azhdarchoidea as the most recent common ancestor of Quetzalcoatlus and Tapejara, and all descendants.

Unwin’s phylogenetic analysis excluded several dozen relevant taxa. When those are added back in, as shown in the large pterosaur tree (LPT, 232 taxa), azhdarchids arise from certain phylogenetically miniaturized Dorygnathus clade specimens (TM 10341) while tapejarids arise from certain phylogenetically miniaturized germanodactylids (Nemicolopterus) and before that, phylogenetically miniaturized scaphognathids (Ornithocephalus brevirostris, BSPG 1971 I 17). Tapejarids and azhdarchids don’t have a common ancestor in the LPT until you go back to Sordes PIN 2585-25. And that was not the intent of Unwin 2003.

Competing hypotheses of relationships
by Naish and Martill 2006; Lu et al. 2008; Pinheiro et al. 2011; and Andres, Clark and Xu 2014 all suffer from the same cherry-picking and taxon exclusion issues. At least Vidovic and Martill 2014 added germanodactylids and dsungaripterids nesting basal to tapejarids, but failed to add dorygnathids and pre-azhdarchids among other relevant taxa basal to azhdarchids. By the way, you heard it here first: germanodactylids and dsungaripterids were ancestral to tapejarids. Thanks are due to Vidovic and Martill for confirming this.

Pterosaur workers are ‘missing the boat’
when they are content to cherry pick traditional genus-based taxa. If they were to employ more specimen-based taxa in an unbiased fashion, as the LPT does, they, too, would recover a wealth of interrelationships otherwise invisible to them.

The clade name Azhdarchia
is hereby defined as TM 10341, Beipiaopterus, their last common ancestor and all descendants. These include Huanhepterus, Ardeadactylus, CM 11426, BSPG 1911 I 31, the flightless azhdarchid JME-Sos 2428, JME Sos 2179, and the traditional azhdarchids, Jidapterus, Chaoyangopterus, Zhejiangopterus, Azhdarcho, Quetzalcoatlus and kin. With origins in the late middle Jurassic, and no matter what size, these are all stork-like waders, gradually getting phylogenetically larger, toothless and ultimately flightless as their distal wing phalanges become vestiges.

The Azhdarchidae.

Figure 1. The Azhdarchidae. Click to enlarge. (That’s a juvenile Zhejiangopterus shown). These are all waders.

We already have a suitable name for the clade Tapejaridae,
which developed elaborate head crests and never stopped flying (so far as is known), based on their elongate distal wing phalanges.

The new skull compared to other tapejarids. Click to enlarge.

Figure 2. Click to enlarge. The rising size of the tapejaridae. These are not waders.

The expansion of the antorbital fenestra above the level of the orbit
in tapejarids and azhdarchids is (I hate to be the only one saying the obvious) a convergent trait. LPT relationships were introduced and published in Peters 2007, and (with a few modifications to incomplete taxa) hold true today. Peters 2007 reported, Major clades typically have a spectral series of tiny pterosaurs at their base suggesting that paedomorphosis was a major factor in pterosaur evolution.” Since then, no other workers have included the vital and relevant tiny pterosaurs in their phylogenetic analyses.

It is also time to trash the clade ‘Pterodactyloidea’
When tiny pterosaurs are added to a phylogenetic analysis (click here) the traditional clade ‘Pterodactyloidea’ divides into two clades that arise from tiny Dorygnathus derived taxa (Fig. 1) and two more that arise from tiny Scaphognathus derived taxa (Fig. 3), for a total of four pterodactyloid-grade taxa. There’s one more semi-pterodactyloid clade,  Darwinopterus and kin, with a large skull and long neck, but also a long tail. And yet another, the anurognathids that do not have a large skull (exception: Dimorphodon) and long neck. However anurognathids do shrink the tail, a pterodactyloid-grade trait that Longrich, Martill and Andres 2018 used to nest anurognathids as the proximal outgroup to their clade ‘Pterodactyloidea’ with the mistakenly reconstructed Kryptodrakon (= Sericiterus) at the base. The LPT lumps and splits all pterosaurs in a logical and tenable fashion.

Figure 1. Scaphognathians to scale. Click to enlarge.

Figure 3. Scaphognathians to scale. Click to enlarge.

From one generation to another
If you were a full professor, would you venture to include taxa suggested by an amateur? So far, none have shown the courage to do so (see below), while outside of pterosaur studies, confirmation of discoveries first announced here has happened several times (e.g. Chilesaurus), without citation. So methods used here work.

Dr. S. Christopher Bennett once told me:
“If you submit that manuscript, it will not get published. And if you somehow get it published it will not get cited.” Uncanny how that prophecy came true… but it doesn’t reflect on the value of the manuscript.

And that’s why
this blog and the website ReptileEvolution.com were launched, outraged at the insanity and insular thinking out there.

PS. As I write this,
Bestwick, Unwin, Butler, Henderson and Purnell (2018) compiled statistics on pterosaur dietary preferences (over 300 pterosaur dietary statements identified from 126 published studies) employing a traditional cladogram with the tiny hand, four-toed crocodylomorph, Scleromochlus as the outgroup, anurognathids basal to eudimorphodontids, wukongopterids basal to ‘pterodactyloids’, cycnorhamphids nesting with ctenochasmatids, pteranodontids nesting with ornithocheirids, and tapejarids nesting with azhdarchids, with loss of resolution at half the nodes. It’s quite disheartening to see this, when we know better… through specimen-based taxon inclusion.

Pity the first author, poor PhD student (U of Leicester) Jordan Bestwick. He is under the tutelage of Dr. David Unwin. You might remember Leicester, was earlier seeking a pterosaur tracker, a student who could somehow find evidence for the invalidated pterosaur forelimb launch hypothesis. Evidently, this is how they operate: Don’t find out for yourself… rather your job is to continue the legacy and dictates of your professor(s).

In addition to the invalid Azhdarchoidea,
Dr. Unwin has promoted:

  1. the invalid ‘uropatagium‘ incorporating pedal digit 5 in basal pterosaurs
  2. the invalid deep chord wing membrane of pterosaurs
  3. the invalid quadrupedal basal pterosaur hypothesis
  4. the invalid pterosaur egg burial hypothesis
  5. the invalid quad-launch hypothesis of pterosaur takeoff
  6. the invalid archosauromorph (Scleromochlus) origin of pterosaurs (see above)
  7. the invalid modular evolution hypothesis to support
  8. the invalid nesting of the Darwinopterus clade basal to
  9. the invalid Pterodactyloidea.

Anyone can test these hypotheses by
adding taxa to current published studies using whatever characters one chooses. Really. That’s all it takes to upset these cherry-picked (taxon exclusion riddled) studies.

Andres B, Clark J and Xu X 2014. The Earliest Pterodactyloid and the Origin of the Group. Current Biology24: 1011–6.
Bestwick J, Unwin DM, Butler RJ, Henderson DM and Purnell MA 2018. Pterosaur dietary hypotheses: a review of idea and approaches. Biological Reviews online pdf
Longrich NR, Martill DM and Andres B 2018. Late Maastrichtian pterosaurs from North Africa and mass extinction of Pterosauria at the Cretaceous-Paleogene boundary. PLoS Biology, 16(3): e2001663.
Lü J, Unwin DM, Xu L and Zhang X 2008. A new azhdarchoid pterosaur from the Lower Cretaceous of China and its implications for pterosaur phylogeny and evolution. Naturwissenschaften. 95 (9): 891–897.
Pinheiro FL et al. (4 co-authors) 2011. New information on Tupandactylus imperator, with comments on the relationships of Tapejaridae (Pterosauria). Acta Palaeontologica Polonica. 56 (3): 567–580.
Peters D 2007. The origin and radiation of the Pterosauria. Flugsaurier. The Wellnhofer Pterosaur Meeting, Munich 27.
Unwin DM 2003. On the phylogeny and evolutionary history of pterosaurs. Pp. 139-190. in Buffetaut, E. & Mazin, J.-M., (eds.) (2003). Evolution and Palaeobiology of Pterosaurs. Geological Society of London, Special Publications 217, London, 1-347.