Monday, January 15, 2018

Hesperornithines: Super-Loons No More?

Among the most remarkable of Mesozoic birds—nay, among the most remarkable of all Mesozoic theropods—were the hesperornithines. These diving birds from the Cretaceous are known to have lived in both freshwater and marine habitats, and the largest could attain body masses comparable to that of a wolf. They swam with strokes from their feet; in fact, the most specialized (and best known) forms had highly reduced forelimbs and were certainly flightless. Even with the great diversity of waterbirds today, and even with new evidence of semi-aquatic specializations in a few non-avialan dinosaurs like Spinosaurus and Halszkaraptor, hesperornithines remain prime candidates for being the most aquatically-adapted dinosaurs known.

Furthermore, unlike many other maniraptoran oddballs (such as the aforementioned Halszkaraptor), hesperornithines are no newcomers to the scientific spotlight, having been known to paleontologists since the 1870s. Together with Archaeopteryx and Ichthyornis, Hesperornis is a member of the "classic trio" that always represents Mesozoic avialans in popular books about paleontology.

Mounted skeleton of Hesperornis, photographed by "Quadell", under CC BY-SA 3.0.

Given the long period of time that we've been aware of their existence, it's not surprising that a good deal has been written about hesperornithines. Naturally, much of this literature has been spent drawing parallels between hesperornithines and extant birds, and two modern groups in particular are widely considered the best functional analogues for hesperornithines: loons (divers) and grebes. Indeed, when paleoartist Matt Martyniuk combed through the literature for hints on how to most plausibly restore Hesperornis, he concluded that hesperornithines could be thought of as "super-loons with grebe feet".

Like hesperornithines, loons and grebes are foot-propelled diving birds. However, they are not the only modern birds that use this method of underwater locomotion; cormorants and some ducks do so, too. Despite the long history of loons and grebes being considered ideal analogues for hesperornithines, this assumption has largely been based on perceived similarity instead of quantitative analysis, with little consideration given to comparison with other diving birds.

In a newly-published study, Alyssa Bell and colleagues performed the first quantitative comparison between hesperornithines and extant diving birds by using morphometrics. The authors took measurements that characterized the shape of the hindlimb bones in a variety of diving birds, including hesperornithines, loons, grebes, cormorants, and diving ducks. They then plotted these measurements against each other (after removing the influence of body size), which allowed them to assess the geometric similarity between each bone in different bird species.

As it turns out, though hesperornithines did overlap with loons or grebes on some measurements, they were generally more similar to diving ducks and especially cormorants! (It's perhaps also worth noting that, despite being frequently discussed in conjunction with one another, loons and grebes often occupied separate regions of the plots in this study.)

Flightless cormorant swimming, photographed by "putneymark", under CC BY-SA 2.0. Better than loons as a functional analogue for hesperornithines?

What does this mean for our understanding of how hesperornithines lived and functioned? It's hard to say. The authors point out that the influence of hindlimb bone shape on locomotion in diving birds has not been well studied, though their findings do suggest that future studies on hesperornithine biomechanics should probably look to cormorants as modern functional analogues instead of (or at least in addition to) loons or grebes.

Bell et al. also investigated the popular inference that hesperornithines had grebe-like lobed toes, instead of the webbed feet common in other diving birds. This idea was first suggested in the 1930s because similarities were observed between grebes and Hesperornis in the shape of the joint surfaces between their feet and their fourth toes. However, Bell et al. found that the same joint morphology is also present in some diving birds that have webbed feet, such as the flightless cormorant. Conversely, grebe-like joint morphology is not found in coots (which have lobed toes).

This does not mean that it is now incorrect to depict hesperornithines with lobed feet, but it does indicate that there is no compelling reason to think that their feet were lobed rather than webbed. Furthermore, the authors point out that the grebe-like joints are really only observed in the most specialized hesperornithines (such as Hesperornis and Parahesperornis), so even if joint morphology correlated with the soft tissues of the toes, lobed toes would have only been present in these later forms (which is consistent with the conclusions of some other recent research).

Western grebe showing off its lobed (not webbed) toes, photographed by Britta Heise, under CC BY 2.0. Hesperornithines have commonly been depicted with similar toes, but the evidence for this may be weaker than previously thought.

What about terrestrial locomotion? Hesperornithines have also been widely considered to have been similar to loons and grebes in how they would have gotten around on land; in other words, it's assumed that they wouldn't have gotten around well at all. Though grebes and the red-throated loon are capable of walking for short distances, it is very energetically costly for them to do so, and other loon species can't walk upright at all, instead having to awkwardly shuffle on their bellies. Cormorants, on the other hand, have no trouble with standing and walking upright on land, even if they won't be running marathons anytime soon. Do the findings of the new paper suggest that we need to rethink this aspect of hesperornithine biology as well?

Bell et al. briefly address this question. Unlike in typical dinosaurs, the hindlimbs of foot-propelled diving birds are often held splayed out to their sides. As such, it is advantageous for diving birds to have a very short femur (thigh bone), because this would allow them to place their feet directly behind their bodies, increasing the efficiency of the kicking strokes that they use for swimming. A possible trade-off of this is that if the femur is dramatically reduced in length compared to the rest of the hindlimb, this may impede the ability of a diving bird to balance upright on land. Indeed, loons have the shortest femora relative to their other hindlimb bones out of all the extant diving birds examined in the study, with grebes (especially the large-bodied western and Clark's grebes) coming in second.

As for hesperornithines, in the smaller and less specialized Baptornis and Brodavis, femur length relative to the foot is most similar to that of cormorants, but femur length relative to the lower leg is closest to that of the grebe genus Podiceps. In contrast, the larger Hesperornis and Parahesperornis had femoral proportions (especially relative to the lower leg) that most closely match those of loons. So if I had to guess, most hesperornithines probably could exhibit at least grebe-like if not cormorant-like gaits on land, but it appears plausible that the very specialized hesperornithids (Hesperornis and Parahesperornis) really were confined to loon-like shuffling.

Western grebe walking on land, photographed by Kevin Cole, under CC BY 2.0. This type of locomotion is very tiring for these specialized diving birds and cannot be sustained for long. Based on evidence from relative proportions of the hindlimb bones, some hesperornithines such Brodavis and Baptornis may have been able to achieve at least this level of terrestrial competence.

The upshot of this paper is that perhaps we should start thinking of hesperornithines as super-cormorants rather than super-loons, and we can no longer assume that they had grebe-like feet. Nonetheless, the most specialized hesperornithines may have been loon-like in their compromised ability to move on land. Of course, it's to be expected that no single group of modern birds serves as a perfect hesperornithine analogue; after all, the later hesperornithines evolved specializations for swimming and diving more extreme than those of any foot-propelled diving bird today. Most of all, this study highlights the need for us to test long-held assumptions using hard data instead of first impressions, or else we risk pigeonholing ourselves in nooks that may not, in fact, provide the best fit.

Reference: Bell, A., Y.-H. Wu, and L.M. Chiappe. In press. Morphometric comparison of the Hesperornithiformes and modern diving birds. Palaeogeography, Palaeoclimatology, Palaeoecology in press. doi: 10.1016/j.palaeo.2017.12.010

Friday, January 5, 2018

Favorite Maniraptor of 2016 Results

Last year was a close race, but the unusual Fukuivenator narrowly won against Tongtianlong, an oviraptorosaur known from a nearly complete skeleton preserved in three dimensions. Coming in third place was Apatoraptor, one of the most completely known caenagnathid oviraptorosaurs.

These polls are inevitably biased in favor of taxa that receive more press. Though a truly level playing field is impossible to achieve, I will try something new this year by providing a concise blurb for each contestant and links to their original description papers. This way at least, none of the taxa will be complete unknowns. That being said, I suspect one particular genus already has this year in the bag...

  • Albertavenator: A troodont from the Late Cretaceous of Canada. Known from a single bone in the skull.
  • Almas: A small troodont from the Late Cretaceous of Mongolia. Known from a partial skeleton including a nearly complete skull. It has been included in several phylogenetic analyses in the past, but wasn't named until last year.
  • Aprosdokitos: A small penguin from the Eocene of Antarctica. Known from a single upper arm bone.
  • Beibeilong: An oviraptorosaur from the Late Cretaceous of China. Known from an embryo long nicknamed "Baby Louie" as well as several eggs. Its adult size would have been comparable to that of Gigantoraptor, the largest known oviraptorosaur.
  • Chupkaornis: A hesperornithine from the Late Cretaceous of Japan. Known from a partial skeleton including parts of the vertebral column and hind limbs. It is the most completely known Asian hesperornithine.
  • Corythoraptor: An oviraptorosaur from the Late Cretaceous of China, one of many recently named from the Nanxiong Formation. Known from a nearly complete skeleton. It had a particularly tall crest on its head.
  • Crexica: A rail from the Miocene of Russia. Known from partial limb bones.
  • Cruralispennia: An enantiornithine from the Early Cretaceous of China. Known from a mostly complete skeleton with preserved feathers. It had several features unusual for an enantiornithine, including strap-like feathers on its legs and wings as well as a rapid growth rate.
  • Daliansaurus: A troodont from the Early Cretaceous of China. Known from a nearly complete skeleton.
  • Diomedavus: A small albatross from the Oligocene of the USA. Known from parts of the limbs and hip as well as a single neck vertebra.
  • Garrdimalga: A large megapode from the Pleistocene of Australia. Known from parts of the limbs and skull.
  • Halszkaraptor: A small dromaeosaurid from the Late Cretaceous of Mongolia. Known from a nearly complete skeleton. It had some unusual anatomical features suggesting that it foraged in the water, including numerous teeth, a long neck, and paddle-like forelimbs.
  • Jianianhualong: A troodont from the Early Cretaceous of China. Known from a nearly complete skeleton with preserved feathers.
  • Junornis: An enantiornithine from the Early Cretaceous of China. Known from a mostly complete skeleton with preserved feathers. The size and shape of its wings suggest that it used bounding flight, similar to many small extant birds.
  • Kumimanu: A large penguin from the Paleocene of New Zealand. Known from a partial skeleton including parts of the limbs, hip, and vertebral column. Among the most completely known giant penguins.
  • Latagallina: A large megapode from the Pleistocene of Australia. Known from several specimens, including one that is nearly complete. The type species was formerly considered a species of Progura.
  • Latenivenatrix: A large troodont from the Late Cretaceous of Canada. Known from several specimens, including a partial skeleton. Many of these were formerly considered specimens of Troodon. The largest troodont known, estimated at over 3 m long.
  • Liaoningvenator: Yet another troodont from the Early Cretaceous of China. Known from a nearly complete skeleton.
  • Maaqwi: A euornithine from the Late Cretaceous of Canada. Known from parts of a forelimb. Its thickened bone walls suggest that it was a diving seabird.
  • Miohypotaenidia: A rail from the Miocene of Russia. Known from partial limb bones.
  • Ostromia: A basal paravian from the Late Jurassic of Germany. Known from a partial skeleton including parts of the limbs and faintly-preserved wing feathers. The holotype has had a convoluted taxonomic history, having been mistaken for a pterosaur and later considered a specimen of Archaeopteryx.
  • Piscivorenantiornis: An enantiornithine from the Early Cretaceous of China. Known from a partial skeleton along with a possible pellet containing fish bones.
  • Serikornis: A basal paravian from the Late Jurassic of China. Known from a complete skeleton with "silky" feathers preserved all over its body.
  • Tsidiiyazhi: A stem-mousebird from the Paleocene of the USA. Known from a partial skeleton. The oldest known member of the "core landbird" clade Telluraves.
  • Vanolimicola: A shorebird-like bird from the Eocene of Germany. Known from a partial skeleton.
  • Zhongjianosaurus: A small dromaeosaurid from the Early Cretaceous of China. Known from a partial skeleton. One of the smallest dromaeosaurids known.

Wednesday, January 3, 2018

Review of 2017

Activity on this blog took a dip last year, most likely on account of me finishing up work for my Master's degree and starting my PhD. Perhaps also reflecting these events, I attended more conferences last year than in the past, among them ProgPal, SVP, TetZooCon, and PalAss. As for visits to zoos, museums, and other notable attractions, the highlight was probably my trip to the temporary Dinosaurs of China exhibit in Nottingham.

Regrettably, other than a handful of Tumblr posts and a parody during April Fools', I didn't draw anything new for the Raptormaniacs comic at all. Though the prospects of doing long-form storylines in the foreseeable future are uncertain, I hope to produce material for at least the Tumblr askblog somewhat more regularly over the coming year.

It would be wrong to say that this blog completely stagnated, however, as it was last year that I decided to start writing more detailed blog posts about new scientific studies on maniraptors. The opportunity to work on these posts was (and will be) highly contingent on my possession of increasingly mythical free time, but on the whole I was personally quite satisfied with the few articles that I was able to produce. As such, they will likely continue to be a recurring feature on this blog in the future.

I may not have made many comics last year, but I did take some time to draw some fun things such as this field guide to Mesozoic Twitter birds.

Speaking of new studies, it's time for the annual overview of maniraptor-related discoveries that were published over the past year. In January, new material of Garganornis was described. A record number of yellow-billed oxpeckers (51-60 individuals) was reported feeding on a single host. Elephant bird DNA was extracted from eggshell material. New studies came out on the coevolution of pygostyle morphology and tail feathers in Mesozoic avialans, the migration of Kirtland's warblers, hind limb morphometrics of phorusrhacids, the cranial osteology of Buitreraptor, the phylogeny of aquiline eagles, correlation between avian footprint and wing morphology, and phalangeal joint kinematics in ostriches. Newly-named maniraptors included the Eocene penguin Aprosdokitos mikrotero and the enantiornithine Cruralispennia multidonta.

Holotype of Cruralispennia multidonta, from Wang et al. (2017).

In February, a new female specimen of Eoconfuciusornis provided insights into its soft tissue and coloration. New material of Gargantuavis was described, as was a Paleocene penguin from the Waipara Greensand. External morphological measurements were documented for a variety of passerines. New studies came out on sentinel behavior in Arabian babblers, group behavior in chimney swifts during approaches to their roost, the function of egg shape in murres and cranial fenestration in neornithines, seed dispersal by mallards, the evolution of gestural displays in icterids and nest shape in passerines, and the macroevolutionary dynamics of bill evolution in neornithines. Newly-named maniraptors included the Miocene anhinga Macranhinga ameghinoi and the Eocene charadriiform-like bird Vanolimicola longihallucis.

Well-preserved Eoconfuciusornis specimen with feather melanosomes highlighted in insets, from Zheng et al. (2017).

In March, the soft tissues of Anchiornis were described using laser fluorescence. The bald head of northern bald ibises was shown to have a thermoregulatory function. Sociality was found to be inversely correlated with brain size in woodpeckers, unlike the pattern seen in some mammal groups. Kea were discovered to produce a specific vocalization that likely functions similarly to laughter. Preserved tendons/ligaments and cartilage in Confuciusornis and their functional implications were reported. New studies came out on craniofacial diversification in domestic pigeons, functional niche partitioning in therizinosaurs, adjustment of migration in a newly-established population of barn swallows, bone histological correlates of flight style in birds, scaling of owl hind limbs, social dynamics in non-breeding common ravens, pectoral girdle morphology in Mesozoic avialans, the phylogenetic position of the São Tomé grosbeak, and sources of conflict in resolving neornithine phylogeny. Newly-named maniraptors included the Miocene rhea Opisthodactylus kirchneri and the Tatamá tapaculo (Scytalopus alvarezlopezi).

Foot pads of Anchiornis revealed under laser fluorescence, from Wang et al. (2017).

In April, the neural basis of species recognition in brown-headed cowbirds was identified. New specimens of Anchiornis were described. A volume on the avian respiratory system was published. New studies came out on the migratory connectivity of semipalmated sandpipers, turn-taking ceremonies in common murres, and cumulative culture in pigeons. Newly-named maniraptors included the dromaeosaurid Zhongjianosaurus yangi, the enantiornithine Piscivorenantiornis inusitatus, and the Miocene rails Crexica crexica and Miohypotaenidia tanaisensis.

Juvenile brown-headed cowbird, photographed by Bill Bouton, under CC BY-SA 2.0.

In May, Egyptian vultures were reported to use soil as cosmetics. Common ravens were shown to remember deceptive behavior from other individuals for over a month. New studies came out on the safekeeping of tools by Goffin's cockatoos, the energetics of foraging flight in Pacific parrotlets, oxygen isotope fractionation between bird bone and drinking water, the diversification of capuchino seedeaters, the evolution of beak shape in waterfowl, the function of slotted wing tips during flight in Eurasian jackdaws, and the balancing mechanism of flamingos while standing on one leg. Newly-named maniraptors included the oviraptorosaur Beibeilong sinensis (based on an embryo long known as "Baby Louie"), the troodonts Jianianhualong tengi and Liaoningvenator curriei, and the dry forest sabrewing (Campylopterus calcirupicola). 2017 could well be considered the Year of the Troodontid for dinosaur paleontology, given that several other new troodont taxa would be named in the coming months.

Pacific parrotlet, photographed by "peterdehaas2317" and modified by "Snowmanradio", under CC BY 2.0.

In June, a nearly complete juvenile enantiornithine was reported preserved in amber. The genetic signature of flightlessness in flightless cormorants was identified. The supposed Miocene auk Petralca was reevaluated as a loon. A new specimen of Buitreraptor was described. Neomorphus cuckoos were suggested to deter predators by mimicking sounds made by peccaries. Seasonally conspicuous plumage was found to prompt risk avoidance in male superb fairy-wrens. New studies came out on the determination of flightlessness in extinct anatids, the function of avian egg shape and double kneecaps in ostriches, the historical range of the Carolina parakeet, the migration of gray-cheeked thrushes, the incubation temperatures of oviraptorosaur eggs, the allometry of pelvic dimensions in birds, and rhythmic drumming in palm cockatoos. Newly-named maniraptors included the Oligocene albatross Diomedavus knapptonensis, the troodont Daliansaurus liaoningensis, the Pleistocene megapodes Garrdimalga mcnamarai, Latagallina naracoortensis, Latagallina olsoni, and Progura campestris, the recently extinct Greater Azores bullfinch (Pyrrhula crassa), and the blue-winged amazon (Amazona gomezgarzai).

Juvenile enantiornithine preserved in amber, from Xing et al. (2017).

In July, common ravens were shown to be capable of planning for future tool use and bartering. Blue petrels were found to be able to recognizing their own eggs by scent. The Liberian greenbul was argued to be a plumage variant of the icterine greenbul. Neornithines were suggested to have undergone a reduction in body size across the K-Pg extinction. Osteomyelitis was documented in an Eocene penguin. Kelp gulls and dolphin gulls were reported to inflict perineal wounds on South American fur seal pups. New Zealand swans were confirmed to have represented a distinct, now-extinct lineage from extant black swans. Japanese tits were discovered to decode novel code sequences using the ordering of specific vocalizations. New studies came out on the osteology of Neuquenraptor, the developmental pathways leading to flightlessness in paleognaths, the biogeographic history of accentors, the bone microstructure of Vegavis, song discrimination in Neotropical passerines, wing development in emus, joint mobility in the avian neck, correlated evolution between sternal keel length and ilium length in birds, camouflage in ground-nesting birds, and the flight style of Sapeornis. Newly-named maniraptors included the Paleocene stem-mousebird Tsidiiyazhi abini, the troodont Albertavenator curriei, and the oviraptorosaur Corythoraptor jacobsi.

Diagram showing the likely soaring capabilities of Sapeornis, from Serrano and Chiappe (2017).

In August, the eggs of Heyuannia were shown to have been blue-green in color. The atlas rib of Archaeopteryx was reported for the first time. "Tasidyptes" was demonstrated to be an invalid taxon, being based on remains of several extant penguin species. Complex plumage patterns were found to be produced mainly with contribution from melanin pigments. The late Pleistocene passerines of Liang Bua in Indonesia were described. New studies came out on seed dispersal by macaws, the bone histology of Daliansaurus and dodos, object categorization by tits, the cranial morphology of Avimimus, the migration of penguins, the dominance hierarchy of North American birds, the kinematics of ostrich knees, skeletal traits shaped by embryonic activity in birds, recognition of individual eggs by house sparrows, the evolution of avian intelligence and senses, and the range of motion in avian hind limbs. Newly-named maniraptors included the hesperornithine Chupkaornis keraorum, the troodont Latenivenatrix mcmasterae, the anchiornithid Serikornis sungei, and the recently extinct kohatu shag (Leucocarbo septentrionalis). Stenonychosaurus was resurrected as a valid taxon, the name Troodon becoming restricted to the original holotype tooth.

The eggs of Heyuannia and their inferred coloration, from Wiemann et al. (2017).

In September, Sapeornis and caenagnathids were shown to have lost tooth sockets through ontogeny. The calls of female common cuckoos were found to misdirect the defenses of their hosts. Okarito kiwis were reported to be capable of surviving in the wild without eyesight. Barn owls were discovered not to experience hearing loss with age. Complex songs from conspecifics were demonstrated to elicit greater aggression in tui. New studies came out on the crafting of hook tools by Goffin's cockatoos, the diversification of anseriforms, the caudal anatomy of Buitreraptor, species recognition in Neotropical passerines, the wing morphing of hovering amazilia hummingbirds, cooperative feeding in African penguins, object exploration in kea and New Caledonian crows, and multitasking in pigeons. Newly-named maniraptors included the painted manakin (Machaeropterus eckelberryi).

Vestigial tooth sockets in (A) Caenagnathasia and (D) Sapeornis, from Wang et al. (2017).

In October, the phylogenetic positions of large flightless neognaths were evaluated. Vegavis, Polarornis, and Australornis were found to form a clade of stem-anseriforms. Selection for longer bills was observed in great tits in the United Kingdom. An alvarezsaurid from the Bissekty Formation was described. Territorial song was documented in female dark-eyed juncos. New studies came out on yellow feather pigmentation in budgerigars, bone fusion in Pterygornis, reconciliation in carrion crows, the preservation of preen gland oils in an Eocene bird, song recognition by golden-crowned sparrow nestlings, heterospecific eavesdropping in ant-following birds, wing allometry in hummingbirds, genome evolution in munias, ecological separation between Mesozoic avialans and pterosaurs, and cooperative breeding in birds. Newly-named maniraptors included the enantiornithine Junornis houi.

Phylogeny of neornithines with an emphasis on large flightless neognaths, from Worthy et al. (2017).

In November, Goffin's cockatoos were found to be capable of using simple keys. A confuciusornithid from the Huajiying Formation was described. A supposed oviraptorosaur from the Santana Formation was reevaluated as a megapraptoran. Whooping cranes were reported to associate with their future mates long before sexual maturity. The aerobic performance of tinamous was shown to be limited by their small heart size. Gastroliths were identified in several specimens of Jeholornis. Increased bill size was documented in snail kites, associated with the invasion of a novel prey item, the island apple snail. The regulatory modules involved in feather development were identified. New studies came out on the hindlimb musculature of foot-propelled swimming birds, the developmental basis of tactile specialization in ducks, song learning in zebra finches, sounds made using flight feathers in crested pigeons, correlation between aerobic power and flight style in birds, the origins of avian secondary temporal fenestrae, the functional anatomy of the Gigantoraptor jaw, the function of chick-carrying in eared grebes, convergent evolution in petroicids, feather structure in Anchiornis, the phylogeny of kingfishers, and genomic diversity in passenger pigeons. Newly-named maniraptors included the oviraptorosaur Avimimus nemegtensis and Guyramemua, a new genus for the chapada flycatcher.

Goffin's cockatoos matching shapes and suitable frames, from Habl and Auersperg (2017).

In December, a preserved retina was reported in an enantiornithine specimen. The fossil feather Praeornis was interpreted as a ribbon-like feather, such as those found on the tail of confuciusornithids and enantiornithines. The anatomical features associated with scent-based foraging in turkey vultures were identified. The ornamentation of male golden-crowned manakins was shown to have arisen through hybrid speciation. Phorusrhacid specimens from the late Pleistocene were described. A tick was found entangled in a Cretaceous pennaraptor feather. New studies came out on the evolution of crowned pigeons, the body mass of the dodo, magnitude coding of space and time by pigeons, tool shape variation in New Caledonian crows, the growth of dental tissues in therizinosaurs, the neuronal basis of imitative learning on swamp sparrows, the perception of male signals by female brown-headed cowbirds, the developmental origins of mosaic evolution in the avian skull, and the phylogenetic positions of Chendytes and the Labrador duck. Newly-named maniraptors included the anchiornithid Ostromia crassipes (based on a specimen long considered an individual of Archaeopteryx), the Cretaceous euornithine Maaqwi cascadensis, the Paleocene penguin Kumimanu biceae, the troodont Almas ukhaa, the semi-aquatic dromaeosaurid Halszkaraptor escuilliei, and the Cordillera Azul antbird (Myrmoderus eowilsoni).

Holotype and skeletal reconstruction of Halszkaraptor escuilliei with various notable aspects of its anatomy highlighted, from Cau et al. (2017).