The Last Terror Birds: A review of Phorusrhacids and their Plio-Pleistocene occurrences

Introduction

66 million years ago, a meteor the size of a mountain slammed into the sea, right off the coast of what is now Mexico. Soot, ash and burning smolders filled the skies, earthquakes struck across the planet, and tsunamis, a hundred meters tall, crashed over the coasts. This was the end of the Mesozoic, and with it, the Age of Dinosaurs—or so we say. Yet some dinosaurs did survive, members of a small, feathered lineage, graced with the ability of flight. It is an amusing twist of fate—or more rightly, of evolution—that some descendants of these furtive, flying animals would abandon these traits once more, turning away from those very adaptations that saved them where their larger cousins vanished, and returning to a shape and lifestyle much more “dinosaurian” than their flighted relatives. These were the ’terror birds’, more rightly the members of the family Phorusrhacidae, whose outsized role in the prehistory of South America is matched only by their ubiquity in the prehistoric pop-culture of our modern day. Few groups of the Cenozoic, mammal, bird or reptile, have captivated the modern imagination as much as the terror birds. Denizens of South- and, for a time, North America, they ranged from the Eocene all the way until the Pliocene and, as we will discuss in this article, probably the Pleistocene as well. 

Fig 1. A red-legged seriema (Cariama cristata), one of the closest living relatives of the terror birds.

Terms of use: This image is licensed under an Attribution-ShareAlike 3.0 Unported. The image is attributed to José Reynaldo da Fonseca and is unedited.

Taxonomy

The family Phorusrhacidae is fairly well-defined and united through several skeletal traits, the most iconic of which is a huge upper jaw ending in a single downfacing point (2), but a host of other features distinguish them from other birds. The closest relatives are the Seriemas (Cariamidae) (2, Worthy et al 2017) along with a number of other extinct families such as the Bathornithidae (2), they are united in the clade Cariamiformes.

Traditionally, the Phorusrhacidae is divided into 5 subfamilies: Brontornithinae, Mesembriornithinae, Patagornithinae, Phorusrhacinae and Psilopterinae (2). Substantial disagreements regarding this taxonomy exists, and relationships between subfamilies is equally controversial.

First up are The Psilopterines, a group of small, lightly built Phorusrhacids, which hardly inspire the name ‘terror bird’ as they may superficially have resembled the seriemas (Cariamidae) and secretary birds (Sagittaridae) (Though up to twice as big as the latter). The family contains its namesake Psilopterus, with the genera Procaraima and Paleopsilopterus also originally included (2), though the former has also been placed within Mesembriornithidae (10) and the latter is quite controversial and may not actually constitute a phorusrhacid at all (4). A recently discovered genus Eleutherornis has also been assigned to the psilopterines (4). This clade is accepted by most phylogenetic studies (1, 10), though 1 considers them to be a family separate from but closely allied to Phorusrhacidae (1). Psilopterines are undoubtedly the oldest known subfamily, depending on the status of Paleopsilopterus dating back to either the Palaeocene or Eocene epochs (2, 4), and potentially the very last survivors of the clade (16).

A size of up from the Psilopterines, we have the Mesembriornithinae. Initially erected based solely on the genus Mesembriornis which is a medium-small genus of terror bird, with a somewhat gracile build (Alveranga & Hofling 2003). An additional two similarly sized genera are suggested to belong to this group now, the aforementioned Procariama and the newly discovered Llallawarvis (10). All genera of this subfamily are known from the Late Miocene and Early Pliocene (2, 10). Much like the psilopterines, the group seems well supported by phylogenetic analyses (10, 1), though 1 once again does not consider it to belong to Phorusrhacidae and instead places it as the immediate sister family.

Most messy are the remaining three subfamilies, which are much closer to the mental image evoked by the term ‘terror birds’ all possessing especially large beaks and considerable sizes. Starting with the Patagornithinae, named after the genus Patagornis, but also comprising Andrewsornis and the particularly well-studied Andalgornis (2), these medium sized phorusrhacids were relatively slim in build though with skulls much greater in size than either the Mesembriornithines or Psilopterines and are known from the middle Oligocene to the Early Pliocene (Alveranga & Hofling 2003). Brontornithinae on the other hand were amongst the largest of the terror birds and the bulkiest, the group consists of the namesake Brontornis as well as Physiornis and Paraphysiornis and these genera were known from the middle Oligocene to the middle Miocene (2). Recent analyses have shown however that Brontornis itself is not a Phorusrhacid at all! In fact, it seems to belong either to a separate branch of the Cariamiformes (Worthy et al 2017) or the Waterfowl (Anseriformes) much like the giant Gastornithidae and Dromornithidae (Agnolin 2021), where exactly this leaves Physornis and Paraphysornis within Phorusrhacidae is unclear. Lastly are the best known of the terror birds, the Phorusrhacines which include famous genera such as Phorusrhacos, Titanis and Kelenken (5) as well as the more obscure Devinsenzia (2), Kelenken probably constitutes the as of yet largest terror bird (5). Phorusrhacines struck somewhat of a middle ground between the gracile Patagornithines and the bulky Brontornithines. As neat as these subdivisions are, phylogenetic analysis both by 1 and Degrage et al 2015 contest the validty of them. Simply put members of these three subfamilies all cluster together and should probably be treated a single subfamily (or in the opinion of 1, a single family), whether any of these three subfamilies form distinct clades is less clear, 1 found monophyly in the Brontornithinae, but paraphyly in the Phorusrhacines (1), whereas 10 found no monphlyly at all (10).

This overview has omitted a number of controversial remains, including the probable phorusrhacid Lavocatavis and additional phorusrhacids which have recently been reassigned to other groups (8). Clearly much more work needs to be done on terror bird taxonomy and phylogenetics before a clear picture can be generated.

Biogeography and origins

Likewise, the origins of the terror birds are controversial. Phorusrhacids are best known from South America where they have a rich fossil record which includes the vast majority of the known genera and species. Conclusive remains dating back to the early Eocene (4) are known, with Paleopsilopterus possibly pushing the date back to the Palaeocene (Alvarez & Hofling 2003), and highly controversial material even suggested from the cretaceous period (2, 8). Until recently it was widely accepted that Terror birds originated from South America and were endemic there until they entered North America during an event known as the Great American Biotic Interchange (GABI) where North and South America formed the landbridge in Panama during the Late Pliocene (4). However, the recent discovery of Lavocatavis from Algeria, and Eleutherornis France and Switzerland suggests that during the Eocene Phorusrhacids were present in Europe and Africa as well (4). Findings from Antarctica have also been documented, though highly disputed (2, 8). This places uncertainty on the point of origin of the Phorusrhacids. Likely they did not originate in Europe as the earliest remains are younger than those of Africa and South America, and no other remains from Laurasia are known (Barring the late remains from North America). Instead, early Phorusrhacids likely crossed into Europe from Africa across the Tethys (4). Between Africa and South America, it is much harder to ascertain a point of origin, the two continents likely split apart in the Early Cretaceous but there seems to have been a corridor of dispersal between the two lasting into the Cenozoic (4) as other groups such as monkeys seem to have crossed during this time. So whilst we don’t know which landmass they evolved on, they likely made it across from one to the other. This hypothesis seems to be favoured over a gondwanan vicariance hypothesis, however, should strongly evidence of phorusrhacids from the cretaceous or Antarctica appear this would change.

Fig 2. Putative Pleistocene Phorusrhacid sites.

Ecology

The size of Phorusrhacids varied heavily as might be expected of such a diverse group. 2 estimates a mass of the smallest species, Psilopterus bachmani at 5kg with a height of about 80cm and the largest estimate for a species at 180kg for Paraphysornis brasiliensis (2) (not including the Brontornithines), though Kelenken guillermoi likely reached larger sizes (5, 10). The degree of size differences within each species is unclear: sex, age or even different spatial or temporal differences in populations are impossible to establish without an abundant fossil record.

Most Phorusrhacids were undoubtedly flightless, their wings were heavily reduced and absolutely could not carry their heavy mass (2). Some controversy exists regarding the smaller Psilopterines, if they were capable of flight it would have been in a very limited capacity as their wing to body ratios are inferior to those of the related sereimas (Cariama spp.) which are only capable of gliding flight (2). However, whether Psilopterines (and indeed early Phorusrhacids in general) could fly at all has large rammifications on their ability to disperse across continents, which would lend credence to the continent hopping to Europe and between South America and Africa.

More controversial is the cursorial abilities of the terror birds, Alverenga & Hofling 2003 suggests that most species were rather slow moving due to their relatively short tarsometatarsus (The lowest leg bone in birds), though noting some members of the family such as Patagornis exhibited a higher degree of cursoriality (2). A more recent mechanical analysis suggests otherwise, it estimated the running speed of Mesembriornis at a staggering 97km/h and Patagornis at an impressive 50km/h, demonstrating that the birds were capable of rapid movement (6). An unclassified large Phorusrhacine was estimated at 50km/h as well, though this number may a slight overestimate according to the authors, even so these were quite clearly not slow-moving organisms (6, 9). Interestingly in a morphometric analysis terror birds clustered with different bird groups in terms of hing leg and pelvic proportions, indicating a wide range of locomotion methods across the subfamilies. The small Psilopterines clustered closely with the bustards (Otididae), indicating they may have been walkers rather than runners and may have been aided by limited flight capabilities, though affinities to wading birds is also noted (22). The rapid Mesembriornithinae on the other hand appear adept runners clustering closely with the ratities (22) and the Patagornithines showing affinities to both these groups (22).

Terror birds also had a characteristically large skull with a downwards pointing tip on the upper beak, much like many birds of prey, these undoubtedly point towards a predatory diet, but their prey species and feeding methods are still debated. 14 found that the cranium of the terror bird Andalgornis best withstood dorso-ventral (top to bottom) stresses, and very weakly lateral (sideways) stress, and this probably was true for all non-psilopterine terror birds, though the skulls of some large terror birds were more compact (9). This indicates that the group weren’t capable of holding down prey like many mammalian predators (14) or that they did so with their feet (14). Instead, it seems they either fed on animals significantly smaller than themselves, which would not put up too much resistance or alternatively used their tip like a hatchet and produced well-targeted strikes to vulnerable parts of the prey species, such as the skull or neck (14). Both the structure of the ear canal and neck would support the use of rapid movements of the skull which would be expected in these strikes or pursuit of swift prey species (10, 19). Little is known of the sensory capabilities of the terror birds, except their auditory capabilities which appear comparable to sereimas and ratites (13). Overall Phorusrhacids appear to be widely different, but show tendencies towards a pursuit predatory lifestyle. 10 suggests a niche partitioning between contemporary phorusrhacids with the large Phorusrhacines preying on megafauna, medium sized patagornithines on medium sized prey in the 10-20kg range and the small psilopterines on small animals (10(2)), but further details need to be resolved this can be fully evaluated.

Decline and Extinction

The decline and extinction of the Phorusrhacids is usually ascribed to the Great American Biotic Interchange, which took place during the early Pliocene period and was an event in which North and South America collided via the Isthmus of Panama.  Particularly the influx of various groups of carnivorous mammals such as big cats (Felidae) and bears (Ursidae) (Cione et al 2015) are often thought to have outcompeted the terror birds. This narrative is not so clear-cut as terror birds are quite rare in the fossil record from the early Pliocene (2), whereas these new groups of hypercarnivores do not become well established until the Pliocene-Pleistocene boundary (Cione et al 2015). This gap could simply be a result of the poor taphonomy of Pliocene South America or be a result of an early immigration event such as the canids and procyonids (Which grew to large sizes in the Pliocene)(Cione et al 2015). A detailed investigation of the decline and extinction of the Phorusrhacids is outside the scope of this article, however regardless of whether they declined due to the GABI or some other event, the known diversity was limited past the Late Pliocene. And yet, terror birds did not go extinct immediately following the GABI, remains from three of four species of terror bird have been described from the time period following this event.

The Last Terror Birds

Fig 3. Two Psilopterus over the fossilised skull of an older, far larger species of terror bird, some time in the Late Pleistocene.

Terms of use: Artwork by Hodari Nundu and Commissioned by The Extinctions

The first and most well-known holdout was Titanis walleri, a phorusrhacine which weighed around 150kg and reached a height of up to 1.9 meters (17). Undoubtedly a large predator, Titanis flies in the face of the narrative that placental carnivores outcompeted terror birds, at least initially, because not only is the genus known only from after the Great American Biotic Interchange, but it also appears endemic to North America. This means the Phorusrhacids must have colonised the continent despite the presence of a largely placental megafauna guild (17).

Exactly how long Titanis persisted has been the topic of much debate, the trouble arises with the sites in which its fossils were found, the Nueces River and Santa Fe river, in Texas and Florida respectively (17). Each of these sites contain species from both Pliocene and Pleistocene faunal assemblages, usually this is not a problem as most taxa ages are simply inferred from other sites, however Titanis has a scarce record and is only known from two other sites, both dated to the Late Pliocene (17). Recently a study by 17 investigated Titanis remains from the putatitve Pleistocene sites and measured the concentrations of Rare Earth Elements (REEs) in the fossils and other fauna from the site. REEs tend to accumulate in the few thousand years following death but will do so at different rates depending on the local geochemistry at the time. In essence this allows us to cluster together fossils from the same time period. At the Nueces river Titanis clustered with Early Pliocene species for REE values and in the Santa Fe River with Late Pliocene, leaving no credible reports of Pleistocene occurences. Nevertheless, the existence of Titanis corroborates a continuance past the GABI. For a long time Titanis was the only known species to survive past this revolutionary event in biogeography, however additional material has been uncovered in the age-old terror bird stronghold of South America.

In 1999 Tambussi et al described an as of yet unnamed species of Phorusrhacine from Uruguay, it lacks an official designation due to the fragmentary nature of its remain, known only from a single tibiotarsus. Nevertheless, based on the size of the bone, it was a large species comparable in size to Titanis and undoubtedly an apex predator (20). Unfortunately, much like Titanis, the dating for this species is fraught with complications. The tibiotarsis was found in the Raigon formation, which does have age uncertainties associated with it (1) but is generally dated to the the Pliocene-Early Pleistocene (20). However, the particular strata in which the phorusrhacine was found has been pinned down to the Late Pliocene-Middle Pleistocene (7, 16) based on the faunal assemblages preserved therein. This would make the species contemporary with, if not somewhat younger than Titanis and provide strong evidence for the continued existence of large terror birds in South America after the GABI.

Another Phorusrhacid fossil from Uruguay was described in 2010 by Alvarenga et al, this time a tarsometatarsus from the Casil Quarry, it is another species of late surviving terror bird (3). Astoundingly this species was found at the same site as Middle-Late Pleistocene species such as Macrauchenia patachonica, Toxodon platensis and Natiomastodon platensis, a tooth of the latter from the same strata was even dated to 17,620 years ago, placing it tantalisingly close to the arrival of humans (Homo sapiens) and the extinction of the South American megafauna in the Early Holocene. An additional tarsometatarsus was reported by 16 from Perico Flaco Creek also in Uruguay which highly resembled the specimen by 3. Little is known about this species, but it appears the individual was comparable in size to Procraiama simplex, which was the largest known species of psilopterine terror bird at an estimated 10kg (2, 3). The taxonomic affinities of the specimen are unclear as the material is restricted to two partial tarsometatarsi. 3 discounted affinities to the Brontornithines and Psilopterines due to morphological differences and noted particular similarities with Titanis and 16 disagreed with this assessment, dismissing any resemblance to Titanis and instead suggests a proximity to Procariama or potentially Psilopterus (16). Additional material may be necessary to shed light on the taxonomy.

A final fossil discovered may or may not provide just that insight. 16 described a humerus of a phorusrhacid from Curupi Creek, Uruguay, nearby sediments were dated to the Late Pleistocene around 96,040 before present (16). This bone undoubtedly belongs to the Psilopterine terror birds and especially Psilopterus, which the specimen has been assigned to under the binomial Psilopterus sp. with a species designation likely to follow when additional material is uncovered (16). Given the fragmentary nature of the fossil, it may be subject to future reclassification in the light of more material. This is particularly true as the current designation would stretch the age of the Psilopterus genus from the middle Oliogocene until the Late Pleistocene, a period of about 30 million years (Alvarenga & Hofling, 16) making it simultaneously one of the oldest and the youngest phorusrhacid genus due to its morphological conservativeness. The relationship between this humerus and the aforementioned tarsometatarsuses is unclear, it is plausible that they belong to the same Late Pleistocene species, as 16 specifically notes the affinity to psilopterines of the tarsometatarus. The humerus does appear to belong to a slightly smaller specimen, but this could be accounted for by intra-specific variation. This appears quite unequivocal evidence of the continued survival of terror birds until the Late Pleistocene with independent datings to both 17,620 and 96,040 years ago of small phorusrhacids and large terror birds until at least the Late Pliocene, but perhaps as late as the Middle Pleistocene.

It is quite remarkable that in the menagerie of bizarre animals encountered by early human settlers of South America, we can count amongst them a terror bird the size of a small bustard. What was this species like? As discussed earlier , it was likely a predator of small animals and may have been capable of limited flight, but beyond that we still know little of this last survivor, or really the infamous terror birds in general. It can be hoped that with the discovery of the Late Pleistocene Psilopterus more resources will be devoted to describing avifauna fossils from South America, as it is an understudied field and perhaps one day we may have a clearer understanding of the ecology of these incredible birds.

References

1.      Agnolin, F. ( 2013). The systematic position of Hermosiornis (Aves, Phororhacoidea) and its phylogenetic implicances. Revista del Museo Argentino de Ciencias Naturales, 15, pp.39-60. DOI: 10.22179/revmacn.15.167

2.      Alvarenga, H. and Höfling, E. (2003). Systematic revision of the Phorusrhacidae (Aves: Ralliformes). Papéis Avulsos de Zoologia (São Paulo), 43(4). DOI: 10.1590/s0031-10492003000400001

3.      Alvarenga, H., Jones, W. and Rinderknecht, A. (2010). The youngest record of phorusrhacid birds (Aves, Phorusrhacidae) from the late Pleistocene of Uruguay. Neues Jahrbuch für Geologie und Paläontologie - Abhandlungen, 256(2). 229-234. DOI: 10.1127/0077-7749/2010/0052

4.      Angst, D., Buffetaut, E., Lécuyer, C. and Amiot, R. (2013). “Terror Birds” (Phorusrhacidae) from the Eocene of Europe Imply Trans-Tethys Dispersal. PLoS ONE, 8(11), p.e80357. DOI: 10.1371/journal.pone.0080357

5.      Bertelli, S., Chiappe, L. and Tambussi, C. (2007). A new phorusrhacid (Aves: Cariamae) from the middle Miocene of Patagonia, Argentina. Journal of Vertebrate Paleontology, 27(2), pp.409-419. DOI: 10.1671/0272-4634(2007)27[409:anpacf]2.0.co;2

6.      Blanco, R. and Jones, W. (2005). Terror birds on the run: a mechanical model to estimate its maximum running speed. Proceedings of the Royal Society B: Biological Sciences, 272(1574). 1769-1773. DOI: 10.1098/rspb.2005.3133

7.      Bossi, J., Ortiz, A. and Perea, D. (2009). Pliocene to middle Pleistocene in Uruguay: A model of climate evolution. Quaternary International, 210(1-2), pp.37-43. DOI: 10.1016/j.quaint.2009.08.011

8.      Cenizo, M. (2012). Review of the putative Phorusrhacidae from the Cretaceous and Paleogene of Antarctica: new records of ratites and pelagornithid birds. Polish Polar Research, 33(3). 239-258.

9.      Chiappe, L. and Bertelli, S. (2006). Skull morphology of giant terror birds. Nature, 443(7114), pp.929-929. DOI: 10.1038/443929a

10.  Degrange, F. (2015). Hind limb morphometry of terror birds (Aves, Cariamiformes, Phorusrhacidae): functional implications for substrate preferences and locomotor lifestyle. Earth and Environmental Science Transactions of the Royal Society of Edinburgh, 106(4), pp.257-276. DOI: 10.1017/s1755691016000256

11.  Degrange, F. and Tambussi, C. (2011). Re-examination of Psilopterus lemoinei(Aves, Phorusrhacidae), a late early Miocene little terror bird from Patagonia (Argentina). Journal of Vertebrate Paleontology, 31(5), pp.1080-1092. DOI: 10.1080/02724634.2011.595466

12.  Degrange, F., Eddy, D., Puerta, P. and Clarke, J. (2019). New skull remains of Phorusrhacos longissimus (Aves, Cariamiformes) from the Miocene of Argentina: implications for the morphology of Phorusrhacidae. Journal of Paleontology, 93(06), pp.1221-1233. DOI: 10.1017/jpa.2019.53

13.  Degrange, F., Noriega, J. and Vizcaíno, S. (2015). Morphology of the forelimb of Psilopterus bachmanni (Aves, Cariamiformes) (Early Miocene of Patagonia). Paläontologische Zeitschrift, 89(4), pp.1087-1096. DOI: 10.1007/s12542-015-0269-1

14.  Degrange, F., Tambussi, C., Moreno, K., Witmer, L. and Wroe, S. (2010). Mechanical Analysis of Feeding Behavior in the Extinct “Terror Bird” Andalgalornis steulleti (Gruiformes: Phorusrhacidae). PLoS ONE, 5(8), p.e11856. DOI: 10.1371/journal.pone.0011856

15.  Gould, G. C. & Quitmyer, I. R. (2005). Titanis Walleri: Bones of Contention: Bull. Fla. Mus. Nat. Hist. 45(4). 201-229.

16.  Jones, W., Rinderknecht, A., Migotto, R. and Blanco, R. (2013). Body mass estimations and paleobiological inferences on a new species of large Caracara (Aves, Falconidae) from the late Pleistocene of Uruguay. Journal of Paleontology, 87(1), pp.151-158. DOI: 10.1666/12-026r.1

17.  MacFadden, B., Labs-Hochstein, J., Hulbert, R. and Baskin, J. (2007). Revised age of the late Neogene terror bird (Titanis) in North America during the Great American Interchange. Geology, 35(2). 123. DOI: 10.1130/g23186a.1

18.  Mourer-Chauviré, C., Tabuce, R., Mahboubi, M., Adaci, M. and Bensalah, M. (2011). A Phororhacoid bird from the Eocene of Africa. Naturwissenschaften, 98(10). 815-823. DOI: 10.1007/s00114-011-0829-5

19.  Tambussi, C., de Mendoza, R., Degrange, F. and Picasso, M. (2012). Flexibility along the Neck of the Neogene Terror Bird Andalgalornis steulleti (Aves Phorusrhacidae). PLoS ONE, 7(5), p.e37701. DOI: 10.1371/journal.pone.0037701

20.  Tambussi, C., Ubilla, M. and Perea, D. (1999). The youngest large carnassial bird (Phorusrhacidae, Phorusrhacinae) from South America (Pliocene–Early Pleistocene of Uruguay). Journal of Vertebrate Paleontology, 19(2). 404-406. DOI: 10.1080/02724634.1999.10011154

21.  Vezzosi, R., (2012). First record of Procariama simplex, Rovereto, 1914 (Phorusrhacidae, Psilopterinae) in the Cerro Azul Formation (upper Miocene) of La Pampa Province; remarks on its anatomy, palaeogeography and chronological range. Alcheringa: An Australasian Journal of Palaeontology, 36(2), pp.157-169. DOI: 10.1080/03115518.2011.597657

22.  Worthy, T., Degrange, F., Handley, W. and Lee, M. (2017). The evolution of giant flightless birds and novel phylogenetic relationships for extinct fowl (Aves, Galloanseres). Royal Society Open Science, 4(10), p.170975. DOI: 10.1098/rsos.170975