Centropus bairdi, Shute & Prideaux & Worthy, 2016

Shute, Elen, Prideaux, Gavin J. & Worthy, Trevor H., 2016, Three terrestrial Pleistocene coucals (Centropus: Cuculidae) from southern Australia: biogeographical and ecological significance, Zoological Journal of the Linnean Society (Zool. J. Linn. Soc.) 177 (4), pp. 964-1002 : 975-980

CENTROPUS BAIRDI SP. NOV. ( FIGS 2–4)

urn:lsid:zoobank.org:act:25375F5E-14C5-4AD6-AE12-332F21F5AEA2

Holotype ( Fig. 2C, H, M)

WAM 09.3.282 (L humerus, complete). Measurements are given in Table 2.

Referred material

WAM 09.3.277 (complete L femur); WAM 09.3.279 (complete R tibiotarsus), and WAM 09.3.278 (R tibiotarsus, shaft with missing proximal and distal ends); WAM 09.3.281 (L tarsometatarsus, missing distal end) and WAM 09.3.280 (L tarsometatarsus, missing distal end). Including the holotype humerus, number of individual specimens = 6, minimum number of individuals = 2. All specimens were excavated from within 3 m of one another and from a similar depth (see Stratigraphy, age, and fauna), but the bones were not articulated or closely associated. All bones are assumed to be from the same species of Centropus because of their congruent size and robusticity, and because collectively they are distinct from those of Ce. phasianinus, Ce. colossus , and Ce. maximus sp. nov. described later in this paper (see Diagnosis). Measurements of long bones are given in Table 2; additional measurements are given in text where applicable.

Type locality

Leaena’s Breath Cave, Nullarbor Plain, Western Australia. The precise location for the site is registered with the Department of Earth and Planetary Sciences, Western Australian Museum , Perth. Leaena’s Breath Cave is one of the three caves comprising the Thylacoleo Caves ( Fig. 1 ). It is formed within the Early Miocene-aged Nullarbor Limestone .

Stratigraphy, age, and fauna

All specimens were excavated from Pit B, Quadrats 1 and 3, Unit 3, in the main chamber of Leaena’s Breath Cave, between 85 and 120 cm below the sediment floor. Leaena’s Breath Cave has a vertical solution pipe entrance approximately 20 m deep, through which the specimens of Ce. bairdi are presumed to have accumulated via pitfall trapping. The roof window of the cave is thought to have been open and accumulating fossils at intervals during the Early to Middle Pleistocene, and was then sealed by a calcrete cap until recently. The holotype was excavated by G. J. P. from Pit B, Quadrat 1, at a depth of 115– 120 cm beneath the sediment floor, on Friday 10 May 2013. Fine-grained sediments in Unit 3 are of reversed magnetic polarity, which along with the overall species composition of the assemblage, indicate an Early Pleistocene age> 780 kyr BP (Matuyama Reversed Chron; Prideaux et al., 2007). This falls within the Naracoortean land mammal age ( Megirian et al., 2010).

Diagnosis

A species of Centropus with a humerus within the length range of that of Ce. phasianinus , but with the shaft less strongly arched dorsally and relatively more stout, the corpus thicker where it approaches the caput humeri, the crista deltopectoralis craniocaudally thicker and with a rounder apex, the impressio m. brachialis shallower and much narrower, the ventral margin of the processus flexorius more ventrally extended with associated greater dorsoventral width of the processus in caudal aspect, the condylus ventralis proximodistally shorter and less spherical, a larger expanse of bone dorsad of the condylus dorsalis, the sulcus scapulotricipitalis wider and flatter, and with leg bones considerably longer and more robust.

The holotype humerus ( Fig. 2C, H, M) differs from those of other large extant and extinct Australian cuculids in both size and morphology: it is very much smaller than the humerus of S. novaehollandiae , and the two extinct Pleistocene species Ce. colossus (see above) and Ce. maximus sp. nov., which is described later in this paper; it is larger and more robust than the humerus of E. scolopaceus , with a shorter, more cranially directed crista deltopectoralis (angled more dorsally in E. scolopaceus ), a thicker crus ventrale fossa, a more proximodistally extended intumescentia humeri, and the processus flexorius directed ventrally rather than distally.

The fossil humerus is longer than that of the medium-sized species Ce. ateralbus from the Bismarck Archipelago, but is almost identical in length and shaft width to the humeri of two of the largest extant Melanesian coucal taxa, Ce. milo ( Solomon Islands) and Ce. violaceus ( Bismarck Archipelago). Centropus bairdi differs from both species by the following features: the proximal and distal widths are smaller; the caput humeri is more prominent proximally; the crista bicipitalis joins the shaft at a sharper angle distally (gradually tapered in Ce. violaceus and Ce. milo ); the crista deltopectoralis is more craniodorsally expanded, with an apex that is more angular (short, robust, and barely projecting in Ce. violaceus and Ce. milo ); the distal end of the bone is more ventrally flared; and the impressio m. brachialis is highly reduced in area. In addition, the fossa pneumotricipitalis of Ce. bairdi has almost no pneumatization whereas Ce. milo is strongly pneumatic.

Etymology

Named in honour of Dr Robert F. Baird, who described Australia’s first extinct species of Centropus in 1985, and who worked extensively on the Quaternary bird fossil record of Australia, including fossil cave faunas in the Nullarbor region.

Description and comparisons

All fossil bones of this species are stained dark brown by the clay-rich sediments in which they were buried. Apart from both tarsometatarsi missing their distal ends and one tibiotarsus missing its proximal and distal ends, the bones are in excellent condition, are not significantly eroded, and preserve anatomical detail well. The bone surfaces lack a porous texture and the tarsometatarsi have completely fused epiphyses, indicating that the elements represent adult individuals.

Humerus: In addition to the diagnostic features described above, the humerus of Ce. bairdi has the following features: the intumescentia humeri is relatively small, occupying around a third of the proximal width in cranial view; the crus dorsale fossae is relatively thicker than for other species; the impressio m. brachialis is narrow and elongate, hence the area for insertion of this muscle is greatly reduced, and this is accentuated by the fact that the shaft is not swollen ventrally around the fossa m. brachialis, unlike in Ce. phasianinus, Ce. violaceus, Ce. milo, Ce. ateralbus , and Ce. colossus .

Femur: Single complete specimen ( Fig. 3C, H, M), with minor erosion of the facies articularis acetabularis on the caput, and of the condylus medialis and condylus lateralis. It is much larger and more robust than the femur of any extant Australian cuculid, including the closest in length, that of Ce. phasianinus , but is considerably smaller than the femur of Ce. colossus described above, and that of Ce. maximus sp. nov. described below ( Fig. 3). Despite its smaller size, its shape and proportions are similar to that of Ce. colossus , being robust and having a straight shaft in cranial aspect (Ce. phasianinus has a gracile, laterally bowed shaft – see Fig. 3E). The junction between the facies articularis antitrochanterica and the cranial surface of the bone is smoothly rounded as in Ce. phasianinus , unlike in Ce. colossus where a distinct ridge separates these surfaces. At its caudal edge, the facies articularis antitrochanterica forms a rounded projection that overhangs the caudal facies (does not project in Ce. phasianinus , cannot be determined in Ce. colossus owing to damage), and distal of this the caudal surface of the shaft is convex (strongly concave in Ce. phasianinus and Ce. colossus ). The fossa poplitea is deeper than in Ce. phasianinus , being bounded laterally by a thickened area of shaft proximal of the trochlea fibularis, and medially by a strongly developed crista supracondylaris medialis. The insertion of the m. gastrocnemialis lateralis appears less distinct than in Ce. phasianinus or Ce. colossus , although it is possible that this region of the femur has been somewhat eroded in this specimen. The femur is shorter but overall more robust than in Ce. violaceus , and lacks a proximal projection of the trochanter femoris (present in Ce. violaceus – Fig. 2F). The impressio ligamentum collateralis lateralis is in a distinct fossa on the lateral facies as in other taxa. The origin of the ligamentum cruciati caudalis is on the lateral wall of the fossa poplitea, directly proximal to the crista tibiofibularis, as in Ce. colossus and Ce. violaceus , but is less distinct than in those taxa. The groove caudally separating the condylus medialis from the condylus lateralis in which the ligamentum cruciati cranialis runs is shallower and less rectilinear than in Ce. colossus . Measurements (mm): for TL, PW, SW, and DW, see Table 1; proximal depth: 11.4; minimum shaft circumference = 19.5.

Tibiotarsus: The tibiotarsus ( Fig. 4A, B ) is large and robust, and is distinguished from those of all extant Australian cuculids by its much larger size. The nearest in size is the tibiotarsus of Ce. phasianinus , which is only around 70% of the length of that of Ce. bairdi and has a little over half the shaft diameter of the new species. Besides size, there are a few morphological differences from the tibiotarsus of Ce. phasianinus : the crista cnemialis cranialis of Ce. bairdi does not project as far cranially, and tapers more gradually towards the shaft at its distal extreme; the crista cnemialis lateralis projects less and is less hooked than in Ce. phasianinus ; and the medial facies of the bone is flattened at its proximal end, from the facies gastrocnemialis to about the mid-length of the shaft, whereas in Ce. phasianinus it is more medially convex. The tibiotarsus of the extinct Ce. colossus is not known, but the size of its femur (see above) indicates that the tibiotarsus would be significantly larger than that of Ce. bairdi . The tibiotarsus of Ce. bairdi is very similar in size and morphology to that of Ce. violaceus , but the muscular impression caudomedially on the shaft mesad of the crista fibularis is less extensive as revealed by its proximally bounding linea intermuscularis medially being well separated from that on the cranial facies (in Ce. violaceus , they are L, left; R, right; TL, total length; PW, proximal width; SW, mid-shaft width; DW, distal width; *, minimum measurement owing to damage.

closer together), the shaft is more robust, and the epicondylaris medialis is more pronounced. Measurements (mm): for TL, PW, SW, and DW, see Table 1; PW including crista cnemialis lateralis: 14.9; width at distal end of crista fibularis: 11.2; depth of cotyla lateralis: 11.1; depth of cotyla medialis: 12.1; minimum shaft circumference = 19.9.

Tarsometatarsus: The tarsometatarsus of Ce. bairdi ( Fig. 4C–E ) is larger than that of any extant Australian cuculid. The more complete of the two specimens ( WAM 09.3.280) preserves much of the shaft length (60 mm), but neither the fossa metatarsi I nor the trochleae are preserved. Even without the distal end, the incomplete bone is longer than the tarsometatarsus of Ce. phasianinus , and is considerably larger at the proximal end and in the shaft. Besides its greater size, the tarsometatarsus of Ce. bairdi is further distinguished from that of Ce. phasianinus by the following features: the excavation of the sulcus flexorius on the plantar surface extends into the distal half, whereas in Ce. phasianinus the plantar surface of the shaft is flat to convex along its distal half; the cotyla lateralis is proportionally wider, approaching the width of the cotyla medialis, and the tuberculum m. fibularis brevis forms a small shelf of bone that protrudes laterally from its lateroplantar corner ( Fig. 4E ) as in Ce. violaceus (in Ce. phasianinus the tuberculum has a lateral margin confluent with the condyle and projects plantarly); the eminentia intercotylaris is offset more plantarly from the dorsal margin; and the tuberositas m. tibialis cranialis is larger. The tarsometatarsus of the extinct Ce. colossus is not known, but on the basis of the much larger femur of this species, it would be expected to be much wider and longer than that of Ce. bairdi . Despite having a more robust femur and tibiotarsus than Ce. violaceus , the tarsometatarsus of Ce. bairdi is more gracile, and is strongly tapered along the proximal two thirds of the preserved length of the shaft distal of the proximal articular surface. The better-preserved tibiotarsus is contralateral to the two preserved proximal tarsometatarsi, precluding an assessment of the articulation of the tibiotarsus and the tarsometatarsus, but their size is congruent. Measurements (mm): for PW and SW, see Table 2; proximal depth including hypotarsus: WAM 09.3.280, 12.33; WAM 09.3.281, 13.5; mid-shaft depth: WAM 09.3.280, 4.3; WAM 09.3.281, not applicable. The more complete specimen, which was excavated from a shallower depth in the excavation ( Table 2) is a little smaller than the other specimen, but within the range of variation of what would be expected of sexually dimorphic taxa such as coucals.

Remarks

Centropus bairdi was intermediate in size between the extant Ce. phasianinus and the extinct Ce. colossus , and was larger than any extant species of Centropus . The humerus of Ce. bairdi is short in comparison to the leg elements ( Fig. 7 ), which along with the reduced degree of pneumatization and dramatically reduced area for attachment of the m. brachialis at the distal end of the humerus, suggests that this species was poorly adapted for flying, and perhaps even for gliding. It may have been truly terrestrial.

Figures

Figure 1. Map of key localities mentioned in the text, showing the range of extant Centropus phasianinus, and showing landbridge connections between Australia and Melanesia at different Pleistocene sea level low-stands (–40 and –130 m) relative to modern sea level; (modified from Sahul and Sunda Shelf Basemap, CartoGIS CAP, 2013, Map 00-101; downloaded from http://asiapacific.anu.edu.au/mapsonline/base-maps/sunda-and-sahul-shelf, 08-05-2015).

Figure 4. Referred material of Centropus bairdi sp. nov. from Leaena’s Breath Cave, Nullarbor Plain, south-central Australia. Right tibiotarsus WAM 09.3.279 in cranial (A) and caudal (B) aspects; left tarsometatarsus WAM 09.3.280 in dorsal (C), plantar (D), and proximal (E) aspects; left tarsometatarsus of Centropus phasianinus MV B.12969 (F) in plantar aspect provided for comparison. Abbreviations: ccl, crista cnemialis lateralis; cf, crista fibularis; cl, condylus lateralis; cm, cotyla medialis; ei, eminentia intercotylaris; fg, facies gastrocnemialis; fdl, canal for tendon of m. flexor digitorum longus; fhl, canal for tendon of m. flexor hallucis longus; fpht, fossa parahypotarsalis medialis; fvp, foramina vascularia proximalia; ii, incisura intercondylaris; ps, pons supratendineus; se, sulcus extensorius; sf, sulcus flexorius; tfb, tuberculum m. fibularis brevis; ttc, tuberositas m. tibialis cranialis. Scale bars = 1 cm.

Figure 7. Simpson log ratio diagram of extinct and extant Centropus bones; zero line represents the standard species channel-billed cuckoo Scythrops novaehollandiae against which all values are calculated. Abbreviations: HL, humerus length; HP, humerus proximal width; HM, humerus midshaft width; HD, humerus distal width; UL, ulna length; CMCL, carpometacarpus length; KD, keel depth; CL, coracoid medial length; FL, femur length; FP, femur proximal width; FM, femur midshaft width; FD, femur distal width; TTL, tibiotarsus length; TTP, tibiotarsus proximal width; TTM, tibiotarsus midshaft width; TTD, tibiotarsus depth; TML, tarsometatarsus length; TMP, tarsometatarsus proximal width; TMD, tarsometatarsus distal width. Gaps in lines reflect absent data owing to incomplete specimens.