Ducula lakeba (Lakeba Imperial Pigeon)

[Deleted]

Hi !

Described from bones from the fijian island Lakeba (speak: Lakemba).
This bird was larger than any other Ducula but was not flightless. It also had relative longer legs than any other Ducula sp.

[another specialist]

A new genus and species are erected for a giant flightless pigeon described from post-cranial fossil bones from four Quaternary sites on Viti Levu, Fiji, in the South Pacific. Lack of cranial material and transformations associated with large size and loss of flight obscure its relationships, but some characters of the tarsometatarsus, tibiotarsus, and humerus are closer to Goura than to any other taxon. The new taxon approaches the dodo and the solitaire in size, but its wings were not as reduced as in these Mascarene birds. A tentatively referred premaxilla suggests that the Fiji bird had an unspecialised bill. It coexisted with a similar-sized, flightless megapode, a terrestrial crocodilian, and a giant iguana, which were the largest animals in prehuman Fiji.

www.rsnz.org/publish/jrsnz/2001/40.php

[another specialist]

A new volant species of Ducula is described from Lakeba Island in the Lau Group, Fiji, that was larger than any extant Ducula, and had more elongate tarsometatarsi than Ducula david, the largest described fossil species. Bones of a Ducula similar in size to the Lakeba pigeon are reported from Viti Levu.

www.rsnz.org/publish/jrsnz/2001/40.php

[another specialist]

This is the complete PDF file where the above info was found

www.rsnz.org/publish/jrsnz/2001/40.pdf

[another specialist]

anotherspecialist said:

This is the complete PDF file where the above info was found

www.rsnz.org/publish/jrsnz/2001/40.pdf

Ducula lakeba new species
(Fig. 7-10; Tables 7, 8)
ETYMOLOGY: The species is named after Lakeba, the island on which it was first found.
DIAGNOSIS: Tarsometatarsus longer than that of the biggest extant Ducula species, D. goliath,
and D. galeata. Proximal width is greater, and shaft length from the proximal end to the fossa
metatarsi I is relatively longer (relative to proximal width) than in D. david of Wallis Island
(Balouet & Olson 1987), the largest fossil species, yet its mid-shaft width is less than in D.
david (Table 7).
HOLOTYPE: A right tarsometatarsus, Fiji Museum, Archaeology Department, bone numbered
197-3-W-1-6, 6. Cast of type, MNZ S38899. For measurements see Table 7.
TYPE LOCALITY: Qara-ni-puqa Rockshelter (site 197), Lakeba Island, Fiji Group. Collected
by Simon Best et al. between December 1976 and September 1977.
PARATYPES: Fiji Museum Archaeology Department, 18 specimens: [Format: "element, number
on bone;" where 197 is "site 197", the following digit (1 or 3) refers to the excavated square,
and the letter (w or v) refers to the layer the specimen was derived from] pR tmt, 197.1.W.2,
11; pL tmt, 197.3.W.1.4, 3; sL tmt, 197.1.W.2, 23; dR tmt, 197.3.W.1, 1 (NW baulk); s+dL
tmt, 197.3.W.1.8, 5; pR tmt, 197.3.W.1.3,47; sLtmt, 197.3.W.1, 2; sL tmt, 197.1.W.2, 10; sL
tmt, 197.1.w.2,40;dLtmt, 197.3.w.l.5,28; s+dRtmt, 197.1.V.2: 3; s+dRtmt, 197.3.w.l.3,2;
s+dLtmt, 197.3.w.3(w), l;dLtmt, 197.3.w.l.3,27;sLtmt, 197.3.w.2,6;s+dRtmt, 197.1.W.2,
36+3; dL tmt, 197.3.w. 1.5, 27; sL tmt, 197.3.w. 1.8, 30. Layer 'w' is the basal layer in the site.
REFERRED MATERIAL: Fiji Museum Archaeology Department, 73 specimens: Part dL hum,
197.1.W.2, 18;sRhum, 197.1.V.2, l;ptdLhum, 197.1.X.2, 4; dL hum, 197.3.W.1.1, 10; pL
hum, 197.3.W.1.8, 8; L ulna, 197.3.W.1.4, 2; sR ulna, 197.1.X.2, 3; sL ulna, 197.3.V.1, 5; R
rad, 197.3.w.v.2, 10;Lrad, 197.3.W.1.6, 3; dR rad, 197.1.W.1, 5; sRrad, 197.3.W.1.8, 7; pR
rad, 197.3.W.1.7, 3; R rad, 197.3.W.1.3, 24; dR rad, 197.3.W.2, 5; dL rad, 197.1.W.2, 12; pL
rad, 197.1.V.2, 5; pR rad, 197.1.W.2, 9; R rad, 197.3.v.3(w), 7; pt R cmc, 197.1.W.2, 22; sR
cmc, 197.3.V.1, 8; d+sR cmc, 197.1.W.2, 7; sR cmc, 197.1.V.2, 21; sR cmc, 197.1.W.2, 14;
s+dL cmc, 197.3.W.1.8, 4; R cmc, 197.3.w.2(E), 18; pL cmc , 197.3.W.1.5, 2; pL cmc,
197.1.W.1, 13; sR cmc, 197.3.V.2, 1; L M2.1, 197.3.W.1.8; sL cor, 197.3.W.1.5, 3; sR cor,
197.3.W.1.7, 7; sR cor, 197.1.W.2, 4; pL cor, 197.3.W.1.1, 54; pR cor, 197.3.W.1.6; pR cor,
197.3.W.1.1, 15; pR cor, 197.3.W.2, 9; R cor, 197.3.W.1.7, 48+5; R scap, 197.3.W.1.1, 30; L
scap, 197.3.W.1.8,40; synsacrum, 197.3.w.2(E), 32; ant. sternum, 197.3.W.1.6, 119; pR fern,
197.1.w.l,8;dRfem, 197.1.v.l(6), 2; sRfem, 197.1.W.2, 15; Lfem, 197.1.V.1, l;p+sLfem,
197.3.W.1.1, l;pRfem, 197.3.W.1.5, l;dLfem, 197.3.W.1.3, l;p+sLfem, 197.1.v.l(c), 1;L
fem, 197.1.v.l,fl46, l;Rfem, 197.3.W.1.3, 25; dLtib, 197.1.W.1, 9; sRtib, 197.1.wBIT, 1;
dRtib, 197.3.W.1.7, 37; dLtib, 197.1.W.2, 6; dR tib, 197.3.W.2, 11; dLtib, 197.3.W.2, 2; dL
tib, 197.3.W.2, 13; dRtib, 197.3.W.2, 12;sRtib, 197.3.W.1.5, 2 1367; pR tib, 197.3.W.2, 7; sR
tib, 197.1.W.2, 5; sRtib, 197.3.T.6(E), 3; s+pLtib, 197.3.W.1.3, 26; s+pLtib, 197.3.W.1.4, 1;
L fibula, 197.3.w.2(E); phal 1.1, 197.3.V.1, 17; phal 1.1, 197.3.W.3, 6;phal 1.1, 197.3.W.1.6,
8; phal 1.1, 197.3.w.2(E), 7; phal 1.1, 197.1.W.2, 43; phal 1.1, 197.1.W.1, 2.
DESCRIPTION: The holotypic tarsometatarsus is stained reddish brown and is complete
except for the loss of the crista medialis hypotarsi plantar of the sulcus hypotarsi. It was
reassembled from two pieces. Unfortunately, during casting of the bone, the trochlea III was
damaged. In addition to the diagnostic features described above for the tarsometatarsus, the
fossa metatarsi I begins at 55% of total length from the proximal end; the fossa infracotylaris
dorsalis is deep and merges with the sulcus extensorius to extend distally to the mid length
point; the tuberositas m. tibialis cranialis is not prominent; the fossa parahypotarsalis
medialis extends to the/o.s.sa metatarsi I.
The femora (Fig. 8) are typical of Ducula, but are larger than in other species (Tables 8-
10). Only two complete femora and a single complete tarsometatarsus of Ducula lakeba are
known, but they suggest that this bird had relatively long tarsometatarsi (femur length 1.23
times tarsometatarsus length at most). In comparison, D. galeata has short tarsometatarsi
(femur length 1.42 times tarsometatarsus length), and all other pigeons have even shorter
tarsi (Table 9).
The tibiotarsi (Fig. 9) of Ducula lakeba share with other Ducula species the following
characters: the tuberosity for the proximal attachment of the retinaculum extensorium tibiotarsi
is at a distance from the condylus medialis equivalent to more than twice the height of that
condyle; the sulcus extensorius is bound by a prominent and sharp ridge laterally; there is a
flat area laterad of the pons supratendineus with no pneumatic foramen and no tuberosity for
the medial attachment of the retinaculum m. fibularis; and the distal width is greater than the
anteroposterior depth of the condyles. The tibiotarsus of D. lakeba is larger than that of D.
galeata or D. goliath (Table 10), the two largest extant species, and the tibiotarsus of D.
david is not known.
The humerus is represented only by small fragments that present no distinguishing
features other than the suggestion that the fossa m. brachialis is relatively deep. Fragments of
ulnae, radii, and carpometatcarpi (Fig. 10) are large versions of Ducula.
The coracoid (Fig. 10) referred to Ducula lakeba is similar to Ducula in that the medioventral
side of the humeral end of the shaft is rounded; the sulcus m. supracoracoideus is smooth and
shallow and little overhung by the proc. acrocoracoid (deep, overhung in Columba as seen in
Fig. 10) and lacks pneumatic foramina; the fades artic. sternalis is mediolaterally expanded
but dorsoventrally compressed; the impressio m. sternocoracoidei is deepest medially and
separated from the medial edge by a thin ridge. This coracoid has a greater shaft width (4.2
mm) than the paratype of D. David (3.1 x 3.3 mm) Balouet & Olson (1987).

[another specialist]

anotherspecialist said:

This is the complete PDF file where the above info was found

www.rsnz.org/publish/jrsnz/2001/40.pdf

[another specialist]

A large Ducula from Viti Levu
Six specimens attest to the former presence of a large species of Ducula on Viti Levu that is
of comparable size to D. lakeba. Unfortunately, sufficiently complete comparable elements
are lacking to provide adequate comparison with Ducula lakeba. The material includes a
complete right humerus (MNZ S37316, Table 8, Fig. 11) from Wainibuku Cave; a L femur
(MNZ S37477) from Udit Tomo; a proximal L femur (FM unreg.) from an unknown point
inside Volivoli Cave, collected by J. C. Balouet; a proximal right coracoid and a fragment of
a left scapula (MNZ S38709) from Vatuma Cave, c. 12 km south of Nadi; and a sternal end
of a coracoid from 70-80 cm depth in the Kulu Bay Lapita archaeological site on Beqa
Island, c. 10 km south of Viti Levu. I can only conclude that Ducula lakeba or a closely
related taxon formerly lived on Viti Levu.



www.rsnz.org/publish/jrsnz/2001/40.php

[Sebbe]

- Worthy, T. H. & Clark, G. 2009. The Early Prehistory of Fiji: Bird, mammal and reptile remains. Terra Australis 31.

Ducula lakeba Lakeba pigeon
Material: total 92 bones. The elements and their catalogue numbers are listed in Worthy (2001b).
Most specimens are from Layer W. Two are from Layer X, 13 from Layer V, and one from Layer
T, i.e. in the basal Lapita layers of the site. This species was described by Worthy (2001b) and
named for the island from which it was first identified. However, it is a volant pigeon and is
expected to have occurred on all the larger islands of the Lau Group. A similar, if not the same
species, occurred on Viti Levu as well. It is slightly larger and more gracile than Ducula david
described from Wallis Island by Balouet and Olson (1987).

press-files.anu.edu.au/downloads/press/p49471/pdf/101.pdf