Two extinct taxa, Moorea Sandpiper Prosobonia ellisi and Tahiti Sandpiper P. leucoptera, once occurred on Moorea and Tahiti, respectively. Four illustrations of Prosobonia from the second and third Cook expeditions (1772–75 and 1776–80) exist, of which one was the model for P. ellisi, whilst two others depict P. leucoptera and one Kiritimati Sandpiper P. cancellata. Considerable confusion exists as to whether P. ellisi is a valid species or an intraspecific variant of P. leucoptera. We examined the Tahiti / Moorea illustrations and original notes by crew of the Cook expeditions. We conclude that P. ellisi should be regarded as a junior synonym of P. leucoptera, as the differences between them may represent age-related, sexual, seasonal or even inter-island variation.
Four illustrations made by artists on the second (1772–75) and third (1776–80) Cook circumnavigations depict shorebirds referred to the genus Prosobonia (Scolopacidae), which is currently considered to comprise five species of very rare or extinct Polynesian endemics. The first was drawn by Johann Georg Adam Forster (1754–94) during Cook's second voyage, in either August 1773 or April–May 1774, when Tahiti was visited. It illustrates what is considered the type of Tahiti Sandpiper P. leucoptera (J. F. Gmelin 1789) (Sharpe 1906b, Hume 2017: 150–151). The second and third illustrations were made by William Wade Ellis and John Webber during Cook's third voyage. Both were made at Moorea (visited between 30 September and 11 October 1777) (cf. Walters 1991: 224, Hume 2017: 151), and much later the bird depicted in the Ellis drawing was described as Moorea Sandpiper P. ellisi (Sharpe 1906b). The fourth illustration was made at Kiritimati, during the third circumnavigation, and shows Kiritimati Sandpiper P. cancellata (J. F. Gmelin 1789) (Latham 1785: 274, Walters 1993, Jansen & Cibois 2020). The other two species were discovered after Cook's expeditions: Tuamotu P. parvirostris (Peale 1849) and Henderson Sandpipers P. sauli (De Pietri et al. 2020).
In addition to these illustrations, there are descriptions by Johann Reinhold Forster (Lichtenstein 1844: 174–176) and a brief description made by William Anderson (Anderson c.1780). As part of a wider study of the history of Naturalis specimen of P. leucoptera (Jansen et al. 2021), we examined these original illustrations and descriptions.
Material and Methods
JJFJJ studied the following material at the Natural History Museum, London, UK (NHMUK) and British Museum, London (BM): Forster's illustration (now held in the R. B. R. Forster collection, Banksian MSS. 6–7, pl. 117), Ellis's illustration (in the collection of William Wade Ellis, 1751–85, Banksian MS. 33, pl. 65) and Webber's illustration (among the John Webber prints, British Museum, Prints and Drawings Dept.), as well as the Solander catalogue (Solander c.1780), at NHMUK, Tring. This catalogue documents the illustrations in the library of Joseph Banks at the time (Whitehead 1978, Medway 1979), and a list of specimens compiled by Jonas Carlsson Dryander (Whitehead 1978, Medway 1979).
Variation in original artwork executed during the second (1772–75) and third (1776–80) expeditions commanded by James Cook, and the description made during the second Cook circumnavigation by J. R. Forster. Only the relevant parts from the Forster description are noted (for the full description see Lichtenstein 1844).
Review of original material
Two illustrations are of birds from Moorea and one from Tahiti. It is unknown if the illustration from Moorea depicts the same bird as that from Tahiti. To understand the morphological and anatomical differences in the illustrations we analysed the original Forster, Ellis and Webber illustrations, and the Forster description in Lichtenstein (1844: 174–176) (Table 1), but we excluded two c.1780 manuscripts.
Anderson.—In Anderson's (c.1780) manuscript describing animals observed on Captain Cook's second and third voyages, Charadrius tardus (Anderson's manuscript name) is mentioned with a short description (our translation): ‘Head/body black above, with a stripe above the eyes, a wing mark, the belly and the undertail white; the breast and rump reddish or brown.' It was mentioned alongside two other species of ‘Plover' (Hooded Plover Thinornis cucullatus from New Zealand and a ‘blackish plover' from Terra del Fuego (possibly Magellanic Plover Pluvianellus socialis). However, from the manuscript it is impossible to determine if the ‘Charadrius tardus’ was actually collected (contra Stresemann 1950: 76, Hume 2017: 151).
Solander.—Herein (Solander c.1780), Tahiti Sandpiper is mentioned as: ‘93 / 1 / Webber Ellis / (Te-te) / Otaheite Eimeo’ (Otaheite = Tahiti, Eimeo = Moorea).
Discussion and Conclusions
The three illustrations examined all differ (Table 1), and none exactly matches the sole surviving mount, RMNH.AVES.87556. The Webber illustration (drawn at Moorea) shows a bird very similar in appearance to the Prosobonia specimen at Naturalis (Jansen et al. 2021). The Ellis illustration, which is not quite finished (i.e. the nails are uncoloured), was used to describe the Moorea taxon. Sharpe (1906b) described Ellis' illustration as differing from that by Forster in having a circlet of rufous around the eye; a double patch of white on the wing-coverts, and the median and greater wing-coverts pale ferruginous, like the rump. However, the Webber illustration, also made on Moorea (Fig. 3), was not mentioned by Sharpe (1906a,b). According to Walters (1991), the Ellis (Fig. 2) and Webber (Fig. 3) depictions are of the same species. However, they differ in at least ten points: bill shape and thickness, ear patch colour, throat colour, tibia feathering, tail pattern, wing-coverts pattern, tail and wing lengths, and leg and underparts colorations. The Webber illustration differs in having a supercilium (absent in Ellis'), only a single patch of white on the wing-coverts (not two), and differently patterned median and greater coverts. The rump is ferruginous, but the same feature is present on the surviving specimen (Jansen et al. 2021).
At that time illustrators were less accurate, especially in details, than now; for example, virtually all depictions of Dodo Raphus cucullatus differ in some respects (Fuller 2002) as do known illustrations of the sole Tahiti Sandpiper specimen in Joseph Banks' collection (Figs. 4–6). It is very unlikely that birds from Moorea were morphologically distinct from those on Tahiti (as these islands are separated by just 18 km). Also, there is evidence of individuals exhibiting patchy white feathers in several Polynesian Acrocephalus warblers (Thibault & Cibois 2017), which could be evidence of a limited gene pool and explain the minor variation in the meagre sample of sandpipers.
We consider the differences between Moorea and Tahiti birds to represent age, sex, season or inter-island variation, rather than evidence of a separate taxon. It is possible that future work testing ancient DNA of archaeological remains on Moorea may resolve this issue, but until then we conclude that Prosobonia ellisi Sharpe, 1906b, is a junior synonym of Tringa leucoptera J. F. Gmelin, 1789.
We especially thank Laurent Raty for help with translating the Forster text, and Erik Åhlander for additional information on Anders Sparrman. Björn Bergenholz, Alison Harding and Giorgio Aimassi helped with literature. Murray Bruce, Vanesa De Pietri and Rick Roe commented on an earlier version. Staff at the British Museum and NHMUK are thanked for access to the original artwork. Paul Scofield and Guy Kirwan improved the manuscript dramatically, and our sincere thanks to them.