Galápagos tortoises are evolutionary icons and are inextricably linked to Charles Darwin and the development of the theory of natural selection. The role that these giant reptiles played in the formulation of Darwin's ideas has been much discussed (e.g., Sulloway 1984, 2009; Caccone et al. 1999), but despite the intensive investigation of Darwin's collecting and collections (Porter 1985), there remains a great deal of confusion regarding the chelonians obtained during the voyage of the Beagle. Giant tortoises clearly made an impression on Darwin, whose Zoology Notes (published by Keynes 2000) discuss Galápagos tortoises at length, commenting on distribution, habitat, diet, locomotion, reproduction, vocalization, and use as food by humans. These notes were subsequently published in Darwin's Journal and Remarks (1839) and Journal of Researches (1845 et seq.). However, the significance of the tortoises as objects of scientific interest was at first underestimated by Darwin, as they were generally believed at the time to be introduced representatives of Testudo indica, the name then applied to the giant tortoises of the Indian Ocean (Sulloway 1984).

A relatively large number of tortoises were captured by the crew of the Beagle and taken on board. FitzRoy (1839) noted 18 small (< 80 pounds) tortoises taken on Chatham Island (San Cristóbal) on 18 September 1835 and another 30 large specimens from the same island in October. In addition, he remarked that tortoise oil and meat was obtained on James Island (San Salvador or Santiago), when Darwin, Benjamin Bynoe (surgeon on the Beagle), Harry Fuller (FitzRoy's steward), and Syms Covington (Darwin's assistant) were dropped off there on 8 October 1835. However, all of these specimens were consumed and their remains dumped overboard after butchering as the Beagle headed west across the Pacific on its return journey (Sulloway 1984).

The absence of adult tortoises from the Beagle collections has never been a serious point of contention. However, FitzRoy famously kept several small tortoises alive on the return voyage, noting (FitzRoy 1839, p. 504) that “a very small one lived upwards of two months on board the Beagle without either eating or drinking” while another “grew three-eighths of an inch, in length, in three months; and another grew two inches in one year. Several were brought alive to England.” Darwin (1839) also reported that the Beagle returned to England with young specimens of tortoises from 3 islands (Hood [Española], James [San Salvador or Santiago], and Charles [Floreana]). It is the fate of these juvenile tortoises, sometimes referred to as “pets”, that has been a source of controversy. This issue has recently been popularized by Chambers (2004), and the salient points are only briefly summarized here.

In the early part of 1837, Darwin brought 4 young tortoises, presumably all still alive at this time, to J.E. Gray at the British Museum to ask if the different island forms could be distinguished. These animals corresponded to Darwin's “pet” from James Island, Covington's from Charles, and FitzRoy's 2 tortoises from Hood (Garman [1917] incorrectly stated that young tortoises had been obtained on Charles, Chatham, and James). Gray expressed the view that the young tortoises could not be distinguished reliably (Darwin 1839; Günther 1875), and at this point the animals fade from history. Although information subsequently obtained from Gabriel Bibron and others indicated to Darwin (1839) that Galápagos tortoises were distinct from those of the Indian Ocean and that there were different varieties or species on different islands, the juvenile tortoises apparently played no further role in the development of Darwin's ideas.

According to Chambers (2004, p. 191), “On March 15 1837, the British Museum's Zoological Accessions book lists Captain FitzRoy as having donated seventeen animal specimens. Accession numbers two and three on the list are shown as being of the species Testudo from Española Island. These are almost certainly the 2 small tortoises whose measurements FitzRoy so carefully noted during the remainder of the Beagle's voyage after leaving the Galápagos.” Chambers (2004, p. 191) further noted of Darwin's and Covington's tortoises, that “the fact that they are not listed as being among the Beagle reptile specimens donated to the British Museum in 1837 suggests that they probably were still living” and further claimed that neither the BMNH nor any other British Institution had a record of 2 Galápagos tortoises being received between 1837 and 1842, when Darwin moved to Downe. The absence of records of these tortoises (indeed Desmond and Moore [1992] were unable to trace any Beagle tortoises) was used by Thomson et al. (1995, 1996) to suggest that “Harriet”, an aged tortoise living in the Australia Zoo might be Darwin's missing pet. Chambers (2004) effectively debunked any connection between Harriet and the Beagle but could not trace Darwin's tortoise. The story was further complicated by Darwin himself, who on 12 April 1874, in response to a query from Albert Günther, said that he brought home no tortoises as FitzRoy and the surgeon [Bynoe] did. He vaguely remembered that specimens had been given to “the Military Institution in Whitehall”. Chambers, believing that FitzRoy's 2 tortoises had been accessioned into the British Museum (now The Natural History Museum) collection, suggested that Darwin and Covington's tortoises may have been the ones sent to Whitehall (Royal United Services Museum, closed in 1968). However, he could find no evidence in the relevant archives of the tortoises ever having gone to that institution. Thus, the ultimate fate of Darwin's “pet” tortoise, and that belonging to his servant Covington, has remained a mystery.

In the course of a comprehensive inventory and evaluation of the herpetological specimens obtained on the voyage of the Beagle (Bauer and McCarthy, in prep.) we re-investigated the chelonian material from the Beagle that was presented to the British Museum. Our findings allow us to correct some previously published misstatements and to announce the rediscovery of Darwin's pet tortoise. Darwin collected some 282 specimens of amphibians and reptiles on the voyage of the Beagle. These were chiefly described by Thomas Bell in volume 5 (Reptiles) of Zoology of the Beagle. However, Bell's work was limited to lizards and frogs. Snakes and amphisbaenians were sent by Bell to Gabriel Bibron in Paris for identification and appear to now be lost (Donoso-Barros 1975). At least 156 specimens have been identified as present in The Natural History Museum (BMNH) collections and several more are in the Muséum National d'Histoire Naturelle (MNHN). Those in the BMNH were presented by Darwin or by Bell on behalf of Darwin between 1837 and 1856 (not all in 1845 as stated by Boulenger [1906]). A small number of additional Beagle specimens of reptiles were presented to the BMNH by Captain Robert FitzRoy and Sir William Burnett (Physician-General of the Navy and technically responsible for the official collections of naval expeditions). In total, there are records of 5 chelonian specimens from the Galápagos having been accessioned into the BMNH collections.

The BMNH General Zoological Register indicates that 3 chelonians from the Galápagos presented by FitzRoy and Burnett were accessioned in March 1837 as 1837.3.15.1, Chelonia, Galapagos, 1837.3.15.2 Testudo, Hood Island, and 1837.3.15.3 Testudo, jun., Hood Island (Fig. 1). The arrival of the first material from the Beagle corresponds to the period when the first general zoology register was initiated by John Edward Gray, then Assistant Keeper (Curator) of Zoology. This register included those animal groups for which Gray was responsible (echinoderms and mollusks in addition to vertebrates) (Wheeler 1996), and these 3 chelonian specimens are the first reptiles, and indeed the very first specimens, to have been entered under this system. None of these specimens can be located today, nor were they cited by Gray (1844) or any subsequent authors dealing with the chelonian collections of the BMNH. It is, therefore, likely that these specimens were lost or misplaced and/or became disassociated from their labels very early after they were accessioned into the collections. The identity of the sea turtle is uncertain, as 3 species of chelonids occur in the Galapagos, but Chelonia mydas agassizi is the most common (Fitter et al. 2000) and most likely to have been collected. The tortoises from Hood would have been Geochelone (or Chelonoidis) hoodensis (or nigra hoodensis or elephantopus hoodensis; see below for a brief discussion of nomenclature of Galápagos tortoises).

View Figure

• Download as Powerpoint
• Download Figure

Contrary to remarks by Desmond and Moore (1992) and Chambers (2004) these are not the only tortoises to appear in the BMNH register. Just 5 months after the accession of FitzRoy's material, the zoology register documents the accession of 2 juvenile tortoises, both indicated as presented by Charles Darwin Esq. (Fig. 2). These 2 animals were both entered in the collection as Testudo Indica jun., as 1837.8.13.1 (James Island, Galapagos) and 1837.8.13.2 (Charles Island, Galapagos)—Darwin's and Covington's pets, respectively. The receipt of these tortoises is likewise noted (albeit without species identification; Fig. 3) in a separate set of notebooks derived from the general register but containing only the relevant herpetological entries. According to Wheeler (1996) most of the entries in these notebooks were made by A.W.E. O'Shaughnessy, but the handwriting in the volumes prior to 1845 seems to be in the hand of Adam White, Assistant in the Zoology Department from 1835 to 1863.

View Figure

• Download as Powerpoint
• Download Figure

View Figure

• Download as Powerpoint
• Download Figure

Like FitzRoy's tortoises, these 2 specimens seem to have disappeared very quickly from the records of the museum. This is particularly odd, as the British Museum curators (keepers), extending from Gray, who met with Darwin upon his return from the Beagle voyage, through Albert Günther to George Albert Boulenger, each compiled at least one catalogue of the chelonian collections. Indeed, Gray's first catalogue was published in 1844, only 7 years after the accession of the 4 juvenile tortoises. Nonetheless, Gray's catalogue makes no mention of any specimens from Darwin, Covington, or FitzRoy. Nor do the later catalogues or monographs of Gray (1855, 1872), Günther (1875, 1877, 1896), or Boulenger (1889) mention such specimens. Interestingly, Thomas Bell (1792–1880), who prepared the reptile portion (volume 5) of the Zoology of the Voyage of H.M.S. Beagle (Bell 1842–1843) also made no mention of these tortoises in his treatment of Darwin's material. Bell himself was a specialist of chelonians and the bulk of his own collection resides in the Oxford University Museum of Natural History. However, it is unclear if Bell even knew of Darwin's tortoise because it may have already been lodged in the BMNH collection before he received the Beagle material from Darwin.

Careful examination of older specimens of tortoises in the collection, however, revealed that one specimen, BMNH 1837.8.13.1, Darwin's tortoise from James Island, was still present in the collection. This specimen had been “hiding in plain sight” for over 170 years. It had been reregistered as BMNH 1874.6.1.6 in the course of Günther's studies on giant tortoises and is entered in the register with the data “history not known, old collection.” The specimen is stuffed, with a straight carapace length of 178 mm (7 inches) (Fig. 4). Although it bears no external markings identifying it as a Beagle specimen, the inner face of the plastron, which is wired to the carapace, clearly bears the data “37.8.13.1 James Island” (Fig. 5). The specimen is entered in the register as a young specimen of Testudo ephippium Günther 1874.

View Figure

• Download as Powerpoint
• Download Figure

View Figure

• Download as Powerpoint
• Download Figure

The rediscovery of Darwin's tortoise definitively demonstrates not only that Darwin and Covington's “pets” did not travel to Australia (contra Thomson et al. 1995, 1996) but that they remained alive in England only for a matter of months at most (contra Chambers 2004). Indeed, if the specimen number was written on the plastron on 13 August 1837, the tortoise must have been cleaned and possibly already stuffed by that date, suggesting that it had been dead for some time. It seems likely that when Darwin brought the juvenile tortoises to Gray at the British Museum early in 1837 they were probably left there to be prepared as specimens.

With the recognition that BMNH 1837.8.13.1 and BMNH 1874.6.1.6 are identical, it was possible to trace the specimen through the nearly 5 decades of tortoise publications. Under Testudo indica, to which the Galápagos tortoise was then allocated, Gray (1844, p. 6) listed 2 specimens without locality or collector/presenter: “f. Young, 7 inches. Nuchal plate none.” and “g. Young, six inches. Nuchal plate none, feet bad.” The larger of these would appear to be BMNH 1837.7.13.1 and it is possible that the smaller, with “bad feet” was Covington's tortoise from Charles Island. As collectors and localities were provided whenever possible, it must be assumed that Gray did not recognize the tortoises as those presented by Darwin and must have also not unhinged the plastron to see the original registration number. Gray (1855, p. 6) repeated essentially the same information about these specimens: “f. Young (stuffed); 7 inches. Nuchal plate none” and “g. Young (stuffed); feet bad; 6 inches. Nuchal plate none”, confirming that they were stuffed mounts. Seventeen years later Gray (1872, p. 3), who finally acknowledged that Galápagos tortoises were distinct from other giant tortoises, referred “two young specimens [presumably Darwin and Covington's animals] and several shells of a black Tortoise in the British Museum without any nuchal plates” to Testudo elephantopus.

Günther (1874, 1875) likewise neglected to examine the interior plastron of Darwin's tortoise when he referred it to his new species, T. ephippium. Günther based his description on a 33-inch-long specimen in the collection of the Museum of Science and Arts, Edinburgh (now National Museums of Scotland [NMSZ] 1932.027.001, a mounted specimen “possibly previously registered in the Old College Collection as NMSZ 1822.058 ‘Large Turtle from South Sea’. Capt. Basil Hall.” fide Herman et al. [1990, p. 5]). Regarding the BMNH specimen he stated, “A very young stuffed example, 7 inches long, in the British Museum is referred to this species on account of its oblong shape and large smooth areolae” (Günther 1875, p. 81). The new registration number, 1874.6.1.6, was first mentioned in print 2 years later (Günther 1877). Boulenger (1889, p. 171) subsequently listed the specimen as the sole T. ephippium in the BMNH collection, “a. Yg. stffd. ________ ?” Following Gray's (1872) mention of a second juvenile Galápagos tortoise, there appear to be no further remarks about the small tortoise (6 inches) with bad feet, which was probably Covington's “pet” from Charles Island. It may be that its condition had deteriorated further and that, in the assumed absence of any data, it was discarded.

Although BMNH 1874.6.1.6 (formerly 1837.7.13.1) has been regarded as a syntype of T. ephippium by some authors (e.g., Herman et al. 1990) and is labeled as such today (Fig. 4), Günther's (1875, 1896) own reference to the Edinburgh animal as “the typical specimen” suggests that it is best regarded as a holotype, with Darwin's tortoise as a referred specimen without type status. Although he originally believed that T. ephippium was native to Charles Island, Günther (1877) subsequently considered Indefatigable (Santa Cruz) as its likely origin. Baur (1889) presented evidence that the type was from Abingdon (Pinta), but Günther (1896) found the best match to Duncan (Pinzon) specimens, and disputed Baur's (1889) synonymy of T. abingdonii with T. ephippium as did Rothschild (1896). Van Denburgh (1914) also accepted Günther's (1896) final conclusion that the species was from Duncan although Garman (1917) accepted Baur's (1889) conclusion and considered T. ephippium as a synonym of T. elephantopus Harlan 1827. Pritchard (1979, 1984, 1996) likewise concluded that the holotype was from Abingdon.

The nomenclatural history of the Galápagos tortoises is particularly complex and confused (Pritchard 1996; Zug 1997; Fritz and Havas 2007) and multiple systems of names are in current use, with different genera (Chelonoidis and Geochelone) being employed and the different island taxa being considered as either full species (e.g., Poulakakis et al. 2008; Turtle Taxonomy Working Group 2009) or subspecies (Caccone et al. 1999, 2004; Fritz and Havas 2007). Further, the uncertainty of the provenance of the holotype of Testudo ephippium leaves open the question of whether the name applies to the tortoises of Pinta or Pinzon and if to the former, if ephippium, or the younger name T. abingdonii Günther 1877 should be used (see Pritchard 1996; Fritz and Havas 2007). Controversy also surrounds the names (nigra or elephantopus) potentially applicable to the Charles Island form (Prichard 1996; Zug 1997). It is particularly unfortunate that BMNH 1837.8.13.2, Covington's tortoise, has apparently not survived. It has been estimated that the Charles Island form was extinct by 1840 or 1850 (Heller 1903; Broom 1929; Steadman 1986), although its genetic line is still extant in individuals translocated to Isabela (Poulakakis et al. 2008). Thankfully, such nomenclatural problems do not apply to Darwin's pet tortoise as the correct specific (or subspecific) epithet coined by Van Denburgh (1907) for the James Island population is unambiguously—and particularly appropriately—darwini.