Taxonomic questions often involve reference to antique documents, but
taxonomy is not a religion. It is a blend of scientific observation and
application of a consensual, evolving, and complex code of
nomenclatural procedures. The present task – i.e. defining the taxon
known as Testudo gigantea is an example of a purely taxonomic
and nomenclatural problem which may secondarily serve as a case history
for broader issues. In this case the issue was initiated by
non-taxonomists apparently unschooled in the rules of zoological
nomenclature and unwilling to abandon a name that they have become used
to. They were supported by reviewers who did not follow the series of
logical steps and preferred to appeal to fiat authority rather than
accept informed scholarship. Disagreements do of course exist among
scientists and scholars on many questions, some of which trivial some
important. Full, honest, respectful and public discussion of these
disagreements does not weaken science, indeed open debate contributes
to its progress. But in the case of entrenched major disagreements, it
is unacceptable that one group of individuals should claim to be the
unique depository of an orthodoxy or ‘revealed truth’ and demand
immunity from challenge. Such problems are not solved by lobbying or
conducting a public vote. The ultimate decision should have nothing to
do with the number of people sharing an opinion. The history of science
is replete with examples where a single person turned out to be correct
in the face of an overwhelming ‘majority’ of people who together
disagreed with a lone voice of truth. One reviewer of our paper was
correct in saying that, intellectually speaking, this subject is a
minefield. Like literal minefields, it is complete with brave soldiers
and cowardly ones. However, those responsible for clearing minefields
should not seek to cover them over with opaque material, but rather to
deal with the mines one by one so that they present no hazard to future
generations. In conflicted situations in science it is not acceptable
that important discussions proceed only behind closed doors, and in
such cases ‘private discussions’ among authors, referees and editors
are not a solution. These discussions should be aired publicly and
every biology colleague who wishes to contribute to the discussion
should be invited to comment on the case, provided that he/she respects
the persons with whom he/she disagrees and does not resort to personal
attacks or calumnies. Critics raised a very broad philosophical
question: do scientists, in this case taxonomists, have a ‘right to
commit error,’ change their minds, and admit past mistakes in taxonomic
and nomenclatural analyses? And is it possible or acceptable for the
perpetrators to correct these errors in the face of new evidence? Are
we bound to follow these errors ad infinitum once they have been
repeated by a few colleagues? If so, and pushing this situation to its
extreme, do we still need a Code of Zoological Nomenclature? Why not
just let ‘usage’, ‘consensus’ or ‘majority’ select the binominals and
trinominals that should stand? If ‘usage’ proves to be based on a wrong
interpretation of previous texts or type specimens, why not just
discard these texts and specimens to protect ‘usage’. Then, we no
longer need the inconvenient old texts, the old specimens, nor even the
museums that conserve them. Such an extreme interpretation may
certainly be much appreciated by those people within or outside our
governments who think that museums and their staff are very costly and
should be terminated. Type specimens would then only be useful when
they correspond to current nomenclatural ‘usage’, but could be
‘suppressed’ when they do not (as was the case in several recent
decisions of the ICZN). Of course, we fully agree that ‘usage’ should
be protected when it is really universal (but this is far from being
the case for Testudo gigantea), and above all when it
concerns not only taxonomists and nomenclaturists, but also laymen, the
mass media, general textbooks and so on; i.e., for names like Drosophila melanogaster, Tyrannosaurus rex or Homo erectus.
It is clearly acceptable to request that the ICZN invalidate nomina
oblita – overlooked, obscure or forgotten senior synonyms of later
names that have been in customary use for a long period. However it is
an entirely different matter when it is discovered that the holotype of
a species is a different taxon from that which it has been presumed to
be. It is especially disturbing when the objectors to correction of
such situations have apparently not troubled to read the details of the
original description, nor examined the actual holotype.
Testudo gigantea Schweigger, 1812 In a key paper describing about two dozen new species of chelonians, Schweigger described a new tortoise, which he named Testudo gigantea,
the Giant Tortoise (1812, pp. 327; 362). Schweigger (1812, p. 327)
added: ‘Vidi animal e collectione regi Li[s]bonensi proventum in museo
Parisiensi’ (‘I saw the animal from the King of Lisbon’s collection at
the Paris Museum’). To summarise the circumstances, Geoffroy
Saint-Hilaire had chosen 10 turtle specimens from the King of
Portugal’s collections in Lisbon to bring to the Paris Museum, where
they arrived in mid-November 1808. A tortoise ‘de plus de 0m 60’
(‘longer than 60 cm’) was present, as noted on a manuscript list by
Lacepède (Daget & Saldanha, 1989, p. 139). It was the type and only
specimen of the new species described by Schweigger (Bour, 2006b). One
very important point, Schweigger stated precisely and unequivocally
‘Habitat in Brasilia’ (‘Inhabits Brazil’). Pritchard (1986), initially
impressed by the stated origin of the Schweigger type specimen
(‘Brasil’), was convinced that Testudo gigantea Schweigger was a synonym of Testudo denticulata Linnaeus, 1766 (today Chelonoidis denticulata),
the (sometimes giant) yellow-footed tortoise of South America which can
reach a straight length of 82 cm (Mittermeier in Pritchard &
Trebbau, 1984, p. 225). This is a somewhat elongate, narrow-shelled
species whose proportions are fully in accord with those indicated by
Schweigger in his description. Bour located the type a few years ago,
and when later he re-read Schweigger’s original text, it became obvious
that it was the 170 very specimen described by this talented student of
Constant Duméril. The description and the given measurements leave no
doubt about its identity. It must be remembered that the specimen
(registered as MNHN 9554) was also described and measured by Duméril
& Bibron themselves (1835, p. 89), without any details about its
origin, under the heading Testudo tabulata Walbaum, 1782, a junior and invalid subjective synonym of Testudo denticulata.
Its presence in the MNHN collections was further confirmed in a
hand-written catalogue dated ca. 1864, with ‘Brésil’ (‘Brazil’) as
locality (registration number: 120). Pritchard’s hypothesis was
Testudo gigantea as interpreted by Duméril & Bibron (1835)
In 1835, Duméril and Bibron associated Schweigger’s description with
another unique specimen, which was obviously distinct. Duméril &
Bibron actually described a new species, but mistakenly attributed it
to Schweigger. Testudo gigantea sensu Duméril & Bibron (1835, p. 120) has the following features which did not fit with Testudo gigantea Schweigger,
1812: ‘carapace bombée; écailles du disque très convexes; une écaille
nuchale; suscaudale double’ (‘shell bulged; scutes of the disc very
convex; one nuchal [cervical] scute; supracaudal scute double
[divided]’). Other details also distinguish this specimen from that
described by Schweigger, including the mention of the broad and rough
scales of the forelimbs, and the great size: according to Duméril &
Bibron, their ‘new’ Testudo gigantea had a shell length (over
the curve) of 130 cm and a depth of 49 cm, versus 75.6 cm and 24.3 cm,
respectively for the ‘old’ or Schweigger’s specimen. The specimen is
still preserved in the Paris Museum collections, with registration MNHN
9566, and it is an Aldabra tortoise.
Testudo elephantina Duméril & Bibron, 1835
The description of Testudo elephantina by
Duméril & Bibron (1835, p. 110) was based on about eight specimens,
from ‘Anjouan, Aldebra [sic], les Comores’ and Bour (1984a, p. 291),
following Rothschild (1915, p. 425), designated as lectotype a large
stuffed male (MNHN 7874) on which the description was mostly based. The
origin of the species was limited to Aldabra by Günther (1877, p. 18),
and the type locality restricted to ‘Malabar, Aldabra’ by Bour (1984a,
p. 291). As outlined by Duméril and Bibron themselves, Testudo gigantea was very close to their new species Testudo elephantina.
This point of view was later shared, for instance, by Günther (1877, p.
22, note) and Boulenger (1889, p. 168). Finally, Rothschild (1897, p.
407), and then Siebenrock (1909, p. 529) combined both nominal species,
Testudo elephantina being considered as a subspecies of Testudo gigantea, but of Testudo gigantea sensu
Duméril & Bibron! Nevertheless, from the beginning of the 20th
century, the valid name for the Aldabra tortoise seemed to have been
definitely settled, or at least most often reduced to the nominal
species Testudo gigantea, and later Geochelone gigantea, with Schweigger as author. Both have been widely used up to the present, although Testudo elephantina was
also regularly used, either as specific or subspecific name, for those
who recognised more than one taxon among Seychelles tortoises. Surely
one thing we can all agree upon (assuming that we have all read
Schweigger’s work) is that the type locality of Testudo gigantea is
identified as Brazil by the original describer. There are no data to
contest this. Whether or not one accepts that MNHN 9554 is indeed
Schweigger’s type does not change this type locality; those who insist
upon designating a neotype for the species would be bound to select a
specimen of a Brazilian tortoise species, of which there are only two.
One of them (Chelonoidis carbonaria) does not reach the size of Schweigger’s specimen. That leaves only Geochelone denticulata, of which MNHM 9554 is a conveniently available example.
A neotype for Testudo gigantea?
Frazier (2006) strongly emphasized the ‘general instability and chaos
regarding the valid name of the Aldabra Tortoise’. He favoured an
‘established nomenclatural system’ (i.e. Testudo gigantea according
to him; although ‘general instability and chaos’, and ‘established
nomenclatural system’ are rather subjective), which is a commonly
proposed argument – and the only one – against the use of Testudo dussumieri Gray, 1831 or Testudo elephantina Duméril
& Bibron, 1835, the types of which are clearly Aldabra tortoises.
Frazier believed that a neotype designation could clarify this
situation, and selected a specimen for this purpose, actually more for
nomenclatural than taxonomic reasons. We wish to stress two points
stated in the Code (1999) which were not taken into account by Frazier.
Recommendation 75B, that ‘before designating a neotype, an author
should be satisfied that the proposed designation does not arouse
serious objection from other specialists in the group in question’, was
not fulfilled. Furthermore, contrary to the wording of Article 75.3.5,
Frazier’s neotype is not ‘consistent with what is known of the former
name-bearing type from the original description’ (e.g. absence vs.
presence of a cervical scute; limbs shielded by tough and very broad
scales vs. only postcranial skeleton, and fragments of skin; from
Brazil vs. from Aldabra). Fortunately, the rediscovery of the holotype
removes any value from the neotype as the type specimen of the same
taxon, so we set aside the neotype according to Article 75.8 of the
Code; Frazier’s action thus becomes void. Although aware of the results
published by Bour (2006b) regarding the identity of Testudo gigantea, Frazier recently (BZN 66:
34–50) decided to request the ICZN to conserve the usage of this name
for the Aldabra tortoise under the plenary power. For the reasons given
above, we do not support this application. True, some of the early
writers in the field of chelonian systematics were frustrating in the
vagueness of their descriptions of new taxa. But Schweigger was not one
of these. He was a brilliant and meticulous man, and no arguments have
been presented to suggest that his description of Testudo gigantea was
faulty in any fundamental way. Schweigger wrote that his new species
was based upon a large tortoise from Brazil in the King of Portugal’s
collection in Lisbon, and since the monarch in question had spent the
early years of the 19th century (1808–1820) in Rio de Janeiro, where he
had received numerous biological specimens collected by Alexandre
Rodrigues Ferreira and party (Wilcken, 2004), this is a fully plausible
type locality. On the other hand, to assume that Dom João received an
Aldabra tortoise (now completely lost) during his stay in Brazil,
without a shred of evidence, circumstantial or otherwise, to back this
up, is an argument that should be disposed of with a slash from Occam’s
Aldabrachelys and Dipsochelys
The remaining question is the choice of the generic name for the Aldabra tortoises and their relatives. Aldabrachelys, as a subgenus of Geochelone Fitzinger, 1835, was coined by Loveridge and Williams to include the Aldabra Tortoise and related species, with Testudo gigantea Schweigger
as type species by original designation (Loveridge &Williams, 1957,
p. 225). In fact the discordance between the intended and the actual
type specimen of Testudo gigantea was not noticed until 1982, when Bour erected a new genus, Dipsochelys, with Testudo elephantina as
type species by original designation, to include the Aldabra tortoise
and related species. Bour (1984a, p. 281) was apparently the first to
resurrect the nominal species T. dussumieri and to recognise
its availability, adding ‘Perhaps provisionally, we will consider this
name as a ‘nomen oblitum’.’ On the other hand, Gerlach & Canning
(1995, p. 133) were certainly the first to coin and use the combination
Dipsochelys dussumieri; the main justification given being to avoid ‘confusion with the phenotypically similar Galápagos complex of Chelonoidis elephantopus (Harlan, 1827)’, which has since been renamed Chelonoidis nigra (Quoy & Gaimard, 1824). Since then, as outlined by Frazier (2006; 2008), Gerlach regularly used Dipsochelys dussumieri for the Aldabra tortoise. Both Aldabrachelys and Dipsochelys could
be considered as being valid candidate names for the Aldabra tortoise.
The latter name is nowadays widely used (e.g. Grzimek, 2003; Bonin et
al., 2006; Roberts, 2007; Cheke & Hume, 2008; Pedrono, 2008;
Vetter, 2008; Wyneken et al., 2008), either as D. elephantina or as D. dussumieri, and we see no reason not to name the Aldabra tortoise Dipsochelys dussumieri. Aldabrachelys was rarely used until recently; a claim to use Aldabrachelys rather than Dipsochelys cannot be made on the grounds of stability. Therefore, the genus Aldabrachelys, with Testudo gigantea as type species, is a junior subjective synonym of Chelonoidis Fitzinger, 1835, which has Testudo boiei Wagler, 1833 (a junior subjective synonym of Testudo carbonaria Spix, 1824) as type species by subsequent designation of Fitzinger, 1843. Dipsochelys is the valid genus name for the Aldabra tortoise and its relatives.
Anonymous. 2008. The life and herpetological contributions of August Friedrich Schweigger. SSAR, Villanova, Pennsylvania.
Bonin, F., Devaux, B. & Dupré, A. 2006. Turtles of the World. 416 pp. A & C Black, London.
Bour, R. 2006a. An unnamed tortoise from the Seychelles Islands. Emys, 13(3): 24–30.
Bour, R. 2006b. Identity of Testudo gigantea Schweigger, 1812. Rediscovery of the type specimen. Emys, 13(4): 12–23.
Cheke, A.S. & Hume J.P. 2008. Lost land of the Dodo: an ecological history of Mauritius, Réunion and Rodrigues. 464 pp. A&C Black, London & Yale University Press. New Haven, Connecticut.
Daget, J. & Saldanha, L. 1989. Histoires naturelles Franco-Portugaises du XIXe siècle. INIP, Lisboa.
Grzimek, B. (Ed.) 2003. Grzimek’s animal life encyclopedia: lower metazoans and lesser deuterostomes. Gale, Farmington Hills, Michigan.
Pedrono, M. 2008. The tortoises and turtles of Madagascar. 147 pp. Natural History Publications, Borneo, Kota Kinabula.
Prichard, P.C.H. 1996. The Galápagos tortoises: nomenclatural and survival status. Chelonian Research Monographs, 1: 1–85.
Pritchard, P.C.H. & Trebbau, P. 1984. The turtles of Venezuela. SSAR, Athens, Ohio.
Roberts, C. 2007. The unnatural history of the sea. Island Press, Washington DC.
Schweigger, A.F. 1812. Prodromus Monographiae Cheloniorum. Königsberger Archiv für Naturwissenschaft und Mathematik, 1, 271–368, 406–468. [facsimile reprint in Anonymous 2008].
Siebenrock, F. 1909. Synopsis der rezenten Schildkröten mit Berücksichtigung der in historischer Zeit ausgestorbenen Arten. Zoologischer Jahrbuch, Supplement 10: 427–618.
Vetter, H. 2002. Terralog 1. Turtles of the world, Vol. 1, Europe, Africa and Western Asia/Schildkröten der Welt 1 Europa, Afrika und Westasien. 96 pp. Chimaira, Frankfurt am Main.
Wilcken, P. 2004. Empire Adrift. The Portuguese Court in Rio de Janeiro. 306 pp. Bloomsbury Publications, London. Wyneken, J., Godfrey, M.H. & Bels, V. 2008. Biology of turtles: from structures to strategies of life. CRC Press, Boca Raton, Florida.