Field of Science

Tortoise Sorting

Indian star tortoise Geochelone elegans, copyright P. G. Palmer.

As a whole group, the 'true' tortoises of the family Testudinidae are easily recognised, with a usually terrestrial habitus (though at least one species, the serrated hinge-back tortoise Kinixys erosa, is a capable swimmer), columnar legs with short heavy-clawed feet, and a relatively high carapace. But relationships within the tortoises have seen a bit of shuffling around in recent years, and nowhere has that shuffling been more obvious than in the genus Geochelone.

Many of you may know Geochelone as the genus including the giant tortoises of the Galapagos Islands and islands in the Indian Ocean, as well as species such as the radiated tortoise G. radiata of Madagascar and the geometric tortoise G. geometrica of South Africa. At its broadest, about half the world's tortoise species have been included in Geochelone. However, the genus has always been poorly defined, marked by its 'primitive' skull (Gerlach 2001), and the lack of features of other tortoise genera such as the plastral hinge of the Palaearctic Testudo species, or the rear-carapace hinge of African Kinixys species. For the most part, Geochelone was simply a home for the bigger tortoises.

Burmese star tortoise Geochelone platynota, copyright Kalyar Platt.

So it is hardly surprising that phylogenetic studies, whether morphological (Gerlach 2001) or molecular (Fritz & Bininda-Emonds 2007), have failed to support a broad Geochelone as a monophyletic group. As a result, recent authors have advocated recognising a number of separate genera: the Galapagos tortoises belong to the genus Chelonoidis, the Seychelles giant tortoises to Aldabrachelys (as confirmed by ICZN 2013), the radiated tortoise to Astrochelys and the geometric tortose to Psammobates. From being the largest recognised tortoise genus, Geochelone has been cut down to a mere two or three species. These are the Indian star tortoise Geochelone elegans, the Burmese star tortoise G. platynota, and, maybe, the African spurred tortoise G. sulcata.

The Indian and Burmese star tortoises are two very similar species that get their names from their colour pattern, with radiating star-like markings on the carapace. The individual scutes of the carapace bulge outwards, giving the animal an overall lumpy appearance. They are both medium-sized tortoises, growing to over 30 cm in length. The Indian star tortoise Geochelone elegans is widespread in dry habitats in India, Pakistan and Sri Lanka, though conservation concerns have been raised about the extent of harvesting of wild tortoises for food and the exotic pet trade. The Burmese star tortoise G. platynota, on the other hand, is critically endangered, having been almost wiped out from its original range in the central dry zone of Burma. As long ago as 1863, Edward Blyth (who, offhand, sported one heck of an impressive beard) was complaining that specimens were difficult to find due to the local people's fondness for eating them (Platt et al. 2011).

African spurred tortoise Geochelone sulcata, copyright Chris Mattison.

The African spurred tortoise Geochelone sulcata is found across the southern part of the Sahara Desert and the Sahel. It is a particularly large tortoise, growing to over 80 cm (Swingland & Klemens 1989); in fact, it is the largest tortoise that is not found on oceanic islands. Spurred tortoises dig burrows that can reach up to 3.5 m in length in which to avoid the full heat of the day. This species is placed by some authors in its own genus Centrochelys, cutting Geochelone down to just the two Asian star tortoises. Molecular studies place G. sulcata as the sister taxon to the Asian species (Fritz & Bininda-Emonds 2007), but they have not been associated in morphological studies. Though widespread, the African spurred tortoise is regarded as vulnerable due to the degradation of its habitat. Concern has also been raised, again, about the collection of wild individuals for the pet trade.


Fritz, U., & O. R. P. Bininda-Emonds. 2007. When genes meet nomenclature: tortoise phylogeny and the shifting generic concepts of Testudo and Geochelone. Zoology 110: 298-307.

Gerlach, J. 2001. Tortoise phylogeny and the ‘Geochelone’ problem. Phelsuma 9 (Supplement A): 1-24.

ICZN. 2013. Opinion 2316: Testudo gigantea Schweigger, 1812 (currently Geochelone (Aldabrachelys) gigantea; Reptilia, Testudines): usage of the specific name conserved by maintenance of a designated neotype, and suppression of Testudo dussumieri Gray, 1831 (currently Dipsochelys dussumieri). Bulletin of Zoological Nomenclature 70 (1): 61-65.

Platt, S. G., T. Swe, W. K. Ko, K. Platt, K. M. Myo, T. R. Rainwater & D. Emmett. 2011. Geochelone platynota (Blyth 1863)—Burmese star tortoise, kye leik. Chelonian Research Monographs 5: 057.1-057.9.

Swingland, I. R., & M. W. Klemens (eds). 1989. The conservation biology of tortoises. Occasional Papers of the IUCN Species Survival Commission 5.

The Sweetest of Lips

Oblique-banded sweetlips Plectorhinchus lineatus, copyright Richard Ling.

In an earlier post on this site, I referred to a fish of the family Lethrinidae being known by the name of "sweetlips". However, as is usually the way with fish vernacular names, there is more than one family of fishes to which this name can be applied. 'Sweetlips' is also the vernacular name for fishes in the Plectorhinchinae.

The Plectorhinchinae is most commonly treated as a subfamily within the family Haemulidae, the grunts (some sources will place 'Plectorhynchidae' as a separate family, and no, that wasn't a typo: read on). Plectorhinchines are distinguished from other subfamily of haemulids, the Haemulinae, by characters including a longer dorsal fin and the presence of at least four prominent lateral line pores under the chin (Johnson 1980). The name 'sweetlips' refers to the prominent lips of mature individuals of two of the genera of plectorhinchines, Plectorhinchus and Diagramma, which are often a distinct colour from the rest of the head. Members of the third genus, Parapristipoma, have the lips not quite so prominent, and are commonly referred to as 'grunts' like the remaining haemulids.

African striped grunts Parapristipoma octolineatum, copyright Juan Cuetos.

The plectorhinchines are found around tropical reefs in the Indo-West Pacific and East Atlantic, with a single species, the rubberlip grunt Plectorhinchus mediterraneus, being found in the in the Mediterranean and Black Seas. No plectorhinchines are found on either side of the Americas. They are nocturnal predators of benthic invertebrates, emerging at night from the secluded crevices and overhangs where they spend the day. Most are medium-sized fish, though the painted sweetlips Diagramma pictum can get up to 90 cm. They are popular with fishers; Smith (1962) referred to them as "among the best if not the best eating fishes of the reef-haunting species". Many species can go through significant changes in coloration as they mature: spotted juveniles may become unicoloured adults, or blotchy babies may mature into stripes. The differences are great enough that juveniles and adults have often been mistaken for separate species.

Juvenile oriental sweetlips Plectorhinchus vittatus, copyright Jan Messersmith. The adult form of this species resembles the oblique-banded sweetlips in the top photo on this post.

But failure to associate parents with their children is not the only way in which this group has been dogged by confusing taxonomy. The name of the type genus has been variously spelled Plectorhinchus or Plectorhynchus, with the family name varying accordingly (it seems that 'Plectorhinchus' is the correct spelling). A surprising number of sources (e.g. Tavera et al. 2012) seem to have it both ways, with the genus being called Plectorhinchus but the higher taxon being called Plectorhynchinae (R. van der Laan et al. confirm the correct family-name spelling). Meanwhile, Smith (1962) argued for the use of the name Gaterin in place of Plectorhinchus, and called the family Gaterinidae. And if you have any interest in the vagaries of taxonomy, settle in: this is going to be a whole thing.

The name 'Gaterin' dates from what is usually known as Forsskål's (1775) Descriptiones animalium, which Fricke (2008) argued should be attributed to Niebuhr (see, right from the first sentence it's confusing). Peter Simon Forsskål and Carsten Niebuhr were members of a Danish scientific expedition in 1761 to 1763 to the Red Sea (though Forsskål himself was Swedish, but that's another story). Forsskål was the expedition's naturalist, while Niebuhr was there as a geographer. The history of the expedition, and of the composition of Descriptiones animalium, has been summarised by Fricke (2008). The expedition was particularly ill-fated; of six original members, Niebuhr was the only one to make it back to Denmark alive. After returning to Denmark, Niebuhr started preparing Forsskål's notes for publication. However, he found this no easy task. Forsskål had not prepared a single manuscript, but made notes on various scraps of paper; in the end, Niebuhr suspected that many of these scraps had gone missing. As an engineer, Niebuhr knew little Latin and even less biology, so he obtained the services of an academic adviser. The identity of this adviser was not divulged in the final publication by Niebuhr himself, but he has since been identified as the Danish naturalist Johann Christian Fabricius. The relationship between Niebuhr and Fabricius was not entirely positive (Niebuhr later stated that his adviser on Descriptiones animalium had been a 'strange fellow'), and Fabricius does not seem to have spent any more time on the Forsskål notes than he absolutely had to. As a result, the final publication that emerged was partly Forsskål, partly Niebuhr, partly Fabricius, and all dog's breakfast.

The name 'Gaterin' is listed by Forsskål/Niebuhr/Fabricius as one of the sub-divisions of the genus Sciaena, and Smith's (1962) revival of the name was based on the assumption that Forsskål intended these subdivisions to represent what we would now call subgenera. As such, Gaterin published in 1775 would clearly be an earlier name than Plectorhinchus published in 1802. Smith further supported this interpretation by pointing out that two names listed by 'Forsskål' as subdivisions of Chaetodon, Acanthurus and Abudefduf, had since been widely accepted as names for separate fish genera. There were no grounds, he claimed, for taking Abudefduf as valid but refusing Gaterin.

As it happens, Forsskål probably never intended either Gaterin or Abudefduf to represent generic names of any kind. It seems that his notes had used local Arabic names to refer to taxa to which he had not yet supplied formal Latin names. When Fabricius compiled these notes, he simply used the Arabic names as formal names, probably because he just didn't care. When 'Forsskål' referred to 'Gaterin' in his introductory paragraph for Sciaena, he was probably referring to the individual species known in Arabia as gaterin rather than any formal group. 'Abudefduf' may have been similarly inadvertent, but long usage as a genus name means that it should probably be retained whatever its original status. No such argument can be marshalled in favour of 'Gaterin', whose usage in place of Plectorhinchus has been minimal.

And I can think of no better response to all that than the expression of this painted sweetlips Diagramma pictum. Copyright John Natoli.


Fricke, R. 2008. Authorship, availability and validity of fish names described by Peter (Pehr) Simon Forsskål and Johann Christian Fabricius in the ‘Descriptiones animalium’ by Carsten Niebuhr in 1775 (Pisces). Stuttgarter Beiträge zur Naturkunde A, Neue Serie 1: 1–76.

Johnson, G. D. 1980. The limits and relationships of the Lutjanidae and associated families. Bulletin of the Scripps Institution of Oceanography 24: 1–114.

Smith, J. L. B. 1962. Fishes of the family Gaterinidae of the western Indian Ocean and the Red Sea with a resume of all known Indo Pacific species. Ichthyological Bulletin 25: 469-502.

Tavera, J. J., A. Acero P., E. B. Balart & G. Bernardi. 2012. Molecular phylogeny of grunts (Teleostei, Haemulidae), with an emphasis on the ecology, evolution, and speciation history of New World species. BMC Evolutionary Biology 12: 57.

The Eater of Light

The Waitomo harvestman Forsteropsalis photophaga, from Taylor & Probert (2014).

A little less than a year ago, I was contacted by a student at the University of Auckland in New Zealand, asking me about some harvestmen that she'd been trying to identify me from the Waitomo cave system. This incited a certain degree of excitement on my part, because I was not entirely unfamiliar with Waitomo's harvestmen. I had first seen specimens from there while doing my MSc back in 2001 or 2002, and had realised then that they represented an undescribed species. However, for various reasons, I had not yet published a description of the species in question. So when Anna contacted me, I decided it was time to bump the Waitomo harvestmen up the to-do list, and I replied to her asking if she would be interested in collaborating on a paper on the Waitomo harvestmen. She agreed, and the resulting paper came out just last week: C. K. Taylor & A. Probert, "Two new species of harvestmen (Opiliones, Eupnoi, Neopilionidae) from Waitomo, New Zealand".

Image of Waitomo cave, from here.

The Waitomo caves may be the world's only tourist attraction centred around an infestation of flies. The caves are home to an abundant population of glow-worms, larvae of the fungus gnat Arachnocampa luminosa. These fly larvae live on the roof of the cave, held in place by a silken hammock, and produce spots of brilliant blue light. It is the spectacle of these lights that draw the tourists, but for the glow-worms they serve a different purpose: the lights attract insects flying in the cave. In flying towards the light, insects become entangled in sticky threads that each glow-worm suspends below its hammock, providing the glow-worm with food. You can see a video of the process here, taken from the BBC's Planet Earth series.

But the glow-worms are not without predators of their own. With their long slender legs, harvestmen are able to carefully tip-toe between the sticky threads and pluck out the glow-worms, as from a luminous buffet (they also eat them at the pupal and adult stages). The harvestmen of Waitomo were studied by Myer-Rochow & Liddle (1988), who identified two species. One, a 'short-legged harvestman' Hendea myersi cavernicola (which actually has decidedly long legs, natch), is endemic to the cave system and was identified by Meyer-Rochow and Liddle as a strict troglobite (i.e. it spends its entire life within the cave). It has a number of features commonly associated with cave-dwelling, such as pale coloration and lengthened legs. It does differ from most troglobites in that it is not blind: while its eyesight is dim, it does retain enough to find its glowing prey.

Meyer-Rochow and Liddle also identified a 'long-legged harvestman' in the Waitomo caves, which they referred to as 'Megalopsalis tumida'. This name refers to a species first described from Wellington, quite some distance to the south, that now goes by the name of Forsteropsalis fabulosa (and has made an earlier appearance at this site, where it was the subject of this post). As it turns out, I identified two species of Forsteropsalis in material from the caves, neither of which was F. fabulosa. I can't be certain which was the species being looked at by Meyer-Rochow and Liddle. For various reasons, I suspect that they may have been looking at examples of both, but, as I've never been able to locate any vouchers for their study, I can't really say (remember, kids, vouchers are important).

Forsteropsalis bona, from Taylor & Probert (2014).

One of these species is indeed very similar to Forsteropsalis fabulosa, and has accordingly been labelled Forsteropsalis bona. Indeed, the two species are similar enough that I can now see that the photograph I used in my earlier post to illustrate F. fabulosa in fact shows an individual of F. bona. The primary difference between the two is in their pedipalps: in F. fabulosa, the patella of the pedipalp has a distinct finger-like process that is much reduced in F. bona. Forsteropsalis bona is not a strict troglobite: specimens have been collected at Waitomo both inside and outside the cave entrance. Instead, it is what is called a troglophile: individuals of F. bona probably use the caves as a cool, damp place to hang out during the day, emerging to forage outside the cave at night. This is the same pattern of behaviour found in New Zealand's cave wetas.

The second species is the beauty pictured at the top of this post. Its species name, photophaga, means 'eater of light', referring of course to its probable predation on glow-worms. This is a stunning animal: the enormous chelicerae typical of New Zealand Neopilionidae are rendered even more eye-catching by the presence of rows of longer spines (offhand, we don't yet know what the females of either of the Waitomo species look like, but they probably resemble other Forsteropsalis females in lacking the long chelicerae of the males). Whether Forsteropsalis photophaga is a troglobite or a troglophile is a bit more uncertain. I'm not aware of it having ever been collected outside the caves, but it doesn't seem to have the obvious modifications for cave-dwelling of Hendea myersi cavernicola (though when t comes to assessing elongated limbs in what is already a long-legged harvestman... how are you going to tell?). At present, I'm guessing troglophile rather than troglobite, but future studies may easily prove me wrong.

Forsteropsalis photophaga is also an intriguing animal from a taxonomic viewpoint. In the past, the two New Zealand harvestman genera Pantopsalis and Forsteropsalis have been pretty easy to distinguish, but F. photophaga has some features that are more reminiscent of Pantopsalis than of Forsteropsalis. Recently, other things have been brought to my attention that suggest that, while Pantopsalis as we currently know it still seems fairly robust, Forsteropsalis is beginning to look decidedly fuzzy around the edges. The relationship between these two genera (if, indeed, they should still be recognised as two separate genera) has still not been resolutely ironed out.


Meyer-Rochow, V. B., & A. R. Liddle. 1988. Structure and function of the eyes of two species of opilionid from New Zealand glow-worm caves (Megalopsalis tumida: Palpatores, and Hendea myersi cavernicola: Laniatores). Proceedings of the Royal Society of London Series B (Biological Sciences) 233: 293–319.

The Velutinidae: Sea Not-Quite-Slugs

Velutinid, probably Lamellaria. Copyright Bill Rudman.

In previous posts on this site, I have introduced you to some of the incredible animals that shelter under the misleadingly unprepossessing name of 'sea slugs'. However, marine mollusc diversity being what it is, it should probably come as no surprise that not all that squashes is slug-y.

The Velutinidae (sometimes referred to in older references as Lamellariidae) are a group of small gastropods that often look rather like sea slugs on the outside, but are misleading it that they actually do have shells. In one subfamily, the Lamellariinae, the mantle has expansive lobes that wrap up around the shell, covering it over. Velutinids are far from being the only molluscs that do this: the glossiness of cowrie shells, for instance, is because live cowries have the shell protected from the elements in this way. Lamellariines differ from cowries, however, in that the mantle lobes are more or less fused to each other and cannot be retracted back. In the other subfamily of Velutinidae, the Velutininae, the shell is not entirely covered by the mantle, as can be seen in the photo below. In both subfamilies, the shell makes it through one or two small loops before broadening out into an abalone-like shape. Lamellariines also differ from velutinines in lacking the marginal teeth of the radula (Beesley et al. 1998).

Velutina prolongata, copyright Dave Cowles.

The velutinids are all specialised predators of ascidians (sea squirts), which they tear into with hard chitinous jaws contained in the buccal mass. The mantle of lamellariines is often coloured to look like the sea squirt they are feeding on, which can make them very difficult to see: to a casual observer, they're just one more squashy blob amongst a whole bunch of squashy blobs. The unfortunate sea squirts are used as nurseries as well as dinner: the female velutinid inserts each egg capsule into a hole that she bites into the sea squirt, with only a narrow neck protruding from its skin through which the velutinid larva hatches.

Velutinid identified as Coriocella nigra, but looking a bit different from other photos online supposed to be this species, from here.

The velutinids are not close relatives of the classic sea slugs, but closer to gastropods such as cowries and periwinkles. They are regarded by most authors as closest to the Triviidae, a group of cowrie-like gastropods that resemble velutinids in their expanded mantle lobes and preferred diet of ascidians. Velutinids and triviids also resemble each other in having an unusual type of larva called an echinospira. Echinospira larvae appear to have two shells, with a mineralised inner shell that is quite separate from a transparent, glassy outer shell. In the Velutinidae, these two shells even coil differently: the outer shell is planispiral, but the inner shell is helical. When the larva metamorphoses, the outer shell is lost. As a result, earlier authors believed that the inner shell corresponded to the true adult shell, while the outer shell was a novelty unique to echinospirae. In recent times, however, a more popular interpretation is that the two shells each correspond to the inner calcareous layer and the outer periostracum of more usual shells, with their growth having become decoupled. The only other gastropods known to possess an echinospira larva are members of the family Capulidae; whether this larval type indicates that all three families form a single clade remains uncertain.

Echinospira larva of Lamellaria perspicua, from Lebour (1935).


Beesley, P. L., G. J. B. Ross & A. Wells (eds) 1998. Fauna of Australia vol. 5. Mollusca: The Southern Synthesis, pt B. CSIRO Publishing: Melbourne.

Lebour, M. V. 1935. The echinospira larvae (Mollusca) of Plymouth. Proceedings of the Zoological Society of London 105 (1): 163-174, pls 1-6.

Bobble-Nosed Trilobites

Cranidium and part of thorax of Onchonotellus sp., from Bao & Jago (2000).

While most fossil invertebrates manage to completely fly under the radar when it comes to popular culture (it's not as if we're drowning under cartoon depictions of euthycarcinoids or eldoniids), one group that will often get a passing nod is the trilobites. Any depiction of early animal life worth its salt is going to feature a couple of these crunchy bugs scurrying about. Nevertheless, the range of varieties of trilobite shown will generally be low, and will usually be something similar to Olenellus or Elrathia. Seeing as trilobites persisted for hundreds of millions of years, it should be no surprise that their actual diversity was much higher.

The fossil shown at the top of this post is a representative of the trilobite genus Onchonotellus. Remains have been assigned to this genus from the late Cambrian and the early Ordovician, though Adrain (2013) expressed some reserve about the genus' monophyly. It has been assigned to the Catillicephalidae, a mostly Cambrian group of trilobites, but again the coherence of this total group is uncertain.

Most known fossils of Onchonotellus are represented by isolated cranidia, the plates that in life covered the trilobites' head. Onchonotellus and other catillicephalids are characterised by an inflation of the glabella, the middle lobe of the cranidium. Generally, the glabella of Onchonotellus is barrel-shaped. In some Onchonotellus specimens, the glabella may almost look spherical in side-view, making this trilobite look like a Bubble O'Bill. In many trilobites, the glabella will bear a series of furrows along the sides, but in Onchonotellus these disappear so that the glabella surface is smooth. In other catillicephalids, the glabella extends right to the front margin of the head, but Onchonotellus does retain a distinct rim around its front. The cheeks on either side of the glabella are relatively broad (Öpik 1967; Shergold 1980). In their time, members of the genus were found around the world, and some species have been highlighted as index fossils, useful in determining the age of rock strata.

So what was the significance of the large glabella? Most researchers have suggested that it probably held some sort of expansion of the digestive system, such as a crop for storing food. Some trilobites in which the glabella became large enough that it actually overhung the front margin have been suggested to be predatory (Fortey & Owens 1999), with the glabella containing a large oesophagus that allowed the trilobite to swallow larger food items. Onchonotellus probably didn't take things that far: not only was its glabella just that little bit smaller, but I haven't found any indication of it possessing the enlarged eyes also found in the predatory forms. The furrows on the glabella of most trilobites may have marked the attachment positions for muscles associated with the oesophagus/crop/whatever, so did the reduction of these furrows in Onchonotellus indicate a correspondingly less muscular pharynx? Perhaps Onchonotellus was a detritus feeder, with an expanded crop allowing it to take in mouthfuls of sediment from which to sieve out tasty organic morsels. Or perhaps it was a scavenger, breaking off lumps from the carcasses of other animals. Whatever it was doing, it was something that involved a big nose that was in actuality a big mouth.


Adrain, J. M. 2013. A synopsis of Ordovician trilobite distribution and diversity. Geological Society, London, Memoirs 38: 297-336.

Bao, J.-S., & J. B. Jago. 2000. Late Late Cambrian trilobites from near Birch Inlet, south-western Tasmania. Palaeontology 43 (5): 881-917.

Fortey, R. A., & R. M. Owens. 1999. Feeding habits in trilobites. Palaeontology 42 (3): 429-465.

Öpik, A. A. 1967. The Mindyallan fauna of northwestern Queensland. Commonwealth of Australia, Department of National Development, Bureau of Mineral Resources, Geology and Geophysics—Bulletin 74, vol. 1: 404 pp., vol. 2: 166 pp., 67 pls.

Shergold, J. H. 1980. Late Cambrian trilobites from the Chatsworth Limestone, western Queensland. Bureau of Mineral Resources, Geology and Geophysics—Bulletin 186: 1-111.