Field of Science


It's time to meet the sweepers.

Smallscale bullseyes Pempheris compressa, copyright John Turnbull.

Sweepers, Pempheridae, are a group of moderately sized marine fish (usually about fifteen to twenty centimetres in length) found around tropical reefs in the Indo-Pacific and western Atlantic. I don't know why they're called sweepers, but in some areas they may be among the most abundant fish on the reef. Distinctive features of the group include a short, high dorsal fin and a long anal fin. The lateral line is also distinctively long, extending past the end of the tail right onto the caudal fin. Perhaps the feature that most stands out about sweepers is their large eyes. The eyes are so big because sweepers are nocturnal; during the day they retreat into protected crevices and caves, emerging at night to feed on minute crustaceans and other small animals (Mooi 2001).

Pygmy sweeper Parapriacanthus ransonneti, from here.

Sweepers are divided between two quite distinct genera. Members of the genus Parapriacanthus have a more 'average fish-like' elongate profile with the body less deep than the head is long. The other genus, Pempheris, has a distinctively deep profile, deeper than the head is long. The exact number of species of pempherid appears to still be uncertain. Pempherids lack the striking markings of other tropical fish and species can appear very similar to each other. What is more, they have two layers of scales on the body, with the outer scales being larger than the inner and deciduous (easily shed), and loss of the outer scales has the potential to change an individual's superficial appearance. Early descriptions of pempherid species are often inadequate for their reliable identification, and new species continue to be described at a quite rapid pace. A recent publication by Randall & Victor (2015), for instance, described no less than thirty-four new species of Pempheris from various locations in the Indian Ocean, close to doubling the number of species in the genus at a stroke. The genus Parapriacanthus is much less diverse, with only about five recognised species.

Orange-striped bullseyes Pempheris ornata in hiding during the day, copyright Peter Southwood.

Because of their relatively small size and retiring habits, sweepers are mostly not that significant economically. At least one species, Pempheris xanthoptera, is fished off the coast of Japan and mostly eaten as fish paste; it is supposed to be quite tasty. Some have appeared in aquaria.

When foraging at night, sweepers communicate with each other by producing popping noises through muscular flexing of the swim bladder wall. Noise production increases in the presence of potential threats, perhaps to warn other members of the school. At least some pempherid species also have bioluminescent glands associated with the posterior part of the gut. The bioluminescent compound is not directly produced by the fish itself but obtained by consuming bioluminescent ostracods. I haven't found whether the function of this bioluminescence is specifically known for pempherids, but similar ventral glows in other fish provide camouflage by breaking up the fish's silhouette when seen from below.


Mooi, R. D. 2001. Pempheridae. Sweepers (bullseyes). FAO Species Identification Guide for Fishery Purposes. The Living Resources of the Western Central Pacific vol. 5. Bony fishes part 3 (Menidae to Pomacentridae) pp. 3201–3204. Food and Agriculture Organization of the United Nations: Rome.

Randall, J. E., & B. C. Victor. 2015. Descriptions of thirty-four new species of the fish genus Pempheris (Perciformes: Pempheridae), with a key to the species of the western Indian Ocean. Journal of the Ocean Science Foundation 18: 77 pp.

Long-legged Harvestmen of Southern Africa

New paper time!

Rhampsinitus conjunctidens, a new species of harvestmen from north-east South Africa, from Taylor (2017).

Taylor, C. K. 2017. Notes on Phalangiidae (Arachnida: Opiliones) of southern Africa with description of new species and comments on within-species variation. Zootaxa 4272 (2): 236–250.

When I first started research for my PhD thesis, *cough* years ago, I asked a number of museums if they could loan me their collections of monoscutid (now neopilionid) harvestmen. The species that I was interested in are found in Australia and New Zealand but when I opened a package of specimens sent to me from the California Academy of Sciences, I found a number of specimens from Africa in the mix. I immediately recognised what they were: not neopilionids, but representatives of another harvestment family, the Phalangiidae.

It seemed an easy enough error to make. Many species of southern African Phalangiidae resemble a lot of neopilionids in that the males have over-sized, elongate chelicerae. I referred to some of these species in the genus Rhampsinitus in an earlier post. To those not familiar with harvestmen diversity (which, let's face it, is the majority of people out there), the two groups can look very similar. True, the phalangiids are all distinctly much spikier than the neopilionids, but that doesn't seem that major a difference. To really see where they diverge from each other, you need to reach underneath the males' genital opercula and pull out their todgers.

Anywho, the specimens sat in storage for much longer than they should have, until I finally got around to looking them over in the latter part of last year. I then decided that it was worth writing them up into a short paper. Not only was there at least one new species among the specimens, they told me some very interesting things about variation within the species. Not only do Rhampsinitus species resemble Australasian neopilionids in their enlarged chelicerae, they resemble them in that individuals of a species vary in how enlarged the chelicerae are.

Major (left) and minor males of Rhampsinitus nubicolus, from Taylor (2013).

Now, I was not the first person to observe this point. Axel Schönhofer (2008) had already provided some detailed examples of variation in males of Rhampsinitus cf. leighi. I did, however, observe that the variation was even greater than Axel seemed to have recognised. Some of the least developed males of the species I was looking at had chelicerae that were pretty much no more developed than those of females. In some ways, the variation was even more remarkable than what I was familiar with in neopilionids. In most of the latter, major and minor males tend to be pretty similar to each other in features other than cheliceral development. In Rhampsinitus, we can see variation in almost all the features related to sexual dimorphism. In the species pictured immediately above, R. nubicolus, major males have massively long pedipalps as well as the long chelicerae; minor males have short, stubby pedipalps like those of a female. We can tell that they are the same species because they are found in the same location and have matching genitalia, but on the outside you would be hard pressed to pick them as such. Just to confuse matters even more, major males of two species may look very different to each other whereas minor males are externally almost identical. Without looking at the genitalia, it is all but impossible to identify which species a minor male belongs to.

As with the neopilionids, we can't yet say for sure what this variation means for the species' behaviour. In many other animal species with comparably varying males, large males will fight to protect and contain females while small males adopt a sneaking behaviour and try to spot females that are not being watched by large males. It seems quite possible that a similar thing is going on with Rhampsinitus. If you're a keen natural historian or behavioralist, there's something here that is crying to be looked into.


Schönhofer, A. L. 2008. On harvestmen from the Soutpansberg, South Africa, with description of a new species of Monomontia (Arachnida: Opiliones). African Invertebrates 49 (2): 109–126.

A Mystery Ammonoid

Münster's (1834) figure of Goniatites hybridus.

Looks like I drew another dud. For today;s semi-random post, I ended up tasking myself to write something about the Devonian ammonoid genus Heminautilinus. But as it turns out, there simply isn't that much to say about this genus, and what there is isn't really worth saying.

Heminautilinus was established as a genus by A. Hyatt in 1884. He diagnosed it as including "species with whorls similar to those of Anarcestes, but with angular lateral lobes in the adults", and designated George de Münster's (1834) Goniatites hybridus as type species on the basis of that author's original figure. The problem is that Münster's figure is apparently not very reliable; the original specimen was only fragmentary and Münster himself expressed uncertainty as to just what section of the ammonoid conch he had on hand. So Hyatt's assumption that Münster's species retained some juvenile features to maturity should not be considered reliable.

As a result, Hyatt's genus seems to have been pretty roundly ignored. Those authors who have made some speculation as to its identity have suggested that it is probably synonymous with some better known genus such as Cheiloceras or Imitoceras. This might present something of an issue because either one of these genera was published more recently than 1884, meaning that Heminautilinus should be considered the senior name. Because there would be little to be gained from replacing a familiar name with one that is all but forgotten, it seems most likely that, even if Heminautilinus' identity could be reliably established, it would be somehow suppressed. As such, Heminautilinus seems doomed to remain in obscurity.


Hyatt, A. 1883–1884. Genera of fossil cephalopods. Boston Soc. Nat. History, Proc. 22: 253–338.

Münster, G. de. 1834. Mémoire sur les clymènes et les goniatites du calcaire de transition du Fichtelgebirge Annales des Sciences Naturelles, seconde série, Zoologie 1: 65–99, pls 1–6.

Blue Moon

Male blue moon butterfly Hypolimnas bolina, photographed by Comacontrol.

My native country of New Zealand is not home to a large diversity of butterflies. Only a couple of dozen or so species are known from the entire country. It would not be unreasonable for a keen butterfly spotter to attempt to track down them all. But one particular species of butterfly generally included in New Zealand lists would a touch of luck: the aptly named blue moon Hypolimnas bolina.

This is because the blue moon is not a regular resident of New Zealand (I've personally never spotted one). It is native to a wide region stretching from Madagascar and India to Japan and northern Australasia where it is usually referred to by the more prosaic name of common or greater eggfly. The only examples found in New Zealand are vagrants who lost their way on southwards migrations. Nevertheless, such vagrants are regular enough for its local appellation to be thought worth coining. Not only does it reflect their rarity, it also describes the appearance of the male, with the wings bearing blue-ringed white spots on a black background.

Two females of Hypolimnas bolina. On the left, a mimetic individual, copyright Greg Hume; on the right, a non-mimetic individual, copyright W. A. Djatmiko.

The appearance of the female is a bit harder to explain because it can vary between individuals. Females of the eggfly genus Hypolimnas are commonly mimics of other, poisonous butterflies of the subfamily Danainae, to which eggflies are only distantly related (both groups belong to the family Nymphalidae but eggflies are placed in the subfamily Nymphalinae). For instance, the diadem or danaid eggfly H. misippus of Africa and Asia (and also introduced into parts of the Americas adjoining the Caribbean) is a mimic of the plain tiger Danaus chrysippus. The chosen model of H. bolina in the western part of its range is the common crow Euploea core and in the region of India almost all females are a remarkably good copy of that species (above left). But as one moves east, one starts seeing females of H. bolina that are not mimics like the individual shown above right; by the time one reaches Australia these make up the greater part of the population. Mimetic females may also vary to resemble different Euploea species, depending on which model is locally present.

Female danaid eggfly Hypolimnas misippus, copyright Raju Kasambe. Males of this species are similar to those of H. bolina but lack the blue rings around the white spots on the wings.

There are about two dozen species of Hypolimnas eggflies found in various parts of the Old World tropics. Hypolimnas misippus is also found in parts of the Americas around the Caribbean where its presence is usually explained as the result of an early introduction (possibly, and somewhat poignantly, in connection to the slave trade). Their vernacular name is probably derived from the unique behaviour (for butterflies) of a number of species whose females stand guard over their eggs, beating their wings over them to protect them from predators until hatching. About two-thirds of Hypolimnas species are mimics. In some of these species, both sexes are mimetic; others resemble H. bolina and H. misippus in that only the females are mimics (Vane-Wright et al. 1977). One might be tempted to ask why this variation exists. One point to be considered is that there are limits on when mimicry is likely to be effective. The mimic needs to be much less abundant than its model, otherwise potential predators may not learn to associate the distinctive coloration with the toxic original. Swinhoe (1896) noted that males of Hypolimnas misippus were very active, aggressively defending their territories from other butterflies, and suggested that this agility might provide males with alternative defences to mimicry. The more sedentary females (especially when egg-guarding) might be expected to benefit more from the passive protection mimicry provides, but mimesis might be expected to disappear in areas where their model is less abundant.


Swinhoe, C. 1896. On mimicry in butterflies of the genus Hypolimnas. Journal of the Linnean Society, Zoology 25: 339–348.

Vane-Wright, R. I., P. R. Ackery & R. L. Smiles. 1977. The polymorphism, mimicry, and host plant relationships of Hypolimnas butterflies. Zoological Journal of the Linnean Society 9: 285–297.

False Spider Mites

Among the enormous diversity of the world's mites, some families are particularly notorious for the damage that they inflict on commercial plant crops. Among such Acari non grata are the spider mites of the family Tetranychidae or the gall mites of the Eriophyidae. But a third, equally notorious group is the false spider mites or flat mites of the Tenuipalpidae.

Red and black flat mite Brevipalpus phoenicis, false-colour SEM by Christopher Pooley.

False spider mites include about 800 known species of more or less flattened, slow-moving mites. They are closely related to the true spider mites and both families have the chelicerae modified into a pair of long, whip-like retractable stylets that are used to pierce and suck fluids from plant tissues. In the case of the false spider mites, though, their commercial infamy comes not only from the direct damage caused by the feeding mites themselves but also from the effects of transmitted viruses. Viruses transmitted by false spider mites include the causative agents of diseases such as citrus leprosis and coffee ring spot and may cause significant reductions in the yield and lifespan of infected plants.

Morphologically, false spider mites differ from true spider mites in the absence of what is called the 'thumb-claw' process, an arrangement of the tarsus of the pedipalp alongside a claw on the seta (hence the family name which means 'slender palp'). The palps are often reduced, with some species having only the barest remnant. Some species also show reduction in the fourth pair of legs, and females of a number of species are six-legged as adults. This merely stands as another example of how mite morphology functions purely to play silly buggers with anything one might learn in basic animal biology.

Hebe stem gall mites Dolichotetranychus ancistrus inside an open gall, copyright Plant and Food Research.

Parthenogenesis is also common in false spider mites. Species found in cooler climes will often overwinter as females, with a new generation of males not appearing until the next spring. In some species, eggs produced parthenogenetically will hatch into males; in others, they will produce females. A few species almost entirely lack functional males. A small group of these species in the genus Brevipalpus is unique among animals in being both parthenogenetic and genetically haploid.

Almost all forms of seed plant seem to be vulnerable to some form of flat mite or another; some mite species are very catholic in their tastes and will latch onto almost anything green and photosynthesising. Others are more discerning. How false spider mites make their way from one host plant to another is little known but they may be passively carried through the air on the wind. Alternatively, they may be inadvertently carried from place to place by feeding herbivores, or by the very human horticulturalists that suffer so much from their presence.


Walter, D. E., E. E. Lindquist, I. M. Smith, D. R. Cook & G. W. Krantz. 2009. Order Trombidiformes. In: Krantz, G. W., & D. E. Walter (eds) A Manual of Acarology 3rd ed. pp. 233-420. Texas Tech University Press.

Oily and Salty Trees

The Annonaceae is another one of those plant families like Acanthaceae that, despite containing a high diversity of speceis, tend to be overlooked because that diversity is mostly tropical. A number of species in the type genus Annona produce commercially significant fruits: custard apples, cherimoyas, soursops and the like. However, these are just a few of the 2400+ species of trees and lianes assigned to this family.

Ylang-ylang flowers Cananga odorata, from here.

Taxonomically, the Annonaceae is well established as distinct, readily recognised by a number of distinctive features. Among these is a characteristic 'cobweb' appearance to the wood structure when seen in cross-section, resulting from prominent rays of xylem connected by narrow cross-bands of parenchyma (Chatrou et al. 2012). Relationships within the family have been much harder to work out, not becoming well established until the advent of the molecular era. Recently, Chatrou et al. (2012) have recognised four subfamilies within the Annonaceae. The majority of species are placed in the subfamilies Annonoideae and Malmeoideae (which together form a clade), but a handful of species are placed in two basal subfamilies: one for the single genus Anaxagorea, and the Ambavioideae. Anaxagorea and the ambavioids differ from the annonoid-malmeoid clade in the structure of their seeds. Seeds of Annonaceae have what is called ruminate endosperm: that is, the surface of the endosperm is not smooth, but divided by wrinkles and grooves (the term 'ruminate' literally means 'chewed'). In Annonoideae and Malmeoideae, the ruminations of the endosperm are shaped like spines or lamellae. In Anaxagorea and the Ambavioideae, the ruminations are irregular in appearance. Molecular analyses place Anaxagorea as the sister taxon to all other Annonaceae.

View into the canopy of a salt-and-oil tree Cleistopholis patens, copyright Marco Schmidt.>

The Ambavioideae, despite not being very diverse, are widespread, with species found in the tropics of Africa, Asia and the Americas. Perhaps the best known ambavioid is the ylang-ylang tree Cananga odorata, native to south-east Asia, whose flowers are used as a source of perfume. Other south-east Asian ambavioids belong to the genera Cyathocalyx, Drepananthus and Mezzettia. The type genus, Ambavia, is native to Madagascar; other ambavioids in the genera Meiocarpidium, Cleistopholis and Lettowianthus are found in continental Africa. Finally, a single genus Tetrameranthus is found in South America. Most species of ambavioid are not systematically economically exploited but a number are locally used as sources of wood. The wood is light and not suitable for structural uses, but can be shaped and finished for utensils and other small items. The West African species Cleistopholis patens, whose Ghanaian name has been translated as 'salt and oil tree' (in reference to the taste of the bark when chewed), provides a fibrous bark that is readily stripped from the tree and is used for such purposes as matting and carrying straps (see here).


Chatrou, L W., M. D. Pirie, R. H. J. Erkens, T. L. P. Couvreur, K. M. Neubig, J. R. Abbott, J. B. Mols, J. W. Maas, R. M. K. Saunders & M. W. Chase. 2012. A new subfamilial and tribal classification of the pantropical flowering plant family Annonaceae informed by molecular phylogenetics. Botanical Journal of the Linnean Society 169: 5–40.

Hydromantes: Salamanders in Different Places

There are times when biogeography is able to throw us some real puzzlers: organisms whose distribution seems to defy expectations. Among these mysteries, special mention must be made of the salamanders of the genus Hydromantes.

Gene's cave salamanders Hydromantes genei courting, copyright Salvatore Spano.

Hydromantes is a genus containing a dozen species from among the lungless salamanders of the family Plethodontidae. Plethodontids are the most diverse of the generally recognised families of salamanders, with over 450 known species found mostly in Central and South America. Hydromantes, however, is a geographically isolated genus in this family with its species found in two widely separated regions: California in western North America, and mainland Italy and Sardinia in Europe. Though some authors have advocated treating the species found on each continent as separate genera, both morphological and molecular studies have left little doubt that the group represents a discrete clade.

Distinctive features of Hydromantes compared to other plethodontids include feet with five, partially webbed toes and a weakly ossified, flattened skull (Wake 2013). Members of this genus capture prey with a projectile tongue which is the most extensive of any amphibian, extending up to 80% of the animal's total body length (Deban & Dicke 2004). There are some differences between North American and European species notable enough for the recognition of separate subgenera (there is something of a gigantic clusterfuck surrounding the names of said subgenera but the details are far too tedious to relate here). The three North American species of the subgenus Hydromantes have bluntly tipped tails that they use as a 'fifth leg' when navigating smooth and/or slippery surfaces, whereas the European species have unremarkable pointed tails. Historically, the North American Hydromantes species have been poorly known, being isolated to restricted ranges. Hydromantes shastae is found in limestone around Lake Shasta whereas H. brunus is found in a small area of mossy talus habitat along the Merced River in the foothills of the Sierra Nevada (Rovito 2010). The third species, H. platycephalus, is found at higher altitudes in the Sierra Nevada, well over 1000 m above sea level. Individuals found living on steep slopes are known to escape predators by tightly coiling their bodies and simply rolling down the slope (García-París & Deban 1995). A molecular analysis of H. platycephalus and H. brunus by Rovito (2010) identified the former species as derived from within the latter, and Rovito suggested that H. brunus may have originated in a remnant population from when H. platycephalus moved into lower altitudes during the Ice Age.

Mt Lyell salamander Hydromantes platycephalus, copyright Gary Nafis.

The seven or eight European species are mostly placed in the subgenus Speleomantes; a single species, Hydromantes genei, is divergent enough to be placed in its own subgenus Atylodes (though most recent studies have indicated that the European Hydromantes overall form a discrete clade). Hydromantes genei and three species of Speleomantes are found in caves on the island of Sardinia; the remaining Speleomantes species on mountains of mainland Italy. Molecular analysis suggests that H. genei became isolated on Sardinia about nine million years ago, with the ancestors of the Sardinian Speleomantes arriving later about 5.6 million years ago when the Mediterranean dried out during what is known as the Messinian Salinity Crisis (Carranza et al. 2008). The absence of any Hydromantes on neighbouring Corsica is something of a mystery, and it has been suggested that they may have been present there in the past before going extinct.

Extinction also seems the most likely explanation for Hydromantes' unusual distribution. The fossil record for the genus is minimal, and provides little information not already available from living species, but molecular dating attempts agree that the division between European and North American Hydromantes happened too recently to be related to the tectonic separation of the two continents. Such a scenario would also leave open the Hydromantes' absence in eastern North America. The description in 2005 of the Korean lungless salamander Karsenia koreana demonstrated the presence of plethodontids in eastern as well as far western Eurasia, and it seems possible that Hydromantes dispersed into Eurasia via the Bering Strait landbridge, becoming widespread across the continent before extinction reduced it to the isolated relicts it is today.


Carranza, S., A. Romano, E. N. Arnold & G. Sotgiu. 2008. Biogeography and evolution of European cave salamanders, Hydromantes (Urodela: Plethodontidae), inferred from mtDNA sequences. Journal of Biogeography 35: 724–738.

Deban, S. M., & U. Dicke. 2004. Activation patterns of the tongue-projector muscle during feeding in the imperial cave salamander Hydromantes imperialis. Journal of Experimental Biology 207: 2071–2081.

García-París, M., & S. M. Deban. 1995. A novel antipredator mechanism in salamanders: rolling escape in Hydromantes platycephalus. Journal of Herpetology 29 (1): 149–151.

Rovito, S. M. 2010. Lineage divergence and speciation in the web-toed salamanders (Plethodontidae: Hydromantes) of the Sierra Nevada, California. Molecular Ecology 19: 4554–4571.

Wake, D. B. 2013. The enigmatic history of the European, Asian and American plethodontid salamanders. Amphibia-Reptilia 34: 323–336.