Ant-like Ichneumons

Female Gelis, copyright Krister Hall.

The ichneumons are one of the best-known groups of parasitoid wasps. The most familiar ichneumons are relatively large for parasitoid wasps, and sometimes even for wasps in general. This can make them somewhat intimidating in appearance, especially considering the likelihood of the long ovipositor of a female being mistaken for a sting by those not in the know. However, not all ichneumons are giants. The photo above shows a tiny ichneumon of the genus Gelis, females of which are wingless and bear a distinct superficial resemblance to ants. This resemblance is likely to afford them some protection from potential predators, and at least one Gelis species, G. agilis, has been shown to release a chemical when threatened very similar to the alarm pheromones of the black garden ant Lasius niger (Malcicka et al. 2015). On the other hand, one might be tempted to wonder if this mimicry may sometimes serve a more nefarious purpose: another species, G. apterus, has been recorded as a parasitoid of the ant-eating spider Zodarion styliferum (Korenko et al. 2013). However, G. apterus has not been recorded to use its ant appearance to lure its host; instead, the female ichneumon uses its ovipositor to pierce the igloo-like silken retreat that the spider occupies during the day. Other species of Gelis are known to be parasitoids of moth cocoons rather than spiders (Gauld 1984), so Gelis' status as an ant-mimic and its choice of host may be simple coincidence.

Phygadeuon exiguus, copyright James K. Lindsey.

Gelis belongs to a world-wide tribe of ichneumons known as the Phygadeuontini (sometimes referred to in older sources as the Gelini), a diverse group including well over 100 genera. Most, but not all, phygadeuontins are also among the smaller ichneumons. The range of hosts attacked by the group is equally diverse, including (among others) moths and lacewing pupae, and spider egg sacs, while some are hyperparasitoids on the pupae of other parasitoid wasps (Gauld 1984). Species of the genus Phygadeuon include parasitoids of wood-burrowing beetles that use the enlarged ends of their antennae to tap at wood in search of hollow burrows within. Some phygadeuontins are external parasitoids, while others are endoparasitoids. The larvae of Gelis apterus can even be regarded as true predators, as they attack not the eggs of their host but its newly-hatched spiderlings (Korenko et al. 2013). A common theme between these diverse hosts, though, is the production by most of them of silken cocoons or other protective structures that the female phygadeuontin is able to pierce with her ovipositor.

REFERENCES

Gauld, I. D. 1984. An Introduction to the Ichneumonidae of Australia. British Museum (Natural History).

Korenko, S., S. Schmidt, M. Schwarz, G. A. P. Gibson, & S. Pekár. 2013. Hymenopteran parasitoids of the ant-eating spider Zodarion styliferum (Simon) (Araneae, Zodariidae). Zookeys 262: 1–15.

Malcicka, M., T. M. Bezemer, B. Visser, M. Bloemberg, C. J. P. Snart, I. C. W. Hardy & J. A. Harvey. 2015. Multi-trait mimicry of ants by a parasitoid wasp. Scientific Reports 5: 8043. doi:10.1038/srep08043.

The Running of the Spiders

Nursery-web spider Dolomedes minor, sitting atop its nursery web. Copyright Konstable.

Spiders are one of the most familiar groups of invertebrates out there. There's no denying this: everybody knows what a spider is. But for various reasons, the classification of spiders tended to lag a bit behind that of other terrestrial invertebrates. Being softer-bodied than insects, they tend not to exhibit the wealth of features that made many insect groups instantly discernible. To the modern arachnologist's eye, the earliest classifications of spiders can verge on the humorous. Latreille (1802), in his Histoire Naturelle des Crustacés et des Insectes, classified the entirety of what would now be called the araneomorph spiders into a single genus Aranea, divided into sections labelled not with formal names but with schematic diagrams of the arrangement of eyes found in that section.

A few decades later, in 1829 (translated into English in Cuvier, 1831), Latreille was to present a more detailed classification of the spiders, in which they were divided into groups largely on the basis of their life habits. The araneomorphs were hence divided between the Sedentariae, those spiders which captured their prey in webs or laid in ambush, and the Vagabundae, those spiders that actively hunted down their prey. The Vagabundae were in turn divided between two sections: the Citigradae or runners, and the Saltigradae or jumpers. Latreille's classification was subsequently more or less abandoned, as his behavioural groupings failed to line up directly with morphological clusters. Almost by accident, however, those taxa included by Latreille in his Citigradae have continued to be associated, and in modern classifications are classified within the Lycosoidea (Jocqué & Dippenaar-Schoeman 2007).

The lycosoids are, indeed, mostly active hunters. Their behaviour is reflected in the vernacular names of a number of the constituent families: the wolf spiders of the Lycosidae (previously featured here and here), the lynx spiders of the Oxyopidae, the prowling spiders of the Miturgidae. But the correspondence to Latreille's 'araignées loups' is not perfect: the Zoropsidae, for instance, are lycosoids that spin extensive webs. Nor are they mere rapacious hunters: many are devoted parents, carrying and/or guarding their egg-sacs to protect them from predators, and in the case of the Lycosidae even providing a certain degree of care for the newly hatched spiderlings.

One group of lycosoids has even gotten a name for parental care. The nursery-web spiders of the Pisauridae construct protective webs for their babies, containing them within a tent constructed by wrapping sheets of silk around suitable vegetation. When I was a child in New Zealand, I used to be fascinated by the nursery webs constructed by the species Dolomedes minor. Like many pisaurids, this species is associated with water, diving into it to hunt for fish and other small aquatic animals. The females would often build their nursery webs by tying together the ends of nearby rushes. Though it seems a little cruel to my adult self, the younger me loved to pull these webs apart to see the eruption of tiny spiders come scurrying out.

REFERENCES

Cuvier, G. 1831. The Animal Kingdom arranged in conformity with its organization, vol. 3. The Crustacea, Arachnides and Insecta, by P. A. Latreille, translated from the French with notes and additions, by H. M'Murtrie. G. & C. & H. Carvill: New York.

Jocqué, R., & A. S. Dippenaar-Schoeman. 2007. Spider Families of the World. Royal Museum for Central Africa: Tervuren (Belgium).

Latreille, P. A. 1802. Histoire Naturelle, générale et particulière des Crustacés et des Insectes, vol. 3. F. Dufart: Paris.

Horny-Arsed Trilobites

Reconstruction of Ceratopyge, from here.

Just a short post for today. The Ceratopygidae are a family of trilobites known from the Late Cambrian and Early Ordovician. The name of the type genus, Ceratopyge, means 'horned rump', and one of the features that has classically defined the family is the presence of one or two pairs of spines on either side of the pygidium, the plate the makes up that hind end of a trilobite. These spines appear to be derived from lateral extensions of one of the anterior segments incorporated into the pygidium. However, there are also some genera without pygidial spines that share other features with the family (such as a narrow rim to the cheeks) and so have also been recognised as ceratopygids. Ceratopygids also possessed narrow spines extending back from the posterior corners of the head. The number of segments between head and pygidium varied between genera: early genera have nine segments, but some later genera have only six (Fortey & Chatterton 1988) (offhand, the drawing above looks to have one too many segments).

Proceratopyge gamaesilensis, from here.

Otherwise, ceratopygids seem to have been fairly generalised trilobites. The eyes were present but not large, and there don't appear to be any features suggesting they were swimmers. The features of the underside of the head are poorly known in ceratopygids overally, but where known, the hypostome (the plate on the underside of the head that would have sat in front of the mouth) is firmly attached to the anterior margin of the head. Trilobites with this arrangement are believed to have been scavengers or predators on small invertebrates (Fortey & Owens 1999). In some later genera, such as Ceratopyge, the glabella in the midline of the cephalon expanded forward, with a corresponding reduction in the width of the anterior margin. As the glabella would have contained the trilobite's stomach, its enlargement may indicate that these later ceratopygids were taking larger prey.

REFERENCES

Fortey, R. A., & B. D. E. Chatterton. 1988. Classification of the trilobite suborder Asaphina. Palaeontology 31 (1): 165-222.

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

Polypodies: In the Fernery of the Senses

Common polypody Polypodium vulgare, copyright Paul Montagne.

I'm not sure if I've ever had cause before to present my concept of the Evil Old Genus. The Evil Old Genus is one that has been used in the past to refer to a massively broader concept than it does currently, and so has been used to refer to many more species in the past than now. This makes dealing with the taxonomy of the genus a major headache, as one has to consider a whole host of now hidden or forgotten combinations. I can't say what would be the most evil of the Evil Old Genera out there, but a definite leader has to be the fern genus Polypodium. When the name was used by Linnaeus way back in the mid-1700s, Polypodium referred to nearly the whole gamut of ferns. Over time, as botanists have come to appreciate that all ferns are not the same, Polypodium has been progressively cut down. Still, it seems that if you go back into the taxonomy of nearly any fern, you'll come up against a 'Polypodium' sooner or later.

At present, Polypodium refers to a group of ferns with creeping, often scaly stems. It is the appearance of these stems that gives them their genus name, meaning 'many feet', as well as the common vernacular name of polypody. The circumscription of the genus can still vary somewhat between authors: some would include about 250 species in the genus, but Smith et al. (2006) restricted Polypodium to only about 30 species found primarily in temperate regions of the Northern Hemisphere, and in Central America. Many of these belong to what is known as the Polypodium vulgare complex. Recognised in the past as a single species Polypodium vulgare, this complex is now recognised as including a number of species found across Eurasia and North America. Ten of these are diploids, but another seven are polyploids. The polyploid species are believed to have originated from hybridisations between the diploid taxa; for instance, the Eurasian Polypodium vulgare sensu stricto is a tetraploid derived from a hybridisation between the diploid species P. glycyrrhiza and P. sibiricum (Sigel et al. 2014). Sigel et al. (2014), investigating the relationships between its diploid species, estimated an early Miocene origin for the P. vulgare complex. A fossil species from the early Oligocene, P. radonii, may belong to the complex or may be closely related (Kvaček 2001).

Appalachian rockcap fern Polypodium appalachianum, copyright Jaknouse.

Distinguishing species of the P. vulgare complex is no easy task, often requiring evaluation of subtle differences in leaf or stem form, or close examination of sporangium morphology. Another feature that has been used in distinguishing Polypodium species, however, is taste: the stems of some species in the complex have distinctive flavours. The Eurasian P. vulgare has been used to impart its bittersweet flavour to confectionary, while the vernacular name of the licorice fern P. glycyrrhiza of North America and eastern Asia is fairly self-explanatory (but like licorice, does it also give you a good run for your money?) The key to Polypodium species in the Flora of North America contains the somewhat unexpected advice that "the reader is cautioned to taste clean rhizomes from uncontaminated soils". And honestly, who could argue with that?

REFERENCES

Kvaček, Z. 2001. A new fossil species of Polypodium (Polypodiaceae) from the Oligocene of northern Bohemia (Czech Republic). Feddes Repertorium 112 (3-4): 159-177.

Sigel, E. M., M. D. Windham, C. H. Haufler & K. M. Pryer. 2014. Phylogeny, divergence time estimates, and phylogeography of the diploid species of the Polypodium vulgare complex (Polypodiaceae). Systematic Botany 39 (4): 1042-1055.

Smith, A. R., H.-P. Kreier, C. H. Haufler, T. A. Ranker & H. Schneider. 2006. Serpocaulon (Polypodiaceae), a new genus segregated from Polypodium. Taxon 55 (4): 919-930.

The Urbaum

Reconstruction of Archaeopteris, from Beck (1962).

It appears that it's been over a month now since I last posted anything at this site. I'm not going to go back and check, but I think this may be the longest hiatus that Catalogue of Organisms has been through since I first launched it nearly eight years ago. I have my excuses all prepared: it's been a busy period, what with trips back home to New Zealand, general job-hunting type stuff, and construction work around the house*. Nevertheless, I have had subjects lined up to present here all that time (nothing to do with construction, I promise you), and so I've found myself looking up material on Archaeopteris.

*An enterprise absolutely guaranteed to transform you into mind-breakingly tedious company for everyone else.

Archaeopteris, I hasten to explain, is nothing to do with Archaeopteryx, though certain parallels could be drawn (albeit with a long bow). Archaeopteryx, of course, is the Jurassic fossil genus that has become renowned as the Urvogel, the original bird. Archaeopteris is a much older fossil, coming from the Late Devonian. And if Archaeopteryx is to be known as the Urvogel, then Archaeopteris can claim to be the Urbaum, the original tree. It was not the earliest arborescent plant: the slightly earlier cladoxylopsid (a distant relative of modern ferns) Wattieza reached a height of at least eight metres (Stein et al. 2007). But Wattieza, with a single central trunk bearing a crown of fronds, would have been more similar to a modern tree fern or palm. Archaeopteris, with substantial side branches arising from its trunk, would have been more similar to the classic image of a modern tree.

Section of Archaeopteris branch, from Beck (1962). The globular structures are sporangia.

When it was first described, from its foliage alone, Archaeopteris was also believed to be an early fern. It wasn't until the early 1960s that fossils were described associating the fern-like foliage to large conifer-like logs that had been described from the same period. The entire tree was estimated to reach heights of at least sixty feet (about 18 metres) (Beck 1962). Archaeopteris was not a fern, but a member of the lineage leading to modern seed plants. As well as its overall habit, Archaeopteris resembled a modern tree in the presence of secondary thickening: a layer of cambium (generative cells) around the outer part of the trunk produced new phloem (nutrient-conducting cells) outside itself and new xylem (water-conducting cells) on the inside, thus allowing the trunk of the tree to expand as it grew (compare that to a tree fern, which gets no broader as it gets taller). However, as well as its fern-like foliage, Archaeopteris still resembled a more primitive plant in one very important regard: rather than producing seeds like a modern tree, it still reproduced through spores. Modified fronds produced clusters of sporangia, with at least some Archaeopteris species showing signs of the production of distinct male and female spore types. Whether these spores produced independent gametophytes in the manner of modern ferns is unknown, and likely to remain so: not only would such gametophytes probably be small and unlikely to be preserved, but they would have few if any features to associate them with the lofty trees.

Archaeopteris also exhibited a few other noteworthy differences from a modern tree. Most recent trees are more or less monopodial: they have a central main shoot from which branches arise laterally as adventitious primordia. Archaeopteris' main mode of growth was pseudomonopodial: instead of lateral branches arising de novo, they developed from the uneven division of the central shoot, with one part continuing upwards and the other part turning outwards. Though the end result would have looked broadly similar, there are some different functional implications. Archaeopteris' growth form may have been more constrained than most modern trees. Because branches were produced in the same spiral as leaves, there could have been a certain fractal-ness to Archaeopteris' appearance, with each major branch being something of a miniature of the tree as a whole (albeit a somewhat lopsided one, as at least some species produced larger leaves on the upper side of branches than on the lower side). Also, a purely pseudomonopodial mode of growth would not allow for the replacement of lost branches or other appendages: Trivett (1993) compared this model of the growth of Archaeopteris to "an inflating balloon or an opening umbrella with its increasingly empty interior". At the same time, she presented evidence that Archaeopteris could have produced a certain degree of adventitious growth, though it may still have been less resilient to damage than recent analogues. There is some circumstantial evidence that Archaeopteris may have sometimes shed leaves or minor branches en masse, though whether this was a seasonal occurrence or a response to stress is unknown.

Despite being potentially more vulnerable to damage than a modern tree, Archaeopteris was undeniably successful. Various species of the genus were found pretty much around the world, and were the dominant large plant wherever they were found until their extinction around the beginning of the Carboniferous. Perhaps resilience was simply less of an issue for Archaeopteris than for modern trees. After all, it lived in a world where there would have probably still been no major herbivores, and the main causes of appendage loss would have been the weather or disease. Also, long-term resilience may have simply not been so important for a tree that probably produced spores by the millions every year. Who knows how many Archaeopteris sporelings or gametophytes there may have been at a time, simply waiting their opportunity to provide a replacement for a fallen senior?

REFERENCES

Beck, C. B. 1962. Reconstructions of Archaeopteris, and further consideration of its phylogenetic position. American Journal of Botany 49 (4): 373-382.

Stein, W. E., F. Mannolini, L. V. Hernick, E. Landing & C. M. Berry. 2007. Giant cladoxylopsid trees resolve the enigma of the Earth's earliest forest stumps at Gilboa. Nature 446: 904-907.

Trivett, M. L. 1993. An architectural analysis of Archaeopteris, a fossil tree with pseudomonopodial and opportunistic adventitious growth. Botanical Journal of the Linnean Society 111: 301-329.

The Rosy Birds

Violet-necked lories Eos squamata, copyright Niels Poul Dreyer.

In taxonomic days of yore, it was a not uncommon practice for new genera to be baptised under the names of classical figures: gods, heroes, emperors, even the occasional prophet (the practice only died down once the barrel of available names became largely empty). In many cases, the connection drawn between the organism in question and its awarded namesake was tenuous at best. In others, it was simply non-existent. But in a favoured few cases, the association fit perfectly.

Eos is a small genus (recent authors have recognised six species) of lories found on islands in eastern Indonesia. They are named, of course, after the Ἠώς ῥοδοδάκτυλος, the 'rosy-fingered dawn', of the ancient Greeks. It takes no great insight to realise why they were so-called: all members of the genus are predominantly coloured in a vibrant red, together with varying extents of blue, purple and/or black. Green is usually absent from their plumage (with some noteworthy exceptions that I'll have cause to mention again), distinguishing them from most closely related parrots such as the rainbow lorikeets in the genus Trichoglossus. Charles Lucien Bonaparte (nephew to the other Bonaparte, and a prominent nineteenth-century ornithologist) stated in 1850 that Eos could be recognised by its "elegant form, small stature, compact, red plumage with more or less blue; compressed, moderate, red bill, with the cere apparent... and longish, not very broad, wedged tail".

Blue-streaked lories Eos reticulata, copyright Doug Janson.

For the most part, Eos species are found on islands between Sulawesi and New Guinea. The black-winged lory Eos cyanogenia is found on islands in Geelvink Bay, in the north-west part of West Papua, but not on the mainland of New Guinea itself. For the most part, no island is home to more than one species of Eos. The island of Seram is an exception, with the endemic blue-eared lory Eos semilarvata found in the central highlands, and the red lory Eos whatchumacallit (see below) closer to the coast (this species is also found on other islands in the South Moluccas). The blue-streaked lory Eos reticulata is found in the Tanimbar group east of Timor. The violet-necked lory Eos squamata lays claim to the North Moluccas, and the red-and-blue lory Eos histrio is found on Talaud and other islands to the north-west of Sulawesi (Juniper & Parr 1998).

Black-winged lory Eos cyanogenia, copyright Lip Kee Yap.

While the taxonomy of the group has been mostly stable in recent years, it was not always so. Bonaparte (1850) snidely commented that some species of Eos had been described "too many times". Hume & Walters (2012) referred to five described species of Eos, all based on isolated specimens since lost, whose identity has been contested. While it is possible that some may represent species now extinct, it is equally possible that they represented unusual individuals of living species. In the absence of examinable type specimens, the identity of most is of academic interest only. The exception is the 'red-and-green lory' Eos bornea, which was originally named Psittacus borneus by old Carolus Linnaeus himself on the basis of a description and plate of a lory supposedly from Borneo published in 1751 by George Edwards (Walters 1998). Edwards' bird, which he had bought as a stuffed specimen from a toyshop in London, was described as dark pink, with a yellow bill, and green patches on the wings and tail. However, no species quite matching Edwards' description is known from Borneo or anywhere else, and it was subsequently suggested that he may had an unusual or a faded specimen of the Moluccan red lory, with the Bornean locality being an error. As such, the name Eos bornea came into use for the red lory, replacing the later-published name 'Eos rubra'. However, Walters (1998) subsequently disputed this identification, recommending the continued use of E. rubra. At present, 'Eos bornea' still seems to be the more commonly used name, and my own sympathies would be more with maintaining the familiar usage than with insisting on strict adherence to the original concept.

Red lories, Eos... let's just say bornea, shall we? Copyright Arnaud Delberghe.

Because of their striking appearance, Eos species have been heavily collected for the pet trade. The have also been widely affected by habitat degradation with the clearing of primary forests. While populations of most species are still regarded as reasonably robust, the IUCN regards all except E. squamata as on the decline. Eos histrio is regarded as actively endangered, having all but disappeared from some of its home islands. In 1999, it was estimated that 1000 to 2000 red-and-blue lories were being captured and exported for the pet trade each year—despite the total population of this species probably being not much more than 20,000 individuals!

REFERENCES

Bonaparte, C. L. 1850. On the trichoglossine genus of parrots, Eos, with the description of two new species. Proceedings of the Zoological Society of London 18 (1): 26-29.

Hume, J. P., & M. Walters. 2012. Extinct Birds. T. & A. D. Poyser.

Juniper, T., & M. Parr. 1998. Parrots: A guide to the parrots of the world. Christopher Helm Publishers.

Walters, M. 1998. What is Psittacus borneus Linnaeus? Forktail 13: 124-125.

In a Pufferfish's Garden

Bullseye puffer Sphoeroides annulatus, copyright Geoffrey W. Schultz.

I don't know if it applies in other parts of the world, but one animal that you are guaranteed to see in the estuary here in Perth is pufferfish. One of the most instantly recognisable fish families, pufferfish (Tetraodontidae) are of course famed for their high toxicity, the determination of some people to eat them despite aforementioned toxicity, and their habit of swallowing air or water when threatened to inflate their distendible bellies. That last feature makes them a favourite of children (or at least of yours truly as a child), because their slow swimming style makes them one of the few fish that can be easily captured by hand (you just have to make sure you don't allow the fish to give you a nasty bite with their beak). The first feature makes them a lot less popular with fishermen who have to experience the frustration of reeling in a line to find that the bait has been taken by a puffer, then trying to remove the puffer from the hook while avoiding the aforementioned beak.

Oceanic puffer Lagocephalus lagocephalus, from Baino96.

There are a little under 200 known pufferfish species worldwide. Most of them are found in coastal marine and brackish waters, but there are also several species found in fresh water in South America, Africa and southeast Asia. Some marine species are also resistant to fresh water and may spend extended periods away from the sea. Some southeast Asian brackish-water Tetraodon species even make regular appearances in the the aquarium trade labelled as 'freshwater' puffers (Yamanoue et al. 2011), though their long-term survival requires more appropriate water conditions. The toxin associated with pufferfishes is not produced by the fish itself, but accumulated through its diet. As such, the exact level of toxicity of a pufferfish may vary according to season.

Grass puffer Takifugu niphobles, copyright OpenCage.

A molecular phylogenetic analysis of pufferfish by Yamanoue et al. (2011) identified four main clades in the family. These clades were also supported by a subsequent analysis by Santini et al. (2013), though the deeper relationships between the clades differed between the analyses. Yamanoue et al. (2011) identified a small number of freshwater clades (only one for each continent with freshwater taxa) and inferred that the transition from marine to fresh water had happened only rarely. Santini et al. (2013), in contrast, supported a higher number of transitions in tetraodontid history, though at least some of the difference between the two studies can be explained by differing definitions of 'freshwater'. For instance, some species of Takifugu usually live in brackish water but spawn in fresh water; Santini et al. counted these as freshwater species, but Yamanoue et al. did not.

Papuan toby Canthigaster papua, photographed by Dwayne Meadows.

One of the major clades identified within the Tetraodontidae includes the genus Lagocephalus, a group of relatively long-bodied puffers including some of the few pelagic puffer species. This genus may be the sister taxon of the remaining puffers (as found by Yamanoue et al.), or it may have a more nested position as sister to a clade including the mostly West Atlantic-East Pacific genera Sphoeroides and Colomesus (as found by Santini et al.). This latter clade includes South America's only freshwater puffer, the Amazon species Colomesus asellus. Santini et al. identified the basalmost tetraodontid clade as an Indo-West Pacific assemblage including the genus Takifugu and related taxa, which Yamanoue et al. had found as sister to the final clade including taxa related to the genus Tetraodon. This last clade includes the African and southeast Asian freshwater puffers (except for a few members of the Takifugu clade that cross into fresh water at times). It also includes the genus Canthigaster, the sharpnose pufferfish. In contrast to the more or less globular form of all other puffers, sharpnose puffers have a laterally compressed body form that superficially looks a bit more like a triggerfish than a puffer. Most Canthigaster species are reef-dwellers, a somewhat unusual habitat for a puffer (the other main group of reef-dwelling puffers being the genus Arothron, also in the Tetraodon clade).

Circular underwater 'nest' constructed by a pufferfish, from Spoon & Tamago.

One of the most remarkable characteristics of any puffer, though, was not discovered until quite recently. In 2012, it was announced that large structures observed off the coast of Japan by underwater photographer Yoji Ookata were in fact the work of pufferfish. These structures, circular and regular geometric patterns in the sea bed about 1.5 metres in diameter, were made by male puffers swimming against the sand. The structures are believed to function in attracting females, and also function as nests in which the females lay their eggs. Rather frustratingly, I haven't found any indication exactly which species of puffer is involved!

Puffer in the process of building a nest, also from Spoon & Tamago.

REFERENCES

Santini, F., M. T. T. Nguyen, L. Sorenson, T. B. Waltzek, J. W. Lynch Alfaro, J. M. Eastman & M. E. Alfaro. 2013. Do habitat shifts drive diversification in teleost fishes? An example from the pufferfishes (Tetraodontidae). Journal of Evolutionary Biology. doi: 10.1111/jeb.12112.

Yamanoue, Y., M. Miya, H. Doi, K. Mabuchi, H. Sakai & M. Nishida. 2011. Multiple invasions into freshwater by pufferfishes (Teleostei: Tetraodontidae): a mitogenomic perspective. PLoS ONE 6 (2): e17410. doi:10.1371/journal.pone.0017410.