Field of Science

The Pisocrinidae: Babyface Crinoids

One question that I haven't yet found an answer to is why the Palaeozoic marine fauna seems to have included so many filter feeders. Cystoids, blastoids, graptoloids... so many of the distinctive taxa occupying this niche would be gone by the period's end, without leaving any clear analogues behind them. What was the cause underlying this abundance? Is it simply a misapprehension caused by the filtering effect of history, with the modern fauna containing fewer major lineages but no fewer actual species? Is it the distorting lens that causes us to tend to assign a higher 'rank' to those lineages arising earlier in time, whatever their practical levels of disparacy? Or was there actually something different about what could be found in Palaeozoic seawater?

Reconstructions of short-armed and long-armed species of Pisocrinus, from Rozhnov (2007).


The Pisocrinidae are one of those distinctive Palaeozoic marine groups, known from around the world during the Silurian and Devonian. As crinoids, they were perhaps not as immediately unfamiliar to the modern eye as some of the other taxa that could be found at that time, but they were certainly different from any modern crinoid. The majority of the crinoids that have ever lived can be assigned to one of two main clades. One, the cladid lineage, includes all the crinoids alive today. Pisocrinids belong to the other major lineage, the disparids, which were prominent for most of the Palaeozoic era but failed to make it past the end of the Permian. Disparids differed from cladids in that their calyx included a single circlet of plates (the inferradials) beneath the circlet of the radials (the large plates making up the main body of the calyx) whereas cladids (at least to begin with) had two such circlets. Many disparid sublineages showed a tendency towards reduction and/or simplification of the calyx. In pisocrinids, most of the calyx was made up of just three plates: two large radials (representing the A and D rays of the basic crinoid calyx) and a greatly enlarged B inferradial. The B, C and E radials were all reduced in size. The arms of pisocrinids mostly lacked lateral pinnules and were undivided; one genus, Cicerocrinus, had bifurcating arms bearing lateral ramules (Moore et al. 1978). The length of the arms varied considerably between species: in some they were quite short and broad, in others they were remarkably long. Because their derived morphology made it difficult to compare pisocrinids to related families, their origins have been regarded as mysterious. Rozhnov (2007) suggested a derivation from an earlier, more typical crinoid family, the Homocrinidae, via paedomorphosis, possibly as a result of the evolution of a longer larval period in the life cycle (he specifically suggested that this extended larval phase may have allowed the ancestors of pisocrinids to spread across the Iapetus Ocean between the then-existing continents of Laurentia and Baltica). A direct pisocrinid-homocrinid connection was not supported in the phylogenetic analysis of disparids by Ausich (2018) but Rozhnov's overall model of pisocrinid paedomorphosis remains a possibility.

Assemblage of Triacrinus, from here.


During the Silurian, pisocrinids were among the most abundant, if not the most abundant, groups of crinoids. They were found in a variety of habitats but were particularly abundant around reefs in deeper waters. At first glance, the non-pinnulate arms of pisocrinids appear poorly suited for filter feeding, and one might be inclined to propose a more tentacular method of obtaining food items. However, such a method would seem unlikely for the short-armed species, whose arms would have been almost entirely inflexible. Even the long-armed species sometimes had arms made up of relatively long segments whose flexibility may have been limited. An alternative possibility, I suppose, is that in life pisocrinids may have had long tube feet that took the place of the missing pinnules. Meanwhile, the absence of the pinnules meant that the arms could be lain tightly alongside each other when the crown was closed. Earlier authors presumed that, because of their preference for deeper waters, pisocrinids were rheophobic (that is, they were found in places where the water lacked a noticeable current). However, Ausich (1977) proposed that they were low-energy rheophilic, seeking locations where a moderate but steady current prevailed. The current would provide a steady supply of organic particles that could be captured by the crown, and the ability to close the arms tight would protect the oral region during occasional bouts of rougher conditions.

REFERENCES

Ausich, W. I. 1977. The functional morphology and evolution of Pisocrinus (Crinoidea: Silurian). Journal of Paleontology 51 (4): 672–686.

Ausich, W. I. (in press, 2018) Morphological paradox of disparid crinoids (Echinodermata): phylogenetic analysis of a Paleozoic clade. Swiss Journal of Palaeontology.

Moore, R. C., N. G. Lane, H. L. Strimple, J. Sprinkle & R. O. Fay. 1978. Inadunata. In: Moore, R. C., & C. Teichert (eds) Treatise on Invertebrate Paleontology pt T. Echinodermata 2. Crinoidea vol. 2 pp. T520–T759. The Geological Society of America, Inc.: Boulder (Colorado), and The University of Kansas: Lawrence (Kansas).

Rozhnov, S. V. 2007. Changes in the Early Palaeozoic geography as a possible factor of echinoderm higher taxa formation: delayed larval development to cross the Iapetus Ocean. Palaeogeography, Palaeoclimatology, Palaeoecology 245: 306–316.

The Most Australian of Plants

Imagine yourself standing in a remote corner of northern Australia. Before you stretches an expanse of rolling hills, extending as far as the eye can see. The hills are covered with a carpet of green. You step forward, eager to explore these open fields. But as you approach them, everything changes. What appeared to be a uniform carpet is actually dense tussocks, each separated by an underlay of bare gravel. And instead of soft, yielding blades, the tussocks offer you nothing but resin and hate. Welcome to spinifex country.

Grassland dominated by Triodia pungens (bright green) and T. basedowii (grey-green), copyright Hesperian.


The spinifexes of the genus Triodia are a uniquely Australian group of plants. Some North American grasses have been assigned to this genus in the past but have since been moved elsewhere. There is also a widespread genus of coastal grasses that formally goes by the name of Spinifex but that is something different again. In many parts of arid Australia (and arid Australia equals most of Australia), spinifexes are the dominant form of plant life. As noted above, they grow in tight tussocks that may reach remarkable sizes and densities: clumps of the largest species may reach 2.5 metres in height and six metres in diameter (Lazarides 1997). Not uncommonly, these largest patches will be ring-shaped due to the centre dying off while growth continues around the edges. The leaf blades are long, needle-shaped, woody and rigid. Speaking from experience, the sharp tips of these blades will break off all too easily, embedding themselves in the flesh of any passers by. And some idea of their rigidity will also be conveyed by the fact that, in the growth season, it was not uncommon to discover macabre shish kebabs made from jumping grasshoppers that had had the misfortune to land on the end of one.

Mature stand of Triodia irritans, showing the tendency of hummocks to grow into circles as the centre dies off. Copyright ANBG photo M. Fagg.


Nearly 70 species are currently recognised within the genus, often differing in their preferred microhabitat. One of the most common species, Triodia basedowii, extends its range across almost the entirety of the continent between 18 and 30 degrees South and west of the Great Dividing Range. This species has a preference for sandplains and dunefields. Other species are far more localised. Barrett & Barrett (2011) described two new species found in association with sandstone cliff faces in the Ragged Range in Western Australia. Triodia barbata was found only in a thin band along the top of the cliff faces and may have had a population of only about 300 individuals. The more abundant (but still not widespread) T. cremnophila was found only on the vertical faces of the cliffs themselves. However, it must be noted that large gaps may exist in our knowledge of the ranges of Triodia species because of the remoteness and difficulty of getting to many of the regions in which they are found (seriously, if you've never been to central Australia yourself, it is difficult to appreciate just how much Absolutely Nothing there is there). Triodia mollis is known from two widely separated regions in northern Western Australia and Queensland with no confirmed records as yet from the entire expanse of the Northern Territory in between.

Preferred habitat of Triodia cremnophila, from Barrett & Barrett (2011). Yes, it only grows on the cliff face. Yes, someone presumably went down the cliff face to get specimens.


Being as woody and harsh as it is, it should come as no surprise that relatively few animals are capable of eating spinifex. Many Australian termites, such as the endemic genus Drepanotermes, are spinifex specialists; workers of Drepanotermes may be seen leaving their nest at night to collect pieces of spinifex blades and carry them back. Pastoralists may refer to 'hard' and 'soft' spinifex varieties but the difference is one of degree only; even the 'soft' spinifexes (usually the resin-producing species) are pretty damn hard by the standards of any other grass. Livestock are sometimes grazed on spinifex when bettter options are unavailable, in which case patches of spinifex may be burnt off to encourage the production of younger, more palatable growth (spinifex burns exceedingly well but also grows back readily from the remnant rootstock). The resin from spinifex also has a history of being used by indigenous Australians as an adhesive when making tools. For the most part, though, the main value of spinifex remains in its role as the dominant vegetation and habitat for the areas where it is found.

REFERENCES

Barrett, R. L., & M. D. Barrett. 2011. Two new species of Triodia (Poaceae: Triodieae) from the Kimberley region of Western Australia. Telopea 13 (1–2): 57–67.

Lazarides, M. 1997. A revision of Triodia including Plectrachne (Poaceae, Eragrostideae, Triodiinae). Australian Systematic Botany 10: 381–489.

The Australian Panda

The world is home to an incredible diversity of snails: there are literally thousands of species, some widespread, some restricted to very small areas. Most, as is the usual way of things, are tiny, barely discernible without very close examination. But then there are some that are very much not—such as the giant panda snail Hedleyella falconeri.

Giant panda snail Hedleyella falconeri, from the Queensland Museum.


Giant pandas are found on rainforest floors in northern New South Wales and southern Queensland, in a range spanning from the Barrington Tops to the D'Aguilar Range. They are Australia's largest land snail, reaching nine centimetres in diameter, about the size of a tennis ball. They have globose, reddish brown shells with a spiral pattern of darker broken bands. Their name bears no relation to any Asian mammals; instead, they were gifted the genus name Panda as a derivation from the Latin word pandere, meaning to stretch out or extend, presumably in reference to their size. The genus name later had to be changed but it survives in the vernacular (as well as in the name of a closely related genus of slightly smaller snails, Pygmipanda).

Panda snails are nocturnal, spending the days in moist spots such as buried in leaf litter or hidden under logs. At night, they roam in search of fallen leaves and fungal fruiting bodies. A study of giant pandas that tracked individual snails found that they wandered more or less randomly, up to about 20 metres over the course of a night, without returning to any particular 'home' site.

A demonstration of the size of H. falconeri, from Pollinator Link.


Like many other land snails, giant pandas are hermaphrodites, able to both fertilise and be fertilised during mating. They may have the largest sperm cells of any mollusc, each over a millimetre in length. Mating usually happens on a February night though observations in captivity suggest it may happen whenever consitions are suitable. The snail lays its hard-shelled eggs in batches of fifteen to twenty in a burrow in the leaf litter*. To continue with the theme, these are also realtively gigantic: close to two centimetres in diameter, comparable in size to those of a small bird. The young snails hatch at about 15 mm in size (I haven't found any reference to the eggs being tended by the parent in any way) and grow slowly. By the time they reach a year in age, they may not have even doubled in size, and it presumably takes several years for them to reach their full extent.

*So it turns out Paazan was right after all: pandas do hatch from eggs.

Pandas are not uncommon within their range and are not generally regarded as a conservation concern. Indeed, their nomadic habits have led to the suggestion that they may be well disposed to re-colonising regenerating forest (Parkyn & Newell 2013). Nevertheless, recent years have seen increasing fragmentation of suitable habitat within their range and this, together with their slow growth rate, means that I can easily imagine them becoming vulnerable if conditions deteriorate. I would hope that appropriate action is taken to ensure that there should always be giant pandas in eastern Australia.

REFERENCE

Parkyn, J., & D. A. Newell. 2013. Australian land snails: a review of ecological research and conservation approaches. Molluscan Research 33 (2): 116–129.

Boreonectes: Diversity Hidden Underwater

The beetle in the photo below (copyright Joakim Pansar) may or may not be Boreonectes griseostriatus. This small diving beetle, a few millimetres in length, has been regarded in the past as widespread with a distribution spanning the Holarctic region. However, in recent years it has become apparent that this single widespread species may actually be a number of more localised species in a skin.


This possibility had been considered for a while. In 1890, a Norwegian entomologist recognised distinct montane and coastal species, noting a tendency for the former to the neatly striped whereas the latter was more blotchy. Later authors, however, rejected this distinction. In 1953, a Russian author expressed the view that B. griseostriatus "varies markedly in many characters; all attempts to establish subspecies and varieties are unjustified, because almost all varieties are connected by transitions" (Angus et al. 2015). In its overall appearance, B. griseostriatus is a a fairly undistinguished small diving beetle. Most of the body surface is densely and finely punctate both dorsally and ventrally, and it lacks some of the modifications found in other diving beetles such as lateral grooves on the pronotum or sucker-hairs on the male tarsi (Angus 2010). This latter feature, offhand, is an adaptation that assists males who have it in clinging to the backs of females during mating. Their functionality would be much reduced in punctate species such as B. griseostriatus because the the uneven surface of the female would prevent the suckers from getting a grip, and phylogenetic studies suggest that their absence in Boreonectes may represent a secondary loss. I don't know if the Boreonectes males do anything to make up for their absence; maybe they just have to grip tighter.

Variation in parameres from male genitalia of the Boreonectes griseostriatus group, from Dutton & Angas (2007).


The complicated nature of B. griseostriatus' identity became really apparent in the 2000s when karyotypic studies on European specimens identified several different chromosomal races, distinct not only in chromosome topography but also in number, that may represent distinct species. The original B. griseostriatus of lowland Sweden possesses a karyotype of thirty pairs of autosomal chromosomes plus the X sex chromosome (sex is determined in this genus by an X0/XX system where males have one copy of the X chromosome and females have two, with no Y chromosome). Boreonectes multilineatus, the Scandinavian montane species, has 28 autosomal pairs. Other species have fewer. It appears likely that a similar thing is happening in Boreonectes to the situation I described in an earlier post for the bat genus Rhogeessa where mutations lead to chromosomes becoming fused or split. It is notable in this regard that Angus (2010) found several specimens of B. ibericus from Morocco that were heterozygous for a chromosomal fusion, so that a single fused chromosome was paired meiotically with distinct chromosomes 1 and 24.

Externally, however, these genetically distinct species remain all but indistinguishable. There may be a tendency for one species to be larger than another, or towards slightly different genital morphologies, but these differences are not distinct enough or consistent enough to provide a reliable guide to identification. Which, if you don't have access to fresh specimens allowing a karyotype spread, is a problem.

REFERENCES

Angus, R. B. 2010. Boreonectes gen. n., a new genus for the Stictotarsus griseostriatus (De Geer) group of sibling species (Coleoptera: Dytiscidae), with additional karyosystematic data on the group. Comparative Cytogenetics 4 (2): 123–131.

Angus, R. B., E. M. Angus, F. Jia, Z.-N. Chen & Y. Zhang. 2015. Further karyosystematic studies of the Boreonectes griseostriatus (De Geer) group of sibling species (Coleoptera, Dytiscidae)—characterisation of B. emmerichi (Falkenström, 1936) and additional European data. Comparative Cytogenetics 9 (1): 133–144.

Rhampsinitus Re-Redux

I've featured the African harvestman genus Rhampsinitus on this site twice before, but I'm going to have another dive into it today. There's still more I can say about this remarkable genus.

Male Rhampsinitus, possibly R. leighi, copyright Peter Vos. The individual ahead of the male is another Rhampsinitus, probably a female; there's also a short-legged harvestmen beneath the male.


There's more I could say about African phalangiids in general, in fact. There's never been a proper phylogenetic study of the long-legged harvestman family Phalangiidae, so we can't speak with confidence about the relationships between the African members of this group and their relatives elsewhere, but it would not be unexpected if the sub-Saharan phalangiids form an evolutionarily coherent group. Many of the family's most striking exemplars are to be found on the African continent: Cristina with their thick, spiky front legs; sleek, flattened Odontobunus, Guruia with their chelicerae like a pair of jar tongs held in a boxing glove. Rhampsinitus' current position as the best-known African harvestman genus is probably due not only to its diversity but also to its more temperate centre of distribution placing it closer to researchers than these other more equatorial genera.

As mentioned in my first post on the genus, there are currently over forty recognised species of Rhampsinitus. As alluded to in my second post, that number might be expected to change in the future. No reliable identification key is currently available for Rhampsinitus, nor is the information available for many species that would allow such a key to be written. A key to the southern African species was provided by Kauri (1961) but, while I did find this key invaluable when I conducted my own tentative foray into rhampsinitology, I couldn't recommend it to a novice. Kauri was simply unaware of the extreme variation that can be found among male Rhampsinitus belonging to a single species. There are only a handful of species for which both major and minor males have been described and, as I explained previously, minor males may not be identifiable to species without examining genitalia.

Probably a male Rhampsinitus vittatus, copyright Nanna.


This, obviously, is a problem for the handful of species that have been described from what appear to be minor males. Some of these, such as Rhampsinitus fissidens and R. hewittius, are probably doomed to remain mysteries at least until someone redescribes their types. Others may be more recognisable. Rhampsinitus qachasneki is an unusually spiny species described from the mountains of Lesotho, with some of the denticles along the front edge of the body multi-pointed. These distinctive denticles, like repurposed muntjak antlers, might reasonably be expected to be present in any major males of this species, if they exist. The challenge may be even greater for the handle of species that have been described from females. Nevertheless, the known female of R. maculatus, another Lesotho mountain species, has a distinctive spotted colour pattern and thick, remarkably hairy pedipalps that might be expected to show their analogues in the unknown males (again, if they exist: we're kind of glossing over the point that some harvestmen species are known to be parthenogenetic, because harvestmen systematics is so heavily predicated on male genital morphology that the idea of an all-female harvestman species is a trifle intimidating*).

*I assume that this is precisely what Zappa had in mind when he got to the end of Thing-Fish.

Then, of course, there's the persistent question of Rhampsinitus lalandei. This was the first species included in Rhampsinitus in 1879 and as such represents the type or sine qua non of the genus. As was not unusual for the time, its author Eugene Simon was a bit vague about where his original specimen(s) had come from, giving the locality as simply 'Cafrerie'. Cafrerie, rendered in English as Kaffraria or Kaffirland, is a geographical designation that has fallen out of favour these days for reasons I would hope to be obvious, but was commonly used during the 1800s to refer to the area around the eastern coast of modern South Africa, particularly around Port Elizabeth. Unfortunately, Simon's description of R. lalandei is not definitive by modern standards—most of the features described could apply to any number of Rhampsinitus species—and Simon's original specimen appears to have been lost. This presents a problem for any who would suggest that this large genus should be divided up as it might become uncertain which division represents the true Rhampsinitus. Starega (2009) suggested that R. lalandei might be the same as R. crassus, a species definitely found in the Port Elizabeth region. However, it should be noted that Simon described R. lalandei as being irregularly armed with denticles dorsally. In the majority of Rhampsinitus species, the denticles on the opisthosoma form very neat transverse rows, but in others they are a bit more messily placed. Rhampsinitus crassus is one of the former species but the description of R. lalandei suggests it may have been one of the latter. So if anyone's looking at harvestmen from around that area, keep your eyes open.

REFERENCES

Kauri, H. 1961. Opiliones. In: Hanström, B., P. Brinck & G. Rudebeck (eds) South African Animal Life: Results of the Lund University Expedition in 1950–1951 vol. 8 pp. 9–197. Almqvist & Wiksells Boktryckeri Ab: Uppsala.

Staręga, W. 2009. Some southern African species of the genus Rhampsinitus Simon (Opiliones: Phalangiidae). Zootaxa 1981: 43-56.

Gar!

Apart from the mostly terrestrial radiation of the tetrapods, the vast majority of today's bony-skeletoned fishes belong to the clade of the teleosts. Way back in the Triassic, the ancestors of this clade went through a process of modification of the jaw skeleton to make it more mobile and adroit in catching small prey, and this together with a tendency towards the lightening of the skeleton and the body's covering of bony scales marked the beginnings of what is now well over 25,000 species. But while they may pale in comparison to this phylogenetic behemoth, there are still non-teleost (and non-tetrapod) bony fishes out there if you look in the right places.

Alligator gar Atractosteus spatula, copyright Stan Shebs.


Most studies on fish phylogeny in the last decade or so have agreed that the living sister group of the teleosts is the Holostei, a clade including only eight living species. One of these is the bowfin Amia calva, an elongate, cylindrical-bodied fish with a long dorsal fin running most of the length of its back. The other seven sepecies belong to the gar genera Lepisosteus and Atractosteus, forming the family Lepisosteidae*. Gars are also elongate like the bowfin, albeit without the long dorsal fin, and have elongate, flattened jaws (tending to be narrower in Lepisosteus than Atractosteus). The tail fin in both bowfins and gars is rounded, not forked. Living holosteans are restricted to North America (including Central America and the Caribbean)** but fossils show them to have been more widespread in the past. They are mostly found in fresh water; some species may tolerate brackish or even salt water but they do not stay there permanently. Bowfins and gars are able to breathe air directly as well as through their gills (indeed, gars are reported to drown if prevented from coming to the surface for several hours) and can therefore survive in more stagnant waters than many other fish. The bowfin averages about half a metre in length; the smaller gar species are also in this range. The largest species, the alligator gar Atractosteus spatula, reaches at least close to three metres. Larger sizes (up to six metres or more!) have been reported for this species but appear likely to be errors or exaggerations; as noted by one authority, "All fishes shrink under the tape measure" (Grande 2010).

*The incorrect alternative spellings Lepidosteus and Lepidosteidae (as well as Lepidosteiformes) have often appeared in the past.

**References to a supposed Chinese gar have long persisted in the literature, based on a description of a "Lepidosteus sinensis" from 1873. This description was based on a drawing rather than an actual specimen, and it is now thought that the fish depicted was probably a belonid (an unrelated long-jawed teleost) rather than a gar.

Bowfin Amia calva sharing a tank with largemouth basses, copyright Bemep.


Modern holosteans are ambush predators, feeding on other fish or aquatic invertebrates. In general, larger species tend to prefer a diet of fish whereas smaller species focus on invertebrates, but all appear to be happy to take whatever they may, whether alive or dead. The alligator gar has been claimed to attack humans but no such attacks seem to have been authenticated; Grande (2010) stated that "swimmers probably have very little to fear from them". As well as their sheer size, this accusation may have been fueled by the alligator gar's apparent tendency in some areas to hang around wharves scavenging garbage. Neither bowfins nor gars are of high importance as food fish for humans though their size and strength gain them some attraction as sport fish*. An industry for the production and marketing of bowfin roe has arisen in recent years following the decline in availability of caviar from Russian sturgeon species; no such market exists for gar eggs, which are toxic to humans. Historically, the thick armour of scales covering the skin of gars was used by Caribbean Indians for making breastplates while individual scales could be used for arrowheads.

*Grande (2010) quotes Eberle (1990) to the effect that gars have "a poor reputation among anglers, who believe [they] would have been better suited as land dwellers had they been able to stand their own reflections in the water".

Shortnose gar Lepisosteus platostomus, copyright Rufus46.


Reproductive habits are best known in the bowfin and the longnose gar Lepisosteus osseus. Male bowfins construct a nest in mats of fibrous vegetation, into which they attempt to induce passing females to spawn. Guarding of the eggs after spawning is the duty of the male alone; the female moves on, perhaps to spawn in another male's nest (the male himself may also court more females). The eggs are adhesive and take about a week and a half to hatch. Following hatching, the fry attach themselves to nearby vegetation by an adhesive organ at the end of their snout and spend some time being nourished by the remains of their yolk sac beofre beginning to forage. The male will continue to guard his fry until they reach about a month of age. Reproduction in longnose gars is similar in the production of adhesive eggs and the early sessile, snout-attached period of the life cycle, but differs in that there is no nest construction or parental care. There is a record of gar eggs being deposited in the nest of a smallmouth bass and the fry being subsequently raised cuckoo-wise by the nest's owner, but it is unclear whether this reflects any deliberate action by the parental gars or simply a fortuitous accident. Gars take up to six years to reach maturity, with males maturing a couple of years earlier than females.

Semionotus bergeri, copyright Ghedoghedo.


The fossil record of holosteans extends back to their divergence from the teleosts in the early to mid-Triassic, with the bowfin and gar lineages apparently diverging from each other not long afterwards. As noted above, both lineages include a diversity of extinct members that somewhat belies their current paucity, such as Macrosemiidae and Semionotidae in the gar lineage, and Ophiopsidae, Ionoscopidae, Caturidae and Sinamiidae in the bowfin lineage. The greatest diversity in both lineages was during the Jurassic and Cretaceous (Brito & Alvarado-Ortega 2013; Cavin 2010) and the two modern gar genera appear to have been separate lineages at least since the late Cretaceous (Grande 2010). Holosteans were also more ecologically diverse in the past. Masillosteus, a gar genus from the Eocene of Europe and North America, had a shorter jaw and flatter teeth than modern jaws, and probably fed on harder-shelled animals such as molluscs and/or crustaceans. The Mesozoic 'Semionotidae', suggested by Cavin (2010) to be paraphyletic to the gars, were even more diverse, including marine as well as freshwater forms, and forms that may have plant feeders or detritivores. In the early Jurassic of eastern North America, one group of semionotids underwent a lake-based radiation that has been compared to the modern cichlids of African rift lakes. Adequately covering the diversity of fossil holosteans would make this post considerably longer than it already is; perhaps one day, I'll get to it.

REFERENCES

Brito, P. M. & J. Alvarado-Ortega. 2013. Cipatlichthys scutatus, gen. nov., sp. nov. a new halecomorph (Neopterygii, Holostei) from the Lower Cretaceous Tlayua Formation of Mexico. PLoS One 8 (9): e73551.

Cavin, L. 2010. Diversity of Mesozoic semionotiform fishes and the origin of gars (Lepisosteidae). Naturwissenschaften 97: 1035–1040.

Grande, L. 2010. An empirical synthetic pattern study of gars (Lepisosteiformes) and closely related species, based mostly on skeletal anatomy. The resurrection of Holostei. Copeia 2010 (2A): iii–x, 1–871.

Canterbury Bells

Bellflowers or harebells are one of the classic plants associated with the English country garden. For today's post, I'll be covering the family of plants that bellflowers belong to.

Fairy's thimble Campanula cochleariifolia, copyright Jerzy Opioła.


The Campanulaceae are a family of over 2300 plant species found almost worldwide (Crowl et al. 2016). The family is, however, divided between five subfamilies that some authors would treat as separate families, in which case 'Campanulaceae' would be restricted to the 600 or so species of the subfamily Campanuloideae. It is this subfamily that includes the bellflowers. The vernacular name, of course, refers to the shape of the flowers produced by these plants, as indeed does the botanical name: Campanula translates as 'little bell'. These flowers are radiately symmetrical with all petals more or less the same size and shape and evenly arranged in a circle. Other subfamilies of the Campanulaceae in the broad sense, the largest of which is the lobelias of the Lobelioideae, produce more bilaterally symmetrical flowers with petals differing in size and/or with some petals closer together than others. Fruits are most commonly a capsule, with the seeds dispersed by wind, but some lobelioids produce fleshy fruits that attract birds. The lobelioids are most diverse in the southern continents, and it is thought that this may have been the original home of the family as a whole when it arose sometime close to the end of the Cretaceous, possibly in Africa. At some time in the early Cenozoic, however, the campanuloids arrived in and underwent a significant radiation in the Palaearctic. This dispersal may be related to the different flower morphology of the campanuloids, as they adapted from the bird, bat and butterfly pollinators of the tropics to the bee and fly pollinators of more temperate habitats.

Glandular threadplant Nemacladus glanduliferus var. orientalis, copyright Stan Shebs.


The genetics of Campanulaceae, specifically of their chloroplasts, should also not go unnoticed. The structure of the chloroplast genome in plants is usually very stable, with few changes in gene arrangement and order. However, at various points in the history of Campunulaceae, large chunks of foreign DNA have been inserted in the original plastid chromosome, with a number of these insertions also associated with inversions in the direction of adjoining sections of the original genes (Knox 2014). This kind of insertion is unique among flowering plants: changes in the gene content of plastids more usually involve genes being transferred out of the plastid. The source of this extra DNA is uncertain: it may have come from the plant's own nucleus, or it may have come from an as-yet-unknown endosymbiont. Also unknown is the functional significance of these rearrangements, if any. Some insertions have clearly resulted in pseudogenes, with their sequences rapidly breaking down through subsequent genetic drift. But others have preserved the structure of functional genes, suggesting continued selection for their retention.

Cyanea duvalliorum, an arborescent Hawaiian lobeliad, copyright Forest & Kim Starr.


The majority of Campanulaceae are small perennial herbs. Two genera of distinctive enough to be assigned to their own subfamilies include annual herbs: the threadplants Nemacladus of southwestern North America, and the little-known Chilean Atacama desert endemic Cyphocarpus. Some members of the Lobelioideae are woody subshrubs, and at some point one of these woody lobelioids managed to make its way to the Hawaiian archipelago where it gave rise to one of the world's most remarkable insular radiations, and the single largest such radiation in plants. Over 120 species of lobeliads are known from the Hawaiian islands, varying from single-stemmed succulents to straggling vines to trees over 18 metres in height. There are inhabitants of lowland forests, of upland bogs, and of rocky cliffs. There are species producing fruit as dry capsules; others produce fleshy berries. So varied are the Hawaiian lobeliads that previous authors have inferred their origin from multiple seperate colonisations, but a study by Givnish et al. (2009) supported a single origin from a single colonist arriving about thirteen million years ago. This would have been before any of the current major Hawaiian islands existed (the oldest, Kaua'i, is a little less than five million years old); the implication is that the ancestor of the Hawaiian lobeliad arrived on a pre-existing island, perhaps corresponding to the modern Gardner Pinnacles or French Frigate Shoals. As the lobeliads diversified, they continued to disperse onto new islands as they arrived, while their original homeland eroded away.

Sadly, a depressing percentage of the species forming this incredible radiation are now threatened with extinction, the victims of pressures such as loss of habitat, the decline of their pollinators and dispersers, or grazing by introduced mammals. The cliff-dwelling pua 'ala Brighamia rockii of Moloka'i is now restricted to five locations with an estimated total wild population of less than 200 individuals. A related species on Kaua'i, the olulu Brighamia insignis, may be extinct in the wild, having last been recorded in the form of a single individual in 2014 (it still survives in cultivation). As we earlier saw with the Hawaiian honeycreepers, there is barely a single section of the Hawaiian biota not marked by tragedy.

REFERENCES

Crowl, A. A., N. W. Miles, C. J. Visger, K. Hansen, T. Ayers, R. Haberle & N. Cellinese. 2016. A global perspective on Campanulaceae: biogeographic, genomic, and floral evolution. American Journal of Botany 103 (2): 233–245.

Givnish, T. J., K. C. Millam, A. R. Mast, T. B. Paterson, T. J. Theim, A. L. Hipp, J. M. Henss, J. F. Smith, K. R. Wood & K. J. Sytsma. 2009. Origin, adaptive radiation and diversification of the Hawaiian lobeliads (Asterales: Campanulaceae). Proceedings of the Royal Society of London Series B—Biological Sciences 276: 407–416.

Knox, E. B. 2014. The dynamic history of plastid genomes in the Campanulaceae sensu lato is unique among angiosperms. Proceedings of the National Academy of Sciences of the USA 111 (30): 11097–11102.