Field of Science

Lilies of Blood

The flora of southern Africa is renowned for being remarkably diverse and, in many cases, remarkably eye-catching. The region is home to more than its fair share of ornamental plants, many of which have become popular garden subjects. Among the remarkable members of the southern African flora are the blood lilies of the genus Haemanthus.

Haemanthus coccineus, copyright Peter Coxhead.

Haemanthus is a genus of 22 known species found in the very southern part of the continent, in the countries of South Africa and Namibia (species from further north that have historically been included in Haemanthus are now treated as a separate genus Scadoxus). It is a member of the belladonna family Amaryllidaceae and, like many other members of that family, grows as a herb from a fleshy bulb that is partially or entirely concealed underground. The plant above ground may be annual or persistent, depending on species. Each individual Haemanthus plant produces very few leaves at a time: two is the most common number (Van Jaarsveld 2020). The leaves are more or less fleshy, often hairy, and may be directed upwards or spread outwards.

In those species that shed their leaves, flower stalks are produced before the next season's leaves appear, in a similar matter to the related naked ladies Amaryllis belladonna. Flowers are produced in dense umbels, subtended by bracts that are often brightly coloured, so at a glance the inflorescence of some species might be taken for a single large flower up to ten centimetres in diameter. Depending on the species, the supporting stalk may vary from over a foot in height to only a few centimetres. The first species to be described bear flowers of a bright red colour, explaining both the genus and vernacular names, but flowers may also be pale pink or white. Species that lack the red colour may be referred to as 'paintbrush lilies' rather than 'blood lilies'. Fruits are soft fleshy berries.

Haemanthus albiflos, copyright Krzysztof Ziarnek, Kenraiz.

Phylogenetic analyses of the genus have identified two major clades, a mostly eastern clade found in regions with summer rainfall and a mostly western clade associated with winter rainfall. A notable outlier is the eastern summer-rainfall species H. montanus which is the sister taxon to the winter rainfall clade. Members of the summer-rainfall clade have white or pale pink flowers; members of the winter-rainfall clade have pale pink to dark red flowers. Members of both clades have been grown as pot plants for their unusual appearance though the scent of the flowers is not regarded as pleasant. Perhaps the most widely grown species is H. albiflos, a species native to both the western and eastern parts of South Africa that bears flowers in umbels up to seven centimetres wide. This species is evergreen, carrying its leaves year-round.


Van Jaarsveld, E. 2020. Haemanthus. In: Eggli, U., & R. Nyffeler (eds) Illustrated Handbook of Succulent Plants: Monocotyledons 2nd ed. pp. 441–443. Springer.

The Oligorhynchiidae

Dorsal view of Oligorhynchia subplana gibbosa, from Cooper (1935).

From Oligochiton, we move onto Oligorhynchia. The Oligorhynchiidae are a family of very small brachiopods known from the Middle and Late Ordovician. They were among the earliest representatives of the Rhynchonellida, a major group of brachiopods that survives to the present day. Rhynchonellidan shells are usually characterised by a strong beak associated in life with a well-developed pedicel. In oligorhynchiids, this beak is suberect and the shell as a whole is an elongate subtriangular shape. The valves of the shell are folded into coarse plicae (ridges). At least towards the base of the shells, the major folds are in what is called an inverted arrangement, with a ridge in the dorsal valve matched by a valley in the ventral valve (Schmidt & McLaren 1965). Other structural features defining the group include small plates projecting into the pedicel opening, distinct vertical dental plates and divided hinge plates in the valve articulation, and the usual absence of a median septum or cardinal process inside the shell (Savage 1996).

The oligorhynchiids first arose in the east of what was then the continent of Laurentia (corresponding to modern North America). They subsequently spread across the Iapetus Ocean to the continents of Baltica and Kazakhstan (Jin 1996). The end of the Ordovician saw their replacement by other rhynchonellid families. Nevertheless, their genetic lineage would continue for some time yet as they have been identified as ancestors of later families: the Trigonirhynchiidae and Camarotoechiidae (Jin 1989). The brief oligorhynchiid spark would blossom into later rhynchonellid success.


Jin, J. 1989. Late Ordovician–Early Silurian rhynchonellid brachiopods from Anticosti Island, Quebec. Biostratigraphie du Paléozoïque 10: 1–127, 130 pls.

Jin, J. 1996. Ordovician (Llanvirn–Ashgill) rhynchonellid brachiopod biogeography. In: Copper, P., & J. Jin (eds) Brachiopods pp. 123–132. CRC Press.

Savage, N. M. 1996. Classification of Paleozoic rhynchonellid brachiopods. In: P. Copper, & J. Jin (eds) Brachiopods pp. 249–260. CRC Press.

The Fate of Oligochiton

Chitons are one of the most distinctive and evolutionarily divergent groups of molluscs alive today. But compared to other groups of molluscs, the fossil record of chitons is rather sparse—or at least sparsely studied. It's not hard to see why. The multi-plated nature of the chiton shell means that it tends to fall apart after death, and the structure of the plates is such that critical features are easily abraded.

(Clockwise from top left) head, intermediate and tail valves of Lepidochitona lioplax, from Dell'Angelo et al. (2011).

Lepidochitona lioplax is one example of a fossil chiton. It was originally described from Oligocene rocks belonging to the Sooke Formation of southern Vancouver Island in British Columbia. Only four moderate-sized valves were initially identified: one head valve, one intermediate, and two tails (so at least two individuals were involved). The valves had a smooth outer surface without a strong distinction in appearance between the central and lateral areas. The insertion plates (lateral projections of the lower surface of the valves that in life anchor them into the surrounding girdle) were very short. The sutural laminae (anterior projections of the lower surface of the intermediate and tail valves that articulate with the valve in front) were low, wide, and divided in the middle by a broad shallow surface. Slits in the lateral insertion plates were numerous, with several in the tail valves and probably two or three on each side in the intermediate valves (Smith 1960). When first described, this species was thought distinct enough to belong in its own genus Oligochiton.

Oligochiton lioplax would then go little reported on until 2011 when Dell'Angelo et al. described an assemblage of chiton fossil from the latest Eocene or early Oligocene of the Lincoln Creek Formation in Washington State. Specimens of lioplax were relatively numerous in this collection and Dell'Angelo et al. were able to examine close to a hundred valves. Their observations would lead to something of a downgrade in the species status. Rather than deserving its own extinct genus, Dell'Angelo et al. felt that lioplax could be comfortably accommodated in the living genus Lepidochitona. Its smooth valves are unusual within Lepidochitona but not unique. The supposed multiple slits in the sides of the valves did not stand up to scrutiny. Instead, intermediate valves of L. lioplax bore only a single slit on each side, in line with other Lepidochitona species. The original inference of multiple slits was an error due to the original specimen being still partially embedded in the surrounding matrix.

Lepidochitona lioplax is one of the earliest known representatives of its genus but its exact significance is obscure. It has been suggested as a direct ancestor of the modern subgenus Spongioradsia but this, again, was based on the supposed slits in the intermediate valves that Dell'Angelo et al. refuted. To know how L. lioplax connects to the big picture of Lepidochitona evolution, we would probably need a better picture of Lepidochitona evolution overall.


Dell'Angelo, B., A. Bonfitto & M. Taviani. 2011. Chitons (Polyplacophora) from Paleogene strata in western Washington State, U.S.A. Journal of Paleontology 85 (5): 936–954.

Smith, A. G. 1960. Amphineura. In: Moore, R. C. (ed.) Treatise on Invertebrate Paleontology pt I. Mollusca 1: Mollusca—General Features, Scaphopoda, Amphineura, Monoplacophora, Gastropoda—General Features, Archaeogastropoda and some (mainly Paleozoic) Caenogastropoda and Opisthobranchia pp. I41–I76. Geological Society of America, and University of Kansas Press.

The Feared Mosquito

It's one of those standard pub-quiz "trick" questions. What animal kills the most people? The hope is that contestants will nominate the 'obvious'—snakes, sharks, bears, whatever—before being blind-sided by the revelation that mosquitoes kill over a million people. They don't kill them directly, of course; their victims die from the diseases they spread*. The statistic also glosses over the point that there are many hundreds of species of mosquito that vary significantly in the nature and severity of their role as disease vectors. Nevertheless, for this post I'm considering the group that includes some of the most notorious vectors: the genus Anopheles.

*For the record, if the question was confined to active killings, the most dangerous animal to humans is other humans. Dogs come a distant second.

Anopheles punctipennis feeding, with the long palps extended in front of the head, copyright Nathan D. Burkett-Cadena/University of Florida.

Anopheles is one of the most divergent genera of mosquitoes, being placed in a distinct subfamily Anophelinae (along with a couple of small related genera) from the bulk of mosquitoes in the subfamily Culicinae. Adult Anopheles can be readily distinguished from culicine mosquitoes by their palps which are about as long as the proboscis (in other mosquitoes, the palps are distinctly shorter). Larvae of Anopheles lack the long respiratory siphons at the end of the abdomen found in other mosquito larvae so they rest parallel with the water surface rather than hanging below it. The genus is found around the world; over 450 named species are currently known (Harbach 2013) with many more waiting to be described. The genus is currently divided between seven subgenera though one of the largest of these, the cosmopolitan subgenus Anopheles, is not monophyletic. The remaining subgenera are better supported with the largest of these, Cellia, being found in the Old World. Between them, the subgenera Anopheles and Cellia account for over 400 of the known Anopheles species. The remaining small subgenera are mostly Neotropical with a single Oriental species being awarded its own subgenus.

Anopheles maculipennis, copyright Ryszard.

Anopheles is of most concern to humans, of course, for its role as a disease vector. As with other mosquitoes, the transmission of disease is done entirely by females taking blood meals to provide nutrients for their developing eggs. Males are not blood feeders, instead feeding entirely on sugar sources such as nectar (females also feed on nectar for their own nutrition). The main disease spread by Anopheles is malaria, but they may also spread malaises such as filariasis and arboviruses (Krzywinski & Besansky 2003). As noted above, species may vary significantly in their importance as disease vectors, even between quite closely related taxa. Many historically recognised vector "species" have proved, on close inspection, to represent species complexes of which some may be vectors and others not. For instance, one of the most important transmitters of malaria, the African A. gambiae, has been divided between at least eight different species (Coetzee et al. 2013). Misidentification of vectors can be a significant issue. For instance, mosquito control regimes in central Vietnam during the 1990s focused on two species, A. dirus and A. minimus, that were each active at different times of year. However, Van Bortel et al. (2001) found that A. minimus was in fact very rare in this area, with specimens previously thought to be A. minimus proving to be another species, A. varuna. Anopheles varuna is not a significant malaria vector, feeding almost entirely on animals such as cattle rather than on humans. Large amounts of resources would have been wasted trying to control a mosquito that was of little concern. What is more, the fact that malaria was not being transmitted by A. minimus raises the possibility that it was being spread by yet another species, one that had managed to escape attention. Remember, kids: bad taxonomy kills.


Coetzee, M., R. H. Hunt, R. Wilkerson, A. Della Torre, M. B. Coulibaly & N. J. Besansky. 2013. Anopheles coluzzii and Anopheles amharicus, new members of the Anopheles gambiae complex. Zootaxa 3619 (3): 246–274.

Harbach, R. E. 2013. The phylogeny and classification of Anopheles. In: S. Manguin (ed.) Anopheles Mosquitoes: New insights into malaria vectors. InTechOpen.

Krzywinski, J., & N. J. Besansky. 2003. Molecular systematics of Anopheles: from subgenera to subpopulations. Annual Review of Entomology 48: 111–139.

Van Bortel, W., R. E. Harbach, H. D. Trung, P. Roelants, T. Backeljau & M. Coosemans. 2001. Confirmation of Anopheles varuna in Vietnam, previously misidentified and mistargeted as the malaria vector Anopheles minimus. American Journal of Tropical Medicine and Hygiene 65 (6): 729–732.

The Running of the Termites

I don't know how many people would profess to have a favourite genus of termites. Which is a shame, because there are some real stand-out examples. Snapping termites, magnetic termites, glue-spraying termites... For my own part, though, I have a particular fondness for the Australian harvester termites of the genus Drepanotermes.

Soldiers and workers of Drepanotermes perniger, copyright Jean Hort.

Nearly two dozen species of Drepanotermes are found on the Australian continent to which they are unique (Watson & Perry 1981). They are arid-environment specialists, being most diverse in the northern part of Australia. My reasons for being so fond of them are, I'll admit, decidedly prosaic. The worker caste of most termite species is very difficult if not impossible to identify taxonomically; one termite worker usually looks very much like another. Drepanotermes workers, however, are different. The name Drepanotermes can be translated as "running termite" and, as befits their name, Drepanotermes of all castes stand out for their distinctly long legs. Soldiers of Drepanotermes also have distinctively shaped mandibles which are sickle-shaped and have a single projecting tooth on the inner margin. They are similar to soldiers of the related genus Amitermes (of which Drepanotermes may represent a derived subclade) but the mandibles of Amitermes tend to be straighter and more robust.

The long legs of Drepanotermes reflect their active harvester lifestyles. Workers will emerge from the nest at night in search of food to carry back home. In the red centre of Australia they will primarily collect spinifex; they will also take fallen leaves, tree bark and the like. Soldiers keep guard while the workers forage. I've found them clustered around a nest entrance of an evening, just their heads poking out to snap at passers-by. Workers may wander up to about half a metre from the nest entrance as they forage. The concentrations of vegetable matter produced by Drepanotermes storing food sources in their nest may form a significant factor in the nutrient profile of areas where they are found.

Alate and soldiers of Drepanotermes rubriceps, copyright Jean Hort.

Depending on species and circumstance, the nests of Drepanotermes may be mounds or entirely subterranean with the latter being the majority option. They prefer compact soils such as clay though they may burrow through looser soils where there is a denser subsoil. Drepanotermes may construct their own nest or move into nests constructed by other termites. One aptly named species, D. invasor, seems to take over pre-existing nests more often than not. Subterranean nests are arranged as a series of chambers about five to ten centimetres in diameter connected by tunnels. These chambers may be arranged vertically, one below another, or they may form a rambling transverse network. Above ground, subterranean nests may be visible as an open circle devoid of vegetation. The ground in these circles is hard as concrete and may remain clear for decades after the actual nest has gone. Walsh et al. (2016) refer to the remains of nests protruding above ground along vehicle tracks after the soil around them has worn down. Local people have a long history of taking advantage of the open space offered by termite nests, such as to move more easily through scrub or as resting or working places.

The alate castes of Drepanotermes tend to be poorly known. Indications are that mature reproductives spend little time in the parent nest before leaving to breed. For most species, breeding flights take place in late summer. Alates may emerge either by day or night. The time of emergence seems to depend on the species; night-flying alates have distinctly larger eyes than day-fliers. Unfortunately, because alates have rarely been collected in association with a nest, we are largely still unable to tell which alates belong to which species.


Walsh, F. J., A. D. Sparrow, P. Kendrick & J. Schofield. 2016. Fairy circles or ghosts of termitaria? Pavement termites as alternative causes of circular patterns in vegetation of desert Australia. Proceedings of the National Academy of Sciences of the USA 113 (37): E5365–E5367.

Watson, J. A. L., & D. H. Perry. 1981. The Australian harvester termites of the genus Drepanotermes (Isoptera: Termitinae). Australian Journal of Zoology, Supplementary Series 78: 1–153.

Sparrows of the West

Recent decades have seen significant shifts in the classification of birds, particularly among the Passeriformes, the perching birds. These shifts have lead to the recognition of a number of major groups that were previously obscured. Among these recent elevations are the New World sparrows of the Passerellidae.

Gambel's white-crowned sparrow Zonotrichia leucophrys gambeli, copyright Gregory Smith.

The New World sparrows are part of a broader radiation known as the nine-primaried songbirds, along with such luminaries as finches, tanagers, and their Old World namesakes. The name 'nine-primaried' refers to the number of well-developed primary feathers (the long outer ones) in the wings; most other perching birds have ten distinct primaries. Though the nine-primaried songbirds have long been recognised as a coherent group, there has been a lot of disagreement over their subdivision. Historically, these subdivisions were strongly influenced by different bill shapes representing different diet specialisations, but recent molecular phylogenies have demonstrated that bill shape is more labile than previously recognised. The New World sparrows were usually regarded previously as a subgroup of the generalist seed-eating family Emberizidae, along with the buntings of the Old World, but molecular phylogenies have asserted the division between the hemispheres. Not all New World representatives of the old Emberizidae have shifted to the Passerellidae: a significant component of the Neotropical fauna (including the finches of the Galapagos islands) have instead proven to be closer to the fruit-eating tanagers of the Thraupidae. As currently recognised, the passerellids are a fairly coherent group of about 140 species distributed around North and South America.

Goldwn-winged sparrow Arremon schlegeli schlegeli, copyright Nick Athanas.

In general, the passerellids are small birds with simple, conical bills. Most are dull brownish in coloration though many are strikingly patterned, particularly around the head. Some are more distinctive: the South American sparrows of the genus Arremon often stand out as particularly colourful. Most passerellids are fairly retiring in their usual habits, foraging at or close to ground level. As noted before, they are mostly generalist feeders. Their short bills are excellently suited for milling the small seeds which make up a large part of their diet. However, they will also take insects and other small invertebrates. One widespread North American species, Ammodramus savannarum, has earned the vernacular name of "grasshopper sparrow" as a result. Notable outliers dietwise are the Neotropical bush-tanagers of the genus Chlorospingus which are primarily berry feeders. These largely greenish birds were previously classified with the Thraupidae as a result before molecular data led to their reassignment.

Common bush-tanager Chlorospingus flavopectus, copyright Becky Matsubara.

Whereas Neotropical members of the Passerellidae are mostly sedentary, North American species are often migratory, moving north with the approach of summer. However, migration is commonly related to environmental conditions. A number of species are migratory in the northern parts of their range but may be found in their southern territories year-round. In some species, migrating populations will leap-frog over resident populations, moving further south than any resident individuals during the winter months. Many passerellid species are strong singers and courting males will often select an exposed branch to sing from in contrast to their usual skulking habits. Other species, particularly those inhabiting open habitats where trees and shrubs are in short supply, may have prominent aerial displays. Males of one of these latter species, the lark bunting Calamospiza melanocorys, moult during the breeding season into black plumage with contrasting white patches on the wings and tail. During the remainder of the year, they are dull in coloration like their females. Nesting is conducted close to ground level like feeding with the nest often being a small cup in the ground concealed under vegetation. Where breeding has been studied in detail, passerellids are commonly what has been called "socially monogamous". Males and females will form what appear to be monogamous pairs with one male remaining close to one female (though construction of the nest and incubation are done by the female alone). However, genetic studies on nestlings have found that chicks are not uncommonly not the child of their apparent 'father', indicating that females have not remained faithful to their mate.

Yellow-striped brush-finch Atlapetes citrinellus, copyright Ron Knight.

Prior to molecular studies, authors had suggested a possible division of North American passerellids between two evolutionary lineages based on ecology and behaviour, the grassland and brushland sparrows. A molecular study of passerellids by Klicka et al. (2014) identified eight well-supported clades within the family. Two further species, the large-footed finch Pezopetes capitalis of Central America and the Zapata sparrow Torreornis inexpectata of Cuba, were not robustly assigned to a clade. Identified relationships were comparable to but not entirely congruent with prior hypotheses. For instance, most 'brushland sparrows' (of the genera Passerella, Zonotrichia and Junco) belonged to a single clade but the remaining 'brushland' genus Melospiza was placed in a clade mostly made up of 'grassland' species. The diverse South American genus Arremon was supported as monophyletic but others were not. In particular, the North American Ammodramus was divided between two widely separated clades. This lead to the resurrection of the genus Ammospiza for a group of saltmarsh-breeding species. Deeper relationships within the family deserve further investigation.


Hoyo, J. del, A. Elliott & D. A. Christie (eds) 2011. Handbook of the Birds of the World vol. 16. Tanagers to New World Blackbirds. Lynx Edicions: Barcelona.

Klicka, J., F. K. Barker, K. J. Burns, S. M. Lanyon, I. J. Lovette, J. A. Chaves & R. W. Bryson, Jr. 2014. A comprehensive multilocus assessment of sparrow (Aves: Passerellidae) relationships. Molecular Phylogenetics and Evolution 77: 177–182.

Steatocranus gibbiceps, the Rapid River Bumphead

The cichlid fishes of the Great Lakes of Africa are rightly renowned as one of the world's most spectacular species radiations. Hundreds of species, occupying a wide range of ecological niches, have evolved in what is, geologically speaking, a short period of time. However, cichlids in Africa are not a phenomenon of the Great Lakes alone and many interesting species may be found in other parts of the continent, some of them belonging to local radiations of their own. Consider, for instance, the Congo River rapids endemic Steatocranus gibbiceps.

Male Steatocranus gibbiceps, copyright Polypterus.

The Congo is one of the largest African rivers with a drainage basin covering one-eighth of the continent (Schwarzer et al. 2011). Downstream of Kinshasa, the river gets funneled into an intermittently deep, narrow channel for a distance of some 300 km before broadening as it approaches the sea. The result is the world's longest stretch of river rapids. Many fish species are found only in this unique region of fast-flowing waters, among them multiple species of the cichlid genus Steatocranus including S. gibbiceps. The genus as a whole is restricted to the Congo basin; a single species previously recognised from the Volta River has since been transferred to its own genus (Weiss et al. 2019). The names Steatocranus and gibbiceps both basically mean the same thing: 'fat head', in reference to a fleshy swelling atop the fish's noggin. The exact size of this swelling varies between individuals, being most prominent in large males. Vernacular names given to Steatocranus species generally reflect this feature, such as bumphead cichlid or buffalo-head cichlid. Half a dozen species have been named within Steatocranus with several more being recognised but not yet formally described, most of them belonging to the radiation within the rapids. Schwarzer et al. (2012) found evidence for extensive historical cross-breeding between species and suggested that hybridisation may have been a significant factor in the genus' diversification.

Steatocranus gibbiceps is the largest species in this genus of moderately-sized fishes, growing up to about nine centimetres in length (Roberts & Stewart 1976). Its fast-current habitat is reflected in its slender body form. It is olive brown in coloration with the scales being light in colour at the centre and darker around the margins. Steatocranus gibbiceps is most clearly distinguished from other described species in its genus by its teeth: the front teeth of both the upper and lower jaws are conspicuously large and truncate. It also has a shorter gut than its congeners. This species appears to be specialised in feeding on freshwater snails which it scoops up and swallows whole, though it will take a broader range of food in captivity. Other species of Steatocranus mostly feed on algae.

Steatocranus species are not buoyant and tend to sit at the bottom of the water (Chase Klinesteker describes their behaviour as 'hopping around the bottom like a goby'). They escape the current by spending time in the hollows and crevices among rocks. Breeding happens within such hollows with dedicated pairs forming and females affixing their eggs to the rocks. Like many other cichlids, Steatocranus gibbiceps are dedicated parents after the eggs hatch. Tending of the fry is mostly the responsibility of the female while the male patrols the territory on the watch for danger. In this way, the baby bumpheads are given the best possible start at life.


Roberts, T. R., & D. J. Stewart. 1976. An ecological and systematic survey of fishes in the rapids of the lower Zaïre or Congo River. Bulletin of the Museum of Comparative Zoology 147 (6): 239–317.

Schwarzer, J., B. Misof, S. N. Ifuta & U. K. Schliewen. 2011. Time and origin of cichlid colonization of the lower Congo rapids. PLoS One 6 (7): e22380.

Schwarzer, J., B. Misof & U. K. Schliewen. 2012. Speciation within genomic networks: a case study based on Steatocranus cichlids of the lower Congo rapids. Journal of Evolutionary Biology 25: 138–148.

Weiss, J. D., F. D. B. Schedel, A. I. Zamba, E. J. W. M. N. Vreven & U. K. Schliewen. 2019. Paragobiocichla, a new genus name for Gobiochromis irvinei Trewavas, 1943 (Teleostei, Cichlidae). Spixiana 42 (1): 133–139.