December 02, 2022

The possible adaptive value of high eyes and ears in the grey rhebok (Pelea capreolus)

The eyes and ear pinnae of ungulates vary in placement, according to habitat and anti-predator strategy.

In the case of the eyes, there is variation in how far to the sides, and how high on the head, they are placed.

In the case of the ear pinnae, there is variation in how high on the head they are placed, and in whether they are oriented upright, sideways or downwards.


An extreme example of lateral placement of the eyes is the pronghorn.

The opposite extreme is Myotragus balearicus (, which evolved on small islands, virtually free of predators.

Turning to the placement of the eyes high on the head, a peculiar example is Hippopotamus amphibius ( The extinct Hippopotamus gorgops was even more specialised ( and

However, there is a similar configuration in warthogs (Phacochoerus spp.,, which are the suiforms with the least affinity for water.

Warthogs live on land, but have a short neck and legs (, and kneel while foraging ( Thus, they risk not spotting danger above the grass.

In their own way, warthogs are unique in having the eyes placed above the 'horns' (


Among wild spp., an extreme example of low placement is Syncerus - in which the function of the ear pinnae seems to be for thermoregulation as much as hearing (see

However, the most extreme species is a domestic one, viz. Bos indicus.


The grey rhebok (Pelea capreolus) emerges as having

  • possibly the most upright ear pinnae of any ungulate,
  • unusually - albeit not extremely - high-placed eyes, and
  • eyes as laterally-placed as in any bovid.

Please see

How can the combination of high eyes and extremely upright ear pinnae in the grey rhebok be explained, in terms of adaptation?

It is true that the habitat of the grey rhebok is generally open, with low vegetation ( This would seem to make high placement/orientation of the sense-organs redundant.

Furthermore, the grey rhebok partly coexists with Damaliscus pygargus pygargus (, one of the most conspicuous of bovids, presumably in adaptation to open environments.

However, an apparent paradox is that the grey rhebok - in complete contrast to D. p. pygargus - has extremely inconspicuous (cryptic) colouration ( and

This is partly explained by the facts that the grey rhebok

The combination of features peculiar to the grey rhebok can be summarised as follows, relative to reedbucks:

I suggest that this peculiar combination is consistent with the habitats and behaviour peculiar to the grey rhebok. Whereas reedbucks tend to be adapted to the tall grass of seasonal marshes, the grey rhebok is dually adapted to mature fynbos and short vegetation.

Also see

Posted on December 02, 2022 21:48 by milewski milewski | 7 comments | Leave a comment

December 01, 2022

Geophytes and other tuberous plants in woodland of Eucalyptus wandoo in southwestern Australia

The southwestern region of Western Australia, with its mediterranean-type climate, is generally flat, and nutrient-poor.

The land surface is ancient, with the most extensive laterite ( and and of any land on Earth at temperate latitudes.

However, there is a minor 'scarp' east of Perth, called the Darling Range (, in association with which some of the slopes are steep enough for the lateritic profile to have been partly eroded (

This has slightly rejuvenated the otherwise senile substrates, allowing a limited occurrence of small, herbaceous plants otherwise unable to compete with the sclerophyllous shrubs typical of the exceptionally phosphorus-poor laterite and siliceous sand (

On the slightly nutrient-rejuvenated substrates, the vegetation includes woodland of Eucalyptus wandoo ( and and The substrate is typically sandy loam over kaolinitic clay.

In this Post, I have compiled a list of the geoohytes in one study area of wandoo woodland, based on a report written for Worsley Alumina Pty Ltd. Perth (1985) Worsley Alumina Project, Flora and fauna studies, phase two. 348 pp.

I took trouble to distinguish between

  • true geophytes,
  • members of the same genera that do not qualify as geophytes, and
  • tuberous plants, of different genera, that do not qualify as geophytes.



Burchardia multiflora
in gulley

Wurmbea dioica
in gully

Wurmbea tenella
on ridge


Caesia parviflora

Chamaescilla corymbosa
in gully

Tricoryne humilis


Dichopogon capillipes
on ridge

Sowerbaea laxiflora
in gully

Thysanotus patersonii


Pauridia occidentalis
in gully

Orchidaceae (partial list):

Caladenia sericea

Cyrtostylis reniformis
in gulley

Pterostylis vittata


Ptilotus manglesii
tuberous taproot


Eryngium pinnatifidum
tuberous taproot
in gulley


Stylidium petiolare
ovoid corm, 0.15 cm diam.


Lagenophora huegelii
numerous root-tubers

Trichocline spathulata
adventitious root-tubers

TUBEROUS but apparently not geophytic:

Thysanotus tenellus

Thysanotus thyrsoideus and


Drosera gigantea
in gully, where seasonally wet
tubers of diam. up to 3.8 cm, as deep as 1 m

Drosera menziesii
seasonally moist and swamp margins

Drosera pallida

NOT GEOPHYTIC/TUBEROUS, despite belonging to the genera/families containing geophytes:

Haemodorum laxum

Platysace juncea
(tuberous taproot, but hemicryptophyte instead of geophyte)

Drosera leucoblasta

Stylidium amoenum
Stylidium affine
Stylidium caricifolium (not usually geophytic although possessing numerous root-tubers)
Stylidium dichotomum
Stylidium diversifolium
Stylidium junceum
Stylidium piliferum
Stylidium schoenoides
Stylidium uniflorum

Craspedia uniflora
(adventitious root-tubers, in gully)

Pelargonium littorale
(not tuberous)

Oxalis corniculata
(not tuberous)

Clematis pubescens
root-tuberous, but foliage perennial

Tuberous with no possibility of geophytic affinity:

Macrozamia riedlei

Posted on December 01, 2022 22:54 by milewski milewski | 7 comments | Leave a comment

Liliaceae and Melanthiaceae: geophytes present in California but not South Africa

South Africa is extraordinarily rich in geophytes, belonging to a bewildering array of families and genera.

However, these exclude two families of geophytes important in California, where the climates are similar to those of Western Cape province.

I refer to Liliaceae and Melanthiaceae.



tunicated bulbs (some spp. lack bulbs)

Calochortus catalinae



bulb of one or more fleshy scales, with or without rice-grain bulblets

Fritillaria biflora
bulb 1.5-2 cm diam., of a few fleshy scales

Fritillaria affinis
bulb of a few scales, and many 'rice-grain' bulblets


scaly bulbs or scaly rootstocks

Lilium humboldtii
bulbs ovoid, oblique 0.5-1.5 cm long


Toxicoscordion fremontii
bulb 3-6 cm long

Toxicoscordion venenosum
tunicated bulb

Posted on December 01, 2022 05:48 by milewski milewski | 1 comment | Leave a comment

November 30, 2022

Non-convergence among orchids in mediterranean-type climates in Australia and California, in the evolution of geophytes

@arethusa @catullus @tonyrebelo @ludwig_muller @jeremygilmore @botaneek @troos @benjamin_walton @afid @otes @grnleaf @bushbandit @meta4 @reiner @em_lamond @felix-insects

Geophytes are defined as perennial, deciduous, herbaceous plants which die down to ground level after each growing season, regenerating vegetatively, year after year.

This regeneration is by means of underground storage organs (, i.e. tubers, derived from either stems or roots.

According to the Raunkiaer classication, geophytes are terrestrial cryptophytes (

In southern and southwestern Australia,

See and and and

The floristic richness of orchids in Australia, under mediterranean-type climates, is remarkable: hundreds of spp. in more than a dozen genera (

Based on the above, it seems reasonable to assume that the geophytic growth-form reflects the seasonality of mediterranean-type climates, viz. those with warm dry summers and cool wet winters.

Hence, it would also seem reasonable to predict that the incidence of geophytic orchids would be similar in California, most of which has a mediterranean-type climate.

However, orchids are not mentioned in the abstract of There are many geophytes in California, but do these include orchids?

With this question in mind, I have attempted to assess the incidence of underground storage organs in orchids in California (

The following is a full list of the genera of orchids in California, according to iNaturalist.

Goodyera oblongifolia
(habitat: coniferous forests at high altitudes)
fleshy creeping rhizome
hemicryptophytic but not not geophytic

fibrous roots
(habitat: forests and dry, open hillsides)
not geophytic

Calypso bulbosa
bulb-like corm

'bulbous caudex' (e.g. and; ?root-tubers in some spp. (descriptions unclear/contradictory)
some possibly geophytic (e.g. and, but descriptions ambivalent

rhizomes absent
(habitat: marshy meadows, including at high altitudes)
Are any species geophytic in California?

creeping, fleshy rhizomes with offshoots
(habitat: streambanks)
not geophytic

fibrous creeping roots and/or rhizomes
(habitat: moist, shady places in forests)
not geophytic

Malaxis brachypoda
solid bulb ('pseudobulbous')
(habitat: moist, high altitudes)

Cephalanthera austiniae
rhizomes, not tubers
(habitat: dense forests at high altitudes)
non-green, myco-heterotrophic (
not geophytic

Zeuxine strateumatica
fleshy, creeping, above-ground rhizome, anchored by fibrous roots
possibly geophytic

coral-shaped rhizomes
myco-heterotrophic, most spp. leafless and rootless
not geophytic


What emerges is that the flora of geophytic orchids in California, even under the mediterranean-type climate, is negligible.

Calypso bulbosa and Platanthera yadonii do seem to qualify, but their categorisation as geophytes might not even occur to anyone, were it not for the search-image arising in the flora of Australia (and, to a lesser degree, southern Africa).

Furthermore, the few geophytic orchids in California - if any - that might correspond to the Australian category, in having root-tubers, are one or more species of Platanthera ( Even in this case, it remains unclear that any tubers are actually derived from roots, and that both the foliage and the inflorescence/infructescence die back during the summer.

What emerges is one of the most categorical ecological/evolutionary differences yet found among the various regions of mediterranean-type climate on Earth: a 'world-class' proliferation of geophytic, root-tuberous orchids in the case of Australia, compared with the virtual absence of such plants in the case of North America.


The genus Spiranthes is shared between California and Australia. It is geophytic in Australia, and at least one species may be geophytic in California. However, note that the genus has hardly penetrated the mediterranean-type climate in Australia.

The genus Zeuxine is also shared between these two continents ( However, in this case the climatic difference is even greater than in Spiranthes.

Posted on November 30, 2022 20:58 by milewski milewski | 6 comments | Leave a comment

Geophytic flora of Wongan Hills Nature Reserve, southwestern Western Australia and

Kenneally 1977

Mediterranean-type climate
Mean annual rainfall 390 mm

Total vascular flora 409 spp.

Geophytic flora 33 spp. (= 8% of flora, consisting mainly of orchids)

Pauridia occidentalis 'ephemeral' (probably actually geophytic), damp areas of gulley, in woodland of Eucalyptus loxophleba over Acacia acuminata


Arthropodium capillipes 'ephemeral' (probably actually geophytic), damp red soil in woodland of Eucalyptus loxophleba over Acacia acuminata

Thysanotus patersonii widespread

Thysanotus sp. rare, hill slope


Caesia parviflora 'ephemeral' (probably actually geophytic), common in damp areas, in woodland of Eucalyptus loxophleba over Acacia acuminata and

Chamaescilla corymbosa damp red soil

Bulbine semibarbata 'ephemeral' (probably actually geophytic), red clay, in woodland of Eucalyptus loxophleba over Acacia acuminata


Wurmbea tenella damp ground

Wurmbea sp. red clay over greenstone rock


18 spp. occur here.
all occur on lateritic substrates, in kwongan vegetation, as well as alluvial soils in the gulley, and in woodland of Eucalyptus salmonophloia
Caladenia 8 spp.
Diuris 1 sp.
Eriochilus 1 sp.
Lyperanthus 1 sp.
Pterostylis 6 spp.
Thelymitra 1 sp.


Stylidium caricifolium laterite

Platysace cirrosa 'tuberous climber', tubers (diam. 5 cm) edible, on red soil of apron of breakaway
Platysace maxwellii kwongan


Drosera aff bulbosa greenstone scree
Drosera macrantha laterite
Drosera macrophylla damp red soil
Drosera menziesii kwongan
Drosera aff stolonifera kwongan
Drosera zonaria kwongan


Ptilotus spathulatus prostrate herb, common on alluvial substrates (is this geophytic?)

On yellow sand, there is tall shrubland of Acacia blakelyi and spp., Callitris arenaria, Banksia prionotes, Banksia attenuata, Hakea platysperma, Xylomelum angustifolium, Allocasuarina campestris, etc. This community contains 6 spp. of Drosera (including geophytes, e.g. D. menziesii), Thysanotus patersonii, and 'quite common' orchids, namely Caladenia 2 spp., Diuris longifolia (, Pterostylis recurva, and Lyperanthus nigricans.


Caesia is a genus shared among Australia, Madagascar, and the Cape Floristic Region of South Africa. It is root-tuberous. However, the development of the tubers is minimal in C. parviflora.

Anigozanthos (Haemodoraceae) does not qualify as geophytic, because the above-ground parts do not die back to ground level in the dry season. Anigozanthis humilis is common in kwongan in Wongan Hills Nature Reserve.

This is an adequate reference for the lack of geophytes in succulent saline vegetation in Western Australia.

Posted on November 30, 2022 10:23 by milewski milewski | 5 comments | Leave a comment

Some aspects of aboriginal diets in southwestern Australia and southern Africa, with particular reference to tubers in mediterranean-type climates

@tonyrebelo @jeremygilmore @yvettevanwijk1941 @sedgesrock @nicky @robertarcher397

In August 1984, I discussed aboriginal diets with the late Sylvia Hallam ( and the late Hilary Deacon (

The following is a transcript of my notes at the time.


The aboriginal population of Australia was remarkably sparse: possibly less than 0.5 million on the whole continent.

Dioscorea hastifolia ( is an indigenous liane. It occurs in a habitat-duality which I have found repeatedly for various spp. of plants: granite outcrops and alluvial banks.

In the relatively nutrient-poor landscapes of southern Australia and southern Africa, both situations - although at opposite ends of the catena - are somewhat enriched, the first by fresh weathering, and the second by concentration via the flow of water.

On alluvial banks, D. hastifolia grew in 'thickets' of Jacksonia sternbergiana ( and, an extremely sclerophyllous, nitrogen-fixing tall shrub or low tree. The liane also climbed on the extremely sclerophyllous cycad Macrozamia riedlei (, another nitrogen-fixer.

In southwestern Western Australia, D. hastifolia occurred in local 'colonies', which were anthropogenically maintained free of the fires otherwise ubiquitous in the area. In these 'colonies', there were permanent 'rabbit-warren' holes dug by aboriginal people, to a depth of several metres in places. The repeated and intensive harvesting was tantamount to a kind of culturing of the food-plant, but did not qualify as domestication, because there was no selective breeding.

Some patches of Dioscorea were fairly extensive, viz. several hundred metres in diameter. Examples occurred at Upper Swan (,_Western_Australia), where D. hastifolia formed fire-protected foci within woodland of Eucalyptus rudis (, which was otherwise regularly burned by the aboriginal people.

Alluvial substrates were particularly supportive of the aboriginal people, partly because of the proximity of Typha domingensis ( and Typha orientalis (, with their edible rhizomes.

The tubers of D. hastifolia have diameter 1.3-4 cm and length up to 60 cm. This large size meant that they were economical to excavate, compared to the other tubers in this flora, all of which are small.

The aboriginal people also ate the unpalatable bulbs of Haemodorum (, an element of the kwongan flora, associated with extremely nutrient-poor sandplains.


Underground storage organs of plants are available for much of the year, and easily found by virtue of the dried above-ground traces. They tend to be unavailable only at that time of year when fleshy fruits are most available, namely autumn (March-May).

In the southwestern Cape of South Africa, the aboriginal hunter-gatherers depended mainly on geophytes.

Several of the 14 South African spp. of Diiscorea reach occur in the Fynbos Biome (, and one ( occurs in the mediterranean-type climate. However, there seems to be no knowledge of any species of Dioscorea in the aboriginal diets. (@yvettevanwijk1941 ?)

Here, corms of Watsonia ( were the main food, in places. This food is starchy, containing less protein than e.g. domestic potato (Solanum tuberosum).

This - the rapid spread of Watsonia as an invasive weed in southwestern Australia notwithstanding (,regrow%20from%20corms%20and%20seeds) - was a slowly-renewing resource, the harvested corms taking up to 4-5 years to be replaced.

Watsonia shoots foliage in late autumn, and grows by depleting the corm, while simultaneously dumping wastes in it, causing it to turn reddish, bad-tasting, and inedible. Then the new corm grows above it, pale and palatable, with minimal tannins, while the above-ground parts die down to leave a good, pale, edible corm in the ground over the summer.

For these reasons, autumn/early winter was really the only time when Watsonia corms were available to the aboriginal hunter-gatherers.


Please see

I find it remarkable that a genus poorly represented in Western Australia compared to South Africa, namely Dioscorea (Dioscoreaceae), was perhaps the most important of the stem-tuberous foods in the mediterranean-type climate in this state, while the category of cormous Iridaceae, so common and diverse in southwestern South Africa, was unavailable under similar climates and on similar soils in southwestern Australia.

There is only one geophytic member of the Iridaceae in southwestern Australia, namely Patersonia babianoides ( However, even this species only marginally/nominally/technically qualifies as a geophyte, because its above-ground parts hardly die down in summer, and the storage organ is merely a corm-like rhizome, hardly qualifying as food for humans, and scarce in the vegetation anyway.

Posted on November 30, 2022 06:03 by milewski milewski | 1 comment | Leave a comment

November 27, 2022

Communities of geophytes in Fitzgerald River National Park, Western Australia, by vegetation type

(writing in progress)

The southwestern region of Western Australia has a mediterranean-type climate, flat topography, generally nutrient-poor soils, and a natural regime of intense wildfires.

Geophytes ( in this region consist mainly of orchids with small root-tubers. However, the incidence of stem-tuberous droseras is significant.

An odd category of geophytes has evolved within the speciose genus Stylidium (

In southwestern Western Australia, the monocotyledonous geophytic flora contributes 7% of the total flora. This is only half the value for the Cape Flora of South Africa, which is environmentally similar ( and

If comparison is made with California, another region of mediterranean-type climate, the main difference is in bulbous plants - which differ greatly in incidence and type in the two regions.

Unlike all other regions of mediterranean-type climate on Earth, Australia lacks any form of fossorial/ subterranean, geophyte-eating rodent (mole-rat or gopher).

In this Post, I summarise the incidence of the various types of geophytes in the main vegetation types in Fitzgerald River National Park (FRNP).

Coastal forest (moort) of EUCALYPTUS PLATYPUS and Melaleuca lanceolata (moonah), on deep calcareous sand

Incidence of geophytes variable (nil in Melaleuca lanceolata over moss-crusted, undisturbed calcareous sand, so perhaps split off Melaleuca lanceolata into another category)
Incidence of geophytes very similar to inland E. platypus, despite the different substrate
Community of geophytes poorer than in woodland of Eucalyptus occidentalis
up to 6 spp. of orchids (only where E. platypus)
Thysanotus patersonii (only where E. platypus)

Low forest (moort/marlock) of EUCALYPTUS conferruminata/lehmannii/macrandra etc. on various substrates
sampled by 11 plots
Clear result, but unknown after wildfire
a few (up to 5) spp. of orchids
Thysanotus patersonii
sometimes no geophytes detected in a given stand, even in spring
no droseras
no geophytic stylidiums
no stem-tubers of any kind

Low forest/woodland (sheoak) of ALLOCASUARINA
Note landform duality, which reflects role of nutrients, P, K
Richest in geophytes where combination of alluvium and fresh weathering?
Similar to the richer communities in woodland of E. occidentalis
Thysanotus patersonii (and T. aff pyg)
Orchids variable, up to 11 spp.
Stylidium crassifolium ( and Stylidium dichotomum (
Drosera spp. (?not as uncommon as in woodland of E. occidentalis)
Chamaescilla corymbosa and Chamaescilla spiralis
Hypoxis lept ( (is this Pauridia glabella?)
Wurmbea tenella (
Glycine clandestina/rubiginosa (geophyte?)( and
Ptilotus spathulatus (
Caesia parviflora uncommon (
Utricularia violacea (geophyte?)(
Craspedia (

Mallee of EUCALYPTUS spp.
Similar to marlock, but additional geophytes present
as in marlock, a few spp. of orchids, and Thysanotus patersonii
In some plots, a stylidium is the only geophyte present
Orchids (0-4 spp. per plot) usually including Pterostylis
Drosera does occur here, including geophyte? (macrophylla,
Stylidium e.g. albomontis ( occurs as geophyte, different from marlock
Ptilotus drummondii ( and Ptilotus holosericeus (, different from marlock
In one plot on well-drained deep soil, Wahlenbergia multicaulis ( other landform was this geophyte found on?)

Mallee-heath of EUCALYPTUS spp. over a diverse lower stratum, in duplex substrates (sand over clay)
most geophytes on well-drained duplex over granite
poor in orchids, as in scrub-heath
in general, as many or more Drosera spp. than orchid spp. per site (true for geophytes?)
more lilies and Haemodorum than in scrub-heath (as statedii by Brown and Hopkins)
never more than 2 spp. of orchids per plot
Stylidium piliferum and Sytilidium schoenoides ( and Stylidium albomontis
Drosera a few spp. including D. menziesii
Anigozanthos humilis (
Haemodorum paniculatum (also in wandoo, but not found in scrub-heath)
Chamaescilla spiralis ( in wandoo, but not found in scrub-heath)
Why no Thysanotus, not even patersonii?
Compared with mallee: shares Stylidium, orchids, and Drosera, but differs in that Thysanotus patersonii of mallee is absent, and the ptilotus (and wahlenbergia) of mallee are replaced in mallee-heath by sparse Chamaescilla and Haemodorum (sand elements)

Woodland (yate) of EUCALYPTUS OCCIDENTALIS on loam
Machaerina occurs where wet in winter
The incidence of geophytes on alluvium under yate is very variable, partly owing to variation in drainage. Geophytes are rare on winter-inundated (which is surprising)(and saline) sites
Overall, richer (in stem tubers?) than mallee, especially in terms of taxonomic diversity (families)
Radical difference according to drainage, but some well-drained plots also poor, for unknown reasons
Stylidium geophytes strangely absent, but other stem-tuberous geophytes present
ANNUALS (cf mallee and moort)
geophytic droseras rare (just D. menziesii in one plot)
Where well-drained, orchids 0-17 (large variation)
Thysanotus patersonii (and T. aff pyramidalis, in one plot)
Wurmbea tenella ( uncommon
Chamaescilla corymbosa ( uncommon
Hyp glabella uncommon
Lagenophora huegelii
Convolvulus erubescens (
Glycine clandestina (geophyte?)(
Ptilotus spathulatus (
Gonocarpus dura
Marsilea drummondii (geophyte?)(

Note: no Wurmbea, Ptilotus spathulatus, Craspedia, or Caesia
Stylidium, Drosera, and orchids are usually present, with similar numbers of spp. per plot
orchids up to 5 spp.
Thysanotus patersonii
Stylidium spin and dichot and piliferum and albomontis and squamellosum
Drosera parvula, stolon, zon, neesii, mac, and menz
Chamaescilla spiralis and Chamaescilla corymbosa
Hypoxis lept
Haemodorum paniculatum (surprise?)
(Poly tenella)
Now cf Cas hueg

Scrub-heath of PROTEACEAE, MYRTACEAE and FABACEAE on deep siliceous sand
Poor in geophytes (0-6 spp. per plot)
All tubers small
Why no Haemodorum?
few orchids (0-4 spp. per plot)
Drosera scorpioides (geophyte?)(
Anigozanthos humilis (not geophyte, but explain)(also found in mallee-heath)(
Thysanotus patersonii
Stylidium piliferum ( and Stylidium albomontis (

Thicket of PROTEACEAE, MYRTACEAE etc. on rocky slopes of quartzite
poor in Drosera
variation in orchids poorly understood, but poverty of drosera noteworthy compared to scrub-heath and mallee-heath
Thysanotus patersonii (rare)
Drosera menziesii
Orchids 0-9 spp.
Stylidium spinulosum ( and schoenoides ( and squamosotuberosum ( and piliferum and albomontis
Poly tenella

Vegetation complex on exposed granite bedrock
Need more plots
Wurmbea tenella
Hypoxis lept
Pauridia glabella ( and and
orchids up to 22 spp.
Stylidium dichotomum (
Thysanotus patersonii
Chamaescilla corymbosa
Drosera mac menz bulbosa
Pol ten
Compared to wodjil: fewer stylidium, no Haemodorum paniculatum, and more orchids in gran exp than in wodjil

Acacia scrub
Inadequate data
(writing in progress)


As in woodland of Eucalyptus occidentalis, the variation in the geophytic community in forest of Eucalyptus platypus is poorly understood. It is not simply a question of salinity.

Now take all moort and marlocks together, and split off all Mel low forest/thickets

Wurmbea (Colchicaceae) is a genus shared between Australia and southern Africa ( It possesses corms (

Caesia (Asphodelaceae) is a genus shared among Australia, Madagascar, and the Cape Floristic Region of South Africa. It is root-tuberous. However, the development of the tubers is minimal in C. parviflora.

Trichocline (Asteraceae) is mainly South American, with one species occurring in Australia ( The tubers are xylopodia (, in at least some spp.

Gardner (1949) states that, in southwestern Australia, Glycine is associated with granite rocks, as are droseras and orchids.

Posted on November 27, 2022 09:14 by milewski milewski | 19 comments | Leave a comment

November 25, 2022

Habitat-distinctions of plants with fleshy fruits in and near Fitzgerald River National Park, southwestern Western Australia

Various species of plants with fleshy fruits occur in/near Fitzgerald River National Park (FRNP,, in southwestern Australia.

The incidence of fleshy fruits, as opposed to other kinds of diaspores, may be related to the nutritional regime of the substrates in question.

In this Post, I summarise the pattern of habitat-preference by these plants. The sources are mainly those of K R Newbey (, together with my own observations.

Occurring patchily on calcareous substrates are

Coastal, calcareous dunes have

The following spp. are shared between calcareous sand and alluvium:

By contrast, alluvium has

Brachyloma geissoloma ( and is poorly documented, but occurs, on alluvium/granite.

The only spp. potentially bearing fleshy fruits on quartzite are two spp. of Ericaceae, viz.

It is noteworthy that no species of Persoonia seems to occur on oligotrophic soils in/near FRNP.

On spongolite there are

On duplex substrates (sand over clay, presumably with mallee-heath) there are

Still to allocate: Billardiera heterophylla ( and


Tetragonia implexicoma


The fruits of these plants (whIch, like Tetragonia, are semi-halophytic) are typically bright-hued but small. Those of Chenopodium spp. are known to be eaten by Gavicalis virescens (, which is probably an important seed-disperser for these plants.

Atriplex semibaccata
Chenopodium baccatum
Chenopodium wilsonii
Enchylaena tomentosa
Threlkeldia diffusa


Alyxia buxifolia


Dianella revoluta


The fruits of these plants are all small.

Leucopogon parviflorus


Nitraria billardierei

Myoporum oppositifolium


These plants are all hemiparasitic.

Exocarpos aphyllus
Exocarpos sparteus
Santalum murrayanum


Alluvium and calcareous sand in/near FRNP, although nutrient-poor by global comparisons, are not as poor as other substrates in the area. Both have considerable floras of fleshy fruits.

The sites richest in spp. with fleshy fruits are likely to be where granitoid-derived alluvia merge with coastal dunes at inlets near the sea (e.g. Culham Inlet, see

Ericaceae with fleshy fruits occur mainly on alluvia in/near FRNP. However, some may occur on granite substrates.

The following ( gives an idea of the nature of ericaceous fleshy fruits in Tasmania. Those in/near FRNP are far more poorly developed: none has conspicuous fruits, and all are as likely to be dispersed by ants and lizards as by volant birds. The genera involved, particularly Styphelia, show a pattern in which

  • the fruits tend to be small, dull, and hidden, and
  • the spp. with the most conspicuous fruits tend to occur in the Eastern States, not Western Australia.

Also see

Posted on November 25, 2022 22:22 by milewski milewski | 16 comments | Leave a comment

November 24, 2022

The parasitic liane Cassytha in southwestern Australia and an ecologically similar region of southern Africa

@alan_dandie @michaelcincotta @insiderelic @scottwgavins @cobaltducks @tonyrebelo @botaneek @troos @fynbosphil @fynbosfia @fynbossie @fynbosfriend @fynbosfae @graham_g @jeremygilmore @arthur_chapman @peterslingsby @benjamin_walton @jayhorn

Most spp.of the genus Cassytha (Lauraceae) occur in Australia, with two outlying spp. occurring in southern Africa.

In and near Fitzgerald River National Park (FRNP,, Western Australia, there are six spp. of Cassytha - probably the greatest diversity of perennial parasitic lianes on Earth.

Strands of various spp. of Cassytha can be found from ground level to 7 m high.

Although all have indehiscent, potentially edible fruits of common design, the size of the fruits varies considerably. Some (particularly Cassytha micrantha and Cassytha glabella) are too small to be considered fleshy fruits, and are probably dispersed and sown by ants.

This means that Cassytha is yet another of the many genera of 'plasticfruits' (, which stretch, within a given genus, from vertebrate-dispersed to non vertebrate-dispersed spp.

Despite their diversity in/near FRNP, all forms of Cassytha are absent from calcareous sand hereabouts, except where limestone crops out, and where Eucalyptus forms dense stands.

In this region, stands of vegetation richest in lianes are the ones poorest in Cassytha, suggesting that this parasitic genus is suited to sites and strata prohibitive to other types of lianes.

The littoral dune of calcareous sand in/near FRNP, vegetated patchily by e.g. Acacia rostellifera, Melaleuca lanceolata, and Chenopodium spp., is free of Cassytha and virtually free of other parasites (whether before or after fire) despite the favourable content of phosphorus in the sand, and Cassytha was also absent from woodland of Eucalyptus occidentalis, containing Santalum on alluvium (visited in November 1991).

Cassytha melantha ( and and and
fruit dull orange-hued, diam. 0.85 cm, seed diam.0.5 cm; fruit mass >0.3 g.

Cassytha pomiformis (
I observed this species in ?Banksia kwongan on deep siliceous sand (certainly SS 1 type) in/near FRNP
no photo available of fruit, which is larger than that of C. glabella
fruit oblong, hairy, fleshy, dull-hued, yellow-green-black, 1.0 X 0.6 cm

Cassytha racemosa (
no photo available of fruit
fruit not necessarily fleshy
fruit oval, ribbed, dull-hued, 0.7 X 0.5 cm
fruit size similar to that of C. glabella
fruit size according to literature: diam.0.4 cm (Weber 1981), 1.0 X 0.4 cm (another ref.)

Cassytha glabella ( and and and
fruit small, variable in shape, greenish or dull orange-hued, meagre-fleshy but succulent (verified by me in FRNP, Nov. 1991), 0.4 X 0.2 cm to 1.1 X 0.3 cm, seed diam. 2.5 mm. Another of my field-notes states fruit dimensions average 0.65 X 0.25 cm, but as little as 0.5 X 0.15 cm, and as much as 1.1 X 0.4 cm.

Cassytha flava (
restricted to kwongan on sandplain
no photo available of fruits
fruit small, inconspicuous, greyish, hairy, 0.5 X 0.4 cm

Cassytha micrantha (
no photo available of fruit
fruit extremely small, inconspicuous, 0.3 X 0.2 cm


Although all its habitats in/near FRNP are subject to fire, Cassytha shows little relationship to fires. It regenerates germinatively, in some spp. possibly from hard, long-lived seed sown before the fire (Weber 1981, Fox 1988).

Although Cassytha (pomiformis?) appears as sparse populations of semi-prostate, dark plants in the early regeneration on open, burnt ground (some stands extending 2 m across bare spaces), it is not particularly prolific at this stage (percent foliage cover <0.1% in regeneration on siliceous sand (SS1 type), two years or less after intense fire) and does not behave as a fireweed, although it does promptly regenerate germinatively.

Any parasite is limited by the regeneration of its hosts and, at least on deep siliceous sands and quartzite alluvium, the hemiparasitic Nuytsia floribunda ( is a very rapid, vegetative regenerator after fire, limiting the opportunities for Cassytha at this stage.

Casytha glabella tends simply to maintain small but significant contribution to the foliage cover of the stand, through the cycle of fire and regeneration.

Cassytha micrantha re-establishes from seedlings within 2 y of fire (Newbey 1987), whereas C. melantha does not for at least two years (Newbey 1987). This difference can most parsimoniously be explained by the speed of regeneration of different strata and life-forms of hosts.

Where the marlocks ( favoured by C. melantha are killed by fire, the flush of pauciennial Alyogyne ( and Kennedia ( dominating the regeneration stands in the first few years after fire is free of this and all other parasites.

Cassytha does not appear to parasitise Nuytsia floribunda. This is true despite the two types of parasites coexisting, and mistletoes being known to parasitise other parasites, including other mistletoes.


All spp. of Cassytha in/near FRNP flower in spring. However, C. melantha finishes in October (Newbey 1987, pers. obs. 1991), whereas other spp. start in Oct.-Nov. and extend to January.

Cassytha glabella may have a particularly protracted flowering season (Newbey 1987), and was seen in fruit in/near FRNP in mid-November 1991 and late April 1992.

Cassytha cannot be considered an autumn fruiter, as are so many bird-dispersed plants of the temperate zone, including Syzygium, Punica, Crataegus/Cotoneaster, Carissa, and Psidium, all fruiting in May 1997 in Claremont-Newlands-Rondebosch, in Cape Town, as I analysed these field-notes.


Cassytha melantha:


In/near FRNP, this species occurs in yate and mallee of Eucalyptus gardneri, in gorges,and on alluvia and coastal dunes. Its habitat tends to be base-rich. It occurs also in moort forest (Eucalyptus platypus).

Cassytha melantha forms the most substantial foliage (strand diameter 2 mm) and fleshy fruit found in this genus in or near FRNP.

Although the fruit does not seem to ripen to a bright hue or a sweet taste, it is too large to be moved far by ants, is not prone to dropping to the ground, and has a somewhat vis ours coating around the seed, suggesting dispersal by birds.

Cassytha melantha parasitises the canopy of tall, mature plants of Eucalyptus on relatively phosphorus-rich soils (where it may exceed 1% foliage cover in some stands). It is absent from oligotrophic landforms, even where Eucalyptus is abundant and reaches 4 m high, such as in mallee-heath in and near FRNP.

This species replaces mistletoes, which are virtually absent from south coastal Western Australia.

Like other parasites, it apparently benefits from physical disturbance of the stands of its hosts. It tends to be common only at the edges of stands of tall eucalypts. It was absent from the stands I examined of Eucalyptus occidentalis on alluvial levees of drainage lines. However, it was present in a stand of the same species at the disturbed edges of a swamp of paperbark Melaleuca.

The fleshy fruit of C. melantha is approximately similar in size to those of its host, Eucalyptus platypus (, as well as the mistletoes it apparently replaces in vegetation subject to canopy-wildfires.

Cassytha racemosa:

Cassytha racemosa is similar in the size and height of the plant, and in the size of the fruit, to Cassytha glabella. It does not contain alkaloids. It flowers in spring, as I noted in Nov.1991 in/near FRNP, and at Sheepwash Creek much farther to the west. It is well into its flowering season in November (pets. obs.; Newbey 1987), but was not fruiting in late Nov. 1991.

In/near FRNP, this species is recorded from coastal dunes, and RMC plots on laterite and limestone, but not in moort (Eucalyptus platypus) or Eucalyptus gardneri. However, in another table (?in Newbey 1987), it is listed from yate, Agonis, Eucalyptus angulosa, Eucalyptus pleurocarpa, thicket, and heath. It is possibly restricted to mature stands of vegetation.

This species may have relatively restricted occurrence in/near FRNP (Newbey 1987). However, I observed this or a similar form in Upland environments (where unrecorded by Newbey 1987), as well as on coastal dunes where a heath stratum was present.

Cassytha glabella:

This species is even more widespread than C. melantha (Weber 1981, Fox 1988). In/near FRNP it occurs in most habitats. It is typical of Eucalyptus pleurocarpa mallee-heath, and found in most other vegetation types including kwongan on siliceous sand (SS1 type) and RMC limestone type. However, it does not occur in Eucalyptus platypus forest.

Cassytha glabella favours Myrtaceae as hosts. In/near FRNP, it forms relatively small plants in low, small-leafed shrubs, often of Melaleuca, in the heath stratum, on oligotrophic soils.

Cassytha glabella flowers in summer. It produces its First fruits at the same time as its larger congener, C. melantha (e.g. late Nov. 1991, and I have a field-note stating end of April 1982, in mallee-heath, where the plant is fine-stranded and yellow). However, it tends not to coexist in the same stands as C. melantha, owing to ecological differences.

This species may be dispersed by small birds (e.g. Zosterops, see notes from Ian Rooke), but appears equally suitable for ants. Did Berg (1975) record it?

Both C. glabella and C. melantha have fruit-pulp that tastes rather resinous, possibly because of the 'resin' content of the Myrtaceae they parasitise.

Habitat differences between C. glabella and C. racemosa are unclear. The latter has fruits of similar size, and parasitises similar hosts, but apparently starts fruiting later in the season.

Cassytha micrantha:


Cassytha micrantha is restricted to coastal southwestern Australia. Nowhere is it common.

In/near FRNP, it occurs on plains, in Eucalyptus pleurocarpa mallee-heath, in kwongan on siliceous sand (SS1 type), in RMC laterite plot, and in RMC limestone plot.

Cassytha micrantha is an extremely gracile, low-growing plant, tending to parasitise wiry sedge-like plants.

Brown and Hopkins (1983) recorded C. micrantha in kwongan in Tutanning Nature Reserve (, on laterite, sand and duplex substrate (sand over clay).

Cassytha micrantha flowers in summer, by mid- to late November (pers. obs., 1991; Newbey 1987), and started to fruit late in November (pers. obs. South Stirling, 1982).


In the Cape Flora of southern Africa, the only species of Cassytha is Cassytha ciliolata ( The fruits have diameter about 0.5 cm, which is not particularly large. However, the succulent fruits turn red, making them brighter-hued than any of the Australian spp. discussed here.

Cassytha pubescens ( does not occur in/near Fitzgerald River National Park. However, it occurs under the mediterranean-type climate in South Australia, making it ecologically as comparable with Cassytha ciliolata of South Africa as are any of the spp. in/near FRNP.

Cassytha pubescens flowers in summer-autumn (Forde 1986).Its fruit has diameter 0.5-0.6 cm (data presumably from Weber 1981), which is larger than that of e.g. Cassytha glabella. However, based on dimensions given in Wikipedia, the fruit of C. pubescens is probably also larger than that of Cassytha ciliolata.

For the relative sizes, please see vs

The fruit of C. pubescens is greenish/reddish/black (Forde 1986). It is eaten by e.g. Dasyornis broadbenti (, Gavicalis virescens (, and Acanthagenys rufogularis (

(This species does not contain alkaloids.)

Observations in iNaturalist show that the hues of the ripe fruit are subtle, rather than bright:

What emerges is that, although C. pubescens of southern Australia and C. ciliolata of southernmost South Africa are intercontinental counterparts, there is a difference in the conspicuousness of the ripe fruits. Although C. pubescens has the advantage of size, the red is so bright in C. ciliolata that its minimum brightness exceeds the maximum brightness of C. pubescens.

Ripe fruits of Cassytha ciliolata:


Many naturalists are familiar with Cassytha, particularly in the Cape Floristic region of South Africa.

However, many may not realise that

  • this genus represents an intercontinental pattern similar to the better-known pattern in Proteaceae, and
  • the bright red hue of the ripe fruits of the species in fynbos, namely C. ciliolata, is odd in the genus, which otherwise has fairly dull-hued fruits, some of them probably dispersed by ants rather than birds.
Posted on November 24, 2022 07:11 by milewski milewski | 19 comments | Leave a comment

November 22, 2022

Misleadingly similar solanums with different relationships to wildfire, in Australia vs southern Africa

@plantperson7654 @wojciech @aguilita @bryanconnolly @mjpapay @troos @botaneek @benjamin_walton @richardgill @sedgesrock @wynand_uys @venturefoth @adriaan_grobler @alastairpotts

Acknowledgement: Many thanks to Jeremy Gilmore @jeremygilmore (see comment in and Tony Rebelo @tonyrebelo for information crucial to this Post.

Solanum symonii of southwestern Australia and Solanum guineense of southwestern South Africa are comparable. This is because both are centred on the mediterranean-type climate, and sandy substrates, mainly coastal.

They are similar in growth-form and fruit-form within the genus, despite belonging to different subgenera.

Solanum symonii is strictly confined to the mediterranean-type (winter-rainfall) climate. Solanum guineense is more versatile, extending somewhat into winter-rainfall semi-arid (Namaqualand) and bimodal rainfall (Eastern Cape). However, this is not a major difference.

Both are short-lived perennials growing to 1.5-2 m high. Solanum symonii tends upright with long leaves, S. guineense somewhat lax with shorter leaves. However, the differences seem unimportant.

Both spp. seem to lack spines on the leaves and stems (for relevance, see

Both have smooth-skinned, succulent berries of approximately oval shape, and about 1.5 cm diameter. Solanum symonii tends to ripen to purple, whereas S. guineense tends to ripen to orange.

However, in the context of the great diversity of fruit-forms in genus Solanum (, this difference seems minor, and is unlikely to indicate a categorical difference in the animals, or range of animals, serving to disperse and sow the seeds.

Given these similarities, why is it that there is such a difference in the number of observations in iNaturalist?

There are currently only 10 observations of S. symonii, compared to 224 of S. guineense.

This may provide a clue to the real difference: that S. symonii is a post-fire (ashbed) specialist, whereas S. guineense bears no particular relationship to wildfire, and tends to occur in vegetation naturally protected from fire.

Rippey and Hobbs (2003, describe S. symonii as a 'fire ephemeral', living only about five years and then disappearing until the next fire.

To understand this fully, it is important to realise that the regimes of wildfire tend to differ in the relevant parts of the two continents. In southwestern Australia, the vegetation is relatively tall and 'resinous', burning intensely every several decades ( In southwestern South Africa, the vegetation is relatively short, burning with limited intensity on a relatively frequent basis - and tending to spare the littoral dunes with which S. guineense is particularly associated.

This difference is partly explained by the more profound and widespread nutrient-poverty in Australia than in southern Africa, exacerbated by the great difference in the incidence of large herbivores ( and and

This difference implies that there is a difference in the nature of the seeds: durable in the soil in the case of S. symonii, compared to non-durable in the case of S. guineense.

Although the agents of dispersal remain to be documented, I suspect that a significant difference is the role of Dromaius novaehollandiae (

Please see and

This species differs fundamentally from Struthio camelus ( in its gastrointestinal system. Emus digest food extremely superficially ( By contrast, ostriches digest their food so thoroughly that few seeds, including the small seeds of solanums, are likely to be defecated intact.

In summary, S. symonii seems to be a natural, indigenous 'fireweed' on a continent run by fire, whereas S. guineense has no particular relationship to fire, instead depending on a combination of relatively phosphorus-rich substrates and relatively intense disturbance by large animals.


SOLANUM GUINEENSE:;jsessionid=4C13F9F3581B712CC175D6B03C19B74F

Scroll through images in

Posted on November 22, 2022 20:23 by milewski milewski | 6 comments | Leave a comment