Dwarf ebonies, part 2: Euclea tomentosa as a substitute for the ericas missing from the Cape Flora

...continued from https://www.inaturalist.org/journal/milewski/61009-dwarf-ebonies-part-1-white-milkwood-as-symbolically-but-not-biogeographically-south-african#

A major ecological pattern in southern Africa is reduction in the height of the natural vegetation (e.g. see https://www.jstor.org/stable/2844553 and https://www.researchgate.net/publication/282619832_Why_was_the_Highveld_treeless_Looking_laterally_to_the_Pampas_for_global_edaphic_principles_beyond_biogeographical_accidents#:~:text=The%20ultimate%20reasons%20for%20treelessness%20in%20the%20natural,partly%20because%20of%20entanglement%20between%20cause%20and%20effect.).

Over the southwestern and central parts of South Africa, the land - whether flat or steep - seems covered by a blanket of small plants conforming to a height of 0.5-1 meters. And this applies over a range of climates and floras, from fynbos (https://en.wikipedia.org/wiki/Fynbos) and strandveld (https://en.wikipedia.org/wiki/Cape_Flats_Dune_Strandveld) through succulent karoo (https://en.wikipedia.org/wiki/Succulent_Karoo) and Nama karoo (https://en.wikipedia.org/wiki/Nama_Karoo) to various types of treeless grassland (https://en.wikipedia.org/wiki/Highveld).

This phenomenon is difficult to describe, because 'stunting' and 'dwarfing' imply a suppression in the sense of a failure of phenotypes to reach a genotypic potential.

Suppression may apply, at least in part, to e.g. the white milkwood (see my last Post), which seems limited by wind on the littoral. However, many of the species common in fynbos, karoo and highveld seem to reach their full expression as plants of waist-height or less.

The phenomenon applies to both woody and herbaceous growth-forms.

Much of the short stature of fynbos and highveld reflects a predominance of graminoids (particularly Restionaceae and Poaceae). However, there are many kinds of low shrubs as well, epitomised by genera such as Erica (Ericaceae), Phylica (Rhamnaceae) and Pteronia (Asteraceae). Furthermore, although shrubs are typically multi-stemmed, many species in southern Africa are single-stemmed, thus resembling miniaturised trees.

The relationship between the ecological pattern described above and the floristic composition of southern Africa is complex. However, a particular surprise is that certain mainly arborescent clades, which elsewhere avoid winter-rainfall or semi-arid climates, have been recruited to the Cape Flora (https://en.wikipedia.org/wiki/Cape_Floristic_Region) in 'dwarfed' form.

One clade showing this kind of aberrance is ebonies (Ericales: Sapotaceae and Ebenaceae).

Not only have ebonies been recruited to climates avoided by them on other continents, but their typical arborescence has been reduced to shrubbiness - in some cases to the point of genotypic 'hardwiring' - in southern Africa.

The result is a category of plants peculiar to southern Africa but not yet fully investigated by botanists: dwarf ebonies. (I realise that 'dwarf' is unsatisfactory but I lack a better word.)

A typical example is the multi-stemmed shrub Euclea tomentosa (https://www.inaturalist.org/taxa/585519-Euclea-tomentosa and https://plants.jstor.org/search?filter=name&so=ps_group_by_genus_species+asc&Query=Euclea+tomentosa).

This species is closely related to Euclea acutifolia (https://www.inaturalist.org/taxa/585511-Euclea-acutifolia) and Euclea polyandra (https://www.inaturalist.org/taxa/568323-Euclea-polyandra and https://plants.jstor.org/stable/10.5555/al.ap.flora.flosa003090369900018), which show the syndrome of dwarf ebonies less clearly.

Euclea tomentosa has such lignified leaves that it is possibly the most sclerophyllous (https://en.wikipedia.org/wiki/Sclerophyll) of all of the 768 species of Ebenaceae worldwide. Not only has it converged in leaf texture with proteas (https://en.wikipedia.org/wiki/Proteaceae), a family typical of mediterranean-type climates in South Africa and Australia, but it exceeds most South African members of that family in degree of sclerophylly.

At the same time, Euclea tomentosa retains the fleshy fruits typical of ebonies (https://www.inaturalist.org/photos/24232778 and https://www.inaturalist.org/observations/16163420). At first sight its fruits seem incongruous with its leaves and to some degree with the vegetation types into which its clade has been recruited.

Fleshy fruits, attractive when ripe to birds and certain mammals, tend worldwide to be associated with non-sclerophyllous foliage. This trend remains clear in southern Africa (e.g. see https://www.jstor.org/stable/2844617).

However, it is in Ericales that the exceptions - in which sclerophyllous plants bear fleshy fruits - tend to occur (e.g. https://www.inaturalist.org/observations/101827824 and https://www.fs.fed.us/database/feis/plants/shrub/arcvis/all.html).

This leads to a possible explanation for the recruitment of Euclea into the Cape Flora.

The family Ericaceae (https://en.wikipedia.org/wiki/Ericaceae) is exceptionally speciose in fynbos, leading to an impression that the southwestern Cape of South Africa is in some sense particularly representative of this family.

However, southern Africa differs from climatically similar southwestern Australia, the southwestern USA, southwestern South America, and the Mediterranean Basin in that its ericaceous flora excludes species with fleshy fruits.

Southern Africa has 753 species of Ericaeae, but all except one belong to the tribe Ericeae of the subfamily Ericoideae (http://biodiversityexplorer.info/plants/ericaceae/index.htm#:~:text=Ericaceae%20(erica%20and%20rhododendron%20family)&text=About%20126%20genera%20and%203995,are%20cultivated%20in%20the%20region.).

So it seems that Ericaceae in the Cape Flora - however species-rich - are unusually specialised to mesic-climate, small-leafed, heathy shrubs lacking fleshy fruits, and are thus far from representative of the family as a whole.

Accordingly, one way to view Euclea - the flowers of which reveal its relatedness to heathers - is as an erica-related clade which substitutes, in southern Africa, for niches occupied elsewhere in the world by Ericaceae.

Thus, a Californian naturalist might see Euclea as a southern African version of a manzanita (Arctostaphylos, https://en.wikipedia.org/wiki/Arctostaphylos). For an Australian, it might seem like a large-leafed version of various genera of Epacridoideae (http://anpsa.org.au/epacris1.html).

However, this does not stand up to closer scrutiny, because:

  • chaparral in California is exceptional for shrubland under winter-rainfall climates globally, in having many fire-prone, sclerophyllous shrubs with fleshy fruits (not only Arctostaphylos but also Xylococcus, Prunus, Rhamnus, Frangula, Heteromeles, Cneoridium and Rhus),
  • most of these shrubs in California are taller than dwarf ebonies,
  • the fleshy-fruited epacridoids of temperate-climate Australia occur mainly along the eastern coast, those occurring in the winter-rainfall area tending to have small, dull-hued fruits and to be restricted to the understory of woodlands with no counterparts in southern Africa, and
  • Gaultheria, one of the few relevant genera in Chile, is restricted to climates rainier than any in South Africa (located well south of the area of sbrubland called matorral).

to be continued in
https://www.inaturalist.org/journal/milewski/61065-dwarf-ebonies-part-3-adaptation-of-diospyros-to-winter-rainfall-and-adjacent-climates-is-unique-to-south-africa#...

Posted on January 07, 2022 10:16 PM by milewski milewski

Comments

Euclea sekhukhuniensis (https://www.researchgate.net/publication/40665257_A_new_species_of_Euclea_Ebenaceae_from_ultramafic_soils_in_Sekhukhuneland_South_Africa_with_notes_on_its_ecology and https://www.inaturalist.org/observations/68714565) has an extremely restricted distribution on ultramafic substrates (https://www.inaturalist.org/taxa/585518-Euclea-sekhukhuniensis). It seems to qualify as a dwarf ebony, but is better-described as a geoxylic suffrutex. The similar species Euclea linearis (https://www.inaturalist.org/taxa/585517-Euclea-linearis) may in part qualify as a dwarf ebony. In subtropical to tropical regions it tends to be stunted and clonal (https://www.inaturalist.org/observations/99339367 and https://www.inaturalist.org/observations/16465057).
However, in a southwestern part of southern Africa E. linearis grows into a tall shrub.

Posted by milewski about 2 years ago

There is evidence of at least two taxa of dwarf ebonies in the Highveld. The first is a form of Diospyros austro-africana (https://www.inaturalist.org/taxa/583880-Diospyros-austro-africana and https://www.inaturalist.org/observations/11187086). Euclea coriacea (https://www.inaturalist.org/taxa/585513-Euclea-coriacea and https://www.inaturalist.org/taxa/585513-Euclea-coriacea), restricted to the Highveld, seems to parallel the adaptations of Euclea tomentosa in a summer-rainfall climate. However, Coates Palgrave states that it can grow up to 10 meters, which exceeds the capacity of E. tomentosa. Found in the same area is Euclea crispa var. ovata (e.g. https://www.inaturalist.org/observations/19145358 and https://www.inaturalist.org/observations/11270835), which closely resembles E. coriacea but may possibly be restricted to a shrubby growth-form. If so, E. crispa var. ovata would be the best-qualified member of its genus as a dwarf ebony in the Highveld.

Posted by milewski about 2 years ago

Euclea racemosa ssp. racemosa (https://www.inaturalist.org/taxa/599245-Euclea-racemosa-racemosa) qualifies as a dwarf ebony (e.g. see https://www.inaturalist.org/observations/10992493 and https://www.inaturalist.org/observations/66238720) for the same reasons as mentioned previously for Sideroxylon inerme.

Posted by milewski about 2 years ago

Some caveats on short stature of plants. For Grassveld and Nama Karoo, short shrubs are almost always resprouters. A combination of temperature (compared to Savannas) [and nutrients in some habitats] and regular fires (annual or biennial) prevents a tree form establishing. In the absence of fire, forests are the norm, and rocky outcrops (fire refugia) often contain larger shrubs or trees, even compared to flatlands metres away. (there is, however, a loophole where trees can recruit - when fires are delayed or overgrazing suppressses fire, that some species exploit in some circumstances - but these are mostly coppicing species with a tree-potential).
Fynbos has the additional constraint of low nutrients, coupled with the decoupling of ideal growth (wet with cold slowing down growth, hot with dry retarding growth), which slows down growth, resulting in a longer fire interval, but nevertheless wiping out larger shrubs. However, unlike Grassveld, Fynbos allows the guild of non-sprouting, massively-reproducing ("fire "annuals"") shrubs, in addition to resprouters. A non-resprouting shrub guild is largely a Fynbos feature.
In both systems, bird dispersal of fruit leaves fruit susceptible to fire damage. Unburied seeds get killed (although there is an "arms race" here: surface seeded species burn "cool", whereas deep soil seed species burn "hot"). I dont know grassland well enough. But for Fynbos bird fruit are largely confined to rocky outcrop species, and parasites (e.g Cassytha - which refuges in rocky outcrops) - where targetted dispersal in fire-safe habitats is an option. (of course, Forest patches with Fynbos are loaded with bird fruit species, as is Strandveld).
Furthermore, given the fire turnaround times, dwarfing of trees is the only way that tree lineage can survive in Fynbos and Grassveld (and arid/desert areas where water may limit tree habits).

This is just by way of background.
Why the Ericeae have been unable to evolve fruit, and why the other subfamilies and tribes with fruit never got to the Cape (or went extinct), and whether Euclea has usurped the niche, are different issues, which you are developing here.

Posted by tonyrebelo about 2 years ago

@tonyrebelo Many thanks for this background. It may be worth noting, as food for thought, that Arctostaphylos combines fleshy fruits and dispersal by vertebrates with extreme dormancy of seeds, which need scarification/scorching for germination. This has led me to suspect that the seeds of certain species of Euclea may also show dormancy and fire-resistance. Also: under a similar climate in southwestern Australia, vegetation dominated by eucalypts is so pyrogenic that four different fire cycles occur concurrently within a given stand (mild and frequent in the heathy understorey; moderate-intensity and infrequent with scorching of the trees; the same plus individual trees burning down from the base up, after the fire sweeps through; and extremely intense and occasional with explosive combustion of the whole canopy). And yet not only did trees persist but they reached as high as 40 meters over extensive areas.

Posted by milewski about 2 years ago

But what is the fire interval: it is far longer. Being longer the plants grow bigger. So 15 year fire interval excludes trees. But 1 (e.g. Savanna) or 50 years (e.g. Australia) allows trees.
But gums are special. In some habitats some species seem to suppress everything else. Do they increase the fire interval by eliminating/reducing other fuel plants?
But even so, in many areas they form mallee rather than trees. Admittedly, by Cape standards much mallee are trees.

Posted by tonyrebelo about 2 years ago

In vegetation dominated by Eucalyptus marginata (https://en.wikipedia.org/wiki/Eucalyptus_marginata) in southwestern Australia, the fire-interval is thought to have been extremely short (as frequent as in fynbos or more so) because there is a heathy understory similar to fynbos. It is currently thought that the fire-interval was as short as two years under aboriginal management for thousands of years. Despite this frequency the trees grew tall and straight, and although eucalypts are dominant there are various other trees in the same stands, belonging to genera such as Banksia (https://en.wikipedia.org/wiki/Banksia_grandis), Allocasuarina (https://honkeynuts.com.au/common-sheoak-allocasuarina-fraseriana/), Xylomelum (https://en.wikipedia.org/wiki/Xylomelum_occidentale) and Persoonia (https://honkeynuts.com.au/upright-snottygobble-persoonia-longifolia/). Furthermore even the monocotyledonous plants included arborescent forms such as Xanthorrhoea (https://www.youtube.com/watch?v=-Mev1nsfM3k and https://www.researchgate.net/figure/The-grass-tree-Xanthorrhoea-preissii-in-unmined-jarrah-Eucalyptus-marginata-forest_fig1_228641779 and https://www.youtube.com/watch?v=vpklazyfLms). Vegetation under a mediterranean-type climate in southwestern Australia is the most arborescent vegetation on any continent under this climate, despite being the most fire-prone (as per frequency and intensity of fires).

Posted by milewski about 2 years ago

Nice1 Thank you. Are you then presuming Euclea "arrived" with Erica and not necessarily down the path with all other sub-tropical evergreen flora.
It raises eyebrows when one states that stunted dune flora is actually forest when composed of Olea, Euclea, etc.

Posted by benjamin_walton about 2 years ago

No doubt about it. Strandveld is forest.

Posted by tonyrebelo about 2 years ago

Mention that to Nicky Schmidt 😶

Posted by jeremygilmore about 2 years ago

The term 'forest' needs to be defined wherever it is used, because it is so relative. One of the best examples of this is an area of rainforest in eastern Australia called the Big Scrub (https://en.wikipedia.org/wiki/Big_Scrub). To a southern African naturalist, 'scrub' means vegetation dominated by shrubs, and rather low shrubs at that. But in Australia there is so much arborescence in various vegetation types, across a range of climates, that the standards are quite different. The Big Scrub was dominated by trees as tall as the tallest indigenous trees in southern Africa (see https://en.wikipedia.org/wiki/Argyrodendron_trifoliolatum and https://en.wikipedia.org/wiki/Toona_ciliata, compared with https://en.wikipedia.org/wiki/Afrocarpus_falcatus). However, Australians found such trees so puny, relative to e.g. Eucalyptus regnans (the tallest plant on Earth, https://en.wikipedia.org/wiki/Eucalyptus_regnans), that they viewed them as mere scrub by comparison, and promptly cleared them to cultivate herbaceous plants. And I suspect that even the word 'big' here refers not to the plants but to the area covered by this unusually extensive patch of rainforest. Vegetation dominated by Sideroxylon inerme in its tallest form would be called 'vine thicket' if it occurred in Australia (https://www.dpaw.wa.gov.au/images/documents/plants-animals/tecs/Monsoon-vine-thickets-Dampier-Peninsula.pdf and https://www.anbg.gov.au/photo/vegetation/rainforesst-vine-thickets.html).

Posted by milewski about 2 years ago
Posted by jeremygilmore about 2 years ago

Another relevant species is Xylococcus bicolor (https://www.inaturalist.org/taxa/79543-Xylococcus-bicolor), a member of Ericaceae that is restricted to the mediterranean-type climate in California and ecologically similar to Arctostaphylos. It possesses thinly-fleshy fruits, complete with pre-ripe display (https://www.inaturalist.org/observations/105118216 and https://www.inaturalist.org/observations/87934800 and https://www.inaturalist.org/observations/105078961 and https://www.inaturalist.org/observations/76252455 and https://www.inaturalist.org/observations/81633006 and https://www.inaturalist.org/observations/78718978 and https://www.inaturalist.org/observations/81650345 and https://www.inaturalist.org/observations/104742485 and https://www.inaturalist.org/observations/104502081 and https://www.inaturalist.org/observations/100467007 and https://www.inaturalist.org/observations/96348492 and https://www.inaturalist.org/observations/95512692), eaten by certain birds and Canis latrans, the latter certainly dispersing and sowing the seeds in its feces. It is comparable with Euclea spp. under mediterranean-type climates in South Africa, including vegetative regeneration from the base. However, it is only semi-sclerophyllous and it remains taller (https://www.inaturalist.org/observations/105181155 and https://www.inaturalist.org/observations/77220558) than dwarf ebonies. Similar comments apply to Comarostaphylis diversifolia (https://www.inaturalist.org/taxa/76423-Comarostaphylis-diversifolia), which is even less similar to dwarf ebonies because of its similarity to Arbutus.

Posted by milewski about 2 years ago

The following is a useful introduction to Arctostaphylos, a genus that approaches analogy with dwarf ebonies on other continents: https://www.ccber.ucsb.edu/news-events/ecology-and-diversity-manzanitas#:~:text=Arctostaphylos%20are%20either%20facultative%20seeders,of%20Arctostaphylos%20are%20obligate%20seeders. A major difference is that 85% of species of Arctostaphylos regenerate germinatively, unlike the vegetative regeneration seen in dwarf ebonies.

Posted by milewski about 2 years ago

The genus Arctostaphylos, many species of which are large shrubs in chaparral vegetation under a winter-rainfall climate in California, differs from all shrubs in fynbos and associated vegetation types in South Africa in combining a) bright-hued, fleshy fruits (https://honest-food.net/manzanita-berries-edible/) and seeds dispersed and sown by birds with b) hard-seededness, seed-dormancy and germinative regeneration after wildfires. The main birds eating the fleshy fruits of Arctostaphylos and passing the seeds intact are Toxostoma redivivum (https://www.inaturalist.org/taxa/14906-Toxostoma-redivivum) and Aphelocoma californica (https://www.inaturalist.org/taxa/506118-Aphelocoma-californica), which have no counterparts in the avifauna of fynbos or associated vegetation types.

Posted by milewski about 2 years ago

But do they have an equivalent of our Bulbul?

Posted by tonyrebelo about 2 years ago

@tonyrebelo That is the surprising difference: Pycnonotidae (https://en.wikipedia.org/wiki/Bulbul), so taken for granted across a range of biomes in southern Africa, seem to have no counterparts in corresponding biomes in California. Bulbuls tend to be smaller-bodied, more frugivorous, and more generalised than Toxostoma or Aphelocoma. Toxostoma redivivum is more than twice as massive as Pycnonotus capensis (https://en.wikipedia.org/wiki/Cape_bulbul) or Andropadus importunus (https://en.wikipedia.org/wiki/Sombre_greenbul), and is a specialised litter-foraging and cover-dependent omnivore that happens to turn to fleshy fruit-pulp seasonally. Aphelocoma californica is a specialised scatter-hoarder, eating seeds of evergreen Quercus as a staple, that also happens to eat fleshy fruit-pulp. So, just as Arctostaphylos is an odd combination of pyrophile and ornithophile, so the birds disseminating it are odd combinations of specialised and generalised.

Besides the various fleshy-fruited Arbutoideae, chaparral in California contains Rhamnus, Prunus, Lonicera, Rhus, Heteromeles (https://en.wikipedia.org/wiki/Heteromeles), and Cneoridium (https://en.wikipedia.org/wiki/Cneoridium#:~:text=From%20Wikipedia%2C%20the%20free%20encyclopedia%20Cneoridium%20is%20a,seed%2C%20but%20can%20also%20be%20accomplished%20by%20cutting.), all of which combine adaptation to wildfire with fleshy fruits.

This syndrome - which we might call 'fleshy-fruited pyrophily' - may seem strange to naturalists in the southwestern Cape of South Africa (particularly given that the species of Rhamnus and Prunus indigenous to this region depend on protection from wildfires), but it is actually present in the Cape Floristic Region in the form of dwarf ebonies, particularly Diospyros glabra and Euclea polyandra.

These dwarf ebonies are evolutionarily convergent with the syndrome so typical of Californian chaparral, but they still fall short in at least two ways. First, the plants are small (typically one meter high) compared to most of the fleshy-fruited shrubs in chaparral (typically three meters high). Second is the question of longevity of seed, which seems not yet to have been investigated in dwarf ebonies. I wonder about Euclea polyandra, a species integrated into fynbos and retaining fleshy fruits but with fruits modified enough that I suspect that this species might be able to form a durable seed-bank in the soil. At very least the dispersal and sowing of E. polyandra would be interesting to study in the field, in a convenient location such as Fernkloof Nature Reserve (https://en.wikipedia.org/wiki/Fernkloof_Nature_Reserve) near Hermanus.

Posted by milewski about 2 years ago

@tonyrebelo Here is Toxostoma redivivum (https://en.wikipedia.org/wiki/California_thrasher), a disperser and sower of 'fleshy-fruited pyrophiles': https://www.youtube.com/watch?v=7sa505cYeZ4. It weighs up to 90 grams, for comparison with the average of 28 grams of that most familiar South African bird, Cossypha caffra (https://en.wikipedia.org/wiki/Cape_robin-chat). Toxostoma redivivum is specialised even compared with another member of the same genus (see https://www.youtube.com/watch?v=Y5lKY-d50QE and https://en.wikipedia.org/wiki/Brown_thrasher), which weighs on average 68 grams and is more comparable with South African Phyllastrephus terrestris (https://en.wikipedia.org/wiki/Terrestrial_brownbul). Also please see https://sora.unm.edu/sites/default/files/journals/auk/v034n04/p0427-p0433.pdf.

Posted by milewski about 2 years ago

Another clade marginally relevant to the topic of dwarf ebonies is Anacardiaceae: Searsia. Here we have another radiation in the Cape Floristic Region, particularly under winter-rainfall climates, in parallel with Euclea but more speciose. All seem to regenerate vegetatively and several species remain diminutive even as fully-grown individuals. As many as 23 species of Searsia occur naturally in the winter-rainfall region of Western Cape Province. Several species certainly penetrate fynbos, and a few may be seen as restricted to fynbos. However, no species is as sclerophyllous as Euclea tomentosa, and no species has unambivalently fleshy fruits with bright hues attractive to seed-dispersing birds. The seeds of most or all species of Searsia seem to be dispersed and sown by vertebrates rather than insects, wind or water, but they tend to be dull-hued when ripe, with poorly-developed fruit-pulp. Searsia in the Cape Floristic Region is generally associated with sites relatively protected from wildfire. This clade occurs in winter-rainfall California in the form of Rhus integrifolia. However, a difference between Searsia and Diospyros is that the latter genus is mainly tropical, making its adaptation to temperate latitudes and winter rainfall odder than that of Searsia and other Anacardiaceae.

Posted by milewski about 2 years ago
Posted by milewski about 2 years ago

Add a Comment

Sign In or Sign Up to add comments