Ecology of fleshy fruits in Oorlogskloof Nature Reserve, southwestern Cape, South Africa

@tonyrebelo @jeremygilmore @venturefoth @benjamin_walton @botaneek @ludwig_muller @sandraf @robertarcher397 @adriaan_grobler @steven_molteno @elmarvrooyen @justinhawthorne @outramps-tanniedi @sedgesrock @yvettevanwijk1941 @rcswart

I have hiked the trails in Oorlogskloof Nature Reserve (https://en.wikipedia.org/wiki/Oorlogskloof_Nature_Reserve and https://www.sa-venues.com/game-reserves/oorlogskloof.php#:~:text=Oorlogskloof%20is%20exactly%20what%20it,with%20water%20and%20pasture%20land.) several times, one of which was on 21-22 June 2000.

My main search-image was for plants with fleshy fruits of various kinds.

The ecology of fleshy fruits is intriguing, for the following reasons. They:

• have evolved independently in an extremely wide range of families, genera, and spp. of plants,
• tend to belong to genera that are biogeographically widespread,
• tend to dominate the vegetation only in environments protected from wildfire, and
• tend to reflect a particular affluence in potassium, which is the only nutrient element consistently rich in fruit-pulp (https://www.jstor.org/stable/2844617 and https://www.jstor.org/stable/2389867 and https://www.inaturalist.org/journal/milewski/63113-why-are-lauraceous-fleshy-fruits-potassium-rich-despite-being-non-succulent#).

The chemical relationship of potassium to environmental regimes predicts that fleshy fruits should be associated particularly with a combination of good drainage, freedom from wildfire, and vegetative regeneration.

Oorlogskloof Nature Reserve is centred on a ravine that is modest in topographical terms, but sufficient to provide some protection from wildfires.

It is also one of the few remants of vegetation on shale-derived soil under a purely mediterranean-type climate in South Africa (https://en.wikipedia.org/wiki/Mediterranean_climate) that has never been farmed. There is no deep sand in the area.

The mean annual rainfall is about 400 mm, able to support

• typical renosterveld (https://en.wikipedia.org/wiki/Renosterveld) on shale, and
• a dry form of fynbos (https://en.wikipedia.org/wiki/Fynbos) on sandstone.

The climate here is dry enough to support a whole community of Aizoaceae (https://en.wikipedia.org/wiki/Aizoaceae). This is equivalent to an edaphically-determined outlier of succulent karoo (https://en.wikipedia.org/wiki/Succulent_Karoo) that forms a fine-scale mosaic with the fynbos associated with sandstone.

The ravine, after which Oorlogskloof Nature Reserve is named, is protected from fire by

• its location adjacent to the Karoo (https://en.wikipedia.org/wiki/Karoo),
• the topographic break of the sandstone scarp, and
• what I infer - based in the nutrient-richness of shale-derived soils - to have been a relatively intense natural regime of herbivory by wild and (pre-European) domestic ungulates.

Despite its extreme northerly location in the area of mediterranean-type climate in South Africa, Oorlogskloof Nature Reserve is a typical example of the vegetation of the arid-eutrophic side of the environmental spectrum in this climate type in South Africa. It is also valuable in providing a contrast with littoral environments, because it is remote from the maritime influence of the coast to the west.

In this Post, I have marked genera and species with non-fleshy fruits, with an asterisk *.

Two vegetation types, occurring in and around this ravine, are dominated by species with fleshy (actually succulent) fruits.

These are

• Olea europaea cuspidata scrub-forest, which forms pockets on deep shale-derived soils, and
• Cassine peragua affinis low forest, which forms pockets among large boulder-slabs in the upper slopes on shale bedrock (https://www.inaturalist.org/observations/87995356 and https://www.inaturalist.org/observations?taxon_id=598069 and https://www.inaturalist.org/observations/11231212 and https://www.inaturalist.org/observations/27743121 and https://www.inaturalist.org/observations/103215930 and https://www.inaturalist.org/observations/11106400).

Neither of these vegetation types occupies more than one hectare per patch. However, both are dominated by plants with fleshy fruits, in the sense that the collective canopy cover of spp. with fleshy fruits exceeds 50% of total collective canopy cover, and the spp. concerned are the tallest and most massive plants in the community.

Both vegetation types have been protected from wildfire, by:

• what I suspect has been relatively intense herbivory in the context of the Fynbos Biome (https://en.wikipedia.org/wiki/Fynbos), and
• topographic position.

The main difference is that Cassine low forest is moister in summer than is Olea scrub-forest.

The two types of vegetation are similar in height (about 7 m).

However, they differ in floristic complexity.

Mature stands of Cassine low forest contain the following spp. that are absent from Olea scrub-forest:

• Kiggelaria africana (https://www.inaturalist.org/observations/98775279 and https://www.inaturalist.org/observations/101325829),
• Viscum capense (https://www.inaturalist.org/observations/103470746 and https://www.inaturalist.org/observations/98875200).
• Asparagus sp./spp.,
• Haemanthus coccineus (https://www.inaturalist.org/observations/109835852 and https://www.inaturalist.org/observations/103195329) and possibly Zantedeschia aethiopica (https://www.inaturalist.org/observations/103162868), and
• several spp. of Celastraceae.

As far as I could determine, Asparagus retrofractus (https://www.inaturalist.org/observations/103596961 and https://www.inaturalist.org/observations/92447508 and https://www.inaturalist.org/observations/88370710) is the main liane within both types of vegetation.

The following spp. occur at the edge of, and in successional stages to, Cassine low forest:

• Myrsine africana (https://www.inaturalist.org/observations/103593493 and https://www.inaturalist.org/observations/103602694),
• Euclea linearis (https://www.inaturalist.org/observations/74716979 and https://www.inaturalist.org/observations/11230879 and https://www.inaturalist.org/observations/11227961 and https://www.inaturalist.org/observations/11228003),
• Maytenus oleoides (https://www.inaturalist.org/observations/103321451 and https://www.inaturalist.org/observations/103190654 and https://www.inaturalist.org/observations/103190638),
• Maytenus acuminata (https://www.inaturalist.org/observations/97138479),
• Searsia angustifolia (https://www.inaturalist.org/observations/103146318),
• Colpoon compressum (https://www.inaturalist.org/observations/103515063 and https://www.inaturalist.org/observations/103190639 and https://www.inaturalist.org/observations/103165612), and
• Asparagus rubicundus (https://www.inaturalist.org/observations/103484522 and https://www.inaturalist.org/observations/103574641 and https://www.inaturalist.org/observations/103593410),
• Searsia undulata (https://www.inaturalist.org/observations/92946483 and https://www.inaturalist.org/observations/133459348 and https://www.inaturalist.org/observations/12338216),
• Diospyros austro-africana (https://www.inaturalist.org/observations/58398356),
• Olea europaea cuspidata (https://www.inaturalist.org/observations/86770456 and https://www.inaturalist.org/observations/38068285 and https://www.inaturalist.org/observations/11228785),
• Gymnosporia buxifolia (https://www.inaturalist.org/observations/103602736 and https://www.inaturalist.org/observations/103190650), and
• Chironia baccifera (https://www.inaturalist.org/observations/89117348 and https://www.inaturalist.org/observations/127217695).

The last five of these spp. also occur at the edge of, and in successional stages to, Olea scrub-forest:

Where the shale-derived soils in the ravine are subject to seasonal waterlogging along the main drainage lines, Searsia lancea (https://www.inaturalist.org/observations/103215914) partly replaces Olea europaea cuspidata.

In the actual drainage lines, *Salix mucronata (https://www.inaturalist.org/observations/137025694), with its non-fleshy fruits, replaces Searsia lancea. The leaves are similar in these two spp. However, S. mucronata is winter- deciduous, whereas S. lancea is evergreen.

It is noteworthy that the flora in the vegetation dominated by plants with succulent fruits is, at the generic level, not of the Cape Flora (https://en.wikipedia.org/wiki/Cape_Floristic_Region). Instead, it is of widespread affinity, in biogeographical terms.

One of the most intriguing species is Euclea linearis (https://www.inaturalist.org/taxa/585517-Euclea-linearis), which has an extremely disjunct distribution between this area and the northernmost parts of South Africa.

In Olea scrub-forest, the following genera are biogeographically widespread (I remind readers that *asterisks show genera and spp. with non-fleshy fruits): Olea, Searsia, Asparagus, *Dodonaea, Diospyros, and Cassine, plus several others that I did not see in my sampling site, but which probably occur in this vegetation type, namely:

• Viscum,
• Lycium (has any species been recorded in this Nature Reserve?),
• Cassytha (Cassytha ciliolata: https://www.inaturalist.org/observations/103434567 and https://www.inaturalist.org/observations/103434450 and https://www.inaturalist.org/observations/103484576), and
• Solanum (see Solanum retroflexum: https://www.inaturalist.org/observations/103195330 and https://www.inaturalist.org/observations/139506968).

Others are at least fairly widespread, biogeographically:
Chironia, Myrsine, Cissampelos, Gymnosporia, and *Otholobium.

The following show the fleshy fruits of, for example, Cissampelos capensis: https://www.inaturalist.org/observations/27859831 and https://www.inaturalist.org/observations/137668488 and https://www.inaturalist.org/observations/140908292 and https://www.inaturalist.org/observations/140510534 and https://www.inaturalist.org/observations/140510534 and https://www.inaturalist.org/observations/83248623 and https://www.inaturalist.org/observations/60837091.

In my sampling site in Cassine low forest: other widespread genera are Podocarpus, Maytenus, *Indigofera, Viscum (probably present, on Maytenus oleoides), Colpoon (marginal at best), and Zantedeschia if this occurs (uncommon at best).

Biogeographically fairly widespread within Africa are Euclea and Kiggelaria.

Of the smaller shrubs found mainly at edges, *Cliffortia, *Phylica, and *Anthospermum are among the most widespread of the fynbos/renosterveld genera.

In the actual drainage lines, floristic composition at the level of genus becomes so cosmopolitan that one might be in southern Asia: *Salix, *Cyperus, *Juncus, *Pennisetum, *Cynodon, ?Searsia, *Conyza, *Inula (introduced), Prosopis (introduced), Olea, *Lebeckia, *Otholobium (compared to Psoralea), *Mentha, *Helichrysum, and *Isolepis. This seems to correlate with eutrophy in the soils.

Note that Euclea, although present in the area as at least two spp., is very subordinate here - contrasting with its commonness (but not dominance) at the coast in the mediterranean-type climate in South Africa. In Oorlogskloof Nature Reserve, Euclea is largely replaced by Olea. Hereabouts, Euclea is merely a sparse dwarf ebony in renosterveld, and although it may play a part in the succession towards low forest and scrub-forest, it does not persist into them in the way Euclea racemosa does in Sideroxylon and even Celtis- Olinia-Apodytes forest, in coastal parts of the southwestern Cape.

The renosterveld widespread in the ravine in Oorlogskloof Nature Reserve contains few plants with fleshy fruits. These belong to the following spp.:

• Chironia baccifera is short-lived enough to grow and fruit in the fire-interval, as scattered small plants.
• Euclea tomentosa (https://www.inaturalist.org/observations/11231771 and https://www.inaturalist.org/observations/11229271 and https://www.inaturalist.org/observations/11307745) occurs in renosterveld rather than scrub-forest, but nowhere dominates.
• Diospyros austro-africana and Searsia undulata seem to be successional, as disturbed renosterveld regenerates back to woody vegetation with O. europaea cuspidata.
• Asparagus capensis (https://www.inaturalist.org/observations/11236718 and https://www.inaturalist.org/observations/74605470 and https://www.inaturalist.org/observations/92772122) occurs in renosterveld; it seems to be one of the few plants with succulent fruits that is restricted to this vegetation type hereabouts, but its fruits are small.
• Muraltia spinosa (https://www.inaturalist.org/observations/103470726 and https://www.inaturalist.org/observations/103437721) occurs in renosterveld (I noted it in particular on a subsequent visit, on 5 August 2001, when M. spinosa was in flower; PLEASE SEE COMMENT BELOW).

The above spp. may become even scarcer if a regime of intense wildfires is perpetuated by germinative regeneration of cohorts of the flammable Dicerothamnus rhinocerotis (https://www.inaturalist.org/observations/59909111 and https://www.inaturalist.org/observations/11106505).

On the sandstone mesa/plateau, the dry fynbos of *Protea laurifolia (and *Protea glabra) is largely devoid of plants with fleshy fruits, despite being obviously nutrient-rich enough to support such plants.

The nutrient-richness results from a combination of sparse rainfall (which prevents leaching) and dust-deposition from the surrounding Karoo. This may explain the considerable component of relatively eutrophic floristic elements. Poaceae are present, as well as Restionaceae. Ericaceae are largely replaced by Aizoaceae, the latter creating their own fire-free patches by growing sparsely on the rock sheets. The nutrient-enhancement is enough to support Trinervitermes (https://www.inaturalist.org/taxa/567385-Trinervitermes-trinervoides), and its specialised predator, viz. Proteles (https://www.inaturalist.org/observations/7390035).

Muraltia spinosa was not seen in the sandstone mesa/plateau, although found in a few places on the purely shale-derived soils of the ravine. I suspect that, if the area were grazed with a corresponding reduction in wildfire, M. spinosa would occur on this substrate as well. CORRECTION: PLEASE SEE COMMENT BELOW

The only plants with fleshy fruits on the sandstone mesa/plateau are

• Maytenus oleoides, which is restricted to microsites protected from wildfire, viz. edges of rock platforms, where it was demonstrably protected from a fire that burned here recently;
• Oftia africana (https://www.inaturalist.org/observations/103244980 and https://www.inaturalist.org/observations/103484537), which I found well-grown with Ehrharta after the fire that occurred less than three years ago, and was not seen in the mature fynbos; and
• Searsia scytophylla (a relatively small plant for its genus, with fleshy but not succulent fruits, https://www.inaturalist.org/observations/129711114 and https://www.inaturalist.org/observations/59552270 and https://www.inaturalist.org/observations/95561629 and https://www.inaturalist.org/observations/141369103 and https://www.inaturalist.org/observations/103434541 and https://www.inaturalist.org/observations/103434449 and https://www.inaturalist.org/observations/103434541 and https://www.inaturalist.org/observations/103321478).

I found a few stunted plants of O. europaea cuspidata at rock edges, where the soil is deep enough to support *Protea nitida (https://www.inaturalist.org/observations/140179266 and https://www.inaturalist.org/observations/11223230 and https://www.inaturalist.org/observations/21035076).

I did not see Colpoon compressum or Cassytha ciliolata in fynbos on the sandstone mesa/plateau.

Finally, let us turn to the western edge of the mesa/plateau, overlooking the Karoo around Vanrhynsdorp. This constitutes a scarp (minor escarpment).

The critical role of wildfire in allowing plants with fleshy fruits is clearly observed at the edge of this scarp. The environment is dry and fairly exposed, but fire-free by virtue of the topographic break and a particular boulderiness.

Here, Maytenus oleoides becomes more common than in the fynbos (and is parasitised by a Viscum other than the common species in the area, https://www.inaturalist.org/observations/32207172). It is joined by a species associated with afromontane forest, namely Maytenus acuminata. The two congeners live side-by-side among the boulders.

The scandent *Secamone alpini (https://www.inaturalist.org/observations/103573664) occurs here, but not in fynbos or renosterveld.

At the edge of this scarp, Podocarpus elongatus (https://www.inaturalist.org/observations/103195287 and https://www.inaturalist.org/observations/103470716 and https://www.inaturalist.org/observations/103602704) enters in the moistest microhabitats, an even stronger demonstration that 'forest' elements can be more accurately described as 'fire-free' elements.

Driving through the Karoo and beholding the harsh-looking scarp south of Vanrhyns Pass, who would guess that Podocarpus (https://en.wikipedia.org/wiki/Podocarpus) occurs in that arid edge?

At this same edge of the scarp, Searsia scytophylla is joined by Searsia undulata and Diospyros austro-africana. Colpoon compressum and Asparagus rubicundus (which can tolerate some wildfire) enter sparsely, as possibly does Haemanthus coccineus. The *Passerina of the fynbos is replaced by arid-adapted shrubs with non-fleshy fruits, e.g. *Euryops, *Felicia, *Eriocephalus. Phylica oleifolia is the in-between zone species, marginal in its fruit type, and apparently found in the fynbos only in relatively protected microsites.

However, the environment is too dry, and the vegetation too low, to be dominated by spp. with fleshy fruits. Such low vegetation is not dominated by plants with fleshy fruits, even in coastal sands to the west - where Euclea forms patches among the passerinas, etc.

Also see https://www.inaturalist.org/journal/milewski/72570-a-description-of-the-fleshy-fruit-of-cassine-peragua-celastraceae#.

Posted on November 18, 2022 08:14 AM by milewski