How should we describe and interpret the fleshy fruit of Carpobrotus edulis (Aizoaceae)?

(writing in progress)

Also see

The fleshy fruit of Carpobrotus edulis (Aizoaceae, and,the%20herb%20or%20kitchen%20garden.) is noteworthy for several reasons.

These are as follows:

  • This presents an example of 'plasticfruits', because it seems unlikely that the genus would be separated from Mesembryanthemum were it not for the shift of its fruits to endozoochory.
  • The species, C. edulis, is unusual in that its flowers and fleshy fruits advertise themselves by means of the same hue (yellow).
  • Carpobrotus has a disjunct distribution on southern continents, with the various spp. retaining the ability to hybridise, despite having been isolated long enough to speciate.

However, describing the fruit of Carpobrotus is complicated, because of

  • unusual involvement of the calyx, in the form of persistent, leaf-like sepals,
  • difficulty of determining which stage of the fruit is most aptly described as 'ripe', and
  • recent hybridisation; two spp. have widely invaded the Northern Hemisphere for anthropogenic reasons, and there has been blurring of species-specific features by interbreeding among various congeners.

The following shows the fruit of Carpobrotus edulis, broken open at a stage when the pericarp and sepals are still succulent:

The following show the progression of development and ripening of the fruit of C. edulis.

Before the fruits are full-size, they (including persistent, succulent sepals) start to turn yellowish.

The persistent sepals may be reddish. However, this does not necessarily make the fruits more conspicuous, because the leaves also tend to feature reddish hues:

When the fruits are full-size, the yellow is conspicuous to the human eye. At this stage, the fruits (including the bright-hued, persistent sepals) are as still as succulent as the leaves:

Then the sepals begin to wither:

Then the rest of the fruit also starts to wither and turn brown:

Finally, the whole fruit and pedicel become brown, with no part of the complex structure remaining succulent, although the moist, pasty, translucent mass in which the seeds are embedded remains fluid:

(writing in progress)

Publicado el 24 de diciembre de 2022 por milewski milewski


@tonyrebelo @jeremygilmore Could you please make comparative comments for C. edulis, based on your own experiences?

Carpobrotus virescens has fleshy fruits ( and

The following show the natural distribution: and

The fruit, at its most appetising-looking stage, is reddish and succulent (including two persistent sepals):

However, having eaten from this species in plenty on several occasions, I have the following complications to point out.

For the appearance of the 'ripe' fruit, see and, and several other photos in the same article.

For anatomical detail of the'ripe' fruit, see and

The information published on the Web implies that the fruit is succulent when ripe, and pleasant to eat at this stage.

This is not in line with my experience.

What I have found is that the succulent, bright-hued stage of ripeness is not palatable, even if the trouble is taken to 'peel' it (which is too fiddly for even aficionados of wild fruits). The sepals, which are succulent and bright-hued, never achieve palatability - beyond those of leaves - at any stage of the process of ripening.

Instead, what I have found myself doing is to seek out the fruit when it has dried out to a brownish, inconspicuous hue. At this stage, which I regard as fully ripe, it can be eaten whole, the pericarp being pleasantly leathery with a vaguely toasty taste, and the internal 'jam', no longer succulent, is mild-tasting. This 'jam' can be squeezed, somewhat like toothpaste, from even the dry fruit if the leathery pericarp is broken.

These fully ripe, 'raisined' fruit tastes noticeably salty, any sweetness being so slight that the main impression is of pleasant 'umami' rather than any kind of fruitiness. There is negligible sourness, no bitterness or aftertaste, and no astringency that I can recall.

The experience is that of eating crusty bread rather than fruit - which is consistent with the Greek origins of the word 'Carpobrotus' (karpos = fruit, brotos = edible, presumably the etymological root of 'bread').

At no stage does the fruit fall off, or become loosely attached. Even when dry and brown, the fruit has to be forcefully plucked.

Summarising the fruit of Carpobrotus virescens:
When fully ripe, and most palatable to humans, it is no longer succulent. It naturally dries to a consistency similar to toasted bread, the inside being a non-juicy 'jam', which tastes pleasant despite hardly being sweet.

My interpretation:

Carpobrotus virescens is adapted for dispersal and sowing by both mammals and large lizards. The large lizard Tiliqua rugosa ( is reputed to be attracted to the fruit at the 'ripe', red stage, and probably eats it mainly at this stage. It is also plausible that kangaroos, wallabies, bandicoots and indigenous rats prefer the fruit at this same, succulent stage. However, from a human viewpoint it seems to me fortuitous that there is a further 'post-ripe' stage, which is fully palatable to Homo.

So, I do not know which stage we should, objectively, regard as 'fully ripe'. My guess is that the mammalian agents of dispersal all share my readiness to eat the 'raisined' fruit, which may no longer be attractive to the lizards. I therefore suggest that we treat the bright-hued, succulent stage as a 'pre-ripe display', and the dried-out, brown, inconspicuous stage as the fully ripe stage. This is based partly on the assumption that all spp. of Carpobrotus are basically adapted to dispersal and sowing by mammals, rather than birds or reptiles.

I have eaten the fruits of C. edulis, but I do not recall the details. If it is like C. virescens, I would call it 'veldbrood' or 'strandbrood', rather than 'suurvy' or 'sour fig'. What do you think?

Publicado por milewski hace 5 meses (Marca)

The following confirm the presence of an edaphic seed-bank in Carpobrotus, and discuss regeneration after fire: and

Publicado por milewski hace 5 meses (Marca)

The following is the only source I have yet found, on the Web, that describes the fruits of Carpobrotus in a non-misleading way:

Publicado por milewski hace 5 meses (Marca)

"Wild Food Plants of Australia" (1988), by Tim Low (Angus and Robertson Publishers, North Ryde, NSW, Australia) is one of the best books I have found in the category of natural history. However, I find even this to be a misleading account of the fruits of Carpobrotus.

On page 66, referring collectively to four spp. of Carpobrotus indigenous to Australia, Low states "Pigface is one of Australia's tastiest wild fruits, its soft wet pulp tasting like salty strawberries or juicy fresh figs". Based on the two spp. I have tasted, I cannot recognise Carpobrotus in this description. In my experience, reality is both subtler and more complicated. If Low's description is apt, then why is it that I have yet to find another naturalist who has ever actually eaten the fresh fruits of any species of Carpobrotus?

Publicado por milewski hace 5 meses (Marca)

@adriaan_grobler I find it puzzling that the common name of Carpobrotus deliciosus calls it 'delicious', yet there is not one photo of the fruit in iNaturalist. Have you ever tasted the fruit of this species?

Publicado por milewski hace 5 meses (Marca)

@yvettevanwijk1941 @koosretief

Have either of you, by any chance, eaten enough of the fruits of any species of Carpobrotus to offer a description, particularly in the fresh vs dry condition?

Publicado por milewski hace 5 meses (Marca)

I can attest that the fruit of Carpobrotus deliciosus are, indeed, delicious. At least to my palate and also the folks who gather them for eating fresh or making preserve in the E-Cape. I was told by E-Cape locals that the common name, Ghaukum. applies to C. deliciosus and not to C. edulis.

C. deliciosus fruit are quite edible once the fruit start colouring a bit and the calyx lobes become withered (the fruit is then still succulent).
You can remove the thin "peel" or just bite off the top and squeeze the fruit pulp and seeds into your mouth.

I don't like C. edulis at all, for the reasons you mention. Locals in the E-Cape seem to agree because I've heard local descriptive names that suggest that the fruit of C. edulis are bitter and slimy.

Publicado por wynand_uys hace 5 meses (Marca)

@wynand_uys Many thanks for your most helpful comment. Would it be true to say that the fruit of C. deliciosus never becomes bright-hued compared to C. edulis (definite yellow for the while fruit including sepals) or C. virescens (definite purplish red for the whole fruit including sepals)?

Publicado por milewski hace 5 meses (Marca)

Looking up my field-notes on Carpobrotus, I found the following entries.

Location: Swanbourne Rifle Range, Perth metropolitan area (,_Western_Australia)

5 Jan. 1983:
Carpobrotus virescens in ripe fruit. I ate about a dozen. Salty taste. I do not know whether the salt is in the plant or on the surface, deposited from aerosol.

March 1982:
In flower-bud is Spyridium globulosum.

7 Sept. 1981:
The dunes here are calcareous. The first littoral dune is colonised by non-sclerophyllous plants. On the second, more humus-enriched, more stable dune, the upper storey becomes Eucalyptus gomphocephala and Banksia sessilis (sclerophyllous and foliar-spinescent), with scattered Grevillea. However, the ground-stratum remains succulent/fleshy/annual, and there remain some Scaevola crassifolia (, Spyridium globulosum, and Olearia axillaris. There is an abundant shrubby daisy, probably Senecio pinnatifolius (, with fleshy pinnate leaves. Annual plants are common: many daisies, grasses, Calandrinia, Erodium, and non-indigenous Pelargonium and Taraxacum officinalis. Some moss is present. There is no particular restriction of the annuals and fleshy/succulent plants to litter. However, these plants seem in places to be limited by mobility of the sand. There is a maximum of perhaps 10 spp. of annuals coexisting here, with a further 5-7 spp. of suffrutescent/perennial herbaceous plants of various kinds. In flower now, in early Sept., are Hardenbergia comptoniana, Clematis pubescens ( and/or linearifolia (, Anthocercis littorea (, and Brachyscome sp. ( Plants with fleshy fruits/arils: Carpobrotus virescens occurs in the first littoral dune. Chenopodium baccatum is common. Acacia cyclops occurs. Lycium ferocissimum (non-indigenous,, now in some fruit, tends to grow in the deeper swales, in the more disturbed areas. There are no geophytes whatsoever (even of non-indigenous spp., and even on a small patch of apparently burnt ground). No orchids and no Haemodoraceae. There is just the usual, non-indigenous Trachyandra divaricata (, which just has thickish velvety roots, with no swollen underground organ as such. This site seems to exemplify the puzzle of a lack of geophytes, and to a lesser extent succulent fruits, on coastal sand in Australia, despite the general softness of the foliage. Animal life is conspicuous here compared with kwongan. Lichmera indistincta ( is abundant, foraging at e.g. the flowers of Hardenbergia comptoniana. Also present are Gavicalis virescens (, Zosterops lateraljs (, and Streptopelia senegalensis ( Lizards can be heard scuttling, and Tiliqua rugosa is present. Oryctolagus cuniculus is abundant.

Publicado por milewski hace 5 meses (Marca)

Correct. I don't think I've ever seen a brightly coloured C. deliciosus fruit, although the calyx lobes do develop some colour before shrivelling (yellow to purple).

Publicado por wynand_uys hace 5 meses (Marca)


What you have pointed out is intriguing: that the fruit of C. deliciosus is (at least to humans) the most palatable in its genus, but far less conspicuous, and thus far less appetising in appearance, to those of less palatable spp. Am I right in thinking that this counterintuitive point has not been stated clearly in the literature? And how can we interpret this w.r.t. dispersal and sowing by non-human agents?

Here is a first attempt:

In Australia, all of the spp. of Carpobrotus have purple-red fruits, because most or all of the non-human dispersers have trichromatic vision (as do we humans). I refer to the emu, lizards, and marsupials (including kangaroos and wallabies?), which see red and green (,trichromats%20among%20closely%20related%20species.).

In southern Africa, the spp. of Carpobrotus have yellow fruits in the case of edulis and acinaciformis, and green fruits in the case of deliciosus. This is hypothetically because the most important natural dispersers are ungulates in the case of the first two, and primates in the case of the third. Ungulates, having dichromatic vision, do see yellow although they cannot distinguish red and green. Primates have trichromatic vision. It pays deliciosus to 'hide' from ungulates because for some reason its dispersal and sowing are better-served by Homo and Papio than by ungulates.

An alternative is that the main disperser of deliciosus is Stigmochelys ( and other tortoises (which, like lizards, see red and green, and, and that the palatability to primates is a 'side effect' rather than a case of 'co-evolution'.

Your further thoughts?

Which do you think is more likely: that C. delicious has 'co-evolved with Homo as a main disperser, or with tortoises - which are known to play an important role for e.g. Muraltia spinosa in comparable habitat - as main dispersers?

Publicado por milewski hace 5 meses (Marca)

The following show the distributions of Muraltia spinosa and Carpobrotus deliciosus:

Publicado por milewski hace 5 meses (Marca)

I have little to offer regarding the evolution of dispersal strategies of Carpobrotum.
I do think that fruit colour (although important in many cases) is only one of several attractants to dispersers.
I'm sure that tortoises also rely on smell to locate a meal and that they'd have no trouble finding ripe C. deliciosus even if there are highly visible fruit of less desirable plants around. That is, if they also prefer deliciosus over edulis. I don't know.

Publicado por wynand_uys hace 5 meses (Marca)

Roadside sellers sell them (Carpobrotus edulis) as dry capsules to eat or make jam with. Presumably they are still filled with a runny fluid and seeds. I like to eat them when they are still yellow, but becoming brown (just before they begin to dry out). I have found them more sweet at this stage.

Publicado por jeremygilmore hace 5 meses (Marca)

@jeremygilmore Many thanks for your comment

Publicado por milewski hace 4 meses (Marca)
Publicado por milewski hace 4 meses (Marca)

Agregar un comentario

Acceder o Crear una cuenta para agregar comentarios.