A new variety of Haworthia Limifolia from Natal (1962)

Reprinted from “The Journal of South African Botany”, vol. XXVIII, part III, July, 1962


M.B.Bayer, College of Agriculture, Cedara.

The typical form of Haworthia limifolia is from somewhere west of Delagoa Bay and was described by Marloth (1908).  Prof. R. H. Compton has kindly sent me specimens from the Umbuluzi River Valley north of Stegi and also a form from near Gollel on the southern border of Swaziland.  Herbarium records show that forms also occur near Barberton;  the confluence of the Pongola and Pivaans River 20 miles east of Paulpietersburg; Ndumu, Zululand;  Kosi Bay, and the Ubombo Mountains in Zululand.  A photograph of a plant collected by Mr. E. R. Harrison 6 miles west of Somkele on the slopes of the Nkonyana Mountains, matches plants collected by me at Ngceba, north of the Black Umfolozi River, between the Umfolozi and Hluhluwe Game Reserves.  This form is also reported to occur on the nTsuzi and Ulumbi hills which flank the Umfolozi Reserve on the north and south.  Two plants have been collected from this area by Mr. H. Dent.  Mr. A. F. Bruyns-Haylett also records it from the Mozaan River Valley between Swaziland and the Pongola River (fig.2.)

Continue reading

Haworthia Revisited – 49. Haworthia limifolia

49. Haworthia limifolia Marl., Trans.R.Soc.S.Afr. 1:409(1908).  Bayer :129(1976).  Bayer :69(1982).  Scott :31(1985).  Smith, Fl.Pl.Afr. 55:24 (1997).  Type: W. Delagoa Bay, Marloth 4678 (PRE):  H. limifolia var. diploidea Res., Feddes Repert.Spec.Nov. 48:(1940).  Type: ex Lisbon.  Not preserved:  H. limifolia var. tetraploidea ibid.  Type: ex hort Lisbon.  Not preserved:  H. limifolia fa marlothiana Res., Bolm.Soc.Broteriana 14:200(1941):  H. limifolia var. marlothiana Res., Mems.Soc.Broteriana: Succ.Afr. 3:93(1943).  Type: ex hort Lisbon.  Not preserved:  H. limifolia var. schuldtiana ibid.  Type: ex hort Lisbon.  Not preserved:  H. limifolia var. stolonifera ibid., Mems.Soc.Broteriana: Succ.Afr. 3:94(1943).  Type: ex hort Lisbon.  Not preserved:  H. limifolia var. stolonifera fa pimentelli ibid.  Type: ex hort Lisbon.  Not preserved:  H. limifolia var. stolonifera fa major ibid.  Type: ex hort Lisbon.  Not preserved:  H. limifolia var. keithii Smith, JS.Afr.Bot. 16:4(1950).  Type: Swaziland, Smith 5693 (EL, NBG).

limifolia: file‑like leaves.

Rosette stemless, slowly proliferous, with or without stolons, 5-7cm φ.  Leaves 12-30, light to very dark green and even brownish-green, opaque, ovate-lanceolate, spreading, scabrid with white or concolorous tubercles, or with confluent transverse ridges, margins and keel scabrid.  Inflorescence slender, to 35cm.  Flowers slender, tepals fused and with revolute tips.

1982 – H. limifolia was originally recorded from west of Delagoa Bay but there does not seem to be any confirmed record of this species in Mocambique itself.  It is reported to be in the vicinity of Maputo.  It is quite well recorded from Swaziland in the eastern mountains. and from Gollel in the south-eastern corner.  Other records include Barberton, Ndumu, Umfolozi and Paulpietersburg.  It is clearly widespread and the populations differ markedly.  It propagates very readily from stolons given optimum growing conditions, as do both H. nigra and H. venosa ssp. tessellata.  There is a record of an unusual white‑tubercled form from the Mozaan River in the south‑eastern Transvaal.  The var. gigantea is a large variety with small raised tubercles as opposed to the transverse ridges characteristic of the species.  The var. ubomboensis is entirely glabrous.  However, it has never been recollected at its purported locality east of Stegi and thus remains a mystery.  The plants are apparently sought after for their purported medicinal and magical properties and may thus be heavily exploited.

1999 – Despite a very wide distribution range, the species is really poorly known.  The herbarium record is fairly comprehensive but the species remains virtually known only from single plant collections from each locality.  The var. ubomboensis is apparently from the locality south of Stegi as indicated to me personally by Capt. Keith.  This is no doubt also the same general vicinity from which the var. keithii also came.  An interesting collection is that of a glabrous light green variant similar to var. ubomboensis from the mountains near Barberton.  This seems to be a more geographically discrete element than the original from south of Stegi.  Also interesting is the olivaceous and stoloniferous variant reported to have been collected at Hoedspruit.  A striated variant in which the tubercles are white and almost confluent in transverse bands was recognised by J. Pilbeam (1983).  This does have some substance and appears to have been collected in the Pongola area and specifically along the Mozaan River, a tributary of the Pongola River.


a.  var. limifolia.

Distribution: 2531(Barberton): Saddleback Mt. (-CC), Stayner in KG33/71 (NBG); Near Pongola (-CC), Ross-Frames 13436 (PRE).  2631(Hlatikulu): Mbuluzi gorge, Blue Jay ranch (-BD), Buitendag 1145 (PRE), Bayer (NBG); Vimy Ridge (-CD), Karstens in PRE39431, in NBG56413.  2632(Bela Vista): Stegi (-CA), Keith in PRE39436, PRE39438; Ubombo Ridge, 13km from Stegi (-CA), Keith in PRE34862, in PRE61130, Smith 5693 (NBG); Ubombo N. of Pongola (-CA), Keith in PRE34863; 16km S. Stegi (-CA), Keith in PRE61128 in PRE61129; Abercorn Drift (‑CC), Keith in NBG69412, Smith 5740 (NBG); Ndumu (-CD), Symons in PRE7532; Kosi Bay (-DD), Cronwright in PRE34875.  2730(Paulpietersburg): Mozaan River near Moolman (-BA), Smith 7214 (NBG); 30km E. Paulpietersburg, Omdraai (-BD), Roos 4362 (PRE).  2731(Pongola): Slopes of Ubombo N. Pongola (-AC), De Wet in PRE34874 Near Pongola (-BC), De Wet in PRE34858; 7km W. Gollel (-BD), Dhlamini in NBG646/59.  2832(Hluhluwe): 8km W. Somkele, Smith 5692a (NBG).

Inadequately located: Zululand, Delagoa Bay, Marloth 4678 (PRE); SuldoSuavo, road to Moamba, Carvealho 755 (NBG).


b. var. gigantea Bayer, JS.Afr.Bot. 28:215(1962).  Bayer :118(1976).  Bayer :112(1982).  Type: NATAL‑2731(Louwsberg): Nongoma (‑DA), Bayer 112 (PRE).     

gigantea: very large.

The location of this variety is obscure and although I have occasionally seen large forms with the same small discrete tubercles which characterize it, I have not been able to determine their geographic origin.

Distribution: 2731(Louwsberg): Nongoma (‑DA), Bayer 112 (PRE).


c. var. ubomboensis (Verdoorn) Smith, JS.Afr.Bot. 16:3(1950).  Bayer :164(1976).  Bayer :69(1982).  Scott :33(1985).  H. ubomboensis Verdoorn, Fl.Pl.Afr. 21:t818 (1941).  Type:  SWAZILAND‑2632 (Bela Vista): 16km S. Stegi (-CA), Keith in PRE26392.

ubomboensis: from the Ubombo Mts.

This variety has its origin in several collections from near Stegi (including probably the var. keithii) which seem to indicate that it has no geographic base.  However, a very similar bright green glabrous form has also been collected by E. van Jaarsveld near Barberton.

Distribution: 2632(Bela Vista): 16km S. Stegi (-CA), Keith in PRE26392, 26392b.



Volume 2, Chapter 3:- Where to Haworthia limifolia

(This chapter was published in Aloe 40:2:41-52, 2003, and I have added an addendum with further explanation and discussion).

My first experience of this species Haworthia limifolia, was with a plant brought from somewhere in Zululand by my late uncle Frank Bayer.  He was involved with malaria control in then Zululand and Northern Natal.  He must have given the plant to my father while we were staying at New Hanover during the period 1950 to 1954.  I have always been attracted to Haworthia, but I did nothing more than admire the plant as it flourished in my mother’s care.  I never really concerned myself with the origin of the plant other than recalling my uncle telling that he had stopped for lunch by the roadside, and his Zulu assistant had come back to the car with the plant.  I had thought he said “near Nkandhla”, while my father later doubtfully remembered “Nongoma”.  It could have been “Ithala”.  I recall now that my uncle was stationed at Vryheid during that period and this is a small way inland from either Ithala or Bivane Dam, which feature later in this article.

Continue reading

Volume 2, Chapter 4:- Haworthia limifolia var. arcana Smith & Crouch

Haworthia limifolia is something of an enigma in that both it and Haworthia koelmanniorum are geographical rather isolated from the rest of the genus.  The latter is confined to a small area around Loskop Dam and Groblersdal in what was the old Transvaal province.  H. limifolia is much more widespread and occurs from the southern Kruger National Park southwards through Swaziland in to Northern KwazuluNatal.  It was described by Rudolph Marloth from a specimen very loosely said to have come from “west of Delagoa Bay”.  No one has ever managed to tie the species down in terms of geographic origin, and the Flowering Plants of Africa (55:24-29, 1997) description of the typical var. limifolia, depicts it as the var. gigantea.  My own fieldwork and observations of material of known field provenance show that this var. gigantea and the var. striata are surely synonomus.  Neither of them could truthfully said to be from “west of Delagoa Bay”

Continue reading

Haworthia limifolia – a conundrum or a lesson?

I was contemplating the problem I see in the way we approach classification, contending that there is no resolution to the endless addition of new names and arguments about their validity and usefulness unless we reach some agreement on method and purpose.  The confusion about names, descriptions and identifications that had arisen in 1947 came about for four reasons.  Small samples and inadequate exploration, too many unqualified experts, lack of insight into what species actually may be, poor methodology.

There is no excuse for this situation to continue.  Sampling is at a very high level and because the other three elements have not been removed, confusion is as great as before.  Many names may have been lost from sight, but there is no lack of new ones to replace them.  Experts are a problem and their qualifications even more so.  In my own experience I have communicated with many highly qualified botanists who have contributed absolutely nothing to the resolution of problems as basic as typification of names, or definition of the word ‘species’.   So what then contributes to qualification?  This is surely an unbridgeable problem when amateurs have dabbled so freely in plant classification with apparent success.  But their success has come from the failure of qualified botanists to put something more substantial in place than a nomenclatural code for binomials and nothing to indicate what those should mean, or who is competent to generate or change them.

Neither of those problems can be solved by me.  My contribution has been to see as much as I can of what others have seen, and then to myself explore in both the arena of the subject matter and the periphery of biology that touches it.  What I have done is to try and relate what I have read and been taught to what I have observed and experienced.  In this way “species” have become a reality as systems of populations and individuals that are spread in space and change with time.  Whether evolution is a fact or not I cannot say, but change certainly is.  Classification since Linnaeus, has always been associated with this change but not necessarily by everyone.  But in the time of Linnaeus, the sampling level was very low and anything new could be, or was, interpreted as a new species and diagnostics were put in place to aid identification.  Hence even the Latin binomial was a diagnostic.  As the sampling level has been raised so has the system come under increasing stress and it is a fallacy to think that only Haworthia is problematic.   However, this is the genus under discussion here and we have to admit that the diagnostics and the methodology do not work.

H. limifolia can be used as an example to illustrate the situation.

It is to all intents isolated from the rest of the genus.  We do have H. koelmanniorum sitting to the far north-west and then H. tessellata further to the south-west.  Otherwise the rest of the genus is in the Cape.  The issue of subgenus need not be seen as a consideration here even though the differences there are as great as any elsewhere in the family.

H. limifolia has a very wide distribution from Komatipoort in the north, westwards and southwards to Vryheid, eastwards and southwards to Gollel and then Stanger in KwazuluNatal.  The whole of Swaziland and a large part of northern KwazuluNatal may harbor this species.  We need to even ask…”Is it a species?”

The fact is that there are very different elements involved in respect of individual appearance and virtually all aspects of that individuality.  The possibility is that whole populations may be derived from vegetatively propagating individual clones.  Some plants are solitary and propagate with difficulty.  Others proliferate as vegetative offsets or by stolons.  The leaf shape and surfaces are highly variable.  The surfaces may be flat or channeled; absolutely smooth, to having raised wave-like wrinkles, to tuberculate with small individual tubercles, or the tubercles may be fused in groups and ridges.  These may be normal leaf-coloured or they may be white.  Flowers are extremely variable and so are the seeds and the capsules.  Flowering time seems to be opportunistic and in cultivation the plants may flower at any time, usually synchronizing during the summer months.

Do we really have any alternative but to see this as one single very large operating systems, despite all the real problems of isolation by distance and hence the associated problem of gene transfer between individuals and populations?  Is it sensible or even practical to insist or suggest that there must be Latin binomials that describe (in the early Linnaean sense) the situation or, more importantly, that there are diagnostic characters by which such single taxa (groups, species, varieties, forms etc) can be identified?

How can the situation we observe in this relatively isolated system and (species?) be extrapolated to the Cape where all three subgenera and the bulk of the species occur?

Recently two authors have proclaimed the value of floral morphology in resolving this ongoing disputation about the species.  What their actual view of what species are has not been spelled out but it appears to be a belief that simple structural (morphological) differences are what define them.  Behind this must lie the unconscious acceptance of the original definition of species that they were systems of interbreeding or potentially interbreeding organisms.  Thus flower structure is assumed to be linked to the breeding system and therefore very important in classification.  This view of species still seems to be the power behind classification although it has long been recognized that in plants inter-fertility may extend across generic difference.

In Haworthia the flowers, even across the sub-genera, are fairly similar and especially in regard to size.  Within the sub-genera, the similarities between what are considered to be species, is remarkable.  There are even exact similarities that can be found in what are truly geographically and vegetative different populations.  Flowering time seems to suggest some kind of additional factor but even here it is evident that breeding can occur between populations with flowering times 6 months apart.  It has to be considered that flowering time differences like this might even be a characteristic of a species.  This has in fact been shown for H. marginata and H. minima that surely can otherwise seen to be different systems.

The pollinating agents are insects and it seems as if flower size may be linked to that fact as it is so consistent across the genera and whatever are thought to be the species.  Here I just present a number of pictures of the profiles and faces of flowers from 5 populations H. limifolia of which some are from different clones in those populations. A difficulty in viewing flowers is that they change with age, and place on the peduncle.  They only last about three days and it is thus difficult to compare flowers of exactly the same age and condition.  Of course these are poor pictures but no essential and detail is lost here by poor focus.  But it is quite obvious that there is bigger difference between the two flowers (profiles) of the plants from Ithala than there is between them and other populations.  Consider too that it can be very difficult to distinguish any of these flowers from those of other species in the subgenus Hexangulares.  This essay is not submitted as conclusive evidence and argument and I will enlarge on this in a following essay.