I commented rather defensively, harshly and disparagingly on Ingo Breuer’s article in Haworthiad 10:1, and now here is Haworthiad 10.2 with the second part. There are also three good letters responding to my request for comment on the requirements for another handbook.
In my frustration with the other authors who also write about Haworthia, it has become obvious to me that I have to look at my own motive and what reward I seek. H. Jacobsen wrote an article (C&SJ(US) 46:230, 1974) entitled ‘Why I wrote books’, but it disappointed me as I do not think he addressed the issue very well. He cited talent as one motive for writing and, as a second and only alternative, the professional scientist writing in pursuit of his own perfection. I do not think that either of these are correct.
It is generally acknowledged that in life we try to create an impression. Calvin (of Calvin and Hobbs) put it ‘God put me on earth to accomplish a certain number of things..’. Here we have two other views. Surely another is simply to earn a living, and there are probably more.
I was brought up with books as a source of information and as a means of communication. At university these were also the crucial issues and as a working (professional?) scientist, writing came with my job. A research project is planned on paper for approval, the results are reported on paper, and if noteworthy they are submitted for publication, critiqued or refereed in the process and then published. A scientist is expected to publish or perish. While he may thought to be pursuing some ideal of perfection, he may have nothing but his own image in mind and do nothing but make a fool of himself. It never really occurred to me that I should do anything else but communicate on paper any new information which came to me. My early experience with Haworthia indicated fairly clearly that they were not well understood. In my first encounter with Col Scott I had no problem in accepting that he would do the writing of a book, and I would do collecting. Unfortunately it came to the point that I did not feel that we were in any kind of agreement about the fundamentals. There was also a problem with access to the Compton herbarium record, on which it is not necessary to elaborate.
I did not think even then that I was the best person to do any kind of taxonomic botany. I wrote at the level I was working at in a botanic garden, with the needs of the garden collections in mind. I had abandoned a career in science or in the public sector as well as any hope of a career in the private sector. The first handbook was very poor and I definitely am ashamed of it. But it was written as a lead-up to the book by John Pilbeam and to hopefully influence the direction being taken by Col Scott. The second handbook was a necessary attempt to improve upon the first, and as a brief concise account of the genus as I had by then come to know it.
The appearance of John Pilbeam’s book, then that of Charles Scott and also of reviews which did not seem to me to make any good sense of the subject, disillusioned me completely.
I cannot say that I retired from Haworthia. The need was for more information and much more extensive collecting, which I could not justify in the work situation. I also felt that while there was not actually a basis for a really comprehensive and truthful statement, there was also not an audience for one at that time. I also took the opportunity to test and explore variation in many other plant genera, often in collaboration with botanists pursuing various and diverse aims.
The Newer Handbook is now the product of many quirks of fate and before it appears I really would like to try and clarify several issues, or have these clarified for me.
Some explanation needs to be sought for the problems of understanding that arise between people involved in the classification and naming of Haworthias. This is apart from the problems of individual ego and motive. We live in a world where often a complicated, incomprehensible, high-tech and untruthful account is valued more than a simple straightforward one. Unfortunately we are not all intellectually gifted and we each have to find our own truth at our own level.
Darwin’ ‘Origins of species’ is considered to have shaken the foundations of western philosophy and religion. Whatever that means and what other implications there might be, different kinds of plants have fascinated various people for one reason or another. So part of our social fabric is that there are individuals and groups that entertain themselves to some degree or another by their interest in plants.
The recognition of kinds of plants and their variations is the foundation of this interest and the communication that is associated with it. The problem is that of different perceptions and opinions; and of distinguishing and discriminating between true and false accounts. The significance of Darwin is that we base our concept of kinds of plants on the basis of species for which we may not have a clear definition. Despite the fact that evolution implies continuous ongoing change, we seem to have a basic assumption that it has all reached a fixed end-point. Generally it is assumed that species are clear distinct entities, while at the same time it is loosely recognised that ‘evolution is still in active progress’ when these distinctions fail. This seems to be the case in Haworthia. Controversy is still the order of the day where no two authors can come to agree on anything.
There are two ways in which the form of a plant may be determined. The one is the genetic basis and the other is the phenetic one. The former is the driving mechanism of evolution and speciation, and the latter is the physical expression as influenced by the environment.
In the earlier days of scientific exploration and discovery it was assumed that species would be fairly well-defined elements. The general definition of the species was that it comprised groups of breeding or potentially interbreeding individuals and the whole emphasis was on the genetic context of isolation. This simplification persists to the present day and the ideal of taxonomic botany is to find a classification which neatly provides the ‘box’ of common parlance, into which each species can be placed. That this is is a total myth, is now recognised implicitly by nearly everyone and yet seemingly ignored or disregarded in our perpetual hope of a better solution and explanation for things.
I am by nature a collector, and all kinds of plants have always interested me. Haworthias came into my life at a very early stage and I am very lucky that circumstances allowed this interest to shape my life. Controversy has been significant in my career and I have been frustrated in not being equiped adequately for the intellectual high-tech scene. The end-point of my career was being demanded to produce a high-tech answer, truthful or otherwise, for the imponderables of vegetation state and change.
The situation with Haworthia has been fairly similar. In my work with insects (noctuid moths), I came to see that classification was not a simple box arrangement, and that physical characteristics were not the necessarily an easy clue to relationships between living things. Despite this, I also tried to find some ideal ordered relationship on which I could build my interest in Haworthia. Initially I tried to find some physical way in which I could explain and separate kinds of Haworthias. It came increasingly apparent that this was not possible and that an explanation at one point of a distribution range for closely related species, was not true at another. I tested this viewpoint very thoroughly in Oxalis which are considerably better endowed with ‘characters’.
My classification therefore abandoned the hope that there were clear boundaries, either genetic or phenetic, between clearly defined and different species. Without recourse to the vague and hackneyed ‘evolution is still in active progress’, I resorted to a classification based on the realities of distribution and geographic relationships.
What this means is that I have recognised that there is a basic genetic flux which has and is following the Darwinian scheme. It has been shaped into a number of different kinds by various isolationary mechanisms over geographic space. Geology, topography, latitude, longitude, and climate as well as biological pressures from other living things are all important.
My concept of a species in Haworthia is that of a group of plants varying, genetically and morphologically, nearly continuously in time and space. The discontinuities are very clear at one level which I recognise as a subgeneric one, not very clear at another which I call the species level, and very vague at still another which I call mostly variety. I cannot see a natural and easily definable level between subgenus and species. The nature of variation below species, and even at the level of the species I formally recognise, is very irregular and erratic.
What I thought a classification had to do was to accommodate all the known records (primarily herbarium specimens), and also in a way which would be likely to include (predict) all the unkown ones. Since 1976, I have with the help of several people like J.D. Venter, G. Marx, P.V. Bruyns, Derek Tribble, R. Kent, D. Cumming and also of Haworthiad and its editors and contributors, been able to test the classification and its predictions. I can only say that I am comfortable to a degree and there are very few new records which suggest to me that my scheme is under stress. Whatever stress there is comes from inaccurate and untruthful statements by other writers and commentators, and from people who seek and insist on order where there does not seem to be any.
I value the comments made by the 3 correspondents in Haworthiad and can say that their requirements will largely be met. One of the difficulties is, however, of adequate illustration. I really do not think that it can be accommodated in a book and envisage rather articles such as were started in the National Cactus and Succulent Journal. I have yet to meet someone who has been prepared for the range of variation which one finds in Haworthia. Populations are not equally variable within themselves and there are some areas where there is no clear pattern whatsoever.
So to comment briefly on Ingo Breuer Part 2.
1. H. glauca var. herrei fa armstrongii. This is a very inaccurate piece of writing and to try and rectify it I quote from the 1976 book and add an extract from the new text.
1976 – ” H. glauca var. herrei fa armstrongii comprises a localised population northeast of Uitenhage ‑ based on a vegetatively propagating clone. J.R. Brown said that this form was close to ‘ H. herrei var. depauperata‘ in general appearance and colour, but easily separated on the basis of having non‑confluent tubercles only on the margins and keel. The one distinguishing feature of fa armstrongii is colour ‑ both in habitat and in cultivation it tends to be darkish green and lacks the glaucous sheen of the typical species. The fa armstrongii can be superficially compared with H. fasciata except that it is caulescent. It also occurs in close proximity to H. fasciata fa browniana. Schneider (1972) showed that fa armstrongii and H. glauca were both distinguished by the absence of waxing in the stomatal chambers, and this too suggests that they belong together. In a very recent publication, Brandham and Cutler (1981), show that the fa armstrongii is a pentaploid of probable hybrid origin, showing closer affinities to H. coarctata than to H. glauca.”
1997 – “There has been some cytological evidence to suggest what relationship the fa armstrongii has to the species. However, the results are conflicting and it is not easy to question this (or scanning electroscopy for that matter). Brandham and Cutler (1981) report H. glauca to be ‘mostly hexaploid….but two .. were aneuploids’. Their discussion is ambivalent as on p515 of the article they suggest there is insufficient evidence to decide its (armstrongii’s) parentage, whereas on p540 they state that it cannot be a variety of H. glauca because it is pentaploid and the species (H. glauca) is hexaploid. The authors refer to two populations of H. glauca studied, but the list of materials cites only one (actually the var. herrei), and one for the fa armstrongii. Riley (1959) reported H. glauca to be tetraploid and thus a pentaploid within the species is to be expected. Finally it is now known that H. glauca is common in the area close to where fa armstrongii was originally collected. A recent visit to the site suggests that it no longer occurs there. Brandham and Cutler also looked at epidermal character where they conclude that fa armstrongii was closer to H. coarctata than to the Steytlerville collection of H. glauca. This result has to be viewed in proper perspective. It is an hypothesis based on a limited range of material and it conflicts with Schneider’s results (quoted above).”
Nowhere did I say I recognised it because of its chromosome number. Breuer does not cite a reference for other studies which report 2N = 14, 40 and 42, and I doubt that these numbers he cites are for armstrongii anyway. This would make his later statement about exclusivity also incorrect. If Ingo is to write technically, I think he should pay more attention to the correct use of technical literature and source citation.
2. H. habdomadis and ilk. Another piece of confounded and inaccurate writing. The first sentence is not reflected in what follows. H. lockwoodii differs in having a flatter relatively unkeeled leaf and the leaf tips generally have to venation whatsoever in contrast to fig.2 of inconfluens where the veins run well in to the leaf-tip.. The problem is that the two species come together in the area immediately south of the Witteberg mountains and according to J.D. Venter they become confused in the field. Breuer does not say what the differences in the inflorescence are to which he refers. I am well aware of the robust peduncles thse species can have and also the stout florets. Unfortunately there are often weak clones, or even populations, where these are not any longer characteristic. The var. habdomadis is a sandstone ecotype and one would expect plants in nutrient poor soils to have correspondingly less vigourous growth and weaker inflorescences. I regard fig.3 of Breuer not to be the var. habdomadis at all, but a weakly denticulate form of H. unicolor var. venteri. Fig. 5 is a glabrous variant of the same (to become arachnoidea var. nigricans). Fig. 6 is a weakly denticulate form of H. unicolor var unicolor (to become H. mucronata var. mucronata). The var. morrisiae is characterised in the field by its bright emerald green colour and by the very rich brown leaf tips. In the new Handbook I have adopted a different arrangement whereby the name H. unicolor is discarded in favour of H. mucronata (to follow Col Scott) and a new relationship is offered for the varieties of H. habdomadis and of the old H. unicolor. It is a very difficult problem.
3. H. koelmaniorum. An acceptable conclusion very recently confirmed in the way it should be and already published anyway – by field observation (by Essie Esterhuyizen).
4. H. limifolia. I do not know how Breuer can so easily reconcile his statement ‘the rest of the forms do not require taxonomic separation’ with the rest of his inadequate comment. I find it irrational. I would prefer to say that ubomboensis should be regarded as a subspecies if it can be shown to have a relatively independent existence in terms of a population structure and distribution. The Pretoria records seem to make it quite clear that it was a single clone from south of Stegi. A recent fascinating collection of a fairly similarly glabrous form (population based) from near Barberton was given to me by Ernst van Jaarsveld. I think that Ingo has actually suggested in private communication, that only one rank below species is desirable and he prefers subspecies. I think the term is too formal and carries too much connotation of scientific and rational import. I like variety as an informal assessment of a relationship which can have wide tolerances, for the wide range of variations which are a fact of life in any well collected plant group. The Barberton collection could better meet the requirements for some formal taxonomic status and it probably has no direct link with the Stegi collection. I understood this to indicate ‘polytopic’ – the independent evolution of two very similar, but geographically isolated elements, from the same genetic stock. The var. gigantea is a little different from the other varieties in having a relatively flat leaf in cross-section, as opposed to the more general triangular cross-section. I have reservation about the var. striata of Pilbeam which may in the wider sense of variety be included in gigantea. I have seen several plants which are similar to the single clonotype and these may have come from the Louwsberg area of Natal. I consider this another attempt to present a new solution to a known problem but based on a much weaker foundations – just pointless. There is simply a need for a thorough field investigation by someone sufficiently competent to express a good opinion.
5. H. magnifica. Here again is an attempt to suggest a solution based on nothing whatsoever. Ingo has no field experience of this group and his five sentences say nothing more or less than that he does not know anything about the species. Not writing anything at all would have been the obvious option.
6. H. margaritifera. More pointless and inaccurate comment. Firstly there are 4 recognisable species involved and minima is indeed minima and no where near the size of maxima (margaritifera). The distribution area is not compact and the sentence ‘many locations have become extinct due to earlier colonisation ‘ is absurdly unnecessary. What is the information that is known today and which implies that maxima and minima should be joined? Is it Col Scott’s distribution maps? I kept poellnitziana alive and with a purpose because I did not know what the situation was. In my writing on the Robustipedunculares I am sure that I have indicated that there is quite enough variation still extant to account for all the old names as well as a good many more.
7. H. truncata. If Ingo ever manages to demonstrate that there are floral diferences between H. maughanii and H. truncata, I predict that he will be a nominee for a Nobel prize of some kind.
8. H. reticulata. It is beyond me to understand how anyone can still write this kind of nonsense, or what excuse there is for it. Ingo is known for his extensive literature collection and presumable he has copies of my articles on H. reticulata and H. herbacea. It is inaccurate to say that it reaches the habitat of H. herbacea on its western, what to say that it reaches the habitat of H. cymbiformis at its eastern border. Is this any basis at all for discarding var. hurlingii, and has it any relevance ?. To me it demonstrates an ignorance and ineptitude which should debar the perpetrator from making any comment at all.
9. H. retusa. I suppose I must concede something. However, there is an explanation for the anomaly which Ingo suggets viz. that I recognised varieties in H. retusa but not in H. turgida. The problem is explained sonewhat in a recent manuscript which I wrote for ALOE. This suggests that H. retusa is really not a role player as an interactive species in the southern Cape. It comprises a few populations in the Riversdale area. These are really derivatives of the greater role player H. turgida which has a very wide distribution and a correspondingly great variation. The elements dekenahii and acuminata did not fit comfortably in retusa and I maintained their identity for that reason. There was no similar problem with the turgida varieties.
10. H. rycroftiana. Ingo writes mostly as if there was some reality in the sections to which he refers. If he had drawn on my writing he would have considered the option of its relationship with H. mucronata (the old unicolor).
11 & 12. H. scabra. The varieties starkiana and lateganiae were not considered discrete just because of ‘distinguishing’ marks. There is a significant difference in point of occurrence as well as in overall growth form. The leaves of the latter are long and slender and sharply angled in cross-section. They are both significant variants but I did not have the assurance that they were only variants of the greater species scabra. The existence of a var. morrisiae does pose a problem related to the whole question of when to recognise varieties or not. I think this is going to remain very arbitrary and I think we all are going to have to accept that the recognition and non-recognition of varieties is part of a communication process.
13. H. unicolor. I sympathise with Ingo for the effort he put into battling his way through this one. Readers will have to wait for the Handbook and see how I have tried to deal with this problem. It is not an easy one at all and involves H. arachnoidea, co-occurrences, and the question of translucence of the leaves.
14. H. venosa. This is also highly inaccurate. Venosa and granulata are not uniform in appearance. The latter is fairly restricted in distribution along the middle Breede River (at least 6 populations), while the latter certainly is not (Karooport near Ceres, to Verlatenkloof near Sutherland – 7 populations recorded). The formation of stolons is not a consistent character and occurs erratically in several different species.
15. Summary.
Ingo’s objectives are said to include pointing out inconsistencies. Yes, I think there are indeed inconsistencies but I think Ingo has just added a whole lot more and not made any useful contribution to their solution. The real inconsistencies may in quite large part be –
a. intrinsic to the genus,
b. deliberate because there is or was no good information to suggest a better solution,
c. related to the communication network of the time,
d. not resolveable by inaccuracies and conclusions based on ignorance.
If Ingo was drawing attention to the need for further work in the field, he could not have done it better by the want of real information in his article. I would say that what the articles have really done is to demonstrate the inaccuracy of our reasoning processes. I would like to ask people like P.I. Forster and A. Bulworth, who appear to have perspicacious minds, how they react to Ingo Breuer’s writing?. The final point regarding the status of Astroloba and Chortolirion could be this. The generic status is really not that important and I do not think that anyone working at this superficial level should even venture a comment. It is just as rational to say that Haworthia should be separated into 3 genera (a solution with which I would concur). The attempt in Taxon to examine the relationships of the genera through cladistics, should be a absolute warning (or a plea) for really ignorant persons not to meddle with things unless they really can make a meaningful and truthful contribution.