(assisted by R.D. Kent and S.A. Hammer)

Let me not detract one iota from an exceptional and remarkable compilation of descriptions. There is no need for me to go through the book in detail because in respect of it being a compilation, it is outstanding.

What concerns me is what it portends and what it holds for the future of Haworthia. The book follows a Vol.1, which was reviewed in my book Thoughts on Haworthia (Spiderwalk, 1999). It is quite clear to me that Ingo Breuer is in a sense re-inventing the wheel. There is very little information in the book new to me and most of the work was available to Smith in 1947 and certainly to me in 1976. A prime problem I thus have with the work is that it is so firmly rooted in the past. It is thus a threat to the present and holds no promise for the future. The work does present problems, and portends disaster.

Some of the implicit interpretations in this work are possibly just as dubious as some of mine and often more so. I see little purpose in generating change for so small a result. I find the foreword by Prof. G.F. Smith disturbing. It demonstrates some of the malaise I describe in other chapters. Smith’s foreword ranks with a similar foreword to Breuer’s Vol. 1. by Prof. Ihlenfeldt which I commented on in Thoughts on Haworthia. I have said from extensive experience with many genera, that Haworthia does not offer any challenges which are not extant in other genera. It is my indictment of taxonomic botany that taxonomists are not better acquainted with natural diversity and the complex reticulate relationships of species.

To suggest that this book is going to help the reader in anyway towards understanding the actual species of Haworthia as Prof. Smith does, is unfounded. Classification is integral to language, organisation of thoughts and any communication. It is necessary for science and not necessarily science in itself as taxonomists seem to take science to be. To say that Breuer’s work comes as a refreshing change is to beg the question “From what?” In the context of the literature of Haworthia, the work by Breuer takes us back sixty years, and Smith is supporting this backward step. As I will point out, the work is fraught with potential disaster. To further say that this work allows the reader to make his or her own decisions on the application of “some of the controversial names” and further “assist in maintaining an acceptable level of stability” is just mis-statement and a poor assessment of the situation. I will set out to show why.

The purpose of a classification is for a single reasonably competent mind to review a group in such a way as to propose a system which unifies nomenclature and circumvents the need for individual judgements. If we each have our own interpretation of names, what purpose does classification then serve? It is like each of us having our own dictionary and own language. Unless authors driven by the communal need set out to respect nomenclatural stability, there will be no such thing. Names of plants are concerned with species. I have said in many places that if we use names, especially Latin binomials, we need to know what a species in fact is. This is something which not many taxonomists, nor aspirants to the task, seem to be aware of. Any botanist surely should know that the only real interpretation of an assemblage of plant specimens is to examine them in their actual field context. This compilation of descriptions by Breuer, however well intentioned, is self-evidently worthless in that respect and all the illustrations nearly so. This is particularly so in respect of the juxtaposed illustrations where Breuer has attempted interpretation.

The name “asperula” can be taken as an example. I discarded it for a very good reason.  von Poellnitz used the name for at least four different elements and Scott for five. I am well aware that Hayashi and Esterhuizen (private communication) are, or were, of the opinion that it should be in the place of my concept of H. magnifica. What Breuer has done is to juxtapose J.R. Browns illustration and concept of what is almost certainly H. pygmaea as Aloe asperula with Salm-Dyck’s neotype and illustration. For him to have made this visual match is in my opinion a bad one even in the context of Haworthia. In my experience of Haworthia, I would say without question that Salm-Dyck’s icon is probably nearest to scabrid forms of H. retusa and, as Hayashi and Esterhuizen actually implied, specifically to a population from east of Riversdale.

Regarding H. emelyae, H. correcta and H. bayeri; it is not absolutely clear if Breuer has done a volte-face. He did intend correcta to supplant bayeri, and now it appears rather that picta may supplant emelyae. The origin of H. emelyae is quite clear from the exchange between Smith and Ferguson. I wrote this up in 1979 (part 4 of the Nat.Cact. Succ. Jl.).  The plants were not in fact collected by Mrs. Ferguson, but by a Mrs. Le Roux, from near VanWyksdorp. The photos by Fourcade that Breuer uses to illustrate emelyae are quite misleading. One is H. reticulata and the other is H. turgida (Gamka River – and probably the lower Gamka which is the Gouritz River). Breuer has again demonstrated this curious inability, exhibited also by Col. Scott, to make visual comparisons. I liken it to the inability of some people to read maps. If one looks properly at the type illustrations of emelyae and correcta, one can conclude that they are more similar to each other, than are the Fourcade to emelyae. Breuer has simply assumed that Fourcade’s identifications were correct.

One of the reasons I regarded emelyae and bayeri as possibly the same species, was because I had two small collections of bayeri from De Rust and from south of Oudtshoorn.  These were so scabrid that the difference between them and the Uniondale plants, was as great or greater than the difference between plants at VanWyksdorp and the predicted variation of emelyae between Rooiberg and Muiskraal (collections by Dekenah, and in recent years my own) and Springfontein (multifolia) and also its var. comptoniana from Georgida.

Breuer has a photograph by G.F. Wagner on page 570N, entitled “H. mirabilis sensu Bayer (mundula)”. I do not know how he arrives at this caption. It demonstrates a complete ignorance of all my writing and the problem I have stated of look-alikes, and demonstrates also the problem of identification which Breuer seems to assume does not exist. The illustration can superficially be taken to resemble mundula I will agree, but it is quite obvious that the plant cannot be and I certainly have never used the name “mundula” for this H. magnifica population at Barrydale. One needs to look at the illustration of mirabilis in Curtis’ Bot Mag. – the neotype – and see how well this compares with the Barrydale illustration in Breuer’s book. Then one must look under at H. emelyae var. beukmannii and see the various illustrations there from which Breuer also draws his epitype for mirabilis.  Unfortunately Smith’s published illustration of his mundula is a very poor rendering of the element as it occurs in nature. Additionally there is a poor comparison of this with the real mundula photographed by Esterhuizen – of one of the less-lined yellowish clones. It can be noted that as a general rule, mirabilis tends to colour red to yellow under stress whereas maraisii (and magnifica) become darker and duskier.

There is actually a remarkable story of change attached to mundula which I should relate in another place. The plants are not now as they were in 1969 when I first saw them, and of course they might not be anything like what Otzen might have seen. It does raise an interesting point about the taxonomic process – is it designed to produce a classification with which we deal with the living phenomena we see, or do we really need a system which allows us to organise a very limited sample of dead and dried specimens with precision?

I class these Barrydale plants as either magnifica or maraisii although in my work it is clear that I have problems with separating maraisii from magnifica. Thus originally I specifically refer to these Barrydale plants as magnifica (see Haworthia Handbook, 1976 p100, under asperula). This is one of the problems I have in trying to discuss Haworthia with anyone.  There is this reticulate relationship and one can find odd plants in populations which more closely resemble other ‘species’ than that to which they belong.

The problem with mirabilis as opposed to mundula is one which I refer to as a weakness in nomenclature. Breuer has thus designated an epitype for H. mirabilis from Skuitsberg whereas I select the Curtis Ker-Gawler illustration and state the similarity to Smith’s mundula. Therefore a local variant becomes H. mirabilis var. mirabilis, and the whole general body of mirabilis has to be treated as the var. triebneriana. In private correspondence Breuer makes it obvious that this aspect of nomenclature has eluded him, by asking me what can have happened to H. mirabilis if I now typify it with a specimen named mundula!  An unpublished manuscript of Breuer’s in my possession suggests that he wanted to uphold the var. beukmannii of mirabilis also from Skuitsberg. This is done in his book. It does not make sense and any change at species level creates a host of complications at subspecies or varietal level (apart from the added complication of rank priority). Also concerning my species mirabilis in a similar way are the two illustrations used by Breuer to depict the var. rubrodentata. This particular variety is actually represented by a few specific slender leaved clones on a particular steep north facing clay slope indicated to me by the original collector G.J. Payne in 1970. There are at least four other localities west of Greyton where mirabilis occurs and Breuer’s IB5129, coming from any of these, is not comparable with the itemised Triebner 1143.

Three last examples of this incapacity to apply visual imagery with any credibility are:-

1. In the case of H. rossouwii where Brown’s 1953 illustration is obviously not comparable with the von Poellnitz lectotype. This particular picture is virtually identical with Breuer’s own illustration of H. emelyae var. major IB5539.

2. The illustration of H. stiemiei can be shown to be ambiguous and controversial and this will emerge from a manuscript of mine dealing with four intergrading elements from the Zuurberg north of Kirkwood. Breuer’s own depiction of this species could be fruitfully compared with elements as far afield as Wolfkloof, Robertson; and many places in the Baviaanskloof, as well as the Elandsriver Valley. While it no doubt could be that Breuer’s illustration may represent the same element that the type of stiemiei does, the actual resemblance of the two, in the sense of the book, is minimal.

3. What is disturbing is the fact that Breuer wants to apply the name denticulata (Bavaainskloof) here, and the visually and chronologically comparable aristata to Barrydale (my mucronata). I am comfortable with the former because I consider this is where aristata and denticulata BOTH COULD actually belong. In fact I am not comfortable at all with the imagery with which Breuer relates his aristata to the lectotype or to Triebner’s plant. I agree there is reasonable merit, but only in the absence of any prior attempt to better interpret the element. Scott’s views and imagery are/were never any better than Breuer’s. Breuer is remiss for having learnt nothing from history.

It occurs to me to now relate an example of the potential confusion that exists in Haworthia by explaining an example of what Breuer might well do. One can argue on the basis of both type illustration and description that bijliana of v. Poelln is actually either Rooivlei (Bredasdorp) H. heidelbergensis, OR H. serrata from northwest Bredasdorp. In terms of Haworthia these would both be valid arguments and I think this should be pre‑empted.  In an unpublished manuscript I uphold H. rossouwii (which is a much later name than H. bijliana), over serrata. But it could be argued that bijliana also is this element.

In Feddes Repert. (1936) v. Poelln. cites the species bijliana from Klawer and Eenriet, and in 1937 cites it again as from Springbok. In this last publication he cites a specimen also from Springbok, another from Bonnievale, another from Ezeljacht Oudtshoorn, and also sinks H. fergusoniae here which in fact was supposed to have come from ‘near’ Grahamstown. Thus Breuer has:

H. bijliana v. Poelln. in Feddes Repert. 27:134 (1929); Bredasdorp, Mrs. van der Bijl. Type van der Bijl s.n. in (B). Not preserved. Lectotype designated Breuer 1999, unpublished photo. icon. (B)  (my copy of the description does not have the word “doubtful” after Bredasdorp as Breuer adds in his book).

Funnily the original illustration of bijliana is in my opinion very like rossouwii, and fergusoniae like that of tenera ‑ whereas later treatment of the names puts them as arachnoidea. Breuer’s treatment thus in many instances demonstrates the inherent self-evident fact that types in Haworthia are, and always will be ambiguous. Types cannot be used to confidently identify and name plants outside of the geographic origins of those types. The point Breuer makes about ambiguity on XIII is actually just some of the juristic jargon which is supposes to lend credibility to nomenclature. It is a fact of life that a type in Haworthia is ambiguous unless it is very clearly illustrated and the locality properly stated. Thus in the case of emelyae it is self-evident from Breuer’s own use of the illustrations, that the lectotype was highly ambiguous.

Without making a point of finding any trivial errors:

On page 784 Breuer has Olifantsdorn for Olifantsdoorn, page 784 Breuer has Conwitz i.s.o. Gouritz and on page 786 Stromsvley for Stormsvlei (this is an interesting error because v.Poellnitz cites a Strydomsvlei when it may have been Stormsvlei (as it happens this is also in connection with H. triebneriana, and potentially damaging because Stormsvlei is the origin of the vars. depauperata and pulchra. All these fall within the broader context of mirabilis and mundula of either Bayer or Breuer).

There is merit in trying to arrive at another interpretation of H. helmiae and I think Breuer in this one case is possibly right. For some reason Scott was adamant that Mrs. Helm had told him she had collected this species at Schoemanspoort, which would have made my interpretation reasonable, but Scott often bent the truth to meet his needs If one looks in any way at the four illustrations that Breuer present for helmiae, one must be struck again at the incongruence of the match. Breuer’s lectotype could hardly be more ambiguous when held against the Fourcade photograph. The point should be made that Breuer seems to attach an omniscience to the Brown and Fourcade illustrations which is wholly at variance with the facts – as noted by myself in 1976.

The re-introduction of the name integra for my rycroftiana is probably acceptable on the basis of the new evidence (if true). I regarded ‘integra‘ as an unknown in 1982 with inadequate evidence to support it. Scott salvaged the name for what in my opinion is arachnoidea var. nigricans. Almost any specimen of any of the similar ‘soft’ Haworthia species from diverse populations, preserved in alcohol, could be taken to be a mucronata variant if so desired or required. This particular one with the dark tips evident to the leaves could in my opinion be from outside the Gamka Valley from a place like Ockertskraal or Muurkees. The species mucronata is any event so variable that the suggestion that integra be upheld as a discrete species from mucronata, is of very doubtful value. As it is I have enough evidence to show that arachnoidea and mucronata can often not be separated.

I think it would be useful to know if Breuer’s IB numbers refer to plants which he had any real familiarity with in terms of their field context. I am aware that Breuer is first in the queue for plants collected and propagated by myself which come to be distributed in the trade. This creates a problem because propagation has always been largely by vegetative means – and mistakes occur in transfer. Thus the trade, and collections, tend to be dominated by proliferous clones and they may be misnamed and wrongly accessioned as to origin. The variability of the population is wholly obscured. Evidence of Breuer’s capacity to match visual images, suggests to me that he would not be able to find any pattern unaided by modern interpretation. Hence the reference to re-invention of a wheel.

There is a problem with a H. gigas – the plants from Amalienstein are familiar to me and they are not quite the same rigid plants with brown spination that is so characteristic of the Laingsburg plants. Treating them as synonymous with my H. arachnoidea var. scabrispina removes the geographic significance of the taxon. It is not in the least significant that they look-alike. Attaching nomenclatural significance to this similarity is wholly unnecessary and misleading.

There will be a problem with aristata as Breuer and I interpret them so differently. There is in fact a problem with gracilis as well which Breuer may not be aware of. However, the dickering over nomenclature here is not going to resolve the problem of what anyone does to understand the relationship between what I have called gracilis, cooperi, blackbeardiana and cymbiformis, and which extends to pringlei, decipiens, xiphiophylla and even mucronata. Using names in any other way is not going to solve any problem and will simply add to the confusion when the problem has to be faced. There is no way that one can actually separate Breuer’s aristata (my mucronata var. mucronata) and his mucronata (my mucronata var. inconfluens) as species. This ONE element is confounded with H. arachnoidea in many places and how anyone is going to find names to accommodate the associated variation in any new arrangement will be fascinating to see.

I have written several manuscripts which deal with specific issues in this context and these are included in this volume of essays. These will show the difficulties which Breuer’s work presents, and will show that Prof. Smith’s use in the foreword of the term “golden age” is for a work which actually helps to usher in a blackout. I agree wholeheartedly with Prof. Smith that Breuer’s work as a librarian is beyond reproach, but for helping to reorganise Haworthia it has very doubtful value. ♦