Latin names definitely mean different things to different people and my contention is that the real essence of these names should, in addition to their many other usages, be in the relation of plants to their origins, relationships, behaviour and imagined future. A classification can only have the authority that experience and knowledge permit, and be really evaluated and understood by persons with the same evidence before them. In coming to closure I have been exploring some more and with my wife Daphne, made two finds which further convince me that we have to come to a classification by agreement. However, the requirement is that species are seen to be highly complex systems with none of the rigidity and inflexibility that nomenclatural rules imply, nor any of the egocentric authoritarianism that a history, of which I have been a part, suggests.

I have recently written two papers. One deals with the H. notrieri complex, and the other with H. pygmaea. In writing the latter I interwove the comments of observers whose opinions I value, and conclude with the proposal that the elements acuminata, splendens, dekenahii, argenteo-maculosa, fusca and vincentii all be included in the super-species H. pygmaea. Concomitantly I suggest the H. turgida and its variants be absorbed in H. retusa. Behind these two propositions, I was intending to re-enforce my view expressed in Haworthia Update Vol. 3 that H. mirabilis too be re-structured to include magnifica, maraisii, heidelbergensis and any associated varieties. The reason of course been the problem of continuity, however that is understood; and I suggest it is best understood by familiarity with the plants in the field and some recognition of biogeographical factors and the role these may play in driving change, adaptation and evolution.

In Update Vol. 3 I illustrate examples from many of the populations known to me, and also draw attention to populations in the lower Breede River valley which are significant. My most recent excursion has in fact been to try and resolve the dramatic juxtaposition of variants that we found there. We have Adam Harrower’s collection from Sandhoogte and Chris Burger’s collection at Buffelsfontein, both south of the Potberg. Then there are my collections from Stoffelsrivier and one nearer to Infanta. A curious addition I have yet to see in the field is a collection by Ismael Ebrahim of SA Biodiversity Institute from southwest of Vermaaklikheid. That collection resembles my Stoffelsrivier collection and so provides the inevitable link to paradoxa.

At two localities near Melkhoutrivier, between Stoffelsrivier and the Infanta collection, we found plants which have to be related to those mentioned above. The plants are highly variable, as I have now come to expect, and individuals can be likened to badia, acuminata and mutica. The surfaces have a curious sheen and may be incipiently spiny as the Sandhoogte and Buffelsfontein plants also are. The paradox is that we are also driven to the conclusion that plants at Ziekenhuis are the same species.

What re-enforces this seemingly improbable statement, apart from my lengthy dissertation on the matter in Update 3, is a second find we made southeast of the Bromberg. This is the locality for what I suggest is H. rossouwii var. elizeae, and also for several variants of what have, in the past, been classified as maraisii and mirabilis or variants thereof. My new find is a remarkable array of large very dark-green (nearly black), plants which dramatically enforce the close association between mirabilis and maraisii as well as with heidelbergensis. These plants include the elements of both “species” as they might have been understood. What is striking is again the huge variation in the population that I have come to expect in this complex. The plants do lack the opaque dots on the under-leaf surfaces while the upper surfaces may be clearly windowed or opaque. They bring forcibly to my mind a sequence of populations from Heidelberg in the east to Verdwaalskloof near Riviersonderend in the west that I suggested forged the interface of mirabilis and maraisii.  Coupled with this is the array of populations both north and south which enforce that continuity and that also lead on to heidelbergensis.

My recommendation is that the nomenclatural code be summarily assigned a secondary role to a dispensation which is more flexibly attuned to the realities of a truly asymmetric species structure with more emphasis on reflecting field relationships. The illusion of reality that the ranks of genus, subspecies and variety provide, should be admitted. Genera should be recognised for their historical value, but for species I would suggest that much more attention be given to the huge asymmetric and asynchronous variability that underlies capacity to change and adapt. With the adoption of H. mirabilis as a superspecies, variants can be indicated by the addition of any other epithet in inverted commas; thus H. mirabilis ‘maraisii‘, H. mirabilis ‘heidelbergensis‘, in the knowledge that there is in fact no clear and infallible distinction. ♦