There is a very curious parallel in the problems and aspects of classification and identification of haworthias and chameleons that I personally find a bit mystifying. Classification of succulent plants and especially Haworthia is often done outside of mainstream science, which means that it is not done by trained professional botanists, who can stand in the intellectual arena with academics and degreed intellectuals. It has been pointed out to me that this is the problem in Haworthia and also why professionals do not want to attempt to resolve the issue because the nomenclature has been so confused and complicated by all the bungling that has taken place. Having been on the fringes, and indeed even failed attempting to cross the bridge into academia, I find it very difficult indeed to reconcile my life experience in the amateur arena with what I encounter from the professional one. This is because I have observed an ever widening gap between the perceptions of the non-science individual compared to that of the professional. My interest in chameleons is simply another aspect of my interest and passion for living things and I am distressed that the knowledge and understanding of these fascinating animals, as with Haworthia, is so clouded by ignorance and confusion.

It is a bit problematic for me to explain why I think classification and plant names is such a big issue and why it is important to me, but fundamental to me is trying to understand and know something about plants and animals as an extension of my life experiences.

Recently a very welcome and valuable booklet, Chameleons of Southern Africa has been authored by Krystal Tolley and Marius Burger and published by Struik Publishers (Cape Town). What makes it so interesting for me beyond my interest in the animals themselves, is the remarks the authors make in the preface in respect of writing the species accounts. The problem of finding identifying characters was a key issue and there is constant reference to confusion. There is of course no statement about what a chameleon species actually is and there is this curious statement, “There are few consistencies of identification characteristics, and we admit that we sometimes initially identify a chameleon by means of ‘gut feeling’.” What makes this so curious is that there is nothing in the booklet to indicate there is any more to the mystery of identification and how a final identification could be made to possibly help the even more than average individual to arrive at own conclusion.

In Haworthia (as indeed is the case in many plant genera) the description of species has been based on a basis of simple visual (morphological) difference, with barely any consideration of variation and in the total absence of a definition of a species beyond the assumption that if they look different they are different (species).

One has to ask what is it that is referred to as ‘gut-feeling’? I have often heard and read that classification and taxonomic statement is a matter of opinion and felt compelled to respond by quoting a writer who made the distinction between mere opinion and informed opinion. “Gut-feel” may vary from the instant response of the ignorant to a fleeting image, to the deeply considered intuitively driven reservation of another who is highly informed and who has a vast bank of experience, images and literature to draw from.

The significant thing is that any decision-making in the chameleon booklet has been facilitated by results from DNA analysis – *please see the footnote. Now I must make it clear that I am not attacking the book or the authors. I want to draw attention to the problems inherent in a methodology which amounts to nothing more than saying “trust me”, and the very serious problem that own judgment is by implication quite useless. As a very knowledgeable observer wrote to me “Our own eyes and brains are seemingly no longer good enough to serve the world – well that is what the RNA-DNA lab technicians are trying to tell us.” There have been several attempts to evaluate the classification of Haworthia by DNA sequencing that do little more than ring alarm bells. Without any pretence at understanding all the convolutions of genetics and the repository of DNA in all its forms, I would have to ask again, what does one do to organize one’s scepticism about the results that are thrust upon one?  If the total facies that an individual organism presents to the eye and senses is not a reasonable representation of the information encoded in the DNA, then we are absolutely helpless.  Could one say that we have been put into this creation without the sensory capabilities to distinguish and discriminate anything?

Recently I submitted photographs of two chameleons to an authority for identification stating that I thought they were not the Karoo Dwarf. The reply I received was “It looks like the Karoo Dwarf. What makes you think it is not?” Apart from the fact that I had already given reasons, the truth is that my correspondent could not possibly have said why he thought it did look like the Karoo Dwarf, thereby shifting the responsibility back on to me.  Are we stuck with the sad and unavoidable fact that unless one can get a DNA result, one cannot know what you are looking at?

What has now happened with respect to Haworthia DNA where we have the same dreadful problem of identification and understanding? I have elsewhere (see Haworthia Update 3 (Ch 3: p92) discussed the papers by Messrs. Treutlein et al  and I summarily dismiss what I think may be unpublished results by another source I must politely keep to myself. There was a rather immodest statement by Messrs. Wallace and Noll published in Haworthiad that promised to unfold the problems of Haworthia classification by DNA analysis. The results have never been disclosed and there is a disconcerting failure by the authors to respond to any inquiry. More recently Dr. Syd Rhamdhani has also been busy sequencing the DNA of a range of Haworthia and species of related genera. The work was initiated in an attempt to relate variability to habitat diversity, but to preclude a total negative result it was deemed advisable to widen the scope to ensure that the methodology was applicable at a broader scale. The results demonstrated, at least from my personal perspective, a complete failure to relate the DNA sequences with the realities of the classification of at least the semi-rational mind of an experienced and informed observer.

Group1A

Group 1Aa

Code	Collector Number	Taxon name	Subgenus	Locality	Source
54	MBB 7128	angustifolia var. baylissii	Haworthia	Wellsgate, E. of Kirkwood	Field
47	MBB 6843	marumiana var. reddii	Haworthia	Inverbolo, Stutterheim	Sheilam
58	MBB sn	truncata v. maughanii	Haworthia		Sheilam

Group 1Ab

Code	Collector Number	Taxon name	Subgenus	Locality	Source
35	JDV 90/57	monticola var. asema	Haworthia	S. of Kruis River	Field
40	JDV 97/128	mucronata var. mucronata	Haworthia	SE of Barrydale	Sheilam
6	SR&TDsn 4	angustifolia	Haworthia	Mountain Drive (Gtown)	Sheilam

Group 1Ac

Code	Collector Number	Taxon name	Subgenus	Locality	Source
32	EvJ 17548	cymbiformis var. transiens	Haworthia	Horee	Field
52	MBB 7032	mutica var. mutica	Haworthia	Rietfontein	Sheilam

Group1B

Group 1Ba

Code	Collector Number	Taxon name	Subgenus	Locality	Source
30	PVB 7047	zantneriana	Haworthia	N. of Redcliffe, Willowmore	Sheilam
62	MBB sn	reticulata	Haworthia	Bosfontein	Sheilam
48	MBB 6875	maraisii	Haworthia	Cogmanskloof, Ashton	Shadowlands
51	MBB 6985	rossouwii var. calcarea	Haworthia	De Hoop

Group 1Bb

Code	Collector Number	Taxon name	Subgenus	Locality	Source
33	GM 623	marxii	Haworthia	Rooi Nek Pass S. of Laingsberg
36	JDV 91/91	cymbiformis var. setulifera	Haworthia	Rainbow Valley	Sheilam
4	SR & TDsn 2	cooperi var. gracilis	Haworthia	Mayfair (Gtown)	Field
18	SR & BBsn Site 5- 27/6	cooperi var. viridis	Haworthia	Kaboega, Dassiekop NE of farmhouse
15	SR & BBsn Site 3- 26/6	cymbiformis var. obtusa	Haworthia	Glen Avon falls	Field
16	SR & BBsn Site 2- 27/6	cooperi var. ‘puberula’	Haworthia	Kaboega, W of Wilgefontein	Field
1	SR&TD 719	cymbiformis var. incurvula	Haworthia	Platosvale	Field
21	SR & BBsn Site 3- 18/10	cooperi var. viridis	Haworthia	Kaboega, De Plaat	Field
24	SR & BBsn Site 1- 19/10	cooperi var. spec.1	Haworthia	Kaboega, Kaboegapoort	Field
23	SR & BBsn Site 1- 19/10	cooperi var. spec.3	Haworthia	Kaboega, Kaboegapoort	Field
25	SR & BBsn Site 1- 19/10	cooperi spec. 4	Haworthia	Kaboega, Kaboegapoort	Field
22	SR & BBsn Site 4- 18/10	cooperi var. viridis Grassland form	Haworthia	Kaboega, W of Wilgefontein	Field
38	JDV 93/45	cymbiformis var. obtusa	Haworthia	Kagasmond	Sheilam
27	SR & BBsn Site 1- 19/10	cooperi spec.2	Haworthia	Kaboega, Kaboegapoort	Field
39	JDV 94/98	decipens var. xiphiophylla	Haworthia	W. of Coega	Sheilam
64	BBsn	aristata	Haworthia	E of Hopewell farm	Field

Group 1Bc

Code	Collector Number	Taxon name	Subgenus	Locality	Source
13	SR & BBsn Site 1- 18/10	cooperi typical	Haworthia	Kaboega, Gertpiets area	Field
26	SR & BBsn Site 2- 19/10	cooperi var. viridis	Haworthia	Kaboega, back of the Dam (Spekboomberg)	Field
44	MBB 6791/4	cooperi var. picturata	Haworthia	Andrieskraal	Sheilam
3	SR & TDsn 1	cooperi var. pilifera	Haworthia	Mayfair (Gtown)	Field

Group 1Bd

Code	Collector Number	Taxon name	Subgenus	Locality	Source
63	MBB sn	parksiana	Haworthia	Mossel bay	Sheilam
50	MBB 6983	rossouwii var. rossouwii	Haworthia	Soutkloof, Napier	Sheilam
67	MBB sn	pulchella var. pulchella	Haworthia	Constable Station	Sheilam
43	MBB 6762	puchella var. globifera	Haworthia	Touwsberg, Little Karoo	Sheilam

Group2A

Group 2Aa

Code	Collector Number	Taxon name	Subgenus	Locality	Source
46	MBB 6829	Poellnitzia rubiflora	NA	Rooikloof	Shielam
71	AJ290302	Poellnitzia rubiflora	NA		Genbank

Group 2Ab

Code	Collector Number	Taxon name	Subgenus	Locality	Source
73	MBB sn	Aloe bowiae	NA	???	Sheilam
9	SR&TD S3-29/8	Aloe speciosa	NA	Plutosvale	Field
10	SR&TD S6-29/8	Aloe africana	NA	Bothas Hill (Grahamstown)	Field
11	SR&TD S7-29/9	Aloe lineata	NA	Stones Hill (Grahamstown)	Field
72	AJ290289	Aloe vera	NA		Genbank

Group 2Ac

Code	Collector Number	Taxon name	Subgenus	Locality	Source
7	SR & TDsn 6	attenuata	Hexang.	Plutosvale	Field
34	JDV 85/145	scabra var. morrisiae	Hexang.	Schoemanspoort	Sheilam
31	EvJ 16840	attenuata var. glabrata	Hexang.	Collywobbles	Sheilam
56	MBB 7179	viscosa	Hexang.	Constantia, W. of Willowmore	Sheilam
65	MBB sn	venosa var. venosa	Hexang.	Breede Riv., Swellendam	Sheilam
41	JDV 1942	attenuata var. attenuata	Hexang.	Soutkloof, Addo	Sheilam
28	Clark & Pienaar 120	venosa var. tessellata	Hexang.	Farm Hughdale 161, Kikvorsberge (Noupoort Region, N. Cape)

Group 2Ad

Code	Collector Number	Taxon name	Subgenus	Locality	Source
66	MBB sn	reinwardtii	Hexang.	Wesley	Sheilam
68	AJ290299	kingiana	Robusti.		Genbank
5	SR & TDsn 3	coarctata	Hexang.	Ecca Reserve (Gtown)	Field
42	MBB 6380	pungens	Hexang.	Braam River	Sheilam
17	SR & BBsn Site 3- 27/6	sordida	Hexang.	Kaboega, E of Olifantskop	Field

Group 2Ae

Code	Collector Number	Taxon name	Subgenus	Locality	Source
19	SR & BBsn Site 2- 28/6	nigra	Hexang.	DePlaat, Kaboega	Field
61	MBB sn	venosa var. woolleyi	Hexang.	Springbokvlakte	Sheilam
74	AJ290298	Gasteria liliputiana	Hexang.		Genbank
45	MBB 6809	fasciata	Hexang.	Gamtoos	Selecta Succ.
57	MBB 7695	glauca	Hexang.	15 km S of Kaboega Farm house, Darlington Dam to Kirkwood	Field
20	SR & BBsn Site 1- 27/6	glauca	Hexang.	Kaboega, W of Wilgefontein	Field

Group 2Af

Code	Collector Number	Taxon name	Subgenus	Locality	Source
59	MBBsn	glauca	Hexang.	????
12	SR sn 30/6/08	glauca	Hexang.	Gertpiet se Kraal	Field

Group 2Ag

Code	Collector Number	Taxon name	Subgenus	Locality	Source
60	MBB sn	Hybrid arachnoidea X blackburniae v. graminifolia	Haworthia	Schoemanspoort	Sheilam
55	MBB 7145	limifolia var. glaucophylla	Hexang.	Three Sisters, Barberton	Sheilam

Group2B

Group 2Ba

Code	Collector Number	Taxon name	Subgenus	Locality	Source
2	SR843	Astroloba foliolosa		Kaboega, De Plaat	Field
69	MBB 6756	Astroloba bullata		??	Sheilam
49	MBB 6952	marginata	Robusti.	N. of Ashton	Sheilam
70	AJ290292	Astroloba foliolosa			Genbank

Group 2Bb

Code	Collector Number	Taxon name	Subgenus	Locality	Source
53	MBB 7096	pumila	Robusti.	Konings River	Sheilam
37	JDV 92/51	blackburniae	Haworthia	Voorbaat	Sheilam
29	EA 1211	kingiana	Robusti.	Great Brak	Selecta Succ.

Group 2Bc

Code	Collector Number	Taxon name	Subgenus	Locality	Source
8	SR&TD S2-29/8	Aloe tenuior		Committees Drift Junction	Field

Dr. Rhamdani has so far generated results from DNA sequencing using the nuclear ITS and chloroplast trnL-F markers. The former, on the evidence of the “species” considered, seems to give a good view of the generic possibilities, with not enough information with respect to the species. The trnL-F spacer results contain several spurious associations if my existing classification (Bayer 1999) and geographic distance are valid considerations. A poster presentation of the results has been published in Alsterworthia International 9(3)13-17.) Dr. Rhamdani has been kind enough to send me the product of an alternative manipulation of his results  that I tend to equate with an attempt to display the relationships of samples as they would be in geographic space. One can see from the groupings (see tables below) that there are gross anomalies that cannot possibly be taken to have classification truth. Group 1Aa for example has H. angustifolia, H. marumiana ‘reddii’ and H. truncata ‘maughanii’ nested together. Group 1Bd includes H. parksiana and H. rossouwii. The presence of H. kingiana in two different groups viz 2Ad and 2Bb has to be dismissed as an identification error where I would stake my reputation on EA1211 rather than AJ290298. Each group can be tediously discussed, but one needs to consider that the actual focus of the exercise was to explore the geographic relationship of DNA sequences below species level. Far too much was expected of a still poorly understood technology.

Coming back to chameleons, I have the problem that, being interested in the creatures, I want to know what they are. Having familiarized ourselves (my wife and I) with the Cape Dwarf (Bradypodium pumilum) we came into contact with the (then) Robertson Dwarf (B. gutturale now Little Karoo dwarf). The distribution given by W R. Branch and the descriptive detail did not seem to match what we were experiencing in the field and we concluded that the Robertson Dwarf and the Karoo Dwarf (B. karroicum) were surely the same species.  We then encountered what we thought was the Western Dwarf (B. occidentale) and were derailed by the opinion given us from a DNA analysis of the “population” observed, that it was a variant of the Cape Dwarf viz. a “renosterveld form”.  What we were told did not satisfy our “gut-feel” however profound and true it may have been or is. Then the booklet was published that confirmed our conviction that the Robertson and Karoo Dwarves were being confused (The true Karoo dwarf is now said to equate the E cape dwarf B. ventrale!), but not throwing any more light on the question of the “Renosterveld form” of the Cape Dwarf.

We then learned of a chameleon in the Gouritz River valley. On seeing these we concluded that on morphological grounds they were not the Little Karoo Dwarf and had to fall within the ambit of the Swartberg and Eastern Cape Dwarves as accepted by Tolley and Burger. These are the chameleons “identified” by an authority as the Karoo Dwarf despite our submission of evidence why it was not. This led us back to the published DNA work and a paper by Tolley et al that discussed specifically the Eastern Cape Dwarves and a number of variants suggested being new species. My experience in Haworthia generated the strong “gut-feel” that this may be quite wrong as wide field experience suggests that species are highly complex systems that defy the conventional taxonomic approach. Further sampling would quite probably generate a host of still other variants and require explanation. It is quite obvious that the chameleons cannot be identified with certainty using morphological characters and this is equally true for Haworthia. My experience is that it is true for many plant genera. Nevertheless, I think that this is because morphological criteria are so heavily relied on with not enough reference to geographic distribution and a rational species definition, as opposed to an open-ended species concept. In another paper by Tolley et al, the two species B. melanocephalum and B. thamnobates are shown to be genetically identical, despite the fact that they are known to be morphologically different and geographically separated. Thus it was a pleasant surprise to find another article (Miller & Alexander (2009) that probes this question and speculates that it is possible that the DNA sequences used in the analysis may not be representative of the full genome. The available results of DNA sequencing for Haworthia suggest to me that those sequences are also not adequately or accurately representing a true or reasonable phylogeny. A problem I then also have is that a phylogram is a simple two-dimensional tree that is being drawn to illustrate the relationship of elements that are spread in at least two-dimensional space derived from change with time. In my opinion far too much is thus inferred from complex statistical procedures that generate a wholly theoretical diagram wholly beyond penetration by ordinary skepticism. I would suggest that the so-named bootstrap and posterior probability values may be subject to the same apparently many vagaries of multivariate statistical procedures that are not perhaps fully understood or even recognized by those who use them.

Dr. Rhamdhani suggested that I mention some of the limitations of the sequencing process that he lists:

1. mostly extant samples are used as these are only available for analysis (except in some cases fossil can be sequenced).

2. small sample size.

3. use of a few selective markers.

4. hybridization and incomplete lineage sorting distorts and complicates patterns.

5. appropriate analytical tools have not been formulated to handle the data and the above issue.

6. manipulation, misuse etc. of species concepts to justify desired results – this can be related to the “gut-feel” issue that this article raises.

I did complain about the problem of probing results published in complex technical papers when the results conflict with my “gut-feel”. The response was that “our eyes and our intuition are frequently tricked by the complexity of the real world…”  This is flatly not true for Haworthia and I strongly suspect that it is not true for chameleons either.  In Haworthia we are tricked by a very simplistic approach and the failure of taxonomic botany to serve interested parties. My kind authority I referred chameleons to also wrote “I know the genetic information may seem very confusing, but it is essential to remember that the genes tell us about genealogy and not morphological or behavioural similarity.” I have said before that the traditional approach of  taxonomy was always to relate morphology, and whatever other information one could acquire, to phylogeny. It is an unlucky fabrication foisted on us by DNA sequencing that this is a new direction for classification. To totally believe that, say, less than a thousand base pairs on a tiny section of mitochondrial DNA is giving what must be seen as a final answer that defies the doubt of an observer is perhaps too much to ask.

An end point in my chameleon inquiry came when Dr. Tolley conceded (as most DNA experts now seem to be doing) “DNA is obviously not a final answer, and everyone that understands well how evolution works, already agrees on that. However, when used wisely, in connection with morphology (which is subject to phenotypic plasticity, strong selection, and local adaptation) it obviously can be useful in understanding the evolutionary history of organisms. We can use DNA to give us clues as to the evolutionary lineages, which sometimes can be the same as what we humans like to call “species”. For me, “species” is partly an artificial construct because evolutionary processes do not suddenly jump, and divide one thing into two species, there are gradual changes, and sometimes (more often than we care to admit), they do not fit in the box we have labelled “species”. She exposes the fundamental problem that no one probably understands properly how evolution works nor do any of us have the wisdom to use DNA information to best advantage.

References

Ashadee, K.M. & Alexander, G. L. 2009. Do dwarf chameleons (Bradypodium) show developmental plasticity.  African Zoology 44:45-54.

Tolley, K.A et al. 2004. Phylogenetics of the southern Africa dwarf chameleons, Bradypodium (Squamata: Chameleionidae).  Molecular Genetics and Evolution 30:354-365

Tolley, K.A. & Burger, M. 2007.  Chameleons of Southern Africa. Struik, Cape Town.

Treutlein, J. et al.  2003a. Phylogenetic relationships in Asphodelaceae (subfamily Alooideae) inferred from chloroplast DNA sequences (rbcL, matK) and from genomic fingerprinting (ISSR). Taxon 52:193 207.

Treutlein, J. et al. 2003b. Evidence for the polyphyly of Haworthia (Asphodelaceae Subfamily Alooideae: Asparagales) inferred from nucleotide sequences of rbcL, matK. ITS1 and genomic fingerprinting with ISSR-PCR. Plant Biology 5:513-521.

Note. The Chamelion DNA study rests on the sequencing of about 40 individual specimens as opposed to the 1000 plus used for the former morphological revision. ♦