In this blog post I will look at how DNA studies are changing how we think about what is and what isn't a Haworthia.
The genus Haworthia has historically been subdivided into 3 groups:
| Genus | Sub-genus | Representative plants |
| Haworthia | Robustipedunculares |  |
| Haworthia | Hexangulares |  |
| Haworthia | Haworthia |  |
In this scheme of things both the soft leaved windowed haworthia and hard leaved Haworthia were both considered Haworthia. I always wondered about the validity of this grouping because, except for the flowers, the hard leaved Haworthia felt closer to gasteria to me.
DNA studies can provide more concrete evidence of relatedness of organisms than morphological traits traditionally used to determine relatedness. Our genomes are like a book filled with blueprints for building our cells, tissues and bodies. Genes are the blueprints. The same is true of plants. Genes and the surrounding sequences mutate (the base pairs in the sequence change) over time and these changes can be used to measure relatedness between organisms. The more base pair differences between organisms the more distantly related they are. For example depending on the precise comparison used human DNA is about 95% identical to chimp DNA (a more closely related primate) but only 85% identical to mouse DNA (more distantly related mammal). By comparing the genetic sequences of Haworthia and it's relatives, scientists can determine what is more closely related to what.
IIn 2003 a study was published where molecular techniques were used to probe the relationships of these plants:
MOLECULAR PHYLOGENETICS
Phylogenetic relationships in Asphodelaceae (subfamily Alooideae) inferred from chloroplast DNA sequences (rbcL, matK) and from genomic finger- printing (ISSR)
Jens Treutlein1, Gideon F. Smith2, Ben-Erik Van Wyk3 & Michael Wink1
1Institut für Pharmazie und Molekulare Biotechnologie, Universität Heidelberg, Heidelberg, Germany. jtreut- le@ix.urz.uni-heidelberg.de; wink@uni-hd.de
2Office of the Research and Scientific Services Director, National Botanical Institute, Private Bag X101, Pretoria, 0001, Republic of South Africa. gfs@nbipre.nbi.ac.za (author for correspondence)
3Department of Botany, Rand Afrikaans University, P.O. Box 524, Auckland Park, Johannesburg, 2006, Republic of South Africa. bevw@na.rau.ac.za
Two independent lines of molecular evidence have been studied to explore phylogenetic relationships in the family Asphodelaceae. Genomic fingerprinting by ISSR (Inter Simple Sequence Repeats) analysis was com- pared to sequence data of the chloroplast genes matK and rbcL. Molecular data indicate that some long-estab- lished taxonomic concepts would have to be re-evaluated. The subfamily Asphodeloideae clusters as a sister group to a distinctly monophyletic Alooideae. However, several Alooideae genera, including Aloe and Haworthia, are apparently not monophyletic. From a molecular point of view, Haworthia can be divided into two distinct groups that agree closely with the current subgeneric classification: a monophyletic group includ- ing species of subgenus Haworthia, and a second polyphyletic group with the subgenera Hexangulares and Robustipedunculares. This second clade includes Poellnitzia, Astroloba, Gasteria and even one Haworthia- like aloe (Aloe aristata). In the polyphyletic assemblage currently classified as Aloe, several smaller clades can be recognised, often reflecting morphological, chemical and geographical discontinuities. The tree aloes (sec- tions Aloidendron and Dracoaloe) and climbing aloes (series Macrifoliae) appear to have separated early in Alooideae, while other groups (e.g., the flavonoid-containing group and a Madagascan group) are embedded within and amongst other genera. Chortolirion clusters with the grass-like aloes (section Graminialoe Reynolds, syn. Leptaloe Berger), A. boylei and A. verecunda, on a well-defined branch. The current taxonom- ic system clearly does not reflect the phylogenetic affinities and relationships amongst the succulent genera Aloe, Astroloba, Chortolirion, Gasteria, Haworthia, and Poellnitzia.
The authors concluded that:
“From a molecular point of view, Haworthia can be divided into two distinct groups that agree closely with the current subgeneric classification: a monophyletic group including species of subgenus Haworthia, and a second polyphyletic group with the subgenera Hexangulares and Robustipedunculares.”
In other words soft leaved haworthia are their own thing. But they did not propose changing the taxonomic groupings so the status quo remained.
About 10 years later more studies were published using more comprehensive and extensive techniques. I won't go into all of the studies but they reach similar conclusions. One of the most comprehensive publications of this kind was a 2014 paper from John Manning's group at the South African National Biodiversity Institute.
Systematic Botany (2014), 39(1): pp. 55–74
© Copyright 2014 by the American Society of Plant Taxonomists DOI 10.1600/036364414X678044
Date of publication 02/05/2014
A Molecular Phylogeny and Generic Classification of Asphodelaceae subfamily Alooideae: A Final Resolution of the Prickly Issue of Polyphyly in the Alooids?
John Manning,1,2,5 James S. Boatwright,3 Barnabas H. Daru,4 Olivier Maurin,4 and Michelle van der Bank4
1Compton Herbarium, South African National Biodiversity Institute, Private Bag X7, Claremont, 7735, South Africa. 2Research Centre for Plant Growth and Development, School of Life Sciences, University of KwaZulu-Natal, Pietermaritzburg, Private Bag X01, Scottsville, 3209, South Africa.
3Department of Biodiversity and Conservation Biology, University of the Western Cape, Private Bag x17, Belville, 7535, Cape Town, South Africa.
4African Centre for DNA Barcoding, University of Johannesburg, P. O. Box 524 Auckland Park,
2006, Johannesburg, South Africa.
5Author for correspondence (j.manning@sanbi.org.za)
Communicating Editor: Allan Bornstein
Abstract—Phylogenetic analysis of plastid (rbcLa, matK, trnH-psbA and the trnL intron) and nuclear (ITS1) sequence datasets in a wide sampling of species of Asphodelaceae: Alooideae provides a generally well-resolved phylogeny. Among traditionally accepted genera only Astroloba and Gasteria are supported as monophyletic. Species of Haworthia are distributed among three clades corresponding to the cur- rent subgenera. Aloe s. l. (including Chortolirion) segregates into six, well-supported clades corresponding respectively to sections Dracoaloe (= Aloidendron), Kumara + Haemanthifoliae, Macrifoliae, Aristatae, Serrulatae, and the remainder of the genus. The first three clades are retrieved as early branching lineages, whereas A. sects. Aristatae and Serrulatae are strongly supported as members of a clade including Astroloba + Haworthia subg. Robustipedunculatae. We examine possible options for recircumscribing the genera of Alooideae as reciprocally monophyletic entities. Although morphological and molecular data are consistent with expansion of Aloe to include all members of Alooideae, we accept and implement an alternative option maintaining historical usage in the group as far as possible. Astroloba and Gasteria are retained as currently circumscribed; Haworthia is restricted to H. subg. Haworthia; the genus Tulista is accepted for members of H. subg. Robustipedunculatae, with the new combination T. minima; and H. subg. Hexangulares is treated as the genus Haworthiopsis with the new combinations H. koelmaniorum, H. pungens, and H. tessellata. The genus Aloe is restricted to the clade comprising the ‘true aloes’, with Aloidendron, Aloiampelos, and Kumara accepted as segregates, the latter broadened to include A. haemanthifolia as K. haemanthifolia. Aloe aristata is segregated in the monotypic genus Aristaloe as A. aristata and Aloe sect. Serrulatae is treated as the new genus Gonialoe with the species G. dinteri, G. sladeniana, and G. variegata.
A link to the manuscipt is here and a paper they submited to the Alsterworthia is here.
Much of this paper deals with re-groupings of the genus aloe which are very complex. But I will focus on the findings as they relate to Haworthia.
Their conclusion as it relates to Haworthia:
"Haworthia is restricted to H. subg. Haworthia; the genus Tulista is accepted for members of H. subg. Robustipedunculatae, with the new combination T. minima; and H. subg. Hexangulares is treated as the genus Haworthiopsis with the new combinations H. koelmaniorum, H. pungens, and H. tessellate."
What they and others before them found is that Hexangulares and Robustipedunculares are more closely related to Gasteria, Astroloba and some aloes (like Aloe aristata which is often mistaken for haworthia).
Because of the findings of these molecular studies the groupings of the genus Haworthia have been changed. What is and what isn't a Haworthia has changed. What used to Haworthia sub-genus Hexangulares is now a new genus under the name Haworthiopsis and what used to be Haworthia sub-genus Robustipedunculares is now a new genus under the name Tulista. Haworthia pumila, which when first discovered was called Aloe pumila, is now Tulista pumila.
Soft leaved windowed Haworthia are now the only true Haworthia and they may be more closely related to the new genus Kumara which contains two former Aloe species: plicatilis and haemanthifolia. So Haworthia retusa may be more closely related to what used to be called Aloe plicatilis (now called Kumara plicatilis) than it is to what used to be called Haworthia pumila (now called Tulista pumila). It's a pretty remarkable paradigm shift. Some plants don't fit neatly into the new categories. For example Haworthia koelmaniorum is placed in the new genus Haworthiopsis but some of the DNA data show it's also close to Tulista.
So to summarize, hard leaf haworthias are no longer considered Haworthia. Haworthia pumila is now Tulista pumila. Haworthia venosa is now Haworthiopsis venosa. Time to change your tags.
| Genus | Representative plants |
| Tulista |  |
| Haworthiopsis |  |
| Haworthia |  |
Briiliant- very clear but clearly representative of a process rather than a process! The three Gonialoe ‘speices’ are clearly geographically distinct taxons of a ‘continuum’ species whilst ultimately, all of the Haworthias which were placed in the old retusaue group will be reduced to two or three, or less! H Koelmaniorum as you say, will probably require generic ‘promotion’ and some other Victorian name or even neologism (Tulistopsis anyone) will be resurrected! Brilliant!