The Polyactium section of Pelargonium mostly features in the news due to the poaching epidemic as tonnes of P. triste tubers are dug up annually and shipped abroad. However, it is also really interesting as it is so incredibly diverse and occupies a number of ecological niches, from fynbos to tropical Africa. Traditionally, the section included tuberous species from the summer-rainfall regions, such as P. luridum, P. schlechteri, as well as three taxa with fimbriate (deeply split) petals, i.e. P. afrum (previously caffrum), P. bowkeri and P. schizopetalum.

The latter were moved to a new section Magnistipulacea by Röschenbleck and Albers [1], with the last three occupying a subsection of their own, Schizopetala. While some taxa in Polyactium were lately reinstated, e.g. P. schlechteri [2], some have curiously disappeared despite synonymy never having been clearly demonstrated, such as P. amatymbicum.

In 1995 [3], P. amatymbicum was considered as a separate taxon, although in vol. I of Pelargoniums of Southern Africa [4], published in 1977, van der Walt says “In the past P. schizopetalum and P. amatymbicum have been regarded as different species. P. schizopetalum is supposed to have less flowers per inflorescence, and more deeply incised leaves with a denser pubescence, than P. amatymbicum […]. Intermediate forms occur and it seems likely that they are conspecific.” On p. 44 of the book, P. amatymbicum is in fact illustrated. Despite having written this in 1977, van der Walt seemingly changed his mind and later refers to P. amatymbicum as a separate taxon in 1995.

It is not known what the views of Röschenbleck and Albers were in this regard in 2014 [1], although Albers who authored the chapter on Pelargonium in Eggli’s Handbook in 2002 [5], maintained P. amatymbicum as a separate taxon. Curiously, in [1], the specimen STEU 1873 from Hogsback, was taken for their DNA analysis. Hogsback is where P. amatymbicum is typically found, about 150 km away from the closest typical P. schizopetalum, in Addo. Thus, we should understand Röschenbleck’s analysis and the resulting conclusions in reference to P. amatymbicum, and not to P. schizopetalum.

The last source, to my knowledge, to provide some more clarity on this issue, was R. Knuth [6]. Although he lists a number of taxa, which are currently considered synonymous with P. luridum, he is remarkably clear about P. amatybmicum on p. 367: “Species affinis est P. schizopetalo, quocum congruit foiliis lobatis er petalis scissis. In P. schizopetalo autem duo lobi infimi foliorum usque ad rachim incisi et folia utrinque pubescentia sunt; umbella huius specie floribus multo numerosioribus praedita est.” Chiefly, the main difference between the two is that in P. schizopetalum, the two lowest lobes of the leaves are incised to the rachis and the leaves are pubescent on both sides. Knuth claims that the umbel of a P. amatymbicum has up to 20 flowers, while P. schizopetalum has up to 10 (data provided in the key, pp. 352-353). In the taxon description (p. 367), Knuth states that the upper leaf surface of P. amatymbicum is glabrous, which van der Walt [1] reduces to P. schizopetalum having “denser pubescence”.

Anyone who has tried to determine Polyactium taxa will know that the leaves are often the character of key diagnostic importance, cf. the key in [7], where the vestiture, margin and lamina shape play major roles. It thus bears looking into Knuth’s claim of P. amatymbicum having a different leaf shape and hairiness to P. schizopetalum.

Fig. 1: P. amatymbicum close to Hogsback.

In 2025, I was lucky enough to visit Hogsback and see P. amatymbicum in leaf (Fig. 1). I visited the area a couple of times previously, but the plants were neither in leaf nor flowering. The plants grow at about 1500 m a.s.l., on grassy slopes, which is remarkably different to P. schizopetalum, which prefers stony situations, in a narrow belt between George and Addo. The margins are indeed either entire or shallowly incised, never all the way to the rachis. Most of the leaves appear to be quite broad, and a systematic analysis of the lamina length/width ratio for P. schizopetalum and P. amatymbicum might show a further clear separation in the future.

Fig. 2: P. amatymbicum in flower, ©Simone Liefferink, CC-BY-NC 4.0, iNaturalist 255763743.

From iNaturalist, we know what that the flowers of the Hogsback population look like (Fig. 2), and indeed they are very similar to P. schizopetalum (Fig. 3), and while the inflorescences look somewhat more numerous, even Knuth doesn’t use this character to clearly divide the two species.

Fig. 3: Sweet’s illustration (lectotype) of P. schizopetalum [8].

Apart from the margins of P. schizopetalum normally being incised to the rachis (Fig. 3), another key character dividing the two taxa is leaf pubescence. According to Knuth, the laminas should be glabrous adaxially. Microscopic analysis confirms that the upper (adaxial) surfaces of P. amatymbicum laminas are covered in sessile glandular hairs only, while those of P. schizopetalum are also covered with long (up to 1.5 mm), patent glandular hairs and long, patent non-glandular hairs in addition (Fig. 4).

Fig. 4: Adaxial lamina surfaces of: left – P. amatymbicum (GER900, Hogsback) and right – P. schizopetalum (GER405, N Knysna).

The petal colour of the two taxa is generally similar (Figs. 2, 3). In P. amatymbicum, the five petals appear to be equally coloured, while in P. schizopetalum, the upper two petals seem to lack intensive markings. However, petal markings in Polyactium/Magnistipulacea do not seem to be consistent, generally, hence this may not be a reliably separating character.  

In conclusion, P. amatymbicum (Eckl. & Zeyh.) Harv. and P. schizopetalum Sweet do seem to show some consistently different traits, meriting the separation of the two taxa to be upheld until a more systematic study possibly demonstrates differently.

	P. amatymbicum	P. schizopetalum
Lamina margin	Incised shallowly	Incised to rachis
Lamina vestiture adaxially	Sessile glandular hairs	Long patent glandular and non-glandular hairs, Sessile glandular hairs
Flowers per inflorescence	<20	<10
Distribution	Hogsback	Narrow mountain belt between George and Addo

 

The sections of Polyactium and Magnistipulacea seem to require closer study. The recent re-emergence of P. schlechteri and the unclear treatment of P. amatymbicum as well as the unclear synonymy of P. woodii, show that some more literature and herbarium research might be necessary. On the other hand, some taxa, such as P. pulverulentum and P. lobatum, where populations with unusual characters, such as fertile stamen number, nectar tube length and flowering time, all normally considered as characters important for the separation of Pelargonium taxa, emerged in the recent years.

By Matija Strlič, Ljubljana, Slovenia

 

Reference

1. Röschenbleck J., Albers F., Müller K., Weinl S., Kudla J. (2014), Phylogenetics, character evolution and a subgeneric revision of the genus Pelargonium (Geraniaceae). Phytotaxa 159, 31-76.
2. Retief E., Craib C., Condy G. (2007), Pelargonium schlechteri. Flowering Plants of Africa 60, 84-88.
3. Maggs G. L., Vorster P., van der Walt, J. J. A. (1995), Taxonomy of the genus Pelargonium (Geraniaceae) – The section Polyactium. 1. Circumscription and intrasectional classification. South African Journal of Botany 61, 53-59.
4. van der Walt J. J. A.: “Pelargoniums of Southern Africa”, Vol. 1, Purnell, Cape Town, 1977.
5. Albers F.: “Pelargonium”, in: Eggli, Illustrated Handbook of Succulent Plants. Dicotyledons, Springer, Berlin, 2002.
6. Knuth R.: “Geraniaceae”, in: Engler, Das Pflanzenreich, Heft 53, IV, 129, Wilhelm Engelmann Verl., Leipzig, 1912.
7. Maggs G. L., Vorster P., van der Walt J. J. A., Gibby M. (1999), Taxonomy of the genus Pelargonium (Geraniaceae): The section Polyactium The subsection Polyactium. South African Journal of Botany 65, 115-143.
8. Sweet R.: Geraniaceae – The Natural Order of Gerania III, t. 232, 1824.

 

Citation and Copyright

© The Author. This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.

ISSN 2464-014X.

This article was first published in Geraniaceae Group News #177. Cite as: M. Strlič: P. amatymbicum (Eckl. & Zeyh.) Harv. Geraniaceae Group News #177 (2025), pp 27-33.