34. Haworthia serrata Bayer, JS.Afr.Bot. 39:249(1973). Bayer :55(1976). non Scott :62(1985). Type: CAPE‑3420(Bredasdorp): Oudekraalkop, Heidelberg (‑BA), Bayer 166 (NBG).
Rosette stemless, seldom proliferous, to 7cm φ. Leaves 20-30, narrow, acuminate, spined along margins and keel, bright yellowish-green with translucent lines above. Inflorescence simple, robust, to 40cm. Flowers white with green venation.
1982 – H. serrata occurs in an area which appears to be a kind of watershed for the genus. This is the area between the Duiwenhoks river draining from Heidelberg to the sea, and the Breede River. Neither H. retusa, H. turgida, nor H. magnifica are known in the area, and only the one variant regarded as a variant of H. heidelbergensis is known 20 km southwest of Heidelberg at Matjestoon. H. serrata has the same general form as H. herbacea but is a more uniformly green plant with larger leaves which have a tendency to bend outward. The flowers are large and white, unlike H. retusa and H. turgida, and more closely resemble those of H. emelyae from north of the Langeberg mountains. Recent collections by C. Burgers from the coastal limestones may throw more light on the relationship of H. serrata with a population at Bredasdorp and H. heidelbergensis, and more collecting needs to be done immediately east of the Breede River too.
1999 – The 1982 discussion is vague regarding the Bredasdorp connection. However, this is because the Bredasdorp area, and particularly the coastal limestones, seem to harbour some unusual elements (see H. mirabilis var. calcarea, H. variegata var. petrophila and H. heidelbergensis var. minor). The reference was specifically to the last named taxon, which is in any case now recognised in this work. H. serrata can curiously best be understood by recognition of non-co-occurrences. There are no other species in the immediate localities known for this species and hence continuance is the only guide. In this perspective it can be seen to relate to both H. heidelbergensis and to H. mirabilis. Seedlings often include weaker forms which resemble either the Bredasdorp or Matjestoon varieties of H. heidelbergensis. Some plants of the Riversdale to Heidelberg populations of that species in turn bear resemblance to H. serrata. There is also a very close resemblance to forms of H. mirabilis var. sublineata. The flower is rather robust for the southern Cape species and thus mention was made of the similarity to H. emelyae var. multifolia. That element also has a connection to H. mirabilis probably through H. heidelbergensis. H. serrata occurs at three discrete localities and is virtually the only Haworthia species in that area between the Breede River and Heidelberg. The general shape of the plant tends to deviate from the retuse-like forms of its close allies and veers towards H. herbacea in that respect. Indeed this is the kind of variability one should expect in terms of Vavilov’s homology or Chaos theory. Col. Scott mistakenly took a form of H. mucronata for this species.
Distribution: 3420(Bredasdorp): Oudekraalkop, Heidelberg (‑BA), Bayer 166 (NBG); Koppies (-BA), Bayer 4902 (NBG); N. Oudekraalkop (-BA), Bruyns 6260 (NBG).
[ed.] Bayer now calls this species H. rossouwii V.Poelln. in Kakteenk. 7: 75 (1938).