Watsonia fulgens

Watsonia fulgens (Andrews)Pers. based on Antholyza fulgens Andrews was regarded as a nomen confusum by Goldblatt (1989) because the type illustration could not be matched to any wild population. Andrews’ description of this plant whch had been introduced to England in 1792 was little more than a diagnosis differentiating it from Antholyza ringens (= Babiana ringens): it had much longer glabrous leaves that remained green until new growth appeared, and bright scarlet, curved trumpet shaped flowers with large spreading lobes.

Ker Gawler (1802) treated it as a distinctive variety of Watsonia iridifolia (Jacq.)Ker Gawl., which is another name of uncertain application. An illustration by Planchon (1856) under W. iridifolia var. fulgens matches a clone that is still widely grown in Melbourne although apparently not commercially available. Planchon noted that it flowered in autumn with a scape to 1-2 metres long, far exceeding the leaves, simple or sometimes branched in vigorous specimens. Plants of this name were being sold in England by 1820 (Loddiges, 1820). In New South Wales, Macarthur (1843) had a plant he called Watsonia iridiflora fulgens and presented material to the Sydney Botanic Gardens in 1831.

The following description is based on accession 180 in my collection:

Evergreen, proliferating, to 150 cm tall. Basal leaves about 4, to 60 cm long, 35 mm wide, bright green with faint glaucous striations and thin green margins. Stem leaves 2, bract-like, slightly inflated. Flowers 24-28 (to 4 open at once) on a brown axis plus 0-2 short branches. Bract acute, to 19 mm long, exceeding the internode, brown-herbaceous. Bracteole subequal, obtuse or notched at apex. Perianth intense orange-red, with a paler star inside throat, alternating red and pale stripes inside tube. Tube to 49 mm long; basal part to 23 mm long; distal part cylindric, curved, to 26 mm long, 8 mm wide at mouth. Ridges absent. Lobes semi-flared with flat margins; outer acute, oblanceolate, to 27 mm long, 11 mm wide; inner elliptic, obtuse, to 28 mm long, 14 mm wide. Stamens closely arcuate with style, anthers 11 mm long, purple with purple pollen. Style branches far exceeding anthers, red with paler stigmas. Capsule cylindric, truncate, to 25 mm long, brown. Seeds with two short wings, 8-10 mm long, dark brown.

Unlike Watsonia tabularis and W. fourcadei, this plant is undamaged by full summer sun in Adelaide as long as it gets enough water. New shoots appear in January while the previous year’s leaves are still green. Flowering is irregular any time from April to September.

There is a superficial resemblance to photos of wild W. zeyheri in colouring: orange-red flowers on a dark axis. But accession 180 is clearly separated from this species by its size, truncate capsules, autumn-spring flowering season, non-thickened leaf margins and the rather characteristic pale star marking in the flowers.

One possible origin could be a garden selection from random hybrids between W. tabularis and W. zeyheri or W. angusta, with strong, hardy growth in cultivation due to F1 vigour. An irregular flowering season is common in Watsonia hybrids between parents with differing phenologies. It also resembles my hybrids of typical W. tabularis pollinated by W. fourcadei in such features as size, flower colour and capsule shape. The four species mentioned in this paragraph are closely related and were treated as the Subsection Angustae in Goldblatt’s revision.

Below is Planchon’s illustration. The prominent leaf venation may be the artist’s interpretation of the striated glaucous bloom emphasising the longitudinal veins.

And the type illustration from Andrews. Assuming it is the same plant as Planchon illustrated, this is less informative. Perhaps it was grown in shaded or otherwise unfavourable conditions, as he described it as only 3 feet tall.

The plant known as Watsonia fulgens has been a “thing” for over 200 years. If it does not match any wild population, perhaps it should be treated as a cultivar. Unfortunately the name has been loosely applied in horticultural literature, for example to W. angusta by Campbell (1986). Watsonia fulgens sensu Montague (1930) was probably a hybrid cultivar; it was described as having pale-rose flowers appearing early in spring. It was distributed by Law Somner (1933) and may have been identical to the Watsonia fulgens described as a deep pink in Brunning’s 1905 and 1918 catalogues.



Andrews, H.C. (1801) Botanist’s Repository 3: t.192.

Brunning, F.H. (1905) Manual of Seeds, Bulbs, Horticultural Sundries. (F.H. Brunning Pty Ltd: Melbourne).

Brunning, F.H. (1918) Winter Flowers, Bulbs, Spring Flowering Sweet Peas. (F.H. Brunning Pty Ltd: Melbourne).

Campbell, E. (1989) Watsonia. In Walters et al. (eds) The European Garden Flora 1: 385-386. (Cambridge University Press: Cambridge).

Goldblatt, P. (1989) The genus Watsonia. (National Botanic Gardens: Kirstenbosch).

Ker Gawler, J.B. (1802) The Botanical Magazine 17: t.600.

Law Somner Pty Ltd (1933) Law Somner Catalogue 1933-34. (Law Somner Pty Ltd: Melbourne).

Loddiges, C.L., Loddiges, G. & Loddiges W. (1820) Catalogue of Plants which are sold by Conrad Loddiges and Sons, nurserymen, at Hackney, near London. (Loddiges: London).

Macarthur, W. (1843) Catalogue of Plants Cultivated at Camden.

Montague, P. (1930) The new watsonias should be freely grown. The Australian Garden Lover 6: 33.

Persoon, C.H. (1805) Synopsis Plantarum 1: 42.

Planchon, J.E. (1856) Flore des Serres et des Jardins de l’Europe. 11: 1.


Almost a belladonna, but not quite

The belladonna lily, Amaryllis belladonna L., is familiar in Australian gardens and also out on roadsides and other public places where it had been planted – or dumped – decades ago. The umbels of large pink flowers appear on unbranched leafless stems at the beginning of autumn.

It has been hybridised with Brunsvigia josephinae (Redouté) Ker Gawl. to produce F1 hybrids that are almost as large and spectacular as the Brunsvigia and are grouped under the name Amarygia tubergenii. They have cartwheel-shaped umbels of many relatively small flowers on longer pedicels.

There are also much more common hybrids, known as Amarygia parkeri (W.Watson)H.E.Moore. Roger Spencer suggested in the Horticultural Flora of South-Eastern Australia that these are actually hybrids with another South African amaryllid, Cybistetes longifolia (L.) Milne-Redh. & Schweick, but was reluctant to complicate the nomenclature further by adopting a new hybrid genus.

An average example of A. parkeri.

It is not easy to distinguish the original belladonna lily from this latter hybrid. The ‘pure’ belladonnas tend to have fewer (less than 13) and larger flowers, which are on even shorter pedicels than in A. parkeri. The various forms of Amarygia parkeri tend to have more flowers in a more symmetrical umbel, and like the Cybistetes they have a conspicuous yellow carotenoid pigment inside the perianth tube.

The acyanic cultivar A. parkeri ‘Hathor’ showing the yellow pigment.

A partial checklist of named Watsonia cultivars

Watsonia is a genus of the Iridaceae with about 53 species in southern Africa. They are perennial herbs growing from corms and producing spikes of showy flowers adapted to pollination by birds or insects. The species are generally interfertile, all being outbreeders with the same diploid chromosome number. Their wide range in size, phenology and flower colour, along with the ease of working with their large simple flowers, make them attractive subjects for collectors and amateur hybridists.

During the early 20th century there was interest in commercial production of named cultivars for home gardens and cut flowers, but the genus has been rather neglected since then. The following checklist is a ‘first pass’ through the referenced publications, with a bias toward those cultivars that have been released in Australia. It is not certain if every cultivar on this list is still extant.

You can download the list as a 210Kb pdf file from this link.

Watsonia ‘Leng’

DOI: 10.13140/RG.2.2.16217.39520

Watsonia coccinea

Watsonia coccinea Herbert ex Baker is native to the South-western Cape in seasonally wet sites on sandy flats.

It has been cultivated in Australia since the 1840s, sometimes under the misapplied name of W. humilis. The earliest recorded importation from the Cape was by Alexander Macleay of Elizabeth Bay via Captain Farquard Campbell in 1838. The specimen illustrated here was purchased from Tesselaars nursery in Victoria in 2002.


In cultivation it grows to 40 cm tall, exceptionally to 1 metre but never with more than 12 flowers. The bright scarlet perianth has an arched, narrow cylindrical tube 4-5 cm long marked internally with six darker red lines, and hooded lobes 24 to 28 mm long.

W. coccinea flowers later than many of the winter-growing watsonias, in late October and consequently the flowers are vulnerable to damage by thrips. It is less useful in the garden than the small forms of W. meriana for this reason, and because it is a “shy bloomer” with some full-sized corms producing only foliage if planted too densely or given less than full sunlight during the winter. It has apparently contributed its flower shape and warm colouration to a few of the cultivars bred by Cowlishaw and Cronin in the early 20th century, crossing with larger plants derived from W. borbonica.


Goldblatt, P. (1989) The genus Watsonia. 148 pp. (National Botanic Gardens: Kirstenbosch) ISBN 062012517

Macleay, A. (1843) Plants received at Elizabeth Bay. (Ms in Mitchell Library, Sydney, 2009/115).

Watsonia tabularis

Watsonia tabularis J.Mathews & L.Bolus has been in cultivation in Australia since the 19th century but does not appear to have contributed to the pedigrees of any hybrid cultivars bred in this country. It is endemic to the Cape Peninsula of South Africa and may be closest to the more widespread and variable Watsonia fourcadei J.Mathews & L.Bolus.

Plants grown from seed recently imported from South Africa have flowers of pale pink with a darker tube as shown in the photo. These represent the high altitude form; plants from lower altitudes differ in having bright orange flowers.


W. tabularis is evergreen, making most growth during autumn and spring then flowering in November to January.


Goldblatt, P. (1989) The genus Watsonia. 148 pp. (National Botanic Gardens: Kirstenbosch) ISBN 062012517

Two advantages of being inedible

A tweet from Patrick Moore to the effect that most of the pesticides present in the food we eat are produced by the crops themselves set me thinking about the “arms race” between plants to avoid being eaten and at the same time encourage herbivores to eat other competing species.

Eating or being eaten can be viewed as a very simple game. In order to survive, any plant or animal, any living organism, must eat. That is, take in from outside the substances that it needs to grow its own body and to provide energy to run its internal processes. It also must not be eaten if it is going to survive for long. Winning this leg of the game consists of convincing any predator that it cannot be eaten, to use the terminology of  Stephens (1993).

Plants have developed the ‘must eat’ game virtually to its limit by now. Housing symbiotic chloroplasts that fix carbon by photosynthesis, absorbing other nutrient elements, and the metabolic pathways that produce the whole plant have been established since the Palaeozoic. Even the later innovations of CAM and C4 photosynthesis have been around for millions of years.

Dennis Stephens (1994) further suggested that the main game among plants is ‘must not be eaten’ because they have not yet evolved as  far as they can go in that department. They are still in an arms race with herbivores, with pathogens and with each other. Plants may develop spines or other physically deterring outgrowths that convince hungry herbivores that they are not edible. They may use nectar to encourage ants to wander over their surface and clean up feeding insects. But most importantly, they may produce any of a vast range of chemicals (secondary metabolites) that make them bad-tasting or toxic to the particular herbivores that threaten them. These are all physical manifestations of the strategy of being inedible.

As an aside, it’s interesting that the development of toxic secondary metabolites is a speciality of the angiosperms or flowering plants that have dominated land vegetation since the Cretaceous age. Ferns can be inedible too, but they are much less rich in this kind of chemistry. And the notably toxic members of the gymnosperms are not the really ancient ones, but those that have diversified since the Cretaceous in competition with angiosperms – the cycads, Ephedra, Taxus and some related conifers.

So there is an evolutionary pressure on plants to not be eaten by convincing herbivores that they are inedible. The most obvious benefit from this – when it succeeds –  is that they do not lose biomass or get killed outright by herbivory.

But there can be a second advantage for the uneatable. Consider a three-way game between two plants and a herbivore, such as sheep grazing on a pasture of grass containing thistles. A thistle’s spines make it unpalatable; either totally inedible, or so hard for the sheep to eat that it will not be nibbled as long as any edible, palatable grass remains around it.

So the harder the sheep graze, the less the grass will compete with the thistles for light, water and ground space. The combination of grazing pressure and their spiny defence against being eaten has given them a powerful strategy in their own competitive game with the grass.

It’s also interesting to consider the strategy of the grass. It might actually derive some benefit from being grazed along with broadleaf weeds that lack the thistle’s defence, since it is better equipped to regrow after grazing than they are. But that’s another story.



Stephens, D.H. (1993) Expanding on Level 5, Sex. Letter tape of 6 May 1993.

Stephens, D.H. (1994) Postulates, Self and the Obsessive IP. Letter tape of August 1994.

Watsonia pillansii

Watsonia pillansii L.Bolus is widespread in the eastern (i.e. summer rainfall) part of South Africa at low and medium elevations. This wide geographic range is associated with variation in ecological requirements and plant size, but the flower colour is generally bright orange to orange-red.

Plants grown from seed recently imported from South Africa have unbranched stems to 1.2 m high bearing up to 22 flowers. They are evergreen, with new shoots appearing in late summer immediately after flowering and before the previous season’s leaves have died. Each flower has a cylindric tube 3.5 to 5 cm long and acute perianth lobes to 24 mm long that flare widely when fully open; the colour in this strain whose exact provenance is unknown is a rather weak orange-juice orange on the lobes and deeper on the outside of the tube. The anthers and pollen are cream.


Watsonia pillansii is related to W. schlechteri in the section Watsonia, subsection Grandibractea.

The species has been in cultivation in Australia since the 19th century. Cultivars that may be selections of W. pillansii include ‘Flame’ (marketed by Lawrence Ball in the 1940s) and ‘Watermelon Shades’ (Cheers, 1997). Watsonia ‘Beatrice’ or the Beatrice Hybrids is a group name for various natural hybrids of W. pillansii (Eliovson, 1968) that were exported to Britain, America and Australia in the early 20th century. The name comes from Watsonia beatricis J.Mathews & L.Bolus, which was a taxonomic synonym of W. pillansii.


Cheers, G. ed. (1997) Botanica. (Random House Australia).

Eliovson, S. (1968) Bulbs for the Gardener in the Southern Hemisphere. (Reed: Wellington).

Goldblatt, P. (1989) The Genus Watsonia. (National Botanic Gardens: Kirstenbosch) ISBN 062012517

Watsonia fourcadei

Watsonia fourcadei J.Mathews & L.Bolus has been in cultivation in Australia since the 19th century but does not appear to have contributed to the pedigrees of any hybrid cultivars bred in this country. It is widespread in the mountains of the southern Cape but absent from the Cape Peninsula where it is replaced by the related W. tabularis J.Mathews & L.Bolus.

Plants grown from seed recently imported from South Africa have flowers in a range of pink shades from pale salmon with a darker tube to the medium pink shown in the photo. They are evergreen, making most growth in mid summer to autumn but flowering in October to December.

The flowers have an arched, narrow cylindric tube about 6 cm long and perianth lobes 26 to 32 mm long incurved to form a cup-shaped limb.



Goldblatt, P. (1989) The genus Watsonia. 148 pp. (National Botanic Gardens: Kirstenbosch) ISBN 062012517

Watsonia schlechteri

Watsonia schlechteri L.Bolus grows in the montane veld of the winter-rainfall parts of the Cape, South Africa. Plants grown from seed imported via Silverhill Seeds first flowered this year and are a good match for the lectotype of W. schlechteri.


The flowers are orange-vermillion with perianth lobes to 23 mm long. The buds end in a slightly downcurved point. There are no staminodal ridges in the perianth tube, a character that distinguishes it from the closely related W. pillansii. However, the fresh leaves of these specimens lack the strongly thickened margins and midvein that are used as another distinguishing character; these are only evident in dried material.

W. schlechteri is one of the smaller watsonias, usually much less than 1 metre tall with leaves about 40 cm long. It flowers in late December to January, resuming growth from offsets soon after flowering while the previous season’s shoot may still be green. Thus it has some leaves all year, or non-flowering plants may be briefly leafless before the new growth starts in late summer. Goldblatt notes that flowering in the wild is conditional on the plants not being shaded out by surrounding vegetation.

Like other watsonias native to high altitudes, it is at risk of damage in the agonisingly hot, dry summers we get here at sea level in Adelaide. The problem is to give the plants sufficient light without excess heat, a tall order on days of 42°C with northerly winds.


Goldblatt, P. (1989) The genus Watsonia. 148 pp. (National Botanic Gardens: Kirstenbosch).

Bonding postulates and the nature of synonyms

Sets of entities, of any kind, can be linked in logic by bonding postulates of the form A ⇒ B (meaning that A is a subset of B, implies B, is within B). The same statement can be written in reverse as B ⇐ A (meaning that B is a superset of A, is implied by A, includes A). A is called the antecedent, and B is the consequent. This does not imply either a causative or a temporal sequence between antecedent and consequent, but simply a logical relationship.

In each case, the set of entities classed as A is completely included within the set of B. A is never found without B although B may occur without A. This situation is described in Boolean algebra as a (1 – b) = 0, or a = ab

Uppercase letters here refer to sets of actual entities, whereas the postulates (in other words, the elective functions or decisions) that define those sets are indicated by the corresponding lowercase letters, following the usage of Boole (1847).

Any relationship that exists between two entities or two postulates can be exposited as a nested hierarchy of A ⇒ B relations. It’s hardly an exaggeration to call this relation the basis of all logical thought.


If A ⇒ B, a pair of conditions holds:

B is necessary for A: A needs B in order to exist, although B can exist without A. eg, water is necessary for plant growth.

A is sufficient for B; the presence of A guarantees B, although B might also exist under alternative conditions not involving A. eg, seeing plants growing is sufficient evidence to assume the presence of water.

Stephens (1994) pointed out that the necessity of B for A and sufficiency of A for B together form a tautology that arises from the way we have circumscribed A and B such that A ⇒ B. For example, if we agree that all dogs are mammals, or dog ⇒ mammal, then being a mammal is one of the necessary qualifications for being a dog, but being a dog is by itself sufficient to qualify an animal as a mammal. This type of tautology is ubiquitous in the systematic classification and naming of plants and animals. Thus species A may be assigned to genus B as one of its members so that A ⇒ B, and that genus is in turn assigned to a family. Thus the classification system of the plant kingdom is a nested hierarchy of A ⇒ B relations with A sufficient for B, and B necessary for A, at each level.

By the same logical process, a taxonomist may assign species M to another species, N, as a synonym if he considers them too similar to merit separate names. A synonymy is an example of what Boole (1854) called an abstract proposition as it is a proposition about species concepts, which are in turn propositions about actual, tangible specimens. Every scientific name of a species refers ultimately to one specimen, known as the type specimen. It will be seen from the paragraphs above that if name ‘M’ is a taxonomic synonym of ‘N’ they cannot be at precisely the same level in the hierarchy: M must be within N as a name applying to a subset of the whole set of individual organisms comprising species N. Therefore two names cannot both be taxonomic synonyms of each other.

The same issue arises with synonyms in ordinary language. There is always an asymmetry in rank, a difference in level between one word and another that is considered to be its synonym. The meaning of the latter must always be a subset within the former. A thesaurus might glibly suggest ‘vehicle’ as a substitute word for ‘car’. But ‘vehicle’ is a more inclusive concept than ‘car’: all cars are vehicles but not all vehicles are cars. Therefore cars are a subset of all vehicles, and the word ‘car’ is within ‘vehicle’ as a synonym.

However, the codes of biological nomenclature were drafted without reference to Boolean algebra. They can add a little confusion since the principle of priority mandates that the earliest-published name be used for the merged species, although this may not be the name associated with the most inclusive set. This arbitrary rule may give the paradoxical impression that a larger M can reside within a smaller N. For instance, many garden plants from China such as the Banksian rose (Rosa banksiae) and the weeping willow (Salix babylonica) were given their botanical names based on the selected horticultural forms first introduced into Europe, but those names must now apply to all wild populations of these species as well.


All the examples above are single bondings where A ⇒ B but not B ⇒ A. This can be expressed in Boolean algebra as a (1 – b) = 0 and b (1 – a) ≠ 0.

However, if A is both necessary and sufficient for B, then B exists if, and only if, A exists. This is the state of equivalence A ⇔ B, meaning that A and B are co-extensive, and either can be called the antecedent or the consequent. This is quite distinct from the taxonomic tautology mentioned above (where the antecedent is necessary for the consequent to be true, and the consequent is sufficient to prove the truth of the antecedent). An example of equivalence would be the relation between the concepts “the 4th of July” and “USA’s Independence Day”; then a statement that “July 4 is Independence Day in the USA” is quite true but adds no new information. If it is agreed that A and B refer to exactly the same things, they may be called nomenclatural synonyms rather than taxonomic synonyms as they differ only in name, not in the sets of entities to which they refer.

On the other hand, a mutual bonding of two non-equivalent entities – that is, A ⇒ B and B ⇒ A where A ≠ B – represents a logical contradiction. They cannot each be contained wholly inside the other if they are different in any way. This is what Stephens (1994) called a double bonding, and may be expressed in Boolean algebra as a (1 – b) = 0 and b (1 – a) = 0.

Therefore every double bonding contains a fallacy. Either one of the bonding postulates is untrue, or they do not both belong to the same logical type in the sense that Whitehead & Russell (1910) used this term, or the same level in the sense of Polanyi (1968).



Boole, G. (1847) The Mathematical Analysis of Logic. (Macmillan: Cambridge).

Boole, G. (1854) An Investigation of the Laws of Thought, on which are Founded the Mathematical Theories of Logic and Probabilities. (Macmillan: London).

Polanyi, M. (1968) Life’s irreducible structure. Science 160: 1308-1312.

Stephens, D.H. (1994) Relationships – Bonding. audio recording of 21 February 1994.

Whitehead, A.N. & Russell, B. (1910) Principia Mathematica. Vol.1 (Cambridge University Press: Cambridge) .