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 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)
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 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).
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) .
The rediscovery of Mendel’s principles of heredity at the beginning of the 20th century inspired a surge of ornamental plant breeding by researchers, commercial nurserymen, and perhaps most importantly by individual gardeners.
John Cronin, Director of the Royal Botanic Gardens in Melbourne from 1909 to 1923, had a personal hobby of experimenting with the improvement of garden flowers. He aimed to demonstrate the application of Mendel’s laws to flower breeding, and encourage gardeners to make their own hybrids. He worked with Dahlia and other genera, but particularly the winter-growing South African watsonias, which he recognised as “everyone’s flower” – easy to grow, attractive, a natural for southern Australian gardens.
In a previous publication I lamented that the exact pedigrees of his Watsonia cultivars were lost with the destruction of his papers after his death in 1923. But now the National Library of Australia has come to the rescue with their wonderful resource of newspaper files at Trove. Cronin was a tireless populariser and communicator, speaking at the evening meetings of horticultural societies around the suburbs of Melbourne and giving interviews to journalists.
In the spring of 1904, while employed by William Guilfoyle at the Botanic Gardens, he crossed a pink Watsonia borbonica with W. borbonica ‘Arderne’s White’. This cross may be represented by the following formula (but note that the order is arbitrary, it is not known which was the pollen parent and which the ovule parent in any of the crosses discussed here):
borbonica × Arderne’s White
He noted that pink flowers were dominant over white in the F1 generation, as has been confirmed by other researchers. In spring 1907 he selected one F1 plant with tall stature, dense branching and large flowers. He crossed this with a purple Watsonia meriana and the widely grown red Watsonia aletroides, and also backcrossed it to W. borbonica ‘Arderne’s White’ to create three lines for further breeding:
1. meriana × (borbonica × Arderne’s White)
2. aletroides × (borbonica × Arderne’s White)
3. Arderne’s White × (borbonica × Arderne’s White)
Cronin’s appointment as Principal of Burnley Horticultural College in 1908 seems to have interrupted this work, and in the following year he succeeded Guilfoyle as Director of the Botanic Gardens. By 1913 he had time to resume his watsonia experiments, and on 20 March sowed seeds from his three 1907 crosses at the Botanic Gardens nursery. Six years is not an inordinately long time to store Watsonia seeds, but there would be some loss in viability which may have unintentionally favoured some genotypes over others. Cronin’s management of the plants was another possible source of selection pressure to produce watsonias adapted to Melbourne gardens: he left the corms in the ground over summer, and gave the plants no fertiliser or watering even though 1913-14 was a drought period.
This generation produced their first flowers in October 1914; Cronin stated that these resembled the 1907 selection in size and colour, and were inbred that year. I interpret this to mean that he produced an F2 generation in each of the three lines by cross-pollinating siblings, since selfing would have produced little or no seed due to incompatibility. Thus,
1. (meriana × (borbonica × Arderne’s White)) × (meriana × (borbonica × Arderne’s White))
2. (aletroides × (borbonica × Arderne’s White)) × (aletroides × (borbonica × Arderne’s White))
3. (Arderne’s White × (borbonica × Arderne’s White)) × (Arderne’s White × (borbonica × Arderne’s White))
Large numbers of these seedlings were raised in the main nursery of the Botanic Gardens. By October 1916 Cronin saw the first flowers of the inbreds, which had a wider range of colours than their parents. Some whites showed up, as would be expected from recombination, including some with flowers of improved size and form compared to the original ‘Arderne’s White’. The watsonias commercially released in the 1920s as the Commonwealth hybrids or “Watsonia Cronini” were selections from this generation.
Line 1 would have produced the many Cronin cultivars with a mixture of characters from W. meriana and W. borbonica. These often have subtle tertiary flower colours due to genes from both species influencing anthocyanin pigment production. Floral bracts are typically well-developed and obtuse, compared to the shorter acute bracts of W. borbonica. Examples include ‘Lilac Towers’, which is the most widely grown Watsonia in Australia today and may be the same as Cronin’s ‘Sydney’, and the one illustrated below which may be his ‘Maitland’.
Line 2 would have yielded flowers with long tubes and small lobes like Watsonia aletroides. The one illustrated here was discussed in a previous post.
It’s significant that Cronin did not use a long breeding program: the cultivars released were no more than three generations away from the original genotypes that had been imported from Africa in the 19th century. As he was working with a perennial that is normally propagated vegetatively, he could stop at the F2 with its fixed heterozygosity. I have bred watsonias four generations on from these and other old cultivars, and can attest that hybrid breakdown soon appears. Some of the resulting plants had interesting extremes of flower shape or colour, many were dwarf or weak in growth, but few were gardenable.
In the spring of 1917 Cronin presented this data to the horticultural correspondent of The Leader, and was lecturing on flower hybridisation to amateur horticultural societies with his new watsonias as exhibits. The following year he gave an interview to The Argus, repeating that his new watsonias were produced by first crossing and then inbreeding on Mendelian lines.
Anon. (1917) Melbourne Botanic Gardens – New colors in flowers – The laws of Mendel. The Leader (Melbourne), Saturday 10 November 1917 pp.13-14.
Anon. (1917) Horticultural society. The Advertiser (Footscray), Saturday 15 December 1917 p.3.
Anon. (1918) Botanic Gardens Experiments. The Argus (Melbourne), no.22,553. Monday 11 November 1918 p. 6.
Cooke, D.A. (1998) Descriptions of three cultivars in Watsonia (Iridaceae) J.Adelaide Bot. Gard. 18: 95-100.
Pescott, E.E. (1926) Bulb Growing in Australia. (Whitcombe & Tombs: Melbourne).
Watsonia humilis Mill. is endemic to the Western Cape area of South Africa. It is winter growing and summer dormant like many other South African irids, restricted to a particular fynbos association on seasonally wet clay and loamy alluvial flats.
Its original range in the Breede River Valley between Tulbagh and Worcester, and the lowlands between Malmesbury, Franschhoek and Gordon’s Bay, has been reduced by 98%. Two populations are known to remain: one of fewer than 50 plants in an urban reserve at Gordon’s Bay and a second rediscovered in 2012 in the Breede River Valley near Wolseley, more than 80 km away. A third population may still exist between Somerset West and Sir Lowry’s Pass despite bulldozing by vandals in 2011.
Although critically rare in its natural habitat, it is secure in ex situ cultivation around the world. At least one strain has been passed between gardeners in Australia since its first importation in the 1830s.
Watsonia humilis is in the section Watsonia, subsection Watsonia of the genus. It is distinguished by its small size, unbranched inflorescence, strongly keeled bracts with outcurved tips and stamen filaments no longer than the perianth tube. Its pale coloured perianth (pale pink to white, never red) suggests that it is adapted to insect pollination unlike its larger and usually red-flowered relatives.
This watsonia can be grown in the open garden here but to keep track of the corms I grow it in 20 cm pots, lifting and dividing each summer. It multiplies vegetatively and I have used it as the ovule parent in hybridization.
The common form in Australian gardens has white perianth lobes and a tube shading to deep pink at the base. The F1 hybrids with bright red flowered W. meriana Mill. have uniformly pale pink flowers, the intensity and hue of the pink (anthocyanin) pigmentation varying slightly between individuals. This result shows that white flowers in this species have a different genetic basis to the recessive acyanic mutants of other species: it suggests incomplete dominance, possibly with more than one genetic locus involved. This is to be expected if white flowers were a stabilised local adaptation in the source population, rather than a passing mutation as in W. borbonica ‘Arderne’s White’. These hybrids lack the outcurved bract tips of W. humilis, which may be the most useful diagnostic character for recognising it.
Goldblatt, P. (1989) The genus Watsonia. (National Botanic Gardens: Kirstenbosch) ISBN 062012517
Goldblatt, P., Manning, J.C., Raimondo, D. & von Staden, L. (2013) Watsonia humilis Mill. National Assessment: Red List of South African Plants version 2013.1. Accessed on 2014/1/28.