Phenology of some Iridaceae

The graph below plots the date on which the first flower has opened on each of six Moraea species in my garden over 18 years. The site is the Adelaide suburb of Warradale, close to sea level at 35° S latitude. All six species are summer dormant, spring flowering corms from South Africa where they evolved in Mediterranean-type climates similar to Adelaide.

This kind of “experiment” is rather in the tradition of Gilbert White, or those other English country parsons who competed to report the first cuckoo of spring in letters to The Times. Among its many limitations,

  • The date of first anthesis may not be the best way to characterise flowering period; a measure of peak flowering may be more meaningful
  • Microclimate was not controllable, although all plants were grown in the open in the same garden
  • Each species was represented by only one genotype
  • There are gaps in the records when a species either failed to flower or was not observed in a particular year

However, the graph seems to justify some comments:

  1. The sequence of flowering shows some constancy, with a ‘late’ group of M. bellendenii, M. ochroleuca and M. setifolia always flowering after the ‘early’ M. aristata, M. flaccida and M. vegeta.
  2. Flowering time did not change in the same direction or to the same extent for all species in each year. Each species has its own response to the weather that it experiences. For example, M. aristata and M. vegeta moved in contrary motion much of the time, but in near unison from 2007 to 2010 (yes, that’s a musical metaphor!).
  3. There may have been a general trend to earlier flowering in the “global warming” period of the late 1990s. Subsequently, flowering times may reflect the southern oscillation with late flowering in the notably cool, wet years of 2010 and 2011.

The mechanism controlling flowering has not been investigated in these Moraea species, but experience suggests it is not determined by photoperiod but by the accumulation of sufficient shoot biomass to support flowering and fruiting.

A second graph gives the corresponding results for seven species of Watsonia with similar phenology to the Moraea.

Again, the sequence of flowering shows some constancy, with W. aletroides consistently the first to flower and W. angusta or W. marginata usually the last. Again, each species reacted in its own way to seasonal conditions. Again, some of them seem to show the effect of a run of hot dry years in the late 1990s but more complex responses to the alternation of La Nina and El Nino periods. The response of W. laccata was especially variable, possibly as (unlike the other six), this species was represented by a population of mixed genotypes from seed.

The combined effects of temperature, insolation and rainfall may be complex, and biomass accumulation may not be easily modelled by the day-degrees method that is useful in higher latitudes where plant growth is closely correlated with temperature. The effects of rainfall and temperature may vary greatly according to the part of the growing cycle that they impact, and cannot be understood by integrating them across the whole growing season.

A cool, overcast winter may delay or inhibit flowering in these irids by slowing their biomass accumulation. As commercial Gladiolus growers have found, flowering fails if winter light intensities are too low. Drought in the period of vegetative growth has the same result. The first elongation of the flowering stem in Moraea flaccida and Watsonia meriana has been shown to coincide with the exhaustion of the old corm from which the plant has grown and the beginning of resource allocation to the formation of the next season’s corm. This is the most vulnerable point in the annual cycle of these irids. Once flowering has been initiated, even a few days of hot dry weather in spring can goad the plants into completing their flowering rapidly by producing fewer and smaller flowers.

(Please click on the graphs if you want to see them full size)

On the scientific method

The defining features of science are the source of its practical value, but they are also its limitations. It takes for granted the separation of subject from object, of an observer from the thing observed, that is the basis of the “common sense” human worldview. Ironically, this same separation is what Buddhists call avidyā or not-knowing, our basic error of failing to recognise phenomena as our own creations. So it might be said that, although science is a method of gaining knowledge, it has been built on a foundation of not-knowing.

Data by itself is not knowledge. Even facts committed to memory are just one step beyond facts stored in books on a shelf. The evaluation of data is as important as the data itself. But to evaluate anything we have to start from at least one datum that is known with certainty. And that presupposes some system of metaphysics (what’s real?) and epistemology (how do we know it?), neither of them prominent in the current intellectual landscape.

The scientific method was often taught in schools as a balance or alternation between deduction from the more general to the specific, and induction from the specific to the general.

But in practice, science proceeds mainly by gathering data and making deductions, rather in the tradition of Sherlock Holmes. Consequently it often appears that each generation of researchers is focussing on smaller and smaller questions in greater and greater detail. There is a lack of any formal process for drawing conclusions from observations and experiments, explicitly transmissible from teacher to student. Inductive thought in science has often been informal, and based on individual inspiration that dares an occasional leap beyond the safe ground of deduction. Think of Kepler’s elliptical planetary orbits, Kekulé’s dream of the benzene ring, Mendel’s digital model of inheritance, or Faraday’s sizeless electron.

A Moraea and its scent

The South African irid Moraea ochroleuca is in flower here at the moment, with its characteristic buttercup-yellow bowl-shaped flowers. The shiny brown veins that radiate from the base of the perianth secrete nectar, and each flower lasts for just two days.

Goldblatt et al. (2005) described its smell as putrid and the pollination syndrome as sapromyophily by carrion flies in the families Calliphoridae, Muscidae, and Sarcophagidae. However, the M. ochroleuca in my garden has a faint smell of yeast, just like the medium used to culture Drosophila in genetics labs. And, sure enough, a species of Drosophila is attracted to these flowers where it remains for long periods apparently feeding on the nectar. I doubt that Drosophila could pollinate this flower, since it stays on the nectaries and would only contact the stigmas, which are held far above the nectaries, by accident. In any case, my M. ochroleuca have never set seed in 12 years despite an attempt at hand pollination, and this is probably due to self-incompatibility as they are all one clone.

Maybe this clone is atypical in its scent. Or maybe we’re describing the same scent in different ways; it’s probably a subjective judgement whether a smell should be called putrid. Humans don’t perceive smells the same way that insects do, and there is lot of variation among humans in judging a perfume. The scent of another sapromyophilous irid, Ferraria crispa, has sometimes been compared to carrion. But to me, it has more of a pungent chemical smell reminiscent of iodine; and my wife says it’s like very stale spices. Neither the Moraea or the Ferraria has the pervasive rotting-corpse stench of Stapelia (Apocynaceae) or Dracunculus (Araceae); it’s necessary to get up close to detect their scent at all.

The presence of nectar is unusual in a sapromyophilous flower; it may be that, having only recently evolved this pollination strategy, Moraea ochroleuca still needs to provide a food reward to the pollinator. Goldblatt et al. also recorded a range of other Diptera including syrphids (hoverflies) as well as honeybees as occasional visitors.

The history of Moraea ochroleuca in Australian gardens is obscure. Unlike many other showy South African irids, the species formerly placed in Homeria were not promoted in 20th century gardening literature or advertised by retail nurseries. However, a few of them were early introductions into botanic gardens, and there may even be some truth in the anecdotes about Boer war veterans bringing back “Cape tulips” as souvenirs. The plant growing at the Adelaide Botanic Garden in 1859 under the name of ‘Moraea grandiflora’ may have been M. ochroleuca. The flower pictured above is heritage garden stock from the Adelaide area, and is a close match to a feral specimen collected by Ray Alcock at Yallunda Flat, AD96452089, that Peter Goldblatt determined as M. ochroleuca.


Francis, G.W. (1859) Catalogue of the Plants under Cultivation in the Government Botanic Gardens, Adelaide South Australia.

Goldblatt, P., Bernhardt, P. & Manning, J.C. (2005) Pollination mechanisms in the African genus Moraea (Iridaceae, Iridoideae): floral divergence and adaptation for pollinators. Adansonia 27: 21-46.