Saturday, August 10, 2013

The History of Portulaca umbraticola in Cultivation

Photo Source: Copyright foto76
Stock Photo - image ID: 100136414,

From time to time I hope to broaden the perspective of this blog by writing about some of the exotic cultivars that are commonly found in Australian gardens and nurseries.

Cultivars of the exotic Portulaca umbraticola are common summer garden plants that are sold throughout the world. They have been in the nursery trade in Australia ever since the early 1980s. The plants are frequently mislabeled as Portulaca oleracea or P. grandiflora, or as hybrids of these species.

The cultivars are frequently sold in nurseries during spring and summer as flowering punnets. Each punnet generally consists of several distinct phenotypes (based on flower colour). Apart from the different flower colours, the plants look pretty much homogeneous. Each plant flowers uniformly and consistently according to its particular phenotype. The plants are remontant and the showy flowers are often produced in abundance throughout summer. The plants can be long-lived providing they are regularly supplied with adequate light, moisture, and nutrients. They can also be successfully "overwintered" if kept on the dry side and out of the frosts.

I purchased my first punnet of these attractive plants from a hardware store some years ago. The reverse side of the label identified the plants as "Portulaca oleracea", which the botanical side of my brain soon began to query. I could not understand how a common, cosmopolitan weed with rather small, yellow flowers and comparatively few stamens could have ever been involved in the development of these cultivars. The leaves, stems and growth habit of the plants showed only a superficial resemblance to P. oleracea. I realized that the species name on the label could not be correct.

Some detective work was of course inevitable, as I then resolved to discover their true identity. The wholesale nurseries that I emailed were very helpful, but they ultimately just confirmed the name that their original suppliers had used. One of the companies checked their sources and believed the plants to be a hybrid cultivar of P. grandiflora or a hybrid between that species and P. oleracea. This is a reasonable presumption, considering that the vegetative parts of the plants look similar to P. oleracea and the floral parts look similar to P. grandiflora.

The use of the botanical name in Australia had clearly been reproduced in good faith from their international suppliers. A subsequent Google search soon revealed that the majority of American nursery sites were still using the name P. oleracea for these plants. A lesser number were using the name P. grandiflora, while a few (mainly horticultural sites) were vaguely classifying them as hybrids between P. grandiflora and P. oleracea.

I contacted Mr David Ferguson at the Rio Grande Botanic Garden in Albuquerque. Dave has an exceptional knowledge of cacti and succulents, based on his many years of experience as a curator and research botanist. It turns out that Dave not only knew the true identity of the cultivars, but he had also brought the mistake to the attention of the horticultural industry some 20-25 years ago! Here is what he told me:

"Usually (until about 18 years ago), all the Portulaca cultivars with flattish leaves were sold here as Portulaca oleracea.  All the ones with cylindrical leaves were P. grandiflora. And they were often confused with one another, even though several species are really involved, and P. oleracea is only sold as something of a novelty herb. I don't know if you've ever seen the "Sunset Garden Books" that they sell in the U.S.? (It use to be the "Western Garden Book", but they added "National" and "Southern Living" versions for a larger audience.) It is something of a "bible" for gardeners looking up commonly cultivated plants. They list a lot of plants, and if they list it, it sort of becomes horticultural dogma here, regardless if correct or not. They listed the cultivars of "Purslane" as Portulaca oleracea.  I wrote them a note pointing out a few misidentifications, including this one, and amazingly they researched it, sent me back an appreciative thank you note, and updated the next version.  Since then all "Purslanes" with big pretty flowers are being called P. umbraticola. The average person just doesn't seem to really notice the differences."

In a later email, Dave confirmed for me that there is no mention of Portulaca umbraticola in the 5th edition of the Sunset Western Garden Book (October 1988), but by the 6th edition (March 1995) the entry describing the 'Wildfire' strain has been edited. Here is the quote from the book:  "A strain called 'Wildfire' has been offered both as P. grandiflora and P. oleracea.  It is actually a strain of P. umbraticola and is popular in the Southwest and Deep South. ...... ".

For the rest of this blog article I will outline the true identity and origins of the Portulaca umbraticola cultivars and provide a detailed history of their development and introduction.

I should first of all stress that these cultivars are technically not true hybrids in the sense that they resulted from the crossing of different species. Only one species with a broad natural distribution was ever involved. The cultivars were most likely produced by way of intra-varietal hybridization and from then on were continued vegetatively by cuttings. In this sense they are essentially cultivars of a species, not "hybrid strains".

When grown from seed, these cultivars are recessive. This means that subsequent generations raised from seed will produce successively fewer fertile seeds. Also the flower colour frequently reverts to plain yellow in the offspring and I suspect the size and number of flowers may also diminish over a number of generations.

The author Urs Eggli places these plants correctly as cultivars of P. umbraticola subsp. umbraticola in the book Illustrated Handbook of Succulent Plants: Dicotyledon. (2001).

The correct naming of the cultivar is prescribed by the Rules and Recommendations of the International Code of Nomenclature for Cultivated Plants. So convention dictates that the name should appear as Portulaca umbraticola subsp. umbraticola 'Cultivar Name' or as Portulaca umbraticola subsp. umbraticola Group Name. The latter may show the Group Name enclosed in brackets and followed by the Cultivar Name enclosed in single quotes.

Eggli states of P. umbraticola : "This species is unique by virtue of the membranous wing ('corona') surrounding the basis of the capsule" and further writes: "A horticultural selection is available as a cultivar 'Wildfire Mixed'."

The wing around the fruit has not been observed in any other species in the genera. It is illustrated clearly in the following link:

P. umbraticola capsules (winged and with small "pimple" in centre). [Source: T. Beth Kinsey.]

Compare this with P. oleracea buds (flattened as if pinched and capsules never winged). [Source: Herbarium of the L. H. Bailey Hortorium, Cornell University.]

The original 'Wildfire Mixed' series was released in the US in 1982-83 by the Pan American Seed company. It is interesting that the intitial releases were packeted seed rather than live plants.

The Ball Seed Company (a subsidiary of Panam) first marketed the seed in 1982 under the name 'Wildfire Mixture'. The Ball Annual Seed Catalogue for that year outlined three colours - yellow, rose, and orange - and provided a photo of a yellow-flowering plant. Their 1983 catalogue contained an identical entry and a photo of 'Wildfire Mixture' which showed the full range of colours available at that time.

Photos of the original 'Wildfire Mixture' in Ball  Seed
Company catalogues from 1982 and 1983.
Source: Ball Horticultural Company archives.

A paper which talks about the development of this first Portulaca umbraticola series is Matthews, J.F., D.W. Ketron and S.F. Zane.1992. "Portulaca umbraticola Kunth.(Portulacaceae) in the United States." Castanea 57: 202-208. (Please click on the following thumbnail).

The segment implies that the original plants used to produce the 'Wildfire' cultivars may have involved a smaller-flowered P. umbraticola subspecies that occurs naturally in the southeastern United States. However the 'Wildfire' series was most likely bred from plants originating in South America, because the flowers of the subspecies found there are much larger, with richer tones and have much greater colour diversity. It is reasonable to presume that the progenitors of the 'Wildfire' series were South American plants that were selectively grown by native plant enthusiasts in the US, who perhaps viewed them as showy and longer-lived "substitutes" for the two native subspecies of P. umbraticola.

The two subspecies that are native to the United States are P. umbraticola subsp. coronata (which is found on granite and sandstone rock outcrops in South Carolina and Georgia and has small, pure yellow flowers), and P. umbraticola subsp. lanceolata (which grows on rock outcrops and sand in southern USA, west of the Mississippi, and has small, yellow flowers tipped with coppery red). Both of the subspecies native to the USA are short-lived annuals with small flowers and have little colour variation in the petals. It seems unlikely that they were ever used to breed the original 'Wildfire' series.

In their revision of the genus Portulaca in Brazil [Acta bot. bras. 24(3): 655-670. 2010], the botanists Alexa Coelho and Anna Maria Giulietti provide the following information: The petal colour of Portulaca umbraticola in South America is found naturally in yellow, white, pink, purple, or orange tones. They add that the variation in flower colour was reported by Legrand (1962), who noted that the variations related to geographical distribution. In South America, pink and orange are most common in Mexico, purple is most common in Argentina, yellow in the Antilles and Guianas, and pink or yellow in Brazil. In Brazil most plants have yellow flowers. They also state that the "styles and stigmas may also vary in colour, being yellow when the petals are yellow, or pink when the petals are pink".

Erica Vale Australia were probably the first horticultural company to import the 'Wildfire' series to Australia. They released it as part of their Simplicity Collection in the Spring of 1983. The details appear in this Sydney Morning Herald article in 1986. The company most likely imported hand-pollinated seed, rather than cuttings, and germinated these to produce their own F1 series. I am not sure if they sold only seed or plants, or both. Note what the article says about the plant's proper identification!

The original 'Wildfire' series was apparently never patented, so other people actively "extended" the series. The new generations of purslane cultivars are essentially the same as the original series, but some have greater flower abundance, larger flowers, and/or improved colour tones, while others are said to be longer lived.

Some of the newer cultivar series names for Portulaca umbraticola include 'Sun Jewels', 'Hot Spots', 'Carnaval', 'Cupcake', 'Pazzazz', 'Yubi', 'Summer Joy', 'Jumbo', 'Sundance' (not to be confused with 'Sundancer' which is a P. grandiflora line), 'Giganthes', 'Rio', and 'Toucan'... and many, many others. They are sold under different names throughout the world, but are all essentially the same plants.

Some of these are being sold in the Australian nursery trade, and are either extensions of the original 'Wildfire Mixed' lineage, or (more likely) have been imported more recently as tissue-cultured clones. As far as I know there have never been any attempts to breed new lines here in Australia. Most of the breeding of new cultivars takes place in America, or, in more recent times, Japan.

As most of the plants are cutting-grown, they are best regarded as individual cultivars rather than as true strains. The word "strain" implies that the plants have been grown as a batch and from seed. The original 'Wildfire' series were recessive when grown from seed. After only one or two generations, fewer viable seeds are produced as sterility returns. There is also a tendency for seedlings to revert to a plain yellow flower form, rather than repeat the parental type. The only way to reliably produce a new batch from seed is to start from scratch and this involves cross-pollinating the original wild selection by hand. This will produce a new batch of F1 intra-varietal cultivars, but unless different stock plants are chosen, the resulting group of F1 cultivars may appear rather similar to previous groups.

It is possible that some lines may be improved and extended by out-crossing F1 hybrids from different breeding lines, particularly if different parent stock has been used to develop each line. It may also be possible to back-cross offspring with F1 plants or the original parent stock. These may be the only choices available to a plant breeder who must work within the limits of recessive genes. New lines of cultivars will ultimately be produced, considering that the gene pool is now global. However one limiting factor to this might be the availability of good, fertile breeding stock, which is in all likelihood highly coveted by a very small number of professional plant breeders and larger companies.

The problem of recessive genes means that the chances of extending a lineage through further breeding becomes very limited. In a regular garden situation, where the potential for inbreeding is high, only chance "sports" will occasionally be found which can then be continued by cuttings.

Some of the wholesale nurseries claim to be using "tissue culture" to multiply their plants for the retail market, but I suspect they may be using this term rather loosely, considering that the plants are generally very easy to grow from small cuttings. Clearly the most commercially-viable way to introduce new lines into Australia is to import flasks of tissue-cultured clones, rather than import cuttings or potentially unreliable seeds.

Dave has pointed out to me that there are a couple of "strains" that do seem to be trending towards new lineages, but it is difficult to know how much selective breeding (as opposed to chance) was actually involved in their development. It is also difficult to predict if these lineages can ever be expanded in future, given the problem of recessive genes and increasing sterility in the F2 and F3 generations.

One of these new "strains" focuses on broken flower color [the 'Duet' series]. Another has focused on flowers with deformed petaloid stamens that make the flower look semi-double [the 'Fairytail' series]. I have heard that another series is being developed in Japan which have corollas that are apically fringed, so that the flowers resemble small carnations.

Above: One of the popular phenotypes from the 'Fairytale"
series of cultivars. Clipart source:

These so-called "strains" mostly consist of random individuals which suggests that they may actually be "sports". They are possibly spontaneous in origin (i.e. they were arrived at by chance) and may be either F2 or F3 generational hybrids, rather than F1. The individual sport is multiplied by vegetative propagation (i.e. it is "cloned" by striking cuttings or by way of tissue culture). It is generally trialed for a period of time to test it for vigour and disease resistance, before being released as a new cultivar name.

An F1, F2, or F3 cultivar can never be reliably replicated from the seed that it produces. Even a selfing is unlikely to produce an identical copy of the parent. A higher success rate might be achieved by crossing different F1 lines, or back-crossing cultivars with the original parent stock. Such crossings could explore the full potential of the gene pool and result in interesting and unexpected variation in the offspring. I have incidentally seen photos of a very weird mutation that was arrived at by the selfing of a 'Fairytale' cultivar. The flowers on this offspring lacked any petals and only have deformed petaloid stamens, loosely arranged in the centre.

Above: The 'Duet' series of P. umbraticola cultivars.
Photo source: Stock photo ID POR028C,

An important point in closing is that the original Pan American series was called 'Wildfire' and the original plan had been to market this series as packeted seed rather than as live plants. However the name was dropped in favour of 'Wildfire Mixed', because the developers found that flower colour was too unpredictable in seed grown plants.

In more recent times the Pan American Seed Company has moved on to the "Toucan" range under their "Hot Summer Survivors" banner. I note that they are selling seed of select colour forms, these being 'Mixed', 'Yellow', 'Fushia', and 'Scarlet'.

I wonder if this means they have finally resolved the problem of unstable flower colour that had plagued the original 'Wildfire' series? I would be interested to hear from anyone who has experience with growing these seeds. I would like to know if the plants have reproduced reliably according to the designated flower colours.

Oddly (and frustratingly) the company are still using the incorrect name "Portulaca oleracea", even though the name had been corrected by the 6th edition of the Sunset Western Garden Book.

Sadly many plant breeders and companies in the nursery trade continue to use the misapplied name to this day.

The mislabeling of horticultural plants is really nothing new. In fact it should not be surprising to anyone who has been involved in the nursery trade for any length of time, especially if they also enjoy the pursuits of botany and horticulture. Erroneous labelling is really just par for the course and basic human error is inevitable. The problem is that once a name has been used incorrectly in horticulture for a long enough period of time, the name becomes ingrained. If a name appears in print, such as on a company plant label or in a photo caption published in a reputable book, the name takes on a certain aura of authenticity and authority. Given enough time and frequency of use, the name can become extremely difficult to extricate from a plant.

The confusion surrounding the use of common names is probably what caused the original P. umbraticola cultivars to be mislabelled as P. oleracea. There is a traditional misconception in the horticultural trade that all of the Portulaca cultivars with flat leaves should be called "purslane", while all of the cultivars with cylindrical or terete leaves should be called "portulaca". The problem is that P. oleracea is also called "purslane" whenever it is sold or grown deliberately as a herb or salad vegetable (e.g. we can buy seeds and punnets of 'Green Purslane' and 'Golden Purslane'). Some food aficionados prefer to use the far more polite and cultured name "Verdolaga" to refer to selected food strains of this species. Yet it does not matter if the plants be wild or cultivated, they all acquire the derogatory name "pigweed" as soon as they run rampant in our gardens or wheat fields! Other genera such as Calandrinia are also sometimes called "purslane", and many gardeners refer to ALL Portulaca species and cultivars as "pigweeds" or "portulacas" generally. To really add to the confusion, some entirely unrelated Amaranthus spp. are also called "pigweed"!

The lesson to be learnt from this is we should never presume that a common name is legitimately and inextricably linked to a single botanical name. The correct botanical name of a plant should always take priority from the outset, and this rule should apply as much to plant breeding as it does to botany.