Wayne's Word		Index		Noteworthy Plants		Trivia		Lemnaceae		Biology 101		Botany		Search

Of all the marvelous wildflowers native to the southwestern United States and Mexico, one of the most fascinating is a tiny stem parasite named Pilostyles thurberi. This little wildflower is rarely mentioned in popular wildflower manuals and is so seldom seen that it doesn't even have a common name. It usually goes by its generic name of Pilostyles (pie-LOS-til-eez). It is rarely seen because it lives completely embedded within the stems of a small, leguminous desert shrub called dyeweed. In fact, you could walk right by a dyeweed shrub without ever knowing that a Pilostyles plant is living inside the stems. Once a year, tiny buds appear on the host stem that resemble small reddish pimples; however, unlike a severe case of acne, the "pimples" burst open into minute reddish blossoms about 2-3 millimeters in diameter.

The genus Pilostyles has a very interesting and widespread distribution, including the Sonoran Desert, Costa Rica, Chile, Australia, Ethiopia and the Near East. Host plants include Psorothamnus and several other legumes in North America, Astragalus in the Middle East, and the Flacourtiaceae in Chile. A remarkable Malaysian relative of Pilostyles, called the "stinking corpse flower" (Rafflesia arnoldii), also grows inside a host stem, and produces the world's largest individual flower.

According to Kuijt (1969), the original colony of Pilostyles thurberi on the stem of dyeweed expands gradually into both downward and upward directions. "Soon the central area, the apparent court of entry, is left barren as no more flowers emerge. Numerous scars remain on this portion as reminders of the flowers produced in the past. Soon the infection ramifies into nearby branches of the host; the floriferous areas eventually become separated, but now progress only acropetally [upward toward the apex]. We can speak of a reproductive zone, therefore, maintaining a certain distance from the shoot apices of the host."

According to Kuijt (1969), Pilostyles exhibits isophasic parasitism, a condition in which the host's apical meristem is permanently infected with an endoparasite that keeps pace with growth of the host. The endoparasite system is carried upward by cells in the subapical region of elongation. Emerging flowers of the mideastern Pilostyles haussknechtii are restricted to Astragalus shoots of a certain age, and actually only to the basal part of that year's growth. "Through its presence in the host's meristematic region the parasite can select, as it were, the places where flowers will be produced along the host branch." Rutherford (1970) found Pilostyles thurberi tissue in the axillary buds of dyeweed, evidence that the infection can be carried into new shoots.

Pilostyles thurberi grows inside the stems of a low, grayish, sweet-smelling desert shrub called dyeweed (Psorothamnus emoryi). There are about 20 other species of Pilostyles, all parasitic on shrubby members of the Pea Family (Fabaceae). The genus is distributed throughout subtropical and desert regions of North and South America, Africa and the Middle East. A Middle East species, Pilostyles haussknechtii is parasitic on some of the shrubby Iranian locoweeds of the genus Astragalus. These locoweeds are the source of gum tragacanth, a water-soluble, polysaccharide gum used as a thickening agent and emulsifier.

The host shrub for Pilostyles thurberi in the Colorado Desert is called dyeweed (Psorothamnus emoryi) because the stems, leaves, and flowers of this shrub are sprinkled with numerous tiny reddish-orange glands that produce a saffron-yellow dye. The dye readily rubs off on your hands, and it was used by local Indians in art work. Although dyeweed is rather common in the Colorado Desert of southeastern California, Arizona and adjacent Mexico, Pilostyles has only been reported from a few areas of this vast desert region.

Unlike most flowering plants, Pilostyles has no roots, stems, or leaves. It is completely dependent upon the dyeweed host for water and vital nutrients (carbohydrates and amino acids). The only visible parts are tiny reddish-brown flowers, about two or three millimeters in diameter when fully opened. The flowers are persistent after the bloom for at least a year, and may be seen throughout the year along the lower branches of dyeweed plants. When the flowers fall from the dyeweed stem, they leave small, circular scars or craters. Early in the year (or even in November, depending upon the fall rains), the flowers appear fleshy and reddish-brown. From a distance they superficially resemble a bad infestation of brownish aphids on the lower branches of dyeweed.

To the casual observer, it appears that the dyeweed host shrub produces two different sets of flowers, the lower reddish ones and the typical purplish pea-like flowers on the upper branches. The entire vegetative part of Pilostyles is reduced to fungus-like filamentous tissue that penetrates the dyeweed stem. Flower buds develop internally from this filamentous tissue and eventually break through the dyeweed stem. In the desert of southeastern California, the buds normally appear around the first of the year (or late December); however, Pilostyles have also been observed in full bloom during November.

Microscopic studies have revealed that several flowers along the dyeweed stem may represent a single individual plant, arising from the same fungus-like infection. Some students have asked Mr. Wolffia why Pilostyles is not considered to be a fungus. It is not a fungus because it produces true flowers and seeds, and it has the typical water-conducting cells (vessel elements) characteristic of true flowering plants. Therefore, it is a flowering plant (division Anthophyta) within the kingdom Plantae.

The Pilostyles flower has no true petals. Instead it consist of four or five scale-like, fleshy sepals and a few outer bracts which collectively resemble petals. Some authors refer to all of the bracts as perianth segments. Each flower has a thick central column with an expanded fleshy disc at the apex. Male and female flowers develop from separate fungus-like plants within the dyeweed stems. According to Rutherford (1970), most species in the Rafflesiaceae are dioecious, with separate male and female individuals in the population. Separate sexes are also typical of date palms, willows, marijuana, and most higher animals. Whether two Pilostyles plants of opposite sexes can inhabit the same dyeweed host certainly seems possible. Job Kuijt (personal communication, 2006) reports several separate Pilostyles infections on a single dyeweed plant.

In male flowers, pollen sacs (anthers) are produced around the central column, just below the fleshy disc at the top. In the photograph masses of powdery pollen can be seen around the upper part of the central column. See photo below:

In female flowers, numerous stigmatic hairs arise from the central column, just below the fleshy disc. You have to look closely because pollen grains adhere to the hairs and resemble the central column of male flowers. At the base of the central column is an expanded, rounded ovary that develops into a many-seeded capsule. The flower is unmistakably female if you can see the swollen basal part with tiny ovules (immature seeds) inside. See photo below:

You are probably wondering what the seeds look like, especially when each tiny flower produces a cluster of at least 100. The seeds are so small that they can easily pass through the eye of an ordinary sewing needle. However, the record for the smallest seeds go to certain orchids and some of the parasitic broom-rapes. In fact, some orchids may produce up to a million spore-like seeds per flower.

Pilostyles is truly an internal parasite (endoparasite) deep within the dyeweed stem. The reproductive structure or flower is the only part that emerges from the host stem. The adaptive advantage of external flowers is undoubtedly pollination and seed dispersal. This is similar to the many kinds of filamentous soil fungi that send up their characteristic fruiting bodies (basidiocarps and ascocarps) containing spores. Exactly how the flowers of Pilostyles are pollinated is still a mystery. The transfer of pollen from male to female flowers may be accomplished by crawling insects, such as ants, or possibly flying insects, such as flies or beetles. Male and female flowers may be a considerable distance apart (more than 5 meters), on separate host shrubs.

Several authorities have suggested possible animals that might carry the seeds of Pilostyles, including ants, termites, birds, and even small rodents. The tiny seeds are coated with a sticky substance that presumably aids in their establishment on dyeweed branches. Upon germination they penetrate the host stem and develop into a branching network of microscopic filaments of water-conducting cells. Exactly how this penetration occurs is still unknown. The fungus-like filaments of vascular tissue later produce minute flower buds that break through the host stem.

Harvester Ants Of The Genus Messor

At this time, my best hypothesis is that seed dispersal (and possibly some pollination) is accomplished by harvester ants of the genus Messor. According to Alex Wild (2005), Department of Entomology, University of California at Davis, Messor pergandei was collected at nearby Clark's Dry Lake and this appears to be the species that Tom Chester and I observed on dyeweed shrubs of the Colorado River flood plain in Anza-Borrego State Park. These ants have a polymorphic worker caste with several size variations. Their job is foraging for seeds of various plants and carrying them back to the nest. Most harvester ants belong to the subfamily Myrmicinae within the family Formicidae. There are other genera and species of harvester ants in the southern California region, but they are monomorphic. One of the most common is the familiar "red ant" or California harvester ant (Pogonomyrmex californicus) that readily stings photographers that sit near its nest. In many areas of cismontane California the latter species has been eliminated by the introduced Argentine ant (Iridomyrmex humilis). Argentine ants have become a serious threat to the coast horned lizard (Phrynosoma coronatum) in southern California. The primary food source for these endangered lizards are native harvester ants. Apparently the horned lizard is not fond of Argentine ants, and is actually attacked by them in enormous swarms. Colonies of Argentine ants need a damp area to survive, and have not invaded some of the dry habitats where native harvester ants and desert horned lizards (P. platyrhinos) live. Of course, they can readily colonize urbanized desert areas inhabited by people. Well-watered gardens with stepping stones and concrete slabs provide the idea living requirements for Argentine ants.

At the 93rd Ecological Society of America Annual Meeting in Milwaukee, Wisconsin (2008), Sedonia Sipes, Aldwin Anterola and Daniel Nickrent presented a paper entitled "A Preliminary Assesment of Pollinators and Floral Scents in the Stem Holoparasite, Pilostyles thurberi (Apodanthaceae)." Both male and female flowers were visited by male Augochloropsis bees and several taxa of wasps. These taxa were observed collecting nectar from flowers of both sexes and appeared to be foraging with constancy on Pilostyles thurberi. A prevalent constituent of the floral scent in both sexes is raspberry ketone (4-(4-hydroxyphenyl)-2-butanone), a known insect attractant. This is the first study to provide preliminary data on the pollinators of any Pilostyles species.

There are about 100 species in the genus Messor. The generic name is derived from the Roman god of crops and harvest, Messor. Colonies can be very large with a single queen. The nest structure is complex with elaborate subterranean graineries where seeds are stored in dry conditions, preventing germination. Along both sides of Hwy S-2 in southeastern Anza-Borrego State Park, naturalist Steve Disparti and I noticed numerous dyeweed (Psorothamnus emoryi) in full bloom during January 2006. The dyeweed shrubs appeared very healthy, some of which were apparently young plants. They were most abundant in the soft sand along the shoulder, perhaps because of run-off water from the pavement. Along the sandy shoulder, rows of dyeweeds were interspersed with thriving colonies of harvester ants (Messor pergandei). Could some of these plants have originated from germinated seeds within the ant colonies?

Tn the Colorado River flood plain of Anza Borrego Desert, flowers of Pilostyles and the host shrub dyeweed (Psorothamnus emoryi) often appear at the same time, and colonies of shiny black harvester ants often occur in nearby mounds or in the accumulated sand under dyeweed shrubs. The ants carry dyeweed flowers back to their nest where they extract seeds from the ripened ovary within each calyx. The flower husks are then carried out of the nest and discarded along the edge of the sand mound. Dyeweed flowers are unmistakable with distinctive rows of orange glands on the calyx. It seems quite possible that the ants could pollinate Pilostyles flowers as they walk along the dyeweed stems en route to the dyeweed blossoms. Since the ants are not plant specific, it also seems likely that they may harvest the seeds of Pilostyles. During favorable years with adequate rainfall, seeds carried underground beneath or around dyeweed shrubs could germinate and invade the host plant near its base. The sticky seeds could also be transferred directly to lower branches of dyeweed shrubs.

Hickman (1974) lists 10 characteristics of flowering plants that are typically pollinated by ants: Six of these are met by Pilostytes flowers. (1) Plants must grow in hot and dry habitats where the frequency of ant activity is high, (3) Plants must be short to provide ant access, (4) Populations must be dense or containing few species, (6) If plants are erect, flowers must be sessile, (7) Pollen volume per flower must be small, and (9) Flowers must be small and of minimal visual attraction. The four remaining characteristics are probably not applicable because the main objective of the harvester ants appears to be the seeds of dyeweed flowers. If harvester ants are streaming over Pilostyles flowers on their way to the branch tips and dyeweed flowers, the following 4 characteristics are probably not that crucial with respect to Pilostyles flowers: (2) Nectaries must be readily accessible to ants, (5) Synchronously blooming flowers must be few per plant, (8) Seeds must be few per flower, and (10) Nectar quantity must be small enough for ants but not enough for other pollinators. Since the ants carry dyeweed flowers back to their nest, the best odds of cross pollination would involve walking over male and female Pilostyles flowers en route to and from the dyeweed blossoms. Whether male and female Pilostyles flowers occur on the same dyeweed plant is unknown, but Job Kuijt (Personal Communication, 2006) has reported multiple infections on a single dyeweed.

The Stinking Corpse Flower: World's Largest Individual Flower

For decades, Pilostyles has been placed in the rafflesia family (Rafflesiaceae), a strange and little-known family of parasitic flowering plants scattered throughout tropical and subtropical regions of the world. Recent mitochondrial DNA studies using the matR gene have transfered Pilostyles and the closely related Apodanthes into the Apodantheaceae within the order Rafflesiales. One astonishing member of this order, Rafflesia arnoldii, produces the largest individual flower on earth, and is truly a wonder of the plant kingdom. It has been called the "giant panda of the plant world" because this rare and endangered species only occurs in the rain forests of Sumatra and Borneo in the Malay Archipelago. Unlike most flowering plants, it has no leaves or stems and grows endoparasitically within the woody stems of its host vine Tetrastigma, a relative of the grape. Unless you saw evidence of an old blossom, you could walk right past a Tetrastigma vine without ever knowing that a complete Rafflesia plant lives inside its stem. Occasionally a large flower bud resembling a pale orange cabbage breaks through the bark of the host vine and expands into an enormous blossom up to 3 feet (0.9 m) in diameter and weighing up to 25 pounds (11 kg). The gigantic unisexual flower has five fleshy red lobes (sepals) spattered with raised white spots.

With an odor reminiscent of a stinking corpse, the blossom attracts carrion beetles and flies which shuttle the pollen from male to female flowers. A single female flower may produce up to four million seeds which are dispersed by a variety of animals, from ants, tree shrews and squirrels, to the feet of wild pigs and Asian elephants. If a seed becomes lodged in a moist crevice on its host vine, it germinates and penetrates the host tissue where it proliferates into a network of microscopic filaments of cells. These fungus-like networks of vascular tissue eventually give rise to a mammoth blossom that pushes out through the host stem, making Rafflesia one of the most intriguing and unusual plants on earth.

The chances of a Rafflesia seed finding a host vine are slim, and massive deforestation has further decreased the odds of this remarkable event. Recently the governments of Indonesia and Malaysia have begun to recognize the marketable curiosity value of this plant and ecological preserves are being established.

Illustration by Graphic Artist E.M. Armstrong.

Copyright (C) ma_suska (2007) Courtesy of Wikimedia Commons.

The precise evolutionary relationships between the Raflessiaceae and other families of dicotyledonous angiosperms have been elusive. A recent article in Science entitled "Floral Gigantism in Rafflesiaceae" by C.C. Davis, M. Latvis, D.L. Nichrent, K.J. Wurdack and D.A. Baum (11 Jan. 2007) has finally shed some light on this enigmatic plant family. The authors presented results of phylogenetic analyses of mitochondrial, nuclear and plastid data showing that Rafflesiaceae are derived from within the Euphorbiaceae (spurge family). Most euphorbs produce minute flowers, suggesting that the enormous flowers of Rafflesiaceae evolved from ancestors with tiny flowers. The authors estimate that for Rafflesia there was a 79-fold (7900 percent) increase in flower diameter in a period of about 46 million years. This is one of the most dramatic cases of size evolution ever reported for eukaryotes.

According to the Jepson Manual Of California Plants (2012), the Rafflesiaceae is polyphyletic, consisting of three main lineages belonging to different orders. Therefore, Pilostyles has been placed in the holoparasitic family Apodanthaceae along with Apodanthes and Berlinianche. Mitochondrial and nuclear DNA sequences confidently place the Apodanthaceae in the order Curcurbitales, where they also fit well in terms of their flower morphology.

One Of The Smallest Flowers In Anza-Borrego Desert

Pilostyles is listed in the California Native Plant Society's Inventory Of Rare and Endangered Vascular Plants of California (1994). Because it is an endoparasite with inconspicuous flowers, it may be more common than originally suspected. You could easily walk by the host shrub (dyeweed) without ever knowing that a complete Pilostyles plant lives inside the stem. Luckily, some of its documented locations are protected within the confines of scenic Anza-Borrego Desert State Park in eastern San Diego County.

Return To The WAYNE'S WORD Home Page Return To The NOTEWORTHY PLANTS Menu