science news logo

science service logo

Warm-Blooded Plants?
OK, there's no blood, but they do make their own heat

Susan Milius

"The dead-horse arum of Corsica looks and smells like the south end of a horse that died going north," says Roger Seymour. He's actually talking about a plant, and a more prosaic soul might add that it belongs to the same family as calla lilies and jack-in-the-pulpits. Seymour is a zoologist, and the plants he studies show an animalistic feature: They can generate body heat. Most plants, including calla lilies and jack-in-the-pulpits, simply assume the ambient temperature because their metabolic reactions hum along so gently that they don't give off bursts of heat. The dead-horse arum, however, belongs to the group of several-hundred plant species scattered among some 10 families that can rev up their own furnaces. That heat can launch strong odors, like those of a dumpster in August. In winter, warm flowers can melt snow.


DEAD GIVEAWAY. The dead-horse arum exudes such a carcasslike fragrance that female blowflies arrive and lay eggs as if the flower were an inviting hunk of carrion. The eggs are doomed, though.


The dead-horse arum outdoes all the others, says Seymour, who's at the University of Adelaide in Australia. The plant's flesh-pink blooms produce more heat than does any other known plant or any animal considered in its entirety. Scientists have measured higher rates of bodily heat production only in the flight muscles of some insects and, possibly, the brown fat of hamsters.

Descriptions of remarkable heat-making plant species date back more than 200 years, but scientists are still discovering new facets of the phenomenon, sometimes hidden in plain sight. Current research about the biochemistry behind plant heat may someday change the way people deal with heat. The pattern of heating power in the botanical family tree intrigues evolutionists searching for traits of ancient flowering plants. And, this winter, two research teams have presented new research on what good this heating does for a plant.


All the plant tissues so far found to warm themselves have reproductive functions, and Seymour sees common themes among the hot species' sex lives. They tend toward large blooms, which have a low surface-to-volume ratio favoring heat retention. In many of these blooms, the female organs mature before the male parts, requiring the plant to briefly kidnap pollinators to make its pollination system work.

Consider the dead-horse arum, Helicodiceros muscivorus. In spring on islands in the Mediterranean, these plants send up blooms with a central, fingerlike projection in front of a rounded dish of tissue, or spathe, several inches wide. When the plant first blooms, the finger radiates heat, which sends out strong aromas. Female blowflies soon swarm over the bloom.

Botanists have speculated that the stench represents step 1 in an entrapment scheme, attracting blowflies under the false pretense that there's nice dead flesh available as a nursery for their eggs. Tests bear that out, reported Marcus C. Stensmyr of the Swedish University of Agricultural Sciences in Alnarp and his colleagues in the Dec. 12, 2002 Nature. Several dominant compounds, called oligosulfides, showed up in both the stink of the flower and in that of a dead seagull. A study of nerve responses of blowfly antennae showed that the flies respond similarly to the composites of compounds that make up the scents, so they don't seem to be able to tell a flower from a dead gull on the basis of smell alone.

In nature, once flies buzz in to explore the dead-horse–arum bloom, many crawl down into the pocket where the spathe narrows to surround the base of the finger. That pocket contains a band of male florets above a band of female florets. Spines and filaments at the entrance to the pocket imprison the flies.

During the first day that a dead-horse arum blooms, female florets have matured enough to receive pollen, but male florets aren't releasing it. The flies, however, may carry pollen they picked up from a previous adventure in another, earlier-blooming plant. As the flies scramble around in the pocket, trying to escape, they dust that pollen onto female florets.

By the next day, the female organs have lost their receptivity, but the male parts have matured. The trapped insects then pick up pollen. The blockade of spines withers, so the flies can at last squeeze up out of the pocket. They then carry the new pollen to the next arum, should they fall for the same trick again.


Seymour reminisces that he first learned about heat-generating flowers several decades ago, when a friend brought the large fingerlike projection of a self-heating flower, Philodendron selloum, as a conversation piece to a California party. The structure was warm to the touch and looked more like a mammal's reproductive organ than a plant's.

Seymour was so taken with the structure that he savaged philodendron blooms in his mother's garden to get specimens for measuring heat generation. Thus began the project that first documented a new twist in a few self-warming plants.

While the dead-horse arum and most other self-heating plants produce heat on a preset schedule, regardless of the air temperature, P. selloum manages something more sophisticated: It regulates its heat generation to keep its flower temperature approximately steady, Seymour and his colleagues reported in 1972.

Growing outdoors, P. selloum keeps its blooms between about 30°C and 36°C. In lab tests, the flowers manage to stay in this range even when experimenters chill the air to 4°C.

Those experiments also revealed that most of the plant's heat comes from a band of tiny, sterile male flowers located between the fertile male and female flowers on the bloom's fingerlike projection. The sterile blooms shut down heat production when air temperature reaches about 37°C.

Like the dead-horse arum, this philodendron in its native Brazil lures insects inside. The philodendron's spathe closes over scarab beetles for only 12 hours. Yet the beetles remain for some 22 hours. While in the bloom, they mate, feed, and brush pollen onto female flowers. At the end of the beetles' stay, they pick up pollen from just-matured male flowers and fly off to another bloom.

The eastern skunk cabbage (Symplocarpus foetidus) in North America and Asia also keeps warm, independent of air temperature, Roger Knutson reported in 1974. The insect-pollinated plant flowers early in the year, sometimes by New Year's Day in mid-Atlantic states. Its bloom, a thick-walled, teardrop-shaped spathe surrounding a fat stub covered with florets, can melt holes in the snow cover (SN: 8/21/99, p. 123).


HOT STUFF. The skunk cabbages in Japan melt snow and show the most-precise temperature regulation yet recorded in a plant.


Skunk cabbages can bloom inside a snowbank and create their own ice caves. "You can break through the snow and look into these fantastic spaces," Seymour says.

In experiments at air temperatures around 15°C, the inner core averaged some 9° higher. When the air temperature dropped to –15°C though, the fingerlike projections reached temperatures 30° higher than the air. "Some mammals can't even do that well," says Seymour.

In 1996, Seymour and his University of Adelaide colleague Paul Schultze-Motel reported that the Asian sacred lotus (Nelumbo nucifera) could also regulate its flower temperature. The species is hardly rare or unfamiliar. It grows widely on both sides of the Atlantic Ocean, but no one had previously tested it for temperature regulation. The Adelaide team found that as environmental temperatures dropped as low as 10°C, flower temperatures stayed between 30°C and 36°C.

The researchers decided to see whether day-night cycling influences the lotus' temperature control, as it does in some other plants. The team covered individual lotus blooms with translucent jackets made from inverted wine-bottle coolers and reversed the normal temperature pattern for night and day. The flowers tracked the temperature instead of the light, Seymour and his colleagues reported in 1998.

The dead-horse arum maintains a relatively stable bloom temperature, but the plant isn't a true temperature regulator, says Marc Gibernau of Paul Sabatier University in Toulouse, France. He, Seymour, and Kikukatsu Ito of Iwate University in Morioka, Japan, found that heating related more to time than air temperature, they report in the December 2003 Functional Ecology.

There's only one other plant that's been identified as regulating heat production. It's a South American species, Xanthosoma robustum, that's related to the dead-horse arum, philodendron, and skunk cabbage. X. robustum has received less attention so far.

Furnace design

The past 5 years have shaken up the study of the chemistry of hot plants by adding a new heat-generating pathway for scientists to probe. Since 1932, physiologists have known about one heat-making pathway, which is a secondary process for respiration (SN: 6/24/89, p. 392). By the 1970s, physiologists had linked the slow-heat burns of arums with a jump in activity in this pathway. An enzyme in the pathway, alternative oxidase or AOX, occurs only in plant cells, where it's located in the cell powerhouses called mitochondria.

Mammal mitochondria can blast out heat, too, but they rely on what's called uncoupling proteins, UCPs. In 1997, a European research team reported similar chemistry in potatoes. Cold prompted activity of a gene making what looked like a version of a mammalian UCP rather than an AOX, said Maryse Laloi of the Max Planck Institute for Molecular Plant Physiology in Golm, Germany.

Ito then began searching for UCPs in skunk cabbages. In 1999, he reported finding genes encoding two UCPs. A temperature drop activated these genes only in the floret-holding stub. The UCPs and AOX seem to function for heat generation simultaneously, he and his colleagues reported at the Plant Biology 2003 meeting in Hawaii in June. "Now, we have to reconsider the functions of two different thermogenic reactions," says Ito.

Since 2001, Ito and his Iwate colleagues—with support from the Japanese government—have been searching for the temperature sensor and other compounds that operate in the skunk cabbage's heat production. The researchers have figured out the basic protocol that the plant follows, says Ito. "We call it 'the skunk cabbage algorithm.'"

Ito has applied for a patent on this protocol and isn't releasing the details. "This sort of biological algorithm could be used as a new brain to control nonbiological devices, such as air conditioners," he says. The standard program controlling an air conditioner was invented more than 60 years ago. The system used by a skunk cabbage, "which is a typical chaotic system, is totally different," he says. Ito's team has recently succeeded in operating an artificial heater with this algorithm. "I really think we can learn a lot from skunk cabbages," he says.

What's hot

The majority of the hundreds of plants known to generate heat sprout from ancient lineages at the base of the botanical family tree, observes Leonard Thien of Tulane University in New Orleans. Self-heating may have been an early innovation that arose soon after the invention of flowering.

Evolutionists are looking at thermogenesis as they reevaluate traits in these old lines. "At the moment, a great deal of discussion is under way to decide upon the state of various characters," says Thien.

For example, heat generation has turned up in certain plants of these ancient lineages of flowering plants: the magnolias, Dutchman's pipes, star anises, custard apples, and water lilies. Heat-generating flowers include the darling of 19th-century aquatic gardens, the Amazon water lily (Victoria amazonica).

Thien says that preliminary results suggest that at least one other ancient family includes a self-heater, but he won't say which one until he double-checks his results next spring.

The most ancient family that Thien has tested is Amborellaceae. Only one member remains, a scrub in New Caledonia, and it shows no sign of heat generation, he and his team report.

Moreover, there's no sharp boundary for the evolutionary disappearance of thermogenesis. The trait does show up in a few lineages of moderately recent origin, such as the arums, the palms, and a related family sometimes called the Panama-hat palms. The Asian sacred lotus represents the highest branch on the botanical family tree that shows heat generation.

The bottom line


JUNGLE TRAP. A Philodendron solimoesense in French Guiana on the first day of blooming (above) starts emitting odors that attract insects. …



…The bloom heats up, and beetles cluster around, saving energy by drawing on the plant's warmth for their nighttime activities. The next evening, when the bloom cools, the beetles fly off.


Study of the evolution of heat generation raises questions about what benefits it might bring, or once brought, to a plant. The trait's absence among the newest plant families suggests that its value has declined as modern plants developed.

Biologists first proposed that heat helps spread the plants' insect-attracting odors. In contrast, one recent finding suggests that heat might make a plant more closely resemble a dead animal because microbial processes in a carcass raise its temperature.

Heating an artificially scented plant restored a fading bloom's capacity to lure insects into its pocket. "This is the first time it has been proved that this heating function of the plant is important, with scent, for guiding the pollinators," says Anna Maria Angioy of the University of Cagliari in Monserrato, Italy. She and her colleagues report their findings in an upcoming Proceedings of the Royal Society of London B: Biology Letters.

However, Seymour suggests another scenario. He points out that some plants keep the heat on after trapping insects in their chambers, so heat itself might serve as a reward for certain pollinators.

To test that idea, he and his colleagues studied Cyclocephala colasi beetles pollinating Philodendron solimoesense in French Guiana. As many beetles do, these produce heat to keep their body temperatures high enough for activity. Beetles active in a warm flower during the evening are using less than half the energy they would have used if they had stayed active out in the open, the researchers report in the Nov. 20 Nature.

The heat-generating flowers "are like nightclubs for beetles," Seymour says. The warm, alluring environment draws an insect in.

During evolution, a floral innovation may have supplanted the nightclub concept. A flower that offers just a sip of nectar or a pollen snack and then sends the pollinator on its way will probably spread its pollen over many more partners than will a plant that traps insects for a whole night. Seymour's take on why heat rewards died out is that "nightclubs were replaced by fast food."



This article speculated on the evolutionary origins of thermogenesis and observed how it predominates in ancient lineages of flowering plants like magnolias and water lilies. But thermogenesis goes back much farther than this, for it also occurs in cycads, nonflowering plants that arose in the Paleozoic. The male cones of some cycads, when mature, may maintain themselves at an astonishing 20° or more above the ambient temperature. This, as in the dead-horse arum, helps volatilize what is (to us) a foul smell. It is considered antisocial in Guam to have male cycads in one's garden, but for certain insects, apparently, this smell is a delicious one, for they are attracted to the hot cones and help distribute their pollen.

Oliver Sacks
New York, NY

The last section of this article briefly contrasts the rewards flowers give pollinators: heat versus pollen and nectar. It should be pointed out that the techniques are essentially identical, in that the plant is rewarding the pollinators with energy. In one case, the energy is passed as chemical energy. In the other case, it's radiant energy. The media are different, but the message is the same: an incentive for the pollinators to behave in a way that benefits the plant. The chemical technique may have a better ratio of energy expended to results produced (pollination and offspring). If so, selection pressures would tend to maximize its occurrence.

George Almond
Ridgewood, NJ


Angioy, A.-M., et al. In press. Function of the heater: The dead horse arum revisited. Proceedings of the Royal Society of London B: Biology Letters. Abstract availabe at

Kikukatsu, I., and O. Yoshihiko. 2003. Thermoscopic identification of heat-generating cells that co-express both SfAOX and SfUCPb genes in the spadix of skunk cabbage, Symplocarpus foetidus. Plant Biology 2003 meeting. July 25–30. Honolulu. Abstract available at

Laloi, M., et al. 1997. A plant cold-induced uncoupling protein. Nature 389(Sept. 11):135–136. Abstract available at

Seymour, M., M. Gibernau, and K. Ito. 2003. Thermogenesis and respiration of inflorescences of the dead horse arum Helicodiceros muscivorus, a pseudo-thermoregulatory aroid associated with fly pollination. Functional Ecology 17(December):886–894. Abstract available at

Seymour, R.S., C.R. White, and M. Gibernau. 2003. Heat reward for insect pollinators. Nature 426(Nov. 20):243–244. Abstract available at

Seymour, R.S., P. Schultze-Motel, and I. Lamprecht. 1998. Heat production by sacred lotus flowers depends on ambient temperature, not light cycle. Journal of Experimental Botany 49(July):1213–1217. Available at

Stensmyr, M.C. … and A.-M. Angioy. 2002. Rotting smell of dead-horse arum florets. Nature 420(Dec. 12):625–626. Abstract available at

Takanori, I., and I. Kikukatsu. 2003. Computer simulation of mitochondrial respiratory activity that controls the homeothermic heat-production in the spadix of skunk cabbage, Symplocarpus foetidus. Plant Biology 2003 meeting. July 25–30. Honolulu. Abstract available at

Thien, L.B., et al. 2003. The population structure and floral biology of Amborella trichopoda (Amborellaceae). Annals of the Missouri Botanical Garden 90:466–490.

Further Readings:

Angioy, A.-M., et al. 2003. Extreme sensitivity in an olfactory system. Chemical Senses 28(May):279–284. Abstract available at

Milius, S. 1999. Skunk cabbage has on-off heat switch. Science News 156(Aug. 21):123. References and sources available at

Seymour, R.S., and P. Schultze-Motel. 1997. Heat-producing flowers. Endeavour 21(No. 3):125–129. Abstract available at

Weiss, R. 1989. Blazing blossoms. Science News 135(June 24):392–394.


Anna Maria Angioy
Department of Experimental Biology
University of Cagliari
Strada Statale 554 - Km 4,500
09042 Monserrato-Cagliari

Marc Gibernau
Universite Paul Sabatier
U.M.R. 5174
118 Route de Narbonne - Bat 4R3
31062 Toulouse Cedex 4

Kikukatsu Ito
Cryobiosystem Research Center
Iwate University
Morioka, Iwate 020-8550

Roger S. Seymour
Environmental Biology
School of Earth and Environmental Sciences
University of Adelaide
Adelaide SA 5005

Leonard B. Thien
Department of Cell and Molecular Biology
Tulane University
New Orleans, LA 70118

From Science News, Volume 164, No. 24, December 13, 2003, p. 379.