Japan's red dragonflies are born yellow. A chemical switch turns the males red, and the most famous one is vanishing.
Every autumn, akatombo (red dragonflies) become the sign of the season across Japan. But the red is not something these insects are born with. Young dragonflies are yellow, and a pigment reaction turns the mature males red. The best-known species, the akiakane, also has a more worrying story: its numbers fell sharply in the 1990s, and researchers have looked hard at what happened in the rice paddies where it grows up. (We'll teach you the Japanese words as we go.)
"Red dragonfly" is a group, not a single insect
赤とんぼ — akatombo (red dragonfly)
Akatombo is the everyday Japanese word for the red-bodied dragonflies of the genus Sympetrum, and Japanese sources count roughly twenty species. The one most people picture is the akiakane (秋茜, Sympetrum frequens). Its life follows a surprising loop: the adults emerge from rice paddies and wet ground on the plains in early summer, move up to cooler high ground for the hot months, and come back down to the lowlands in autumn. Mature males turn red on the abdomen, but not on the head or thorax, while females stay yellowish-brown, and some show only a slight red tinge on the back of the abdomen.
Yellow pigment, red pigment: the same molecule in two states
成熟 — seijuku (maturation)
In 2012, Ryo Futahashi and colleagues published a study in PNAS, "Redox alters yellow dragonflies into red", that worked on Sympetrum frequens, Sympetrum darwinianum and Crocothemis servilia. The color comes from two ommochrome pigments, xanthommatin and decarboxylated xanthommatin, which sit as pigment granules inside epidermal cells of the abdomen. In the reduced form they look red, and in the oxidized form they look yellow. The team measured that 90% to 100% of the pigment was in the reduced form in mature males, against roughly 55% to 75% in mature females. Then they tested it directly: after injecting a reducing agent, yellow immature males and even yellow mature females turned red within several hours. So the red of seijuku is not a new pigment being built, but a chemical state being switched. The authors note that male-specific coloration like this plays roles in mating and territoriality.
The sharp decline of the akiakane
田んぼ — tanbo (rice paddy), 農薬 — nōyaku (agricultural chemicals)
The akiakane's larvae grow up in the water of the tanbo, which is why its fate is tied to rice farming. A review in Environmental Science and Pollution Research by Kosuke Nakanishi, Hiroyuki Yokomizo and Takehiko Hayashi of Japan's National Institute for Environmental Studies asked whether the sharp declines of the 1990s were caused by two insecticides, fipronil and imidacloprid, used in nursery boxes for rice seedlings. They point out that these insecticides are highly toxic to dragonfly nymphs, that early-stage nymphs are exposed to relatively high concentrations right after the seedlings are transplanted, and that the introduction of the insecticides in the late 1990s overlapped with the sharp declines. They concluded that the use of these insecticides, especially fipronil, was a major cause of the akiakane's decline, with a high degree of certainty. They also say the evidence cannot rule out that some farming practices, such as midsummer drainage or crop rotation, which can limit the survival of aquatic nymphs, played a role too. Warming has also been suggested in Japanese sources, for example by keeping lowland paddies too warm in the season when the dragonflies return, but that explanation is less settled.
Why a dragonfly's color is more than decoration
The same insect gives you two lessons. First, color can be a state rather than a fixed trait: a reaction can move an animal from a yellow, immature look to a red, mature signal in a few hours, and the amount of reduced pigment tracks sex and age. Second, even a symbol as familiar as the akatombo depends on ordinary farmland practices. The nursery song Akatombo, written by Miki Rofu (1889 to 1964), who is from Tatsuno in Hyogo, has a line about a dragonfly resting at the tip of a pole, and that is often taken to be an akiakane. If you are in Japan in autumn, the red dragonflies you see on lowland paddies and paths are the ones that spent the summer on higher ground and came back down.
Common questions
Q. Are Japan's red dragonflies born red?
A. No. Young dragonflies of the genus Sympetrum are yellow, and the males turn red as they mature.
Q. What makes the red color?
A. Ommochrome pigments (xanthommatin and decarboxylated xanthommatin) in the epidermis of the abdomen. In the reduced form they look red and in the oxidized form they look yellow, and a 2012 PNAS study showed that injecting a reducing agent turned yellow individuals red within several hours.
Q. Do female red dragonflies also turn red?
A. Mostly not. In the akiakane the females stay yellowish-brown, and some show only a slight reddening on the back of the abdomen. The study found 90% to 100% of the pigment in reduced form in mature males, against roughly 55% to 75% in mature females.
Q. Why has the akiakane declined?
A. A review by researchers at Japan's National Institute for Environmental Studies concluded that rice nursery-box insecticides, especially fipronil, were a major cause of the sharp declines in the 1990s, with a high degree of certainty, while noting that some farming practices may also have contributed.
Q. When can you see red dragonflies in Japan?
A. Adults are seen from June to December. The akiakane emerges on the plains in early summer, spends the summer on cooler high ground, and returns to the lowlands in autumn, which is when most visitors notice them.
Where to find Japan's wildlife in the wild: our region-by-region wildlife guide. Or explore all 47 prefectures by recorded species in Ikimono Quest, built from open biodiversity data.
Written by naturalists. The science here reflects the published study "Redox alters yellow dragonflies into red" (Futahashi, Kurita, Mano and Fukatsu, Proceedings of the National Academy of Sciences, 2012, 109:12626-12631) on Sympetrum frequens, Sympetrum darwinianum and Crocothemis servilia, and the review "Were the sharp declines of dragonfly populations in the 1990s in Japan caused by fipronil and imidacloprid? An analysis of Hill's causality for the case of Sympetrum frequens" (Nakanishi, Yokomizo and Hayashi, Environmental Science and Pollution Research). Life-history details come from Japanese prefectural and museum sources. Species identification is of real, extant insects native to Japan. Nature is full of exceptions — that's what makes it worth studying.