Remote sensing of thermal radiation (infrared wavelengths) has been reported only in some terrestrial animals and is known to have significant physiological and ecological meaning. In aquatic animals, however, it has not even been discussed because water almost completely absorbs infrared (IR) wavelengths, and such sensitivity has been regarded as useless in aquatic environments. Here we report, for the first time, an anomalous behavior—swimming toward an IR-radiating source—in an aquatic animal, the giant jellyfish, Nemopilema nomurai. Experiments were performed with laboratory-reared juvenile medusae of Nemopilema in a small experimental chamber. In all 52 trials, the majority (average: 78.3%; variation: 63.3%– 94.3%) of the medusae gathered on the IR-irradiated half of the chamber. Near-IR (850/940 nm) and visible light neither attracted the medusae nor induced taxis. A series of results indicated that the juveniles of Nemopilema tended to flock toward a mass of warmer water. Although the exact physiological and ecological meaning of this behavior is uncertain, it might be possible to consider infrared taxis as one of the behaviors in marine ecosystems. It may thus be useful to re-examine the swarming behavior reported in some jellyfish species from the viewpoint of IR taxis. Infrared (IR) sensitivity is known to be mediated by the so-called pit organs, which enable crotaline and boid snakes to apprehend homeothermic prey (1–7) and vampire bats to detect IR radiation from blood-rich locations (8, 9). Moreover, similar mechanisms have been reported in some insects such as fine debris species of beetle (10–13), bloodsucking bugs (14), and some butterflies (15, 16). In aquatic animals, however, little is known about the IR sensitivity. This is the first report to describe an anomalous IR-tactic behavior in the giant jellyfish, Nemopilema nomurai, living in an aquatic environment. Nemopilema nomurai Kishinouye 1922 is one of the largest scyphozoan jellyfish, measuring 1–2 m in diameter and weighing 100–200 kg. Massive blooms of this species have occurred almost annually since 2002, severely affecting coastal fisheries in the Sea of Japan. The life cycle of N. nomurai has been well established by us and our coworkers (17, 18), so the juvenile medusae are easily available. We thus used laboratory-reared juveniles (extended bell diameter, 8–15 mm) to examine the IR sensitivity of N. nomurai. The experimental chamber (width 30 cm, length 24 cm, height 16 cm) used to test IR sensitivity was made from transparent acryl resin plate (5 mm thick). It was filled with seawater to a depth of 3.5 cm and 20 to 35 juvenile individuals were added. The distribution of the medusae in the chamber was monitored by photographing them from above with a photographic strobe every 30 or 60 min. The IR wavelengths were obtained from a tungsten-filament lamp (40 W) by passing the light through a long-pass filter (50% cut-on at 990 nm; LIO-990, Asahi Spectra Co. Ltd.). The tungsten-filament lamp was fixed in a lightproof aluminum housing with a window, where the long-pass filter was mounted. All possible heat sources, except for the IR-wavelength source, were excluded from around the experimental chamber. Near-IR wavelengths, 940 and 850 nm, were obtained from two types of 1.2-W LEDs, LB12WP01 and LC12-WP01 (Ebisu Electronics Co. Ltd.), reReceived 1 August 2011; accepted 20 October 2011. * To whom correspondence should be addressed. E-mail: ohtsu@ life.shimane-u.ac.jp Reference: Biol. Bull. 221: 243–247. (December 2011) © 2011 Marine Biological Laboratory