<p>Tirumala limniace and T. septentrionis, migrant danaines, fly over to Japan every year from the further south. In the summer of 1995, numerous T. limniace and a few T. septenrionis arrived at Amami-ohshima Isl., and aggregated together with Ideopsis similis, a resident danaine there. Although their overwintering aggregations in their residential regions (e.g. Taiwan) have been reported, "aestivating aggregations" have little been noted. The present paper reports such a case and describes the behavior of the aggregated butterflies during the hot season at their northern temporary residence (Amami-ohshima). 1. From 4 July to 2 September in 1995, the migrant Tirumala limniace and T. septentrionis and the resident Ideopsis similis were observed forming an aggregation at the northern part of Amami-ohshima. 2. The two migrant species were collected as specimens (i. e. removed; 253 [160 ♂ ♂ 93 ♀ ♀] of T, limniace and 4 ♂ ♂ of T. septentrionis in total) almost every day by one of the authors, while none of I. similis, which was as numerous as T. llimiace, was removed, 3. They showed routine behaviors daily-visiting flowers or mating in the cool hours of early morning and late afternoon, but resting during the hot hours around midday. The weather in July-September was mostly sunny or cloudy (sometimes with showers) with the highest temperature over 30℃ , and no typhoon came in the study season. 4. Their habitats consisted of two sites, wooded areas for resting and flower beds for feeding. In late July they moved from the first site (station A) to the second site (station B) distanced about 30m from the former. 5. They mostly rested on some aerial roots of Ficus microcarpa and F. elastica (Moraceae) or small twigs of Jatropha hastata (Euphorbiaceae). Usually T. limniace rested higher than /. similis in position. 6. Nectar source flowers were Euphorbia milii var. splendens and Jatropha hastata (Euphorbiaceae). 7. Ten pairs of T. limniace in copula were observed in the field. The presence of spermatophore examined by fingers indicated most T. limniace females were fertile already. Due to the absence of foodplants, however, oviposition did not occur. On the other hand, despite of the presence of foodplant (Tylophora tanakae), I. similis did not lay their eggs there either. 8. Why did they make such a notable aggregation in 1995 at Amami-ohshima? The arrival of such an unusual large number (over 280) of T. limniace is likely to have been the fundamental cause. Although from April to July of the year, remarkable numbers of migrant or visiting butterflies, including newly recorded species in Japan, flew to the Nansei Islands from the Asian continent, Taiwan, and the Philippines, the number of T. limniace on Amami-ohshima was outstanding. 9. The departure site of the two species of Tirumala is inferred to be somewhere in Taiwan for two reasons. (1) The T. limniace was identified as ssp. limniace Cramer distributed in Taiwan and the Asian continent, but not as ssp. orestilla Fruhstorfer found in Luzon Isl. The T. septentrionis, which is not distributes in Luzon Isl., was ssp. septentrionis Butler occuring in Taiwan and the Asian continent. (2) The two Tirumala species are common in Taiwan in June and July together with the Chestnut Tiger (Parantica sit a), which is confirmed to migrate to Japan by markrelease-recapture method . The two Tirumala species are often found together with the Chestnut Tiger at their arrival sites in Japan. 10. In conjunction with other records of migrant butterflies in Japan, the two Tirumala species likely arrived at Amami-ohshima in the end of June, and move northward to reach the northern region of the island. Then they aggregated at this site to spend the hot summer days. In addition, possibly they were attracted by Ideopsis similis, which had aggregated there as precedent visitors. 11. One female of T. limniace bred in cage to obtain eggs lived for</p><p>(View PDF for the rest of the abstract.)</p>