(34(2):213-222)Introduction, Propagation and Liberation of a Pupal Parastoid, Tetraslichus brontispae, for the Control of the Coconut Leaf Beetle in Taiwan

本所於民國72 年11 月自關島引進可可椰子紅胸葉蟲（Brontispa longissima）之蛹寄生蜂Tetrastichus brontispae Ferr. 供可可椰子紅胸葉蟲之防治。該蜂引進本省後先在本所檢疫室進行檢疫、飼育以及生物特性等觀察。經繁殖3 世代後即在高雄澄清湖與屏東林邊兩地釋放。農試所在室內飼育該蜂，發現其發育期之長短隨溫度之高低而有不同，於室溫20℃時，從卵發育至成蜂羽化，最短需17 天，最長達32 天。各蟲期分別為卵期2-3 天，幼蟲期5-6 天，蛹期10-11 天。平均雌蜂壽命約10.8 天，雄蜂約3.6 天。初步飼育觀察該蜂在室內之產卵寄生率平均為60.13 ±8.0 %。T. brontispae 蜂於民國72 年1 月18 日開始，先後10 次釋放於澄清湖畔之椰樹林，至同年8 月其寄生率為18.8-79.2 %。另於同年2 月8 日釋放於林邊，在3 個月內釋放7 次其寄生率為9.3-44.2 %。該蜂在兩地之釋放後均能寄生於紅胸葉蟲之蛹，順利繁衍，而發揮其生物防治效果，使受害之可可椰子樹逐漸恢復生機。本文報導該蜂之引進本省經過，室內增殖、部分生物特性觀察及其釋放於高、屏地區後，對可可椰子紅胸葉蟲棲群之抑制效果。 The coconut tree, Cocos nucifera L., grown in the tropical and subtropical regions of the world, is the best known palm with high economic values. In Taiwan, it can be grown on the east coast and in areas south of Taichung on the west. The number of coconut trees planted in 1960 was estimated at 12,000 trees and was gradually increased to a total of 60,000 trees in 1984. The coconut leaf beetle, Brontispa longissima Gestro (Chrysomelidae), a serious pest of coconut trees, was first detected in Pingtung in July 1975. Since then, it has spread north to Taitung and then to Hualien. Since coconut trees are tall, reaching a height of 15-20 m., and are commonly planted along highways and around fish ponds, it renders the use of chemical pesticides for control of this pest economically impractical and environmentally unsafe. Therefore, the biological control approach was chosen as a viable alternative for solving this serious pest problem. Through the cooperative efforts of Dr. R. Muniappan, University of Guam, Tetrastichus brontispae Ferriere (Eulophidae), was field collected and shipped to Taiwan. The first shipment of B. palauensis (Esaki and Chtjo) parasitized by T. brontispae was received via air mail on November 3, 1983 and the second shipment was hand delivered on November 5, 1983. Upon receipt, the shipments were screened and studied in a quarantine laboratory at Taiwan Agricultural Research Institute. The packing material was autoclaved at 120°C, 15 PSI for 30 mm. and discarded. B. palauensis adults which had escaped parasitization by T. brontispae were also destroyed by autoclaving. Under the room temperature of 21°C, T. brontispae required 17 to 32 days to complete its life cycle from egg to adult. The durations of egg, larval and pupal stages were 2 to 3, 5 to 6, and 10 to 11 days, respectively. The longevity of adult male and female was 3.6 and 10.8 days, respectively. T. brontispae is a pupal parasitoid which apparently preferred to oviposit in one- day old pupae of B. longissima. The average percent parasitism in the laboratory tests was 60.1±8.0%. After three generations had elapsed under tightly controlled quarantine conditions,T. brontispae was field released at coconut patches in Chenchinhu and Linbien. A total of 10 releases of 11,456 T. brontispae adults in Chenchinhu was made during January-July 1984 and seven releases of 4,881 adults in Linbien during February-June 1984. The percent parasitism recorded from field recoveries made in Chenchinhu and Linbien was 21.2 to 79.2% and 9.3 to 36.2%, respectively. The population dynamics of B. longissima was also investigated at both Chenchinhu and Linbien. Based on the comparison of the pre- and post-release data, the results showed that the population densities of B. longissima in Chenchinhu fluctuated from an average of over 190 larvae per plant to 60 larvae per plant during September-December 1983. However, after the release of T. brontispae in January 1984, the population densities of B. longissima were kept below an average of 40 larvae per plant, except April and May 1984 , when there was an upsurge of B. longissima. This upsurge was subsequently suppressed after 4 releases of T. brontispae to a level with less than 10 larvae per plant during August-December 1984. A similar trend of this population fluctuation at much lower levels was also observed on B. longissima adults. This indicated that T. brontispae prevented the majority of larvae of B. longissima from developing into adults. In Linbien, however, T. brontispae appeared to be unable to suppress B. longissima populations. Although the percent parasitism ranged from 15 to 40%, the effectiveness of T. brontispae at this location remains unproven and further studies will definitely be needed. The ability of T. brontispae to disperse in field conditions was demonstrated when it was repeatedly recorded at Niaushun, Fengshan, Wujaen and Tashue, approximately 2 to 8 km from the release site in Chenchinhu. Hence, it strongly indicates that T. brontispae has become well established in Taiwan and gradually dispersed by itself among coconut growing areas. It also indicates that T. brontispae does possess the desirable attributes in host searching and dispersal to become an effective biological control agent of B. longissima