Microplitis croceipes (Cresson) females were observed between 06:00 and 20:00 hours in cages containing spring host plants infested with Heliothis virescens (F.), tobacco budworm, larvae. Cages contained Geranium dissectum L. and Trifolium incarnatum L. in plantings of equal area, each infested with 50 H. virescens larvae. A time budget for the various behaviors exhibited by the wasps was obtained. In order of decreasing time spent, M. croceipes females were observed to exhibit standing still, searching on plant surface, preening, oriented flight, general flight, and walking. The proportion of time spent preening increased over the day. Oriented flight, general flight, searching, and walking peaked in the morning between 09:00 and 12:00 hours. Oriented flight, general flight, and walking showed minor increases in the afternoon. Microplitis croceipes females spent significantly more time preening, orienting to, and walking and searching on clover than on geranium. However, there was no difference in time spent standing still on geranium or clover. Significance of these findings for development of effective augmentative release strategies is discussed. KEY-WORDS : behavior, time budget, field cage.Parasitoid foraging is widely studied (see Hassell & Waage, 1984 ;Van Alphen & Vet, 1986 for reviews) because of the importance of parasitoid searching success and fitness to their practical application in biological control of pest insects. Most studies, however, are conducted under laboratory conditions and, thus, do not provide a reliable indication of wasp behavior under field conditions. The labor intensive nature of behavioral observations and the small size and high mobility of parasitoids make it difficult to observe active parasitoids for extended periods under field situations (Casas, 1989). However, use of field cages erected over natural vegetation allows observation of parasitoids under conditions similar to field environments. Direct observation of parasitoid behavior provides insight into the processes that lead to parasitization rates usually observed in the field. In contrast, measuring field parasitization rates is an indirect way of studying parasitoid behavior.(l)

Flight responses of the parasitoid, Microplitis croceipes (Cresson) (Hymenoptera: Braconidae) to two phenological stages of spring host plants (crimson clover, Trifolium incarnatum L., hairy vetch, Vicia villosa Corbiere, and cutleaf geranium, Geranium dissectum L.) and summer host plants (cotton, Gossypium hirsutum L., sorghum, Sorghum bicolor L., and soybean Glycine max (L.) Merr.) of Heliothis virescens (F.), were measured in a dual choice situation in a wind tunnel. Preflowering clover was more attractive than preflowering geranium, followed by preflowering vetch. Clover was preferred over geranium in flowering stages as well. No within species differences were detected for the two phenological stages. Preflowering and flowering cotton and soybean were preferred over preflowering and flowering sorghum. The findings could be important in pre‐determining release sites for M. croceipes in the field.

Responses of experienced Microplitis croceipes Cresson females in a wind tunnel choice situation were tested at three periods. Training consisted of exposing females to cotton or soybean leaf material with host (Heliothis virescens) larvae for parasitization or without host larvae. Control parasitoids received no initial exposure. Testing periods were: within 1 hour after exposure; 3 days after exposure or after exposure to reverse training. Pairwise comparison of choices of trained wasps versus controls showed a significant difference in only one case. Pairwise comparison of cotton versus soybean exposed wasps within each training level showed significance for “rewarded” wasps in periods 1 and 3 with highest numbers of wasps flying to the source‐plant of initial training. Results suggest the occurrence of associative learning of M. croceipes on host‐host plants, and indicate a time dependent component in the learning process and irreversibility of this conditioning. Zusammenfassung Zum Lernverhalten von Microplitis croceipes Cresson (Hym., Braconidae) Mittels Wahlversuchen in einem Windkanal wurden die Reaktionen von geprägten Microplitis croceipes‐Weibchen in drei verschiedenen Zeitabschnitten nach der Prägungsphase getestet. Die Prägung bestand darin, daß den Weibchen Baumwoll‐ oder Sojabohnenblätter mit und ohne Wirtslarven (Heliothis virescens) angeboten wurden. Die Kontrolltiere wurden keiner Prägung unterzogen. Die Windkanalversuche wurden wie folgt durchgeführt: 1 Stunde nach der Prägung, 3 Tage danach oder nachdem die Weibchen nicht der Prägungspflanze, sondern der anderen Art ausgesetzt wurden. Der paarweise Vergleich der Ergebnisse der Wahlversuche bei den geprägten und den Kontrolltieren ergab nur in einem Fall einen signifikanten Unterschied. Der paarweise Vergleich von Parasitoidenweibchen, die auf Baumwolle bzw. Sojabohne geprägt waren, ergab in den Uberprüfungsperioden 1 und 3 einen signifikant höheren Anflug an die Pflanze, auf welche die Weibchen geprägt wurden. Die Ergebnisse lassen darauf schließen, daß bei M. croceipes eine Prägung auf Wirt und Wirtspflanze vorliegt, daß diese Prägung zeitabhängig ist und daß es sich dabei um eine irreversible Prägung handelt.