Thripinema khrustalevi (Chizov et al.) isolate Chile, a parasite of Frankliniella australis (Morgan), was collected from Cestrum parqui (L’Herit.) in the Aconcogua Valley of central Chile. Percent infection of males and females of F. australis by T. khrustalevi in the flowers of C. parqui was estimated biweekly for two years at three locations, including the La Campana National Park. Males were less infected than females. Numbers of F. australis adults and larvae in the flowers of C. parqui were greatest in the winter and early spring when parasitism of the adult females was low. Populations declined in late spring when parasitism was high. Parasitism remained high through the summer and fall, and populations of thrips remained very low. Highest parasitism of the females and males of F. australis was 84 and 60%, respectively.
Nematodes infesting Thysanoptera were first recorded from Europe by Uzel (1895) and from North America by Russell (1912). Neither species was described. Sharga (1932) described a nematode as Tylenchus aptini from Aptinothrips rufus (Gmelin) collected in England. Lysaght (1936) proposed the taxon Anguillulina aptini for this species. Wachek (1955) transferred the species to the genus Howardula. Nickle and Wood (1964) reported Howardula aptini infecting Frankliniella vaccinii Morgan and Taeniothrips vaccinophilus Hood collected in blueberry.
Siddiqi (1986) created a new genus, Thripinema, based on host range and the oval shape of the parasitic female. His revision of the genus included renaming the nematodes described by Sharga (1932), Nickle and Wood (1964), and Reddy et al.(1982) as T. aptini, T. nicklewoodi, and T. reniraoi, respectively. Chizov et al. (1995) described Thripinema khrustalevi from the thrips hosts, Thrips trehernei Prisner and T. physopus L. Tipping et al. (1998) described T. fuscum from Frankliniella fusca (Hinds) in peanut. Teulon et al. (1997) reported an unidentified species of Thripinema infecting Thrips obscuratus (Crawford) in New Zealand.
Loomans et al. (1997) reviewed the biology and life cycles of Thripinema species. One heterosexual generation develops in the haemocoel of the adult thrips abdominal and thoracic cavities. The oviparous parasitic female releases eggs into the haemocoel and the motile vermiform larvae feed on the thrips’ internal fluid. Mature larvae exit the host by boring through the wall of the midgut. From the midgut, the larvae move to the pyriform rectum where they reside for a period before issuing from the anus (Sharga, 1932; Lysaght, 1936) or the ovipositor (Reddy et al., 1982). Little is known about the free-living stage. The female attaches to and enters the thrips host by boring through the soft intersegmental membranes (Lim et al., 2001). Loomans et al. (1997) noted that there was no record of a parasitized male thrips, but Tipping et al. (1998) noted that males of F. fusca collected in peanuts were frequently infected by T. fuscum. In laboratory experiments, male F. occidentalis were parasitized by T. nicklewoodi as readily as the females (Lim et al., 2001).
Sharga (1932) reported that infection of A. rufus by T. aptini in the UK was lowest in March (12.3%) and highest in July (37.5%). Lysaght (1937) reported a maximum of 64% in late August and early September, and infection levels less than 10% from November to March. Reddy et. al. (1982) determined from weekly samples that percent infection of a population of Megaluriothrips sp. by T. reniraoi in India ranged between 45 and 63%. Infection of T. trehernei by T. khrustalevi in Russia ranged between 10 and 50% (Chizov et al. 1995). The highest percent infection of F. fusca females by T. fuscum in Florida observed by Tipping et al. (1998) and Funderburk et al. (2002) was 68%.
Surveys of the natural enemies of thrips in Chile were begun following the accidental introduction of Frankliniella occidentalis (Pergande) in 1992. Nematodes were found infecting a native species of thrips, Frankliniella australis (Morgan). Morphological and morphometric studies show that the nematode is closely related to Thripinema khrustalevi. In this paper we consider the nematode as T. khrustalevi isolate Chile. The nematode was first discovered infecting thrips collected in the La Campana National Park. Cestrum parqui (L’Herit.) (Solanaceae) is a native plant species that serves as a reproductive host for populations of F. australis and F. occidentalis in Chile. We report the infection rates of F. australis in the flowers of C. parqui from three locations in the Aconcogua Valley of Central Chile. Population densities of thrips and other natural enemies of thrips also are reported.
Materials and Methods
Estimating Weekly Infection Rates
The three locations sampled were: (1) La Campana National Park near the access gate to Sector Ocoa, (2) near the city of Quillota, and (3) near the city of Llay Llay. Flowers of C. parqui were present on most sample dates from May 1999 to June 2001. From 15 to 20 females of F. australis were dissected from each of four samples of about 10 inflorescences randomly collected on each sample date at each location, and the number parasitized was determined. Densities of male thrips were always less than densities of females, and estimates of parasitism of males were not possible on some dates. Further, only 5 males were dissected from each of four samples on dates when males were present in adequate numbers. Parasitism of males was estimated on 12, 12, and 20 sample dates at the Llay Llay, Quillota, and La Campana locations, respectively. Adults of F. occidentalis also were dissected. Percent infection of males and females at each location transformed to arcsine square root were subjected to ANOVA and significant differences were determined using a t-test.
Thrips were dissected under a stereomicroscope at ×35 in distilled water and considered infected if any stage of the nematode was detected. The swollen females, eggs, and juvenile stages were easily detected during dissection. These stages are present in thrips within several days of infection by the infective-stage females (Lim et al., 2001). Prior to swelling, an infective-stage female in the absence of other stages of nematodes sometimes went undetected during dissection. To determine how much parasitism was being underestimated with the dissection procedures, 20 females on 9 to 12 dates depending on the location were maintained individually for 7 days and then dissected. This method provided percent infection that was consistently about 10% greater than the estimate of percent infection from females dissected immediately after collection.
Estimating Densities of Insect Populations
Four samples of five open flowers were collected biweekly from October 1999 to June 2001 at each of the locations listed above. Flowers were placed in vials containing soapy water and returned to the laboratory for processing. Samples were examined under a stereoscope at ×35, and the numbers of thrips and natural enemies determined.
Results and Discussion
Adults of F. occidentalis collected from the flowers were never infected by T. khrustalevi isolate Chile. Similarly, Funderburk et al. (2002) reported that infection of F. occidentalis by T. fuscum in peanut flowers was rare. Adults and larvae of F. occidentalis were readily infected under laboratory conditions (unpublished data). The reason that these field populations of F. occidentalis were not parasitized by T. fuscum or T. khrustalevi is not clear. Thripinema nicklewoodi is an important parasite of F. occidentalis in the western USA (Loomans et al., 1997), and the species has potential for biological control if introduced into Chile or other geographic regions.
Adults and second instars of F. australis collected from the flowers of C. parqui were parasitized. The ovaries of the infected females were reduced and eggs absent. After entering the thrips host, the females of the new isolate change into the characteristic swollen shape of the genus, and the ovary becomes convoluted with numerous flexures. Up to several hundred nematodes were observed in individual females of F. australis. All stages of the nematode were observed in infected males and females. Males and females of T. fuscum were observed exiting the thrips via the anus as reported by Lysaght (1937) for T. aptini, Nickle and Wood (1964) for T. nicklewoodi, and Tipping et al. (1998) for T. fuscum.
Open flowers of C. parqui were present on nearly all sample dates from May 1999 to June 2001 at the Quillota and La Campana locations (Fig. 1). Flowering was suppressed during the winter and there were several sample dates in July and August 2000 when population densities and % infection of thrips could not be estimated because of a scarcity of open flowers at these locations. Temperatures were lower at the Llay Llay location, and open flowers were not available for sampling for several months each winter. About 96% (range 95 to 97% between the locations) of the adult thrips in the flower samples were F. australis, with the remainder being F. occidentalis. No natural enemies other than T. khrustalevi were observed that could have contributed significantly to suppression of populations of thrips. The only predator in the flower samples was Orius insidiosus (Say), but only four adults and nymphs were found total in all of the samples. A total of 13 adults of Ceranisus menes (Walker) (Hymenoptera: Eulophidae), a larval parasite of thrips, were collected.
Numbers of F. australis increased greatly in the C. parqui flowers each year in October/early November when levels of % parasitism of the adult females was very low (Fig. 1). Adults and larvae were abundant through the remainder of each spring. Populations of adults and larvae decreased to near extinction in February, persisting in low numbers until the following spring. Parasitism of the adult females increased each January and was high at each location through the summer and fall until the beginning of winter in May/early June. Parasitism of males was almost always less than parasitism of females (data not shown). On dates when parasitism of males and females was estimated, % parasitism of males vs females averaged 9.2 (range 0 to 35) vs 15.2 (range 0 to 45), 11.3 (range 0 to 60) vs 16.8 (range 0 to 69), and 4.5 (range 0 to 35) vs 16.8 (range 0 to 42.5) at the Llay Llay, Quillota, and La Campana locations, respectively. Parasitism of the females was significantly greater at the La Campana location, according to the t-test (error df = 158, P < 0.0001) and the difference approached significance at the Llay Llay location (error df = 94, P = 0.08).
Reported cases of suppression of thrips populations by natural enemies were lacking until recently (Parrella and Lewis, 1997), and the population attributes of thrips including rapid colonization and growth were believed to possibly outstrip the capacities of natural enemies to regulate populations (Mound and Teulon, 1995). Funderburk et al. (2000) and Ramachandran et al. (2001) showed that O. insidiosus suppressed rapidly colonizing and developing populations of thrips in pepper. Funderburk et al. (2002) showed that T. fuscum was an important natural enemy affecting the population dynamics of F. fusca in peanut. Several other investigators reported high levels of infection during summer and fall for other species of Thripinema (Sharga, 1932; Lysaght, 1937; Reddy et al., 1982) which may indicate that natural suppression of thrips populations under local conditions by species of Thripinema is widespread. We conclude in this study that T. khrustalevi isolate Chile is an important natural enemy of F. australis in C. parqui.
The authors thank K. B. Nguyen (Department of Entomology and Nematology, University of Florida) for the identification of the nematode Thripinema khrustalevi, and L. E. Mound (Division of Entomology, Commonwealth Scientific and Industrial Research Organization) for verifying the identification of Frankliniella australis. Stuart Reitz (USDA-ARS) and Russell Mizell (North Florida Research and Education Center, University of Florida) made comments on the manuscript. Ruby Mariana Silva Cutbill provided encouragement and permitted the sampling on her eco-tourism ranch in Quillota, Chile. We are thankful to the park rangers at the La Campana National Park for their many assistances. This article is Florida Agricultural Experiment Station Journal Series number R-08739.