Creighton and Fassuliotis (1981) were the first to give a detailed report of the laboratory culture of a mermithid of terrestrial insects. Preparasitic juveniles of F. leipsandra and the larvae of the cucumber beetle, Diabrotica blateata, were placed in holes in soil in clay pots. The pots were sealed and held for parasite emergence. Ratios of 1000 nematodes per 100 host larvae were sufficient for maximum recovery of postparasitic nematodes. The method has the potential of producing between 80,000 and 240,000 nematode eggs per experimental unit. From this basic procedure, Creighton and Fassuliotis (1982) developed an in vivo mass rearing system for F. leipsandra. As eggs were laid, they were concentrated, sterilized by immersion for 10 minutes in 0.28% sodium hypochlorite, and rinsed. The eggs were transferred to a Baermann funnel and allowed to mature and hatch. First instar larvae of the banded cucumber beetle were exposed to the preparasites in clay pots as previously described. Twenty-four pots were placed in a plastic box and held 4-5 days at 24-27'C. The contents of the clay pots were then transferred to porous baskets. The baskets were placed at one end of a larger container with sand at the other end. As the host larvae matured, they migrated to the sand and pupated. After about 10 days the nematodes began to emerge from their hosts. The postparasites were then transferred to a container of aerated water until the nematodes molted to the adult stage and began to deposit eggs (7-10 days). Each larger container (37 x 27 x 10 cm) produced an average yield of 1.9 X 106 nematode eggs. The system can produce about 5 x 106 eggs week- I at a cost of $0.40 (US$) per I 000 eggs including supplies and labor for producing the host insects (1981).
Although mermithids are relatively common in aquatic insects, especially in the Simuliidae, Chironomidae, Ceratopogonidae, and Culicidae, many of these host species cannot be successfully maintained in laboratory colonies or cannot be maintained in large numbers. Also many mermithid species undergo egg diapause or have an asynchronous hatch which prevents their mass production. As a result, few aquatic mermithid species have been studied extensively. The midge parasite, Hydromermis conopophaga, has been readily maintained in colony because the host, Tanytarsus midges, can be easily maintained. However, H. conopophaga has never been mass produced or released in the field (Poinar, 1979). Similarly, the mermithid Limnomermis rosea, a parasite of chironomid larvae, has been reared through several generations but has not been released in the field.
Muspratt (1947) was the first person to develop, maintain, and report a laboratory culture of a mermithid nematode. He collected 0. muspratti (Agamomermis) from naturally infected host mosquito larvae and placed the nematodes in containers filled with moist sandy soil. The soil was allowed to dry out over a period of several months. After 11-12 months some of the soil was placed in a container of water and first instar mosquitoes added. Seventy to 80% of the host larvae contained nematodes. The original work by Muspratt though very basic provided the direction for future work with mosquito mermithids. Subsequent studies with 0. muspratti have refined the rearing technique and defined biological limitations (Petersen, 1977). A mass production system for 0. muspratti has failed to develop, even though a suitable laboratory host system is available, because the eggs of this species failed to hatch at the same time although they appeared to mature. Because of the nonsynchronous hatch, cultures were observed to produce infective stage nematodes after periodic floodings for over 5 years (Petersen, 1981).
The mermithid, S. peterseni, a parasite of Anopheles mosquitoes, has been maintained in colony for over 10 years (Woodard and Fukuda, 1977). Though this nematode appears to be an infective parasite, and is easily maintained in laboratory colony, economical mass production systems have been difficult to achieve because the only hosts are Anopheles mosquitoes, which are less suitable for mass rearing systems than are certain Aedes and Culex species.
The most efficient and effective in vivo rearing system for mermithids is that developed for R. culicivorax. The primary techniques were developed in 1971 (Petersen and Willis, 1972a). This initial system called for the exposure of 20,000 first-instar Cx. quinquefasciatus larvae in 136 x 52 x 5 cm galvanized 'Lrays to preparasites at a parasite-host ratio of 12: 1. The hosts were then fed a regimented diet for 7 days. After this period the culture trays were drained and the mosquitoes concentrated. Pupae of uninfected mosquitoes were separated using a chilled water procedure, and the infected mosquitoes were placed in 36 x 25 x 10 cm trays and held for nematode emergence. Postparasitic nematodes were then washed and 10-15 g placed in paraffin-coated aluminum pans (22 x 33 x 5 cm) which contained clean, coarse, sterile sand and water. After 3 weeks the free water was removed and the cultures stored an additional 4-15 weeks before use. When preparasitic nematodes were needed the cultures were flooded with chlorine-free water to stimulate nematode hatch. Nematode cultures averaged 1.91 million preparasites (45-60% of total yield) if they were flooded for the first time when they were 11- 19 weeks old. Total yields were highest (5.32 million preparasites) when cultures were flooded for the first time when 8-10 weeks old and then at 3-4 week intervals thereafter (Petersen, 1978b).
It has since been determined that yields can be increased by having a density of 24 postparasites CM-2 in the culture trays. Yields of preparasites were tripled by simply setting up three cultures (22 x 33 x 5 cm), each containing 5 g of nematodes, instead of the usual method of one culture containing 15 g of postparasites (Petersen, 1980). The longevity of cultures could be extended for up to 9 weeks if the cultures were allowed to mature at ambient temperature and then held at 5-IOC. Also, immature cultures survived best at 15-20'C (Petersen, 1979b). Further, Sterling and Platzer (1978) reported that by adjusting the pH of water to 4.5 in R. culicivorax cultures, the nematophagous fungus C. anguillulae could be effectively controlled. Chapman and Finney (1982) reported that mature eggs of R. culicivorax can be stored for 12 weeks at IO'C and 21 weeks if stored during early development at 15'C and later transferred to IO'C. They also reported that eggs of R. culicivorax placed in a damp fluoroform product could be shipped with only a 10% loss. The technique has the advantages over the standard shipment of sand cultures of being about eight times lighter, and costing less to ship in addition to reducing shipping losses.
In an actual test of the mass propagation system, sufficient R. culicivorax were reared to treat 144,000 m2 of breeding area in El Salvador. The necessary inoculum required the exposure of 1.6 million first-instar Cx. quinquefasciatus to 137 million preparasites (I : 14 ratio) each week for 6 weeks. The system produced an average of 13.7 g (about 2000 g- 1) of postparasitic nematodes per rearing tray (20,000 mosquitoes), a total of 6392 g for the 6-week period and 435 cultures (Petersen et al., 1978a).