Journal of Vector Ecology 107. Tianyun Su 1 and Mir S. Mulla 2

Transcription

1 Vol. 30, no. 1 Journal of Vector Ecology 107 Toxicity and effects of microbial mosquito larvicides and larvicidal oil on the development and fecundity of the tadpole shrimp Triops newberryi (Packard) (Notostraca: Triopsidae) Tianyun Su 1 and Mir S. Mulla 2 1 Coachella Valley Mosquito and Vector Control District, Trader Place, Indio, CA 92201, U.S.A. 2 Department of Entomology, University of California, Riverside, CA 92521, U.S.A. Received 18 August 2004; Accepted 16 February 2005 ABSTRACT: Tadpole shrimp (TPS), Triops newberryi (Packard) (Notostraca: Triopsidae), is a predatory agent for immature mosquitoes breeding in aquatic habitats. This biological control agent could be used with other larvicides in mosquito control programs. In order to elucidate compatibility of the TPS and commonly used mosquito larvicides, studies were initiated to investigate effects of Bacillus thuringiensis ssp. israelensis (B.t.i.) de Barjac, Bacillus sphaericus Neide and Golden Bear larvicidal oil on growth, longevity, and fecundity of TPS in laboratory and field. The exposure of TPS to high dosages of B.t.i. and B. sphaericus in the laboratory or in the field did not have significant adverse effects on growth, longevity, and fecundity. These results indicate that TPS are compatible with microbial larvicides and can be used jointly in practical mosquito control programs. The larvicidal oil GB-1111, on the other hand, caused almost 100% mortality within 48 h after treatment in the laboratory even at the low dosage of 0.38 gallons/ac. The impact of oil in the field was less severe, as significant adverse impact on population density was noted at 1 gallon/ac when water temperatures were warmer, or 2 gallons/ac when water temperatures were cooler. Almost 100% mortality within 48 h was noted at 2 gallons/ac when water temperatures were warmer or at 4 gallons/ac when the water temperatures were cooler. These results indicate incompatibility between this larvicidal oil and TPS at the higher rates of the label range, and joint use of TPS and larvicidal oil at dosages of 1-2 gallons/ac or greater, depending on water temperatures, should be avoided. Journal of Vector Ecology 30 (1): Keyword Index: Tadpole shrimp, Triops newberryi, Bacillus thuringiensis ssp. israelensis, Bacillus sphaericus, larvicidal oil, toxicity, biological control, mosquito control. INTRODUCTION In the Coachella Valley of southern California, flood irrigation in agriculture creates widespread ephemeral habitats for floodwater mosquitoes, primarily Psorophora columbiae Dyar and Knab, as well as stagnant water mosquitoes if water stands for longer periods. Because of the variable amount of irrigation water, irregularity of flooding, and unpredictable duration of inundation, mosquitoes in these intermittently flooded and temporary habitats are difficult to control. Application of repetitive control measures, such as microbial and chemical larvicides or non-recycling pathogens, is costly in terms of inspection, monitoring, and treatment. Predacious organisms such as larvivorous mosquitofish are not suitable for biological control in these ephemeral habitats. Other predacious invertebrates, such as aquatic beetles, are not effective either, because they need some time to locate and colonize the habitats after each flooding. Rapid and synchronous development of mosquitoes in ephemeral habitats is usually finished before the predacious organisms invade, reach the right size, and grow to adequate population densities to yield good control. A successful predator of immature mosquitoes in these ephemeral habitats should have the attributes of being easily and permanently established with little or no augmentation, a high reproductive potential, and resistance to biological and physicochemical constraints imposed on it in intermittently flooded habitats. More importantly, the organisms should hatch and develop synchronously with their immature mosquito prey and be able to colonize and recycle in mosquito-infested habitats. Tadpole shrimp (TPS) are fresh water crustaceans adapted to temporary bodies of water in arid regions. The species, Triops longicaudatus Le Conte [designated as T. newberryi (Packard) by Sassaman et al. (1997)], has been the subject of considerable investigation because of its importance as a pest of seeded rice paddies (Grigarick et al. 1961) and as a potential control agent of weeds in transplanted rice fields (Takahashi 1977). The potential of TPS as a biocontrol agent for mosquitoes was indicated decades ago when Maffi (1962) noted that T. granarius (Lucas) decimated Anopheles gambiae Giles larvae in temporary breeding sources around huts in a village in Somalia. Adequate information on the potential use of TPS for mosquito control was not available until recent years when laboratory and field studies were conducted by Tietze and Mulla (1989, 1990, 1991), Fry and Mulla (1992), Fry et al. (1994), Fry-O Brien and Mulla (1996a, b), and Su and Mulla (2001, 2002a, b, c). Biological traits of TPS, such as fast nymphal growth, early maturation and high reproduction capacity (Fry-O Brien and

2 108 Journal of Vector Ecology June 2005 Mulla 1996a, Su and Mulla 2001), make it suitable for the control of some mosquitoes that share ephemeral habitats with TPS. The biological features of preconditioned TPS eggs with developed embryos are of special interest, because they are dessication-resistant (Carlisle 1968) and serve as the only viable stage in the absence of water. Once hydrated, these dormant eggs hatch quickly (Su and Mulla 2002a). As a bethedging survival strategy, these eggs undergo installment hatching, as do eggs of flooded water mosquitoes, upon each inundation (Fry-O Brien and Mulla 1996a, Su and Mulla 2002a). In a regularly irrigated or flooded habitat, a multigenerational assemblage of eggs known as an egg bank (Su and Mulla 2002b) is built up, where mature TPS resulted from each hydration add more eggs to the egg bank if the water stands long enough for maturation and oviposition. Triops newberryi is a native notostracan in flood-irrigated date gardens and a few other ephemeral aquatic habitats in the Coachella Valley and other arid regions of California as well as the southwest U.S.A. Information regarding optimal hatching conditions of TPS eggs (Su and Mulla 2002a) and an optimal rearing regimen to maximize growth, longevity, and fecundity were achieved in laboratory studies (Su and Mulla 2001). Horizontal and vertical egg distribution in soil was determined in several fields in the Valley, and hatchability of field eggs was studied in the laboratory (Su and Mulla 2002b). Field introductions of desiccation-resistant eggs and active shrimp into a date garden were carried out, shrimp populations were successfully established, and good mosquito control was achieved (Su and Mulla 2001c). Microbial larvicides such as Bacillus thuringiensis ssp. israelensis (B.t.i.) de Barjac and Bacillus sphaericus Neide, the IGR methoprene, larvicidal oils, and a few other materials, are currently used for controlling immature mosquitoes in the Coachella Valley and elsewhere in California and the U.S.A. For integrated mosquito control, TPS could be used in conjunction with other larvicides to provide extended periods of control in certain habitats. In order to investigate compatibility of TPS with the currently used mosquito larvicides and to protect TPS, we initiated laboratory and field studies on the toxicity and effects of B.t.i., B. sphaericus and a larvicidal oil on growth, survivorship, longevity, and maturity of Triops newberryi. MATERIALS AND METHODS Laboratory studies Microbial larvicides: Microbial mosquito larvicides such as B.t.i. and B. sphaericus are extensively used in practical mosquito control programs throughout the world. Do these microbial agents exert any adverse effects on the growth, survivorship, and especially fecundity of TPS if employed jointly for controlling mosquito larvae? To answer this important question, the following study was initiated. To acquire juvenile TPS, top soil containing TPS eggs was collected from the Aquatic and Vector Control Facility in the Agricultural Experimental Station of University of California at Riverside. Soil was processed and TPS eggs were hatched in the insectary by procedures described by Su and Mulla (2001). The juvenile shrimp with carapace length (CL) at carinal suture about 2 mm were transferred individually to individual rearing pans (30 x 19 x 5 cm), each containing 200 g soil (sandy loam devoid of TPS eggs) as substrate and 1000 ml of distilled water. Each rearing pan was supplemented with 0.5 g rabbit pellets (Brookhurst Mill, Riverside, CA) and 20 of 2 nd - 3 rd instar mosquito larvae every other day. Larvae surviving and not consumed were removed and discarded before a new batch of larvae was introduced to prevent adult emergence. Juvenile TPS in the pans were subjected to three treatments. The first treatment was control, where no microbial larvicide was added. The second treatment was B.t.i., where suspension of B.t.i. technical material (lot# W5, 7,000 ITU/mg, Abbott Laboratories, now Valent BioScience, Co., North Chicago, IL) was added at 3 ppm (3 ml of 0.1% suspension per pan) on a weekly basis until all TPS died. This dosage was 100 times of LC 90 (0.03 ppm) of susceptible 4 th instar larvae of Cx. quinquefasciatus Say (Su and Mulla 1999). The third treatment was B. sphaericus, where suspension of B. sphaericus TM (lot# W5, 2,000 ITU/ mg, provided by the same company as B.t.i. TM) was added at 8.5 ppm (8.5 ml of 0.1% suspension per pan) on a weekly basis, which was 100 times the LC 90 (0.085 ppm) of susceptible 4 th instar larvae of the same species (Su and Mulla 1999). The dosage and frequency of treatment were intentionally higher than in field applications. Fifteen replicates were made for each treatment and control. Tests were conducted at temperature C and photoperiod 16L: 8D. Carapace length as a growth parameter was measured and mortality was recorded every day until all shrimp died, and the size at death as indicated by CL was determined. After the death of all TPS, rearing water was drained, soil at the bottom of rearing pans containing fresh TPS eggs was dried at C, and eggs in the dried soil were counted according to the procedures of Su and Mulla (2001). The comparisons in growth, longevity, and fecundity were made among control and B.t.i. and B. sphaericus treatments by ANOVA test (Abacus Concepts, Inc. 1987). Larvicidal oil: Larvicidal petroleum oils such as GB (Witco Corp., Oildale, CA) still play an important role in mosquito control especially when late instar larvae and pupae are present in large numbers, as B.t.i. and B. sphaericus do not control these late stages. This test was conducted under the same laboratory conditions as the previous one. Live nymphal TPS (CL 2-3 mm) used in this test were collected from Pyramid Date Garden in the Coachella Valley, California. Five nymphal TPS with CL 2-3 mm were placed in 1,000 ml of distilled water in enamel pans (30 x 19 x 5 cm). Each rearing pan was supplemented with 0.5 g rabbit pellets as organic enrichment and 200 g of dried soil collected from the same date garden. The pans were treated (only one treatment) with larvicidal oil GB-1111 at dosages of 38, 76, 52, and 304 µl/ pan, equaling 0.75, 1.5, 3, and 6 gallons/ac, respectively, which covered the labeled dosage range of < 3-5 gallons/ac for controlling mosquito larvae and pupae in the field. The control was left untreated. Ten replicates were made and 50 active TPS in total were used per treatment and control. Mortality

3 Vol. 30, no. 1 Journal of Vector Ecology 109 of nymphal TPS was checked 24 h and 48 h after treatment. Since all shrimp died at 48 h at all treatment rates (see below), the test was concluded at 48 h after treatment. Field studies Microbial larvicides: Field studies on the impact of B.t.i. and B. sphaerisus on TPS were conducted in mesocosms or ponds at the Aquatic and Vector Control Research Facility in the Agricultural Experimental Station of University of California at Riverside, California. In each test of B.t.i. or B. sphaericus, a series of 12 bare ground dirt ponds were used, each with 27 m 2 (3.7 x 7.3 m) surface area and 30 cm water depth when flooded. These ponds in the past have been used for research on the biology and control of mosquitoes. Shrimp eggs occur in large numbers in the dry soil where an egg bank has been well-established. Egg hatch occurs soon after flooding when water temperature and photoperiod conditions are optimal. The B.t.i. test was initiated on July 28, A series of 12 ponds located in a sunlit open area were used. The pond bottom was dried completely prior to flooding by using water from a reservoir. Water depth was approximately 30 cm (about 8,100 liters per pond) and water levels were kept relatively constant by float valves on the water lines feeding each pond. In order to boost nutrient level in the pond water, the ponds were enriched before flooding with rabbit pellets at 2,000 g/ pond. A minimum-maximum thermometer was placed in one of the ponds for measuring water temperature. Tadpole shrimp hatched on flooding were sampled on day 4 after flooding, before B.t.i. treatment, and on days 7, 14, and 21 after treatment. For B.t.i. treatment, VectoBac G (200 ITU/mg, lot# N8, Valent BioSciences, North Chicago, IL) was applied at 10 and 20 lb/ac (label rate: lb/ac), equaling 30 and 60 g/pond, respectively. Four replicates were made for each treatment rate and control. Granules of B.t.i. were broadcast over the water surface as evenly as possible. Only a single treatment was made. For sampling TPS, we used a long-handled D net (30 x 25 cm) to sweep the bottom along each long side of the rectangular pond, with two samples taken from each pond. Collected TPS along with any debris, sand, and soil were transferred to an enamel tray (40 x 24 x 6 cm) containing 1 liter of water from the pond. The exact number of TPS in the tray was counted if the number was less than 100 or estimated by counting approximately half if the number was greater than 100. Up to 20 shrimp, depending on number of shrimp available in the tray, were randomly picked and preserved in 75% alcohol for further processing in the laboratory. Before laboratory processing, preserved shrimp were kept in a refrigerator (5-7 C) and samples were processed on the same day of collection. Size of the shrimp was estimated by measuring carapace length at carinal suture (Su and Mulla 2001), and the percentage of gravids was determined by examining modified appendages of the 11 th pair of limbs, i.e. the ovisacs, where reddish, shining mature eggs were present in gravid individuals. Average population density (no./net) and size (CL mm) of the shrimp as well as their standard errors were calculated and comparison was made among treatments and control by using one-factor ANOVA at 0.05 level. Percentage of gravids was analyzed among treatment and control by Chi square test at 0.05 level (Abacus Concepts 1987). The B. sphaericus test was started on September 5, 2003, when another series of 12 dirt bottom ponds were used. Half of those ponds were located in a semi-shaded area. Pond dimensions were the same as those used in the B.t.i. test. In order to enhance dryness of the soil at the pond bottom and egg hatch after flooding, the top 1 inch of the bottom was raked 3 days before flooding, because it has been shown that eggs in drier soil exhibited a higher hatching rate (Fry and Mulla 1992). Ponds were enriched and flooded in the same way as in the B.t.i. test. A minimum-maximum thermometer was placed in one of the ponds to monitor water temperature throughout the test period. On day 7 after flooding, TPS were sampled in the same way as in the B.t.i. test, then ponds were randomly assigned as control, treatment at 10 lb/ac, and treatment at 20 lb/ac (label rate: 5-20 lb/ac) of VectoLex CG (50 ITU/mg, lot# N8, Valent BioSciences, North Chicago, IL), each with 4 replicates. Only a single treatment Figure 1. Growth of tadpole shrimp (carapace length at carinal suture) treated weekly for 5 weeks with 100 times LC 90 for Cx. quinquefasciatus of two microbial larvicides in the laboratory.

4 110 Journal of Vector Ecology June 2005 was made. Tadpole shrimp were sampled on days 7, 14, and 21 after treatment. Samples were processed and data were analyzed in same way as in the B.t.i. test. Larvicidal oil: Two tests were conducted to evaluate toxicity of larvicidal oil (GB-1111 ) on TPS under different water temperature regimens. The first test was initiated on October 3, 2003, when the same series of ponds as in B.t.i. test was enriched and flooded in the same way as in previous tests. On day 7 after flooding, tadpole shrimp were sampled and counted in the same way as in previous tests, then the ponds were assigned randomly as control and treatments using 2 gallons/ac (50 ml/pond) and 4 gallons/ac (100 ml/pond) of larvicidal oil (label rates: < 3-5 gallons/ac). During oil application, a squeeze bottle was used as an applicator to generate a jet spray. Oil was applied to the water surface by walking along the edges of the ponds to evenly distribute the oil film and thoroughly cover the water surface, which was important for good efficacy in mosquito control. Each regimen of control and treatments was replicated 4 times. On days 2 and 5 after treatment, TPS were sampled, processed, and data were analyzed in the same way as in previous tests. The second test was initiated on September 27, 2004, when the same series of ponds as in B. sphaericus test was used. On day 7 after flooding, tadpole shrimp were sampled 24 h 48 h Figure 3. Mortality of tadpole shrimp exposed to various dosages of larvicidal oil (GB-1111) in laboratory. Significant difference in mortality was indicated between the control and treatments by a Chi square test at 0.05 level. and counted in the same way as in previous tests, then the ponds were assigned randomly as control and treatments using lower dosages of larvicidal oil: 1 gallon/ac (25 ml/pond) and 2 gallons/ac (50 ml/pond). Oil was applied in the same way as in previous test. Each regimen of control and treatments was replicated 4 times. On days 3 and 7 after treatment, TPS were sampled, processed, and data were analyzed in the same way as in previous tests. RESULTS Laboratory studies Microbial larvicides: Tadpole shrimp were treated with technical material of B.t.i. and B. sphaericus at 100 times the LC 90 of Cx. quinquefasciatus on a weekly basis until all shrimp died (about 5 weeks). In these experiments, the shrimp growth profile indicated by CL, longevity (days post-hydration) and fecundity (eggs/tps), did not show any significant negative impact by the treatments of either B.t.i. or B. sphaericus (P > 0.05 by ANOVA) (Figures 1 and 2). Larvicidal oil: After being exposed to larvicidal oil administered at various dosages ranging from 0.38 to 6 gallons/ac, the cumulative mortality of nymphal TPS was % at 24h and 100% at 48h after treatment (Figure 3). The cumulative mortality in control (no oil treatment) was low, 4.4% and 6.6% at 24 h and 48 h after treatment, respectively. The oil treatment killed live TPS before they matured and could lay their eggs. Figure 2. Growth, longevity, and fecundity of tadpole shrimp treated weekly for 5 weeks with 100 times LC 90 for Cx. quinquefasciatus of two microbial larvicides in the laboratory. No significant difference in each parameter was indicated among control and treatments by one-factor ANOVA at 0.05 level. Field studies Microbial larvicides: In the B.t.i. test using a single treatment in the ponds, TPS density was fairly low (< 10 shrimp/net) on treatment day (4 days after flooding), then increased and peaked on day 7 after treatment, fluctuating within shrimp/net. The overall density declined naturally and progressively thereafter until day 21 after treatment or day 25 after flooding, when the test was terminated. Throughout the test period, no significant

5 Vol. 30, no. 1 Journal of Vector Ecology 111 Figure 4. Effect of B.t.i. single treatment on growth and fecundity of field tadpole shrimp populations. For tadpole shrimp density and size, no significant differences were indicated among control and treatments by one-factor ANOVA at 0.05 level. For percentage of gravids, as well, no significant differences were noted by a Chi square test at 0.05 level. differences in shrimp density were noted among control and treatments (P > 0.05 by ANOVA) (Figure 4). The average shrimp size indicated by CL increased gradually from approximately 3-4 mm on treatment day, to 5-6 mm on days 7 and 14, and to 6-7 mm on day 21 after treatment. As in population density, no significant differences in average shrimp size were noted among control and treatments (P > 0.05 by ANOVA) (Figure 4). The percentage of gravid shrimp, a parameter used to evaluate fecundity of shrimp, was not significantly different among control and treatments (P > 0.05 by Chi square test). Percentage of gravid shrimp increased gradually from day 0 (treatment day) to day 21 after treatment, ranging from 30% to 60% (Figure 4). During the test period, water temperatures ranged from 21.5 to 24.5 C for minimum and from 25.7 to 28.3 C for maximum. In the B. sphaericus test, the overall shrimp density was much higher than that in B.t.i. test, assumingly, because bottom soil was raked up and further dried prior to flooding. On treatment day (day 7 after flooding), average shrimp density was greater than 200 shrimp/net in every pond. On days 7, 14, and 21 after treatment, average shrimp density declined progressively in treatments and control to around 50 shrimp/ net, when the test was terminated (Figure 5). As in B.t.i. test, the B. sphaericus treatment did not exert any significant impact on population density of the shrimp (P > 0.05 by ANOVA) (Figure 5). Shrimp did not grow well because of high density and CL remained approximately 5 mm throughout the test period. However, there were no significant differences among control and treatments (P > 0.05 by ANOVA). Fecundity as indicated by percentage gravid shrimp was also relatively lower (25-45%) than that in the B.t.i. test because of smaller size of the TPS. As in previous comparisons, no significant differences in gravid percentage were noted among control and treatments by Chi square test (P > 0.05) (Figure 5). During the test period, water temperatures in this test fluctuated from 19.5 to 22.2 C for minimum, and from 23.7 to 25.3 C for maximum. Larvicidal oil: In the first test of larvicidal oil GB-1111, average initial shrimp density was moderately high (70-80 shrimp/net), on treatment day, or day 7 after flooding. A low dosage of larvicidal oil (2 gallons/ac), significantly reduced population density on days 2 and 5 after treatment, while the high dosage (4 gallons/ac) almost completely eliminated TPS on day 2 after treatment (P < 0.05 by ANOVA) (Figure 6A). The shrimp that survived oil treatment at low dosage grew at a similar rate as did controls, as indicated by CL measurements (P > 0.05 by ANOVA). This was also somewhat true for fecundity as indicated by percentage of gravid shrimp (P > 0.05 Chi Square test) (Figure 6A). This test was conducted when the water temperatures ranged from 17.5 to 20.1 C for minimum and from 21.7 to 23.5 C for maximum. In the second test, average initial shrimp density was high ( shrimp/net) on treatment day. A low dosage of larvicidal oil (1 gallon/ac), significantly reduced population density on days 3 and 7 after treatment, while the high dosage (2 gallons/ac) almost completely eliminated TPS on days 3 and 7 after treatment (P < 0.05 by ANOVA) (Figure 6B). The shrimp that survived oil treatment grew to a larger size, as indicated by CL measurements (P < 0.05 by ANOVA), and showed higher percentage of gravids (P < 0.05 by Chi Square test) (Figure 6B). This test was conducted when the water temperatures were warmer than those in the first test, ranging from 22.5 to 23.8 C for minimum and from 25.7 to 28.3 C for maximum. DISCUSSION The microbial mosquito larvicides B.t.i. and B. sphaericus have been well-recognized for their efficacy, target specificity, and environmental compatibility and have served as the most

6 112 Journal of Vector Ecology June 2005 Figure 5. Effect of Bacillus sphaericus single treatment on growth and fecundity of field tadpole shrimp populations. For tadpole shrimp density and size, no significant differences were indicated among control and treatments by one-factor ANOVA at 0.05 level. For percentage of gravids as well, no significant differences were noted by a Chi square test at 0.05 level. important and commonly used larvicides against mosquitoes since the mid-1980 s for B.t.i. and mid-1990 s for B. sphaericus. Pro-toxins in the formulated materials are ingested by mosquito larvae and are activated by the specific enzymes secreted in the midgut of mosquito larvae. Activated toxins induce the pathological disruption of the gut epithelium, resulting in death of the larvae. High levels of safety for both these larvicides to immature stages of a wide varieties of aquatic invertebrates has been well-documented, even though some groups such as chironomid midges may be impacted by B.t.i. at high dosages. Tadpole shrimp, sharing fresh water habitats with some mosquito species, are omnivorous. Food ingested by TPS consists of algae, bacteria, protoctista, rotifers, bits of detritus, and micro- as well as macroinvertebrates available in ephemeral aquatic habitats. Movements of the legs of the shrimp serve as a food-gathering mechanism. Food materials from the substrate are strained non-selectively from the water or scraped from the substrate by the setaceous appendages (especially by the basal endites), concentrated and agglutinated in a median ventral groove running most of the length of the body. This stream of food moves forward, activated chiefly by the gnathobases and is further agglutinated at the anterior end by a sticky secretion produced by labral glands. Mastication occurs outside the digestive tract in an atrium formed by the overhanging labrum. Tadpole shrimp probably feed continuously, but all the food reaching the mouth is not necessarily ingested, the excess being sloughed off (Smith 2001). Considering this feeding behavior and food constituents, TPS did have the opportunity, as did mosquito larvae, to ingest particles from B.t.i. and B. sphaericus preparations in the treated water. In these studies, we have found that B.t.i. and B. sphaericus are safe to the TPS even at 100 times the concentration effective against Cx. quinquefasciatus in weekly treatments for about 5 weeks in the laboratory and at the higher level of recommended dosages in the field. The results clearly indicate that there is a good compatibility between microbial mosquito larvicides and the TPS in mosquito control, which could be used jointly in operational mosquito control programs. It will be desirable to employ reduced dosages of microbial larvicides in the presence of the TPS to achieve an extended period of control (Fry-O Brien and Mulla 1996b). Petroleum oils as surface-active larvicides have been used for mosquito control for more than a century in the United States. Golden Bear 1111, a light viscosity, highly refined petroleum, contains 98.7% of active ingredients consisting of aliphatic petroleum hydrocarbons and 1.3% of inert material. This product spreads quickly and evenly over the water surface, kills larvae, pupae and emerging adults by physical suffocation, and may repel ovipositing mosquitoes (Beehler and Mulla 1996). It is applied primarily in swamps, floodwater areas, and other stagnant waters where mosquitoes develop. It should be applied at normal rates of 3 gallons/ac or less. Under special circumstances, such as extremely dense vegetation or water with high organic content, rates to a maximum of 5 gallons/ac may be used. No emergence of resistance in mosquitoes to petroleum oils has been reported to date. However, use of oils has been limited recently by adverse effects on some non-target macro-invertebrates. Mulla et al. (1979) concluded that petroleum hydrocarbons used in mosquito control have little effect on benthic fauna. Detrimental effects were shown to be more pronounced in organisms which surface for oxygen, and these groups usually recovered one to three weeks after application. Tadpole shrimp, called branchiopods (gill foot), have pairs of leaf-like appendages used for swimming, food collection, and respiration (consumption of dissolved oxygen). The shrimp

7 Vol. 30, no. 1 Journal of Vector Ecology 113 do spend substantial time at the bottom of their habitats when water temperature is low, but they swim mostly on their backs with gill-like appendages close to the water surface when the water is warm and dissolved oxygen in the water is low. It is reasonable to assume that the shrimp are more vulnerable to larvicidal oil treatment when dissolved oxygen in water is lowered by warm water temperatures, organic pollution, etc. According to mode of action of larvicidal oil GB-1111 and respiration mechanism of the tadpole shrimp, it is understandable that the dosage of 2 gallons/ac of larvicidal oil significantly lowered TPS density, while 4 gallons/ac completely eliminated TPS when water temperatures were cooler as in the first test. This toxicity was more pronounced when water temperatures were warmer as in the second test, when dosages of 1 gallon/ac significantly lowered TPS density, while 2 gallons/ac almost completely eliminated TPS. It is thus advisable to use oil at the rates of 1 gallon/ac or less in ephemeral mosquito habitats with breeding of the TPS. Fortunately, in field operational programs, larvicidal oils are not often used in habitats supporting TPS and mosquitoes together. Acknowledgments The enthusiastic support and encouragement for these studies by the Board of Trustees and General Manager Donald E. Gomsi at Coachella Valley Mosquito and Vector Control District are gratefully acknowledged. We also appreciated valuable assistance of Field Supervisor James Saulnier by providing test materials. We are grateful to the assistance during field studies by Yongxing Jiang and Albert H. Lee at Department of Entomology, University of California at Riverside. REFERENCES CITED Figure 6A. Effect of mosquito larvicidal oil on growth and fecundity of field tadpole shrimp populations. For tadpole shrimp density, different letters on the same sampling day represent significant differences among control and treatments by one-factor ANOVA at 0.05 level. No differences were noted in tadpole shrimp size among control and treatments by one-factor ANOVA at 0.05 level. Different letters on the same sampling day for percentage of gravids indicate significant diferences by a Chi square test at 0.05 level. Abacus Concepts, Inc StatView + Graphics. Abacus Concepts, Inc., Berkeley, CA, 234 pp. Beehler J.W. and M.S. Mulla Larvicidal oils modify the oviposition behavior of Culex mosquitoes. J. Vector Ecol. 21: Carlisle, D. B Triops (Entomostraca) eggs killed only by boiling. Science 161: 279. Fry, L.L. and M.S. Mulla Effects of drying period and soil moisture on egg hatch of the tadpole shrimp (Notostraca: Triopsidae). J. Econ. Entomol. 85: Fry, L.L., M.S. Mulla, and C.W. Adams Field introductions and establishment of tadpole shrimp, Triops longicaudatus (Notostraca: Triopsidae), a biological control agent of mosquitoes. Biol. Contr. 4: Fry-O Brien, L.L. and M.S. Mulla. 1996a. Optimal conditions for rearing the tadpole shrimp, Triops longicaudatus (Notostraca: Triopsidae), a biological control agent against mosquitoes. J. Am. Mosq. Contr. Assoc. 12 (Part I):

8 114 Journal of Vector Ecology June 2005 Figure 6B. Effect of mosquito larvicidal oil on growth and fecundity of field tadpole shrimp populations. For tadpole shrimp density and size, different letters on the same sampling day represent significant differences among control and treatments by one-factor ANOVA at 0.05 level. Different letters on the same sampling day for percentage of gravids indicate significant diferences by a Chi square test at 0.05 level. Fry-O Brien, L.L. and M.S. Mulla. 1996b. Effects of tadpole shrimp, Triops longicaudatus (Notostraca: Triopsidae), on the efficacy of the microbial control agent Bacillus thuringiensis var. israelensis in experimental microcosms. J. Am. Mosq. Contr. Assoc. 12: Grigarick, A.A., W.H. Lange, and D.C. Finfrock Control of the tadpole shrimp, Triops longicaudatus, in California rice field. J. Econ. Entomol. 54: Maffi, M Triops granaries (Lucas) (Crustacea) as a natural enemy of mosquito larvae. Nature 195: 722. Mulla, M.S., G. Majori, and A.A. Arata Impact of biological and chemical mosquito control agents on nontarget biota in aquatic ecosystems. Residue Rev. 71: Sassaman, C., M.A. Simovich, and M. Fugate Reproductive isolation and genetic differentiation in North American species of Triops (Crustacea: Branchiopoda: Nortostraca). Hydrobiologia 359: Smith, D.G Phyllopodous Branchiopoda. In: Pennak s Freshwater Invertebrates of the United States, Porifera to Crustacea. 4 th Edition, pp John Wiley & Sons, New York. Su, T. and M.S. Mulla Field evaluation of new waterdispersible granular formulations of Bacillus thuringiensis ssp. israelensis and Bacillus sphaericus against Culex mosquitoes in microcosms. J. Am. Mosq. Contr. Assoc. 15: Su, T. and M.S. Mulla Nutritional and edaphic factors affecting growth, longevity and fecundity of the tadpole shrimp Triops longicaudatus (Notostraca: Triopsidae), a potential biological control agent of immature mosquitoes. J. Vector Ecol. 26: Su, T. and M.S. Mulla. 2002a. Factors affecting egg hatch of the tadpole shrimp Triops longicaudatus (Notostraca: Triopsidae) Le Conte, a potential biological control agent of immature mosquitoes. Biol. Contr. 33: Su, T. and M.S. Mulla. 2002b. Spatial distribution and hatch of field eggs of the tadpole shrimp Triops longicaudatus Le Conte (Notostraca: Triopsidae), a biological control agent of immature mosquitoes. J. Vector Ecol. 27: Su, T. and M.S. Mulla. 2002c. Introduction and establishment of tadpole shrimp Triops newberryi (Notostraca: Triopsidae) in a date garden for biological control of immature mosquitoes in the Coachella Valley, southern California. J. Vector Ecol. 27: Takahashi, F Triops spp. (Notostraca: Triopsidae) for the biological control of weeds in rice paddies in Japan. Entomophaga 22: Tietze, N.S. and M.S. Mulla Prey-size selection by Triops longicaudatus (Notostraca: Triopsidae) feeding on immature stages of Culex quinquefasciatus. J. Am. Mosq. Contr. Assoc. 5: Tietze, N.S. and M.S. Mulla Influence of tadpole shrimp, Triops longicaudatus (Notostraca: Triopsidae), stocking rate on Culex tarsalis development in experimental field microcosms. J. Am. Mosq. Contr. Assoc. 6: Tietze, N.S. and M.S. Mulla Biological control of Culex mosquitoes (Diptera: Culicidae) by the tadpole shrimp, Triops longicaudatus (Notostraca: Triopsidae). J. Med. Entomol. 28: