Abstract California pastured cattle were treated with 250 ml of a 15% mixture of fatty acids (C8–C9–C10) or 125 ml of 2% geraniol in a mineral oil carrier to assess impacts on horn flies, Haematobia irritans (L.) (Diptera: Muscidae) over two summers. Horn flies were netted from cattle every 3–4 d for 2 wk before treatment, 2 wk during treatment (four treatments, with flies collected before each treatment), and 2 wk after treatments ceased. Blood meal weights were estimated by hemoglobin assay of excised abdomens. Other females were dissected to determine the number of active ovarioles and the stage of primary follicle development. Depending on year and herd, pretreatment males contained an average of 0.6–1.0 mg of blood, while females contained 1.7–2.7 mg. Pretreatment egg development (least developed oocytes were stage 1 and fully developed eggs were stage 5) averaged 3.7–4.3, and number of active ovarioles averaged 18.1 to 19.6/female. During treatment periods, significant reductions in blood weight were noted for females, but usually not for males, and females also often exhibited reduced mean oocyte stage and number of active ovarioles. Peaks in proportions of young nulliparous females (oocyte stages 1 or 2) were seen during some repellent application periods. This suggested older females had been killed or driven off from the local population by the treatments, and flies on cattle included more young flies that likely were recent arrivals. The repellent-oil mixture thus impacted blood feeding, reproductive fitness, and probably age structure in the field. horn fly, repellent, fecundity, fitness, sublethal effects The horn fly, Haematobia irritans (L.) (Diptera: Muscidae), is a serious blood-feeding pest of pastured cattle in Europe, North and South America, and parts of Asia and North Africa (Foil and Hogsette 1994, Oyarzun et al. 2008). Of several potential control methods, the most widely used are insecticides applied to adult cattle to reduce adult fly numbers, and this has caused high levels of resistance to evolve in many areas (Foil and Hogsette 1994, Oyarzun et al. 2008). Repellents are another management option, including very low toxicity or plant-derived compounds that have potential for use against a number of veterinary pests (Ellse and Wall 2014, George et al. 2014), including horn flies (Mullens et al. 2009, Lachance and Grange 2014, Zhu et al. 2014, Mullens et al. 2017). Repellents may come from a variety of sources and chemical groups and have been widely tested for protection against certain groups of medical or veterinary pests, particularly mosquitoes and ticks (Debboun et al. 2014). Generally, such tests on humans involve reducing biting rates in the lab or, less often, in the field. With veterinary pests, since animals cannot report biting, the main field criterion is visually to document reduced numbers of pests on test animals relative to controls or periods before treatment (e.g., Lachance and Grange 2014, Zhu et al. 2014, Mullens et al. 2017). In a landmark pair of studies, Jensen et al. (2004) showed interesting and repeatable differences in horn fly loads among individual cattle in the field, which could have been due to differences in the type or concentration of natural chemical volatiles the cattle produced. Birkett et al. (2004) followed that up by documenting chemicals in air immediately adjacent to those individual cattle. They then showed via electroantennograms (EAG) and wind tunnel experiments that certain chemicals elicited EAG responses and either positive or negative flight responses by horn flies in a wind tunnel. A small-scale field test by Birkett et al. (2004) to try to manipulate field fly numbers on cattle via chemical enhancement was inconclusive, but the case was made that natural products from cattle also had potential use as attractants or repellents. Besides influencing fly counts on animals, repellents may manifest their activity in less obvious ways, and mechanisms may remain obscure. For example, repellents may interfere with host location, orientation, or blood feeding (see Debboun et al. 2014), and reduced blood meal size might result in subsequent downstream impacts such as reduced survival or fecundity (Xue et al. 2007, Barnard and Xue 2009). If these more subtle aspects are considered at all, they are nearly always investigated via laboratory studies (Hao et al. 2008, Kumar et al. 2011, Sugiharto et al. 2016), rather than being documented in the field. The present study explores impacts of two promising repellents, fatty acids and geraniol, on horn flies, documenting repellent-related changes over time in blood feeding and reproductive parameters under actual field conditions. In the process, we supply field data that will be relevant to a better understanding of horn fly population biology. Materials and Methods Study Site Details of the site and applications, together with weather patterns and fly population trends, are provided in Mullens et al. (2017). Briefly, we used cattle herds employed in teaching programs at California Polytechnic State University, Pomona, California (34°03′N, 78°02′W). The animals were handled often by students, so cattle were unusually tame, which was an advantage for spraying them and collecting flies. Herds of 15–35 mature cattle, plus 8–15 calves, were held on irrigated pastures. For this part of the study, we used one group in summer and fall 2010 (treated periodically as below). In summer and fall of 2011, we used two groups; one was treated periodically as below and a second group was held 1.6 km away as an entirely untreated herd. Treatments The first repellent tested was a 15% fatty acid (5% of C8, 5% of C9, and 5% of C10) mixture in mineral oil (FA) (Emery Oleochemicals, LLC; Cincinatti, OH). The second repellent was a 2% geraniol mixture in mineral oil (GER) (FASST Products, LLC; Rockville Center, NY). Both were applied as a spray directly to cattle using a calibrated backpack sprayer to apply 250 ml (fatty acids) or 125 ml (geraniol) per adult animal, or half that amount per calf. Untreated control animals (n = 4) were identified by ear tag number on each date and separated from the rest of the herd for the brief treatment periods. Those cows were not directly treated with the repellents, although it is likely they periodically touched other treated cattle later when they rejoined the treated cattle in the open pasture. This preserved some flies in the herd. Cattle to be treated were moved into a holding pen and were sprayed individually as completely as possible (except for the face), including belly and legs. Once an animal was treated, it was released back into the main pasture. Tests consisted of a 2-wk pretreatment period, a 2-wk treatment period, and a 2-wk postreatment period for each repellent and in each year (2010 and 2011). Horn flies were collected using a sweep net from 3 to 6 adult cows in the herd and the flies were pooled. This was done every 3–4 d for 2 wk before treatment (four collections). A fifth pretreatment collection was made immediately before the cattle were first treated. During the treatment period, four repellent treatments were made at 3–4 d intervals, with flies collected as above immediately prior to each spraying and 3–4 d after the last spray. By 3–4 d posttreatment, fly numbers had regained at least moderate numbers on the treated herds (Mullens et al. 2017), and flies were collected for assay from both sprayed and unsprayed cattle. In 2010, additional fly collections were made daily for 4 d following the last treatment day using the FA. The same cattle were used for the consecutive GER and FA repellent treatments, but a full month passed between the FA and GER applications, so any repellent residual should have been negligible. The estimated duration of significant suppression of horn fly numbers following either repellent was only 1–3 d (Mullens et al. 2017). Fly Collections Collected flies from multiple cows on a given herd and date were anesthetized immediately after collection using CO2, poured into a labeled, capped 50 ml plastic tube using a funnel, placed on dry ice in a cooler, and transported to the laboratory. In the laboratory, each tube received a small moistened piece of tissue, the cap was replaced tightly and sealed using Parafilm (Dixie/Marathon; Greenwich, CT), and the tube was snugly wrapped in aluminum foil. The tubes for each date were placed into a labeled ziplock plastic bag with another moist paper towel and this was placed into a freezer at −20°C. This prevented drying out of the flies and kept them in good condition when they were removed from the freezer for subsequent sexing, assay, and dissection. Blood Meal Calibration To determine blood meal size, we used a hemoglobin assay (Briegel et al. 1979). Briefly, Drabkin’s reagent (Sigma-Aldrich #D5941-6VL; St. Louis, MO) was added to 1 liter of deionized water in a dark bottle, and 0.5 ml of 30% Brij L23 solution (Sigma-Aldrich #B4184) was added to create the Drabkin’s solution. Initially, we added known volumes (0.1–2.1 µl, approximating blood weights of 0.1–2.1 mg) of defibrinated cattle blood (Hemostat Laboratories; Dixon, CA) to 500 µl volumes of Drabkin’s solution in a 2 ml microcentrifuge tube. Cattle blood is only slightly heavier than water (specific gravity 1.046–1.05). This was homogenized using a vortex mixer (Mini Vortexer, Fisher Scientific; Pittsburg, PA) for 10 s and a 200 µl amount then was placed into an ultramicro cuvette (BrandTech Scientific Products #759220; Essex, Connecticut). A Spectronic 20 spectrophotometer (Spectronic Instruments, Inc.; Fitchburg, WI) was calibrated using a microcuvette with Drabkin’s solution only (recalibrated every five samples) for absorbance at 540 nm wavelength. Absorbance of samples containing blood was regressed against the known blood weight. Once the relationship between blood weight and absorbance was validated using blood alone, calibration was done using blood-fed horn flies. Live horn fly pupae were obtained from the USDA-ARS Knipling-Bushland Livestock Insects Laboratory in Kerrville, Texas. One-day-old flies (never fed blood) were weighed individually to the nearest 0.001 mg (Sartorius MP2 Microbalance, Sartorius AG; Goettingen, Germany) to get a prefeeding weight. Those individuals then were allowed to take a defibrinated cow blood meal from a saturated pad for 30 min. and reweighed, so blood weight ingested could be determined. There were eight unfed controls. Most flies were frozen immediately after they fed and were weighed (0 h), although some were held for either 3 h or 6 h before freezing. We initially tested several factors: 1) blood weight ingested, 2) fly sex, 3) time elapsed since blood feeding and before freezing (0, 3, and 6 h), and 4) whether the gut was dissected from the abdomen or whether the abdomen and its gut contents were tested whole. For most flies, whole abdomens were removed and processed. The entire abdomen (or dissected gut) of individual labeled flies was removed and placed into a 2 ml microcentrifuge tube with 500 µl of Drabkin’s solution and a series of ceramic and zirconium disruptor beads (Kraemer Industries, Inc.; Piscataway, NJ). Tubes were then placed into a Hammer Genie tissue disruptor (Scientific Industries, Inc.; Bohemia, NY) and run for 30 s, then off for 30 s (to avoid friction heating). That on–off process was repeated 3 times, and the tubes then were centrifuged for 2 min at 10,000 rpm. A 200 µl aliquot of supernatant was removed and tested for absorbance at 540 nm as above. The relationship between known blood weight within the fly abdomen and absorbance was examined using regression as below. Processing Field-Collected Flies For each field collection date and herd, abdomens of 30 male and 30 female horn flies were assessed for blood weight via absorbance, which was then converted to an estimate of amount of blood ingested. Another group of 30 females from each collection was dissected under a dissecting microscope in physiological saline (0.7% NaCl). Abdomens were carefully opened, exposing the ovaries and associated structures on a glass microscope slide (Fig. 1). The stage of the primary follicles was noted using a scale of 1 (immature oocytes without yolk) to 5 (mature egg) (Krafsur and Ernst 1986), as slightly modified from Schmidt (1972) and Kuramochi and Nishijima (1984). Due to its subjective nature (potential for different individuals to grade parity differently) and difficulty in assessing parity in gravid flies (Krafsur and Ernst 1986), we did not attempt to determine parity. The ovaries then were carefully teased apart using fine minuten pin probes and the number of active ovarioles (Schmidt 1972) was recorded. Fig. 1. View largeDownload slide Dissected abdominal contents of wild-captured horn fly female at 25×, showing blood in gut and two egg cohorts—one ready to lay (stage 5) and the subsequent one with yolk occupying about 1/3 of the oocyte length (stage 3). Fig. 1. View largeDownload slide Dissected abdominal contents of wild-captured horn fly female at 25×, showing blood in gut and two egg cohorts—one ready to lay (stage 5) and the subsequent one with yolk occupying about 1/3 of the oocyte length (stage 3). Statistical Analysis Statistical analyses were done using Minitab version 14 (Minitab, Inc.; State College, PA). Spectrophotometer absorbance at 540 µm wavelength first was regressed against the known amount of blood added to the Drabkins solution. An initial multivariate analysis of impacts on absorbance was done using a general linear model. Independent variables were blood weight, fly sex, time since feeding, and gut dissected versus whole abdomen. Ultimately, we then regressed absorbance values from individual, processed horn fly abdomens against the known blood weights the flies had ingested. This was used to convert absorbance values to estimated blood weights of field-collected flies. Analyses were constructed around the three 2-wk periods for each repellent tested in 2011 and 2012: 1) before repellent treatment, 2) during repellent treatment, and 3) after repellent treatment. Analysis of variance was used to test whether estimated blood weights (for each fly sex) or female fecundity (number of active ovarioles) differed among the 2-wk periods for each repellent and year. Tukey’s HSD test was used to separate the 2-wk period means if the overall F value was significant. For oocyte stage, we used a Kruskal–Wallis test to determine whether the 2-wk periods differed. If so, pairwise Mann–Whitney U tests were used to separate the period values from each other for a given repellent and herd. Finally, the proportion of young females in each group of 30 flies (those with stage 1 or 2 oocytes) on each date was transformed using arcsin of the square root and a final analysis of variance was done to test if it might vary among treatments or treatment periods. The treatments were FA, GER, or untreated. Year was nested within treatment for the FA and GER, while the control had the appropriate periods to match the FA and GER periods. The treatment periods were before, during, or after treatment. Tukey’s HSD test again was used to separate means if the overall F value was significant. Results Using blood alone with Drabkin’s solution, the fit between absorbance at 540 µm and blood volume was nearly perfect for 10 volumes within the range of 0.1–2.1 µl with an r2 value of 0.954 (t = 14.46; df = 1, 9; P < 0.001). But, the more appropriate test was the ability of the assay to detect blood in a fly, which had other fly tissue in addition to the cattle blood. An initial multivariate analysis of variance (MANOVA), using flies that ingested a known blood weight, indicated blood weight was responsible for nearly all the accountable variance (adjusted r2 = 58.6%) (F = 109.98; df = 1, 91; P < 0.001). Whether the gut was dissected from the abdomen also was significant (F = 17.31; df = 1, 91; P < 0.001) and dissecting out the gut gave better readings compared with processing the entire abdomen with the gut and blood inside. However, dissecting the guts from flies entailed both a large increase in time required and risk of damage to the gut and loss of blood, which was irrecoverable. So, the decision was made to process abdomens intact, allowing processing of far larger numbers of flies. In the MANOVA, fly sex was not significant once differences in ingested blood weight were accounted for (F = 2.41; df = 1, 91; P = 0.124). Time since feeding (F = 10.05; df = 2, 91; P < 0.001) was significant. The longest time since blood feeding (6 h) had about 30% lower absorbance, while 0 and 3 h did not differ. This was attributed to digestion of the blood over 6 h. Flies on cattle take blood meals an average of 24–38 times per day (i.e., an average of once every 37–60 min) (Harris et al. 1974), so the elapsed time since experimental feeding was not considered to be a major concern. Ultimately, we calibrated intact abdomens of flies irrespective of fly sex, using flies that had freshly fed (0 h). The fit between absorbance and known blood weight in the abdomen (r2 = 0.475) was not as good as pure blood in reagent (r2 = 0.954) but still was highly significant (F = 63.41; df = 1, 68; P < 0.001). The fitted equation used to convert absorbance to predicted blood weight was blood weight = (absorbance − 0.0375)/0.0492. The means (SE) for blood ingested for each sex, the mean oocyte stage, and mean number of active ovarioles over time for each herd are shown in Fig. 2 (2010 treated herd), Fig. 3 (2011 treated herd), and Fig. 4 (2011 separated control herd). Across the entire period, males contained far less blood than females did for each herd and year. The sex differences (mean ± SE) were highly significant for 900 flies tested from treated cattle in 2010 (males 0.72 ± 0.03 mg; females 1.92 ± 0.04 mg) (F = 527.16; df = 1, 898; P < 0.001), for 720 flies from treated cattle in 2011 (males 0.86 ± 0.03 mg; females 2.11 ± 006 mg) (F = 376.44; df = 1, 718; P < 0.001), and for 719 flies from control cattle in 2011 (males 0.53 ± 0.02 mg; females 1.78 ± 0.05 mg) (F = 581.26; df = 1, 717; P < 0.001). Even within a fly sex, there was quite a bit of variation in blood weights among dates, but males seldom ingested an average of more than 1.3 mg, and female averages were nearly always less than 3 mg. Fig. 2. View largeDownload slide Estimated blood ingestion (A), oocyte development stage (B), and number of active ovarioles (C) of horn flies netted from pastured cattle at Pomona, California in 2010. Repellent treatment periods with a 15% fatty acid mixture (FA) or 2% geraniol (GER) shown. On treatment days, flies were always collected before a treatment occurred. N = 30 flies per sex per date. Fig. 2. View largeDownload slide Estimated blood ingestion (A), oocyte development stage (B), and number of active ovarioles (C) of horn flies netted from pastured cattle at Pomona, California in 2010. Repellent treatment periods with a 15% fatty acid mixture (FA) or 2% geraniol (GER) shown. On treatment days, flies were always collected before a treatment occurred. N = 30 flies per sex per date. Fig. 3. View largeDownload slide Estimated blood ingestion (A), oocyte development stage (B), and number of active ovarioles (C) of horn flies netted from pastured cattle at Pomona, California in 2011. Repellent treatment periods with a 15% fatty acid mixture (FA) or 2% geraniol (GER) shown. On treatment days, flies were always collected before a treatment occurred. N = 30 flies per sex per date. Fig. 3. View largeDownload slide Estimated blood ingestion (A), oocyte development stage (B), and number of active ovarioles (C) of horn flies netted from pastured cattle at Pomona, California in 2011. Repellent treatment periods with a 15% fatty acid mixture (FA) or 2% geraniol (GER) shown. On treatment days, flies were always collected before a treatment occurred. N = 30 flies per sex per date. Fig. 4. View largeDownload slide Estimated blood ingestion (A), oocyte development stage (B), and number of active ovarioles (C) of horn flies netted from untreated pastured cattle at Pomona, California in 2011. N = 30 flies per sex per date. Fig. 4. View largeDownload slide Estimated blood ingestion (A), oocyte development stage (B), and number of active ovarioles (C) of horn flies netted from untreated pastured cattle at Pomona, California in 2011. N = 30 flies per sex per date. Oocyte stages indicated most females had moderately to well developed primary oocytes, usually with mean scores between 3.3 and 4.6. Numbers of active ovarioles also varied among dates and varied between about 16.3 and 21.7. Of primary interest was whether blood ingestion, oocyte stage or number of active ovarioles varied between pretreatment, treatment, or posttreatment periods in each repellent trial. Those comparisons are shown in Table 1. In 2010, male blood weights did not differ among treatment periods for either FA or GER. This was also true for male blood weights during the FA treatment in 2011 and for the separated control herd during the GER period in 2011. Male blood weights were lower during the GER treatment in 2011 and also for the control herd during the period the treatment herd was being sprayed with the FA. Table 1. Responses of horn fly adults (mean ± SE) to cattle repellent treatments in pastures in southern California Response Time interval Herd (yr) Trt Variable P value Pretrt Trt Postrt Trt (2010) FA Blooda (male) 0.151 0.65 (0.05)a 0.77 (0.04)a 0.73 (0.06)a FA Blood (female) <0.001 2.04 (0.08)a 1.69 (0.06)b 2.29 (0.99)a FA Oocyte stageb <0.001 4.32 (0.07)a 3.46 (0.09)b 4.27 (0.09)a FA Active ovarioles <0.001 19.61 (0.28)a 17.92 (0.22)b 19.93 (0.26)a GER Blood (male) 0.893 0.71 (0.05)a 0.74 (0.05)a 0.70 (0.05)a GER Blood (female) <0.001 2.73 (0.11)a 1.92 (0.09)b 2.23 (0.11)b GER Oocyte stage 0.076 3.72 (0.11)a 3.81 (0.12)a 4.08 (0.12)a GER Active ovarioles 0.217 19.63(0.25)a 19.68 (0.25)a 20.29 (0.34)a Trt (2011) FA Blooda (male) 0.063 0.95 (0.05)a 0.83 (0.05)a 0.77 (0.06)a FA Blood (female) <0.001 2.30 (0.10)a 1.69 (0.07)b 2.35 (0.11)a FA Oocyte stage <0.001 3.95 (0.09)b 3.79 (0.11)b 4.47 (0.08)a FA Active ovarioles 0.002 18.19 (0.30)b 17.96 (0.35)b 19.49 (0.31)a GER Blood (male) <0.001 1.00 (0.05)a 0.63 (9.05)b 0.86 (0.06)a GER Blood (female) <0.001 2.24 (0.14)a 1.38 (0.08)b 2.11 (0.09)a GER Oocyte stage <0.001 3.98 (0.11)a 3.38 (0.14)b 4.19 (0.10)a GER Active ovarioles 0.068 19.58 (0.26)a 18.69 (0.29)a 19.21 (0.31)a Cont (2011) None1c Blooda (male) 0.009 0.57 (0.04)a 0.42 (0.03)b 0.59 (0.05)a None1 Blood (female) <0.001 1.99 (0.07)a 1.49 (0.07)b 1.82 (0.09)a None1 Oocyte stage 0.508 4.04 (0.08)a 4.11 (0.37)a 4.20 (0.09)a None1 Active ovarioles 0.014 18.86 (0.20)b 18.93 (0.23)b 19.80 (0.28)a None2 Blood (male) 0.670 0.60 (0.04)a 0.64 (0.05)a 0.65 (0.06)a None2 Blood (female) 0.139 1.73 (0.06)a 1.73 (0.08)a 1.94 (0.11)a None2 Oocyte stage 0.064 3.95 (0.10)a 3.51 (0.13)a 3.78 (0.14)a None2 Active ovarioles 0.075 18.09 (0.21)a 18.78 (0.22)a 18.56 (0.28)a Response Time interval Herd (yr) Trt Variable P value Pretrt Trt Postrt Trt (2010) FA Blooda (male) 0.151 0.65 (0.05)a 0.77 (0.04)a 0.73 (0.06)a FA Blood (female) <0.001 2.04 (0.08)a 1.69 (0.06)b 2.29 (0.99)a FA Oocyte stageb <0.001 4.32 (0.07)a 3.46 (0.09)b 4.27 (0.09)a FA Active ovarioles <0.001 19.61 (0.28)a 17.92 (0.22)b 19.93 (0.26)a GER Blood (male) 0.893 0.71 (0.05)a 0.74 (0.05)a 0.70 (0.05)a GER Blood (female) <0.001 2.73 (0.11)a 1.92 (0.09)b 2.23 (0.11)b GER Oocyte stage 0.076 3.72 (0.11)a 3.81 (0.12)a 4.08 (0.12)a GER Active ovarioles 0.217 19.63(0.25)a 19.68 (0.25)a 20.29 (0.34)a Trt (2011) FA Blooda (male) 0.063 0.95 (0.05)a 0.83 (0.05)a 0.77 (0.06)a FA Blood (female) <0.001 2.30 (0.10)a 1.69 (0.07)b 2.35 (0.11)a FA Oocyte stage <0.001 3.95 (0.09)b 3.79 (0.11)b 4.47 (0.08)a FA Active ovarioles 0.002 18.19 (0.30)b 17.96 (0.35)b 19.49 (0.31)a GER Blood (male) <0.001 1.00 (0.05)a 0.63 (9.05)b 0.86 (0.06)a GER Blood (female) <0.001 2.24 (0.14)a 1.38 (0.08)b 2.11 (0.09)a GER Oocyte stage <0.001 3.98 (0.11)a 3.38 (0.14)b 4.19 (0.10)a GER Active ovarioles 0.068 19.58 (0.26)a 18.69 (0.29)a 19.21 (0.31)a Cont (2011) None1c Blooda (male) 0.009 0.57 (0.04)a 0.42 (0.03)b 0.59 (0.05)a None1 Blood (female) <0.001 1.99 (0.07)a 1.49 (0.07)b 1.82 (0.09)a None1 Oocyte stage 0.508 4.04 (0.08)a 4.11 (0.37)a 4.20 (0.09)a None1 Active ovarioles 0.014 18.86 (0.20)b 18.93 (0.23)b 19.80 (0.28)a None2 Blood (male) 0.670 0.60 (0.04)a 0.64 (0.05)a 0.65 (0.06)a None2 Blood (female) 0.139 1.73 (0.06)a 1.73 (0.08)a 1.94 (0.11)a None2 Oocyte stage 0.064 3.95 (0.10)a 3.51 (0.13)a 3.78 (0.14)a None2 Active ovarioles 0.075 18.09 (0.21)a 18.78 (0.22)a 18.56 (0.28)a Time intervals are 2-wk periods, with 3–5 sampling points (dates) per period at 3–4 d intervals. Flies netted from 3–6 cows per date, with n = 30 flies of each sex processed per date. Flies were always collected before treatments were applied on a given treatment day (i.e., 3–4 d had passed since the latest treatment). Row means followed by the same online letter are not significantly different (P > 0.05) using ANOVA (blood, active ovarioles) or Kruskal–Wallis test (oocyte stage). FA = 15% fatty acid mixture (C8–C9–C10); GER = 2% geraniol mixture. ANOVA, analysis of variance. aEstimated blood weight in mg/fly. bStage = oocyte stage (1 = earliest follicle development to 5 = mature eggs). cNone1 = untreated but time same as FA treatment periods; None2 = untreated but time same as GER treatment periods. View Large Female blood weights, on the other hand, were distinctly (P < 0.001) reduced during the repellent spray periods for the 2010 and 2011 FA, as well as during the 2010 and 2011 GER treatments. Female blood weights also were reduced for control cattle during the FA treatment period in 2011, but did not differ during the GER treatment period. Where significant treatment effects on blood weights were noted, blood weights rebounded in the posttreatment period, which usually was not different from the pretreatment period. Relative to pretreatment, the FA treatment period showed female blood weights reduced by 17.2% in 2010 and 26.5% in 2011, while the GER reduced blood weights by 29.7% in 2010 and 38.4% in 2011. Blood ingestion by the separated control herd during the FA treatment period also was reduced by 25.1% relative to the pretreatment period, while controls in the GER period did not differ for the period before versus during spraying of the treatment herd. The number of active ovarioles was very significantly reduced (P < 0.001) for FA-treated cattle in 2010, although, relative to pretreatment, that reduction was only 3.5%. There was no reduction relative to the pretreatment period for the GER cattle in 2010, or for the FA cattle or GER-treated cattle in 2011. The GER treatment, however, did not significantly reduce ovariole number. Ovariole number in the control cattle was not reduced in 2011 periods where the treated cattle were being sprayed. Mean oocyte stage was reduced (P < 0.001) during the treatment period for treated cattle during the FA treatment period in 2010 and 2011 and the GER treatment period in 2011. There was no reduction in oocyte stage for the GER-treated cattle in 2010 or for control cattle in 2011 for either repellent. If repellent treatment reduced oocyte stage, the mean oocyte stage increased at least to pretreatment levels during the posttreatment period. We did not determine parity of dissected flies. However, flies with primary follicles in stage 1 or 2 were considered to be young nullipars (Krafsur and Ernst 1986), since gravid flies already have the secondary follicles matured to stage 3 at the time earlier maturing eggs are laid (Fig. 1). This therefore provided a supplemental perspective to mean oocyte stage. We looked specifically for periods with larger proportions of young flies, as might occur during repellent treatment periods if older flies were killed or driven away. Fig. 5 shows this for treated herds (before, during, and after treatments) in 2010 and 2011, as well as for the control herd in 2011. Fig. 5. View largeDownload slide Percentage of young nullipars (oocyte stages 1 and 2) in collections of female horn flies from pastured cattle in Pomona, California. Fig. 5. View largeDownload slide Percentage of young nullipars (oocyte stages 1 and 2) in collections of female horn flies from pastured cattle in Pomona, California. Overall 14.6% of all females in the study were at oocyte stage 1or 2. The comprehensive analysis indicated that the treatments did not differ (F = 0.94; df = 3, 61; P = 0.429) and there was no treatment * period interaction (F = 0.18; df = 6, 61; P = 0.983). The periods (before, during, and after treatment) did differ, however (F = 6.10; df = 2, 61; P < 0.01). Pooling treatments and years, young nullipars were significantly more prevalent (F = 7.31; df = 2, 72; P = 0.001) when treatments were being applied (23.0% young) versus lower prevalence of 12.2% before treatment and 5.9% after treatment; pretreatment and posttreatment did not differ from each other. Relative to pretreatment, the percentage of these young flies in the field population approximately doubled when the FA treatments were being applied in 2010 (from 15% to 27%) and when the GER treatments were being applied in 2011 (from 14% to 30%). A similar increase also was noted, however, in the 2011 control herd during the period when the treated herd was being sprayed with GER (from 13% to 26%). In that case, it contrasts somewhat with the overall mean oocyte stage, for which the medians did not differ among treatment periods (Table 1). A separate analysis examined the effect of repellents on prevalence of young flies but without the control herd, using the two treatments (FA and GER), 2 yr (2010 and 2011), three treatment periods, and interactions. Only period was significant (F = 4.39; df = 2, 41; P = 0.019). In the subsequent one-way ANOVA pooling treatments and years, there were 12.0% young females in the pretreatment period, 23.2% in the treatment period, and 6.7% in the posttreatment period. Pretreatment did not differ from treatment or posttreatment periods, while the treatment and posttreatment periods differed (F = 4.84; df = 2, 48; P = 0.012). Discussion A mineral oil (carrier control) treatment was not included in the study because we lacked a pasture to separate a herd for that purpose. The light misting of mineral oil likely had some effects itself on horn fly numbers itself, although such carrier oil effects were probably less than about 24 h (Mullens et al. 2009, Lachance and Grange 2014, Zhu et al. 2014). In horn fly laboratory bioassays, the fatty acids in a dry clay carrier had distinct effects against adult horn flies for about 3 d (Mullens et al. 2009), so the repellents in the present trials were regarded as having effects beyond those of the oil carrier. Nevertheless, the present treatments properly should be regarded as involving the oil-repellent mixtures, rather than measuring effects of the fatty acids or geraniol themselves. Blood ingestion by horn flies has been examined several times, but using two very different basic methods of feeding (blood-soaked pad or feeding from cattle) and assessing blood ingestion in different ways, including visual assessments, weight assessments, and/or biochemical quantification. In perhaps the most realistic feeding trial on a live host, Bruce (1964) placed a single group of young (10 h post-emergence) horn flies on a caged bovine, allowed them to feed for an unstated period of time, and compared their weights to those of previously separated, unfed flies from the same cohort. He estimated weights of blood ingested at 1.04 mg per fly for males and 1.15 mg for females. A single group of older adult horn flies then was captured from an animal in the field (exact age unknown), starved for 16 h, similarly allowed to feed on an animal, and the group weight was compared with flies that had not been allowed to feed. Those flies ingested an average of 1.23 mg for males and 2.19 mg for females. Most other studies have been done in a laboratory, feeding horn flies using blood-soaked pads. Harris et al. (1966) fed horn flies on a pad soaked with a blood mixture consisting of 1 part blood, 1 part saline plus bovine muscle extract, and antibiotics. Using the same blood mixture formula, Harris and Frazar (1970) estimated daily “blood” loss from the pad by weight, correcting for control pad weight loss due to evaporation. Ingestion per fly was estimated by this rough measure at 12 mg/d for males and 17 mg/d for females, and the authors estimated that females ingested about 1.4 mg of blood material per meal. Harris et al. (1974) scored blood volumes visually on a scale from 0 to 8. Flies feeding on cattle fed 4 times as often and took about 10 times longer to feed than did flies offered a blood-soaked pad. Flies fed on blood both day and night, and females fed more often than males did. Kuramochi and Nishijima (1980) fed horn flies on pads soaked with blood containing sodium citrate to block coagulation. They then used a combination of gravimetric methods and a second method of adding amaranth to blood, estimating blood weight using a spectrophotometer. With body weights of about 4.0–4.5 mg, the flies (mixed sexes) ingested 1.4–1.7 mg per feeding. Kuramochi and Hori (1984) fed horn flies for different periods of time and used the same methods. They estimated females that were allowed 150 s to fully engorge ingested an average of up to 2.8 mg at each feeding, while males took less (about 1.7 mg). Using the same methods as Kuramochi and Nishijima (1980), Kuramochi (1985) later confirmed that females in the laboratory took larger meals from blood-soaked pads. Over the duration of the gonotrophic cycle (6–9 d depending on holding temperatures from 28°C to 32°C), females ingested 51–65% more blood than males did. Kuramochi (2000) fed citrated blood from a variety of animals to horn flies and estimated daily ingestion (via the amaranth method) and survival. Flies survived twice as long using cattle blood and ingested 10–13 mg/d, with females again taking more blood than males. More recently, Cupp et al. (2010), using the Drabkin’s method but dissecting guts from horn flies, estimated ingestion (mixed sexes) at 0.9–1.4 µl per fly (their control flies). Allingham et al. (1994) tested ivermectin in blood meals for impact on the related buffalo fly, H. irritans exigua DeMeijere (Diptera: Muscidae). In the process, they visually estimated that males took larger blood meals than females for their very first feeding, but did not quantify that or provide specifics. As opposed to the lab studies above, our study provides substantial field data on blood ingestion by horn flies from cattle. In general, our estimates of blood contained in their abdomens are roughly in the range of those estimates provided in earlier literature, but tend to be slightly higher. Pretreatment averages (before repellent treatments began) in the present study were 0.5–1.0 mg for males and 2.0–2.7 mg for females. Given the very frequent feeding by horn flies from cattle, 24 times per d for males and 38 times per d for females (Harris et al. 1974), our blood estimates no doubt reflect multiple feeding bouts. The present study shows field-collected males contain 2–4 times less blood than females, which agrees with Bruce (1964) and several laboratory studies above and probably reflects the demands of oogenesis in females. We saw no evidence of greater male blood ingestion during an initial blood meal, as noted in the laboratory by Allingham et al. (1994), but our ability to detect such an effect in the field, if it existed, was likely low. Relative changes in blood ingestion were of particular interest in the repellent trials. Blood weights distinctly declined (by 17–38%) for females, but usually not for males, during treatment using either repellent. Those blood weights rebounded just as distinctly after treatments ceased. It must be emphasized that, with the exception of the last FA treatment in 2010, these estimates were taken from horn fly populations on cattle that had not been sprayed for 3–4 d, when fly numbers had substantially rebounded from counts immediately after spraying (Mullens et al. 2017). Good numbers of flies were on the cattle, but normal feeding still appeared to be affected. So our estimates of the effects of repellents on blood ingestion could be considered both conservative and chronic over the 2-wk treatment time frame. Further, each treated herd had a small number (n = 4) of unsprayed animals, but even with them in the field, we still observed population-level reductions in horn fly blood feeding and fitness. The unsprayed cattle no doubt soon made some contact with bodies of treated cattle in the herd. Significantly, horn flies also are well known to move readily among animals in a herd. In the early days of pyrethroid use, when such treatments were maximally effective, Harvey and Brethour (1979) wiped out all horn flies in a cattle herd 1 d after treating a single animal in the herd. Similarly, Krafsur and Ernst (1986) saw clear impacts on horn flies occupying control cattle, merely caused by those cattle getting into a pasture adjacent to cattle treated with stirofos ear tags. So, it is likely that flies on the control cattle did get exposed to repellents either by those cattle rubbing against treated ones or by flies moving to treated cattle. We cannot explain the drop in fly blood weights in the control herd in 2011 during the period when the FA was being applied to a herd 1.6 km away. However, the related metrics (oocyte stage and number of active ovarioles), which often also responded to treatments (see below), were unaffected for that control fly population and time. Horn flies do not exhibit gonotrophic harmony or concordance; i.e., females do not synchronously develop a single egg batch at a time or in response to a single blood meal (Schmidt 1972; Krafsur and Ernst 1983, 1986; Kuramochi and Nishijima 1984). Rather, they feed repeatedly in a single gonotophic cycle and mature older follicles while simultaneously adding yolk to less mature follicles (Fig. 1). This contributes to the relative preponderance of field flies with more mature oocyte development in the proximal follicle, as we found and as also noted by other field researchers (Krafsur and Ernst 1983, 1986; Kuramochi and Nishijima 1984). Unlike those authors, we did not try to assess parity specifically, as the follicular relics (“yellow bodies”) can be difficult to see in gravid flies (Krafsur and Ernst 1986). Further, it is very difficult to train different people to perform the dissections and judge parity in the same way, and the number of flies dissected in the present study made it quite difficult for one person to do all of them. Nevertheless, we usually were able to detect very clear responses to repellent treatment periods, except for the GER treatment period in 2010. The GER treatment is not immediately fatal to horn fly adults hit by the spray, as the FA treatment can be (Mullens et al. 2017), so perhaps the effect on oogenesis is less repeatable for GER. Oocyte stage differences were not seen during the season for the control herd. Krafsur and Ernst (1986) similarly observed a distinct drop in older (vitellogenic) flies and an increase in younger flies after the control herd being studied herd came close to a treated herd (stirofos ear tags). The mechanism of reduced oocyte stage in a treated horn fly population is likely via killing or repelling flies from recently treated cattle, followed by a relative influx of younger nullipars emerging from local dung pats or dispersing from other areas. If younger flies ingest less blood, that could contribute to the drop in blood ingestion that we often observed after treatment, and this may have occurred in the unreplicated studies of Bruce (1964). However, we know of no studies that have shown definitively that younger flies ingest less blood and consider the repellent explanation to be more likely. Horn flies are excellent dispersers. This would be expected in a fly that evolved with free roaming wild bovid herds. It would be common for the herd to have moved a long distance in the 2 wk or more it takes for horn flies to emerge from natal pats. Sheppard (1994) documented movement of marked flies up to 5 km in 4 h, and Kinzer and Reeves (1974) saw marked flies move up to 10–12 km overnight. Horn flies tend to disperse as young flies, staying with a herd once they find it; Guillot et al. (1988) showed newly arriving dispersing flies were 89% nullipars, and about half of those were previtellogenic. In the present study, proportional representation of young nulliparous females (stage 1 and 2 oocytes) was higher in the period when treatments were being applied. Oddly, this was also true for the unsprayed herd 1.6 km from the treatment herd. Unfortunately we had only 1 yr of data from a completely unsprayed herd, while we had 2 yr of repellent data. The increase in young females lends some support to the idea that repellent effects influenced age structure. Another mechanism for an increase in physiologically young flies, however, might be the reduced amounts of blood taken by females on treated cattle, as we have shown, and subsequent possible impacts of those reduced blood meals on rate of oogenesis. Future analyses of pterin data (Mullens et al., in preparation) hopefully will allow a more detailed treatment of horn fly age structure changes relative to these repellents. Trends in horn fly fecundity have been examined a number of times, and our estimates of number of active ovarioles are close to expected for that species. Fecundity is related to size (Schmidt and Blume 1973); Lysyk (1991) showed that the largest horn flies reared in the laboratory could contain as many as 32 ovarioles per female, but most published estimates are somewhat less. Laboratory studies by Schmidt (1972) showed an average of 19 active ovarioles each, and older females had fewer active ovarioles. Thomas and Kunz (1985) showed eggs laid per female in summer often were in the range of 10–17 eggs/d, with a maximum of 22 eggs/d. Kuramochi and Nishijima (1984) looked at marked flies released onto field cattle and recaptured. They showed those flies had an average of 22–23 active ovarioles each. Krafsur and Ernst (1983) showed it was common for females to mature up to 6 eggs/ovary at a time, with an average of 18.4 functional ovarioles per fly. It is notable that the number of active ovarioles also responded at times to periods of repellent treatment, such as the 2010 FA application. However, such responses were not as marked, and were not as repeatable, as the other two metrics (blood weights and oocyte stage). Consequently, the effects of repellent treatments on fecundity appear to be less common and severe than reductions in those other parameters. One mechanism for this might be if flies feeding on a repellent-treated host compensated for reduced blood volume per meal by taking more frequent meals, but this cannot be determined from our data. The present studies show that, even if populations were sampled 3–4 d after cattle treatment, these repellent-oil mixtures were impacting blood ingestion and reproductive fitness in horn flies in the field. Consequently, such longer-term or sublethal effects should be considered as contributors to the pest management potential of regularly applied repellents for strategic horn fly management. Acknowledgments These studies were supported by USDA-CSREES grant 2009-51100-05769. We thank Cal Poly Pomona, especially Dr. B. Sandelin and herdsman A. Estep, for allowing us to use their cattle in the study. A. Estep and several Cal Poly and UC Riverside staff and students were very helpful with cattle handling and treatment. We are grateful to Dr. M. K. Rust (UC Riverside) for the use of his scale and to Dr. K. Lohmeyer (USDA-ARS, Kerrville, Texas) for generously sending fly pupae for use in the initial blood feeding calibration. References Cited Allingham, P. G., Kemp D. H., Thompson D. R., and Rugg D.. 1994. Effect of ivermectin on three field populations and a laboratory strain of Haematobia irritans exigua (Diptera: Muscidae). J. Econ. Entomol . 87: 573– 576. Google Scholar CrossRef Search ADS Barnard, D., and Xue R.-D.. 2009. Bloodmeal mass and oviparity mediate host avidity and DEET repellency in Aedes albopictus (Diptera: Culicidae). J. Med. Entomol . 46: 1235– 1239. Google Scholar CrossRef Search ADS PubMed Birkett, M. A., Agelopoulos N., Jensen K.-M. V., Jespersen J. B., Pickett J. A., Prijs H. J., Thomas G., Trapman J. J., Wadhams L. J., and Woodcock C. M.. 2004. The role of volatile semiochemicals in mediating host location and selection by nuisance and disease-transmitting cattle flies. Med. Vet. Entomol . 18: 313– 322. Google Scholar CrossRef Search ADS PubMed Briegel, H., Lea A. O., and Klowden M. J.. 1979. Hemoglobinometry as a method for measuring blood meal sizes of mosquitoes (Diptera: Culicidae). J. Med. Entomol . 15: 235– 238. Google Scholar CrossRef Search ADS Bruce, W. G. 1964. The history and biology of the horn fly, Haematobia irritans (Linnaeus); with comments on control. North Carolina Agric. Exp. Sta. tech Bull . 157: 33. Cupp, M. S., Cupp E. W., Navarre C., Zhang D., Yue X., Todd L., and Panangala V.. 2010. Salivary gland isoforms differentially regulate blood uptake of horn flies fed on control- and thrombostasin-vaccinated cattle. J. Med. Entomol . 47: 610– 617. Google Scholar CrossRef Search ADS PubMed Debboun, M., Frances S. P., and Strickman D.. 2014. Insect repellents: Principles, Methods and Uses , 2nd ed. CRC Press, Boca Raton, FL. Ellse, L., and Wall, R. 2014. The use of essential oils in veterinary ectoparasite control: a review. Med. Vet. Entomol . 28: 233– 243. Google Scholar CrossRef Search ADS PubMed Foil, L. D., and Hogsette, J. A. 1994. Biology and control of tabanids, stable flies and horn flies. Rev. Scien. Tech. Off. Internat. Epizooties 13: 1125– 1158. Google Scholar CrossRef Search ADS George, D. R., Finn, R. D., Graham, K. M., and Sparagano, O.A.E. 2014. Present and future potential of plant-derived products to control arthropods of veterinary and medical significance. Par. Vect . 7: 28. Google Scholar CrossRef Search ADS Guillot, F. S., Miller J. A., and Kunz S. E.. 1988. The physiological age of female horn flies (Diptera: Muscidae) emigrating from a natural population. J. Econ. Entomol . 81: 555– 561. Google Scholar CrossRef Search ADS PubMed Hao, H. L., Wei J. R., Dai J. Q., and Du J. W.. 2008. Host-seeking and blood-feeding behavior of Aedes albopictus (Diptera: Culicidae) exposed to vapors of geraniol, citral, citronellal, eugenol, or anisaldehyde. J. Med. Entomol . 45: 533– 539. Google Scholar CrossRef Search ADS PubMed Harris, R. L., Frazar E. D., and Graham O. H.. 1966. Resistance to Ronnel in a strain of horn flies. J. Econ. Entomol . 59: 387– 390. Google Scholar CrossRef Search ADS Harris, R. L. and Frazar E. D.. 1970. Intake of blood by adult horn flies reared in the laboratory. Ann. Entomol. Soc. Amer . 63: 1475– 1476. Google Scholar CrossRef Search ADS Harris, R. L., Miller J. A., and Frazar E. D.. 1974. Horn flies and stable flies: feeding activity. Ann. Entomol. Soc. Amer . 67: 891– 894. Google Scholar CrossRef Search ADS Harvey, T. L., and Brethour J. R.. 1979. Treatment of one beef animal per herd with permethrin for horn fly control. J. Econ. Entomol . 72: 532– 534. Google Scholar CrossRef Search ADS PubMed Jensen, K.-M. V., Jespersen J. B., Pickett M. A., Pickett J. A., Thomas G., and Wadhams L.. 2004. Variation in the load of the horn fly, Haematobia irritans (Diptera: Muscidae) in cattle herds is determined by presence or absence of individuals. Med. Vet. Entomol . 18: 275– 280. Google Scholar CrossRef Search ADS PubMed Kinzer, H. G., and Reeves J. M.. 1974. Dispersal and host location of the horn fly. Environ. Entomol . 3: 107– 111. Google Scholar CrossRef Search ADS Krafsur, E. S., and Ernst C. M.. 1983. Physiological age composition and reproductive biology of horn fly populations, Haematobia irritans irritans (Diptera: Muscidae) in Iowa, USA. J. Med. Entomol . 20: 664– 669. Google Scholar CrossRef Search ADS Krafsur, E. S., and Ernst C. M.. 1986. Phenology of horn fly populations (Diptera: Muscidae) in Iowa, USA. J. Med. Entomol . 23: 188– 195. Google Scholar CrossRef Search ADS PubMed Kumar, S., Singh A. P., Nair G., Batra S., Seth A., Wahab N., and Warikoo R.. 2011. Impact of Parthenium hysterophorus leaf extracts on the fecundity, fertility and behavioral response of Aedes aegypti L. Parasitol. Res . 108: 853– 859. Google Scholar CrossRef Search ADS PubMed Kuramochi, K. 1985. Studies on the reproductive biology of the horn fly, Haematobia irritans (L.)(Diptera: Muscidae). II. Effect of temperature on follicle development and blood meal volume of laboratory-reared flies. App. Ent. Zool . 20: 264– 270. Google Scholar CrossRef Search ADS Kuramochi, K. 2000. Survival, ovarian development and bloodmeal size for horn fly Haematobia irritans irritans reared in vitro. Med. Vet. Entomol . 14: 201– 206. Google Scholar CrossRef Search ADS PubMed Kuramochi, K., and Hori, K. 1984. Blood meal size of the horn fly, Haematobia irritans (L.) estimated by measuring the feeding time. Res. Bull. Obihiro Univ . 13: 287– 292. Kuramochi, K., and Y. Nishijima. 1980. Measurement of the meal size of the horn fly, Haematobia irritans (L.) (Diptera: Muscidae), by the use of amaranth. Appl. Ent. Zool. 15: 262–269. Kuramochi, K., and Nishijima Y.. 1984. Studies on the reproductive biology of the horn fly, Haematobia irritans (L.)(Diptera: Muscidae). I. The ovarian development of flies collected from the field. Appl. Entomol. Zool . 19: 331– 340. Google Scholar CrossRef Search ADS Lachance, S., and Grange, G. 2014. Repellent effectiveness of seven plant essential oils, sunflower oil and natural insecticides against horn flies on pastured dairy cows and heifers. Med. Vet. Entomol . 28: 193– 200. Google Scholar CrossRef Search ADS PubMed Lysyk, T. L. 1991. Use of life history parameters to improve a rearing method for horn fly, Haematobia irritans irritans (L.)(Diptera: Muscidae), on bovine hosts. Canad. Entomol . 123: 1199– 1207. Google Scholar CrossRef Search ADS Mullens, B. A., Reifenrath, W. G., and Butler, S. M. 2009. Laboratory trials of fatty acids as repellents or antifeedants against house flies, horn flies and stable flies (Diptera: Muscidae). Pest Manag. Sci . 65: 1360– 1366. Google Scholar CrossRef Search ADS PubMed Mullens, B. A., Watson D. W., Gerry A. C., Sandelin B. A., Soto D., Rawls D., Denning S., Guisewite L., and Cammack J.. 2017. Field trials of fatty acids and geraniol applied to cattle for suppression of horn flies, Haematobia irritans (Diptera: Muscidae), with observations on fly defensive behaviors. Vet. Parasitol . 245: 14– 28. Google Scholar CrossRef Search ADS PubMed Oyarzun, M. P., Quiroz, A., and Birkett, M. A. 2008. Insecticide resistance in the horn fly: alternative control strategies. Med. Vet. Entomol . 22: 188– 202. Google Scholar CrossRef Search ADS PubMed Schmidt, C. D. 1972. Classification of the physiological development of laboratory-reared female horn flies, Haematobia irritans. Ann. Entomol. Soc. Amer . 65: 695– 701. Google Scholar CrossRef Search ADS Schmidt, C. D., and Blume R. R.. 1973. Laboratory-reared horn flies: relationship between width of head and weight of pupa in both sexes and between these measurements and number of ovarioles. Ann. Entomol. Soc. Amer . 66: 1307– 1308. Google Scholar CrossRef Search ADS Sheppard, D. C. 1994. Dispersal of wild-captured, marked horn flies (Diptera: Muscidae). Environ. Entomol . 23: 29– 34. Google Scholar CrossRef Search ADS Sugiharto, V. A., Grieco J. P., Murphy J. R., Olsen C. H., Colaciccho-Mayhugh M. G., Stewart V. A., Achee N. L., and Turell M. J.. 2016. Effects of preexposure to DEET on the downstream blood-feeding behaviors of Aedes aegypti (Diptera: Culicidae) mosquitoes. J. Med. Entomol . 53: 1100– 1104. Google Scholar CrossRef Search ADS Thomas, D. B., and Kunz S. E.. 1985. Effect of season and density on the fecundity and survival of caged populations of adult horn flies (Diptera: Muscidae). J. Econ. Entomol . 78: 106– 109. Google Scholar CrossRef Search ADS Xue, R.-D., Ali A., and Barnard D. R.. 2007. Effects of in vivo exposure to DEET on blood feeding behavior and fecundity of Anopheles quadrimaculatus (Diptera: Culicidae). Exp. Parasitol . 116: 201– 204. Google Scholar CrossRef Search ADS PubMed Zhu, J. J., Brewer G. J., Boxler D. J., Friesen K., and Taylor D. B.. 2014. Comparisons of antifeedancy and spatial repellency of three natural product repellents against horn flies, Haematobia irritans (Diptera: Muscidae). Pest Manag. Sci . 70: 1553–1560. © The Author(s) 2017. Published by Oxford University Press on behalf of Entomological Society of America. All rights reserved. For permissions, please e-mail: firstname.lastname@example.org.
Journal of Medical Entomology – Oxford University Press
Published: Mar 1, 2018
It’s your single place to instantly
discover and read the research
that matters to you.
Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.
All for just $49/month
Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly
Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.
Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.
Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.
All the latest content is available, no embargo periods.
“Hi guys, I cannot tell you how much I love this resource. Incredible. I really believe you've hit the nail on the head with this site in regards to solving the research-purchase issue.”Daniel C.
“Whoa! It’s like Spotify but for academic articles.”@Phil_Robichaud
“I must say, @deepdyve is a fabulous solution to the independent researcher's problem of #access to #information.”@deepthiw
“My last article couldn't be possible without the platform @deepdyve that makes journal papers cheaper.”@JoseServera