CASE STUDY: Impact of Horn Flies, Haematobia irritans (L.) (Diptera: Muscidae), on the Behavior of Beef Steers

INTRODUCTION

Horn flies (Haematobia irritans L.; Diptera: Muscidae), are the most economically damaging pest of cattle on pasture (Steelman, 1976; Byford et al., 1992). Significant decrease in weight gains of calves have been attributed to the stress, annoyance, blood loss, and reduced feed efficiency resulting from horn fly infestation (Campbell, 1976; Quisenberry and Strohbehn, 1984). Horn flies also cause damage to livestock by their indirect role in disease such as keratoconjuctivitis and infections of filarial parasites (Haufe, 1987). Total losses accredited to horn fly infestation are reported to be $876 million annually (Kunz et al., 1990). Steelman (1990) reported a decreased of 8.1 kg in calf weaning weights for every 100 horn flies present on the cow. This decrease is likely contributable to a decrease in cow milk production (Kunz et al., 1990). Yearling beef heifers with insecticide treatment for horn flies had 14% greater weight gain than untreated heifers (DeRouen et al., 2003). These effects may be due to changes in animal behavior that reduce grazing and increase energy loss due to efforts of the animal to fend off horn flies (Harvey and Brethour, 1979). The objective of this study was to evaluate the well-being and behavior of beef steers experiencing horn fly infestation before and after application of an insecticide ear tag.

MATERIALS AND METHODS

An observational case study was conducted at Kentland Farm, Blacksburg, Virginia (37°11' N, 80°35' W) from July 6 to 11, 2006. Angus-cross steers (n = 4; 543 ± 19 kg) were grazing on endophyte-free tall fescue (Festuca arundinacea Shreb.). Cattle were treated with a commercially available Pyrethrin insecticide pouron (1% lambdacyhalothrin) 2 wk before the study. However, the horn fly infestation returned at a level suggesting they were pyrethroid resistant. Cattle were then monitored for a period before and after application of an organophosphate insecticide ear tag (Co-Ral Plus; 20% diazinon and 20% coumaphos, Bayer Healthcare, Shawnee Mission, KS) on d 0 at 1030 h. Methods were approved by the Virginia Tech Animal Care and Use Committee.

Direct visual observations of animal behavior were recorded during morning (0830 to 0930 h) and afternoon (1530 to 1630 h) grazing sessions starting on the afternoon of d - 2 through the morning of d 3. The observer monitored behavior from a distance to avoid disturbing the behavior of the steers and the flies. Cattle behaviors that are well established as fly-avoidance activities were recorded (Harvey and Launchbaugh, 1982; Dougherty et al., 1993a, b). Behaviors included head throws, tail flicks, ear flicks, leg stomps (front and rear), and skin twitches (panniculus reflex) and were defined according to Dougherty et al. (1993a, b) and Mullens et al. (2006). Briefly, a head throw was defined as tossing back of the head at an angle such that the animal's nose crossed an imaginary plane across the animal's chest. A tail flick was recorded when the tail moved from the resting position (straight down) to the side. If the tail recrossed the resting position and move to the other side an additional event was recorded. Ear flicks were recorded when the ear moved as a single twitch or rotated for a few seconds in an agitated fashion. Leg stomps were defined as lifting and dropping of the leg (not for locomotion). Contraction of the cutaneous trunci muscle causing visible skin twitches were recorded from the side facing the observer. One event was recorded if a single twitch or an uninterrupted shiver over an area was observed. Activities were recorded as number of events per minute. Grazing behavior parameters recorded were bite rate (bites/min) and number of feeding stations per minute. Feeding station was defined as the area (half-cylinder shape within reach of the mouth) an animal grazed with its front feet stationary (Ruyle and Dwyer, 1985). When the animal moved its front feet to a new area it was considered another feeding station. During an observation period, each steer was monitored for two 1-min periods for fly-avoidance activities and for two 1-min periods for bite rate and number of feeding stations.

Total daily number of bites was recorded with the use of a bite counter developed at the Grazing Management Laboratory at the National Agricultural Research Center for the Hokkaido Region in Japan as described by Kawamura et al. (2006). These bite counters have been validated for use on cattle (K. Umemura, unpublished data). The device was attached to a collar worn around the neck of the steers (n = 2) and counted grazing bites (but not ruminating mastications) on 10-min intervals. Bite counters were worn for the entire experimental period; however, only data from d -1, 1, 2, and 3 are reported because of the periods of time when animals had to be removed from the paddocks on d -2 (to put the bite counters on) and 0 (when ear tags were applied).

High-resolution digital photographs were taken (Olympus Camedia, Model c-50, Olympus America Inc., Center Valley, PA) of each steer from both lateral views of the animal during each period (0830 to 0930 h and 1530 to 1630 h) in which behavioral observations were conducted. Images were later analyzed with computer software (Photoshop Elements, Adobe, San Jose, CA) that allowed for magnification of the images. The number of flies present on both sides of each steer was counted and the 2 numbers were added together to give the total fly burden on the animal at each sampling time. When horn fly numbers are excessive, the use of recorded images has been suggested because of the speed at which observations can be recorded (Castro et al., 2005). Castro et al. (2005) reported that recorded images of flies on cattle are highly correlated to a direct counting method in the field by trained observers.

Air temperature (°C) and relative humidity (%) were recorded during the data collection periods (Table 1) using a weather station (WeatherWatch2000 Weather Station, Campbell Scientific Inc., Logan, UT) located within 0.2 km of the experimental paddocks.

Data were analyzed using the GLM procedure (version 9.1.3, SAS Inst. Inc., Cary, NC) with Tukey's adjustment for means separation. The model included main effect of day. Transformations (square root (x + 0.5) or log base 10 + 1) were performed on all data (except bite rate) to meet the assumptions of the ANOVA. Significance is reported for transformed values whereas untransformed means are reported. Correlations between observation variables and fly numbers were analyzed using the CORR procedure of SAS. A significance level of a = 0.05 was set for analyses with trends defined as 0.10 > a > 0.05.

RESULTS AND DISCUSSION

The number of flies present differed (P < 0.0001) between the days before application of insecticide ear tag and after the tag was in place (Table 2). Before ear tags were applied, when fly numbers were greatest, steers flicked their tail more (P < 0.0001) than after ear tags were in place and fly numbers decreased. This is in agreement with previous reports of greater rates of tail switching in untreated cattle compared with those treated with insecticide (Harvey and Launchbaugh, 1982). Number of tail flicks has been suggested as a means for producers to monitor intensity of pest infestations (Mullens et al., 2006). Steers were observed stomp- ing more (P = 0.005) on d 0 before ear tag treatment than on d 2 and 3, whereas no other differences between days were observed. Number of head throws by steers before ear tagging was greater (P = 0.003) than the number of head throws observed after tagging. Harvey and Launchbaugh (1982) found that head tossing never exceeded 1/min and averaged 0.06/ min for insecticide-treated steers compared with 0.45/min for those un- treated. The number of skin twitches observed before ear tags were applied was greater (P = 0.0004) than after ear tag application. Number of ear flicks by steers before ear tags were applied did not differ (P = 0.31) from the number observed after tagging. The lack of difference observed may not be related to fly numbers: application of the ear tag itself may have been the cause of steers flicking their ears after d 0 regardless of whether flies were present in excessive numbers or not. Harvey and Launchbaugh (1982) reported rates for ear flicks, skin twitches, and leg stomping were low and varied widely in cattle that were treated and untreated for horn flies.

Bite rate before ear tags were applied differed (P = 0.02) from bite rate after ear tag application, with rates increasing somewhat after ear tags were applied. Infestation with stable flies (Stomoxys calcitrans L.) has been shown to decrease bite but increase bite mass as compensation to maintain DMI (Dougherty et al., 1993b). The number of feeding stations per minute grazed by steers before ear tag treatment was (P < 0.0001) than the number of stations visited after treatment. Steers walked more, moving to new feeding stations because of agitation by the horn flies, not because they exhausted the resource of the feeding station they were in. This could lead to a decrease in energy efficiency by expending more energy to travel to a new resource before necessary.

There tended (P = 0.09) to be a day effect for total bites per day recorded by the bite counters. On d 3 the total bites (29,800) were greater (P = 0.02, Figure 1) than bites on d -1 (25,247) and tended to be greater than on d 1 (27,010; P = 0.09) and 2 (26,648; P = 0.07) as well (SEM = 885). Total bites per day have not been previously reported for cattle experiencing excessive horn fly infestation. Figure 1 illustrates the pattern of grazing bouts over the course of d -1 and 3. Steers seemed to be grazing more in the early morning hours before sunrise before ear tag treatment, perhaps in an attempt to avoid horn flies that are more active during daylight hours. The late evening grazing bout (1800 to 2100 h) did not differ greatly between d -1 and 3. However, after ear tagging, steers had 4 major grazing sessions during the daylight hours whereas before ear tag application, grazing bouts were shorter and more erratic, most likely due to annoyance from the horn flies.

Correlations

Bite rate was negatively correlated to the number of flies on the steers (r = -0.33, P = 0.04, Table 3). Dougherty et al. (1993b) found that correlation of fly populations and biting rate varied with the location of stable flies on the animal's body. Stable flies that rested on the head were positively correlated (r = 0.28, P = 0.03) with bite rate whereas flies on the hind legs were negatively correlated (r = -0.26, P = 0.04) with bite rate. Horn fly populations in the current study were found on many areas of the body, predominantly along the back and shoulders, and down the sides to the prepuce (Figure 2).

The number of feeding stations in the present study was positively correlated with number of fly population (r = 0.58, P < 0.0001). Dougherty et al. (1993b) also reported a positive correlation (r = 0.34, P < 0.01) between movement to new feeding stations and stable fly populations, particularly with flies occurring on the animals head. Head throws (r = 0.20, P = 0.003) and tail flicks (r = 0.54, P = 0.0003) were positively correlated (P = 0.02) to the number of flies present on the steers (Table 3). A positive correlation trend was observed between number of flies present to the frequency of leg stomps (r = 0.30, P = 0.06) and skin twitches (r = 0.29, P = 0.07)

IMPLICATIONS

Decreasing fly burdens can promote animal well-being by decreasing the display of fly-avoidance activities and allowing animals to graze for longer periods without interruption. Past studies have reported no impact of horn flies on total grazing time; however, the other components of DMI (bite rate and bite mass) were not evaluated. In the present study, the negative correlation of bite rate with fly numbers and the increase in bite rate after horn fly load decreased suggests that DMI might have been lower when horn fly infestation was higher. However, steers might have increased bite size to compensate for a decrease in bite rate, hence consuming the same amount of DM as after being treated. Further research to estimate DMI and performance of treated and untreated animals is warranted.

© 2008 American Registry of Professional Animal Scientists Provided by ProQuest LLC. All Rights Reserved.