Measurement of ammonia concentrations in horse stalls

The effect of treating stall floors with an ammonia-absorbingcompound (AAC) on aerial ammonia levels was tested during two 14-day periods in a four-stall barn. In the first period, two stalls were treated with the AAC and the other two stalls remained as the non-treated controls. In the second period, the stalls were switched. Four geldings were used in each period and were randomly assigned to the stalls. The horses remained in the stalls at all times except for 30 min of turn-out per day at which time the stalls were cleaned. Ammonia levels at the stall floor were measured by a Bruel and Kjacr 1302 gas monitor on day 0 before the horses were admitted to the barn and on days 1, 3, 5, 7, 10 and 14. Ammonia diffusion tubes were attached to the horses' halters on days 1, 7, 11 and 13 for 8 hours to determine the level of ammonia exposed to the horse. On day 14, ammonia concentrations near the stall floor were 25% lower in treated stalls than in control stalls (P<0.1). Ammonia concentrations measured at the halter were much lower than those at the floor, but horses in treated stalls had lower ammonia exposure than horses in control stalls (P<0.1).     

Effect of sodium bisulfate on skin and hooves of horses

OBJECTIVE: To evaluate the safety of sodium bisulfate for use in horse barn environments by determining its irritant effect on skin and hooves. ANIMALS: 6 female mixed-breed ponies. PROCEDURE: Sodium bisulfate was applied to clipped intact skin of 6 ponies to evaluate its irritant effect after single (48 hours) and repetitive (6 h/d for 10 days) applications; similar areas of skin were used as untreated control sites. In addition, sodium bisulfate was applied to the sole of both front hooves of each pony and covered with wet gauze, and the entire hoof was covered with adhesive tape for 48 hours. RESULTS: Contact with moistened sodium bisulfate for 48 hours had no effect on pony skin. Contact with sodium bisulfate for 6 hours on 10 consecutive days did not cause gross changes but did cause mild to moderate microscopic changes including epidermal necrosis, hyperkeratosis, capillary congestion, edema, and diffuse mixed inflammatory cell infiltrate. All changes were limited to the epidermis and superficial dermis. Gross changes in hoof sole, signs of lameness, and increase in digital pulse pressure or pulse intensity were not detected. CONCLUSIONS AND CLINICAL RELEVANCE: Duration of contact with sodium bisulfate in this study was in excess of that expected under typical husbandry conditions. Despite this fact, gross changes in skin and hooves were not detected. Microscopic lesions were confined to the epidermis and superficial dermis. Results suggest that contact with sodium bisulfate under these conditions is safe.     

Detection of Salmonella organisms and assessment of a protocol for removal of contamination in horse stalls at a veterinary teaching hospital

OBJECTIVES: To assess methods of detecting environmental contamination with Salmonella organisms and evaluate a cleaning and disinfection protocol for horse stalls in a veterinary teaching hospital. DESIGN: Original study. SAMPLE POPULATION: 37 horses with diarrhea likely to be caused by Salmonella infection and their stall environments. PROCEDURES: Fecal samples were collected from horses daily during hospitalization; samples were obtained from stall sites after cleaning and application of disinfectants. Fecal and environmental samples were cultured for Salmonella spp and tested via polymerase chain reaction (PCR) assay to detect Salmonella DNA. RESULTS: 1 horse died and 2 were discharged prior to sample collection. Fecal samples from 9 of 34 horses yielded growth of Salmonella organisms on bacteriologic culture, and 23 yielded positive results via PCR assay on > or = 1 occasion. Among environmental samples from 21 stalls, salmonellae were detected at > or = 1 stall site on 6 of 78 occasions, and > or = 1 stall site yielded positive results via PCR assay on 69 of 77 occasions. Salmonella DNA was detected more frequently in samples of stall drains, cracks, and corners. Salmonella spp were cultured from samples of 3 stalls after both initial and second cleaning and disinfection cycles, but no organisms were detected in samples obtained after use of a peroxygen disinfectant. CONCLUSIONS AND CLINICAL RELEVANCE: Results suggest that stalls in which horses with salmonellosis were housed should only be used to accommodate newly hospitalized horses after samples (collected after 2 cycles of cleaning and disinfection) from drains, cracks, and corners yield negative results on bacteriologic culture.     

Evaluation of the relationship between injuries and size of gestation stalls relative to size of sows

OBJECTIVE: To determine whether there is a relationship between sow injuries and size of gestation stalls relative to sow size. DESIGN: Prospective study. ANIMALS: 267 pregnant sows. PROCEDURE: Sows were randomly selected from 4 swine farms. Sow and stall measurements were obtained, and injuries were scored on the basis of location, number, and depth. Ratios of stall length to sow length and stall width to sow height were calculated. RESULTS: High injury scores were associated with low ratios of stall length to sow length and stall width to sow height. CONCLUSIONS AND CLINICAL RELEVANCE: A small increase in stall dimensions could reduce injuries and improve well-being of sows considerably.     

Role of horse flies in transmission of wquine infectious anemia from carrier ponies

Equine infectious anemia virus was transmitted from an acutely ill and an inapparently infected pony to uninfected ponies by the interrupted feeding of horse flies (tabanids). Transmission from acutely ill ponies was not accomplished following: (1) the interrupted feeding of a single horse fly, (2) bites of horse flies that had fed on an acutely affected pony 24 hours earlier, (3) bites of horse flies that had oviposited after feeding on an acutely affected pony, or (4) the inoculation of larval material derived from horse flies that had fed to repletion. It was concluded that horse fly transmission of equine infectious anemia virus is mechanical only and that infected horses that are free of clinical signs can be a source of virus for insect transmission.     

Local airborne particulate concentration is associated with visible tracheal mucus in Thoroughbred racehorses

Reason for performing study: Accumulations of tracheal mucus assessed by endoscopic examination are associated with poor performance in racehorses. The air quality in horses' stalls may contribute to this visible tracheal mucus. Objectives: To determine whether the concentration and number of airborne particulates in stalls are associated with visible accumulations of tracheal mucus and with the number of inflammatory cells in tracheal aspirates. Methods: We studied 107 racehorses from 3 stables, in 3 different months, and measured airborne particulate matter 3 times daily in each of the stalls. On each monthly visit, horse airways were examined endoscopically and assigned a mucus score, and tracheal lavage was performed. Bivariate procedures, general estimating equations and linear mixed models were applied to estimate the association between PM and the presence of accumulations of mucus and number of inflammatory cells. Results: Stable, stall, month and PM were all significantly associated with the presence of accumulations of tracheal mucus, which had an overall prevalence of 67%. The odds of horses having visible accumulation of mucus were increased when horses occupied enclosed stables or stalls with higher particulate concentrations, and when concentrations of larger particles (≤10 µm in diameter) were elevated. Sixty‐eight percent of tracheal wash samples contained more than 20% neutrophils. Increased numbers of neutrophils were associated with the concentration of smaller particles (≤2.5 µm in diameter). Potential relevance: Careful consideration of stable construction and management practices focused on maintaining the lowest possible dust concentrations throughout the day should reduce the prevalence of visible accumulations of tracheal mucus, potentially improving racing performance.     

The epidemiology of Parafilaria bovicola in the Transvaal Bushveld of South Africa

A total of 20 375 flies collected off cattle on 12 farms over 36 months were identified and examined for 3rd stage P. bovicola. The 3 vector species accounted for 64.1% of the flies collected and were the only fly species found to be infected. Musca lusoria was clearly the dominant vector fly, although large numbers of Musca sp. A appeared regularly between February and April each year. This phenomenon, coupled with high numbers of M. lusoria throughout most of the year, led to an increase in the numbers of vector flies from their lowest level in June to a peak in February-April. Of the 13 070 vector flies examined for 3rd stage larvae only 64 (0.52%) were positive; of these 41 were M. lusoria and 17 Musca sp. A. No positive male flies were found. Incubation of wild-caught flies for up to 13 days at 27 degrees C noticeably increased the larval recovery rate. Flies were found to be infected mainly from August-March. Infected M. lusoria were recorded from July-March and infected Musca sp. A from January-May. Only 6 infected M. xanthomelas were collected and this was during the period August-December, when most ovipositional blood spots occur on cattle. It is concluded that P. bovicola transmission in the Bushveld is not correlated with peak periods of bleeding but rather with high numbers of vector flies, the various species augmenting each other so that transmission may take place almost throughout the year.     

Evaluation of the efficacy of a topically administered combination of imidacloprid and permethrin against Phlebotomus perniciosus in dog

The phlebotomine sand fly Phlebotomus perniciosus is one of the main vectors of Leishmania infantum, responsible for human and canine leishmaniasis in the Mediterranean Basin. The objective of this study was to evaluate the repellent and insecticidal efficacy of imidacloprid 10% (w/v)/permethrin 50% (w/v) spot-on against sand flies (P. perniciosus) on dogs. The dogs used in this trial were laboratory-bred beagles: eight were impregnated with the solution (treated group), while the other eight were left untreated (control group). On day 0 the animals in the treatment group received 0.1 ml/kg body weight of the combination imidacloprid/permethrin spot-on. Dogs were exposed for 1h to about 100 female sand flies at weekly intervals for a period of 4 weeks, on day 1, 7, 14, 21, and 28 after applying the product. The repellency criterion was based on the feeding rate of sand flies in the treated compared to the untreated group. The insecticidal efficacy criterion was based on comparison of the survival rate of sand flies between the two groups. The product had an insecticidal efficacy on female sand flies of 53.2% (day 1), 49.4% (day 7), 15.1% (day 14), 13.2% (day 22), and 2.9% (day 29). The product showed a repellent effect of 97.7% (day 1), 96.3% (day 7), 96.5% (day 14), 92.7% (day 22), and 74.0% (day 29). Within the first week of application the insecticidal effect was significant; however it did not surpass 50%. On the other hand, the product showed a potent anti-feeding effect of over 90% during the first 3 weeks of this trial. Therefore, the application of this product every 3 weeks would be a good tool to significantly reduce sand fly bites over the period of transmission of vectorial diseases such as leishmaniasis and several arbovirosis such as Toscana virus.     

Efficacy of doramectin 0.5% w/v Pour-On for control of the horn fly, Haematobia irritans

In 1998, three groups of cattle at three locations in Lousiana were treated with Dectomax^® (0.5% doramectin) Pour-On and horn fly populations were monitored. Acceptable levels (less than 50 flies per side) of horn fly control were observed from 4 to 8 weeks. Differences in the length of control among the three sites were most likely affected by immigration of adult flies from untreated groups. In 1999, acceptable horn fly control was obtained for 13 weeks by the use of two treatments of doramectin Pour-On.     

Occurrence of second and third instars of Gasterophilus intestinalis and Gasterophilus nasalis in stomachs of horses in Kentucky.

The occurrence of 2nd and 3rd instars of Gasterophilus intestinalis and Gasterophilus nasalis was determined in 476 horses during the 22-year period from 1951 to 1973. Overall, G intestinalis infected 98.7% of the horses and averaged 168/horse; whereas G nasalis infected 80.7% of the horses and averaged 52/horse. Aggregate average total numbers for G intestinalis ranged from a low of 50 in September to a high of 229 in March, and for G nasalis, from a low of 14 in September to a high of 82 in February. Horses were infected by 2nd or 3rd instars of both species on a year-round basis. Differentiation between the instars provided insight into the dynamics of the infections. For G intestinalis, the data indicated (1) 2nd instars from the previous year's generation of flies continue to arrive in the stomach until April; (2) arrival of the current year's crop of 2nd instars starts in July; and (3) 3rd instars from the previous year's generation of flies are not voided in appreciable numbers until August, when numbers of the current year's crop of 3rd instars appear in the stomach. For G nasalis, the data indicated (1) arrival in the stomach of the current year's generation of 2nd instars starts in July and terminates in January and (2) 3rd instars from the previous year's generation of flies are voided over a period beginning in March and extending into August, when the current year's crop of 3rd instars begin to appear. Acquisition of infections of horses with G intestinalis is nearly a year-round process, except for April, because of the persistence of viable larvae in eggs for a prolonged period after development and fly activity has ceased in the fall. In contrast, infection with G nasalis is interrupted between December and May, because eggs hatch naturally after a short period of development, and fly activity ceases in the fall.     

Cattle fly control using controlled-release insecticides

Field trials were carried out from 1980 to 1984 on the use of a controlled-release pesticidal fly control technique on farms in Sussex, England, with a recurrent infectious keratoconjunctivitis problem related to fly attack. Pesticide impregnated p.v.c. ear-tags provided control of biting and disease-carrying fly species throughout the season. Over 90% control of flies, compared with untreated herds, was attained using 2 tags impregnated with cypermethrin, fenvalerate or flucythrinate, although control using a single tag was unsatisfactory. In all cases infectious keratoconjunctivitis was almost completely eliminated.     

Effects of mid-season avermectin treatments on pyrethroid resistance in horn fly (Diptera: Muscidae) populations at three locations in Louisiana

Between 1999 and 2002, the effect of mid-season doramectin treatments on the level of resistance in pyrethroid-resistant horn fly populations was examined at three separate Louisiana State University Agricultural Center research stations. The cattle were treated with pyrethroid ear tags in all years at all farms, and each farm received a mid-season doramectin treatment in 1 year. The number of weeks of control at Red River was 11 weeks higher in the year following the mid-season treatment of doramectin. At Macon Ridge, the number of weeks of control was 2 weeks higher in the year following the doramectin treatment. No change was observed at St. Joseph. The LC50s for fly populations tested at Macon Ridge and St. Joseph were found to increase for pyrethroids from the spring populations to the fall populations between 2000 and 2002. The LC50s for fly populations at Red River followed the same trends except in 2000, the year when the doramectin treatment was administered. Flies collected pre and post-treatment each year from St. Joseph and Red River were assayed for two alleles (kdr and skdr) associated with target site resistance to pyrethroids. Flies collected pretreatment at Macon Ridge in 1999 also were assayed for the kdr and skdr, and this population of flies had a frequency of 85.6% R-kdr alleles. At St. Joseph and Red River there was a general decline in the frequency of homozygous susceptible skdr (SS-skdr) and homozygous susceptible kdr (SS-kdr) individuals, as well as a general increase in homozygous resistant skdr (RR-skdr) and homozygous resistant kdr (RR-kdr) individuals, during the 4-year study. At both sites, the frequency of R-kdr alleles increased significantly in flies collected in the fall compared to flies collected in the spring with the exception of Red River in 2000, when dormacetin was applied. The frequency of the R-kdr alleles was significantly higher in flies collected in the fall compared to flies collected in the spring in the following year at both sites in two out of three comparisons. The frequency of R-skdr alleles was significantly lower in fly populations tested in the spring compared to fly populations tested in the fall at both farms in years when doramectin was not applied but there were no differences in the years when doramectin was applied. The frequency of R-skdr alleles was significantly higher in fly populations tested in the fall compared to in the spring the following year during all three comparisons at Red River and in one of three comparisons at St. Joseph.     

Molecular investigation of transplacental and vector-borne transmission of bovine haemoplasmas

The present study was carried out in a herd with concurrent infections of Mycoplasma wenyonii and 'Candidatus M. haemobos', to investigate if transplacental and/or vector-borne transmission is possible for one or both bovine haemoplasma species. For this purpose blood samples were collected from 38 mother animals and their newborn calves; as well as from 17 uninseminated cows twice three months apart. In addition, 311 mosquitoes and blood-sucking flies (Diptera: Culicidae, Tabanidae, Muscidae) were cought near the animals. DNA was extracted from all samples, followed by real-time PCR analysis. In 10.5% of neonate calves, that were born to cows harbouring both haemoplasmas, M. wenyonii and/or 'Candidatus M. haemobos' positivity was detected. Copy numbers in positive samples from cows and their calves indicated that - in comparison with M. wenyonii - 'Candidatus M. haemobos'-bacteraemia had usually lower levels. In samples of uninseminated cows the rate of infection with the latter species decreased. These findings may explain why M. wenyonii was significantly more frequently detected in blood-sucking flies, than 'Candidatus M. haemobos'. In conclusion, molecular evidence is provided for the first time on the transplacental transmission of bovine haemoplasmas. Regarding their spread by blood-sucking arthropods, new potential vectors were identified, i.e. the horn fly (Haematobia irritans), the stable fly (Stomoxys calcitrans) and two species of horse flies (Tabanus bovinus, T. bromius).     

Cyromazine pour-on for the prevention of cutaneous myiasis of sheep

The results of trials on eight farms to assess the efficacy of two pour-on formulations containing cyromazine for the prevention of cutaneous myiasis of sheep are presented; data from trials on sheep with larval implants and on sheep kept in cages with adult flies are also reported. The incidence of cutaneous myiasis was reduced by between 87 per cent and 100 per cent for eight weeks when a formulation containing 6 per cent w/v cyromazine was used at an application rate of 60 to 85 mg of active ingredient/kg bodyweight. When a formulation containing 10 per cent w/v was used at an application rate of 50 to 100 mg/kg, the incidence of the condition was reduced by between 90 per cent and 100 per cent for eight weeks. Studies of sheep with larval implants, using the formulation containing 10 per cent w/v cyromazine at 50, 100 and 150 mg/kg bodyweight gave variable results with some animals at each dose rate having lost protection by the seventh week. When sheep were treated with the formulation containing 10 per cent w/v cyromazine at 50 or 100 mg/kg and exposed to adult flies in fly-proof cages they were completely protected for nine and eight weeks, respectively.     

Development of treated targets for controlling stable flies (Diptera: Muscidae)

Electrocution techniques were used to determine if treated targets similar to those used for tsetse control could be developed for stable fly control. In a series of two experiments, a half blue and half black (UK) 1 m2 target constructed of trigger cotton poplin was determined to be acceptable for development studies. In the first experiment, an average of 350 stable flies per hour (maximum 794 flies in 1 h) was collected using the UK target. A time-delayed circuit trial using untreated UK targets demonstrated that stable flies remained on or around the targets for at least 30 s. Two experiments were conducted with time-delayed circuits and UK targets treated with 0.1% lambda-cyhalothrin. In the first experiment, the number of flies collected using the 30 s on/off treated target treatment was not different from the number of flies collected using the other treatments. In the second experiment, the number of flies collected using the 30 s on/off treated target treatment was not different than the untreated target continuous or 30 s on/off treatments, but was significantly lower than the treated target continuous treatment. The number of flies collected with UK trigger targets was significantly higher than that for alsynite cylinder traps in two experiments. The mean number of flies collected during 22 1h assays using targets was 6.1-fold higher than that for alsynite traps, and the mean number of flies collected during 40 3 h using the targets also was 6.1-fold higher than that for alsynite traps. The results of this study indicate that treated cloth targets may be a viable addition for stable fly control programs.     

Claw trimming routines in relation to claw lesions, claw shape and lameness in Norwegian dairy herds housed in tie stalls and free stalls

We assessed the prevalence of claw lesions, abnormal claw shapes and lameness in relation to most-recent claw-trimming routines in Norwegian dairy herds housed in tie stalls and free stalls. Equal-sized groups were randomly sampled from both tie and free stalls in each of the three most animal-dense regions in Norway. The study population consisted of 2551 cows of the Norwegian Red breed housed in 54 tie stalls and 52 free stalls. Fourteen educated claw trimmers performed claw trimming and recording of claw lesions once during the spring of 2002. A multivariable model including cluster effects and individual-cow factors was fit for each claw lesion and abnormal claw shape.

In tie-stall herds with routine trimming 39.9% of the cows had one or more lesions or abnormal shapes in front or hind claws versus 52.8% in herds with no routine trimming. Hind-claw results in tie stalls with concrete stall base: herds trimmed occasionally had more haemorrhages of the white line (OR = 2.8) and corkscrewed hind claws (OR = 3.6) versus herds trimmed routinely; herds never trimmed had more heel-horn erosions (OR = 2.6) versus herds trimmed routinely and less haemorrhages of the white line (OR = 0.3) and the sole (OR = 0.2) versus herds trimmed occasionally.

In free-stall herds with routine trimming 76.8% of the cows had one or more lesions or abnormal shapes in front or hind claws versus 68.9% in herds with no routine trimming. Hind-claw results in free stalls with concrete stall base: herds never trimmed had less haemorrhages of the white line (OR = 0.3) and the sole (OR = 0.3) versus herds trimmed routinely; and also less haemorrhages of the white line (OR = 0.3) and white-line fissures (OR = 0.3) versus herds trimmed occasionally. Hind-claw results in free stalls with rubber-mat stall base: herds trimmed occasionally had less heel-horn erosions (OR = 0.5) and more dermatitis (OR = 5.4) versus herds trimmed routinely.

The routine claw trimming performed in Norwegian free stalls has not had the desired effects on claw lesions and abnormal claw shapes. Routine trimming in tie stalls, however, seems to have prevented claw disorders.     

Prevalence of lameness among dairy cattle in Wisconsin as a function of housing type and stall surface

OBJECTIVE: To determine the prevalence of lameness as a function of season (summer vs winter), housing type (free stalls vs tie stalls), and stall surface (sand vs any other surface) among lactating dairy cows in Wisconsin. DESIGN: Epidemiologic survey. ANIMALS: 3,621 lactating dairy cows in 30 herds. PROCEDURE: Herds were visited once during the summer and once during the winter, and a locomotion score ranging from 1 (no gait abnormality) to 4 (severe lameness) was assigned to all lactating cows. Cows with a score of 3 or 4 were considered to be clinically lame. RESULTS: Mean +/- SD herd lameness prevalence was 21.1 +/- 10.5% during the summer and 23.9 +/- 10.7% during the winter; these values were significantly different. During the winter, mean prevalence of lameness in free-stall herds with non-sand stall surfaces (33.7%) was significantly higher than prevalences in free-stall herds with sand stall surfaces (21.2%), tie-stall herds with non-sand stall surfaces (21.7%), and tie-stall herds with sand stall surfaces (12.1%). CONCLUSIONS AND CLINICAL RELEVANCE: Results suggest that the prevalence of lameness among dairy cattle in Wisconsin is higher than previously thought and that lameness prevalence is associated with season, housing type, and stall surface.     

High mortality in egg layers as a result of necrotic enteritis

A new facility was designed to hold 1.8 million birds in 10 houses; chickens were placed in five of the houses, and the remaining five houses were under construction when this outbreak occurred. An increase in mortality was reported in five houses; however, mortality in house 7 was quite high. Well-fleshed birds were suddenly found dead without a significant drop in egg production. The middle and distal intestines were distended with gas, congested, thin walled, atonic, and bluish or pale in color with sloughed mucosa in some places. Necrotic enteritis was diagnosed as the cause of increased mortality. The ingesta in the crop occasionally contained flies. The 4-wk mortality in house 7 was 6.55% with a loss of 10,898 chickens. The 4-wk mortality rate in the other houses ranged from 0.54% to 1.98%. The houses affected with necrotic enteritis were treated for coccidiosis with amprolium because low numbers of the oocysts were present in the intestinal specimens of some of the chickens. Household bleach was added to the water at a dilution of one part bleach to 1040 parts water to control bacterial contamination. The fly (Musca domestica) population was out of control. Clostridium perfringens was isolated from the alcohol-washed macerated flies caught from houses 4 and 7. Dead flies were often seen in the feed troughs. The chickens may possibly have had C. perfringens infection as a result of consumption of dead flies or their secretions/excretions. The alcohol-washed, macerated, clarified fly extract from the affected houses caused death in 11 inoculated mice and paralysis in one mouse. Similarly, illness and mortality were present in four mice inoculated with clarified intestinal contents. The bacterium isolated on anaerobic culture was identified as C. perfringens by polymerase chain reaction. The disease was brought under control after straw was added and mixed in with the litter. As a result, the litter temperature increased, causing a decrease in the fly population. This study suggests that flies in the poultry houses acted as mechanical transmitters of C. perfringens and that the development of necrotic enteritis was by ingestion of bacteria present in the flies and their secretions/excretions.     

Use of a real-time polymerase chain reaction-based fluorogenic 5' nuclease assay to evaluate insect vectors of Corynebacterium pseudotuberculosis infections in horses

OBJECTIVE: To develop and use a sensitive molecular assay for detecting the phospholipase D (PLD) exotoxin gene of Corynebacterium pseudotuberculosis in an attempt to identify insect vectors that may be important in transmission of clinical disease in horses. SAMPLE POPULATION: 2,621 flies of various species. PROCEDURE: A real-time polymerase chain reaction (PCR)-based fluorogenic 5' nuclease (TaqMan) system (ie, TaqMan PCR assay) was developed for the detection of the PLD gene in insects. Flies were collected monthly (May to November 2002) from 5 farms in northern California where C. pseudotuberculosis infection in horses is endemic. Three of the 5 farms (which housed a total of 358 horses) had diseased horses during the study period. A total of 2,621 flies of various species were tested for the PLD gene of C. pseudotuberculosis. RESULTS: Evidence of bacterial DNA for the PLD gene was detected in skin biopsy specimens from clinically affected horses and from 3 fly species collected from farms where affected horses were housed. Farms with a high incidence of diseased horses had a high proportion of insects carrying the organism. High percentages of flies with positive results for the PLD gene were observed in October, when most clinically affected horses were observed. CONCLUSIONS AND CLINICAL RELEVANCE: Our results are consistent with the hypothesis that C. pseudotuberculosis may be vectored to horses by flies. Three potential vectors were identified, including Haematobia irritans, Stomoxys calcitrans, and Musca domestica. The organism can be identified in up to 20% of house flies (Musca domestica) in the vicinity of diseased horses.