Development of an animal repellent selection, efficacy and presentation

CSIRO PUBLISHING www.publish.csiro.au/journals/ajea Australian Journal of Experimental Agriculture, 2006, 46, 851 856 Development of an animal repellent selection, efficacy and presentation P. J. Murray A,C, A. C. Burns B and J. R. Davy B A School of Animal Studies, The University of Queensland, Gatton, Qld 4343, Australia. B PO Box 1500, Edinburgh, SA 5111, Australia. C Corresponding author. Email: peter.murray@uq.edu.au Abstract. Under some circumstances it may be desirable to repel herbivorous pests (e.g. goats and wallabies) from plants (e.g. horticultural or forestry seedlings) rather than to kill them. These circumstances may include using these animals as a resource at a later time, and there may also be welfare, public perception or legislative issues where repelling animals from particular plants or area is preferable to killing them. The first experiment tested the efficacy of 5 different materials (tiger fecal extract, maggot brew the liquid exudate from a flyblown goat, capsaicin, dog urine, and burnt goat hair, skin and carcass + fat) to repel groups of goats in a feedlot pen from 1 trough of feed when another trough of feed was available to them to eat. Each repellent was tested on 5 groups of mixed age female goats for periods of 3 days. All 5 groups were exposed to the 5 repellents. Both tiger fecal extract and maggot brew reduced intake of feed from the protected feed trough and significantly (P<0.05) delayed normal feeding behaviour by more than 5 h from that trough. The repellents became less effective with repeated use even with different groups of goats. The reduction in effectiveness was indicated by higher intakes of feed and earlier feeding from the protected trough. This habituation to the repellents, where the effectiveness is reduced with repeated exposure, is of concern. In the second experiment, we selected the most effective repellent (tiger fecal extract) and tested its efficacy under similar conditions, after the repellent had been mixed with a carrier (bentonite) in an endeavour to increase the duration of its effectiveness. This repellent was significantly more effective in repelling groups of 3 goats in a feedlot pen from 1 trough of feed for 3 days, when another trough of feed was available to them to eat. Additional keywords: animal repellent, feeding behaviour, feed intake, tiger fecal extract. Introduction Arnould et al. (1998) tested the repellent effects on sheep of dog, wolf, pig and sheep fecal odours, as well as pentane extracts containing the volatile constituents of pig, dog and wolf feces. Their experiments indicated that odours associated with dog and wolf feces have a higher repellent effect than those from pig or sheep. The test animals were bred from a line of animals that had had no previous experience with dogs or wolves for several generations. In a similar study, Abbott et al. (1990) reported the repellent effect of dung from the African lion (Panthera leo) on European red deer (Cervus elaphus). The products of digesting animal meat and fat and the resultant oxidative and microbiological breakdown were thought to generate a general carnivore odour. The repellence of lion dung is attributed to this general carnivore odour, as European red deer react to the smell of the African lion even though lions have not predated red deer for centuries. The response of goats (Capra hircus) to fecal odours of the Bengal tiger (Panthera tigris tigris), Siberian tiger (Panthera tigris altaica), African lion (Panthera leo), brown bear (Ursus arctos), the puma (Felis concolor) and the ungulates Dorcas gazelle (Gazella dorcas), the white bearded gnu (Connochaetes taurinus albojubatus), and conspecifics were studied by Weldon et al. (1993). Their experiments showed that feeding was suppressed where carnivore fecal scents were present. The above studies indicate that feces affect the foraging efficiency of an animal, and the effect is more pronounced when the feces are associated with a predator that invokes an innate avoidance response. The time that an animal avoids the area that it believes contains a predator may be measured in minutes to days, and this time is dependent on individual, social and ecological factors. Thus, a repellent based on predator fecal extracts has the potential to keep animals away from an area (containing plants), but only if the predator odour generates an innate avoidance response in the species being repelled. Urine extracts might be applicable but the research has shown both negative and positive results when used as a repellent (Nolte et al. 1994; Novellie et al. 1982). Fecal extracts from a large cat CSIRO 2006 10.1071/EA06004 0816-1089/07/060851

852 Australian Journal of Experimental Agriculture P. J. Murray et al. (e.g. tiger, lion, or leopard) would appear to offer a solution to the problem. The hypothesis tested in this study was that goats (as a herbivore model) would be repelled by tiger fecal extract, maggot brew the liquid exudate from a flyblown goat, capsaicin, dog urine, and burnt goat hair, skin and carcass + fat. The first experiment was undertaken to determine which of these were most effective in keeping goats away from feed when an alternative feed source was available, for what period the odour was effective, what behavioural responses were invoked by the repellent, and how quickly the animals habituated to the presence of the odours. The second experimental hypothesis tested the effectiveness of encapsulating tiger fecal extract (an effective repellent identified in the first experiment) in a granular form using bentonite to repel goats from an alternative source of feed in a feedlot to determine whether the encapsulation affected the efficacy of the repellent. Materials and methods The University of Queensland Gatton campus has a 180-steer feedlot that has 18 pens, 9 pens on each side of a central 3-m wide laneway. Each pen is about 22 by 12 m with a covered feed area (about 3-m wide) along the fence farthest from the central laneway. Six of the pens near 1 end of the feedlot were fenced to contain goats. Animals In both experiments a combination of mixed age feral and Boer cross does were studied. In the first experiment female goats were kept in the feedlot pens in groups of 6. Included in each group of goats were a mature feral doe, and 5 Saanen Boer does (3 weaners and a doe with a kid at foot). In the second experiment adult female Boer Saanen goats were used (3 adult does in each group). Feed In both experiments the feed offered was a chaffed mixture of lucerne and grassy hay. Feed was offered from 2 feed troughs where the total amount of feed offered was 4% of the combined liveweight of animals in the feedlot pen. That meant that more feed than sufficient to maintain liveweight was offered to the goats each day. Feed troughs were of identical size and construction. If the repellent were effective then the goats would only eat 1 trough of feed, i.e. half of the amount offered, which equated to 2% of their liveweight close to their maintenance level of intake. Repellents The repellents tested in experiment 1 were: (i) Capsaicin, which is an irritant to animals including the goat and, therefore, was tested as a repellent. Blairs 3AM Reserve (imported into Australia by Deathsauce Australia) contains 1 500 000 2 000 000 scoville units, as a result of the ingredients capsaicin (8-methyl-N-vanillyl-trans-6-nonenamide) and related capsaicinoids, and was used as a source of capsaicin. (ii) Maggot brew provided the distinctive, sweet off smell of the liquid component of a flyblown carcass. It was obtained from maggots, possibly of the fly species Chrysomya megacephala, C. rufifacies, C. saffranea, Lucilia cuprina, Musca domestica, M. vetustissima, and Sarcophaga spp., which had infected the head, liver, heart, lungs and 2 kidneys of an adult goat that was left outside under shade in a 76 L bin for 5 days, after which the liquid component (maggot brew) was collected. (iii) Tiger fecal extract was produced using the procedure by Hara and Radin (1978), where lipids and fatty acids are extracted from Panthera tigris sumatrae feces. In brief, fresh feces were placed in a large beaker, to which was added a mixture of hexanes and isopropanol (3 : 2, 3L), and the resultant slurry was stirred for 1 h. The bulk of the insoluble material (horse hair, skin, bone and plant material) was removed by filtration followed by washing of the solids with additional isopropanol. The filtrate was then filtered by vacuum through a bed of celite to remove insoluble particulate matter. The brown filtrate was concentrated in vacuo to remove the hexane and isopropanol. This resulted in a brown gummy liquid, which was then taken up into ethanol and reconcentrated to ensure that the only solvent remaining in the product was ethanol. A 31% dilution (with water) of this material was used in the experiments. (iv) Dog urine was collected by catheter from dogs kept within the Small Animal Clinic at the University of Queensland. These dogs had been eating a high protein diet. (v) Burnt goat hair, skin and carcass were produced by singeing, with a gas burner, the skin of a recently slaughtered goat such that both the hair and the skin were charred. The blackened remnants were then rinsed with 100% ethanol into a Winchester. Raw goat carcass was also burnt using the same process and the remnants rinsed with 100% ethanol into the same Winchester. In both experiments the repellents were poured into each repellent trough (150-mm PVC pipe, the same length as the feed trough, split longitudinally and capped at each end) that was hung 20 mm above the feed in the trough, at the start of each 3-day test period. In the first experiment, each time the repellents were applied to the repellent trough, 60 ml of solution was used. In the second experiment, 250 g of the encapsulated tiger fecal extract was used for each 3-day test period. Experimental design To determine the effectiveness of the 5 repellents, a Graeco-Latin Square design was used to balance (i) any effect of the preceding treatment (repellent) on the goats response to the repellent being tested, (ii) any effect of the pen in which repellents were being tested, and (iii) any differences in response by the 5 groups of goats. The goats were kept in the feedlot pens for a week prior to the start of the experiment and during this time were fed an equal mixture of chaffed lucerne and grassy hay at 4% of their combined liveweight. All repellents were offered to all of the pens of goats in a predetermined sequence, as it was possible that some of the groups could become habituated to new repellent smells, depending on the sequence that they were exposed to the repellents, so that they learnt, for example, to associate repellent smells with more feed. The goats behaviour was influenced by the pen in which they were confined (e.g. edge effect), hence, the groups of goats were moved for each 3-day treatment such that each group of goats spent time in each pen during the experiment. Two feed troughs were placed inside each pen away from the central laneway. The troughs, 1 containing feed and 1 with feed and a repellent trough containing the repellent, were a minimum of 8 m apart. At each change of treatment, the goats were removed from the pens and the feed troughs emptied of any residue and cleaned before new feed was added. In a similar manner, at each change of treatment the repellent troughs were cleaned before the next dose of repellent was added and the different repellent troughs that were used for each of the repellents were not swapped during experiment 1 but were moved to the appropriate pens. The efficacy of the repellent was determined as: (i) the reduction in the total amount of feed eaten in each pen; and (ii) change in feeding behaviour in each pen. Video tape recorders attached to cameras with infrared lights, were focused to record the activity of the goats in each pen at both feed troughs, the water trough and under the covered area at the end of each pen. The goats activity was recorded from the videotapes every 5 min

Development of an animal repellent Australian Journal of Experimental Agriculture 853 and this data was pooled hourly. The goats activities were classified according to their behaviour and location within the pen according to the following criteria: (i) number of goats feeding from the control trough and the trough protected with repellent; (ii) grazing within 4 m of the control trough and the trough protected with repellent; (iii) standing or sitting within 4 m of the control trough and trough protected with repellent; (iv) drinking from the water trough; (v) standing, sitting or lying under the shelter at the end of the pen; (vi) actively moving around the pen; and (vii) not in sight (outside of view from camera). The activities of the goats in the feedlot were analysed hourly to determine if there were subtle changes to their behaviour and the data from the time repellents were placed in the troughs until 2100 hours, when virtually all behaviours ceased, was aggregated. This data was analysed both as actual records of behaviour and as a proportion of the total behaviours recorded in the feedlot pens during those time periods. Statistical analyses of the data were performed using ANOVA in the GLM procedure of SAS version 8.2 (SAS Institute Inc., Cary, NC, USA). Animal ethics The experiments described in this report were conducted with approval from The University of Queensland Animal Ethics Committee under permit numbers SAS/421/02/R and SAS/039/05/R. Results No goats were injured or exhibited extreme adverse responses (e.g. extreme fear or flight response) to the repellents during the experiment. Experiment 1 A consistently strong goat group effect and a consistent trough effect were observed in the course of experiment 1. There was generally less intake of feed from the treated troughs (Table 1). Over the 3 days, tiger fecal extract (P = 0.0123), dog urine (P = 0.0484) and, to a lesser extent, maggot brew (P = 0.0678) showed a repellent effect in relation to the control. On a day by day basis, tiger fecal extract showed a repellent effect on days 1 and 3, dog urine on day 1 and maggot brew on day 3 (Table 1). No individual repellent was identified as having a significant effect on day 2, however, a general repellent effect was evident when the intake from all treated troughs on day 2 was averaged. As experiment 1 progressed, the goats ate more feed from both troughs, and left less feed residue. During Period 4 it rained overnight on 2 nights and this may have been the reason for the reduced feed intake and higher feed residues. The goats in all pens had, on average, between 8 and 9 feeding sessions from the trough containing repellents for each daylight hour during the first day of exposure to repellents in experiment 1. The goats exposed to maggot brew had fewer than 8 feeding sessions from the repellent trough for the first 6 h of exposure to the repellent. For goats exposed to tiger fecal extract, it took 5 h of exposure before goats had 8 or more feeding sessions from the trough protected by repellents. Goats exposed to capsaicin Table 1. Experiment 1. The average amount (kg) of chaff eaten by goats from the control trough or the trough protected by 1 of 5 repellents (treated) for each day, and the average intake over the 3days Repellent Trough intake (kg) Pooled Control Treated s.e.m. Day 1 Burnt goat 4.10 4.19 Capsaicin 4.20 3.92 Dog urine 4.28 3.87 0.136 Maggot brew 4.13 3.92 Tiger fecal extract 4.28 3.82 Day 2 Burnt goat 4.16 4.15 Capsaicin 4.25 4.01 Dog urine 4.10 3.80 0.125 Maggot brew 4.15 3.95 Tiger fecal extract 4.16 3.90 Day 3 Burnt goat 4.12 4.18 Capsaicin 4.19 4.11 Dog urine 4.05 3.92 0.105 Maggot brew 4.26 3.91 Tiger fecal extract 4.29 3.91 Days 1, 2 and 3 Burnt goat 4.13 4.17 Capsaicin 4.21 4.02 Dog urine 4.14 3.86 0.094 Maggot brew 4.18 3.93 Tiger fecal extract 4.24 3.88 and burnt goat had more than 12 feeding sessions from the repellent trough in the first hour of exposure to these repellents. There were significant (P<0.05) differences in the proportion of time goats spent at each feed trough in their pens and these differences were strongly influenced by the type of repellent used to protect 1 of the troughs (Table 2). The number of feeding sessions per hour during daylight hours changed during the day with peaks immediately after new feed was supplied to the goats, a reduced number of feeding sessions per hour during the middle of the day, and another period of increased feeding sessions from mid to late afternoon. Capsaicin was relatively ineffective as a repellent by both feed intake and behavioural measurements. Using capsaicin as a control to estimate normal feeding behavioural patterns, when feed was offered in 2 troughs, it was possible to see that the highest levels of feeding activity were immediately after new feed was supplied to the goats. There were a reduced number of feeding sessions per hour during the middle of the day and goats ate from both troughs consistently during daylight hours.

854 Australian Journal of Experimental Agriculture P. J. Murray et al. Table 2. Experiment 1. The proportion of the total behaviours of goats observed within 4 m of either the control trough or the trough protected with 1 of 5 repellents (treated) for each hour from 0900 to 2000 hours Time (hours) 0900 1000 1100 1200 1300 1400 1500 1600 1700 1800 1900 2000 Burnt goat Control 0.24 0.29 0.21 0.17 0.18 0.12 0.15 0.14 0.13 0.13 0.03 0 Treated 0.27 0.20 0.11 0.07 0.17 0.18 0.26 0.28 0.26 0.16 0 0 Capsaicin Control 0.26 0.17 0.19 0.11 0.16 0.17 0.16 0.17 0.15 0.11 0.02 0 Treated 0.26 0.21 0.13 0.11 0.17 0.18 0.27 0.23 0.23 0.23 0.05 0.01 Dog urine Control 0.30 0.32 0.26 0.27 0.19 0.19 0.24 0.20 0.24 0.14 0.01 0 Treated 0.16 0.11 0.12 0.09 0.12 0.13 0.22 0.23 0.17 0.09 0.04 0 Maggot brew Control 0.28 0.37 0.30 0.16 0.19 0.19 0.11 0.11 0.09 0.12 0.01 0 Treated 0.05 0.03 0.08 0.06 0.11 0.09 0.23 0.29 0.32 0.25 0.04 0.01 Tiger feces Control 0.34 0.27 0.28 0.20 0.23 0.16 0.12 0.13 0.11 0.14 0.01 0 Treated 0.08 0.12 0.08 0.11 0.13 0.18 0.28 0.35 0.32 0.17 0.01 0 It is apparent from the increase in the number of feeding sessions from the troughs protected by both tiger fecal extract and maggot brew that the efficacy of these repellents decreased the more times they were used. Examining the efficacy of the repellents within each 3-day period of exposure, it was apparent that the efficacy of repelling the goats from feed was reduced for every day of exposure to the repellents. During the 12 h period of time from when the repellents were placed in 1 of the feed troughs until dusk, when the goats became inactive, there were no significant differences (P>0.05) in the overall proportion of time goats spent feeding at each trough, standing and sitting within 4 m of the troughs, drinking or under the shelter at the end of the pen. However. during this period of time goats with either burnt goat or tiger fecal extract repellents in their pens spent significantly (P<0.05) more time moving around their pens or out of sight of the camera. Looking at the behaviours of the goats at the 2 troughs as a proportion of the total behaviours recorded indicated that both the maggot brew and tiger fecal extract repellents influenced feeding behaviour of the goats at the 2 feed troughs in each pen and their behaviour within 4 m of each of these 2 troughs (Table 2). The proportion of the goat behaviours at the 2 troughs were only significantly (P<0.05) different from 1100 to 1200 hours and 1600 to 1700 hours when burnt goat repellent was used. At no time during the period from when the capsaicin and dogs urine repellents were put out until dusk were the proportion of goat behaviours at the 2 troughs significantly different (P>0.05). The proportion of the goat behaviours within 4 m of each of the 2 troughs were only significantly (P<0.05) different from 1000 to 1100 hours, 1100 to 1200 hours and 1200 to 1300 hours when dog urine repellent was used (Table 2). At no time during the period from when the capsaicin and burnt goat repellents were put out until dusk were the proportion of goat behaviours at the 2 troughs significantly different (P>0.05). The proportion of goat behaviours within 4 m of each of the 2 troughs in the pens where maggot brew and tiger fecal extract was used were significantly (P<0.05) different for at least 6 h between 0900 and 2000 hours (Table 2). Experiment 2 Each group of does would have had a full gut when they started in the feedlot and, therefore, would not have been very hungry on the first day in the feedlot. The goats in replicate 1 escaped the first evening and therefore started days 1 and 2 with a full gut from grazing. There were significant (P<0.01) differences in the intake of feed from the 2 troughs for each day for 3 of the replicates. Because of the markedly different response of 1 of the goats (P90) in replicate 3 to the repellent, this doe was examined closely at the end of the experiment and was found to have a nose problem; the cartilage in her nose was soft and the nose

Development of an animal repellent Australian Journal of Experimental Agriculture 855 Table 3. Experiment 2. Summary of the goat feeding behaviour at the feed troughs containing tiger fecal repellents and their behaviour within 4 m of this protected trough Replicate Day Behaviour 1 1 No goats were observed within 4 m of the feed trough containing repellent. 2 No goats were observed within 4 m of the feed trough containing repellent. 3 One goat approached the feed trough containing repellent at 1053 hours sniffed once and left; 1653 hours sniffed once and left; 1737 hours sniffed once and left. 2 1 No goats were observed within 4 m of the feed trough containing repellent. 2 No goats were observed within 4 m of the feed trough containing repellent. 3 No goats were observed within 4 m of the feed trough containing repellent. 3 1 One goat (P90) approached the feed trough containing repellent at 0912 hours and fed intermittently until feed residues were collected, second goat fed briefly with P90, third goat did not go near feed trough (>3 m). 2 Goat P90 fed for 135 min (20 feeding sessions of 1 to 17 min duration); a second goat fed for 46 min (6 feeding sessions of 1 to 17 min); the third goat only nibbled feed once for less than 30 s. 3 Goat P90 fed for 95 min (19 feeding sessions of 1 to 13 min duration); a second goat fed for 99 min (19 feeding sessions of 1 to 13 min); the third goat fed for 93 min (16 feeding sessions of 1 to 11 min). 4 1 No goats were observed within 4 m of the feed trough containing repellent. 2 One goat was bullied by the other goats and was not allowed to feed from the control trough; it approached the repellent trough and sniffed at 1024 hours, 1659 hours, but did not feed. 3 The bullied goat was not allowed to feed from the control trough; it approached the repellent trough and sniffed at 0653 and 0732 hours then nibbled feed at 0918 hours (for 19 min) on the last morning of the experiment. The other 2 goats joined her at 0918 and 0925 hours and nervously fed until 0952 hours (the end of experiment). was progressively widening, which is indicative of cancer or a bacterial infection and doe P90 appeared to have either no sense of smell or greatly reduced sense of smell. The total amount of feed offered daily for the 4 groups ranged from 6.28 to 7.02 kg (because of differences in the cumulative liveweights of animals in each group). The percentage of the maximum amount of feed remaining in each trough each day for the 4 replicates indicates that for 3 of the 4 replicates the encapsulated tiger fecal extract was extremely effective in repelling the goats from the protected feed trough. The small percentage (<5%) of intake from the protected feed troughs in replicates 1 and 2, and days 1 and 2 of replicate 4 is thought to be an artefact of the feed drying in the sun rather than consumption by goats as the video recordings show that no feed from these troughs was eaten by goats. The records of the daily behaviour of goats in the feedlot within 4 m of the feed trough containing repellent are given in Table 3. In 2 of the replicates (1 and 2) goats did not feed from the feed trough protected by repellent for the 3 days that goats were in the feedlot and goats in 1 of these replicates did not go within 4 m of this feed trough. In replicate 3, with the goat with the presumed minimal sense of smell, the goats did not eat from the protected trough until the second day of the replicate, and 1 of the goats only ate briefly from the protected trough on the second day (Table 3). Discussion Experiment 1 The experiment demonstrated that it was possible to use either tiger feces extract or maggot brew to repel goats confined in a relatively small area from feed for a number of hours and in doing so affect their feeding behaviour and associated feed intake. On numerous occasions, particularly when the repellents were first placed in the protected feed troughs, the goats were clearly suspicious of the tiger fecal extract and maggot brew repellents. With tiger fecal extract behaviour was cautious (approach very slowly, sniff then leave quickly), whereas with the maggot brew the goat behaviour made it apparent that the repellent was offensive to their sense of smell (approach slowly, sniff, snort, shake head and leave quickly). There are a number of issues that this research has highlighted. These include some evidence that goats with close confinement and with both regular and relatively intense exposure will become relatively quickly habituated (within several weeks) to the repellents such that they become much less effective. It is not clear whether this is an issue with a once only or short-term exposure but may be a problem for repeated or long-term exposure. Another major issue was the smell of the effective repellents. Although the tiger fecal extract was not as offensive as the maggot brew, both were unpleasant to handle. It is likely that both could be sterilised to reduce any risk of bacterial effects, presumably without reducing their efficacy as repellents, but their smell has several problems. Use of these repellents requires an appropriate respirator, and there is the problem of storing and transporting the repellents and the risk that the repellents, when being used, may be attractive to non-target animal species. The smell of rotten meat and tiger feces may be a strong attractant a range of predators including tigers depending on where the repellents are used.

856 Australian Journal of Experimental Agriculture P. J. Murray et al. It was unexpected to find that burnt goat was ineffective as a repellent. One possible explanation may be that the amount of burnt material in the burnt goat repellent was relatively small and masked by the much greater volume of ethanol that was used as the carrier. It was also unexpected that dog urine would have limited effectiveness as a repellent given the predator prey relationship between dogs and goats. However, it is possible that a greater volume or a more concentrated solution of dog s urine may have been more effective. There was no evidence that the capsaicin was at all effective as a repellent. Experiment 2 It appeared, as evidenced by her lack of response to the repellents, that Doe P90, in Replicate 3, had minimal sense of smell. This was indicated by abnormal nose structure (determined after the experiment) and a minimal response to the repellent. Significantly, the second doe in replicate 3 would very tentatively follow her to the protected feed trough but the third doe only spent time eating from the protected trough at the end of the third day. This behaviour supports the idea that the goats did not like the smell of the repellent and will avoid protected feed. However, extreme hunger or in this case a doe that did not appear to be worried by the smell of the repellent resulted in other goats following her lead and eating from a protected trough. The results clearly indicate that mixing the tiger feces extract with bentonite, a product known to absorb and then slowly release organic compounds (Ashworth 1978), significantly increased the efficacy of the repelling effect of tiger fecal extract from less than half a day in experiment 1 to at least 3 days in experiment 2. The smell of the encapsulated tiger fecal extract was less offensive than the liquid form of the repellent, and was much easier to handle and distribute. Conclusions We have identified and, in a crude form, extracted material from tiger feces, which when encapsulated with bentonite was effective in repelling goats from an alternative feed source for at least 3 days. We believe that the extension in effectiveness was due to the volatiles in the repellent, which evaporated after a few hours in the first experiment, being released at a much slower rate in the second experiment, thus extending their effectiveness. The results of the 2 experiments appear to confirm this theory but further research is required. The study has focused on goats as a model herbivore. There may be differences in the efficacy of these repellents for other herbivores. Further studies are underway to identify the active ingredients that act as repellents and to determine if these ingredients are effective with other herbivores. Acknowledgments The authors acknowledge Drs Bhesh Bhandari, Peter Torley and Theresa Hay, who encapsulated the repellent in bentonite for the second experiment utilising the School of Land and Food Sciences laboratory at The University of Queensland. References Abbott DH, Baines DA, Faulkes CG, Jennens DC, Ning PCYK, Tomlinson AJ (1990) A natural deer repellent: chemistry and behaviour. In Chemical signals in vertebrates 5. (Eds DW MacDonald, D Muller-Schwarze, SE Natynczuk), pp. 599 609. (Oxford University Press: Oxford) Arnould C, Malosse C, Signoret J-P, Descoins C (1998) Which chemical constituents from dog feces are involved in its food repellent effect in sheep? Journal of Chemical Ecology 24, 559 576. doi: 10.1023/A:1022321104758 Ashworth J (1978) Reactions of Ammonia with Soil II. Sorption of NH 3 on English Soils and on Wyoming Bentonite. Journal of Soil Science 29, 195 206. Hara A, Radin NS (1978) Lipid extraction of tissues with a low-toxicity solvent. Analytical Biochemistry 90, 420 426. doi: 10.1016/0003-2697(78)90046-5 Nolte DL, Mason JR, Epple G, Aronov E, Campbell DL (1994) Why are predator urines aversive to prey? Journal of Chemical Ecology 20, 1505 1516. doi: 10.1007/BF02059876 Novellie P, Bigalke RC, Pepler D (1982) Can predator urine be used as a buck or rodent repellent? South African Forestry Journal 123, 51 55. Weldon PJ, Graham DP, Mears LP (1993) Carnivore fecal chemicals suppress feeding by alpine goats (Capra hircus). Journal of Chemical Ecology 19, 2947 2952. doi: 10.1007/BF00980594 Received 25 November 2005, accepted 4 April 2006 http://www.publish.csiro.au/journals/ajea