The effects of the intravenous injection of Clostridium perfringens type D epsilon toxin were compared in 18 goat kids and 10 lambs. This toxin produced neurological signs within 0.5-3 hours of intravenous injection in (1) all of 6 kids given doses of 250, 185 or 120 mouse lethal doses 50% (MLD50)/kg body weight, (2) 2 of the three kids given 60 MLD5ofkg, and (3) all of 5 lambs given 250 or 120 MLD50/kg. Six kids and 3 lambs given 45, 30 or 15 MLD50/kg, 1 lamb given 60 MLD50/kg, and 3 kids and 1 lamb given saline (controls) all remained clinically normal. Gross post-mortem changes were observed only in the kids and lambs that showed clinical signs. In the kids these changes consisted of severe acute interstitial and alveolar oedema of the lungs. However, only 2 out of 5 lambs that presented clinical signs showed pulmonary oedema. No histological changes were observed in the brain of any of the kids inoculated with epsilon toxin. In the brain of 4 out of the 5 lambs given doses of 120 or 250 MLD50/kg, there were histological lesions consisting of perivascular proteinaceous oedema and haemorrhages. These results show that goat kids and lambs are equally susceptible to the intravenous injection of epsilon toxin, but that they differ in the histological response of the central nervous system (CNS) to the toxin. The effects of intraduodenal administration of C. perfringens cultures and culture products in goats were evaluated in order to develop a reliable experimental model of enterotoxaemia in this species. Five conventionally-reared, 11 to 16 week old Angora kids were dosed intraduodenally with whole cultures of C. perfringens type D, while 5 similar animals were dosed with C. perfringens type D filtered culture supernatant and a third group of 5 kids were dosed with C. perfringens type D washed cells. Two kids were used as controls and received sterile, non-toxic culture medium intraduodenally. All animals received starch solution into the abomasum. All the kids (5) inoculated with whole culture and 3 of 5 dosed with culture supernatant and with washed cells developed CNS signs. Diarrhoea was observed in 2 of 5 kids inoculated with whole culture, in all (5) of those dosed with culture supernatant and in 3 of 5 of those that received washed cells. The most striking post-mortem findings consisted of lung oedema, necrotising pseudomembranous colitis and perivascular proteinaceous oedema of the brain. The protocol thus provided a reasonable model of naturally-occurring enterotoxaemia in goats, producing a range of clinical signs and post-mortem changes similar to that observed in the natural disease. Perivascular proteinaceous oedema was described in naturally-occurring clostridial enterotoxaemia in 2 goats. The 2 cases were selected from the files of two diagnostic laboratories in Australia, on the basis that the clinical syndrome and history were consistent with a diagnosis of enterotoxaemia, and that diagnosis was confirmed by epsilon toxin identification in intestinal contents. The brain lesions consisted of acute perivascular proteinaceous exudate in the brain, and in 1 of the goats, bilaterally symmetrical foci of encephalomalacia were observed. The histological demonstration of the brain lesions confirms that C. perfringens type D enterotoxaemia may produce histologically detectable brain changes in goats. Although this is not a consistent finding in goat enterotoxaemia, it is a valuable indicator of this disease when it occurs, as it is in sheep. Indirect and competitive ELISA techniques (I-ELISA and C-ELISA, respectively) were developed and their ability to detect antibodies to C. perfringens epsilon toxin in goat serum was compared. Different dilutions of a hyperimmune goat serum, in serum from a colostrum-deprived kid, were used as positive controls, while sera from 11 colostrum-deprived kids were used as negative controls. The epsilon toxin antibodies in the hyperimmune serum were also measured by mouse neutralisation test (MNT). The correlation coefficient between both the I-ELISA technique and MNT was 0.99, while the same coefficient for the C-ELISA was 0.98. Both the I­ ELISA and C-ELISA proved to be rapid, simple, sensitive and specific for detecting antibodies to C. perfringens epsilon toxin in serum of goats. A vaccination trial was performed with a commercial enterotoxaemia (pulpy kidney disease) vaccine, using 29 Angora goats. The animals were allocated to four groups (1 to 4) and at the beginning of the experiment (day 0), animals in groups 1, 2 and 3 received 1 dose of vaccine. Animals in groups 2 and 3 received a booster of the same vaccine at days 28 and 42 after the first vaccination, respectively. Group 4 remained as an unvaccinated control. All the animals were bled on days 0, 17, 28, 42, 59, 70, 86, 98 and 128, and the sera were processed for C. perfringens type D epsilon anti-toxin by an I-ELISA technique. A wide range of individual antibody titres was found after vaccination. A great majority of the animals in groups 2, 3 and 4 had serum antibodies below the protective level at day 98 after the beginning of the trial, and most animals in all the groups had serum antibodies below that threshold at day 128. These results agree with previous reports indicating that antibody titres in goats are short-lived and support the recommendation that goats must be re-vaccinated between 2 and 3 months after a first double vaccination. Another trial was performed using 20 goats to compare the antibody responses to epsilon toxoid vaccines prepared with different adjuvants, i.e. aluminium hydroxide (ALV), incomplete Freund's adjuvant (FAV) and liposomes (LIPV). The animals were allocated to 4 groups (1 to 4) at the beginning of the trial. Animals in group 1 were vaccinated with ALV, while animals in group 2 received FAV and those in groups 3 and 4 were vaccinated with LIPV. Animals in groups 1 to 3 received 3 doses of the corresponding vaccine at 3 week intervals, while those of group 4 received only 1 dose of vaccine at the beginning of the trial. A blood sample was obtained from all goats at the beginning of the trial and then weekly until the end of it, 7 weeks later. The samples were analysed for epsilon toxin antibodies by an I-ELISA technique. No major clinical abnormalities were observed in the animals after vaccination with the exception of those that received the FAV, which experienced a transient lameness that lasted for a few minutes. The highest and longest lasting antibody response was observed in the animals vaccinated with FAV, although they presented moderate to severe inflammatory tissue reactions at the injection site. Moderately high antibody responses were obtained with the ALV, with which only minor local reactions were observed. Almost no antibody response was obtained with the LIPV; no local reactions being observed. Enterotoxaemia in goats is mainly characterised by enterocolitis, and it has been suggested that the poor efficiency of commercial vaccines in preventing this disease is due to the local action of C. perfringens toxin/s within the intestine, where circulating antibodies might not exert their action. Five goat kids were vaccinated with an incomplete Freund's adjuvant C. perfringens type D epsilon toxoid vaccine on 3 occasions at 3 week intervals. Four similar additional kids were vaccinated with a commercial enterotoxaemia vaccine at the same times. Five other non-vaccinated kids were used as controls. All the animals were challenged intraduodenally 1 week after the last vaccination with C. perfringens type D filtered culture supernatant intraduodenally. At the time of challenge, the serum epsilon toxin antibodies in the Freund's-adjuvant vaccinated kids ranged between 2.45 and 230 International Units (IU)/ml, while the kids that received the commercial vaccine had anti-epsilon toxin levels which varied between 0.22 and 1.52 IU/ml. The epsilon anti-toxin levels of the non-vaccinated kids were below 0.03 IU/ml. No clinical or post-mortem alterations were observed in the kids that received the Freund's adjuvant-vaccine. Three of the four kids that received the commercial vaccine developed mild, pasty diarrhoea with a slight reddening of the colonic mucosa being the only post-mortem change observed. All non-vaccinated kids developed severe diarrhoea, respiratory distress and central nervous system signs, and were killed humanely between 6 and 24 hours after challenge. Post-mortem changes consisted of pseudomembranous colitis, lung oedema and perivascular oedema of the brain. Moderate to high serum levels of anti-epsilon antibody seem to protect goats against both the systemic and the intestinal effects of C. perfringens type D toxins. A polymerase chain reaction (PCR) was used to identify the genes encoding the major toxins of C. perfringens in faeces of goats. When pure cultures of C. perfringens types A, B, C, D and E were used as templates in the PCR, amplicons were observed on the agarose gel as bands at approximately the 247 (alpha primers), 1025 (beta primers), 403 (epsilon primers) and 298 (iota primers) bp level of the DNA marker. When used to identify different types of C. perfringens in samples artificially spiked with these microorganisms, the PCR detected as few as 1-1.5 x 102 CFU/g of the 5 types of C. perfringens tested. Gastro-intestinal contents and faeces were collected from 20 goats at slaughter and processed by PCR for the epsilon toxin gene alone. Several positive results were obtained from the first 5 goats that were slaughtered and sampled a few days after their arrival at the abattoir, but only a few samples gave positive results during the following weeks, after the goats had been fed a concentrated ration containing monensin. A possible role of this drug in control of enterotoxaemia is therefore suggested. The PCR technique allowed the identification and typing of C. perfringens strains in faeces of goats, without recourse to other techniques such as the MNT. Information about goat enterotoxaemia was extracted from a questionnaire that was mailed to goat farmers in Australia. Another questionnaire to collect information on frequency of enterotoxaemia diagnosis in goats was also sent to veterinary diagnostic laboratories throughout Australia. 366 (92.0 %) out of 398 owners said that they gave some sort of supplementary feeding, while 32 (8.0 %) said that no supplementary feeding was given to their goats. Out of 397 goat owners 235 (59.2 %) stated that they gave at least one dose of vaccine against enterotoxaemia during 1995, while 162 (40.8 %) did not vaccinate their goats during that year. One hundred owners (43.0 %) gave more than one dose of enterotoxaemia vaccine during 1995, while 132 (57.0 %) gave only one dose of vaccine during the same period. Only 4 owners (1.0 %) out of 381, considered that enterotoxaemia was the main cause of disease in the herd. The information collected from the diagnostic laboratories indicated that 701 goats were necropsied during the previous 3 years and that 48 cases of suspected enterotoxaemia were established. Only 10 (20.8 %) of these cases were confirmed as true cases of the disease. To improve the knowledge about the incidence of enterotoxaemia in goats, post-mortem examinations of all goat mortalities should be encouraged together with improvement of the diagnostic techniques presently available for enterotoxaemia. Farmers should vaccinate their goats against enterotoxaemia and the use of a single dose of vaccine is a practice that should be discouraged. An attempt was made to grow caprine and ovine endothelial cells in vitro, to challenge them with C. perfringens epsilon toxin, in order to compare their susceptibility to this toxin (Appendix B). No morphological alterations were observed in the cell cultures even after challenging them with doses as high as 1,200 MLD50/ml of epsilon toxin. An indirect immunofluorescence technique using FITC­ conjugated rabbit antisera to alpha actin and von Willebrand factor, was performed to confirm that the cells grown were in fact endothelial cells. It was found that approximately only 5 % of the cells were endothelial cells. Because pure cultures of endothelial cells could not be obtained, the results were not statistically analysed, since the target population was not obtained.