Control of Rhipicephalus microplus is predominantly mediated by the application of acaricides, but the rapid acquisition of resistance by this species and environmental pollution resulting from discarded acaricides, necessitates the discovery of new control measures. Due to the fact that Rhipicephalus spp. are genetically diverse and often have more than one host, it has been difficult to identify a common protective vaccine candidate able to target all species of this genus. Only one anti-tick antigen, Bm86, has been commercialized to date and is sold as GAVAC® and GAVACPlus® in South America. In an attempt to identify protective antigens, a protein termed subolesin was identified using expression library immunisation. RNAi studies showed that subolesin knockdown causes the degeneration of tick guts, salivary glands, reproductive tissues and embryos. Subolesin additionally mediates tick gene expression, impacts the innate immune response and affects tick infection by Anaplasma, Ehrlichia, Rickettsia, Babesia or Theileria spp. The R. microplus EF-1α homolog was identified as a subolesin-interacting protein via yeast two-hybrid and co-affinity purification experiments. RNAi experiments have suggested that EF-1α is another possible anti-tick vaccine candidate since it exhibits a similar phenotype as subolesin upon knockdown. The aim of the present research was to express R. microplus EF-1α in the yeast, Pichia pastoris and to exploit the yeast two-hybrid system in an attempt to identify its protein-binding partners. This will provide insight into understanding the translational machinery of this species and of ixodid ticks. Recombinant EF-1α was expressed as a 24 kDa protein, validated by western blotting. A highly representative cDNA library was produced from R. microplus mixed lifestages mRNA, fractionated and cloned into a two-hybrid prey vector. No definitive hits were obtained during the two-hybrid screen of reporter genes, as E-values attained after tblastx and PSI-BLAST analysis were higher than the required limit of 1 x 10-4. / Dissertation (MSc)--University of Pretoria, 2013. / Biochemistry / unrestricted

The control of ticks, and to some extent tick-borne diseases, over much of South Africa is currently dependent on acaricides and will probably remain so for the foreseeable future. Resistance to these chemicals by ticks thus poses a major threat to the livestock industry especially as these chemicals constitute an ever-diminishing resource with fewer being discovered and the cost of their development becoming prohibitive. In order to determine the extent of acaricide resistance in the eastern region of the Eastern Cape Province one-, two- and three-host ticks were collected from cattle at 58 dip-tanks over a period of 2 years. The one-host tick selected was Rhipicephalus (Boophilus) microplus, the two-host tick Rhipicephalus evertsi evertsi and the three-host tick Rhipicephalus appendiculatus. The ticks were tested for resistance to three compounds, namely amitraz, cypermethrin and chlorfenvinphos. The Shaw Larval Immersion Test detected emerging resistance to amitraz in the one-host tick R. (Boophilus) microplus at two dip-tanks and resistance at a third. It also revealed resistance in this tick to cypermethrin at one dip-tank, and emerging resistance to chlorfenvinphos at eight dip-tanks and resistance at two. The two-host tick R. evertsi evertsi was susceptible to amitraz and cypermethrin at all dip-tanks, but showed emerging resistance to chlorfenvinphos at seven dip-tanks and resistance at four. The three-host tick R. appendiculatus was susceptible to amitraz and chlorfenvinphos at all dip-tanks and demonstrated emerging resistance to cypermethrin at one. With the exception of R. (Boophilus) microplus, in which emerging resistance to amitraz was detected at one dip-tank by the Reproductive Estimate Test, all three tick species at all dip-tanks at which sufficient numbers of ticks had been collected, were susceptible to the three acaracides in both the Egg Laying Test and the Reproductive Estimate Test. Thus despite its fairly long and widespread use in the eastern region of the Eastern Cape Province very little or no resistance to amitraz was detected in three tick species regularly encountered on cattle in this region, namely R. (Boophilus) microplus, R. evertsi evertsi and R. appendiculatus. On the other hand resistance to chlorfenvinphos was detected in both R. (Boophilus) microplus and R. evertsi evertsi at a number of dip-tanks even though it, or other organophosphorous-based compounds, had probably not been used for tick control in the region for a number of years. The localities at which ticks were collected had already been mapped and the localities at which acaricide resistant ticks were encountered were mapped during this study. The three tick species that were targeted for acaricide resistance testing were widespread throughout the study region, but no pattern of geographic distribution for the acaricide resistant strains of these species that were detected, emerged. The rapidity of selection for acaricide resistance appeared to be closely related to the life cycles of the three ticks and the number of days that they theoretically would spend annually on their preferred host animals. Thus a greater number of acaricide resistant strains were encountered amongst the one-host tickR. (Boophilus) microplus and the two-host tick R. evertsi evertsi than the three-host tick R. appendiculatus. The fist two ticks both complete more than one life cycle a year and hence spend a longer time on their cattle hosts than the three-host tick R. appendiculatus, which completes only one life cycle a year and in addition is a rapid feeder in all its stages of development. To counter selection for acaricide resistance it is proposed that regular testing for resistance should be carried out, and that as soon as emerging resistance is detected in ticks on cattle at a particular dip-tank, that the acaricide in use at that dip-tank should be changed to a compound belonging to a completely different group of chemicals. Copyright / Dissertation (MSc)--University of Pretoria, 2009. / Veterinary Tropical Diseases / unrestricted

O estabelecimento de isolados de Rhipicephalus (B.) microplus resistentes a drogas é um passo fundamental para pesquisas sobre mecanismos de ação das drogas e desenvolvimento de novos acaricidas. Diferentes tratamentos de pressão de seleção foram efetuados sobre uma população de campo com fator de resistência (FR) à ivermectina (IVM) 1,37 visando obter experimentalmente uma cepa resistente. O tratamento das larvas com IVM não resultou em infestações de bezerros. O tratamento das adultas, por duas gerações consecutivas não alterou o valor do FR da população. A este último procedimento foi associado o tratamento do hospedeiro com IVM, resultando em aumento significativo do FR. Infestações utilizando larvas provenientes de posturas de alta eclodibilidade parece ter contribuído ainda mais para o aumento do FR observado na última geração. Em resumo, foi possível estabelecer pela primeira vez, em laboratório, uma cepa resistente à IVM que apresentou um aumento estatisticamente significante do valor do FR da última geração analisada (F10) em relação à inicial (F1). / The establishment of drug resistant Rhipicephalus (B.) microplus isolates is a fundamental step on researches about drugs mechanisms of action and on the development of new acaricides. Different selection treatments were effectuated upon a field population with an ivermectin (IVM) resistant ratio (RR) 1,37 aiming to obtain experimentally a resistant strain. The larval treatments with IVM do not reach calf infestations. The adult treatments, in two consecutive generations do not changed the population RR. To this last protocol was added the IVM host treatment, increasing significantly the RR. Infestations using larvae obtained from the high-eclodibility oviposition seemed to have contributed much more to the RR increase observed on the last generation. Finally, it was possible to establish for the first time, on laboratory, an IVM resistant strain that showed a significantly higher RR value from the last generation (F10) in relation to the first one analyzed (F1).

Host-parasite interactions are complex, involving host behaviour, host condition, and possible antagonism or mutualism between co-occurring parasites. The white-footed mouse (Peromyscus leucopus) is a small, abundant, generalist rodent, and the primary host for the black-legged tick (Ixodes scapularis) in its larval stage. Since larval ticks do not actively pursue their hosts (they wait for possible hosts to come to them), individual differences in mouse exploratory behaviour might be expected to correlate with tick burden. However, mice may also prevent tick attachment with defensive behaviour such as grooming. Yet, the relationships between tick parasitism and host exploratory and grooming behaviours have never been assessed. I analysed data from a capture-mark-recapture study conducted over 5 summers (2016-2020) in eastern Ontario, Canada, in which 1,035 mice were captured a total of 4,333 times. The probability of tick parasitism occurring and being more intense was significantly higher when the mouse was also parasitized by fleas, suggesting co-occurrence of these two parasites on host mice. A total of 510 mice were subjected to an open-field test to quantify exploratory and grooming behaviours. Exploration had a negative relationship with tick presence and a positive relation with flea presence. Most interestingly, there was a significant “tick × flea” interaction on exploratory behaviour such that fleas were positively associated with exploration only when ticks were absent. Surprisingly, there was no relationship between grooming behaviour in the open-field test and parasite presence, although grooming increased with mouse activity. This study shows that co-occurring parasite species (ticks and fleas) may interact to affect their host’s exploratory behaviour. Alternatively, individual differences in exploratory behaviour of hosts may differentially affect their susceptibility to being infested with ticks, fleas, or both. Future manipulative studies should use experimental design to determine the causality of the parasitism-behaviour relationships observed.

The tick size regime on systematic internalisers (SIs) was seen as a necessary action to level the playing field between SIs and other trading venues, with the hopes to improve the market composition and market quality in favour of regulated markets. However, the previous literature objects to the view as SIs may have gained the first-mover advantage from their previous tick size exemption. This thesis aims to examine whether the MiFID II tick size regime implementation for SIs on June 26, 2020, alters the market composition and improves the market quality at Nasdaq Stockholm. We consider 45 Swedish stocks with the highest daily average turnover to conduct difference-in-difference regressions. We find that the market quality worsens at Nasdaq Stockholm, while the market composition remains unaffected by the SI tick size regime implementation. The quoted spreads, effective spreads and price impact increase at Nasdaq Stockholm following the SI tick size regime. Impatient traders who trade on information may have rerouted their orders to Nasdaq Stockholm after the event since SIs can no longer offer the avoidance of the price-time priority. Therefore, SIs may have an important role in attracting informed traders who consume rather than supply liquidity. However, the interpretation of the results is conditioned on an inflow of traders from SIs to Nasdaq Stockholm, which we cannot explicitly measure due to the order flow not being adjusted for orders less than the standard market size.

The aim of present master’s thesis was to compare the results of serological methods for diagnosing borreliosis that are commonly used in Spadia laboratories (ELISA, immunoblots) in terms of recommendation on how and when to indicate and interpret them. The theoretical part is focusing on the characteristics and history of borreliosis, microbiological description of Borrelia, immune system and pathogenesis of the disease as well as the therapy and prevention. The experimental part is focusing on the analysis of results obtained from common examinations of antibodies to Borrelia made in Spadia Lab laboratories from January 1st 2014 to December 31st 2015. Screening of antibodies to Borrelia made by ELISA in IgM and IgG was done for all samples according to recommendation of CDC. In 2014 the ELISA screening was done using ELISA kits from Euroimmun and Evolis sample processors whereas in 2015 it was done using DiaSorin’s CLIA kits on Liaison analyser. Positive results were then confirmed by Westernblot or lineblot alternatively if the physician did not ask otherwise. It must be remembered that ELISA and Westernblot belong among serological methods that are using antibodies, i.e. substances produced by the immune system. The immune system plays the key role in protecting the body against infection and the antibodies are its important tool. Serological methods belong among immunoassay methods, which is still not standardized. Diagnosis of infections cann‘t be based only on antibody testing. It is necessary to assess the results in the context of the entire clinical picture, history and in the case of antibodies it is recommended retesting with an interval.

This observational study of 100 dogs naturally infected with Babesia rossi determined whether severity of parasitaemia was associated with outcome of infection and documented the relative distribution of parasitised red blood cells (pRBC) in capillary and venous circulation. The association between increased parasitaemias and outcome with a clinically compromised circulation was also investigated. Outcome was defined as either hospitalisation with death, or hospitalisation with eventual recovery or treatment as an outpatient. Dogs were enrolled if large babesias were found on stained thin capillary blood smears made from an ear prick. Thin venous smears were prepared from jugular or cephalic blood. Parasitaemias were manually counted and expressed as the percent pRBC. Ten dogs died, 50 recovered after hospitalisation and 40 were treated as outpatients. Venous sampling site did not affect venous parasitaemia (P = 0.6). Both capillary and venous parasitaemias of dogs that died were significantly higher than those of dogs that recovered after hospitalisation (P = 0.002) and dogs that were treated as outpatients (P < 0.0001). When assessing the whole group, capillary parasitaemia (median 0.61%, range <0.05-71.6%, interquartile range (IQR) 0.22-3.75%) was significantly higher than venous parasitaemia (median 0.14%, range 0-30.6%, IQR 0.046–0.52%) with P < 0.0001. The 21 dogs with a clinically compromised circulation were more likely to die (P <0.0001) and had significantly higher capillary (median 5.98%, range 0.09-71.6%, IQR 2.44-19.41%) and venous (median 2.81%, range <0.05-30.6%, IQR 0.17-9.03%) parasitaemias than the 79 dogs with a clinically normal circulation (capillary median parasitaemia 0.38%, range <0.05-12.87%, IQR 0.16-1.42%; venous median parasitaemia 0.096%, range 0-6.13%, IQR <0.05-0.33%; P < 0.0001). This study shows that high parasitaemia is significantly associated with death in B rossi infected dogs. Unfortunately, there was a wide overlap in the parasitaemias of the three outcome groups with the result that neither capillary nor venous parasitaemias appear prognostically useful. The previous clinical suspicion that capillary parasitaemias are usually higher than venous parasitaemias is confirmed. Thus capillary samples are the most appropriate diagnostic samples. Prior observations that a clinically compromised circulation is associated with death are confirmed. This association provides a rapid means of identifying patients in need of intensive monitoring and treatment. Despite the highly significant association between compromised circulation and higher parasitaemia, it is thought unlikely that parasite burden is the sole trigger for circulatory collapse. Copyright 2006, University of Pretoria. All rights reserved. The copyright in this work vests in the University of Pretoria. No part of this work may be reproduced or transmitted in any form or by any means, without the prior written permission of the University of Pretoria. Please cite as follows: Bohm, M 2006, The comparative assessment of capillary and venous Babesia rossi parasitaemias on thin blood smears and their association with disease manifestation, MMedVet dissertation, University of Pretoria, Pretoria, viewed yymmdd < http://upetd.up.ac.za/thesis/available/etd-05042007-154527 / > / Dissertation (Master of Veterinary Medicine (Small Animal Medicine))--University of Pretoria, 2006. / Companion Animal Clinical Studies / unrestricted

The pathology of canine cerebral babesiosis was examined at the gross, histological and ultrastructural levels. Gross lesions could be categorised as either global or regional. Congestive brain swelling , diffuse cerebral congestion and diffuse cerebral pallor were classified as global lesions. Multifocal haemorrhage and malacia were classified as regional lesions. Oedema was inconsistently present and could be either focal or diffuse. The majority of histological changes were observed in both cerebral babesiosis and control cases. Regional lesions were unique to cerebral babesiosis and had specific histological features. Highly localised endothelial injury was the primary lesion. Early lesions were multifocal and strictly associated with the microvasculature. Intermediate lesions, with perivascular haemorrhage and neutrophil infiltration, were suggestive of reperfusion injury. Advanced lesions were locally extensive and similar in appearance to haemorrhagic infarction. It is likely that the pathogenesis of regional lesions is by a process of microvascular infarction, as venous thrombosis could not be demonstrated. Ultrastructural evidence for adherent contact between erythrocytes and capillary endothelium was demonstrated. Endothelial cell necrosis occurred early in the development of lesions, before neuronal and glial injury. It is postulated that endothelial injury is the primary event in the development of regional lesions and secondary lesions develop as a consequence of microvascular infarction. / Die patologie van die serebrale vorm van bosluiskoors in honde is ondersoek. Die letsels is makroskopies, histologies en elektronmikroskopies beskryf. Letsels kon makroskopies in twee groepe verdeel word: Globale letsels en gelokaliseerde letsels. Kongestiewe brein swelling, diffuse serebrale kongestie en serebrale anemie kom voor as globale letsels in serebrale babesiose. Multifokale bloeding en nekrose kom voor as gelokaliseerde letsels. Edeem was nie konsekwent teenwoordig nie, en was algemeen of verspreid. Die meeste algemene histologiese veranderinge was in beide serebrale en kontrole gevalle teenwoordig. Gelokaliseerde letsels waarin spesifieke hisotpatologiese veranderinge voorgekom het, was kenmerkend van serebrale babesiose. Die primere letsel is hoogs gelokaliseerde beskadiging van endoteelselle. Beskadiging van die kapillere bloedvate ontstaan vroeg in die ontwikkeling van letsels. Verdere ontwikkeling van die letsel word gekenmerk deur peri-vaskulere bloeding en neutrofiel infiltrasie wat aanduidend is van reperfusie beskadiging. Volontwikkelde letsels is plaaslik-ekstensief en het die voorkoms van hemoragiese infarkte Dit is waarskynlik dat mikrovaskulere infarksie 'n rol speel in die patogenese van die letsels, aangesien veneuse trombose nie ontstaan nie. Noue kontak tussen rooibloedselle en kapillere endoteel is elektronmikroskopies bevestig. Endoteelselnekrose ontstaan voordat tekens van beskadiging geidentifiseer kan word in neurone of gliaselle. Dit blyk dat kapillere endoteelselbeskadiging die primere letsel by die ontstaan van gelokaliseerde lese Is is, en dat sekondere lesels ontwikkel as gevolg van mikrovaskulere infarksie. / Dissertation (MSc)--University of Pretoria, 2000. / Paraclinical Sciences / Unrestricted

Several studies have observed that stocks tend to drop by an amount that is less than the dividend on the ex-dividend day, the so-called ex-dividend day anomaly. However, there still remains a lack of consensus for a single explanation of this anomaly. Different from other studies, this dissertation attempts to answer the primary research question: How can investors make trading profits from the ex-dividend day anomaly and how much can they earn? With this goal, I examine the economic motivations of equity investors through four main hypotheses identified in the anomaly’s literature: the tax differential hypothesis, the short-term trading hypothesis, the tick size hypothesis, and the leverage hypothesis. While the U.S. ex-dividend anomaly is well studied, I examine a long data window (1975 to 2010) of Thailand data. The unique structure of the Thai stock market allows me to assess all four main hypotheses proposed in the literature simultaneously. Although I extract the sample data from two data sources, I demonstrate that the combined data are consistently sampled. I further construct three trading strategies: “daily return,” “lag one daily return,” and “weekly return” to alleviate the potential effect of irregular data observation. I find that the ex-dividend day anomaly exists in Thailand, is governed by the tax differential and is driven by short-term trading activities. That is, investors trade heavily around the ex-dividend day to reap the benefits of the tax differential. I find mixed results for the predictions of the tick size hypothesis and results that are inconsistent with the predictions of the leverage hypothesis. I conclude that, on the Stock Exchange of Thailand, juristic and foreign investors can profitably buy stocks cum-dividend and sell them ex-dividend while local investors should engage in short sale transactions. On average, investors who employ the daily return strategy have earned significant abnormal return up to 0.15% (45.66% annualized rate) and up to 0.17% (50.99% annualized rate) for the lag one daily return strategy. Investors can also make a trading profit by conducting the weekly return strategy and earn up to 0.59% (35.67% annualized rate), on average.

The most important tick-borne diseases (TBDs) occurring in Zambia that affect domestic animals, particularly cattle and small ruminants, are theileriosis (East Coast fever and Corridor disease), anaplasmosis (gall sickness), babesiosis (red water), and heartwater (cowdriosis). Of these, theileriosis is the most important, causing significantly more deaths than the other tick-borne diseases combined. Despite their importance, little is known about the occurrence and prevalence of haemoparasites in cattle in the communal areas of Zambia. Clinical signs and post mortem lesions are pathognomonic of mixed tick-borne infections especially babesiosis, anaplasmosis and East Coast fever (ECF). The main objective of this study was, therefore, to screen selected communal herds of cattle for tick-borne haemoparasites and identify the tick vectors associated with the high cattle mortalities due to suspected TBDs in the local breeds of cattle grazing along the banks of the Chambeshi River in Mungwi, Zambia. East Coast fever is endemic to the district of Mungwi, Northern Province, Zambia and vector control using acaricides has proved to be very costly for the small scale farmers. Also, Mungwi experiences increased cattle mortalities between December to March and May to July. All age groups of cattle are affected. A total of 299 cattle blood samples were collected from July to September 2010 from Kapamba (n=50), Chifulo (n=102), Chisanga (n=38), Kowa (n=95) and Mungwi central (n=14) in the Mungwi District, Northern Province, Zambia. Ticks were also collected from the sampled cattle from April to July 2011. DNA was extracted and the parasite hypervariable region of the small subunit rRNA gene was amplified and subjected to the reverse line blot (RLB) hybridization assay. The results of the RLB assay revealed the presence of tick-borne haemoparasites in 259 samples occurring either as single or mixed infections. The most prevalent species present were the benign Theileria mutans (54.5%) and T. velifera (51.5%). Anaplasma marginale (25.7%), Babesia bovis (7.7%) and B. bigemina (3.3%) were also detected in the samples. Nine percent of the samples tested negative for the presence of haemoparasites. In a number of samples (4%) the PCR products failed to hybridize with any species-specific probes but hybridized only with the genus-specific probes which could suggest the presence of a novel species or variant of a species. Of the four Theileria species known to occur in Zambia (T. parva, T. mutans, T. velifera and T. taurotragi), T. parva is the most economically important, causing Corridor disease in the Southern, Central, Lusaka and the Copper-belt provinces, while causing ECF in the Northern and Eastern provinces of Zambia. In our study, only one sample (from Kapamba) tested positive for the presence of T. parva. This was an unexpected finding; also because the tick vector, Rhipicephalus appendiculatus, was identified on animals from Kowa (14%), Chisanga (8.5%), Chifulo (6%) and Kapamba (1.4%). We can only speculate that the RLB hybridization assay may not have been able to detect the parasite in the animals sampled due to a too low parasitaemia. The samples should also be subjected to the T. parva specific real-time PCR assay to determine a more accurate T. parva prevalence in cattle in the Mungwi district, Northern Province. In Zambia, Babesia bovis and B. bigemina are recognized as being of economic importance in cattle. In our study, B. bovis was present in 7.7% of the sampled animals and B. bigemina in 3.3% of the animals. We detected B. bovis in all of the five sampled areas with the highest detection in Mungwi central (14.3%) and Kowa (10.5%). As expected, the tick vector Rhipicephalus (Boophilus) microplus was identified from animals from all of these areas. Babesia bigemina was only reported from Kowa (10.5%). The most abundant ticks identified from the sampled animals from Kowa were Rhipicephalus (Boophilus) decoloratus (36.3%) and Rhipicephalus (Boophilus) microplus (8.8%). These tick vectors have been implicated in the transmission of B. bigemina. Our findings are in concordance with results obtained by other authors who have speculated that an increase in the detection of B. bovis may indicate that B. bovis is becoming endemic in this part of the country. This could be due to uncontrolled movement of cattle that frequently occurs within Zambia. Heartwater (cowdriosis) is caused by Ehrlichia ruminantium, a rickettsial disease that affects domestic and wild ruminants. In Zambia, heartwater is mainly a disease of cattle, although outbreaks in sheep and goats have been reported and recorded. In our study, only one sample (from Kapamba) tested positive for the presence of E. ruminantium even though Amblyomma variegatum ticks were identified from 52.9% of the sampled animals from all study areas. The cattle sampled in our study are not regularly dipped and no game has been spotted in cattle grazing areas. It is possible that these cattle may have attained a state of endemic stability to heartwater. It is also possible that the RLB hybridization assay may not have been sensitive enough to detect E. ruminantium infections if the parasitaemia was very low. Samples should also be subjected to the E. ruminantium-specific pCS20 real-time PCR assay to determine more accurately the E. ruminantium prevalence in cattle in the Mungwi district, Northern Province. Anaplasma marginale (the causative agent of bovine anaplasmosis) has previously been shown to be present in all the provinces of Zambia and is the only Anaplasma species of importance to cattle in Zambia. In our study, 25.7% of the sampled cattle tested positive for A. marginale; it was detected in all areas except Chisanga. Amblyomma variegatum was identified from 52.9% of the sampled cattle, and Rhipicephalus (Boophilus) microplus from 12.1% of the cattle. Rhipicephalus (Boophilus) microplus has been incriminated as being a vector of A. marginale. Furthermore, three samples (from Kowa) tested positive for the presence of Anaplasma centrale. To our knowledge, no vaccination regime using A. centrale is being conducted in the Mungwi district of Zambia. The presence of A. centrale is, therefore, an interesting finding. The results of our study suggest that the cause of cattle mortalities in Mungwi during the winter outbreaks is mainly due to A. marginale, B. bovis and B. bigemina infections. This was confirmed by the results of the RLB hybridization assay, clinical manifestation of the disease in the affected cattle (own observation) and the tick species identified on the animals. It appears that in Mungwi, babesiosis due to B. bovis mostly infects cattle above one year of age. Calves appear to be less affected by B. bovis infection. There is need for further epidemiological surveys in Mungwi district, Northern Province, Zambia to get a better understanding of the epidemiology of these tick-borne haemoparasites affecting cattle. We conclude that integrated control policies should be developed to take account of multi-species pathogen communities that are commonly associated with clinical and sub-clinical TBD infections in Zambia. / Dissertation (MSc)--University of Pretoria, 2012. / Veterinary Tropical Diseases / unrestricted