SUMMARY Spain is the leading producer of stone fruit and citrus of the European Union. The most important viral diseases associated with these crops are sharka in stone fruit, caused by Plum pox virus (PPV; Potyvirus) and citrus tristeza, caused by Citrus tristeza virus (CTV; Closterovirus); both are transmitted by aphids in a non-persistent and semi- persistent manner, respectively. Nowadays there is a vast amount of information about the epidemiology of both viruses in adult orchards, however, there are no studies on their epidemiology in nursery blocks. In addition, nurseries' specific features such as a small planting pattern and the presence of plants with juvenile characters and succulent shoots make controlling the natural spread of both these viruses more difficult. The main objective of this Thesis is to study the factors involved in determining the epidemiology of PPV and CTV in nursery blocks, in order to design viable control strategies. Moreover, bearing in mind that any control strategy requires reliable detection methods, serological and molecular methods have been validated for an early and accurate viral detection of infected plants in the nursery prior to the commercialization of plant material. Three experimental nursery plots were established in 2006 in different locations across the province of Valencia. The first nursery plot was laid out in the village of Llíria close to an adult Japanese plum orchard with a PPV-incidence of 90%; another one in Carlet in an area with a PPV-incidence of 5% and a third experimental nursery plot was done with the Citrus macrophylla citrus rootstock in Moncada, an area with an incidence of 85% of CTV. All three plots were sampled to determine the viral incidence in time intervals and analyzed by ELISA-DASI and Spot real-time RT-PCR for PPV and by Tissue print-ELISA in the case of CTV. The most susceptible Prunus rootstocks to natural infection by PPV in an environment of high inoculum pressure were Adesoto 101 and Mariana GF8-1. Nemaguard and Mirobolan 29C rootstocks showed a medium susceptibility, while Cadaman and Garnem were resistant to PPV-D having the virus detected only by real- time RT-PCR. Just Adesoto 101, Mariana GF8-1 and Nemaguard rootstocks were infected in the plot under low inoculum pressure. Furthermore, there was confirmation of the C. macrophylla rootstock's high susceptibility to natural infection with CTV. The dynamics of aphid populations were established based on samples captured with Moericke yellow traps and the sticky plants method. May turned out to be the month with the highest number of captures of aphid population in Prunus nurseries and in the C. macrophylla plot. The estimated average number of visitant aphids per plant in Llíria during May for two consecutive years was 38.1 while in the plot of Carlet was 36.7. Meanwhile, in the C. macrophylla plot, the number of visitant aphids per plant during May was 387.9. The dominant aphid species throughout the year by both sampling methods was Aphis spiraecola (over 90% in traps Moericke and approximately 55% in sticky plants), A. gossypii and Hyalopterus pruni were other species often found too. Percentage-wise, the number of PPV-viruliferous aphids determined by squash real- time RT-PCR which flew over and/or landed on the Prunus nursery plants in an area with a high inoculum pressure during May (30.32%) was over three times higher than those determined in an area with low inoculum pressure (7.98%). It was determined that the visits of the aphids to nursery plants and to the different Prunus rootstocks followed no pattern. In addition, no significant differences were found in the number of PPV- viruliferous aphids that visited the different Prunus rootstocks tested in both plots. It was concluded that there is no link between the number of aphid visitants and the leaf area of a nursery plant. 17.50% of the aphids which landed and/or flew over the experimental C. macrophylla plot were CTV-viruliferous. In Llíria's plot Nemaguard and Mariana GF8-1 rootstocks were used to evaluate the effects of the application of mineral oil to the viral incidence. The treatment consisted of the application of horticultural mineral oil (85% Ultrafine Sunspray [EC] P/V) at a rate of 1% every 10-12 days during the first year (2006) from the sprouting of plants to leaf fall. During the second year (2007), the treatments were made with the same oil every seven days during the spring's peak flight time of aphids. In the plot located in Carlet the treatments were evaluated with mineral oil in the six Prunus rootstocks studied. The treatments did not stop the viral presence in the experimental nurseries, but reduced significantly the PPV-incidence in the Mariana GF8-1 rootstock in an area with high inoculum pressure. The repeated application of mineral oil not only did not damage the vegetative development of the treated plants but it could also replace other pesticide treatments. In C. macrophylla plot the use of mineral oil did not reduce the incidence of CTV. The Spot and Tissue-print real-time RT-PCR techniques were validated and compared with ELISA-DASI and Tissue print-ELISA, respectively, for routinely use in detection. The coincidence in results between Spot real-time RT-PCR and ELISA- DASI was almost perfect for PPV detection and the coincidence between Tissue-print real-time RT-PCR and Tissue print-ELISA was, according to Cohen's kappa index, substantial for CTV detection. The diagnostic parameters calculated indicate that molecular techniques were more sensitive and are therefore recommended when very precise results in negative individuals (mother plants, material base and material for export) are required. However, serological techniques used showed greater specificity and are recommended for routinely inspection. The combination of serological and molecular techniques is recommended for almost total analysis accuracy. It has been proved that the Spot real-time RT- PCR technique can be successfully used in compound samples of up to ten plants without losing accuracy in detection of PPV throughout the year. By contrast, the ELISA-DASI showed significant differences in detection accuracy depending on the time of year and the number of plants analyzed being spring and summer the best times for detection of PPV followed by autumn and it is not recommended use of compounds in winter. Latent models have been applied for first time in plant pathology for estimating the different diagnostic parameters of Spot real-time RT-PCR and ELISA-DASI techniques. It has been shown that they can be applied to the estimation of the parameters for the diagnosis of new and existing validated techniques for detection of plant pathogens. As a general conclusion, this Thesis provides a basic understanding of the epidemiology of sharka and tristeza diseases in nursery blocks, which may be applied to the design of control strategies. Furthermore, it suggests the use of environmentally friendly control methods with zero residues such as the employment of less susceptible or resistant PPV Prunus rootstocks and the use of mineral oils. Also, the molecular techniques Spot real-time and Tissue-print RT-PCR have been validated for a reliable and accurate detection of PPV and CTV in nursery plants. The use of both these techniques provides a better guarantee for the virus-free status of analyzed plant material than other techniques employed so far.