Effect of the Pietrain breed used as terminal boar on Post-weaning Multisystemic Wasting Syndrome (PMWS) in the offspring in four PMWS-affected farms

The aim of this study was to determine the potential risk factors for PMWS at the individual pig level and assess the effect of the Pietrain paternal genetic background of the animals in a cohort study. The survey was set up in four PMWS-affected farms with 2 repetitions (batches) per farm. A representative sample of 60 pigs per batch, stratified according to the paternal genetic background (Pietrain: yes vs. no), was randomly selected after farrowing. The representative cohort was divided into 8 batches and the pigs were individually monitored from birth to slaughter. Survival analysis was used to determine the factors related to the time to PMWS. The litter-cluster effect was taken into account using the marginal Cox model (robust estimation of the covariance matrix) and the gamma shared frailty model which were compared.No protective effect of the Pietrain breed on the time to PMWS and the proportion of affected pigs in the offspring was found in this study. Piglets showing low circovirus type 2 (PCV2) titres at 7 weeks-old with no subsequent seroconversion and piglets from PCV2 negative sows were most likely to be affected by PMWS (HR=7.0 and 2.8, respectively). Active infection of the pregnant dams with parvovirus was related to an increased risk of PMWS in the offspring (HR=2.3). Neck injuries due to poorly performed injections in the dams were associated with an increased risk of PMWS with the marginal model (HR=2.1). Oxytocin injection (dams) during farrowing was protective against PMWS in the offspring (HR=0.6).     

Some determinants of the spatio-temporal fire cycle in a mediterranean landscape (Corsica,France)

Based on recent needs to accurately understand fire regimes and post-fire vegetation resilience at a supra-level for carbon cycle studies, this article focusses on the coupled history of fire and vegetation pattern for 40 years on a fire-prone area in central Corsica (France). This area has been submitted since the beginning of the 20^th century to land abandonment and the remaining land management has been largely controlled by frequent fires. Our objectives were to rebuild vegetation and fire maps in order to determine the factors which have driven the spatial and temporal distribution of fires on the area, what were the feed backs on the vegetation dynamics, and the long-term consequences of this inter-relationship. The results show a stable but high frequency of small fires, coupled with forest expansion over the study period. The results particularly illustrate the spatial distribution of fires according to topography and vegetation, leading to a strong contrast between areas never burnt and areas which have been burnt up to 7 times. Fires, when occuring, affect on average 9 to 12% of the S, SE and SW facing slopes (compared to only 2 to 5% for the N facing slopes), spread recurrently over ridge tops, affect all the vegetation types but reburn preferentially shrublands and grasslands. As these fire-proning parameters have also been shown to decrease the regeneration capacity of forests, this study highlights the needs in spatial studies (both in terms of fire spread and vegetation dynamic) to accurately apprehend vegetation dynamic and functionning in fire-prone areas.This revised version was published online in May 2005 with corrections to the Cover Date.     

A dynamic simulation model for powdery mildew epidemics on winter wheat*

A system dynamic model for epidemics of Blumeria graminis (powdery mildew) on wheat was elaborated, based on the interaction between stages of the disease cycle, weather conditions and host characteristics. The model simulates the progress of disease severity, expressed as a percentage of powdered leaf area, on individual leaves, with a time step of one day, as a result of two processes: the growth of fungal colonies already present on the leaves and the appearance of new colonies. By means of mathematical equations, air temperature, vapour pressure deficit, rainfall and wind are used to calculate incubation, latency and sporulation periods, the growth of pathogen colonies, infection and spore survival. Effects of host susceptibility to infection, and of leaf position within the plant canopy, are also included. Model validation was carried out by comparing model outputs with the dynamics of epidemics observed on winter wheat grown at several locations in northern Italy (1991–98). Simulations were performed using meteorological data measured in standard meteorological stations. As there was good agreement between model outputs and actual disease severity, the model can be considered a satisfactory simulator of the effect of environmental conditions on the progress of powdery mildew epidemics.     

A model estimating the risk of Fusarium head blight on wheat*

A dynamic simulation model for the risk of Fusarium head blight on wheat was elaborated based on systems analysis. The model calculates a daily infection risk based on sporulation, spore dispersal and infection of host tissue of the four main species causing the disease (Gibberella zeae, Fusarium culmorum, Gibberella avenacea, Monographella nivalis). Spore yield and dispersal are calculated as functions of temperature, rainfall and relative humidity, while the main factors affecting the infection rate are temperature, wetness and the host growth stage. The model also calculates a risk for mycotoxin production by G. zeae and F. culmorum in the infected head tissue. First validations against field data, collected in some wheat‐growing areas in northern Italy and not used in model elaboration, produced satisfactory results.     

Estimating the potential development of Diaporthe helianthi epidemics in Italy*

Diaporthe helianthi is the causal agent of a severe sunflower disease but, in Italy, disease outbreaks are sporadic with no significant losses. The present work investigates the role of meteorological conditions on the potential development of D. helianthi epidemics in Italy, using the French model Asphodel, which simulates the effect of air temperature, relative humidity and rainfall on ascospore maturation and dispersal, infection establishment, disease onset and severity during the period of host susceptibility. Meteorological data measured in eight stations distributed from north to south Italy, over a 5‐year period (1995–99), was used as model input. Results showed that meteorological conditions in Italy are frequently favourable for D. helianthi infections on sunflower, and severe epidemics are possible. Therefore, climatic conditions are not a limiting factor for disease development in the Italian sunflower‐growing areas. The lack of disease epidemics in Italy may be related to differences in the pathogen populations compared with the French ones.     

Response of snails and slugs to fragmentation of lowland forests in NW Germany

Habitat fragmentation is a major cause for species loss, but its effect on invertebrates with low active dispersal power, like terrestrial gastropods, has rarely been studied. Such species can not cross a hostile habitat matrix, for which the predictions of island theory, such as positive relations between species richness and patch size, should apply. In order to test this prediction, we studied gastropod species diversity by assessing gastropod assemblage characteristics from 35 sites in 19 fragments of deciduous old-growth forests in the Lower Rhine Embayment, Germany. Assemblages differed between larger (≥700 ha) and smaller forests (<400 ha), those of large forests held a higher percentage of forest species. Although α-diversity was similar between the two forest size classes, small forests often comprised matrix species, resulting in a higher β-diversity. Edge effects on the species richness of matrix species were noticeable up to 250 m into the forest. Hierarchical partitioning revealed that distance to disturbances (external edge, internal edges like roads) explained most assemblage variables, whereas forest size and woodland cover within a 1 km radius from the sites explained only a few assemblage variables. Densities of two forest-associated species, Discus rotundatus and Arion fuscus, decreased with forest size. Yet, forest size was positively correlated with richness of typical forest species and densities of Limax cinereoniger. The latter species seems to need forests of >1,000 ha, i.e., well above the size of most fragments. In conclusion, the prediction is valid only for forest species. The response to fragmentation is species specific and seems to depend on habitat specialization and macroclimatic conditions. Jean-Pierre Maelfait: Deceased.     

Cytotoxic capabilities of bovine lymphocytes after lectin stimulation

Bovine peripheral blood mononuclear leukocytes (PBML) were stimulated in vitro with the mitogenic lectins concanavalin A (Con A) and phytohemagglutinin. Their cytotoxic capabilities were evaluated in a 51Cr release assay. Lectin-activated bovine effector cells did not mediate antibody dependent cellular cytotoxicity (ADCC) nor direct killing against cultured tumor target cells. Nevertheless, activation of PBML with lectins consistently generated effector cells able to mediate lectin-dependent cellular cytotoxicity. Cultivation of Con A stimulated-PBML for 3 to 4 weeks in the presence of lymphokines-containing IL-2 generated cells with the ability to mediate lysis without using Con A-coated target cells. However, cytotoxic cultures capable of mediating direct lysis of target cells were not able to mediate ADCC.