2.1.2. Colony level infection dynamics
The temporal dynamics of N. apis infections in temperate climates have been described by many authors (White, 1919; Borchert, 1928; Michailoff, 1928; Bailey, 1955; Furgala and Hyser, 1969; Furgala and Mussen, 1978), with a similar pattern in both northern and southern hemispheres (Doull and Cellier, 1961). In short, the typical infection exhibits low prevalence during the summer, a small peak in the autumn, and a slow rise of infection during the winter. In the spring, infections increase rapidly as brood rearing starts while flight possibilities are still limited. There are few data on the temporal dynamics on N. apis infections in tropical or sub-tropical climates, but the infection appears to be present but with low impact on colony fitness (Wilson and Nunamaker, 1983) and probably without the pronounced temporal dynamics observed in temperate climates (Fries et al., 2003). For N. ceranae, few long-term studies have been performed on the temporal dynamics of this infection in honey bee colonies. Studies from central Spain suggest much less variation in infection prevalence over the season for N. ceranae compared to what has been described for N. apis (Higes et al., 2008a). However, a clear seasonal effect on disease prevalence, with higher prevalence and infection levels in the early season has been documented in eastern USA (Traver and Fell, 2011; Traver et al., 2012) and in untreated colonies in maritime Canada (Williams et al., 2010; 2011). There is an urgent need for long-term studies of the temporal dynamics of N. ceranae infections under different climatic conditions.