RESUMEN
Transmission to a new host is a critical step in the life cycle of a parasite. Variation in the characteristics of the transmission process, for example, due to host demography, is assumed to select for different variants of the parasite. We have experimentally tested how variation in the time to transmission (early or late after infection) and exposure to adverse conditions outside the host (immediate or delayed contact with new host) interact to determine the success of the infection in the next host, using the trypanosome Crithidia bombi infecting its bumblebee host, Bombus terrestris. These two experimentally manageable steps mimic the processes of within- and among-host selection for the parasite. We found that early transmission led to higher infection success in the next host as did immediate contact with the new host. However, there was no interaction between the two parameters as would be expected if early-transmitted variants, resulting from rapid multiplication within the host, would be less adapted to the conditions encountered during the between-host transfer or infection of the next host. Furthermore, typing the genetic variability of the parasites with microsatellites showed that the four different transmission routes of our experiment selected for different degrees of allelic diversity of the infecting parasite populations. The results support the idea that variation in the transmission process selects for different genotypic variants of the parasite. At the same time, the relationship of allelic diversity with infection intensity suggested that the coinfection model of May and Nowak (1995) may be appropriate, where each parasite is able to infect and multiply independent of others within the same host.