In Sensor Networks, the lack of topology information and the availability of only one communication channel has led research work to the use of randomization to deal with collisions of transmissions. However, the scarcest resource in this setting is the energy supply, and radio communication dominates the sensor node energy consumption. Hence, redundant trials of transmission as used in randomized protocols may be counter-effective. Additionally, most of the research work in Sensor Networks is either heuristic or includes unrealistic assumptions. Hence, provable results for many basic problems still remain to be given. In this paper, we study upper and lower bounds for deterministic communication primitives under the harsh constraints of sensor nodes.