Calcium ions act as second messenger in many cell types. They transfer extracellular signals (e.g. from hormones) to targets within the cell, like Ca2+-dependent enzymes or transcription factors. Since a number of different effectors and cellular targets exist, it has been suggested, that specific information is encoded in the amplitude, frequency and waveform of the Ca-signal and decoded again, later on, by cellular targets. After stimulation, the calcium concentration in the cytosol of hepatocytes, for example, can display complex dynamic behavior including spiking and bursting oscillations. Using the information-theoretic measure Transfer Entropy (Schreiber 2000) we studied the properties of this signal transduction under different conditions. Therefore, we coupled a simple Ca2+-dependent enzyme activation process to a model of calcium oscillations (Kummer 2000) and to experimentally measured calcium time series. We simulated the system stochastically to account for random fluctuations in the case of low particle numbers. To approximate the rate of information transfer we analyzed the resulting time series for different levels of activation and different numbers of particles using kernel density estimation.