The purpose of this study is to diagnose the effect of condensation on propagating signals in the tropics in a relatively simple general circulation model based on a spectral primitive equation model. The study was performed in two stages: validating the model used by comparing long runs of this simple GCM with observational data, and then performing short runs of the model in perturbation form to study the response to a short-lived heating anomaly in the Indian Ocean. The tools used for an alysis are time-mean states of there presentative variables and Hovmöller diagrams and Wheeler-Kiladis space-time spectra in order to distinguish the propagating signals in time and to detect their spectral peaks in the wave number-frequency domain. During the study two aspects of realistic models were explored, using increasingly realistic basic states and adding physical processes which in this case was the latent heating due to condensation. The analysis of both effects was conducted by running the model with and without the condensation and by using different basic states for each experiment. As we progressed from the simpler to the more complex basic states it was observed that the condensation has an increasing effect. In the simplest experiment with a resting basic state two signals were present: a slower signal with a period of 12-13 days and a faster with a period of 7-8 days. In this case adding condensation heating made little difference to either signal,but then with a 3-d basic state the condensation heating amplifies the slower signals,and a period of 30-40 days also emerges,consistent with the behaviour of the simple GCM. The results presented indicate that both the condensation heating and the complexity of the basic state have an effect on the propagation of the signals. While further analysis is needed, the results lead us to the conclusion that rather than modifying the propagation characteristics gradually, condensation heating acts to select slower modes.