This paper focuses on the comparison of cloud amounts derived from an atmospheric general circulation model (AGCM), Satellite-observed clouds, and Ground-based cloud observations. This is distinctly different from Earth Radiation Budget Experiment (ERBE)-type comparisons because it does not mix potential errors in the cloud amount with those in the radiation code embedded in the model. Long term cloud climatologies were to compare global cloud amounts and regional seasonal cycles. The results obtained were surprising in many respects. The AGCM successfully reproduced the signatures of the warm pool and North Pacific seasonal cycle cloudiness but failed in the low stratus region off the coast of South America, a known problem for AGCMs. The data sets also reproduced the anomaly signature associated with El Ni¿o in the warm pool region, but the model amounts were lower. Global results had a similar success rate, with the model generally producing lower total cloud amounts to compared to the satellite and in situ measurements. Also, an attempt was made to compare cloud vertical distributions between the data sets. Because of the inherent differences in the measuring processes among the three data sets, the cloud height may need to be validated using the corresponding radiation fields. Unfortunately there were also some large discrepancies between the two observed cloud data sets. We conclude that the character of the observed cloud data sets, while tremendously improved over the last decade, must be substantially enhanced before they will be useful in validating AGCMs by any but the crudest levels of comparison. ¿ American Geophysical Union 1995