Methane hydrate is formed in a sand pack that undergoes cooling-heating cycles over a range of temperature. Five cycles are designed so that hysteresis can be observed in the sand pack. Each cycle has a different melting temperature which leads to varying intensity of temperature relaxation effect on the hysteresis. Evidence of hysteresis is observed in three separate temperature readings of thermocouples. Formation of hydrates is dependent on the thermal cooling rate of the sand pack, and the melting temperature of the previous cycle. A temperature increase is observed in the whole system, and this increase is driven by temperature peaks indicating significant hydrate formation near the thermocouples. These peaks have important effects on the whole system. By comparing each cycle's temperature peaks, hysteresis is clearly observed at the temperature readings of the short thermocouple. The same hysteresis pattern follows for the location of the temperature peaks. When significant hydrate formation occurs in the sand pack, a steepening of the pressure decline is observed, indicating a rapid loss of free gas in the system. The pattern that is observed in the temperature peaks is also identified in the pressure profiles, thus linking the gas saturation to hydrate formation. The time derivative of pressure corroborates these findings. A new model is proposed for the prediction of secondary hydrate formation time as a function of the melting temperature the porous medium experienced.