A study was made of the effects of the heat-conducting paths from the temperature sensor to the sample and from the sample-temperature to the reference-temperature sensors on the determination of heat capacity in a temperature-modulated DSC (TMDSC) of the heat-flux type. These effects are usually assumed to be small. This is shown not to be so, and calibration procedures are developed. Using a simple model calculation, it is demonstrated that the phase angle difference between the sample temperature and the heat-flow rate should be taken into account when determining heat capacity. The results are experimentally verified for aluminum samples with various masses in quasi-isothermal experiments for a wide range of frequencies of temperature modulation for a Mettler-Toledo 820 DSC using no reference pan to simplify the analysis.