Long-term changes in cloud opacity and cloud amount are indicative of potential responses and feedbacks to a changing global climate. A practical method of detecting changes in shortwave cloud transmission, distinct from cloud amount or IR effects, is developed and applied to several existing long-term (>20-year) data records. The difference between clear and cloudy solar irradiance has been widely used to investigate the effects of clouds on the surface energy budget. In those studies, several methods were used to objectively detect the presence of clouds by using either sophisticated remote sensing instrumentation or multiple simultaneous ambient irradiance observations. However, the available longer time series of total solar irradiance data exist in the absence of other side-by-side observations. The method presented here utilizes only the high-time resolution (1- to 3-min) surface irradiance data obtained from a single pyranometer to assess presence of cloud influences and then determine the extent of the cloud’s influence on the irradiance. Cloud presence is determined from the temporal variance and relative magnitude of the data while cloud transmission is found by using the ratio of observed irradiance to that that would be expected for clear skies. This ratio is interpreted as a total cloud transmission with the advantage that the measurement uncertainty due to the absolute calibration of the instrument is reduced because the same instrument is used to determine both clear and cloudy irradiances. This approach is particularly beneficial in the analysis of long-time series of pyranometer irradiance data where calibration histories are uncertain, which is typical. The cloud presence determination and the cloudy to clear irradiance ratios result in a long-time series of nearly instantaneous cloud transmission values. The method is applied to four 25-year irradiance records from globally remote and climatically diverse sites. The character and variation of the frequency-of-occurrence for binned cloud transmissions and temporal means are examined. A significant trend in annual mean cloud transmission at one of the sites is found and the fractal dimension of the frequency-of-occurrence for cloud transmission can be followed as potentially significant variations are detected. Indistinct boundaries between thin cloud and aerosol effects and their potential coexistence inhibit rigorous interpretation of long-term variations in these two quantities.