Few discussions in geology can occur without reference to geologic time. Geologic time is often discussed in two forms:

Think of relative time as physical subdivisions of the rock found in the Earth's stratigraphy, and absolute time as the measurements taken upon those to determine the actual time which has expired. Absolute time measurements can be used to calibrate the relative time scale, producing an integrated geologic or "geochronologic" time scale.
It is important to realize that with new information about subdivision or correlation of relative time, or new measurements of absolute time, the dates applied to the time scale can and do change. Revisions to the relative time scale have occurred since the late 1700s. The numerically calibrated geologic time scale has been continuously refined since approximately the 1930s (e.g., Holmes, 1937), although the amount of change with each revision has become smaller over the decades (see fig. 1.5 and 1.6 of Harland et al.) and a few numerical estimates were available previously (but often for the duration of the entire scale rather than its individual subdivisions).

In addition, like any good scientific measurement, every dated boundary has an uncertainty associated with it, expressed as "+- x millions of years". These can not be included in the diagram for practical reasons, but can be found in Harland et al., 1990, along with a detailed description of the history of earlier-proposed time scales and the terminology, methodology and data involved in constructing this geological time scale.

Because of continual refinement, none of the values depicted in the table below should be considered definitive, even though some have not changed significantly in a long time and are very well constrained (e.g., the Cretaceous/Tertiary boundary has been at 65+-1 Ma for decades, and has been tested innumerable times, with almost all dates somewhere between 64 and 66 million years). The overall duration and relative length of these large geologic intervals is unlikely to change much, but the precise numbers may vary as a result of new data.

(c) 1996 macrae@geo.ucalgary.ca