It is widely used in dating fossils or archaeological samples containing organic material such as wood, charcoal, bone, shells, etc.
Conversely, contamination by newer plant matter carried by flowing water or intruding plant roots may result in a date that is much too young. The original technique was based on counting the number of individual radioactive decay events per unit of time, using a device similar to a Geiger counter.
Archaeologists are acutely aware of these and other potential difficulties, and take extreme care in the selection and handling of objects to be dated. In the 1970s a new technique was developed called Accelerator-based Mass Spectrometry (AMS), which counts the number of carbon-14 atoms directly.
In this way, calibration tables have been developed that eliminate the discrepancy.
Despite its usefulness, radiocarbon dating has a number of limitations.
Carbon-14 dating cannot be applied to materials that have no C dates are less than that figure.
It is sometimes thought possible to extend the dating range a few half-lives, so one occasionally sees dates as old as 70,000 years or more.
(Since humans have only existed in the Americas for approximately 12,000 years, this is not a serious limitation to southwest archaeology.) Radiocarbon dating is also susceptible to contamination.
If the ground in which an object is buried contains particles of coal or other ancient sources of carbon, radiocarbon testing may indicate that the object is far older than it really is.
This can be done very accurately, although some samples may be difficult to work with.
Beyond this, the accuracy of the date depends on the reliability of the assumptions used in interpreting the measurements (see below).
Carbon-14 dates usually appear to be reasonably accurate whenever they can be checked against historical records.