Radiation from the sun is constantly striking the earth’s atmosphere. Every day, about 21 pounds of nitrogen is converted to Radioactive C14. That C14 then slowly decays to become normal nitrogen. Through tests in the lab the half-life of C14, that is the amount of time for half the C14 to decay, is about 5730 years. About .0000765% of the atmosphere today is made of C14. Since plants breathe the atmosphere it is assumed that they also contain .0000765% C14. All animals either eat plants or something that eats plants, so it is assumed that all animals also contain .0000765% C14. By measuring the amount of C14, scientists can measure backward and estimate the age of the object being tested.

Other than those mentioned above, there are some fundamental assumptions being made. If holes were poked in a bucket, 1 per inch, and water was poured into the bucket at a constant rate, the water would reach a point where the flow of water in would equal the flow of water out. This point is called equilibrium. The amount of time that it would take for the amount of C14 being produced to equal the amount of C14 that is decaying is approximately 30,000 years. If C14 hasn’t reached equilibrium yet, there would be more C14 today than 40 years ago. This means that if the earth is 6,000 years old, as indicated in many parts of the world, carbon dating would be inaccurate. The same problems apply to other forms of dating such as Potassium Argon. As you can see, carbon dating is not entirely scientific.

Here are a couple of wild dates that were obtained using radiometric dating methods.

Shells from living snails were carbon dated as being 27,000 years old. Science vol. 224, 1984, pp. 58-61

A freshly killed seal was carbon dated as having died 1300 years ago! Antarctic Journal vol. 6, Sept-Oct. 1971, p.211

"One part of the Vollosovitch mammoth carbon dated at 29,500 years and another part at 44,000.
--Troy L. Pewe, Quaternary Stratigraaphic Nomenclature in Unglaciated Central Alaska, Geological Survey Professional Paper 862 (U.S. Gov. printing office, 1975) p. 30.

"One part of Dima [a baby frozen mammoth] was 40,000, another part was 26,000 and the "wood immediately around the carcass" was 9-10,000.
--Troy L. Pewe, Quaternary Stratigraaphic Nomenclature in Unglaciated Central Alaska, Geological Survey Professional Paper 862 (U.S. Gov. printing office, 1975) p. 30

"The lower leg of the Fairbanks Creek mammoth had a radiocarbon age of 15,380 RCY, while its skin and flesh were 21,300 RCY.
--In the Beginning Walt Brown p.. 124

"A geologist at the Berkeley Geochronology Center, [Carl] Swisher uses the most advanced techniques to date human fossils. Last spring he was re-evaluating Homo erectus skulls found in Java in the 1930s by testing the sediment found with them. A hominid species assumed to be an ancestor of Homo sapiens, erectus was thought to have vanished some 250,000 years ago. But even though he used two different dating methods, Swisher kept making the same startling find: the bones were 53,000 years old at most and possibly no more than 27,000 years— a stretch of time contemporaneous with modern humans."
--Kaufman, Leslie, "Did a Third Human SSpecies Live Among Us?" Newsweek (December 23, 1996), p. 52.