Several specific questions about RadioCarbon Dating using AMS

I've been following the discussion but haven't contributed as my specialties within C14 dating are sample selection and interpretation of the results. I leave the mechanics to the physicists and other bright folks.But you'd be amazed at how complex, and how little understood, sample selection can be! As an example, in one area I work we have a lot of abalone and mussel shells in the archaeological sites. The oldest abalone shell dated so far is about 5900 BP (n=390), while mussel shells take you back beyond 9400 BP (n=403). What this means is if you select just abalone shells for your samples you are introducing a horrible bias and missing the oldest dates.

To answer Percy's question: almost all of the sites I work in have some degree of "bioturbation," which is disturbance by burrowing animals. The smaller items, such as mussel shell, can be moved about vertically. One site we examined had almost zero correlation between sample age and depth.But the abalone/mussel problem I mentioned in my previous post reflects a cultural phenomenon--it seems the prehistoric inhabitants did not start using abalone until about 5900 BP, while they began using mussels by at least 9400 BP. Hence, if you date abalone shells you simply can't identify the earliest components in these sites. This is just one of the problems that can come from poor sample selection.

Dating marine organisms of any kind can produce dates artificially old as they usually contain reservoir carbon, which has been isolated from the atmosphere.The source you cited was 1977, but since then there has been a lot of research done on marine shells and other marine organisms. One of the ways to correct for the reservoir effect is to date shells collected from the area you're working, but you have to have shells from the pre-atomic bomb era. From dating a series of shells of known ages you can come up with a calibration. A website with this information is http://calib.qub.ac.uk/marine/We test these calibration rates all the time. A few years back I came across a feature containing abalone, mussel, and charcoal in a context that showed they were contemporaneous. So, I dated all three materials and when all calibrations were applied the dates fell with a range of 14 years. That gives quite a bit of confidence in our local calibration figures.Another problem though is organisms that have both marine and terrestrial diets--including people and any dogs/coyotes that scavenge from their meals. Using the C14/N15 ratios obtained from the samples (usually bone in this case), we can estimate the percent of marine organisms in the diet and apply an appropriate correction for the reservoir effect.All of this is built into our main tool for calibrating or recalibrating dates--Calib. 7.1, available on the web at http://calib.qub.ac.uk/calib/calib.htmlAnother problem with dating shells is dating freshwater shells that may be absorbing carbon from old limestone or other mineral deposits which is in the groundwater. Those certainly can date too old. I have tended to avoid dating those entirely.