It's an ongoing discussion what the "best" way is, but in my opinion it's always a combination of various methods that leads to a appropriate estimation.

Besides the methods described in standard osteology books - here's a really good collection of relevant websites/literature on skeletal age estimation: http://theolduvaigorge.tumblr.com/post/18708065972/this-should-keep-you-busy-for-a-while-open

I agree with Michael. You can opine collectively considering various methods.

In addition to using different methods, I find it is also important to consider the specific skeleton and the information on the individual's life history this provides.

There is an issue of the American Journal of Physical Anthropology that has a set of papers devoted to aging and sexing of the skeleton. It is from 1985, Volume 68, issue 1.

I enjoyed reading other responses, and agree that using several methods is best. While the 1985 AJPA papers are good, newer methodology has been devised since then, while other methods (such as cranial suture ossification) have been de-emphasized in more recent literature (such as White & Folkens, 3rd edition - 2010). My two cents on the matter are: 1) It most strongly depends on the age group of the individual you're ageing, 2) How complete is the skeleton, as some areas tend to be more accurate than others, and 3) Whenever possible, the result of said ageing should be compared with the wider sample context of the site.

To expand on these points:

1) You wouln't check dental wear on a 5 year old, just like you wouldn't check dental development on a 40 year old. Every age group, give or take a few years, has a few ageing methods that specifically apply to it, just like it has others that are completely useless to apply. In the methodology section of my PhD (due next year), I discuss and discriminate which are best to prioritize at every life stage.

2) If you're working with forensic and definitely archaeological remains, you're bound to find yourself with a fragmentary mess, and in order to sensibly age any individual, it is good to know which methods carry more weight (again, at any life stage). I would always trust pubic symphysis morphology (I prefer the Suchey-Brooks system over the Todd system) over the auricular surface. Assuming I find only half a pubic symphysis, and I calculate 29-33 years at death, I would still place more emphasis on it, even if I had a complete auricular surface that looked as if the person was in their mid-40s.

3) What I mean about comparing the particular ageing methods in one individual over others at the site is this - Imagine that, you only have teeth on which to base your estimate for a particular individual and all other remains at a site are adults. You judge dental wear and estimate the person died in their early 20s because you're just beginning to see some dentin on the M2s. However, if you check out other skeletons from the site and notice that they have surprisingly unworn posterior dentition for their age, then you know you're likely underestimating age on the skeleton you're working on.

I emphasize #3 because back in 2007 I was working in China and found a skeleton with most epiphyses unfused but still with a dental age of around 29-33 years. Of course it turned out that it was a pathological case and the individual suffered from an endocrine disorder, but it has made me cautious about ageing ever since.

Hope this helps!

Environment and diet can influence tooth wear patterns, making it difficult to rely on general charts or scales, particularly when working with archaeological populations. Having many skeletons for comparison helps, as does using multiple methods on multiple points on the skeletons.

I still think that the best method to get an estimate of the appropriate adult age is serialization, using the technique of morphological changes in the auricular iliac facet.

Have you tried transitional analysis from Bolsen et al. 2002?

Please check American Journal of Physical Anthropology, september issue (vol 152, issue 1, 2013).

All the methods listed by Mauricio - and the fact that you need to weight some depending on the quality of remains and potential age stage - are great.

I've been working with an assemblage recently where it has been very difficult to age individuals with any accuracy, and I've sort of been wondering the point of it all. In many archaeological populations with a relatively low age at death (50-60s), I wonder how much difference it makes whether an individual was 35-50 or 50+ or any more specific range (although I'm reticent to go more specific than YA, MA, OA). For me to feel confident aging was actually adding to the body of data I'm collecting, I'd like to see some further evidence that being a slightly different age had any effect culturally on the individual's life.

(Kind of playing devil's advocate with the osteological paradox a bit.)

I agreed............

Recent study

Tooth abrasions and sutural aging methods are not reliable.