I did manage to dig up the full paper. [my notes in brackets]
Geochronology of the Carboniferous ,Permian, and Triassic - Forster, Warrington 1985
mem.lyellcollection.org/content/10/1/99.full.pdf
For starters, thank you for this. I kept trying to get it from the lyell collection, but it always asked me to pay; now I can actually read the paper in full as well. Thanks!

For what it's worth, this is kind of the discussion I was
hoping we could have, actually getting into the meat of the papers and trying to form an understanding of what's going on there.
So, just for reference, here are the exclusion criteria:
1. The stratigraphic position of the dated rocks or minerals must be known. Ideally, samples for radiometric dating should be from extrusive igneous rocks intercalated with fossiliferous sediments which are precisely dated by independent biostratigraphic means. Such rocks are, however, often altered and may have suffered partial loss of radiogenic daughter products. Intrusive igneous rocks, as well as reworked volcanics, metamorphic rocks and most detrital sediments, give only maximum or minimum ages but tend to yield more reliable radiometric data because they are more likely to have formed stable, closed isotopic systems.
2. The radioisotopic data should be as precise and unambiguous as possible. Ages should be derived from completely undisturbed systems and be confirmed by isochron techniques or by a number of separate conventional K-Ar, Rb-Sr or fission track determinations. If the chemistry of the rock allows, several different dating methods should be used on rocks and minerals from the same geological environment. The data should be internally consistent and should not be in conflict with the known geological sequence.
3. Complete analytical details should be published. It is important that the quality of any results can be assessed independently and that the original data can, if necessary, be recalculated using new techniques or regressions. Details of the decay constants used and the manner in which the errors have been assessed are essential.
4. The petrography of samples used for whole-rock determinations should be described so that the suitability of the samples can be independently assessed. In the case of determinations from minerals, only those from species normally regarded as reliable age indicators (Dalrymple & Lanphere 1969; Faure 1977) should be used
So... what's the problem here? The first issue is sort of a no-brainer; if you're trying to date a specific event in history, you need the rocks from that event, otherwise you might end up dating rocks from literally any time period. The second is, again, a no-brainer: the authors were trying to get the most comprehensively accurate dates possible, and excluding ones that are missing data or for which
known problems are likely to play a role. The third one is simply a matter of being able to
check the data. The fourth is, again, a matter of precision. So... this can't be the issue here.
The majority of the published ages have been discarded because of lack of analytical
information or lack of stratigraphical constraint, or because of the obviously
discrepant nature of the data presented. Single, unconfirmed, conventional age
determinations have frequently been used to provide key marker points on a radiometric
time-scale; such data are useful in terms of geochronological reconnaissance but cannot, until confirmed by further research, be used to define points on a time-scale.
So many of them were rejected because the raw data was not published along with them. This is entirely reasonable if you're trying to form a comprehensive meta-analysis. (The unbolded there is a sentence you left out which I think is fairly relevant). In fact, looking through that section a little further, what you see a lot of is them criticizing other metareviews for including such data. I'm not much of a science historian, and it's hard to find details on anything but the most up-to-date information, but it wouldn't surprise me if the field made immense strides in the 20 years between 1965 and 1985, just like it wouldn't surprise me if it had made significant strides in the 30 years between 1985 and today.
Conclusions:
It is clear from the foregoing discussion that the data available for each of the
periods under consideration show immense variation in both quantity and quality...
Well, they kinda do. I'm not sure about geology, but in medicine, there's always a
ton of research that is, well, absolute crap. 30-person pilot studies on the effects of things we know don't work that somehow come up with positive results anyways, and then are touted by quack organizations, are fairly common and depressingly useful for fooling a lay audience who doesn't understand how such studies are analyzed. That's
why they use the criteria above to filter out research they cannot verify effectively.
It is interesting to note that at the time of this publishing, 1985, the field of Geochronology and the modeling of a major portion of "geologic time" was admittedly in poor condition, with only 9% of relevant publications even considered worthwhile of review. Are we really supposed to believe that by roughly a decade or two later these models had become 'beyond dispute', as they are characterized today? Is that true? Or did the Evolution industry's public relations/propaganda wing just get increasingly more persuasive?
I don't know about Geochronology. Finding the
history of a scientific discipline is something I'm really quite
terrible at, and it's generally easier to find the current, the cutting edge, the
modern stuff, because that's usually the stuff that matters the most. However, I can say that in a related field, Climatology, within 20 years we went from a portion (admittedly, a small one) of the scientific literature predicting a coming ice age to a massive disciplinary consensus formed around where the earth was heading and what was causing it. We went from a field with less than 100 published papers per decade to one with thousands published every year. Science, as a rule, moves forward, with iteratively better and more accurate results. We absolutely should think that within 20-30 years, the conclusions drawn from an ever-increasing body of evidence would grow more and more robust.
"The stratigraphic position of the dated rocks or minerals must be known. Ideally,
samples for radiometric dating should be from extrusive igneous rocks intercalated with
fossiliferous sediments which are precisely dated by independent biostratigraphic means."
[Here it is stated unequivocally that the fossils are the ultimate indicator of how old a particular rock layer should be.
@RickG care to voice an opinion on this interpretation? I'm not sure that you're on point here. I'm running a little short on time right now and would like to finish this post; suffice it to say I'll come back to this.
Isn't it interesting how Forster and Warrington are reprimanding geochronologists for not including all their data? What have I been claiming this whole time?
Not quite. The data you're accusing them of excluding (entire dates) is not the same data Forster and Warrington are accusing them of excluding (specific details as to how the date was gathered and the radiological data involved in measuring it). It's the difference between them saying "this date doesn't work, let's chuck it" and them saying "we don't
really have to publish every detail of our methodology". Both are clearly problematic, let's be clear, but the former is a serious issue of systemic corruption, while the latter just makes the dates hard to analyze in a metareview.
With this in mind, shall we revisit two earlier quotes from the literature from different authors? Do they still sound like "quote mines" ?
"Finally, it is rarely a good solution to eliminate many apparently inferior results just because one date or one formation appears to be highly reliable. These remarks would appear to be somewhat obvious, but experience shows that these rules are frequently rejected." (Odin 1985 pp.42-43)
"The uncritical use of such dates is a good illustration of why one should not nail down chronostratigraphic schemes with singular dates, no matter how "excellent," while ignoring other, analytically sound, and equally acceptable data that may not agree with our preconceptions." (Haq et al. 1988 p. 602)
I don't know. I'd like to see the context first. There's probably more I should say here, and I will get back to this again later, but I'm late for my train.
On a more personal note, I feel this bears repeating: good on you for reading the paper and engaging it. I'm not sure I agree with your interpretation of the data, but this is a huge step up from making claims about it backed up by nothing more than the abstract. Thank you.
