The following information was sent to me by e-mail:
Radiometric dating is predicated on the assumption that throughout the earth's history radioactive decay rates of the various elements have remained constant. Is this a warranted assumption? Has every radioactive nuclide proceeded on a rigid course of decay at a constant rate? This has been challenged by studies involving Carbon (C)-14.
At the temperature or pressure, collisions with stray cosmic rays or the emanations of other atoms may cause changes other than those of normal disintegration. It seems very possible that spontaneous disintegration of radioactive elements are related to the action of cosmic rays and the rate of disintegration varying from century to century according to the intensity of the rays. The evidence for a strongly increasing change in the cosmic ray influx is most favorable especially in light of the decay of the earth's magnetic field.
Most geochronologists maintain that pleochroic haloes give evidence that decay constants have not changed. Crystals of biotite, for example, and other minerals in igneous or metamorphic rocks commonly enclose minute specks of minerals containing uranium or thorium. The a-(alpha) particles emitted at high velocity by the disintegrating nuclides interact, because of their charge, with electrons of surrounding atoms which slow them down until they finally come to rest in the host material at a distance from their source that depends on their initial kinetic energy and the density and composition of the host. Where they finally stop to produce lattice distortions and defects there generally occurs discoloring or darkening. Each of the 8 a-particles emitted during the disintegration of U238 to Pb206 produces a dark ring in biotite. Each ring has its own characteristic radius in a given mineral (in this case biotite). This radius measures the kinetic energy, hence the probability of emission of the corresponding a-particle and also the half-life of the parent nuclide according to the Geiger-Nuttall law. The Geiger-Nuttall law is an empirical relation between half-life of the a-emitter and the range in air of the emitted a-particles. If the radii of these haloes from the same nuclide vary, this would imply that the decay rates have varied and would invalidate these series as being actual clocks. Are the radii in the rocks constant in size or are there variable sizes?
Most of the early studies of pleochroic haloes were made by Joly and Henderson. Joly concluded that the decay rates have varied on the basis of his finding a variation of the radii for rocks of alleged geological ages. This rather damaging result was explained away saying that enough evidence of correct radii for defferent geologic periods and sufficient variation in the same period have been obtained that one is forced to look for a different explanation of such variations as were observed by Joly.
Measurements were later made in an excellent collection of samples with haloes. It was found that the extent of the haloes around the inclusions varies over a wide range, even with the same nuclear material in the same matrix, but all sizes fall into definite groups. The measurements are, in microns, 5,7,10,17,20,23,27, and 33.
More recent studies have been made by Robert V. Gentry. Gentry also finds a variation in the haloes leading him to conclude that the decay constants have not been constant in time.
Gentry points out an argument for an instantaneous creation of the earth. He noted form his studies of haloes: "It thus appears that short half-life nuclides of either polonium, bismuth, or lead were incorporated into halo nuclei at the time of mica crystallization and significantly enough existed without the parent nuclides of the uranium series. For the Po218 (half-life of 3 minutes) only a matter of minutes could elapse between the formation of the Po218 and subsequent crystallization of the mica; otherwise the Po218 would have decayed, and no ring would be visible. The occurrence of these halo types is quite widespread, one or more types having been observed in the micas from Canada (Pre-Cambrian), Sweden, and Japan." The argument seems hard to refute.
So, then, careful scientists have measured variations in halo radii and their measurements indicate a variation in decay rates. The radioactive series then would have no value as time clocks.
The following quotation also suggests a cause for a change in the decay rate:
"An earlier increase [in neutrino levels] must have had the peculiar characteristic of resetting all our atomic clocks. This would knock our C-14, potassium-argon, and uranium-lead dating measurements into a cocked hat! The age of prehistoric artifacts, the age of the earth, and that of the universe would be thrown into doubt."
F. Johnson, *J.R. Arnold, and *R.F. Flint, "Radiocarbon Dating," in Science, February 8, 1957, p. 240. This is significant because it is known that neutrinos do interact with the nucleii of atoms, and it is also believed that much of the energy of supernovae is carried away by neutrinos.
Isochrons are an attempt to avoid the need for an absence of daughter element initially in computing radiometric ages. The idea is that one has a parent element, X, a daughter element, Y, and another isotope, Z, of the daughter that is not generated by decay. One would assume that initially, the concentration of Z and Y are proportional, since their chemical properties are very similar. Radioactive decay would generate a concentration of Y proportional to X. So we would obtain an equation of the form
Y = c1*X + c2*Z
By taking enough measurements of the concentrations of X, Y, and Z, we can solve for c1 and c2, and from c1 we can determine the radiometric age of the sample. A good general introduction to isochrons from an evolutionary perspective can be found at http://www.talkorigins.org/faqs/isochron-dating.html
Let's apply this to potassium argon dating, where X is K40, Y is Ar40, and Z is probably Ar36. If the concentration of K varies in a rock, that it is unlikely for the concentration of added argon 40 to vary in a way that will yield an isochron. But if the concentration of K does not vary, then one can still get an isochron if the concentration of the non-radiogenic isotope Ar36 of the daughter product varies. So let's call an isochron a "super-isochron" if the concentration of the parent element varies from one sample to another. Let's call it a "wimpy isochron" otherwise. The question is, what percentage of isochrons are super-isochrons, and how do their dates agree with the conventional dates for their geologic period? I would think that it may be rare to have a super-isochron. If one is dealing with minerals that exclude parent or daughter, then one cannot get an isochron at all. If one is dealing with minerals that do not exclude parent and daughter elements, then most likely the parent element will be evenly distributed everywhere, and one will have a wimpy isochron that cannot detect added daughter product, and thus may give unreliable ages. Whole rock isochrons may also tend to be wimpy, for the same reason. Even super isochrons can yield ages that are too old, due to mixings, however.
False K-Ar isochrons can be produced if a lava flow starts out with a lot of excess Ar40 which becomes well mixed, along with potassium. Then while cooling or afterwards, a mixture of Ar36 and Ar40 can enter the rock, more in some places than others. Other isotopes of argon would work as well. I believe that this will produce a good K-Ar isochron, but the age calculated will be meaningless.
There is another way that false isochrons can be produced. For a wimpy isochron, say a K-Ar isochron, we can assume that initially there is a uniform concentration of K everywhere, and concentrations of Ar40 and Ar36 that form an isochron. Then a lot of Ar40 enters, uniformly, through cracks in the rock or heating. This will retain the isochron property, but will make the isochron look too old.
My reasoning was that if the lava is thoroughly mixed, then the concentration of parent material should be fairly constant. If the concentration of parent substance is not constant, it could indicate that the lava is not thoroughly mixed. Or it could have other explanations. If the lava is not thoroughly mixed, it is possible to obtain an isochron from the mixing of two different sources, in which case the radiometric age is inherited from the sources, and does not necessarily yield the age of the flow.
Someone pointed out to me that many Rb-Sr isochrons are super isochrons. I find this information very interesting, and thank him for it. I'd be curious to know which strata they occur in, as my main interest is the geologic column of Cambrian and above. My impression is that these are not on this part of the geologic column. And how well do the dates correlate with others for the same formation?
There are also mixing scenarios that can produce even super isochrons having invalid ages. And geologists admit in any event that isochrons can sometimes give false ages.
Here is a mixing scenario for false isochrons. Consider this possibility: There are two sources of lava, A and B. Suppose these mix together so that at point 0 we have only A, at point 1 we have only B, and in between we have varying concentrations. Half way between there is a mixture of half A and half B, for example. Suppose X is a parent substance, Y is its daughter, and Z is a non-radiogenic isotope of the daughter. Suppose A has a little X and lots of Y and not much Z, all uniformly distributed, and B has some mixture of Y and Z, all uniformly distributed. Then this varying mixture of A and B, with all A at 0 and all B at 1, produces a good isochron. There is no way this mixture can be distinguished from a similar case in which A has lots of X and little Y, and B is the same as before, and a lot of time passes.
It is claimed that mixing can often be detected. If this is so, then the question remains, for super isochrons on the geologic column which can be shown not to be caused by mixing, how do they correlate with other methods, and with the expected dates for their geologic period?
My understanding is that isochrons measure the time since a rock was last well mixed. For a lava flow, this could be the time of the flow. Or it could be that several flows all come from the same well-mixed magma, and might yield a joint isochron giving the time of the flow. It seems to me that a single lava flow might not mix well, and thus the age obtained would be that of the magma and not the time of the flow. So this points out another problem with interpretation of isochrons.
I'm also curious to know how much of the geologic column is datable by super isochrons for which no mixing can be shown.
Atlantic sea floor dating
One often hears about K-Ar dates of the Atlantic Ocean bottom which increase from zero at the mid-Atlantic ridge to about 150 million years at the edges. This is taken as proof that the continents began separating about 150 million years ago. However, this can be explained by assuming that argon rises to the top of the magma, so magma deeper down looks younger. The magma deeper down would have come to the surface later, and thus would be nearer to the mid-Atlantic ridge. Or if the continents split quickly, the observed pattern of dates could be explained by a decreasing concentration of Ar40 in the water. In any event, I don't see how the lava in the center of the Atlantic could have a young age in the conventional view, since it would have cooled rapidly under a lot of water, and would have retained its argon, making it look old.
We now make some comments about dating the meteorites. Since I have not had as much time to study this, I will just list some points that must be considered. Many parent to daughter ratios for many meteorites give radiometric ages of about 4.5 billion years. This gives support to an ancient age for the meteorites, assuming constant decay rates. However, in interpreting these results, some facts need to be kept in mind. The first is that these results are not obtained by a simple parent to daughter ratio. Instead, some estimate of the amount of daughter initially present in the meteorite has to be made in order to compute a radiometric age. Thus one has a "fudge factor," and in fact, a different fudge factor for each method. So one has to be sure that these fudge factors are properly used, and not simply adjusted in order to obtain an agreement among the dates.
The importance of this is underlined by the fact that these same fudge factors are used to estimate uranium and thorium dates on earth. Thus the estimate of initial concentrations of lead isotopes could also affect the 4.5 billion year age computation for the earth.
We noted above that there also seems to be a fudge-factor built into potassium-argon dating, namely, the branching ratio estimate. This causes the correlation between K-Ar dates and other dates on meteorites to come into question, as well.
Now, at least for uranium-lead dating, a kind of isochron has been observed among five meteorites containing uranium and a number which do not, which gives a rational basis for assuming how much daughter product was present initially. See http://www.talkorigins.org/faqs/faq-age-of-earth.html . The obvious question to ask in regards to this is how the meteorites were chosen for this isochron, and whether there are other meteorites and other bodies from the solar system that do not fit. If so, this calls this interpretation into question. In addition, there is just one point on this isochron for all of the meteorites that do not contain uranium. Is this obtained by averaging, or do they all have exactly the same ratio of lead isotopes? If the former, then this could indicate that the points of this isochron have considerable scatter, further calling the age computation into question. A point from the earth is also on this isochron. This is from a sedimentary deposit. But since uranium is much more water soluble than lead, it seems questionable to use this point as reprsenting the ratio of lead isotopes on earth, since it may be impoverished or enriched in uranium. In addition, if other sediments yield different ratios of isotopes, why was only this one chosen? Another question that needs to be asked is whether this isochron could have been produced by some kind of a mixing process, since such processes can produce isochrons not representing a true age. It also needs to be determined whether the daughter products for methods other than uranium-lead dating also yield isochrons among the different meteorites.
The above discussion concerns dating techniques based on simple parent to daughter ratios. There are other dating techniques such as isochrons and discordia which avoid the need to estimate initial daughter product concentrations. Therefore, it should be determined how many correlations remain in meteorite dating when only such techniques are applied.
Of course, in the traditional view, the matter out of which the solar system was formed would have been very old at the start, in any event, and so the radiometric ages obtained from meteorites or from the earth do not necessarily tell us anything about the age of the solar system or the age of the earth.
My point is not to refute the meteorite dating, since it may be sound, assuming a constant decay rate. However, on seeing the lack of evidence for large-scale evolution, the many problems with radiometric dating on the geologic column, and the many plausible evidences for catastrophe which often seem to be interpreted away by science, I have become somewhat skeptical of any area of science having to do with origins, and so have come to question even the assumptions behind the dating of the meteorites.
An evolutionist said his experience is that whenever he looks into a creationist source, it blows up on him. My experience is that whenever I look into an evidence for evolution or (now) the reliability of radiometric dating on the geologic column, it blows up on me, too.
I don't deny that there is some degree of plausibility to radiometric dating, although I have to wonder if many field geologists secretly have their doubts about it. My concern is instead to know how much stamina the evidence has against other evidence that may call it into question. My conclusion for the geologic column is, not much.
Gentry's radiohaloes in coalified wood
Here is some more material from my web site bearing on the question of the age of the geologic column:
It is also of interest in regard to radiometric dating that Robert Gentry claims to have found "squashed" polonium haloes as well as embryonic uranium radiohaloes in coal deposits from many geological layers claimed to be hundreds of millions of years old. (See the Oct. 15, 1976 issue of Science.) These haloes represent particles of polonium and uranium which penetrated into the coal at some point and produced a halo by radioactive decay. The fact that they are squashed indicates that part of the decay process began before the material was compressed, so the polonium had to be present before compression. Since coal is relatively incompressible, Gentry concludes that these particles of uranium and polonium must have entered the deposit before it turned to coal. However, there is a very small amount of lead with the uranium; if the uranium had entered hundreds of millions of years ago, then there should be much more lead. The amount of lead present is consistent with an age of thousands rather than millions of years. It's hard to believe, according to conventional geological time scales, that this coal was compressed any time within the past several thousand or even hundred million years.
Here is a quote from Coffin, page 306, about Gentry's findings:
"Coalified wood from Triassic and Jurassic sediments (225- to 135-million-year conventional geologic age) contains radiohaloes. Published lead-206/uranium-238 ratios for their inclusion centers may be expressed in terms of uranium-lead radioisotope ages ranging between 236 thousand and 2.9 million years. No presently available experimental evidence would exclude the possibility that essentially all the lead-206 in the halo centers was introduced together with the uranium (either directly or as parent polonium-210 or lead-210) and thus did not accumulate from uranium."
In fact, a couple of the haloes have ages consistent with an origin thousands of years ago.
Thus the amount of lead with the uranium is consistent with an age in the hundreds of thousands to millions of years range, much too small for conventional geologic time. And it is reasonable to assume that almost all of this lead came with the uranium, rather than being a result of decay, suggesting that the true age could be much younger than this.
Note that this phenomenon of squashed haloes appears in different coal deposits in different geologic formations, and all give about the same U-Pb ages. The squashing is in the vertical direction, and I can't think of any way this could happen at a time later than the burial of the logs or whatever under a lot of sediment. Coal is not water soluble (at least, coal cars aren't covered, and no one seems to worry about thunderstorms dissolving the coal away), and wood is waterproof, so one would expect that coalified wood would also be waterproof. Coal has small pores. If it had cracks, they would have to be small, since the cell structure is still visible. And if there was a flow of water, it would be more likely to remove soluble uranium than insoluble lead, making the date older. But it is possible that small cracks exist and that uranium could be deposited by a flow of water at some more recent date.
If there were such cracks, we would expect uranium to be entering at regular intervals, and to give a range of ages up to about 225 million years or even higher due to lead being introduced with the uranium. But note that all of the haloes give young ages. The fact that all the ages are so young suggests that the coal is young, too.
It seems most likely that the uranium entered at the same time as the polonium. The fact that so many of the polonium haloes are squashed indicates that the polonium entered before the wood was covered with sediement. I think the most reasonable explanation is that this coal has an age at most a few millions of years old, possibly much younger, and that the geologic time scale is in error. Some of the haloes have ages of 200,000 or 300,000 years, so the true age would have to be this or younger. This applies to several geologic periods. In fact, a couple of the haloes have such low ratios as to imply an age in the thousands of years.
Another possible objection made by an evolutionist is that the radon 222 that results from uranium decay is an inert gas and may have escaped, resulting in little lead being deposited. This would make the observed haloes consistent with an old age for the coal. However, the fact that these uranium haloes are embryonic (very faint) also argues for a young age. In addition, not all of the radon would be on the surface of the particles of uranium. That which was inside or bordering on coal would likely not be able to escape. Since radon 222 has a half-life of about 4 days, it would not have much time to escape, in any event. Such haloes were also found in shale, with young U/Pb ages as well, and it may be less likely for the radon to escape from shale.
Carbon 14 dating
The following material is from http://www.rae.org/ch04tud.html
(It looks like C14 dating is the ``bad boy'' of radiometric dating.)
Dr. Libby, the discoverer of the C14 method, which won for him a Nobel prize, expressed his shock that human artifacts extended back only 5000 years, a finding totally in conflict with any evolutionary concept. Older dates were found to be very unreliable (CRSQ , 1972, 9:3, p.157). By this time tens of thousands of C14 dates have been published from tests performed by various laboratories around the world. In the annual volumes in which the dates are published, concerns have been expressed about many relatively young dates that violate established geological age notions. One example given was Ice-Age materials that were dated by C14 to fall within the Christian era (CRSQ , 1969, 6:2, p.114). In his book on prehistoric America, Ceram notes a classic case of the difficulties that befall C14 dating. Bones 30,000 years old were found lying above wood dated at 16,000 years (Ceram, 1971, p.257-259).
Another classic C14 problem was noted for Jarmo, a prehistoric village in northern Iraq. Eleven samples were dated from the various strata and showed a 6000-year spread from oldest to most recent. Analysis of all the archaeological evidence, however, showed that the village was occupied no more than 500 years before it was finally abandoned (Custance, 1968, Mortar samples can be given normal C14 tests since mortar absorbs carbon dioxide from the air. Mortar, however, from Oxford Castle in England gave an age of 7,270 years. The castle was built about 800 years ago. The kind of contamination is unclear. Living trees near an airport were dated with C14 as l0,000 years old, because the wood contained contamination from plane exhaust (CRSQ , 1970, 7:2, p.126; 1965, 2:4, p.31). p.19).
[I wouldn't be surprised if these last 2 examples have simple explanations.]
C14 analysis of oil from Gulf of Mexico deposits showed an age measured in thousands of years - not millions. Data produced by the Petroleum Institute at Victoria, New Zealand, showed that petroleum deposits were formed 6,000-7,000 years ago. Textbooks state that petroleum formation took place about 300,000,000 years ago (Velikovsky, 1955, p.287; CRSQ , 1965, 2:4, p.10). Fossil wood was found in an iron mine in Shefferville, Ontario, Canada, that was a Precambrian deposit. Later the wood was described as coming from Late Cretaceous rubble, which made it about 100 million years old instead of more than 600 million years old. Two independent C14 tests showed an age of about 4000 years (Pensee , Fall 1972, 2:3, p.43).
The last major glacial advance in America was long dated at about 25,000 years ago. C14 dates forced a revision down to 11,400 years. The United State Geological Survey carried out studies that gave a C14 date as recent as 3300 years ago, but no text treats such a puzzling find that falls well within historic times (Velikovsky, 1955, p.158-159; CRSQ , 1968, 5:2, p.67). Here is a remarkable example of C14 difficulties in a book published by Stanford University Press. Six C14 ages were determined from a core in an attempt to date the formation of the Bering Land Bridge. The dates ranged from 4390 to 15,500 Before Present.
The first problem was that the results were so disarranged from bottom to top of the core that no two samples were in the correct order. Then the oldest date was discarded because it was 'inconsistent' with other tests elsewhere. Next the remaining dates were assumed to be contaminated by a fixed amount, after which the authors concluded that the delta under study had been formed 12,000 years ago (Hopkins, 1967, p.110-111). ... Even more astonishing is this cynical statement made at a symposium of Nobel Prize winners in Uppsala, Sweden, in 1969: If a C14 date supports our theories, we put it in the main text. If it does not entirely contradict them, we put it in a footnote. And if it is completely 'out of date,' we just drop it (Pensee , Winter 1973, p.44).
As for the contamination issue, someone asserted that any C14 date of 30,000 years or more is due to contamination. If this is so, then why do they say the method is accurate to 50,000 years? If any C14 date has ever yielded a value over 30,000 years, this implies that such contamination is not ubiquitous. Of course, it could be that older measurement techniques were less accurate. Now, 30,000 years is about 5 half lives of C14, which means that a contamination of 1/32 (slightly less) would be required to achieve this date for a sample of infinite age. This is a substantial contamination.
Anyway, as for C14 dating in general, it seems clear that many, many results are much too young according to the standard view, and that explaining away one or two of them does not appreciably diminish the problem.
Here is another instance of an anomalously young carbon 14 date:
At the 1992 Twin Cities Creation Conference, there was a paper presented called "Direct Dating of Cretaceous-Jurassic Fossils (and Other Evidences for Human-Dinosaur Coexistence)". Among other things, the results of carbon-dating of Acrocanthosaurus bones are given.
The authors noted that dinosaur bones are frequently ("as a rule") found with a black carbon residue of some sort on the bones. The authors speculated that this residue could be the leftovers of the decayed skin and flesh: they quote the Penguin Geology Encyclopedia's definition of "carbonization": "Carbonization; the reduction of organic tissue to a carbon residue. An unusual kind of fossilization in which the tissue is preserved as a carbon film. Plants are commonly preserved in this manner, soft-bodied animals more rarely." Since this material is organic, it can be used to carbon-date the fossils.
The authors describe in detail the measures taken to ensure that no other source of carbon contamination was present inside or outside the bones. When the bones were ground up and carbon-dated, the dates they received from the lab from different methods were 9,890 to 36,500 years BP (before present).
Some have claimed that this bone was covered with shellac, causing the carbon 14 date to be young. Concerning this issue, one individual sent me the following information:
The papers of Miller's that are cited by Lepper are:
Fields, W., H. Miller, J. Whitmore, D. Davis, G. Detwiler, J. Ditmars, R. Whitelaw, and G.Novaez, 1990, "The Paluxy River Footprints Revisited," in _Proceedings of the Second International Conference on Creationism held July 30-August 4, 1990, Volume 2, technical symposium sessions and additional topics_, edited by R.E. Walsh and C.L. Brooks, pp. 155-168, Christian Science Fellowship, Pittsburgh.
Dahmer, L., D. Kouznetsov, A. Ivenov, J. Hall, J. Whitmore, G. Detwiler, and H. Miller, 1990, "Report on Chemical Analysis and Further Dating of Dinosaur Bones and Dinosaur Petroglyphs," same proceedings, pp. 371-374.
The above two articles are the ones that purportedly refer to carbon 14 dating of a dinosaur bone covered with shellac. The article I referred to is the following:
"Direct Dating of Cretaceous-Jurassic Fossils (and Other Evidences for Human-Dinosaur Coexistence)" (1992 Twin Cities Creation Conference).
In this paper, the authors describe in detail the measures taken to ensure that no other source of carbon contamination was present inside or outside the bones.
The fact that these are separate papers, and the fact that every attempt was made to avoid contamination, suggests that these are two different incidents. I also received the following information from another person:
As far as I can ascertain from the paper, the researchers responsible specifically mention that the dinosaur bones being dated were not coated with shellac (page 10). Otherwise, the details of the material at your website are as in the paper, and the comment about a black carbon residue around fossilised dinosaur bones is referenced in their paper to a secular source, so it is not simply their observation. The comments from the Penguin Geology Encyclopedia merely add to their case.
However, of the results they give in their paper, I personally would only be comfortable with the AMS results obtained on the same sample in two different laboratories - the one at 25,750+/-280 years BP and the other at 23,760+/-270 years BP. The other results were obtained on unspecified equipment or via the less reliable older beta technology and generally appear not to have been cross-checked in another laboratory.
Again I confirm that the claim about the shellac appears to be totally false and merely a smokescreen to avoid the implications of an uncomfortable radiocarbon date.
So, based on all of this information, it looks like there were two separate incidents, and the one I referred to involved a dinosaur bone that was not covered with shellac, but still gave a young carbon 14 date.
Finally, some more quotes about carbon 14 dating from http://www.parentcompany.com/handy_dandy/hder12.htm A survey of the 15,000 radiocarbon dates published through the year 1969 in the publication, Radiocarbon, revealed the following significant facts:27 a. Of the dates of 9671 specimens of trees, animals, and man, only 1146 or about 12 percent have radiocarbon ages greater than 12,530 years.
b. Only three of the 15,000 reported ages are listed as "infinite."
c. Some samples of coal, oil, and natural gas, all supposedly many millions of years old, have radiocarbon ages of less than 50,000 years.
d. Deep ocean deposits supposed to contain remains of the most primitive life forms are dated within 40,000 years.
I think it is interesting that so few specimens have old dates, suggesting a rapid increase in the amount of carbon 14 in the atmosphere.
On the same subject, some fossils from the Paluxy River are "anomalous" as well. Carbonized (burnt) wood was discovered in Cretaceous limestone, and dated to 12,800 to 45,000 YBP.
Coffin gives quite a bit of evidence from increases of C14 ages with depth that the concentration of C14 has increased rapidly in recent years, making C14 dates too old, especially after about 4000 years ago. The fact that C14 is still increasing in the atmosphere shows that the earth recently went through some kind of a catastrophe, and this increase is even admitted by some evolutionists.
It has been claimed that Carbon 14 dating was revolutionized in 1969 or so. But it remains to establish how much in error the old dates were. It seems to be a common pattern that when dating methods are revised, we are told how inaccurate the old methods were, but are not told how inaccurate the current methods are.
A number of people requested references for my statements about young carbon 14 dates for coal and oil and fossils. Here is what I found at http://www.christiananswers.net/q-aig/aig-c007.html Consider this: if a specimen is older than 50,000 years, it has been calculated that it would have such a small amount of C14 that for practical purposes it would show an infinite radiocarbon age. So it was expected that most deposits such as coal, gas, etc. would be undatable by this method. In fact, of thousands of dates in the journals Radiocarbon and Science to 1968, only a handful were classed "undatable" - most were of the sort which should have been in this category. This is especially remarkable with samples of coal and gas supposedly produced in the Carboniferous period 300 million years ago! Some examples of dates which contradict orthodox (evolutionary) views:
Coal from Russia from the "Pennsylvanian," supposedly 300 million years old, was dated at 1,680 years. (Radiocarbon, vol. 8, 1966).
Natural gas from Alabama and Mississippi (Cretaceous and Eocene, respectively) should have been 50 million to 135 million years old, yet C14 gave dates of 30,000 to 34,000 years, respectively. (Radiocarbon, vol. 8, 1966. Many of the earlier radiocarbon dates on objects such as coal and gas, which should be undatable, have been attributed to contamination from, for example, workers' fingerprints, creationist researchers are currently working on the construction of an apparatus, using existing technology, to look for very low levels of C14 activity in, for example, coal after excluding contamination. Such low-level activity would not be expected on the basis of old earth theory, and so is not looked for at present.)
Bones of a sabre-toothed tiger from the LaBrea tar pits (near Los Angeles), supposedly 100,000-one million years old, gave a date of 28,000 years. (Radiocarbon, vol. 10, 1968)
Tree ring chronologies
Tree ring chronologies are also used to give a history of the earth stretching back over 8000 years. Are these accurate? Here is some information from http://www.creationscience.com/onlinebook/faq/radiocarbon.shtml
One way to infer how the atmospheric concentration of carbon-14 changed in the past is by tree-ring dating. Some types of trees, that grow at high elevations and have a steady supply of moisture, reliably add only one ring each year. In other environments, multiple rings can be added in a year. 4 The thickness of a tree ring depends on the tree's growing conditions, which will naturally vary from year to year. Some rings may even show frost or fire damage. By comparing sequences of ring thicknesses in two different trees, a correspondence can sometimes be shown. Ring patterns will correlate strongly for two trees of the same species that grew near each other at the same time. Weaker correlations (or less confident matches) exist between trees of different species growing simultaneously in different environments. Claims are frequently made that wood growing today can be matched up with some scattered pieces of dead wood so that tree-ring counts can be extended back more than 8,600 years. This may not be true.
These claimed "long chronologies" begin with either living trees or dead wood that can be accurately dated by historical methods. This carries the chronology back perhaps 3,500 years. Then the more questionable links are established based on the judgment of a tree-ring specialist. Standard statistical techniques could establish just how good the dozen or more supposedly overlapping tree-ring sequences are. However, tree-ring specialists refuse to subject their judgments to these statistical tests, and they have not released their data so others can carry out these statistical tests. 5
There are some general problems with constructing a chronology by piecing together records of tree rings from different trees. When trying to find the best solution to a problem like this, there are generally a huge number of possible solutions. So one uses a heuristic program to try to find a good one. There may also be many other solutions that are nearly as good. In fact, there may be others that are even better. So it's not clear to me that there is one clear-cut chronology based on tree ring dating. It was claimed that carbon 14 levels were not considered at all in constructing this chronology. I'd like to have his reference for that. In such a case, one typically defines a goodness function for each solution, and this could incorporate the desire to maintain a nearly constant carbon 14 level in the atmosphere. Add to this the fact that different trees can respond differently to the same climatic condition, and the fact that trees sometimes have more than one ring (especially if there is more than one rainy season per year) and one has even more uncertainty. Without a very thorough examination of the data, it's hard to know how to interpret the result. I'd be interested to know what the authors of this work say about the existence of other chronologies, and how much less of a good fit they are.
In such an optimization problem, it is difficult to know if one has the true solution, so not much weight should be given to the chronology obtained. It's not enough just to eyeball it and say it looks convincing. It should be subjected to several optimization procedures and one should also optimize for shorter chronologies as well to see how much (if any) the quality suffers.
Someone gave me some information about constructing tree ring chronologies by piecing together sub-chronologies. But in a problem like this, sometimes one can get a better solution in the end by taking a sub-optimal choice along the way. So the described procedure will not necessarily find the best chronology.
The following message was sent to me by e mail on February 11, 1998:
As one who has taught dendrochronnology, I have a few opinions on this particular subject. Also, one of my graduate students went to work for Ferguson in his lab at U of A, and in fact was the curator of his work after his death, and is presently probably the only one who knows anything about how he [Ferguson] produced the bristlecone chronology. Another of my graduate students gave a seminar to the lab on dendrochronology of fossil trees and had ample opportunity to analyze the procedures there, and to work with Ferguson for a while. I can say on pretty firm grounds that the Bristlecone chronology before 4000bp is fraught with problems and unanswered questions. While Ferguson was alive, he never allowed anyone to analyze his original data or the bases for the many suppositions that went into the establishment of the chronology. Thus the chronology was not subjected to the normal rigors of science. This is regrettable, because I believe he was a careful and sincere scientist. Of course one could always excuse Ferguson for not revealing the bases of his decisions (for example, the most important rings in any chronology are the "missing rings" which have to be added by the investigator). But suffice to say the chronology before 4000bp is entirely dependent on C14 dates of the wood, and is thus tautologous. This does not mean it is meaningless or necessarily wrong, just that I wouldn't base too much on it.
Another article discussing difficulties with tree ring chronologies can be found at