S&V - Page 115
The last denial in date has been published in January 2005 in the scientific review the Thermochimica Acta, via an article stating a new method of dating developed by Raymond Rogers. This American chemist, retired from the national laboratory of Los Alamos, University of California, is a ex-member of Sturp (research project on the shroud of Turin), famous organization which gathered forty scientists sindonologists from 1976 to 1996, date of its dissolution.

In possession of microscopic dusts collected on the linen with adhesive tapes, but perhaps also of samples taken on the shroud for the dating, the researcher estimates indeed those would come from a later mending of the linen cloth, which would have distorted the dating measurements. To support this assertion, Rogers compared the rate of vanillin - a chemical compound present in the lignin of linen fibres - of a sample of linen cloth cut out for the dating of 1988 with that of the dusts collected with adhesive tapes on other places of cloth. Whereas this sample has no vanillin trace, the dusts, them, have some, though in quantity lower than the rate measured by Rogers on medieval textiles. The vanillin, which is degraded with time, would allow - always according to the author - to estimate the age of a vegetable support. However, the fact that the « official«  sample has no vanillin trace would re-open the question of the reliability of this sample for the dating of the shroud. In the end, the vanillin rate measured at other places of the relic would not correspond to that awaited for a cloth of the Middle Ages, but would reflect its two millennia of age.

S&V - Page 116
« The least that one can say is we have no benefit of hindsight to judge viability and precision of this very new method", estimates the physicist Patrick Berger, member of the Zetetician Circle (the zetetic is the science which studies - in a rational way - the paranormal phenomena). « Moreover Rogers does not provide in its article the data sources of its method and its degree of uncertainty. It refers to works of one of his colleagues, Stanley Kosiewicz, without quoting any publication, and it compares collected dusts with cloth samples, therefore with materials of different nature. And then, his equations are packed full of errors. » he concludes.

In addition, it seems astonishing that Rogers had samples of 1988 radiodatation. Why doesn't he indicate the source of it? The cardinal of Turin has half of it and does not wish to part with it at ends of analysis. As for other half, it was distributed between the three laboratories. To suppose that the latter did not date - and thus destroyed - totality from the given material, they did not have the right to give the remainder up to whoever.

VSD - Page 24
The arguments advanced by the protesters are supported, among other things, by two remarks. In the first place one finds in the measurements published by the three laboratories some peculiarity which should be present only if the samples used did not have the same origin. This - which is indisputable and was highlighted by several specialized mathematicians - is obviously enough to wonder about the reliability of the study. And like the shroud several times were restored, one can wonder whether the date 1260-1390 is not an « average » resulting from measurements carried out on a very old part of the linen cloth, and other measurements coming from a mending.

VSD - Page 28
Some mathematicians initially denounced the statistical handling of the results obtained by the three laboratories of Oxford, Tucson and Zurich: the range of dates obtained by Oxford being completely apart from the intervals of Tucson and Zurich, these two last are initially amalgamated; then the two ranges obtained are again gathered into only one to lead to the final result 1260-1390.

Many of other anomalies were noted, in particular the doubtful choice of the area for the samples, on an edge of fabric very handled at the time of the public exhibitions of the relic and close to a carbonized zone. For certain observers, measurements are not to call into question, but the samples were not representative.