[Dr. James Roth was the twenty-second witness called by the defence. He testified on Thursday, April 21, 1988.]
Dr. James Roth, the laboratory manager of Alpha Analytical Laboratories, testified as to the results of tests done on the numbered samples removed from Auschwitz and Birkenau by Fred A. Leuchter. Roth had obtained his doctorate from Cornell University in analytical chemistry.
Roth testified that he received samples from Fred Leuchter in his capacity as an Analytical Chemist at Alpha Analytical Laboratories. Roth directly supervised the tests performed on the samples and the preparation of the test report. The purpose of the tests was to determine total iron content and total cyanide content in the samples. The identification numbers assigned to the samples were those designated by Leuchter. (33-9274 to 9276)
Iron tests were conducted on three of the samples, namely, samples 9, 29 and 32. Results of the tests for total iron content were essentially the same for all three samples. Sample 9 contained 7,580 mg/km; sample 29 contained 6,280 mg/km and sample 32 contained 6,170 mg/km. (33-9276, 9291, 9292)
Iron was normally present in brick and mortar and the quantities found in the brick samples tested were fully within the acceptable ranges for brick type. Red bricks were red because of the iron, although even white bricks had these levels of iron present. (33-9306)
Cyanide was analyzed in a total of 32 samples of which 31 were brick material and one was a gasket material. The minimum trace level for cyanide was one milligram per kilogram of material. Tests results which did not detect cyanide were designated on the report as "ND," meaning "not detected."(33-9276 to 9278)
Roth testified that the test results indicated the following: sample 1 showed no detection; sample 2 showed no detection; sample 3 showed no detection; sample 4 showed no detection; sample 5 showed no detection; sample 5 duplicate test showed no detection; sample 6 showed no detection; sample 7 showed no detection; sample 7 spike recovery test indicated 119 percent; sample 8 showed no detection; sample 8 duplicate showed 1.9 milligrams per kilogram; sample 9 showed 6.7 milligrams per kilogram; sample 10 showed no detection; sample 11 showed no detection; sample 12 showed no detection; sample 13 showed no detection; sample 14 showed no detection; sample 15 showed 2.3 milligrams per kilogram; sample 16 showed 1.4 milligrams per kilogram; sample 16 spike recovery test indicated 96 percent; sample 17 showed no detection; sample 18 showed no detection; sample 18 spike recovery test indicated 100 percent; sample 19 showed no detection; sample 19 spike recovery test indicated 120 percent; sample 20 showed no detection; sample 20 duplicate showed 1.4 milligrams per kilogram; sample 21 showed 4.4 milligrams per kilogram; sample 22 showed 1.7 milligrams per kilogram; sample 23 showed no detection; sample 24 showed no detection; sample 25 showed 3.8 milligrams per kilogram; sample 25 duplicate showed 1.9 milligrams per kilogram; sample 26 showed 1.3 milligrams per kilogram; sample 26 spike recovery test indicated 140 percent; sample 27 showed 1.4 milligrams per kilogram; sample 28 showed 1.3 milligrams per kilogram; sample 29 showed 7.9 milligrams per kilogram; sample 30 showed 1.1 milligrams per kilogram; sample 30 duplicate showed no detection; sample 31 showed no detection; sample 32 showed 1,050 milligrams per kilogram. (33- 9278 to 9287) A bar graph of the sample results which Roth had examined and determined to accurately represent the test results was entered as Exhibit 154. (33-9288)
The tests were performed by taking a representative sample of the material that was received by the laboratory, placing it in a flask that could be sealed, adding a low concentration of acidic solution, specifically sulphuric acid, then warming the sample in that solution while in the process passing gas through it. Air passed through the solution and the acidic environment volatilized the cyanide and formed hydrogen cyanide gas. This gas was then passed through a solution of sodium hydroxide. Any hydrogen cyanide would react with the sodium hydroxide to form sodium cyanide. After a period of time required to assume complete removal of any cyanide in the sample, the solution was analysed colour metrically for the presence of cyanide. (33-9280)
This process was repeated with each of the samples, with duplicates on certain selected samples and with spot samples in which known amounts of cyanide were added to check recovery. Cyanide spike recovery tests performed on several of the samples all indicated that the analyses and the techniques and methods by which the samples were analyzed were valid. (33- 9281 to 9287)
Prussian Blue (ferro-ferri-cyanide)
Roth was shown Exhibit 144, a colour photograph of the blue staining on the wall of Delousing Facility No. 1 at Birkenau from which sample 32 had been removed. He indicated that the blue colour was what was commonly referred to as "Prussian blue." (33-9289) The chemical definition of Prussian blue was ferro-ferri-cyanide. (33-9297) Prussian blue was an iron cyanide produced by a reaction between iron and the hydrogen cyanide. It was a very stable compound which stayed around a long time. If hydrogen cyanide came into contact with bricks or mortar containing iron, it was fully conceivable that a reaction of the iron and hydrogen cyanide would take place, leaving behind the Prussian blue. (33-9290) In porous materials such as brick and mortar, the Prussian blue could go fairly deep as long as the surface stayed open, but as the Prussian blue formed, it was possible that it would seal the porous material and stop the penetration. If all surface iron was converted to Prussian blue, the reaction would effectively stop for lack of exposed iron. (33 9291)
Roth testified that the iron/cyanide reaction capabilities of samples 9 and 29 were no different from that of sample 32. If samples 9 and 29 had been exposed continually everyday for two years to 300 parts per million of hydrogen cyanide, Roth testified that he would expect to see the formation of the iron cyanide compounds; the so called "Prussian blue" material, in detectable amounts. The reaction of the two substances was an accumulative reaction; the reaction continued with each exposure. One way for this reaction not to occur would be a lack of water. These reactions, in many cases, required water or vapour in order to occur. However, in rooms of normal temperatures and normal humidity, there would be plenty of moisture present for this type of reaction to take place. (33-9293, 9294)
Prussian blue did not normally disappear unless it was physically removed. To be removed from a porous material like a brick it would have to be removed by sandblasting or grinding down the surface or by the application of a strong acid such as high levels of sulphuric, nitric or hydrochloric acid. It would be more difficult to remove from porous surfaces because of the fact that the formation would have taken on depth. (33-9297, 9298) This ended the examination-in-chief of Roth, and his cross-examination commenced.
Roth testified that he did not take the samples or have any control over the sample taking. He agreed that cyanide radicals could exist in forms other than Prussian blue and that the absence of Prussian blue did not necessarily mean that cyanide radicals were absent. To Pearson's suggestion that a good control sample would have been one where Prussian blue was not present in order to determine if any cyanide radicals were present there, Roth pointed out that there were many samples where no cyanide was detected. (33-9301, 9302)
Roth testified that in order to have Prussian blue, iron must be present and accessible to the cyanide. (33-9301) He agreed that the presence of Prussian blue almost guaranteed that the ferri-cyanide complex was present. (33-9302) How deep Prussian blue would penetrate was totally dependent on many factors, such as the porosity of material and what moisture existed in the area. (33-9303) Asked if a building was blown up with dynamite and the surface blown off, the Prussian blue might thereby be removed, Roth replied that if just the surface was removed and the rest of the material was left, the answer would be yes. The Crown stated this was not what was suggested; the suggestion was that in an explosion the surface of the brick would come off. Roth replied that normally bricks would break up. "Now, if that's removal of the surface, yes." (33-9304)
Roth refused to answer a question dealing with the amount of hydrogen cyanide required to kill insects as opposed to human beings as he felt this was not his area of expertise. (33-9304) He agreed that he would not want to be around 300 parts per million of hydrogen cyanide. (33-9305)