Auschwitz: The End of a Legend

By Carlo Mattogno


TABLE OF CONTENTS

  1. INTRODUCTION
  2. THE CREMATORY OVENS OF AUSCHWITZ-BIRKENAU
    ACCORDING TO JEAN-CLAUDE PRESSAC
  3. THE CREMATORY OVENS OF AUSCHWITZ-BIRKENAU
    ACCORDING TO CREMATION TECHNOLOGY
  4. GENESIS AND DEVELOPMENT OF THE "FINAL SOLUTION"
  5. CREMATORIES II AND III OF BIRKENAU
  6. BUNKERS 1 AND 2
  7. CREMATORIES IV AND V
  8. CONCLUSION

APPENDIX
NOTES

GLOSSARY


AUSCHWITZ:
THE END OF A LEGEND

by Carlo Mattogno

All rights reserved. No part of this book may be reproduced, or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording, or by an information storage and retrieval system without written permission from the publisher except for the inclusion of brief quotations in a review, in which case the following attribution shall appear: "From Auschwitz: The End Of A Legend, by Carlo Mattogno, Russ Granata, publisher." A copy of the review should be sent to the publisher.

Translated from the Italian, AUSCHWITZ: FINE DI UNA LEGGENDA Considerazioni storico-techniche sul libro Les crématoire d’Auschwitz. La machinerie du meurtre de masse di Jean-Claude Pressac, Copyright © 1993 by Carlo Mattogno.

Translator: Anne Sharp
First English language edition Copyright © 1994
by Russ Granata, publisher
First Printing 1994

Library of Congress Cataloging in Publication Data
Mattogno, Carlo
AUSCHWITZ: THE END OF A LEGEND
Critique On Pressac’s Auschwitz
by Carlo Mattogno. 1st ed.
Includes index, appendix and glossary
CIP 94-75562
ISBN 0-9640716-0-6: Softcover


Chapter One

Jean-Claude Pressac

Jean-Claude Pressac is the author of a large-format book on the Auschwitz-Birkenau complex entitled Auschwitz: Technique and Operation of the Gas Chambers, published in 1989 by the Beate Klarsfeld Foundation, 515 Madison Avenue, New York, NY 10022. That work, which was acclaimed at the time of its publication as the definitive proof of the existence of homicidal gas chambers at Auschwitz-Birkenau, brought Pressac praise as spécialiste incontesté des recherches sur les techniques de l’extermination nazie ("unquestionable specialist in the research of Nazi extermination techniques") and as expert incontesté, sinon unique ("unquestionable expert, if not unique") [1] in this field.

But praise from shallow journalists aside, to the expert eye, Pressac reveals a surprising ignorance of the chemical-physical properties of Zyklon B and its use for the purpose of disinfestation, as well as the structure and functioning of crematory ovens. [2] This double incompetence in the two essential aspects of the problem inevitably led Pressac to unfounded conclusions in his 1989 work. Despite this, that book is valuable for its considerable documentation and for a critical spirit uncommon in the traditional historiographic field, where, regarding the sources, a systematized theological dogmatism rules. Pressac should furthermore be acknowledged for his courage in overcoming, or at least attempting to overcome, the traditional historiographic methodology in this field, which he justifiably labels as:

a history based for the most part on testimonies, assembled according to the mood of the moment, truncated to fit an arbitrary truth, and sprinkled with a few German documents of uneven value and without any connection with one another. [3]

That book provided enough arguments for historical Revisionism to be considered crypto-revisionist, evidently even by its own publisher, since it has been practically impossible to obtain.

Another book by Jean-Claude Pressac, entitled Les crématoires d’Auschwitz: La machinerie du meurtre de masse [The Crematoria of Auschwitz: The Machinery of Mass Murder] published in Paris, 1993, should have complemented his earlier book by virtue of the vast amount of documentation he encountered in Moscow, particularly the archives of the Bauleitung (the Auschwitz construction management), which were left intactes ("intact") in the hands of the Soviets (p. 1) [!1].

But in fact, reading his Les crématoires d’Auschwitz, one senses an uncomfortable reversion: Jean-Claude Pressac returned to the worst cliches of the worst traditional historiography. This was inevitable: In the 80,000 (eighty thousand!) documents at Moscow, and in the entire archives of the Bauleitung [!2], Pressac found NO PROOF of the existence of one single homicidal gas chamber at Auschwitz-Birkenau. For example, concerning Crematory II at Birkenau, no "criminal trace" (Pressac’s term) is dated prior to 31 March 1943, the date of the official consignment of the crematory to the administration of the camp. Now that is, to say the least, just a little bit strange for an extermination plant that was supposed to have functioned:

as a homicidal gas chamber and incineration installation from [the] 15th of March 1943, before its officially coming into service on [the] 31st of March, to [the] 27th of November 1944, annihilating a total of approximately 400,000 people, most of them Jewish women, children and old men. [4]

Thus, for over twenty months of alleged extermination activity in this supposedly homicidal crematory, for an extermination of 400,000 people, the archives of Moscow do not even contain one single "criminal trace"! And the same goes for the other crematories at Auschwitz-Birkenau.

This must have disturbed Pressac, who must have found himself in the difficult position of making these documents say what they do not say. This need explains Pressac’s cranky methodology, which is characterized by his indiscriminate use of sources, and by arbitrary and unfounded deductions inserted into the body of the text within a dense web of notes so as to give the impression of coming from historical documentation. The connection between the various documents appears forced, and the interpretation of those documents is contorted to make it seem as if they support the existence of homicidal gas chambers.

Pressed by Revisionist research which demonstrates the impossibility of mass extermination at Auschwitz-Birkenau from the technical point of view, Pressac plays not only at diminishing the numbers of victims, but also with the intentions of the SS. The number of presumed homicidally gassed victims, which in 1989 was "about 900,000," [5] of whom 750,000 were supposedly killed at Crematories II and III alone, [6] is here reduced to only 630,000 (p. 148). Both figures are completely arbitrary. Furthermore, the homicidal gas chambers have become "little" and thus of small extermination capacity. In effect, Pressac has been forced to "equilibrate" the capacity of the homicidal gas chambers to that of the crematory ovens. In his 1989 work, his ratio of corpses gassed to corpses cremated had been much higher.

All these changes have naturally required jarring contradictions with respect to his earlier Auschwitz: Technique and Operation of the Gas Chambers. But this is unimportant to an author, who seems to accept or reject figures and arguments depending on his whim.

To complete the picture, Pressac has again enormously exaggerated the capacity of the crematories of Auschwitz-Birkenau as he did in 1989, arriving at conclusions which are technically and thermotechnically senseless, due to his apparent ignorance of essential aspects of cremation.

The subject of homicidal gas chambers has caused Pressac no little difficulty, not only due to the absolute lack of proof on this subject in the Moscow documents, but above all, because the documentation on ventilation installations in the basement of Crematories II and III show undeniably that homicidal gas chambers were not planned, and were not installed. We shall subsequently see by what means Pressac has attempted to overcome this difficulty.

The critique presented here is essentially based upon a scientific study of the crematory ovens, and of the presumed homicidal gas chambers at Auschwitz-Birkenau, which has involved over five years of research with the very valuable collaboration of Engineer Dr. Franco Deana of Genoa, and Engineer H.N. of Danzig. That work consists of two volumes entitled Auschwitz: i forni crematori [Auschwitz: The Crematory Ovens], Auschwitz: le camere a gas, [Auschwitz: The Gas Chambers], which are being published in Italy. The present work is a synthesis of these studies. The interested reader will be able to find among the citations of those two detailed studies, many references which are not included in this critique.

The Crematory Ovens of Auschwitz-Birkenau study to be published in America, extends historiography in this field wherein researchers may find in that particular work, references other than those appearing in this critique. The Table of Contents of The Crematory Ovens of Auschwitz-Birkenau is shown as Document No. 1 of the appendix herein.

Chapter Two

THE CREMATORY OVENS OF AUSCHWITZ-BIRKENAU
ACCORDING TO JEAN-CLAUDE PRESSAC

Capacity: The Suppositions

A scientific study of the crematory ovens of Auschwitz-Birkenau must confront and resolve two fundamental thermotechnical problems: that of capacity, and that of yield. Capacity is the number of corpses cremated within a time frame (reference: one day of activity). Yield is the relation between heat produced and heat used: specifically, fuel consumption. Jean-Claude Pressac does not confront either of these two problems scientifically, limiting himself simply to a series of statements as to the capacity of the ovens (which he erroneously calls "yield"), sprinkled here and there throughout his book. These statements, under analysis, yield the following argument:

1. The mobile oil-heated Topf two-chambered oven which was installed in Dachau at the end of 1939 had a capacity of two corpses per hour (p. 7). Thus, the cremation of one corpse in one chamber lasted one hour.

2. The Topf two-chambered "Auschwitz model" oven heated by coke was of a design different from that of the Dachau oven. This was the result of a change in the first two-chambered Topf oven at Buchenwald which was originally heated with combustible oil, into a coke-heated oven, via the installation of two gasogenes in the rear (p. 12). Thus, the above-mentioned capacity of two corpses/hour does not apply to this oven.

3. The installation of compressed air (Druckluftgebläse) reduced the duration of cremation (pp. 13 and 68).

4. The "Auschwitz model" oven had a capacity of 30 to 36 corpses in ten hours (p. 13).

5. The ovens were used 21 hours a day, because their functioning required three hours rest (p. 13).

6. The three two-chambered ovens of Crematory I at Auschwitz had a capacity of 200 to 250 corpses per day (pp. 49, 80).

7. The two Topf three-chambered ovens heated by coke at Buchenwald (of which one was also adaptable for heating with combustible oil) resulted in:

un rendement incinérateur supérieur d’un tiers à celui calculé à partir de l’expérience acquise sur les fours bimoufle (p. 39).

[an incineration capacity larger by one-third, their calculations based on the operation of the twin-chambered ovens.]

8. The capacity of the five three-chambered ovens of this model installed in Crematories II/III in Birkenau was 800 corpses per day (p. 39) or 1,000 per day (p. 80).

9. The capacity of each of the two eight-chambered ovens installed in Crematories IV and V at Birkenau was 500 corpses per day (p. 80).

10. During the first experimental cremation in Crematory II on 4 March 1943, 45 corpses of hommes gras [fat men] were cremated; 3 for every chamber, and the cremation lasted 40 minutes (p. 72).

11. The "official" capacity of the crematories was as follows:

Crematory I : 340 corpses per day
Crematory II : 1,440 " " "
Crematory III : 1,440 " " "
Crematory IV : 768 " " "
Crematory V : 768 " " "

Pressac comments:

Ces chiffres officiels sont de la propagande mensongères et pourtant sont valables. Leur validité apparente repose sur le fait que la durée d’incineration de deux enfants de 10 kg et d’une femme de 50 kg est égale a celle d’un homme de 70 kg, ce qui introduit un coefficient multiplicateur variant de 1 à 3, et rend aléatoires tous chiffres de rendement crématoire. (pp. 80-81)

[These official figures derive from false advertising but are nevertheless valid. Their apparent validity rests on the fact that the duration of the incineration of two children weighing 10 kg each and of a woman weighing 50 kg is equal to that for a man weighing 70 kg, which introduces a factor of multiplication varying from 1 to 3, and renders all crematory capacity figures uncertain.]

Capacity: The Facts

This reasoning is completely unfounded from both the technical and the documentary points of view. In this regard, we note the following:

1. The reference cited by Pressac is a letter from the Topf firm dated 1 November 1940 to the SS-Neubauleitung KL Mauthausen (note 9 on page 97). This document is a letter attached to a "cost estimate" (Kostenanschlag) of:

1 koksbeheizten Topf-Doppelmuffel-Einäscherungs-Ofen mit Druckluft-Anlage,
1 Topf-Zugverstärkungs-Anlage.

[One Topf two-chambered coke-heated crematory oven equipped with a compressed air ventilation system (DruckluftAnlage), one Topf draught booster system.] [7]

The oven offered is not the Dachau oven but rather the one installed in Crematory I at Auschwitz. This is evident not only from the "cost estimate" mentioned above but also from Topf technical design D57253 attached to the letter dated 10 June 1940, concerning precisely the first two-chambered oven of Crematory I at Auschwitz. This drawing is published by Pressac as Document 6.

Concerning the capacity of this model oven, one reads in the above letter:

Unser Herr Prüfer hatte Ihnen bereits mitgeteilt, dass in dem vorher angebotenen Ofen stündlich zwei Leichen zur Einäscherung kommen konnen.

[Our Mr.Prüfer had already communicated to you that in the oven presented above, it is possible to cremate two corpses per hour.] (Our italics)

2. As stated above, it is evident that the capacity of two corpses per hour refers not to the Dachau oven but to the "Auschwitz model" oven since "the oven presented above" is precisely that model.

3. The source cited by Pressac is the Topf letter of 6 January 1941 to the SS-Neubauleitung KL Mauthausen (note 25 on page 98). That the installation of compressed air reduced the duration of cremation is an arbitrary assumption by Pressac without any foundation in the text (or in reality). The text states:

Bei beiden Öfen haben wir berücksichtigt, dass die Generatorgase den Einäscherungsgegenstand von oben und unten angreifen, wodurch eine schnelle Einäscherung bewirkt wird.

[In both ovens the arrangement is such that the body being incinerated is attacked from above and below, thereby effecting a rapid cremation.] [8]

This letter refers to the two-chambered oven of the Auschwitz model, mentioned in technical drawing D57253, and to the coke-heated oven (drawing D58173) which was never installed, so that the "rapid cremation" (with respect to the civilian ovens) is nothing but the duration of one hour indicated by Prüfer in the 1 November 1940 letter. This "rapid cremation" depended upon the placement of the grill, made of fire-resistant clay, with respect to the opening of the connection to the gasogenes.

4. Pressac’s citation from the Topf letter of 14 July 1941 to the SS-Neubauleitung KL Mauthausen is correct, but Pressac apparently hasn’t the slightest idea of the meaning of this document. [9] This letter speaks of the incineration of 30 to 36 corpses in about ten hours in a two-chambered oven, corresponding to an incineration time of 33 to 40 minutes per corpse. These results could only be obtained under optimal conditions with the aid of an intake draft system (Saugzuganlage). The installation’s typical limit of efficiency for adult corpses was 40 minutes principal combustion in the cremation chamber, plus another 20 minutes of post-combustion in the ashpan underneath. This was altogether one hour, which even in the 1970s represented the minimum duration obtainable from gas ovens, as resulted from cremation experiments conducted in England. [10] The duration of 33 minutes (plus 20 minutes of post-combustion) could only be obtained in exceptional cases, and only for a short time. These data apply almost uniquely to the oven at Gusen, a Topf two-chambered mobile oven which was originally oil-heated and then transformed into a coke-heated oven like the first oven at Dachau with the installation of two lateral gasogenes (illustrated in Document 7 of Pressac). Because of local technical difficulties those data apply only theoretically to the ovens in Crematory I at Auschwitz.

The first cremation occurred on 15 August 1940 (p. 13). After only three months, on 22 November, the Bauleitung sent the Hauptamt Haushalt und Bauten in Berlin a letter which stated:

Der vergangene Betrieb des Krematoriums hat gezeigt, dass schon in der verhältnismässig guten Jahreszeit die Ofenanlage mit 2 Kammern zu klein ist.

[The past functioning of the crematory has shown that even in the relatively favorable times of the year, the oven with two (combustion) chambers is too small. (Therefore insufficient — Author)] [11]

According to Pressac, from March to December 1940 there were 2,000 deaths at Auschwitz (p. 146), an average of 8 per day; thus, the crematory at Auschwitz had difficulty in cremating 8 corpses per day! The letter in question is part of the Moscow documents from the Bauleitung of Auschwitz, but Pressac does not even mention it. His motive in excluding it is easily understandable.

5. The gas-generating ovens heated with coke required a daily rest for the cleaning of the furnaces, since the residue from the coke which melted and adhered to them, over a long period of time, would impede the passage of primary combustion air through the bars of the grill, causing poor operation of the crematory ovens. From a letter by Engineer H. Kori at the KL Lublin of 23 October 1941, [12] one deduces that the crematory ovens in the concentration camps were used only twenty hours at a stretch.

6. Accepting the data in the Topf letter dated 14 July 1941, the capacity of a two-chambered oven, over 21 hours of activity, would be:

30 corpses/10 hours × 21 hours = 63 corpses;
36 corpses/10 hours × 21 hours = 76 corpses.

so that the capacity for three ovens would be 63 × 3 = 189 and 76 × 3 = 228 corpses per day. Pressac unjustifiably estimates an excess of 200 to 250 corpses per day. We say unjustifiably, since from the very beginning, the data supply a maximum capacity for an oven with two chambers.

7. In a letter sent to Ludwig and Ernst-Wolfgang Topf dated 15 November 1942, [13] Engineer Prüfer indicates that the three-chambered ovens he designed which were installed in the crematory at Buchenwald had a yield greater by one-third than that which he expected. Here Pressac, who normally confuses capacity with yield, commits the opposite error by using yield for capacity. In effect, the greater yield depended upon a thermotechnical advantage of which Prüfer himself was not aware (maybe because he had designed the three- and eight-chambered ovens during his "free time" (Freizeit), as he writes in a letter to Topf dated 6 December 1941). [!3] [14] But this has nothing to do with capacity. Instead, Pressac interprets this as duration of cremation of the three-chambered oven as reduced by one-third compared to the two-chambered oven, which is technically meaningless, since the theoretical and effective heat availability of the two-chambered oven was greater than that of the three-chambered oven (about 210,000 kcal/h/chamber as opposed to 163,000 kcal/h/chamber; strictly in terms of combustion capacity of the grill, 30 kg/h/chamber as opposed to 23.3 kg/h/chamber).

Pressac arbitrarily places the closure of Crematory II at 22-23 May (p. 80).

8. But even if, for the sake of argument, the Pressac interpretation were correct, then it would follow that the maximum capacity of a three-chambered oven would be:

36 corpses/10 hours × 21 hours × 3/2 = 113.4 corpses per day,

therefore the capacity of five ovens would be:

113.4 × 5 = 567; 567 × 4/3 = 756 corpses per day.

But Pressac mentions an effective capacity of 800 corpses per day, which is then magically transformed into 1,000. Thus Pressac is not even consistent in his own technically errant presuppositions.

9. Pressac does not even attempt to justify in some way the capacity that he attributes to the eight-chambered oven, which is as technically unfounded as the capacity he attributes to the three-chambered oven.

10. The cremation of 45 fat adult corpses—three per chamber—in the five ovens of Crematory II at Birkenau in 40 minutes (reference from the witness H. Tauber) can be taken seriously only by those who have not the vaguest idea of the structure and operation of these ovens. First of all, the cremation time of one adult male corpse was an average of 60 minutes; secondly, the small combustion capacity of the grill with two furnaces, which was designed for the cremation of only one corpse at a time in each chamber, would have been insufficient to maintain a chamber temperature of 600°C (less than that required for a positive combustion of the heavy hydrocarbons which develop during the gasification of a corpse—at least 700°C) even assuming the simultaneous cremation of two corpses in each chamber; the simultaneous cremation of three corpses is a fortiori thermotechnically impossible.

11. Pressac’s reasoning, according to which all the capacity figures from the ovens at Auschwitz-Birkenau are aléatoire (uncertain) due to the presumed presence of small-sized corpses, is in reality a simple anticipated alibi: Not able to understand thermotechnical phenomena with which he is forced to deal, he does not want others to understand, and therefore decrees that any solution to the problem of capacity of the ovens is "uncertain". Even here, Pressac is mistaken. We have confronted and resolved the problem on the basis of the percentage of infants and children presumed homicidally gassed at Birkenau, by their age and average weight. The result is that the capacity of the ovens, for the assumed presence of infant and child corpses, would have been increased by twenty percent. Moreover, Pressac contradicts his own assertion, since he accepts H. Tauber’s story as true: in effect, that the cremation of 9 adult corpses in 40 minutes corresponds to a capacity of 1,417 adult corpses per 21-hour working day. [!4]

THE COKE

Pressac mentions absolutely nothing about the consumption of coke at Auschwitz-Birkenau.

THE OVENS

Before explaining the question of the crematory ovens, it is opportune to rapidly examine Pressac’s historical and technical assertions on this subject to furnish other elements with which to judge his competence and the value of his conclusions.

Pressac’s claim: The Volckmann-Ludwig system crematory oven went off the German market toward the end of 1934 (p. 4). Pressac begins his "recapitulative chronology" with the Volckmann-Ludwig patent (p. 110). He even presents a technical drawing as Document 2, which has nothing to do with the theme he has developed, evidently only to impress those who exalt him as expert incontesté on the subject of cremation. [15]

In fact: The H. R. Heinicke company, holder of the Volckmann-Ludwig patent, at that time had its headquarters at Chemnitz. They installed fifteen other ovens of this type in Germany between 1935 and 1940. [16]

Pressac’s claim: From the W. Müller oven of Allach, the SS deduced that a cremation without casket permitted a cremation time reduction to half an hour, and that 100 kg of coke were enough to cremate twenty corpses in one day (p. 6).

In fact: In a gasogene oven, heated with coke, the casket delayed corpse-water vaporization by 5 to 6 minutes, acting in a way as a thermal shield until breaking apart by the effect of the flames. Simultaneously, the heat produced by the casket, which raised the temperature of the chamber to 1,100°C, accelerated the vaporization process, therefore cremation without a casket did not take less time than cremation with a casket.

Regarding the consumption of coke in the gasogene ovens, incomparably the most important fact to be found in the specialized German literature of the time, is the cremation experiment conducted by Engineer Richard Kessler, one of the top specialists on cremation during the 1920s and 1930s. This experiment occurred 5 January 1927 in the Gebrüder Beck, Offenbach system oven, at the crematory of Dessau. [17] The results of the experiment, displayed in two thermotechnical diagrams, for each of the eight corpses cremated one after the other, were an average consumption of 29.5 kg of coke, plus the casket. These diagrams are of exceptional importance in understanding the operation of the gasogene crematory ovens. With the oven at thermal equilibrium (in a hypothesis of twenty consecutive cremations), the consumption of coke would have been reduced to 23 kg, plus the casket. A casket averaging 40 kg produced an actual quantity of heat equal to that produced by 15 kg of coke, therefore a cremation without a casket required about 38 kg of coke, and with 100 kg of coke three corpses could be cremated, not twenty. The "deduction" is evidently not that of the SS, but rather that of Pressac, and it is a very poor deduction.

Pressac’s claim: For him, the function of the intake draft was to:

augmenter la quantité de gaz de combustion et d’éviter ainsi une dépense de combustible supplémentaire lors de l’incinération des cadavres ‹glacés›. (p. 29)

[increase the quantity of combustion gas and thus avoid the waste of additional fuel when incinerating ‘frozen’ corpses.]

In fact: Here our author confuses the abilities of the Volckmann-Ludwig gas oven with the coke ovens of Auschwitz-Birkenau. Actually, in the gasogene oven heated by coke, the intake draught, impacting directly on the draft of the furnaces of the gasogene, caused an increase in the combustion capacity of the grill (the amount of coke burned hourly in the furnace), and consequently an increase in the consumption of coke.

Pressac’s claim: In Crematory I at Auschwitz, which had a chimney 15 meters high,

Koehler ajouta un carneau de liaison de 12 mètres pour obtenir une longueur de tirage de 27 mètres. (p. 40)

[Koehler added an exterior flue twelve meters long to obtain an intake of twenty-seven meters.]

In fact: In reality, the force of the intake of a chimney is determined by the height and cross-section of the flue above the grill. The working formula given by Engineer W. Heepke in his classic work on crematories [18] is precisely Z = 0.6 × H (for a fume temperature of 250°C), where Z is the force of the intake and H the height of the flue above the grill of the furnace. The length of the smoke conduit can only have a negative influence on the intake because too long a conduit would cool the fumes excessively.

Pressac’s claim: Pressac attributes the last page of a Kostenvoranschlag (preliminary cost estimate) by Topf for Auschwitz dated 1 April 1943 in the amount of 25,148 Reichsmark, published by R. Schabel, [19] to the planned Crematory VI,

fondé sur le principe de l’incinération à ciel ouvert. (p. 69)

[based on the principle of open-sky incineration],

In fact: Pressac’s interpretation (à ciel ouvert = "cremation pits") is unsustainable since the document in question mentions:

1 gusseiserner Rauchkanalschieber mit Rollen, Drahtseil und Handwinde

[one cast iron damper with pulleys, steel cable and hand winch],

and a smoke conduit infers a closed combustion chamber on one side and on the other a chimney; installations that would not be recommended for an à ciel ouvert (open-air) combustion.

THE FLAMES

Pressac’s claim: Pressac has the audacity to accept the story told by various eyewitnesses of flames coming out of the chimneys (of Crematories II and III) (p. 91).

In fact: This is technically impossible. Any uncombusted gas emitted from the chambers would either be burned in the smoke conduits if there were the necessary ignition temperature and combustion air, or, should these two conditions not exist, they would emerge from the oven uncombusted. In the first case, completely combusted gas would be emitted from the chimney (particularly nitrogen, carbon dioxide, water vapor, and a minimum amount of sulfur dioxide); in the second case, only smoke would emerge.

THE PITS

Pressac’s claim: Even the story of the cremation pits, similarly accepted by Pressac, is technically meaningless. The cremation of corpses in pits by the process described by the eyewitnesses is impossible due to the lack of oxygen in the lower portion of the pit. In 1871, the attempt to cremate the dead soldiers from the Battle of Sédan, by opening mass graves, filling them with tar and setting them on fire, resulted in charring of the uppermost layer of corpses, the baking of the intermediate layer, and no effect on the bottom layer. [20]

In fact: The Pressac technical drawings of the crematory ovens contain structural errors due to his lack of thermotechnical knowledge. Here are some examples:

Plan of the modified Dachau oven (p. 14): The connection of the two gasogenes to the chambers is incorrect (the products of combustion of the gasogenes were emitted in the posterior part of the chambers and discharged directly into the smoke conduit).

Plan of the three-chambered oven at Buchenwald (p. 28): The connection system of the gasogenes to the chambers is wrong (the two gasogenes were connected only to the two lateral chambers; the products of combustion penetrated the central chamber through the three inter-chamber openings that were found in the inner wall of the lateral chambers).

The plan of the rustique (rustic) three-chambered oven (p. 37) is a plan of the probable disposition of the two simplified three-chambered oven (p. 50): The oven only had one gasogene. The Kostenanschlag of 12 February 1942 mentions only one horizontal grill Planrost, [21] not two. The connection system of the two gasogenes to the three chambers through three connection apertures is wrong; the discharge system of the combusted gases is wrong. Due to the draft of the chimney, most of the combusted gases would have passed through the point of least resistance, that is, through the chamber closest to the smoke conduit.

Plan of the initial eight-chambered oven (p. 78): The discharge system of the combusted gases is wrong; the external chamber of every couple of chambers was connected to the horizontal smoke conduit through a vertical conduit placed in the wall of the posterior part of the chamber. Pressac places this conduit between the two chambers.

Plan of the reinforced eight-chambered oven (p. 78): The discharge system of the combusted gases is wrong; the discharge conduit situated by the gasogenes (at the right of the plan) did not exist.

Chapter Three

THE CREMATORY OVENS OF AUSCHWITZ-BIRKENAU
ACCORDING TO CREMATION TECHNOLOGY

Here we briefly review the results of our thermotechnical study on the Auschwitz-Birkenau crematory ovens.

1. Coke Consumption

The theoretical consumption of coke for an emaciated adult corpse in the two-chambered oven, is 27.8 kg, and for a normal corpse is 22.7 kg, according to the method calculated by Engineer W. Heepke [22]; the most thorough to be found in the German technical literature of the period. The real consumption of coke for the cremation of 677 corpses (presumably emaciated) in the two-chambered oven at Gusen, during a cremation period of 13 days, at an average of 52 corpses per day with the oven in permanent heat balance, was 20,700 kg [23], averaging 30.5 kg per corpse.

The consumption of the two-chambered oven, adjusted to reflect these experimental data, is 30.5 kg per emaciated corpse and 25 kg for a normal corpse.

The consumption of the three-chambered oven (with a reduction of one-third) is 20.3 kg for an emaciated corpse and 16.7 kg for a normal corpse.

The consumption of the eight-chambered oven (with a reduction of one-half) is 15.25 kg for an emaciated corpse and 12.5 kg for a normal corpse.

In the following table we summarize the coke consumption of the single crematory ovens at Auschwitz-Birkenau:

Type of oven

Emaciated Corpse

Normal Corpse

Two-Chambered Oven

30.5 kg

25 kg

Three-Chambered Oven

20.3 kg

16.7 kg

Eight-Chambered Oven

15.25 kg

12.5 kg

2. Capacity

The average cremation time of a continuously operating oven was about forty minutes of principal combustion (in the chamber), obtainable with the aid of the installation of an intake draft system (data relative to the Gusen oven).

The average time of a cremation without an intake draft system (taking into account the combustion capacity of the furnace grill) was sixty minutes, as is evident from the statement by Engineer Prüfer (in the 1 November 1940 letter), as well as from the diagrams published by Engineer R. Kessler concerning the principal combustion in the chamber (considering the structural differences of the Gebrüder Beck oven compared to those at Auschwitz-Birkenau). Since the Auschwitz-Birkenau ovens lacked draft intake installations (p. 81), the average time for a cremation (principal combustion in the chamber) was one hour. The continuous operation of the ovens was 20 hours per day, at the most. Therefore, the capacity of the single crematories was as follows:

Site

Corpses per day

Crematory I

120 Corpses per day [!5]

Crematory II

300 Corpses per day

Crematory III

300 Corpses per day

Crematory IV

160 Corpses per day

Crematory V

160 Corpses per day

Totals:

1,040 Corpses per day

Supposing the reality of homicidal gassings, considering the percentage of small-sized bodies among the corpses, as well as average weight as a function of age, the daily capacity would have increased by 6/5: Column I. In Column II, we include the data from the 28 June 1943 letter which Pressac considers valid in the cremation hypothesis of infants and children, as follows:

Site

I

II

Crematory I

144

340

Crematory II

360

1,440

Crematory III

360

1,440

Crematory IV

192

768

Crematory V

192

768

Totals:

1,248

4,756

Since in twenty hours the ovens altogether could burn (based on the combustion capacity of the single oven’s grill) 23,200 kg of coke, the average coke consumption for each corpseaccording to Jean-Claude Pressacwould be (23,200 / 4,756 =) 4.88 kg, which is thermotechnically impossible.

3. The Reason For Existence Of Crematories

The decision to build three more crematories at Birkenau was made on 19 August 1942 (p. 49), after Himmler, during his inspection of Auschwitz on 17 and 18 July 1942, had ordered that the actual forecast for the KGL (prisoner of war camp) at Birkenau be increased from 125,000 to 200,000 prisoners (p. 44). It also came during the terrible typhus epidemic in the summer of 1942, which caused decimation in the Auschwitz-Birkenau camp: In the male sector alone, from 1 to 19 August, 4,113 deaths were registered, [24] on the average 216 per day. In the third trimester of 1942, the mortality was 20.5% of the average camp population, [25] which did not exceed 25,000 units. The capacity of the crematories was therefore quite adequate for the camp population established by Himmler, and provided for a possible future typhus epidemic.

4. The Number Of Cremations In 1943:
The Estimate Of The SS.

The Aktenvermerk (file entry for the records) of 17 March 1943 [26] (mentioned by Pressac on p. 119) shows the coke consumption estimate for the four crematories at Birkenau. The operational time of the crematories is estimated at 12 hours. The letter indicates the combustion capacity of the grills of the ovens; therefore one is able to calculate the number of corpses that could possibly be cremated, namely, about 362 emaciated adult corpses per day. From 1 March to 17 March the average mortality at Birkenau was 292 prisoners per day, [27] which, in terms of coke consumption for cremation, represents 80% of the estimate by the SS. This means that this estimate was calculated on the basis of the actual average mortality plus a 20% security margin. Since the average of the presumed gassed according to the Kalendarium of Auschwitz during this period is about 1,100 per day, if the story of the homicidal gas chambers were true, during this period there would have been an average daily death toll of about 1,400, approximately four times the actual estimate of the SS.

5. The Number Cremated In 1943: Coke Consumption

From 1 March to 25 October 1943 a total of 641.5 tons of coke was supplied to the crematories of Auschwitz-Birkenau. [28] In this period, the number of natural deaths among the prisoners was about 27,300, that of the presumed gassed about 118,300, [29] thus altogether about 145,600. For the prisoners deceased of natural causes, there results an average coke availability of (641,500 / 27,300 =) 23.5 kg per corpse, a figure which is quite compatible with the consumption of the ovens. For the presumed homicidally gassed, plus the prisoners deceased of natural causes, instead there results an availability of (641,000 / 145,600 =) 4.4 kg, which is thermotechnically impossible.

The estimate of the SS of 17 March 1943, and the quantity of coke supplied to the crematories from March to October 1943, demonstrate that the crematories cremated only the corpses of the registered prisoners deceased of natural causes and that, consequently, there was no mass homicidal gassing.

6. The Cremation Capacity of the Crematories in 1943

From 14 March to 25 October 1943, the crematories at Birkenau were able to function only for a total of about 400 days. The maximum number of cremations theoretically possible (with the capacities taking into account corpses of babies and children) is about 100,000, but the number of corpses to be cremated (presumed homicidally gassed plus registered prisoners) is about 142,000. Thus, the cremation capability of the crematories rendered the cremation of presumed homicidally gassed prisoners impossible; therefore, there was no mass homicidal gassing.

In Auschwitz: Technique and Operation of the Gas Chambers, Jean-Claude Pressac states that from April to October 1943 the crematories at Birkenau cremated 165,000 to 215,000 corpses with 497 tons of coke, [30] which means that for him it was possible to cremate a corpse with an average of 2.6 kg of coke!

According to Rudolf Höss, Crematories II and III could cremate 2,000 corpses per day, Crematories IV and V, 1,500 per day. [31] Hence the average consumption of coke per corpse was respectively 3.5 kg and 1.8 kg!

7. The Duration of the Fire Resistant Walls
of the Crematory Ovens

In his 1989 book, Jean-Claude Pressac furnishes the following numbers of corpses cremated at Auschwitz-Birkenau:

Site

Cremations

Crematory I

10,000 Cremated [32]

Crematory II

400,000 Cremated [33]

Crematory III

350,000 Cremated [34]

Crematory IV

6,000 Cremated [35]

Crematory V

15,000 Cremated [36]

Subtotal:

781,000 Cremated

Cremation Pit 1942

107,000 Cremated

Cremation Pit 1944

50,000 Cremated

Subtotal:

157,000 Cremated

Total:

938,000 Cremated

These numbers refer exclusively to those presumed homicidally gassed and do not include the corpses of the registered prisoners who died of natural causes.

In his 1993 book, Jean-Claude Pressac reduces the number of presumed homicidally gassed to 630,000, and furnishes a total general death toll of 775,000, rounded off to approximately 800,000 (p. 148).

This revision of the number of those alleged to have been homicidally gassed has no relation to the Moscow documents studied by Pressac. The reduction is dictated exclusively by his realization that the Birkenau crematories in 1943 (and mostly in Spring and Summer of 1944: see No. 8) could not have cremated the corpses of the presumed homicidally gassed even with the inflated capacity numbers he adopted. To eliminate the contradiction, he decrees that the number of the deported brought to Auschwitz-Birkenau at this time according to the Auschwitz Kalendarium (about 53,000) and, consequently, of the alleged homicidally gassed (about 42,000) is excessive (p. 147). Now Pressac, on the basis of simple conjecture, expects to "correct" in one little page, the recent study by Franciszek Piper on the number of victims of Auschwitz-Birkenau, [39] which is the most in-depth and documented Exterminationist work, compiled from the documents in possession of the museum of Auschwitz. Therefore, from the point of view of the supporters of the reality of homicidal gas chambers, the reference work remains that of Piper, and the affirmations of Pressac are mere unfounded conjectures.

All the same, even the new number of cremations adopted by Pressac is technologically impossible. He reduces the number of those allegedly cremated in the open air in 1942 from 107,000 to 50,000, and no longer in "cremation pits" but on pyres. For 1944, he does not furnish any numbers. Thus we take for valid those given in his 1989 book: 50,000. Therefore, of the 775,000 cremated, about 100,000 were cremated in the open, and the remaining 675,000 in the crematories.

The engineer R. Jakobskötter, speaking in 1941 of the Topf ovens heated with electricity in the crematory of Erfurt, states that the second oven was able to perform 3,000 cremations, while the normal duration of the refractory walls of the ovens was 2,000 cremations. [40] The Gusen oven lasted for 3,200 cremations, [41] after which it was necessary to dismantle it and replace the fire-resistant walls. [42] The duration of one chamber was therefore about 1,600 cremations. Now, even supposing that the ovens of Auschwitz-Birkenau were used to the extreme limit of 3,000 cremations per chamber, altogether they would have been able to cremate about 156,000 corpses (according to Pressac, the total number of the victims among the registered prisoners was 130,000 (p. 146)), while the cremation of 675,000 corpses would have required at least four complete substitutions of the fire-resistant walls of all the chambers. This means that for Crematories II and III alone, 256 tons of fire-resistant wall material would have been necessary (not counting that for the gasogenes), with a labor time (based on that required at Gusen) of about 7,200 hours.

All the same, in the archives of the Bauleitung that were left "intact" by the SS of Auschwitz, which Pressac has examined in their entirety, there is not a trace of these enormous projects, which means that they were never carried out, for in these records, there exists, to give an example, "un réglement de comptes administratifs acharné" ["a tenacious administrative settlement of accounts"] between the Bauleitung and Topf of even a meager payment of 828 marks (p. 59).

The cremation of 675,000 corpses is technologically impossible, and consequently no mass extermination was perpetrated at Auschwitz-Birkenau.

8. The Deportation and "Extermination"
of the Hungarian Jews

Jean-Claude Pressac, embarrassed by the technical impossibility of a mass cremation of the Hungarian Jews alleged to have been gassed, plays at reduction, declaring that of the approximately 438,000 deportees to Auschwitz-Birkenau, 146,000 were able to work and therefore were saved; the 292,000 remaining were incapable of work, and were gassed (p. 147). He refers to the statistical estimates of G. Wellers, who represents Establishment orthodoxy on the matter, and according to whom the number of people gassed was 410,000 (p. 147).

From these, Pressac calculates that with a total capacity of Crematories II, III, and IV, and of the "cremation pits" of 3,300 corpses per day, with the possibility of an extension to 4,300 (Pressac does not say how), "les SS pouvaient anéantir en 70 jour jusqu’à 300.000 personnes" (p. 148) ["the SS could annihilate up to 300,000 people in seventy days."]

Regarding the first point, since Pressac does not furnish any proof of the transfer of (146,000 - 28,000 registered [prisoners] =) 118,000 Hungarian Jews from Auschwitz, with the same logic one can claim that (438,000 - 28,000 =) 410,000 Hungarian Jews were transferred from Auschwitz, and thus did not undergo extermination.

Regarding the second point, we notice immediately the capacity indicated by Pressac is technically impossible: Crematories II, III, and V, would have been able to cremate at the most 900 corpses, while the "cremation pits", as we have indicated, are a technical absurdity.

Nevertheless, the deportation of the Hungarian Jews took place from 15 May to 8 July 1944, in a time frame of 54 days, not 70; therefore, even assuming the maximum capacity of 4,300 corpses per day (54 × 4,300), it would have been possible to cremate 232,200 corpses, not 292,000. As a matter of fact, after eliminating the pauses between the various waves of deportation and the actual days of deportation and arrival of deportees to Auschwitz, which were 39, [43] the installations at Birkenau would have been capable of cremating 39 × 4,300 = 167,700 corpses.— And where would the remaining 124,300 corpses have been put?

It is pointed out also that the aerial photographs taken by the Allied military on 31 May 1944, [44] at the crucial time of presumed extermination, on the day of the arrival at Birkenau of about 15,000 deportees, and after 14 days of intense arrivals (184,000 deportees, averaging 13,000 per day) and with an extermination toll (according to Pressac’s hypothesis) of at least 110,000 homicidally gassed, which would have had to average 7,800 per day, every single day for 14 consecutive days; after all of that, the photographs do not show the slightest evidence of this alleged enormous extermination: No trace of smoke, no trace of pits, crematory or otherwise, burning or not, no sign of dirt extracted from pits, no trace of wood set aside for use in pits, no sign of vehicles or of any other type of activity in the crucial zones of the courtyard of Crematory V nor in the earth of Bunker 2, nor in Crematories II and III. These photographs [45] constitute irrefutable proof that the story of extermination of the Hungarian Jews is historically unfounded.

Chapter Four

GENESIS AND DEVELOPMENT OF
THE "FINAL SOLUTION" [46]

Jean-Claude Pressac states that the last stage of the "final solution"

ne fut decidée par les autorités SS de Berlin qu’à partir de mai-juin 1942, pour être ensuite concretisée techniquement par les SS de la Bauleitung D’Auschwitz et les ingénieurs de la firme J.A. Topf und Söhne d’Erfurt. (p. 2)

[was not decided on by the SS authorities in Berlin until after May-June 1942, and only subsequently concretized technically by the SS of the Bauleitung of Auschwitz and the engineers of the firm J.A. Topf and Sons of Erfurt.]

Regarding this, Pressac reports:

début juin 1942, Himmler ayant convoqué Höss à Berlin, l’informa du choix de son camp comme centre d’anéantissement massif des Juifs. Le chef des SS avait retenu Auschwitz parce que sa situation ferroviaire était favorable et que le camp serait bientôt pourvu d’un extraordinaire crématoire pouvant incinérer 1.400 corps par jour (épisode que Höss place faussement à l’été 1941 comme d’ailleurs Eichmann après l’avoir lu dans les écrites de ce dernier). L’action commencerait le 1er juillet et il fallait qu’a cette date tout soit prêt pour l’exécuter. (p. 41, our italics)

[at the beginning of June 1942, Himmler summoned Höss to Berlin, and informed him of the choice of his camp as center for the massive annihilation of the Jews. The chief of the SS had chosen Auschwitz because of its favorable situation close to the railway, and because the camp would soon be provided with an extraordinary crematory capable of cremating 1,400 bodies per day (an episode that Höss wrongly places in the summer of 1941, which Eichmann also does after having read Höss’s writings). The action would begin 1 July and everything would have to be ready to execute it by this date.]

In reality, it is Pressac who wrongly places in 1942 an event which, according to the chronological and logical developments of the events referred to by Höss, could only have taken place in 1941.

We summarize the following development:

[Höss] In the summer of 1941—at the moment I am not able to cite the exact date—I was suddenly summoned to Berlin to the Reichsführer [Himmler], by his Aide.

[Himmler] The Führer has ordered the final solution of the Jewish question, and we SS must execute this order.

I chose Auschwitz, both for its optimal communications position and because the adjacent land can be easily isolated and camouflaged.

You will learn further details from Sturmbannführer Eichmann, of the RSHA, whom I will send to you shortly.

[Höss] Soon thereafter, Eichmann came to me at Auschwitz, where he revealed to me the plan of action for the various countries.

Hence, we went on to discuss the means of effecting the extermination plan. The means could only be gas...

Eichmann promised that he would inform himself on the existence of an easily produced gas which would not require any particular installations, and that he would later inform me on the matter.

We went to inspect the area to establish the best location, and ascertained that it was the farm [the future Bunker 1] situated at the northwest corner of the future third sector of buildings, Birkenau. [!6]

Eichmann then returned to Berlin, to report to him [Himmler] the content of our conversation.

At the end of November in Eichmann’s office in Berlin a conference was held on the entire Jewish Section, at which I was invited to participate...

We were not told when the action would begin, nor had Eichmann been able to find the appropriate gas.

In the autumn of 1941, by means of a secret order given to the prisoner of war camps, the Gestapo separated all the politruks, the commissars and some political officials and sent them to the closest concentration camp, to have them liquidated.

At Auschwitz small transports of these men were continuously arriving; they were then shot in the gravel pit by the buildings of the [tobacco] monopoly, or in the courtyard of Block II [Block 11].

On the occasion of one of my service trips, my substitute, Hauptsturmführer Fritzsch, on his own initiative, used gas to exterminate these prisoners of war; he filled the cells located in the basement to overflowing with Russians and, protecting himself with a gasmask, had Cyklon B [Zyklon B] infused, which provoked the immediate death of the victims.

On Eichmann’s next visit, I reported to him on the use of Zyklon B, and we decided that it would be the gas used for the imminent mass slaughter.

The killing of the Russian prisoners of war with Cyklon B, which I referred to above, continued, but no longer in Block II, because after the gassing, the entire building would have to be aired out for at least two days. As a result, the mortuary room of the crematory near the hospital was used, after the doors were rendered gas proof, and gas emission holes were opened in the ceiling.

I wouldn’t know at what period exactly the extermination of the Jews began; probably already in September 1941, but perhaps only in January 1942. [47]

It is thus clear that the presumed summons of Höss to Berlin came before the first alleged homicidal gassing in the Bunker of Block 11 (and before the successive gassings in the Leichenhalle of Crematory I at Auschwitz); but since Pressac places this event entre le cinque et la fin décembre 1941 (p. 34), it is just as clear that the summons dates back to the summer of 1941, not to 1942.

Himmler’s second motive for choosing Auschwitz does not find any verification in the "Annotations" (Aufzeichnungen) and in the sworn testimonies of R. Höss, but is the simple fruit of the fantasy of Jean-Claude Pressac, which is equal to imagining both the precise date of the presumed summons (at the beginning of June), as well as the date of the beginning of extermination (1 July).

One may ask why Pressac would begin his book with these manipulations. The answer is simple: Not having found any evidence in Moscow of criminal aims in the plans of the crematories of Birkenau, and being forced to admit that these crematories were initially foreseen sans chambres à gas homicides (without homicidal gas chambers) (p. 53), he had to postpone the alleged decision to exterminate the Jews by one year, since otherwise the planning of four crematories without gas chambers in the very place destined to be the principal center of such an extermination would have appeared too unlikely. Pressac falls into contradiction nevertheless, because he places the beginning of the homicidal activities in Bunker 1 at the end of May 1942 (p. 39), that is, before R. Höss knew or received the alleged extermination order from Himmler (beginning of June). Here, it is opportune to remember that the beginning of the activity at Bunker 1 was placed in January 1942 in the first edition of Kalendarium of Auschwitz; [48] in the second edition, it is moved to March.[49] Pressac finally moves it to the end of May. In all three cases this was done without any proof of any kind. In addition, since the second edition of Kalendarium of Auschwitz places the beginning of the activity in Bunker 2 at 30 June 1942, [50] it is evident that R. Höss theoretically must have been summoned to Berlin at the beginning of June, and it matters little that R. Höss never specifies it: Jean-Claude Pressac decrees it authoritatively.

Thus it is clear that the declaration of R. Höss on the presumed summons to Berlin by Himmler in June 1941 upsets Pressac’s reasoning from the very beginning.

Granting this, let us now follow this reasoning in its successive logical and chronological development.

As indicated, Pressac places the first homicidal gassing between 5 December 1941 and the end of that month (p. 34). Regarding this, Pressac writes:

D’après Höss (qui n’y assista pas) la mort aurait été immédiate. D’autres parlent d’un gazage ayant duré deux jours avec introduction d’une seconde quantité de toxique parce que la première n’avait pas tué tout le monde. De l’acide cyanhydrique, se vaporisant à 27°C, utilisé dans un sous-sol non encore chauffé en plein hiver silésien e une méconnaissance de la dose létale pourraient expliquer la durée anormale de ce gazage. (p. 34, our italics)

[According to Höss (who did not take part), death would have been immediate. Others speak of a gassing having lasted two days, with the introduction of a second toxic quantity because the first did not kill everyone. Hydrocyanic acid, vaporizing at 27°C, used in an as yet unheated basement in full Silesian winter and a misreading of the lethal dose could explain the abnormal duration of this gassing.]

It is known that this presumed event was supposed to have occured, according to the Auschwitz Kalendarium, on the basis of various eyewitnesses, between 3 and 5 September 1941. [51] The Polish historian S. Klodziński, with the help of responses to a questionnaire on the first alleged gassing which he sent to 250 former prisoners of Auschwitz who were registered before September 1941, alters the date of this gassing to between 5 and 9 September 1941. [52] Pressac, who in 1989 still followed the Auschwitz Kalendarium literally, [53] moves the date by at least three months: On what basis? On the basis of our 190-page study Auschwitz: La prima gasazione, [54] in which we show that this event has no historic foundation since it is not supported by any document, but is instead contradicted by the available documents, and finally, that all the testimonies on this subject are contradictory on all their essential points. Pressac, instead of accepting this inevitable conclusion, assumes one of our polemical observations is true, [55] then decrees that this supposed event not only has a historic reality, but de nos jours (at the present time) (p. 34) its official date is that which he has indicated.

Here Pressac gives one of many examples of his captious methodology. The story of the introduction of a second quantity of Zyklon B comes from the testimony of M. Kula (deposition of 11 June 1945), [56] which, however, places the first homicidal gassing with absolute certainty in August 1941:

"According to my information, the first gassing took place the night of the 14-15 and the day of the 15th of August 1941, in the Bunkers of Block 11. I remember this date exactly, because it coincides with the anniversary of my arrival at the camp, and because then the first Russian prisoners of war were gassed." [57]

M. Kula is the witness from whom Pressac takes the story of the four wire-latticed columns (quatre colonnes grillagées) for introducing Zyklon B in the supposed gas chambers of Crematories II and III (p. 74). [58]

Pressac’s explanation for the "abnormal" duration of the supposed gassing is the cold temperature and ignorance of the lethal dose. These claims are already discredited in our book on the basis of testimony from the same witnesses:

(witness Glowacki: "there was a tremendous heat";

witness Kielar: the air was very "suffocating, hot").

They are also refuted by practical experiences during disinfestation of the barracks with Zyklon B during the years 1940 to 1941 at times when the local temperatures were between -4°C and -8°C, during which the gas developed to its maximum extent after one or two hours. [59] Finally, in this work we have shown that the lethal dose of hydrocyanic acid for human beings was perfectly well known since the 1930s, on the basis of Schädliche Gase, Dämpfe, Nebel, Rauch und Staubarten (Verlag von Julius Springer, Berlin 1931), the classic by Ferdinand Flury and Franz Zernik. This is in direct contradiction to the statements of Jean-Claude Pressac (see above). [60]

Pressac’s thesis involves another contradiction, because this supposed event, predating by at least five months the decision to exterminate the Jews, evidently has no connection with this, any more than do the successive gassings in the Leichenhalle of Crematory I of Auschwitz, from January 1941. All the same, Pressac affirms that at the end of April, difficulties having arisen, it was decided de transférer ce genre d’activité à Birkenau (p. 35); in other words, it was decided to put into operation Bunker 1, which however was linked with the extermination of the Jews.

Here Pressac therefore ruptures the consecutive logic (i.e. logically admissible, but historically false) which he put forth in his 1989 book:

Because the lethal dose for humans was not known, the SS had made a botched trial gassing in the basement of Bunker 11 of the Stammlager on 3, 4, and 5 September 1941, the victims being 850 Soviet POWs and other prisoners.

It was subsequently seen to be more convenient to gas people as required in the very place where all corpses inevitably had to go eventually: the morgue of Krematorium I.

But trials to perfect the technique could not be carried out in this crematory attached to the camp, hence the idea of establishing Bunker 1 in an isolated location on the edge of Birkenau wood. [61]

As regards the "final solution," Pressac inflicts the final blow to the traditional interpretation of the Wannsee conference declaring in this connection:

Le 20 janvier [1942], se tenait à Berlin la conférence dite de Wannsee. Si une action de ‹refoulement› des Juifs vers l’Est fut bien prévue avec l’évocation d’une élimination ‹naturelle› par le travail, personne ne parla alors de liquidation industrielle. Dans les jours et les semaines qui suivirent, la Bauleitung d’Auschwitz ne reçut ni appel, ni télégramme, ni lettre réclamant l’étude d’une installation adaptée à cette fin. (p. 35, our italics)

[January 20, 1942, the so-called Wannsee conference was held in Berlin. If an operation to ‘expel’ the Jews towards the East was indeed anticipated with the evocation of a ‘natural’ elimination by work, no one spoke then of industrial liquidation. In the days and weeks that followed, the Bauleitung of Auschwitz received neither appeal, nor telegram, nor letter calling for an installation adapted to this end.]

The story of this supposed "final solution," begun with a verbal order from Himmler, could only have been concluded with another verbal order:

Fin novembre [1944], sur ordre verbal d’Himmler, les gazages homicides furent arrêtés. (p. 93)

[The end of November (1944), on the verbal order of Himmler, the homicidal gassings were stopped.]

Needless to say, there is no proof of the existence of this "verbal order". [62]

Chapter Five

CREMATORIES II AND III

Jean-Claude Pressac affirms that a capacity of 1,440 corpses per day was foreseen for the new crematory destined for the principal camp, which later became the prototype of Crematories II and III at Birkenau (p. 28). As we have seen, that is precisely what would have induced Himmler to choose Auschwitz for carrying out the extermination of the Jews (p. 41); nevertheless he specifies that:

bien que le crématoire II ait servi de catalyseur pour la choix d’Auschwitz dans la liquidation des Juifs, il ne se rattache pas directement à cette extermination, mais est considéré comme un moyen supplétif occasionnel; le crématoire III n’est projeté qu’en complément du II, pour faire face à un effectif de 200,000 détenus, et n’est ‹criminalisé› que pour les besoins de la burocratie SS. (pp. 54-55)

[Although Crematory II has served as a catalyst for the choice of Auschwitz in the liquidation of the Jews, it is not linked directly to this extermination, but is considered as an occasional supplementary means; Crematory III is planned only as an extension of II, to deal with an effective 200,000 prisoners, and is only ‘criminalized’ by the needs of the SS bureaucracy.]

Crematory III had a "sanitary orientation" (p. 50), equal to Crematory II, of which it was naturally the extension; the structures of Crematory II and III were not planned with the intention of homicidal gassing (p. 63), and none of the four crematories at Birkenau initially anticipated homicidal gas chambers (p. 53). Crematories IV and V instead "belonged to Bunkers 1 and 2" (p. 50); they were "destined to Bunkers 1 and 2" (p. 52), "connected directly to Bunkers 1 and 2" (p. 54).

So according to Pressac, Crematories II and III initially had a normal hygienic-sanitary function, while Crematories IV and V (which were not equipped with homicidal gas chambers) had a criminal function because they were supposed to have cremated alleged homicidally gassed cadavers of Bunkers 1 and 2.

This results in the meaningless conclusion that the Auschwitz Bauleitung technicians assigned thirty chambers (presumed capacity: 2,880 corpses per day) to the normal health measures of the camp, and only sixteen chambers to mass extermination. In other words, they expected more corpses from the natural mortality of the camp than from mass extermination!

Another Pressac conclusion, even more revealing, is that Auschwitz was chosen by Himmler to commence extermination of Jews because of the design of the new crematory, which Pressac presumes could cremate 1,440 corpses per day. The Bauleitung technicians however, instead of making this crematory and its twin Crematory III the fulcrum of the extermination, turned to two other crematories of distinctly lower capacity!

Regarding the genesis of the other three Crematories (III, IV and V), Pressac writes:

Le 19 [août] est à considérer comme la date où fut entérinée la décision de construire trois autres crématoires à Birkenau, dont deux étaient liés directement au processus criminel d’anéantissement des Juifs. (p. 49)

[The 19th (August) is to be considered the date when the decision to build three other crematories at Birkenau was ratified, of which two were linked directly to the criminal process of annihilation of the Jews.]

But 19 August is also a date on which a terrible typhus epidemic raged at Auschwitz-Birkenau that decimated the prisoners of the camp.

Pressac himself admits that:

la pression de l’épidemie de typhus, avec chaque jour ses 250 à 300 morts parmi les détenus, ainsi que les civils et SS les accompagnant dans l’au-delà, conjuguée aux incessantes arrivées de convois juifs, poussa Bischoff, sur demande de Höss, à précipiter la réalisation du programme crématoire et à le doubler. (p. 50)

[the pressure of the typhus epidemic, with its 250 to 300 deaths among the prisoners each day, as well as the civilians and SS accompanying them to the hereafter, plus the incessant arrivals of Jewish convoys, pushed Bischoff, on Höss’s orders, to expedite the crematory program.]

Actually, the decision to construct four crematories at Birkenau depended exclusively on the terror which the typhus epidemic provoked in the SS, [63] in light of the proposed expansion of the camp’s population by about ten times.

Pressac shows that the installations and the precautions of the SS, which were aimed at halting the typhus epidemic, bore the designation special (Sonder-) in the SS terminology.

Doctor Wirths:

prévoyait un retour du typhus si des ‹mesures spéciales› (Sondermassnahmen) pour améliorer la situation sanitaire n’étaient pas prises d’urgence. (p. 82, our italics)

[foresaw a return of the typhus if some ‘special measures’ (Sondermassnahmen) to improve the sanitary situation were not taken urgently.]

Pressac explains correctly that:

les termes ‹mesures spéciales/Sondermassnahmen› et ‹mesures de construction spéciales/Sonderbaumassnahmen› désignent des dispositions liées aux questions sanitaires ou aux bâtiments s’y rapportant (par exemple, alimentation en eau, mesures d’hygiène appliquées aux détenus, etc. (p. 107, note 256, our italics)

[the terms ‘special measures/Sondermassnahmen’ and ‘special measures of construction/Sonderbaumassnahmen’ designate the dispositions relating to the sanitary issues or to the buildings if involved (for example, water supply, hygienic measures applied to the prisoners, etc.]

With the construction of the disinfection and disinfestation installation of the Zentralsauna:

les SS voulaient contrer ‹définitivement› toute résurgence du typhus à Birkenau. (p. 69)

[the SS wanted to ‘definitively’ counter any resurgence of the typhus at Birkenau.]

Now the Zentralsauna was in fact part of the Sonderbaumassnahmen (p. 107, note 256) and, as with the crematories, was involved in the Durchführung der Sonderbehandlung (implementation of such special measures) (p. 61).

If, therefore, the construction of the new crematory had a purely sanitary purpose (by Pressac’s admission), it should have been undertaken immediately "en raison de la situation crée par les ‹actions spéciales›", as is stated in a document of the end of July 1942 (p. 47), a time of full typhus epidemic; it is clear that these "special actions" (Sonderaktionen) were linked to the fight against the epidemic and had no criminal connotation.

And, if the same document mentions 4 Stück Baracken für Sonderbehandlung der Häftlinge in Birkenau (p. 46, our italics) ["4 barracks for the special treatment of prisoners at Birkenau"], it is equally clear that even this "special treatment," reserved for the registered prisoners of the camp, referred simply to the "mesures d’hygiène appliquées aux détenus" ["sanitary measures put into practice for the prisoners"], to use the words of Jean-Claude Pressac.

And, if finally on 26 August 1942, during a full typhus epidemic, Zyklon B was picked up at Dessau für Sonderbehandlung, it is still clear that it served disinfestation purposes (p. 47). [!7]

Self-contradictingly, Pressac instead claims that the Sonderbehandlung was a convenient term that designated la liquidation par le gaz des inaptes juifs à Birkenau (the liquidation by gas of the unfit Jews at Birkenau.) (p. 46)

He further specifies that:

l’acte de mise à mort en lui-même était dit ‹traitement spécial› ou ‹transfert de population juive› alors que la globalité de l’opération, incluant sélection, transport des inaptes et gazage homicide, s’énonçait ‹action spécial›, terme n’étant pourtant pas spécifiquement criminel car pouvant s’appliquer à une opération qui ne l’était pas. (p. 46)

[the act of putting to death itself, was called ‘special treatment’ or ‘transfer of Jewish population’, since the totality of the operation, including selection, transfer of the unfit and homicidal gassing, was termed ‘special action’, a term not specifically criminal, since it could apply to an operation that was not.]

Pressac himself mentions a case in which the term Sonderaktion had no criminal meaning; after the strike (in an extermination camp!) of the civilian workers on 17 and 18 December 1942, the Gestapo made a Sonderaktion consisting of the interrogation of the workers to find out what had produced the refusal to work (p. 63).

Neither Sondermassnahmen, nor Sonderbaumassnahmen, nor Sonderbehandlung, nor Sonderaktion had a criminal significance and Jean-Claude Pressac does not introduce a single document to support the contrary. Therefore, his deductions have no foundation.

What Pressac writes about the supply of Zyklon B to Auschwitz is really a bit incredible. He pretends that the SS-WVHA knew nothing about the typhus epidemic that raged at Auschwitz-Birkenau in 1942. This epidemic required enormous quantities of Zyklon B for the purpose of disinfestation. He maintains that the administration of the camp, which did not want the WVHA to know of the epidemic, could not request Zyklon B without betraying itself. This is how he extricates himself:

Une astuce fut trouvée. Mettre sur le dos des Juifs les effarantes quantités de gaz employées. L’autorisation de transport accordée le 26 août le fut pour ‹traitement spécial›. Bien que les responsables du SS-WVHA de Berlin sussent la finalité du ‹traitement›, ils en ignoraient les modalités, c’est-à-dire les quantités de toxique utilisées. Ce qui permit de leur faire croire que la majorité du Zyklon-B livré servait aux gazage homicides dans les Bunker 1 et 2, alors que 2 à 3% suffisait. Ainsi, 97 à 98% du gaz pouvait être consacré à l’épouillage. (p. 47)

[A ruse was devised. The blame for the bewildering quantities of gas used was placed on the Jews. The shipment authorization given on 26 August was for ‘special treatment’. Although the authorities of the SS-WVHA [!8] of Berlin knew the end result of the ‘treatment’, they were unaware of the specifics, that is, the quantities of toxin used. Since 2% to 3% was enough, this allowed them to make others think that the majority of the Zyklon B delivered was needed for homicidal gassings in Bunker 1 and 2. In this manner, 97% to 98% of the gas could be used for delousing.]

The goal of this reasoning is quite clear: Since the purpose for the ordering Zyklon B are apparently twofold—on the one hand the "special treatment" (Sonderbehandlung: procurement of 26 August) and the "resettlement of Jews" (Judenumsiedlung: 2 October), terms that Pressac interprets in the criminal sense; on the other hand, the disinfestation (Gas zur Desinfektion des Lagers [gas for disinfesting of the camp]: 29 July) [64] there would exist two types of procurements, bureaucratically defined: one for gassing Jews, the other for disinfestation of the camp. But in this case, the quantity of Zyklon B for a homicidal purpose would be enormous (the request of 2 October alone refers to 5 tons gross), in contradiction to the thesis advanced by Pressac in his 1989 book, that only 2-3% of the Zyklon B supplied to Auschwitz was used for the purpose of homicide. [65] To overcome this contradiction, Pressac has found nothing better than to assert that the request for Zyklon B, supposedly for the purpose of homicide (Sonderbehandlung, Judenumsiedlung), was in reality only a camouflage for a request with a sanitary purpose! Just how far the WVHA was unaware of the epidemic of typhus that raged at Auschwitz can be deduced from the fact that Doctor Wirths, who on 6 September 1942 had been

nommé médicin-chef de la garrison du camp d’Auschwitz afin d’enrayer l’épidémie de typhus, (p. 116)

[named chief doctor of the garrison of the camp at Auschwitz in order to retard the epidemic of typhus,]

came from the Inspectorate of the Concentration Camps, [66] that is, from the Amtsgruppe D of the WVHA.

Realizing the naivete of his reasoning, Pressac seeks to render the picture he has drawn more credible by fabricating the purpose of Pohl’s visit of 23 September 1942 to Auschwitz:

Le chef du SS-WVHA, le géneral de corps d’armée SS Pohl, se présenta à l’improviste dans la matinée du 23 septembre 1942 à Auschwitz pour savoir ce qui s’y passait et où filaient les tonnes de Zyklon-B accordées. Pohl se rendit d’abord à la Bauleitung, se fit expliquer l’implantation générale du camp, décrire les batiments achevés, en cours (dont les quatre crématoires de Birkenau) et en projet. A sa question sur le Zyklon-B, il lui fut répondu qu’avec ce produit, on détruisait à la fois les poux et les Juifs. (p. 59, our italics)

[The Chief of the SS-WVHA, General of the SS Army Corps Pohl, presented himself unexpectedly in the morning of the 23rd of September at Auschwitz to learn what was going on, and where the assigned tons of Zyklon-B were going. Pohl went first to the Bauleitung, and had the general set-up of the camp explained to him, and the buildings that had been erected, those under construction (including the four crematories of Birkenau) and those planned described to him. His question on Zyklon-B was answered that with this product the Jews and the lice were destroyed at the same time.]

Pressac’s source for the above is the diary of Johann Paul Kremer published in Auschwitz vu par le SS (Edition du Musée d’Etat à Oswiecim) [Auschwitz Viewed by the SS (Edition of the State Museum at Oswiecim, 1974)], pages 233 and 234 (notes 182 and 183 on p. 105). In reality, these two pages from the end of page 233 to the end of page 234, read as follows:

La matin, l’Obergruppenführer Pohl est arrivée avec sa suite au Foyer des Waffen SS. Devant la porte, une sentinelle. Pour la première fois, on me présente les armes. Le soir, à 20 heures, dîner au Foyer des officiers SS en compagnie de l’Obergruppenführer Pohl: un véritable festin. On nous a servi du brochet frit à volonté, du vrai café, une excellente bière et des sandwiches.

[In the morning, Obergruppenführer Pohl arrived with his entourage, at the residence of the Waffen SS. In front of the door a sentinel. For the first time I am presented arms. The evening at 20 hours; dinner at the SS Officers Club in the company of Obergruppenführer Pohl: a real feast. We were served fried pike all we wanted, real coffee, an excellent beer and some sandwiches.]

That is all. The rest is the product of Pressac’s imagination. On p. 117 he twice contradicts himself, by writing that Pohl had gone to Auschwitz:

surtout préoccupé de construire une grande station d’épuration des eaux à Broschkowitz (au nord de la ville d’Auschwitz) pour réduire le risque typhoïdique. (p. 117, our italics)

[above all, concerned with constructing a large water purification station at Broschkowitz (north of the town of Auschwitz) to reduce the risk of typhus.]

Therefore, the WVHA already knew of the typhus epidemic and Pohl did not inspect Auschwitz to ask "où filaient les tonnes de Zyklon-B accordées" ["where the tons of Zyklon B given them were going"].

Jean-Claude Pressac’s fundamental thesis is that Crematories II and III, planned and constructed as simple sanitary installations, were subsequently turned into instruments of crime:

S’imposa fin octobre 1942 l’idée, somme toute évidente, de transférer l’activité ‹gazeuse› des Bunker 1 et 2 dans une pièce de crématoire, équipée d’une ventilation artificielle, comme cela avait été pratiqué en décembre 1941 dans la morgue du crématoire I. (p. 60)

[Towards the end of October 1942, the idea occured, an obvious one when all was said and done, of transferring the ‘gassing’ activity of Bunkers 1 and 2 into a room of the crematory, equipped with artificial ventilation, as had been practiced in 1941 in the morgue of Crematory I.]

That is an arbitrary statement by Pressac, not supported by any document. He adds an equally unfounded assertion, that:

en novembre 1942, les SS de la Bauleitung résolurent d’équiper les crématoires de chambres à gaz homicides. (p. 66)

[In November 1942, the SS of the Bauleitung resolved to equip the crematories with homicidal gas chambers.]

According to Pressac, initially the intentions of the SS were to:

utiliser pour les gazages la ‹Leichenkeller 1› du crématoire II dès qu’elle serait opérationelle ou, si l’expédition des matériels requis tardait, se rabattre sur la ‹Leichenhalle› du crématoire I aprés avoir installé sa désaération définitive, déjà livrée, et capable d’extraire 8.300 m3 d’air par heure de toutes les pièces du bâtiment, dont environ 3.000 m3 par heure de sa ‹Leichenhalle›. (p. 61)

[use the ‘Leichenkeller 1’ of Crematory II for the gassings as soon as it was operational. Or, if shipment of the required material was delayed, to fall back on the ‘Leichenhalle’ of Crematory I after having installed its final ventilators, already delivered. These ventilators were capable of extracting 8,300 m3 of air per hour from all the rooms of the building, and about 3,000 m3 per hour from its ‘Leichenhalle’.]

This appears clearly irrational, even from Pressac’s perspective. Granted, on the one hand, that the Bauleitung could have continued to use Bunkers 1 and 2 as it had done until then for mass extermination of the Jews while they were waiting for the requested materials to arrive for the homicidal transformation of Leichenkeller 1 of Crematory II; but on the other hand, the gassings had already been transferred to the end of April 1942 from Crematory I at Birkenau because:

un gazage imposait d’isoler totalement la zone du crématoire, ce qui perturbait l’activité du camp,

[gassing demanded total isolation of the crematory zone, which disturbed the activity of the camp],

and in addition:

il était impraticable lorsque les travaux étaient en cours, (p. 35)

[it was impracticable, while work was in progress], (which occurred fairly often.)

The idea that the SS were thinking again of moving the supposed homicidal gassings to Crematory I occurs to Pressac because a note dated 27 November 1942 gives the order to install the ventilation system of Crematory I (p. 60). Nevertheless, he states at the same time that:

sa morgue étant désaérée mécaniquement, des gazages homicides avec un toxique gazeux pouvaient y être pratiqués. (p. 23)

[homicidal gassings with a toxic agent could be carried out there because its morgue was mechanically ventilated.]

(Referring to the temporary ventilation system mounted in Crematory I by the Boos company between 23 February and 1 March 1941 [p. 18].)

Therefore, if the temporary ventilation system could already support the homicidal gassings, why install a permanent system? If instead the permanent system was indispensable for homicidal gassings, why wasn’t it installed immediately instead of being left in a storage room? It had been shipped from Topf 16 April 1942. Then the idea to use Crematory I for mass gassings was discarded, and the Bauleitung concentrated on the criminal transformation of Crematory II and III:

Transférer les gazages homicides dans les crématoires II et III semblait simple sur le papier, mais l’était beaucoup moins du fait que le bâtiment, conçu par Prüfer et amélioré par Werkmann, n’avait pas été envisage à cette fin. Le rez-de-chaussée, avec la salle des fours et ses pièces de service, n’avait besoin d’être modifié. Mais le sous-sol devait etre amenagé pour que puissent s’y pratiquer les ‹actions spéciales›. (pp. 63-64, our italics)

[Transference of the homicidal gassings to Crematories II and III seemed easy on paper, but was much less so because of the fact that the building, planned by Prüfer and improved by Werkmann, had not been envisioned for this purpose. The ground floor, with its oven room and service rooms, did not need to be modified. But the basement had to be arranged so that the ‘special actions’ could be carried out there.]

There is no doubt that beginning with the end of 1942, the basement of Crematory II had undergone various transformations with respect to the initial project. There is also no doubt that the oven room had not undergone any modification in number or capacity with respect to the initial plan. How to explain this inconsistency? If Crematory II had been planned as a simple sanitary installation, adequate for the natural death toll of the camp, its transformation into an instrument of mass extermination would have required a corresponding increase in the capacity of the ovens: in other words, the installation of additional ovens. But that did not happen. Therefore, all that remains is to excessively inflate the real capacity of the ovens and, contradictingly, to infer that the ovens could handle even a mass extermination without difficulty even though they were designed for hygienic purposes. It suffices for Pressac to declare that Crematory II could really cremate 1,440 corpses in 24 hours (capacity that could be defined as technically preposterous) to overcome the contradiction.

The reality is quite different. The installation in Crematory II and III of a 210 m2 gas chamber (the area of Leichenkeller 1), in which it would have been possible to gas 1,800 victims without difficulty (the eyewitnesses speak even of 3,000), would have required 75 chambers instead of the existing 15 for the cremation of the corpses in one day. The time required to cremate the bodies of the victims, would have taken five days, presenting a serious obstacle to an extermination process. The fact, therefore, that the oven room was not transformed, demonstrates that the changes made in the basement were not of a criminal nature.

The picture of the changes made in the basements of Crematories II and III once more relies on Plan 2003 of 19 December 1942, which Pressac considers a bavure architecturale [!9] (an architectural trace), in which the slide (Rutsche) for the corpses no longer appears:

l’escalier nord devient le seul accès possible aux morgues, ce qui implique que les morts devront descendre l’escalier en marchant. (pp. 64-65)

[The north stairway becomes the only possible access to the morgues, which implies that the dead will have to descend the stairs walking.]

Actually, Plan 2003 was nothing more than a proposal to transfer the basement access to the street side (Verlegung des Kellerzuganges an die Strassenseite) [67] and not a plan to eliminate the slide. Therefore, the absence of the slide is basically a simplification of a part of the design which is technically irrelevant. The elimination of the slide would have been technically irrational (unless the lift were used to transport the corpses to the mortuary rooms), since the natural mortality at the camp continued. In fact, the slide was constructed according to the original design both in Crematory II and in Crematory III. This was independent of the fact that:

le plan 2003 arriva trop tard sur les chantiers 30 [Crematory II] and 30a [Crematory III] (p. 65)

[Plan 2003 arrived too late at construction sites 30 (Crematory II) and 30a (Crematory III)],

as Pressac claims. That, however, could not be valid for Crematory III (which was in a less advanced construction stage) but, more precisely, depended on the logical necessity of an easy access for the corpses to the mortuary chambers.

The initial plan of the SS (November 1942) was to install in Crematories II and III two homicidal gas chambers operating alternately:

Les SS envisagèrent aussi que les deux morgues fussent utilisées en chambres à gaz, croyant alors à tort que le fort rendement prévu des cinq fours trimoufle permettrait une marche alternative. Dans cette configuration, un vestiaire extérieur était indispensable, donnant directement sur l’escalier de service, qui desservait les deux salles par le vestibule central. De plus, améliorer la ventilation de la Leichenkeller 2 (seulement desaérée) s’imposait, par adjoncton d’une aeration. Apres que les fours eurent été testés et leur rendement mieux estimé, cette solution fut rejetée parce qu’elle aboutissait à produire au sous-sol des monceaux de cadavres que les fours du rez-de-chaussée auraient mis trop longtemps a incinérer. (p. 66, our italics)

[The SS had also envisioned that the two morgues be used as gas chambers, at that time wrongly believing that the large yield foreseen for the five three-chambered ovens would permit an alternative operation. In this configuration, an exterior changing room was indispensable, opening directly on the service stairs that connected the two rooms through the central vestibule. Moreover, it was further essential to improve the ventilation of Leichenkeller 2 by the addition of blower type ventilation. At the time it was only ventilated by aspiration, that is, by drawing air out of the room. After the ovens had been tested and their yield better estimated, this solution was rejected because it ended up producing a heap of corpses in the basement which the ovens on the ground floor would have taken too long to incinerate.]

Here Pressac ensnares himself in another series of insuperable contradictions. On one hand, the plan of the dualistic homicidal gas chamber, which depended on an overestimate of the capacity of the ovens, could not have been done in November 1942, because the two three-chambered ovens of the crematory of Buchenwaldthat were of the same model as the ovens in Crematories II and III of Birkenauwent into operation on 23 August and on 3 October 1942 respectively (p. 39). In November therefore, the oven’s real capacity was perfectly known, having already been in activity a total of four months. Furthermore, the plan of the dualistic homicidal gas chamber could not have been abandoned in November 1942 in consequence of the real capacity resulting from the testing of the ovens, because the first test of the ovens in Crematory II took place, according to Pressac, on 4 March 1943 (p. 72).

What remains sure is that Jean-Claude Pressac admits the irrationality of a plan which envisioned, in the basement of the crematories, a capacity of corpses enormously greater than that of the ovens on the ground floor. In fact, he even states that these gas chambers were divided in two to balance the process of extermination. The capacity of the ovens being (in his opinion, lacking in any technical foundation) still inferior to that of the gas chambers:

La recherche d’un meilleur agencement se poursuivit même après la mise en exploitation. Ainsi, fin 1943, afin de ‹régulariser› la marche des crématoires II et III, l’administration du KL fit deviser leurs chambres a gaz en deux, ne consacrant plus que 100 m2 au gazage, pour tuer et incinérer 500 à 700 arrivants inaptes (comprenant beaucoup d’enfants) en vingt-quatre heurs. (p. 67, our italics)

[The search for a better arrangement continued even after the start of the operation. Thus, at the end of 1943, so as to ‘standardize’ the running of Crematories II and III, the administration of the KL had their gas chambers divided in two, devoting no more than 100 m2 for gassing, to kill and incinerate 500 to 700 unfit arrivals (including many children) in twenty-four hours.]

The source of this information is the deposition of H. Tauber, which however speaks only of Crematory II. Therefore, Pressac’s attribution of this supposed modification to Crematory III, as well, is unwarranted. In his book of 1989, Pressac comments regarding H. Tauber’s story concerning the division of the gas chamber and of the successive gassings as follows:

"One of the very few contestable points in the deposition." [68]

Needless to say that there is no proof of this supposed division, either documentary or architectural.

It is evident that since the maximum real capacity of Crematories II and III of 360 corpses per day (considering the presence of the corpses of children), Pressac admits a fortiori the pointlessness of a supposed extermination plan by the Bauleitung.

The final project of the SS, effectively accomplished, according to Pressac, was the transformation of Leichenkeller 1 into a homicidal gas chamber, and of Leichenkeller 2 into an undressing room. That would mean that Crematories II and III were no longer provided with mortuary rooms. So, one may ask, where did the SS expect to deposit the corpses of the registered prisoners deceased of natural causes which had to be cremated? The question is even more legitimate in that for each of the planned Crematories II and III, we remember that originally there were envisioned three mortuary rooms, exclusively for sanitary purposes, for a total area of 671 m2. [69]

In support of his thesis, Pressac quotes a series of bavures (traces or suspicions) which we shall deal with subsequently. Nevertheless, the "definitive" proof is connected to the ventilation system of the crematories.

VENTILATION SYSTEM

The initial ventilation project of the new crematory included:

A blowing ventilator (No. 450) for the B-Keller (the future Leichenkeller 1) with a capacity of 4,800 m3/h;

An aspirating ventilator (drawing air out) (No. 450) for the B-Keller with a capacity of 4,800 m3/h;

An aspirating ventilator (No. 550) for the L-Keller (the future Leichenkeller 2) with a capacity of 10,000 m3/h;

An aspirating ventilator (No. 550) for the oven room with a capacity of 10,000 m3/h;

An aspirating ventilator (No. 375) with a capacity of 3,000 m3/h for the autopsy room (p. 30).

Since Pressac indicates even the volume of the respective rooms (p. 30), it is possible to calculate the number of air exchanges estimated within one hour:

4,800 / 483 = 9.93 exchanges for the B-Keller;

10,000 / 966 = 10.35 exchanges for the L-Keller;

10,000 / 1,031 = 9.69 exchanges for the oven room;

3,000 / 300 = 10 exchanges for the autopsy room.

Subsequently, the capacity of the ventilators was increased as follows:

pumping ventilator for the B-Keller: 8,000 m3/h
(=16.56 air exchanges per hour);

aspirating ventilator for the B-Keller: 8,000 m3/h
(= 16.56 air exchanges per hour);

aspirating ventilator for the L-Keller 13,000 m3/h
(=13.45 air exchanges per hour);

aspirating ventilator for the oven room: 12,000 m3/h
(= 11.64 air exchanges per hour)

aspirating ventilator for the autopsy room: 4,000 m3/h
(= 13.33 air exchanges per hour) (p. 38).

The capacity of the ventilators mentioned by Pressac is not certified by any document. He obviously calculated them on the basis of the power of the motors. These are shown in the D59366 Topf plan of 10 March 1942 (Pressac’s Documents 13-15) which, by its own date, refers to a period in which the crematory was being planned exclusively for hygienic purposes.

Pressac states that Leichenkeller 1 of Crematories II and III was actually equipped with ventilators with a capacity of 8,000 m3/h of air (p. 74 and 118), and even mentions the invoice of the ventilation system for Crematory III: invoice No. 729 of 27 March 1943 (p. 105, note 184).

He leaves understood that the increased capacity of the ventilators from 4,800 to 8,000 m3/h was effected to compensate for the arrangement of the ventilation system planned and built for a normal mortuary room. In fact he states, in relation to the Gasprüfer, which we will deal with later, that:

les SS voulaient vérifier si la puissance de ventilation de la Leichenkeller 1 compenserait sa disposition d’origine, aération haute et désaération basse prévues pour une morgue, et qui aurait dû être inversée pour une chambre à gas, requérant aération basse et désaération haute. (pp. 71-72)

[The SS wanted to verify if the power of the ventilation of Leichenkeller 1 would compensate for its original disposition, that is, high ventilation and low air aspiration which was anticipated for a mortuary, whereas the intention was to convert it to a gas chamber, requiring low ventilation and high air aspiration.]

Finally, by this time Leichenkeller 2, having become a changing room, no longer required a ventilation system; the ventilation systems were installed in Crematories II and III but the ventilators’ motors were not. (pp. 79, 80)

The study of the ventilation systems of Crematories II and III actually provides definite proof that Leichenkeller 1 was not transformed into a homicidal gas chamber. First of all, the Topf invoice No. 729 dated 27 March 1943 [70] cited by Pressac mentions that a ventilator with a capacity of 4,800 m3/h was required for the B-Raum, the supposed homicidal gas chamber, and that a ventilator with a capacity of 10,000 m3/h was needed for the L-Raum, the supposed changing room. The same capacities are indicated by the invoice No. 171 of 22 February 1943 for Crematory II. [71]

In his preceding 1989 work, Pressac publishes a table which summarizes "Dimensions and volumes of the Krematorium II and III Leichenkeller" on the basis of the plans of the crematories: Leichenkeller 1 measured 30 m in length, 7 in width and 2.41 in height; therefore, it had an area of 210 m2 and a volume of 506 m3. Leichenkeller 2 was 49.49 m long and 7.93 m wide and 2.30 m high, so its area was 392.5 m2, its volume was 902.7 m3. [72] Consequently, for the supposed homicidal gas chamber, the SS had foreseen 4,800 / 506 = 9.48 air exchanges per hour, while the supposed changing room 10,000 / 902.7 = 11 air exchanges per hour: Thus the gas chamber was less ventilated than the changing room! But that’s not all. In Engineer W. Heepke’s classic work on the planning of crematories, one reads that for morgues/mortuaries it was necessary to provide a minimum of five air changes per hour, and in case of intense utilization, up to ten exchanges. [73] It is therefore evident that the ventilation system of Leichenkeller 1 was planned and built for a morgue. As a means of comparison, seventy-two air exchanges per hour were foreseen [74] for the disinfestation gas chambers with the Kreislauf system, the plans of which Pressac publishes in documents 16 and 17. Additionally, we notice that seventeen air exchanges per hour were foreseen for the first ventilation plan from Topf for Crematory I of Auschwitz (p. 18). This was almost twice that of the alleged homicidal gas chamber of Crematories II and III! These plans were for the autopsy room and morgue, without any homicidal purposes, with a purely hygienic goal.

Concerning the ventilation of Leichenkeller 2, it is true that the motor destined for this area does not figure in Plan 2197 of 19 March 1943, but that does not mean that it was decided not to install it at all. The work done in the crematories demonstrate the contrary. In Crematory II, the ventilation system of the alleged homicidal gas chamber in Leichenkeller 1 was installed between 22 February and 14 April 1943; the ventilation system of Leichenkeller 2’s supposed changing room was installed between 15 and 28 March; the ventilation system of Leichenkeller 2 of Crematory III was installed between 12 and 22 April, as Pressac informs us in his 1989 book. [75] Now, if the absence of ventilator motors had been based on the SS decision to convert Leichenkeller 2 into a homicidal gas chamber, it is difficult to understand why they had the ventilation systems installed in the two crematories of Leichenkeller 2 after they had decided that Leichenkeller 2 did not need a ventilation system! It is therefore evident that the SS had the ventilation system of Leichenkeller 2 installed because they intended to use it. That the motors were not installed immediately was solely coincidental.

What has been presented here is already more than sufficient to cancel the criminal character of the bavures (traces or suspicions) listed by Jean-Claude Pressac, which are in fact framed in a completely different context.

Pressac correctly states that the struggle against typhus at Auschwitz was aided by the disinfestation systems (p. 84). Since the appearance of the first cases of typhus, the SS had begun to think of expanding the existing systems; even with the introduction of new technologies (meeting of 30 June 1942) (p. 83). The compelling necessity of new disinfestation systems is confirmed by the design of the Zentralsauna (24 November 1942), [76] which, by its purpose of fighting the typhus epidemic, was part of the Sonderbaumassnahmen and came under the Durchführung der Sonderbehandlung. All this leads us to recall that in the meantime, at the end of 1942, the SS had decided to install several temporary disinfestation gas chambers. The first of these were in Crematories II and IV, [!10] which were in a more advanced phase of construction. That explains in a historical and logical manner all the bavures mentioned by Pressac, from which we shall take a brief respite.

The term Sonderkeller applied to Leichenkeller 1 (p. 60) enters into the terminology. Sonder- applied to the fight against typhus.

The term Vergasungskeller designates a disinfestation basement. In the explanatory report on the construction of KGL Birkenau dated 30 October 1941, the two Zyklon-B Entlausungsbaracken (disinfestation installations) subsequently built, BW5a and 5b are equipped with a Vergasungsraum. [77]

The plan to preheat Leichenkeller 1 (p. 73) makes sense for a disinfestation gas chamber because it would permit shorter gassing times (the duration of a gassing using 20 g of hydrocyanic acid per square meter required 45 minutes at a temperature of 25°C to 35°C, but three hours at a temperature of 0 to 5°C). [78] It would be much less for a homicidal gas chamber, for a mass extermination in which the bodies of the victims would heat up the area sufficiently; therefore a preheating would be absolutely superfluous. [!11]

The presence of a gas-proof door (p. 80) is perfectly normal in a disinfestation gas chamber.

The request for 10 Gasprüfer (p. 71) (see appendix, Documents No. 7 and 8), supposing that it really deals with Anzeigegeräte für Blausäure Reste (implements for the indication of residues of hydrocyanic acid) (p. 72), is perfectly normal for a disinfestation gas chamber.

Instead, Pressac writes, as enthusiastically as naively:

Ce document constitue la preuve définitive de l’existence d’une chambre à gaz homicide dans le crématoire II. (p. 72, our italics)

[This document constitutes the definitive proof of the existence of a homicidal gas chamber in Crematory II.]

This document can be at best an indication, not a definitive proof, of the existence of a gas chamber. But that this gas chamber was homicidal, however, is a simple arbitrary affirmation by Jean-Claude Pressac.

On this subject Pressac adds a very important explanation:

Des essais avec introduction préalable de Zyklon-B furent pratiqués. La mesure du gaz cyanhydrique résiduel aurait été effectuée par une methode chimique, et non avec les dix détecteurs de gaz, demandés trop tardivement pour être livrés à temps. (p. 73, our italics)

[Some experiments with the preliminary introduction of Zyklon-B were made. The measurement of the residual hydrocyanic gas would have been done by a chemical method, and not with the ten gas detectors requested too late to be delivered in time.]

Although this document jibes perfectly with our thesis, in our opinion, it is a forgery; of proper design, but poorly executed. To break it down:

a. The Gasprüfer, in the German technical terminology, [79] were simple analyzers of the combustion gases.

b. To prove the existence of hydrocyanic acid gas residue, there was only one chemical method, and no détecteurs (detectors) based on physical properties. [80]

c. The apparatus that was used for this test was called Gasrestnachweisgerät für Zyklon. [81]

d. This apparatus was required equipment at all of the disinfestation installations, including those at Auschwitz.

e. Since these devices were available in the disinfestation installations at the camp, it would not have made sense to request them from a company that did not produce them. They could have been obtained from the above-mentioned installations or ordered directly from the companies which made or distributed them (the same ones that distributed the Zyklon B).

f. Since the crematory administrations did not even request gas masks with the special filter "J" (for hydrocyanic acid), as with the Gasprüfer, it is clear that it was possible to procure them at the camp, where it could also procure the Gasrestnachweisgeräte für Zyklon.

Our conclusion: the Gasprüfer were simple analyzers of the combustion gases of the crematory ovens.

Regarding the presence of 14 showers in Leichenkeller 1: According to Pressac, this is a bavure because these showers were false (p. 80), and were used therefore to deceive victims of alleged homicidal gas chambers; that these showers were false is a simple arbitrary statement by Pressac.

The mention of "a wooden blower" (Holzgebläse) destined for Leichenkeller 1 (p. 70) is for Pressac a bavure technique (technical trace) since:

prouvait que l’air extrait n’était plus celui d’une morgue, chargé de miasmes, mais de l’air mélangé à un produit aggressif ne devant être aspiré que par une soufflerie ne pouvant être corrodée, donc entièrement de bois (de préférence en cyprès). Le toxique gazeux utilisé dans les chambres à gaz homicides était de l’acide cyanhydrique à forte concentration (20 gr/m3) et les acides sont corrosifs. (pp. 70-71, our italics)

[it proved that the air extracted was no longer that of a morgue, permeated by miasmas, but air mixed with a corrosive substance which could be vented only by a non-corrodible fan, made entirely of wood (preferably cypress). The gaseous toxin used in the homicidal gas chambers was concentrated hydrocyanic acid (20 gr/m3) and the acids are corrosive.]

Nevertheless the above-mentioned wooden blower was later replaced with a metal one, as is clear from the Aktenvermerk of 25 March 1943, [82] which reads:

Anstelle des Holzgebläses für die Entlüftungsanlage des Leichenkellers I wird ein Schmiedeeisengebläse als Ausführung gewählt.

[instead of the wooden blower for the exhaust system of Leichenkeller I, a wrought-iron-type blower was adopted into the final design].

Pressac must therefore explain why, given that hydrocyanic acid (as he says) is corrosive, the engineers of the Bauleitung replaced a wooden blower with a metal one, and why the Degesch engineers proposed a metallic apparatus for the disinfestation gas chamber with the Kreislauf system, like those that appear on Documents 16 and 17 of his book. Why would they have done this—so that they could have been "corroded" by hydrocyanic acid?

In these rooms, standard or normal, which had a volume of 10 m3, one can of Zyklon B of 200 g (HCN contents) was used in order to produce a gas concentration of 20 g/m3. Pressac states, as always without any proof, that this was the concentration of the alleged homicidal gas chambers. But previously, he claimed that the concentration of gas used in homicidal gas chambers of Birkenau was 12 g/m3. [83] We shall later learn the reason for this increase.

Drahtnetzeinschiebvorrichtung does not signify "dispositifs d’introduction en treillis de fil de fer" ["introduction devices made of wire netting"] (p. 79), but rather, insertion devices (the verb einschieben, means in fact, "insert, to slide into"; for example, one "inserts" a drawer in a closet).

The device for the introduction of Zyklon B in the alleged homicidal gas chambers would be called Einwurfvorrichtung (Pressac himself speaks in fact of déversement, "pouring out," of the Zyklon B in the alleged homicidal gas chambers [p. 89]). The Holzblenden, obturateurs de bois, "wooden obstructors" (p. 79), cannot be what Pressac maintains, i.e. wooden covers of the alleged introduction devices of the Zyklon B: These devices would be called in fact Holzdeckel, precisely covers, not obstructors.

Pressac states that the above-mentioned devices were found in "morgue 1" (p. 79), that is, in Leichenkeller 1, the alleged homicidal gas chamber. In reality, in the inventory of Crematory II, [84] these devices are attributed to Leichenkeller 2, the alleged changing room (for the supposed homicidal gas chamber): Did the SS want to gas the victims in the "changing room"? But that’s not all! These devices do not figure at all in the inventory of Crematory III: [85] Then how did the SS think they could introduce the Zyklon B into the gas chamber?by affably asking the victims to carry in the cans of Zyklon B and open them after the gas-proof door closed behind them?

Therefore, these devices could have been anything but what Jean-Claude Pressac claims.

The designation cave à déshabillage "cellar for undressing" (p. 74), Auskleidekeller, attributed to Leichenkeller 2, is perfectly normal, from the moment that a temporary disinfestation gas chamber was installed in Leichenkeller 1.

Jährling’s bavure shows once again, as if there were any more need for it, with what kind of distorted logic Pressac has fabricated his "criminal traces". The passage merits citation in its entirety; but first, it is necessary to give a brief historical setting. Nineteen Zyklon B gas disinfestation chambers with the Kreislauf system were designed for the Aufnahmegebäude (reception building of the new prisoners) of the main camp. The Kreislauf systems were not at first installed.

At the end of 1943, it was decided to transform eight of these rooms into short wave disinfestation chambers, in line with a new process developed by the Siemens company. Work began in February 1944 (p. 88). At the same time, it was decided to put the eleven remaining rooms into operation by installing the Kreislauf system. The Boos company, which should have performed this work, raised objections. The Testa company (Tesch und Stabenow), distributor of Zyklon B, had also taken an interest in the affair, as had Dr. Wirths, who remembered that according to an ordinance in effect, the Zyklon B had to be replaced by another gas: Arginal, the use of which required an adaptation of the Zyklon B gas chambers. (pp. 88-89).

A cette occasion, l’employé civil Jährling commit une extraordinaire bévue dans une lettre destinée à la Testa. Il désigna les chambres a gaz d’épouillage de ‹Normalgaskammer› mot souligné et mis entre guillemet, comme s’il existait des chambres a gaz ‹normales› et d’autres ‹anormales›. L’appellation fut reprise par la Testa, qui affirmait d’abord que la conversion a l’Arginal n’était obligatoire que dans le cas d’installations nouvelles, et insistait surtout pour que le personnel s’occupant des chambres à gaz normales à l’acide cyanhydrique fût particulierement bien formé, sous-entendant que leur fonctionnement était nettement plus complexe que le simple déversement de Zyklon B dans des chambres a gaz ‹anormales›. (p. 89, our italics)

[On this occasion, the civil employee Jährling committed an extraordinary blunder in a letter destined for the Testa company. He designated the delousing gas chambers as ‘Normalgaskammer’, a word underlined and put in quotation marks, as if there existed ‘normal’ gas chambers and others which were ‘abnormal’. The appellation was taken up by the Testa company, which first stated that the conversion to Arginal [!12] was only obligatory in the new installations, and insisted that the personnel who handled the normal gas chambers with the hydrocyanic acid should be particularly well trained: implying that their operation was distinctly more complex than the simple pouring of Zyklon B into the ‘abnormal’ gas chambers.]

If Pressac had familiarized himself even a little with the disinfestation gas chambers using hydrocyanic acid, he would have known that a Normalgaskammer was a gas chamber according to the norms; that is, a standard Degesch room with a Kreislauf system. A disinfestation room not conforming to the norm was a simple auxiliary gas chamber (behelfsmässige Blausäurekammer). [86]

Therefore, Jährling simply wanted to underline that the envisioned transformation of the operating system referred to gas chambers planned as hydrocyanic gas chambers with the Degesch-Kreislauf, normal, and not to chambers without such a system, abnormal, like that of the BW 5b system of Birkenau. That’s all.

Chapter Six

BUNKERS 1 AND 2

Before examining Jean-Claude Pressac’s statements on Bunkers 1 and 2, it is well to specify that this designation (like those of "red house" and "white house") is not found either in the German documents or in the reports of the clandestine resistance movement of the period at Auschwitz; it has been created by post-war eyewitnesses.

Pressac states that Bunker 1, destined for mass extermination, went into operation at the end of May 1942 (p. 39), that is, as we have seen, even before R. Höss received from Himmler the alleged order for extermination of the Jews.

There exists no document on the existence of Bunker 1 (and 2) as homicidal installations. What Pressac says on the subject, as if it were certified historical truth, is in reality the simple result of extrapolation of testimonies which contradict each other on all essential points. [87] According to Pressac, Bunker 2 began its activity at the beginning of June 1942 (p. 41). Pressac describes the genesis of the installation as follows:

Non loin de Bunker 1, s’élevait une seconde fermette, crépie de chaux et d’une superficie de 105 m2. La transformer en chambre à gaz était simple, l’opération ayant déjà été realisée au Bunker 1, et un demi-millier de personnes pourraient y tenir. Mais Höss voulut améliorer la ventilation. Il consulta Bischoff qui lui montra un article du Dr. G. Peters, le directeur de la Degesch (entreprise fabriquant le Zyklon-B), où était décrite une installation d’épouillage au Zyklon-B avec huit petites chambres a gaz de 10 m3 disposees côte à côte. (pp. 41-42)

[Not far from Bunker 1 arose a second little whitewashed farmhouse with an area of 105 m2. To transform it into a gas chamber was simple. The operation had already been realized at Bunker 1, and it could have held half a thousand people. But Höss wanted to improve the ventilation. He consulted Bischoff, who showed him an article by G. Peters, the director of the Degesch Company (the enterprise fabricating the Zyklon B), which described a delousing installation using Zyklon B with eight little gas chambers of 10 m3 arranged side by side.]

The article, as Pressac himself informs us, had been requested by the Boos Company:

afin de s’en inspirer pour équiper le futur bâtiment de reception des detenus du Stammlager d’une batterie de 19 cellules d’épouillage à gaz semblables. (p. 42)

[in order to use it as a guide for equipping the future prisoner reception building of the Stammlager with a battery of nineteen similar delousing gas cells.]

Therefore the article concerned the nineteen hydrocyanic acid disinfestation rooms with the Kreislauf system planned for the Aufnahmegebäude. The date of the request is 1 July 1942 (p. 103, note 135), that is, one month after the presumed beginning of the alleged activity of Bunker 2. That this article (in which the plan for a Degesch gas chamber with the Kreislauf system published by Pressac as Document 16-17, which has been previously discussed) is presented as having been shown by Bischoff to Höss for the purpose of furnishing Bunker 2 with a ventilation system, is not attested to by any document, and is the pure fantasy of Jean-Claude Pressac, who states, moreover, that no mechanical ventilation system was installed in Bunker 2.

Finally:

furent construites dans la blanche chaumière [Bunker 2] quatres petites chambres à gaz d’environ 50 m2 (sic), placées en parallèle, sans ventilation mécanique, mais orientées au mieux dans le sens du vent (nord-sud à Birkenau). (p. 42, our italics)

[Four little gas chambers of about 50 m2 [sic] were constructed in the white thatched house (Bunker 2). These were placed in parallel, without mechanical ventilation, but oriented in the best direction for the wind (North-South at Birkenau).]

The aim of Pressac’s pretensions are obvious. One of the criticisms made of his 1989 book regarding the alleged homicidal chambers is that, even though the Germans were at the forefront of technology in hydrocyanic disinfestation chambers, thanks to their invention of the Degesch-Kreislauf System, according to Pressac, the Germans’ alleged homicidal gas chambers were supposed to have been technologically unsophisticated. Pressac must therefore establish in some manner a connection between the two types of systems, which he does in a crafty manner, asserting on the one hand that the transformation of Bunker 2 into a homicidal gas chamber had been done:

en s’inspirant des installations d’épouillage montées par la Degesch de Frankfurt/Main (cellules placées en parallèle), (p. 115, our italics)

[emulating the delousing systems installed by the Degesch Company of Frankfurt/Main (cells placed in parallel)],

and furthermore, that the first homicidal gassing in Crematory II had been done with the introduction of 6 kg of Zyklon B, which:

représente une concentration d’environ 20 g d’acide cyanhydrique par m3, identique a celle preconisée par les dirigeants de la Degesch dans leurs cellules d’épouillage. (p. 119, our italics)

[represents a concentration of about 20 g of hydrocyanic acid per m3, identical to that authorized by the directors of Degesch for their delousing cells.]

Thus, the Bauleitung engineers would have drawn the least significant element from the article of G. Peters (and E. Wüstiger): the arrangement of the gas chambers en parallèle (in parallel). Of the presumed Bunker 2 only the foundation remains which shows subdivision in seven locations the height of these locations, nothing is known. Therefore the volume calculated by Pressac (4 chambers of 50 m3 each) is simply the fruit of his imagination.

Concerning the concentration of hydrocyanic acid: since the volume of Leichenkeller 1 (506 m3) would have been reduced to about 406 m3, after the removal of about 100 m3 occupied by the 1,492 bodies of the victims and the reinforced cement columns, the concentration obtainable with 6 kg of Zyklon B [!14] would have been (6,000 / 406 =) about 14.8 g/m3, not 20.

Not bad: Pressac states authoritatively that the concentration of hydrocyanic acid in the alleged homicidal gas chamber was 20 g/m3. Here the second factitious connection between the Degesch disinfestation gas chambers and the alleged homicidal gas chambers is created from thin air.

In the beginning, the SS had not foreseen changing rooms for Bunkers 1 and 2; the victims undressed en plein air (in the open air), but then:

Bischoff demanda dans son second rapport le montage, près des deux Bunker, de quatre baraques-écuries de bois comme vestiaire pour les inaptes. Chaque baraque coutait 15,000 RM. La demande fut formulée ainsi: ‹4 Stück Baracken für Sonderbehandlung der Häftlinge in Birkenau/4 Baraques pour [le] traitement spécial des detenus à Birkenau.› (pp. 45-46, Pressac’s italics)

[Bischoff requested in his second report the construction, close to the two Bunkers, of four wooden hut-stables for changing rooms for the unfit. The cost of each hut was 15,000 RM. The request was formulated thus: ‘four huts for the special treatment of the prisoners at Birkenau’.]

The report in question was written at the end of July 1942, during a full scale typhus epidemic. As we have explained, the "special treatment of the prisoners" did not have a criminal significance, but was a health measure coming under the sanitary provisions taken by the SS to arrest the epidemic. There is no need to add that the relation of these huts to Bunkers 1 and 2 is a purely arbitrary opinion by Pressac and, as usual, has no documentary foundation.

The Badeanstalten für Sonderaktionen mentioned in the Aktenvermerk of 21 August 1942 (p. 52) had the same function; each had to be equipped with two three-chambered ovens of the simplified models, evidently to cremate infected corpses of prisoners who died of typhus.

Pressac believes he has found a bavure in a plan of the area of interest at Auschwitz-Birkenau:

indiquant que la zone où se situaient les Bunker 1 et 2 et leurs fosses d’enfouissement était classée ‹Sperrgebiet/zone interdit› (legend of Document 21 and p. 52)

[indicating that the zone where Bunkers 1 and 2 and their burying pits were situated, was classified a ‘prohibited area’.]

But this document bears the date of 2 June 1943, at which time the two bunkers had ceased their alleged activity two and a half months earlier, and the so-called "cremation pits" (which Pressac transforms for the occasion into fosses d’enfouissement, "burying pits") had been covered over and the earth leveled: What, therefore, did the SS want to hide in this zone?

Actually, the Sperrgebiet refers to the entire white area within the oblique hatching, and therefore includes the entire zone of the camp of Birkenau. The Sperrgebiet is clearly related to various Lagersperren (camp closures) decreed by Höss due to typhus: 10 July 1942 (p. 115), 23 July (p. 116), 8 February 1943 (p. 118). In June 1943, typhus still raged in the Gypsy camp at Birkenau, and in Sector BI cases of typhus were reported until the end of July (pp. 120 and 121).

In May-June of 1944, during the deportation of the Hungarian Jews to Auschwitz,

le Bunker 2 fut réactivé pour la circonstance pour de petits groupes, dont les corps étaient brûlés dans une fosse d’incinération de 30 m2. (pp. 90-91)

[Bunker 2 was reactivated on occasion for small groups, whose bodies were burned in an incineration pit of 30 m2.]

This is decidedly irrational. The SS would have supposedly equipped an extermination installation providing "a half thousand" corpses at a time with a cremation area, sufficient at the most for 50 corpses; that is, only a tenth of the actual needs. It is necessary to add furthermore, that the eyewitness M. Nyiszli speaks of two "cremation pits" which measured 50m × 6m (600 m2 altogether), and served 5,000 to 6,000 corpses per day. [88] In his preceding book, Pressac considers this witness credible. His only fault: He multiplied the numbers by 4! [89] Yet, in the specific case, Pressac mentions a burning surface 20 times less than that declared by Nyiszli, and a cremation capacity, deducible from the area, 100 to 120 times less!

On page 147, suddenly a second pit appears, plus petite (smaller) than the first. Pressac introduces this to increase the capacity of Bunker 2 slightly, so as to justify technically the alleged extermination of the Hungarian Jews. This does not change anything we have demonstrated above.

Chapter Seven

CREMATORIES IV AND V

Pressac states that Crematories IV and V were dependent upon Bunkers 1 and 2 (p. 50), and assigned to them (p. 52).

This logistic arrangement was, to say the least, an unhappy one, given that the distance of the crematories (road distance) from the supposed Bunker 1 was about 800 meters, and from the supposed Bunker 2 about 900 meters. Therefore, the corpses would have to have been transported to the crematories by truck. If one considers that in Crematory I (according to Pressac) a more rational extermination procedure had already been worked out, one subsequently begun in all four of the other crematories of Birkenau—the placing of the homicidal gas chamber in the crematory—the planning of two "criminal" crematories not only without gas chambers, but even 800 to 900 meters away from the alleged homicidal gas chambers, is decidedly senseless.

Describing the genesis of these crematories, Pressac writes:

Quant au crématoire IV (et V), son premier dessin d’août 1942 n’en montrait que la partie incinératrice. A la mi-octobre, la Konrad Segnitz, chargée de sa toiture, le représenta avec ses dimensions définitives, la salle du four étant prolongée d’une vaste morgue de 48 sur 12 mètres (576 m2), indiquant son utilisation ‹en bout de chaîne›: le déshabillage et le gazage des victimes se situaient toujours au Bunker 2, mais les cadavres produits étaient déposés dans la morgue du crématoire IV pour y être incinérés. Puis, les SS cherchèrent à placer une chambre à gaz (chauffée avec un poêle) au centre du bâtiment, ce qui lui aurait donné la disposition logique suivante:

Vestiaire – Chambre a gaz – Sas – Salle du four a 8 moufles (p. 67, our italics)

[As for Crematory IV (and V), its first drawing of August 1942 showed only the incinerator portion. In mid-October, the Konrad Segnitz company, assigned to do its roofing, depicted in its final dimensions that the oven room was an extension of a huge morgue, 48 by 12 meters (576 m2), indicating that its function ‘at the end of the sequence’, which was the undressing and the gassing of the victims, was always situated in Bunker 2, but the corpses produced were deposited in the mortuary of Crematory IV to be incinerated. Then the SS tried to place a gas chamber (heated with a stove) at the center of the building, which would have given it the following logical arrangement:

Changing room – Gas chamber – Airlock – Oven room with 8 chambers.]

The drawing of the Segnitz company is Plan 1361, dated 14 October 1942. [90] "Then," according to that statement, the SS at that time, tried to install a homicidal gas chamber at the center of the building heated by a stove. This is false because the presence of a stove at the center of the building appears in Plan 1678 from 14 August 1942, and Pressac comments on this as follows:

The presence of a stove in the uncompleted room of Drawing 1678 is a formal indication that it was used for gassing. [91]

Thereupon, Pressac expounds the subsequent development of the plans for Crematories IV and V:

Mais le vestiaire manquait. Edifier une baraque-écurie à l’extérieur compensait cette absence et donnait:

Vestiaire – chambre a gaz – Morgue – Sas – Salle du four a 8 moufles.

Les crématoires IV et V ayant un rendement incinérateur moitié moindre que celui des II et III, leurs chambres à gaz devaient être plus modeste. Les SS conjuguèrent leur besoin de chambres à gaz de faible capacité (100 m2) pour ‹traiter› de petits groupes de victimes à l’idée de marche alternative et établirent ainsi le 11 janvier 1943 le plan définitif du crématoire IV (et V). (p. 67)

[But the changing room was missing. The building of a hut-stable compensated for this absence and gave:

Changing room – Gas chamber – Mortuary – Airlock – Oven room with 8 chambers.

Crematories IV and V, having incineration yields half of those of Crematories II and III, were to have more modest gas chambers. The SS combined their need for gas chambers of lower capacity (100 m2) for ‘treating’ small groups of victims with the proposal for an alternative operation and thus established, on 11 January 1943, the final plan of Crematory IV (and V).]

The simplified plan laid out by Pressac includes this sequence: a changing room which serves two homicidal gas chambers (Nos. 1 and 2), each for 500 "unfit," a corridor, a mortuary room, an Sas (airlock) chamber, and the oven room (p. 67).

He adds that:

cette conception nécessitait la construction d’un vestiaire extérieur, qui n’était pas indispensable par beau temps, les victimes se déshabillant dehors (été 1944), mais l’était en hiver. Pour éviter de le bâtir, les SS attribuèrent à la salle centrale une double fonction, de vestiaire et de morgue, en alternance. (p. 68)

[this conception necessitated the construction of an outdoor changing room, which was not indispensable in nice weather, the victims undressing outside (summer 1944), but was necessary in winter. To avoid constructing it, the SS assigned to the Central Hall a double function of changing room and mortuary, alternately.]

In summary, the criminal structure of Crematories IV and V, établi par les techniciens et les ingénieurs de la Bauleitung ["established by the technicians and the engineers of the Bauleitung"], revealed itself as "aberrant" (p. 68), only because "les techniciens et les ingénieurs" ["the technicians and the engineers"] of the Bauleitung, after having furnished (according to Pressac) Bunkers 1 and 2 with two changing shacks each, now, unexplainably, had to éviter (avoid) building a single shack near Crematories IV and V! For what reason? Impenetrable mystery!

Pressac states that Crematories IV and V were each equipped with two gas chambers of 100 m2 each, in total 200 m2, which could hold altogether 1,000 people, with a density of five people per square meter. But in his book Auschwitz: Technique and Operation of the Gas Chambers, he writes:

The floor area of the block of three gas chambers was 240 m2 (4,800 m3 [!15]). 2,400 people could therefore be squeezed in at a density of ten per square meter [!] [92] (our italics)

Yet the third gas chamber suddenly reappears on page 147. For what reason, we shall see later. In the above-mentioned book, Pressac admits that the extermination system of Crematories IV and V was even more "aberrant," even with his forcibly inflated oven capacities:

It would take four [or] five days to cremate 2,400 bodies. [93]

Considering the maximum real capacity of the ovens, the cremation of 2,400 corpses would have required over twelve days. Inversely, to cremate 2,400 corpses in the course of one day would have necessitated 100 chambers instead of the existing eight.

The gassing technique imagined by Pressac is this:

Le premier gazage fut catastrophique. Un SS devait, masque sur le visage, monter sur une petite échelle pour accéder a une ‹fenêtre›, l’ouvrir d’une main et de l’autre verser le Zyklon B. Sa prestation tenait du numéro d’équilibriste et devait être répétee six fois. (p. 76, our italics)

[The first gassing was catastrophic. An SS man, mask on face, had to climb on a little ladder to access a ‘window’, open it with one hand, and with the other, pour the Zyklon B. This feat was like a balancing act, and had to be repeated six times.]

Pressac forgets to add that the SS juggler would also have had to plead affably with the victims not to push him backwards, or grab him, or pull him in, while holding himself with one hand on the ladder. He would have had to extend his other hand inside the window (perched at 1.70 meters above the pavement) to pour a can of Zyklon B into the gas chamber!

Pressac’s narration continues:

Lorsque les portes étanches furent ouvertes pour évacuer le gaz, on s’aperçut que l’aération naturelle etait inefficace et il fallut percer d’urgence une porte dans le couloir nord pour provoquer un courant d’air. (p. 76, our italics)

[When the air-tight doors were opened to evacuate the gas, it was perceived that the natural ventilation was ineffective and it was urgently necessary to open a door in the north corridor to induce an air current.]

The story of ventilation in Crematories IV and V is one of those tales which illustrates the silliness of Pressac’s argumentation. In his book Auschwitz: Technique and Operation of the Gas Chambers, Pressac pretends to see this door "urgently" opened in the north wall of Crematory IV in a photograph [94] which shows only the south side of Crematories IV and Vnot in (on) Crematory IV, which appears clearly in the foreground, but rather in (on) Crematory V, which is in the background partially obscured by trees. The south wall of Crematory V is so indistinct that one can make out a door in connection to the corridor only with a great deal of determination, and examination of the original photograph [95] shows that Pressac has mistaken for a door, shade produced by trunks of trees delimited at the bottom by lighter ground.

Prüfer, arriving at Birkenau the 18th or 19th of May,

constata avec une tristesse feinte que la garantie du four du crématoire IV était expirée et qu’il ne pouvait plus réparer un four édifié avec des matériaux de second choix, estima néanmoins que ses chambres à gaz étaient encore utilisables à condition de les ventiler mécaniquement, empocha une commande de deux installations de désaération pour les crématoires IV e V se montant a 2.510 RM, et repartit le 20. (pp. 79-80)

[stated with a feigned sadness that the warranty for the oven of Crematory IV had expired, and that it was no longer possible to repair an oven built with second class materials. He judged that the gas chambers were nevertheless still usable, on the condition that they be mechanically ventilated. He pocketed an order for two ventilation systems for Crematories IV and V, amounting to 2,510 RM, and departed on the 20th.]

The source indicated by Pressac in note 247 (p. 107) is a "lettre et devis Topf du 9 juin 1943" ["letter and estimate from Topf of 9 June 1943"]. But in Auschwitz: Technique and Operation of the Gas Chambers he affirms, regarding this same source:

The author would point out that NOTHING in this letter indicates that the air extraction systems proposed for Crematories IV and V were for the gas chambers, and that they could on the face of it, only be for the furnace rooms. [96] (capital letters by Pressac)

Given that the ventilation systems were so urgent and essential for the good operation of the alleged homicidal gas chambers, one would have expected them to be installed immediately; here is instead what occurred:

Concernant ces dernières, la Topf, qui avait trouvé difficilement un moteur électrique adéquat, expédia quand même une des deux désaérations en petite vitesse le 21 decembre. Elle fut stockée au Bauhof le 1er janvier 1944 et laissée ainsi jusqu’en mai 1944. (p. 88)

[Concerning these last, Topf, who had found an adequate electric motor with difficulty, nevertheless quickly shipped by freight train one of the two ventilation systems on 21 December. It was put into storage at the Bauhof on 1 January 1944, and left as such until May 1944.]

Regarding this, Pressac adds:

L’installation de désaération, en magasin depuis janvier, fut montée en mai au crématoire V, dont le four fut jugé capable de fonctionner correctement. Pour les deux chambres à gaz et le couloir, représentant un volume de 480 m3 presqu’égal à celui des morgues des crématoires II et III, Schultze avait prévu une désaération de même puissance: une soufflerie No. 450 avec un moteur de 3,5 CV extrayant 8,000 m3 par heure. (pp. 89-90)

[The ventilation system, which had been in storage since January, was installed in May in Crematory V, whose oven was judged capable of functioning correctly. For the two gas chambers and the corridor, representing a volume of 480 m3, Schultze had anticipated a ventilation of the same capacity, almost equal to that of the mortuaries of Crematories II and III: A ventilator, No. 450 with a 3.5 CV motor, extracting 8,000 m3 per hour.]

Leichenkeller 1 (the alleged homicidal gas chamber of Crematories II and III) measured 483 m3 (p. 30) and had a ventilator capacity of 8,000 m3 per hour (p. 38), corresponding to 16.56 air changes per hour. Schultze had planned a ventilator for the three alleged homicidal gas chambers of Crematory V, which measured 480 m3, with a capacity of 8,000 m3/h of air, corresponding to 16.66 air exchanges per hour. Therefore, the two systems had the même puissance [same capacity].

We have already pointed out that the volume of Leichenkeller 1 was 506 m3, and not 483 m3, and that the ventilators of its system had a capacity of 4,800 m3/h of air, not 8,000, which corresponds to 9.41 air exchanges per hour, and not 16.56. Therefore, concerning Crematory V, according to Plan 2036, of 11 January 1943, [97] the three areas supposedly transformed into gas chambers measured respectively:

(12.35 m × 7.72 m = 95.3 m2) × 2.20 m = 209.6 m3

(11.69 m × 8.40 m = 98.2 m2) × 2.20 m = 216.0 m3

(11.69 m × 3.50 m = 40.9 m2) × 2.20 m = 90.0 m3

Totals: 234.4 m2 515.6 m3

Here Pressac finds himself faced with another difficulty: Since the combined volume of the two gas chambers (which he mentions on pages 67 and 68) is 425.6 m3, the ventilator, with a capacity of 8,000 m3/h of air, would correspond to 18.8 air exchanges per hour. In other words, engineering specialists from the Topf company are supposed to have equipped ground-level rooms, provided with doors and windows (and which therefore would have been more easily ventilated), with a ventilating system proportionally more powerful than those of basement rooms, which were more difficult to ventilate! Then Pressac introduces the third gas chamber, and decreases the total volume from 515.6 to 480 m3 to deceptively obtain two ventilation systems de même puissance (of the same power).

On page 90, Pressac presents a plan which shows the:

désaération des chambres à gaz du crématoire V, conçue par Karl Schultze en juin 1943 et monté en mai 1944.

[ventilation of the gas chambers of Crematory V, designed by Karl Schultze in June 1943, and installed in May 1944.]

The source is not indicated, because it does not exist. This plan is in fact the simple fruit of Pressac’s imagination. Furthermore it is a mistaken fruit, because the letter from the Topf company of 9 June 1943 [98] mentions:

die Ausführung der gemauerten Entlüftungskanäle

[the construction of walled ventilation ducts],

while Pressac’s plan shows bare pipes.

Chapter Eight

CONCLUSION

With this, we have arrived at the close of our critique of these Auschwitz books by Jean-Claude Pressac.

The 1979 declaration of French historians on Hitlerian extermination policies closed with the following axiom:

Il ne faut pas se demander comment, techniquement un tel meurtre de masse a été possible. Il a été possible techniquement puisqu’il a eu lieu. Tel est le point de départ obligé de toute enquête historique sur ce sujet.

[We must not ask ourselves how technically such a mass murder was possible. It was possible technically because it took place. Such is the obligatory point of departure for all historic investigation on this subject.] [99]

Jean-Claude Pressac, however, wanted to study the crematory ovens and the alleged homicidal gas chambers of Auschwitz-Birkenau technically, although he lacked the required technical competence to undertake such a study. Nevertheless, Pressac had to accept the Revisionists’ methodological principle, according to which, where testimonies and technics disagree, it is the latter which must prevail. He has applied that principle by reducing the number of the alleged victims of homicidal gassing, due precisely to its incompatibility with the capacity (craftily inflated by him) of the crematory ovens. In this manner, he has opened an irreparable leak in traditional historiography, because technology reveals the material impossibility of mass extermination at Auschwitz-Birkenau. If therefore, Pressac wants to be coherent in his technical stance, all that remains for him is to accept this conclusion. If he does not accept it, he can only go backwards, declaring, in acceptance of the 1979 appeal of the French historians, that one must not ask how such alleged mass extermination was technically possible.

In any case, one thing is certain: These Auschwitz books by Jean-Claude Pressac represent the end of a legend.


NOTES

1.

L’Express, 23-29 September 1993, pp. 78 and 80.

2.

These examples are sufficient to illustrate the level of technical competence of Jean-Claude Pressac:

He thinks that "the temperature has to be raised to 27°C for hydrocyanic acid to evaporate" (J.C. Pressac, Auschwitz: Technique and Operation of the Gas Chambers, p. 375), ignoring that evaporation of hydrocyanic acid can occur even below its boiling point (25.6°C), even at temperatures below 0°C [See in this connection: G. Peters, Die hochwirksamen Gase und Dämpfe in der Schädlingsbekämpfung: Sammlung chemischer und chemischtechnischer Vorträge [Stuttgart: Verlag von Ferdinand Enke, 1942], pp. 85-88].

Regarding the crematory ovens, Pressac presents an "Operation plan of a Topf oven with three chambers which was built in two models in Crematories II and III" (J.C. Pressac, Auschwitz: Technique and Operation of the Gas Chambers, op. cit., p. 492) based on the deposition of H. Tauber, in which the gases of the gasogenes pass around the chambers: The technical basis of this plan is a translation error! (Pressac’s two translators have translated the Polish preposition przez, (through), as around. The translation error is found on p. 489. The Polish text says: przez obie boczne retorty [through the two lateral chambers] (Archiwum Państwowego Muzeum w Oświęcimiu, hereafter: APMO, Dpr.-Hd, 11a, p. 133).

3.

Pressac, Auschwitz: Technique and Operation of the Gas Chambers, p. 264.

4.

Ibid., p. 183.

5.

Ibid., p. 97.

6.

Ibid., p. 183.

7.

Kostenanschlag of Topf for KL Mauthausen of 1 November 1940. Bundesarchiv Koblenz (hereafter: BK), NS4 Ma/54.

8.

Letter from Topf to SS-Neubauleitung KL Mauthausen of 6 January 1941. BK, NS4 Ma/54.

9.

Letter from Topf to SS-Neubauleitung KL Mauthausen of 14 July 1941. Staatsarchiv Weimar, LK 4651.

10.

"Factors Which Affect the Process of Cremation: Third Session" by Dr. E.W. Jones, assisted by Mr. R.G. Williamson. Extracted from: The Cremation Society of Great Britain Annual Cremation Conference Report, 1975.

11.

APMO, D-Z/Bau, nr. inw. 1967, p. 65.

12.

Letter of H. Kori G.m.b.H. to SS-Sturmbannführer Lenzer, Lublin, of 23 October 1941. Archiwum Państwowego Muzeum na Majdanku, sygn. VI-9a, vol.1.

13.

APMO, BW 30/46, p. 18.

14.

APMO, BW 30/46, p. 6.

15.

On the Volckmann-Ludwig oven see:
- Dipl. Ing. Volckmann, Hamburg, "Ein neues Einäscherungsverfahren", Zentralblatt für Feuerbestattung, 1931;
- Kurt Prüfer, "Ein neues Einäscherungsverfahren. Eine Entgegnung", Die Flamme, 40 Jg., 1931;
- Richard Kessler, "Der neue Einäscherungsofen System Volckmann-Ludwig", Zentralblatt für Feuerbestattung, 1931;
- Friedrich Helwig, "Vom Bau und Betrieb der Krematorien", Gesundheits-Ingenieur, 54. Jg., Heft 24, 1931;
- H. Wolfer, "Der neue 'Volckmann-Ludwig' - Einäscherungsofen im Stuttgarter Krematorium", Gesundheits-Ingenieur, 55. Jg., Heft 13, 1932.

16.

H.R. Heinicke, VL-Kremationsofen Bauart Heinicke. Summary of the sales kindly furnished by the H.R. Heinicke company of Stadthagen.

The two Volckmann-Ludwig ovens installed in the crematory of Dortmund in 1937 are described in Herman Kamper, "Der Umbau der Leichenverbrennungsöfen und die Einrichtung von Leichenkühlräumen auf dem Hauptfriedhof der Stadt Dortmund", Gesundheits-Ingenieur, 64. Jg., Heft 12, 1941.

17.

R. Kessler, "Rationelle Wärmewirtschaft in den Krematorien nach Massgabe der Versuche im Dessauer Krematorium", Die Wärmewirtschaft, 4. Jg., Heft 8-11, 1927.

18.

W. Heepke, Die Leichenverbrennungs-Anstalten (die Krematorien) (Halle a.S.: Verlag von Carl Marhold, 1905), p. 71.

19.

R. Schnabel, Macht ohne Moral: Eine Dokumentation über die SS (Frankfurt/Main: Röderberg-Verlag, 1957), p. 351.

20.

Dr. H. Frölich, "Zur Gesundheitspflege auf den Schlachtfeldern", Deutsche Militärärztliche Zeitschrift, I, 1-4, 1872, pp. 109-110.

21.

Kostenanschlag auf Lieferung von 2 Stück Dreimuffel Einäscherungs-Öfen und Herstellung des Schornsteinfutters mit Reinigungstür of Topf, 12 February 1942. APMO, BW 30/34, pp. 27-33.

22.

W. Heepke, "Die neuzeitlichen Leicheneinäscherungsöfen mit Koksfeuerung, deren Wärmebilanz und Brennstoff verbrauch", Feuerungstechnik, 21. Jg., Heft 8/9, 1933.

23.

List of the cremations of the crematory at Gusen (26 September-12 November 1941). Öffentliches Denkmal und Museum Mauthausen, Archiv, B 12/31.

24.

D. Czech, Kalendarium der Ereignisse im Konzentrationslager Auschwitz-Birkenau 1939-1945 (Reinbeck bei Hamburg: Rowohlt Verlag, 1989), p. 281.

25.

H. Langbein, Menschen in Auschwitz (Vienna: Europaverlag, 1987), p. 74.

26.

APMO, BW 30/7/34, p. 54.

27.

The numbers of the deceased are based on the Leichenhallenbuch [27a] and on the Sterbebücher [27b] for the male camp, on the documents of the clandestine resistance movement [27c] revised and corrected for the female camp.

27a.

APMO, sygn. D-Au-I-5.

27b.

APMO, Φ 502-4.

27c.

APMO, Ruch Oporu, t. II, sygn. RO/85, pp. 62, 62a.

28.

APMO, D-Au-I-4, segregator 22, 22a.

29.

This number is derived from the Auschwitz Kalendarium.

30.

J.C. Pressac, Auschwitz: Technique and Operation of the Gas Chambers, p. 227.

31.

Commander at Auschwitz: Autobiographical memoirs of Rudolf Höss (Turin: Einaudi, 1985), p. 180.

Kommandant in Auschwitz. Autobiographische Aufzeichnungen des Rudolf Höss ed. Martin Broszat (Munich: Deutscher Taschenbuch Verlag, 1981), p. 171.

32.

J.C. Pressac, Auschwitz, p. 132.

33.

Ibid., p. 183.

34.

Ibid., p. 183.

35.

Ibid., p. 236.

36.

Ibid., p. 236.

37.

Ibid., pp. 162 and 213.

38.

Ibid., pp. 236 and 390.

39.

F. Piper, "Estimating the Number of Deportees to and Victims of the Auschwitz-Birkenau Camp", Yad Vashem Studies 21 (Jerusalem, 1991), 49-103; Auschwitz. Wie viele Juden, Polen, Zigeuner ... wurden umgebracht (Krakow: Universitas, 1992).

40.

R. Jakobskötter, "Die Entwicklung der elektrischen Einäscherung bis zu dem neuen elektrisch beheizten Heissluft einäscherungsofen in Erfurt", Gesundheits-Ingenieur, 64. Jg., Heft 43, 1941, p. 583

41.

The crematory oven of Gusen went into operation on 29 January 1941. From February to October 1941 in the Gusen camp, 3,179 prisoners died. H. Marsalek, Die Geschichte des Konzentrationslager Mauthausen. Dokumentation (Vienna: Österreichische Lagergemeinschaft Mauthausen, 1980), p. 156.

42.

Bescheinigung über besondere Berechnung geleistete Tagelohn-Arbeiten, 12 October - 9 November 1941. BK, NS4 Ma/54.

43.

R.L. Braham, The Politics of Genocide: The Holocaust in Hungary (New York: Columbia University Press, 1981). See the summary table of the deportations on p. 602 of vol. 2.

44.

Records of the Defense Intelligence Agency (RG 373). Mission 60 PRS/462 60 SQ, CAN D 1508, Exposure 3055, 3056.

45.

J.C. Ball, Air Photo Evidence (Delta, B.C., Canada: Ball Resource Services Ltd.).

46.

See in this connection: Carlo Mattogno, La soluzione finale: Problemi e polemiche (Edizioni di AR, Padua 1991).

47.

Commander at Auschwitz, pp. 171-174; Kommandant in Auschwitz, pp. 157-159.

48.

D. Czech, Kalendarium der Ereignisse im Konzentrationslager Auschwitz-Birkenau, in: Hefte von Auschwitz, Wydawnictwo Państwowego Muzeum w Oświęcimiu, 3, 1960, p. 49.

49.

D. Czech, Kalendarium der Ereignisse im Konzentrationslager Auschwitz-Birkenau 1939-1945, p. 186.

50.

Ibid., p. 239.

51.

Ibid., pp. 117-119.

52.

S. Klodziński, "Pierwsza zagazowanie więzniów i jenców w obozie oświęcimskim", Przegląd Lekarski, I, 1972.

53.

Pressac, Auschwitz: Technique and Operation of the Gas Chambers, p. 132.

54.

Carlo Mattogno, Auschwitz: La prima gasazione [Auschwitz: The First Gassing] (Edizioni di AR, Padua 1992).

55.

Ibid., p. 159: "Since moreover the first gassing, according to the judge Jan Sehn, was an execution of condemnees to death selected by the commission presided over by Mildner, which arrived at Auschwitz ‘in November 1941’ and concluded its work ‘after a month’, the first gassing in any case could not have occurred before December."

56.

Ibid., p. 85.

57.

Ibid., p. 84.

58.

Pressac even presents his own drawing of such columns (Auschwitz: Technique and Operation of the Gas Chambers, p. 487).

59.

Mattogno, Auschwitz: La prima gasazione [Auschwitz: The First Gassing], pp. 131-132. The reference is to the article by G. Peters and W. Rasch, "Die Einsatzfähigkeit der Blausäure-Durchgasung bei tiefen Temperaturen", Zeitschrift für hygienische Zoologie und Schädlingsbekämpfung, 1941.

60.

Ibid., pp. 28-29 and 36-37.

61.

Pressac, Auschwitz: Technique and Operation of the Gas Chambers, pp. 184.

62.

The Auschwitz Kalendarium places this alleged order under 2 November 1944 (op. cit., p. 921).

63.

The mortality indicated by Pressac, based on the Sterbebücher, is lower than the actual, because these registers contain only a small portion of the deaths that took place among the female prisoners.

64.

NO-2362, NO-2363; Kalendarium der Ereignisse im Konzentrationslager Auschwitz-Birkenau, p. 259.

65.

Pressac, Auschwitz: Technique and Operation of the Gas Chambers, p. 188.

66.

Auschwitz vu par les SS [Auschwitz Seen by the SS] (Musée d’Etat à Oświęcim [Oświęcim State Museum], 1974), p. 337.

The WVHA was informed monthly of the number of prisoner deaths in all the concentration camps, including Auschwitz (PS-1469).

67.

Pressac, Auschwitz: Technique and Operation of the Gas Chambers, p. 302.

68.

Ibid., p. 484.

69.

See note 72.

70.

APMO, D-Z/Bau, nr. inw. 1967, pp. 246-247 (see appendix, Document No. 2).

71.

APMO, D-Z/Bau, nr. inw. 1967, pp. 231-232 (see appendix, Document No. 3).

72.

Pressac, Technique and Operation of the Gas Chambers, p. 286.

73.

Heepke, Die Leichenverbrennungs-Anstalten, p. 104 (see appendix, Document No. 4).

74.

This follows, among other things, from the article by G. Peters and E. Wüstiger mentioned by Pressac on pages 41 and 103, from which he also draws Documents 16-17. The title indicated by J.C. Pressac: "Entlausung mit Zyklon Blausäure in Kreislauf-Begasungskammern", (Zeitschrift für hygienische Zoologie und Schädlingsbekämpfung, Heft 10/11, 1940) (note 134 on p. 103) is wrong; the exact title is: "Sach-Entlausung in Blausäure-Kammern" (Zeitschrift für hygienische Zoologie und Schädlingsbekämpfung, 1940, pp. 191-196). On page 195 one reads:

Ventilator mit Motor. Für diesen ist eine Leistung von 12 cbm je Minute bei einem stat. Druck von 80 mm WS ausreichend, um sowohl eine äusserst rasche Gasentwicklung als auch eine genügend rasche Luftung (72-facher Luftwechsel je Stunde) des begasten Kammerinhalts zu bewirken.

[Ventilator with motor. For this a capacity of 12 m3 per minute with a static pressure of 80 mm of water column is sufficient to produce a very rapid development of the gas as well as a sufficiently rapid ventilation (72 air exchanges per hour) of the contents of the gassed chamber.] (See appendix, Documents No. 5 and 6.)

75.

Pressac, Auschwitz: Technique and Operation of the Gas Chambers, p. 370.

76.

Ibid., p. 68.

77.

APMO, nr. neg. 1034/7, p. 5.

78.

F. Puntigam, H. Breymesser, E. Bernfus, Blausäuregas kammern zur Fleckfieberabwehr. Sonderveröffentlichung des Reicharbeitsblattes (Berlin, 1943), p. 31.

79.

"Hütte" des Ingenieurs Taschenbuch (Berlin: Verlag von Wilhelm Ernst & Sohn, 1931), vol. 1, p. 1013, No. 3, with specific and exclusive reference to the "Rauchgasanalyse" (analysis of combustion gases) (p. 1011) (see appendix, Document No. 9).

80.

Blausäuregaskammern zur Fleckfieberabwehr, p. 21.

81.

Letter of the Tesch & Stabenow company of 29 July 1942 to the Waffen-SS Kriegsgefangenenlager Lublin, Verwaltung (see appendix, Documents No. 10 and 11).

82.

APMO, BW 30/25, p. 8.

83.

Pressac, "Les Carences et Incoherences du 'Rapport Leuchter'" [The Deficiencies and Incoherences of the ‘Leuchter Report’], Jour J, 1988, p. III.

84.

Übergabverhandlung of Crematory II, 31 March 1943. APMO, BW 30/43, p. 12.

85.

Übergabverhandlung of Crematory III, 24 June 1943. APMO, BW 30/43, p. 24.

86.

Blausäuregaskammern zur Fleckfieberabwehr. This work describes with great accuracy two types of gas chambers: the standard ones with the Kreislauf system and the Behelfsmässige Blausäurekammern (pp. 62-68).

87.

On page 59, Pressac writes regarding prisoners working at cremating corpses buried in common graves:

Ils étaient devenus, involontairement, les seuls témoins, en dehors des SS, des signes extérieurs du massacre des Juifs car, parmi les détenus qui participèrent à ce ‹nettoyage›, aucun ne fut laissé en vie. (our italics)

[They had become, involuntarily, the only witnesses, outside of the SS, of the outward signs of the massacre of the Jews, for among the prisoners that had participated in this ‘cleansing’, none were left alive.]

How, then, is the fact explained that the eye witnesses of the alleged extermination activity of the Bunkers were still left alive?

88.

M. Nyiszli, Médecin à Auschwitz: Souvenirs d’un médecin déporté, traduit et adapté du hongrois par Tibère Kremer [M. Nyiszli, Doctor at Auschwitz. Memories of a Deported Doctor, translated and adapted from Hungarian by Tibere Kremer] (Paris: Julliard, 1961), pp. 96-98.

89.

Pressac, Auschwitz: Technique and Operation of the Gas Chambers, p. 479. Actually M. Nyiszli is a false witness. See in this connection our study «Medico ad Auschwitz»: Anatomia di un falso ["Doctor at Auschwitz": Anatomy of a Fake] (Parma: Edizioni La Sfinge, 1988).

90.

Pressac, Auschwitz: Technique and Operation of the Gas Chambers, p. 397.

91.

Ibid., p. 392.

92.

Ibid., p. 384.

93.

Ibid., p. 384.

94.

Ibid., pp. 416-417.

95.

APMO, nr. neg. 20995/465.

96.

Pressac, Auschwitz: Technique and Operation of the Gas Chambers, p. 386.

97.

Ibid., p. 399.

98.

APMO, BW 30/27, p. 18.

99.

Le Monde, 21 February 1979, p. 23.


!1.

Page numbers with no other indications refer to: Les crématoires d’Auschwitz: La machinerie du meurtre de masse by J.- C. Pressac.

!2.

Pressac states that these archives were left intact because the second and last director of the Bauleitung, SS-Obersturmführer Werner Jothann ignored the "‘explosive’ contents of the documents" since l’aménagement homicide des crématoires (the fitting out of the crematories for homicide) was carried out under the direction of SS-Sturmbannführer Karl Bischoff (p. 1). But on page 88, Pressac contradicts himself, stating that Bischoff "avait pris la tête de l’Inspection des constructions ‹Silésie›, mais gardait le contrôle de la Bauleitung d’Auschwitz" ["had been in charge of the inspection of ‘Silesia’ constructions, but retained control of the Bauleitung of Auschwitz"] (our italics).

!3.

The three-chambered oven, using preheated air in the lateral chambers, had an approximate yield greater by 1/3 than that of the two-chambered oven, that is a lesser fuel consumption by 1/3; in compensation, the volumetric transition velocity of the fumes in the central chamber was greater than their velocity of combustion, hence they burned in the smoke conduits. This, along with a careless use of the intake draft systems, caused the damage of Crematory II at the end of March 1943.

!4.

H. Tauber states that the SS was annoyed because: "according to the calculation and plans for this crematory, five to seven minutes were alloted to burn one corpse in a chamber!" (Jean-Claude Pressac, Auschwitz: Technique and Operation of the Gas Chambers, p. 489). The testimony of H. Tauber is full of thermotechnically unfounded statements of this type.

!5.

This is the maximum theoretical capacity. The existing documents show that the effective capacity was much less.

!6.

The future section BIII at Birkenau.

!7.

In the Auschwitz construction plan of 28 October 1942, a disinfestation facility Entwesungsanlage of 1,000 m2 für Sonderbehandlung was foreseen and planned specifically for the hygienic-sanitary treatment of the prisoners; it was endowed with heat, showers, and disinfestation installations (Heiz-Brause- u. Desinfektionsanlage) and had a cost of 73,680 RM. Another Entwesungsanlage smaller (262.84 m2) was destined for the guard troops (für die Wachtruppe). Zusammenstellung des Bauvorhaben Kriegsgefangenenlager Auschwitz (Durchführung der Sonderbehandlung, 28 Oktober 1942. Photocopy in: Florian Freund, Bertrand Perz, Karl Stuhlpfaffer, "Der Bau des Vernichtungslager Auschwitz-Birkenau", Zeitgeschichte, Heft 5/6, May-June 1993, p. 207).

!8.

SS-WVHA = (Wirtschafts-Verwaltungshauptamt: Economic and Administrative Offices of the SS)

!9.

With the term bavure Pressac means:

toute indication relevée dans un document quelconque (écrit, plan, photo) relatif à un emploi anormal des crématoires et ne pouvant s’expliquer que par le gazage massif d’êtres humains. (p. 60)

[Any indication noted in any document whatsoever (writing, plan, photo) relating to an abnormal use of the crematories that could only be explained by the massive gassing of human beings.]

!10.

That is not unusual. Disinfestation installations were planned in the crematories at Majdanek, [!10a] Dachau, [!10b] and Stutthof. [!10c]

!10a.

Plan of 23 October 1941 (Archiwum Państwowego Muzeum na Majdanku, sygn. VI-9a, vol.1)

!10b.

Plans of the "Baracke X" (crematorium) of March 1942 (NO-3884, NO-3885, NO-3887)

!10c.

Plan of the crematorium of 29 May 1945 (Pressac, Technique and Operation of the Gas Chambers, p. 561)

!11.

The body of a normal adult, standing, produces 1.72 kcal per minute. [!11a] 1,800 bodies produce therefore 3,096 kcal per minute. The heat of vaporization of hydrocyanic acid is 6.67 kcal/mole; since its molecular weight is 27.03, the vaporization heat of 6 kg of hydrocyanic acid is (6,000 × 6.67) / 27.03 = 1,480 kcal, less than half of the heat produced by 1,800 bodies in one minute.

!11a.

F. Flury, F. Zernik, Schädliche Gase, Dämpfe, Nebel, Rauch und Staubarten, Verlag von Julius Springer, Berlin 1931, p. 29.

!12.

The gas in question called Arginal (Alkylformiat) was used together with Cartox for the disinfestation of silos against calandra granaria (grain weevil), a fearful grain parasite (H.W. Frickhinger, Schädlingsbekämpfung für Jedermann, Leipzig: Heilingsche Verlagsanstalt, 1942; G. Peters, Die hochwirksamen Gase und Dämpfe in der Schädlingsbekämpfung: Sammlung chemischer und chemischtechnischer Vorträge, Stuttgart: Verlag von Ferdinand Enke, 1942, pp. 37-38 and 55-57).

!13.

The indication of 50 m3 is obviously a printing error.

!14.

This datum, in reference to the first alleged homicidal gassing in Crematory II, is pure invention by Jean-Claude Pressac, because in this connection, there exists no document, and no witness affirms that on such an occasion, 6 kg of Zyklon B were used. Pressac draws this datum from R. Höss, who speaks in general of 5-7 cans of Zyklon B, of 1 kg each (NI-034, NI-036).

!15.

Printing error, for 480 m3.


APPENDIX

Preface to the List of Documents

The technical documents included in this supplement [or appendix] refer to two important aspects of the presumed machinerie de meurtre de masse [machinery of mass murder] treated in this study: that of the ventilation of the Leichenkeller [basement morgue] of Crematories II and III, and that of the Gasprüfer [gas testers].

According to Pressac, Leichenkeller 1 is the presumed homicidal gas chamber. Documents 2 and 3 refute the affirmations of Jean-Claude Pressac, according to which the capacity of the ventilators of the ventilating system for Crematories II and III of Birkenau was 8,000 m3 of air per hour: the actual capacity was 4,800 m3 of air per hour, corresponding to 9.48 exchanges of air per hour.

These documents show, moreover, that the capacity of the Entlüftung exhaust ventilator of Leichenkeller 2 (the presumed dressing room) was 10,000 m3 of air per hour, corresponding to 11 exchanges of air per hour. The consequence is that, paradoxically, according to the Pressac thesis, the Zentralbauleitung engineers of Auschwitz and the Topf engineers provided a lower number of air exchanges for the homicidal gas chamber than for the dressing room!

The number of air exchanges planned for these locations is in reality that which Engineer Wilhelm Heepke, one of the most specialized German engineers in the field of crematories, prescribed for morgue rooms planned for intense use (Document 4). Therefore, the Leichenkeller were projected and constructed as mortuary chamber morgues.

The warm-air circulation disinfestation chambers constructed by DEGESCH (DEGESCH-Kreislauf-Anlage für Entlausung mit Zyklon-Blausäure) had in effect a ventilator with a capacity of 12 m3 of air per minute, corresponding to 72 exchanges of air per hour (Documents 5 and 6).

With this falls also the affirmation of Pressac according to which Leichenkeller 1 was supposed to have been transformed into a homicidal gas chamber. The fact that Crematories II and III, which were projected and constructed as plain hygienic-sanitary installations, entered into function with the same number of ovens and with the same capacity of the ventilators of Leichenkeller 1 projected from the very beginning, demonstrates that they were not transformed into machinerie de meurtre de masse [machinery of mass murder].

On 26 February 1943, the Zentralbauleitung of Auschwitz requested Topf to send ten Gasprüfer (Document 7). The Topf company supposedly responded with a letter dated 2 March 1943 which speaks of Anzeigegeräte für Blausäure Reste [instruments for indicating remnants of hydrocyanic acid, or prussic acid] (Document 8). Jean-Claude Pressac attributes to this document the value of a definite proof of the existence of a homicidal gas chamber in Crematory II.

The Gasprüfer was a device for the analysis of burnt gases functioning according to physics (Document 9). The test kit for Zyklon B residual gas was called Gasrestnachweisgerät für Zyklon [apparatus which shows evidence of residual gas] (Document 11) and it functioned according to chemical methods; this was distributed by the same company which supplied the Zyklon B. Gasrestnachweisgerät came in a small wooden box containing:

1) eine Flasche mit Lösung I (2,86 g Kupferazetat im Liter)
2) eine Flasche mit Lösung II (475 cm3 bei Zimmer temperatur gesättigte Benzidinazetatlösung und 525 cm3 Wasser)
3) Mischgefäss mit 2 Marken zum Abmessen gleicher Raumteile
4) Papprolle mit Fiesspapierstreifen
5) Farbmuster (Papierstreifen in Reagenzglas)
6) sechs leere dickwandige Reagenzgläser mit Korken.

[ 1) a bottle with Solution 1 (2.86 g of copper acetate per liter)
2) a bottle with Solution 2 (475 ml at room temperature of a saturated solution of benzidine acetate in 525 ml of water)
3) a mixing utensil with two markers for measuring equivalent room areas
4) a cardboard roll with strips of blotting paper
5) color test pattern (strips of test paper in a test tube)
6) six empty thick-walled test tubes with cork stoppers].

The test for the residual gas (Gasrestprobe) was accomplished by mixing in the mixing utensil the required proportions of Solutions 1 and 2. In the solution thus obtained, one immersed the lower parts of six strips of blotting paper, each one of which was then introduced into a test tube which was quickly secured with a cork stopper. The person performing the test, wearing a gas mask, entered the testing area with the test tubes which were opened at various locations exposing the strips of blotting paper which were moistened with the test solution. The paper strips reacted in the presence of hydrocyanic acid gas, taking on a blue coloration which becomes more intense with a higher concentration of the gas. (A. Sieverts, A. Hermsdorf, "Der Nachweis gasförmiger Blausäure in Luft", Zeitschrift für Angewandte Chemie, 34. JG., 1921, pp. 4-5).

Document 11 is a photograph of a Gasrestnachweisgerät which was found by the Soviets at Auschwitz after the liberation of the camp.

Document 12 is an Allied aerial photo of Auschwitz 31 May 1944. Also published in The Ball Report (Toronto 1993), p. 5.

LIST OF DOCUMENTS

Document No. 1: The table of contents of the study entitled, The Crematory Ovens of Auschwitz-Birkenau, by Carlo Mattogno with the collaboration of Engineer Dr. Franco Deana, of Genoa, Italy.

Document No. 2: J.A. Topf & Söhne, Erfurt. Billing No. 729 of 27 May 1943 addressed to Zentralbauleitung der Waffen-SS und Polizei Auschwitz concerning the Crematory III of Birkenau. APMO, D-Z/Bau, Nr. inw. 1967, pp. 246-247.

Document No. 3: J.A. Topf & Söhne, Erfurt. Billing No. 171 of 22 February 1943 addressed to Zentralbauleitung der Waffen-SS und Polizei Auschwitz concerning Crematory II of Birkenau.

Document No. 4: Wilhelm Heepke, Die Leichenverbrennungs-Anstalten (die Krematorien). Verlag von Carl Marhold, Halle a.S., 1905, p. 104.

Document No. 5: G. Peters, E. Wüstiger, "Sach-Entlausung in Blausäure-Kammern", Zeitschrift für hygienische Zoologie und Schädlingsbekämpfung, Heft 10/11, 1940, p. 195.

Document No. 6: P. Puntigam, H. Breymesser, E. Bernfus, Blausäurekammern zur Fleckfieberabwehr. Sonderveröffentlichung des Reicharbeitsblattes (Berlin, 1943), p. 50.

Document No. 7: Telegram of the Zentralbauleitung Auschwitz to the Firm J.A. Topf & Söhne, Erfurt, of 26 February 1943. APMO, BW 30/34, p. 48.

Document No. 8: J.A. Topf & Söhne, Erfurt. Letter of 2 March 1943 to the Zentralbauleitung der Waffen-SS und Polizei Auschwitz. J.C. Pressac, Les Crématoires d’Auschwitz: La maschinerie du meurtre de masse (Paris: CNRS Editions, 1993), Document 28.

Document No. 9: "Hütte" des Ingenieurs Taschenbuch (Berlin: Verlag von Wilhelm Ernst & Sohn, 1931), vol.1, pp. 1010-1013.

Document No. 10: Tesch & Stabenow, Hamburg. Letter of 29 July 1942 to the Waffen-SS Kriegsgefangenenlager Lublin, Verwaltung. Archiwum Państwowego Muzeum na Majdanku, sygn. I d 2, vol. 1, p. 107.

Document No. 11: APMO, photograph from Nr. neg. 625. The Gasrestnachweisgerät für Zyklon test kit.

Document No. 1

The Contents of a forthcoming publication entitled:

THE CREMATORY OVENS OF AUSCHWITZ-BIRKENAU

Introduction
1. Structure and Operation of the Gasogene Crematory Ovens Heated with Coke.
2. Heat Balance of a Gasogene Crematory Oven Heated with coke.
3. The Duration of the Cremation Process in Gasogene Crematory Ovens Heated with Coke.
4. The Topf Crematory Oven with Two Chambers Heated with Coke.
5. The Topf Crematory Oven with Three Chambers Heated with Coke.
6. The Topf Crematory Oven with Eight Chambers Heated with Coke.
7. Heat Balance of the Topf Crematory Ovens of Auschwitz Birkenau.
7.1 Heat Balance of the Oven with Two Chambers for Two Normal Corpses.
7.2 Observations on Heat Balance.
7.3 Heat Balance for Two Emaciated Corpses.
7.4 Heat Balance for Two Corpses with an Average Loss of Weight.
7.5 The Coke Consumption of the Crematory Oven at Gusen.
7.6 Heat Balance for the Ovens with Three and Eight Chambers.
7.7 Thermotechnical Considerations on the Ovens with Three Chambers.
8. The Duration of the Cremation Process in the Topf Ovens of Auschwitz-Birkenau.
8.1 The Documents.
8.2 The Cremation Experiments of Engineer Kessler.
8.3 The List of the Cremations of the Crematory at Gusen.
8.4 The Simultaneous Cremation of Several Corpses in a Chamber.
8.5 The Capacity of the Topf Crematories of Auschwitz Birkenau.
9. The Number of the Cremated in the Ovens of Auschwitz Birkenau.
9.1 The Number of the Cremated According to J.C. Pressac.
9.2 The Activity of the Crematories of Auschwitz-Birkenau.
9.3 The Duration of the Refractory Covering of the Ovens at Auschwitz-Birkenau.
9.4 The Number of the Cremated in 1943: the Estimate of the SS.
9.5 The Number of the Cremated in 1943: the Consumption of Coke.
9.6 The Deportation of the Hungarian Jews to Auschwitz.
10. The Cremation Pits of Birkenau.
10.1 The Witness Filip Müller.
10.2 The Combustion Process in a Cremation Pit.
10.3 Human Fat Recovery System.
10.4 The Procedure of Human Fat Recovery.
10.5 Mass Cremation Arrangements for Epidemics and Battlefields.

Document No. 2:

Billing No. 729 of 27 May 1943 from J.A. Topf & Söhne, Erfurt. Addressed to Zentralbauleitung der Waffen-SS und Polizei Auschwitz concerning Crematory III of Birkenau. APMO, D-Z/Bau, Nr. inw. 1967, pp. 246-247: page1 | page 2.

Document No. 3:

Billing No. 171 of 22 February 1943 from J.A. Topf & Söhne, Erfurt. Addressed to Zentralbauleitung der Waffen-SS und Polizei Auschwitz concerning Crematory II of Birkenau: page1 | page 2.

Document No. 4:

Wilhelm Heepke, Die Leichenverbrennungs-Anstalten (die Krematorien). Verlag von Carl Marhold, Halle a.S., 1905, p. 104.

Document No. 5:

G. Peters, E. Wüstiger, "Sach-Entlausung in BlausäureKammern", Zeitschrift für hygienische Zoologie und Schädlingsbekämpfung, Heft 10/11, 1940, p. 195.

Document No. 6:

P. Puntigam, H. Breymesser, E. Bernfus, Blausäurekammern zur Fleckfieberabwehr. Sonderveröffentlichung des Reicharbeitsblattes (Berlin, 1943), p. 50.

Document No. 7:

Telegram of the Zentralbauleitung Auschwitz to the Firm J.A. Topf & Söhne, Erfurt, of 26 February 1943. APMO, BW 30/34, p. 48.

Document No. 8:

J.A. Topf & Söhne, Erfurt. Letter of 2 March 1943 to the Zentralbauleitung der Waffen-SS und Polizei Auschwitz. J.C. Pressac, Les Crématoires d’Auschwitz: La machinerie du meurtre de masse (Paris: CNRS Editions, 1993), Document 28.

Document No. 9:

"Hütte" des Ingenieurs Taschenbuch (Berlin: Verlag von Wilhelm Ernst & Sohn, 1931), vol.1, pp. 1010-1013.

Document No. 9 (second part):

"Hütte" des Ingenieurs Taschenbuch (Berlin: Verlag von Wilhelm Ernst & Sohn, 1931), vol.1, pp. 1010-1013.

Document No. 10:

Tesch & Stabenow, Hamburg. Letter of 29 July 1942 to the Waffen-SS Kriegsgefangenenlager Lublin, Verwaltung. Archiwum Panstwowego Muzeum na Majdanku, sygn. I d 2, vol. 1, p. 107.

Document No. 11:

Picture of "Gasrestnachweisgerätes für Zyklon", found by the Soviets in the Auschwitz camp after its  liberation. APMO, Nr. neg 625.

 

GLOSSARY

Aide: A military officer acting as assistant to a superior

AL: Arbeitslager, work camp

Aktenvermerk: File entry

Alimentation: Allowance

Aleatorie: Coincidental

Amtsgruppe: Official group

Anzeigegeräte: Indicators

APMO: Archives of the Polish Museum at Oświęcim [Auschwitz]

Arginal [Areginal]: A Gas

Aspiration: Exhaust process

Aufnahmegebäude: Admittance building

Aufzeichnungen: Notes, records

Auschwitz: German spelling for the Upper Silesian town, [Polish name Oświęcim, c. 45,000 inhabitants] located 2 km Southwest from the former large complex which is known as Auschwitz-Birkenau

Auskleidekeller: Disrobing basement

Badenanstalten: Swimming, bathing facilities

Baracken: Barracks

Bauhof: Construction yard

Bauleitung: Construction administration, office in charge of construction

Bavure: Trace

Behelfsmässig: Temporary, makeshift, improvised

Berücksichtigt: In consideration

Bescheinigung: Receipt, certificate

Blausäure: Hydrocyanic acid

Blausäuregaskammern: Hydrocyanic acid gas chambers

Blausäurekammern: Hydrocyanic acid chambers

Boos: Name of a German manufacturer (Friedrich Boos)

Brennstoffverbrauch: Fuel consumption

Buchenwald: Location of a German concentration camp

Bundesarchiv: German archives

Bunker: Shelter

Chemischtechnischer: Chemical-technical

Crypto-revisionist: Secret, or underground revisionist

Czech (Danuta): Author of Kalendarium der Ereignisse im Konzentrationslager Auschwitz-Birkenau 1939-1948

Degesch-Kreislauf: Circulatory system type Degesch

Drahtnetzeinschiebvorrichtung: Meshed-wire slider

Drahtseil: Wire rope

Dreimuffel-Einäscherungs-Öfen: Three-chambered crematory ovens

Druckluftgebläse: Compressed air blower

Druckluft-Anlage: Compressed air installation

Einäscherungsofen: Crematory oven

Einäscherungsverfahren: Cremation procedure

Einsatzfähigkeit: Operational capability

Einwurfvorrichtung: Insertion installation

Entlausungsbaracken: Disinfestation barracks, barracks for delousing

Entlüftungsanlage: Exhaust equipment

Entlüftungskanäle: Exhaust channels

Erfurt: A German city

Europaverlag: A German publishing establishment

Feuerbestattung: Cremation

Feuerungstechnik: Cremation technics

Fleckfieberabwehr: Typhus prevention

Gasentwicklung: Gas development

Gasogene (Gazogene): An apparatus producing gas for fuel by burning coke, charcoal or wood

Gasprüfer: Analyzer of the com bustion gases

Gasrestnachweisgeräte: Residual gas indicators (indicating instruments)

Generatorgase: Generator gases

Gestapo: Geheime Staatspolizei - Secret State Police later incorporated into the Reich Main Security Office headed by Heinrich Muller

Gesundheitspflege: Health care

Gesundheits-Ingenieur: Health engineer

Gleichschaltung: Co-ordination

Gusseisern: Wrought iron

Hamburg: Northern German port city

Handwinde: Hand winch

Hauptamt: Main office

Hauptfriedhof: Main cemetery

Hauptsturmführer: Captain of SS

Häftlinge: Prisoners

HCN: Hydrocyanic acid

Heisslufteinäscherungsofen: Hot air crematory oven

Holzblenden: Wooden shutters

Holzdeckel: Wooden lid(s)

Holzgebläse: Wooden blowers

HUTA: Acronym for Hoch und Tiefbau AG (a construction firm)

Judenumsiedlung: Jewish resettlement

Kammerinhalts: Room contents

Kellerzugang: Cellar or basement entrance

KGL: Prisoner of war camp

KL: War camp

Koblenz: A German city

Koksbeheizt: Heated by coke

Koksfeuerung: Coke burner

Konzentrationslager: Concentration camp

Kori: A German manufacturer of combustion plants

Kostenanschlag: Cost estimate

Kostenvoranschlag: Preliminary cost estimate

Krakow: A Polish city

Kreislauf: Circulation

Kreislauf-Begasungskammern: Circulatory gas chambers

Kriegsgefangenenlager: Prisoner of war camp

Lagergemeinschaft: Camp community

Leicheneinäscherungsöfen: Corpse crematory ovens

Leichenhalle: Mortuary

Leichenhallenbuch: Mortuary book or register

Leichenkeller: Mortuary basement

Leichenkühlräume: Corpse cooling rooms

Leichenverbrennungsöfen: Corpse cremating ovens

Leichenverbrennungs-Anstalten: Corpse cremating faciliities

Luftwechsel: Air exchange

Miasma: Noxious

Militärärztliche: Of a military physician nature

Nazi: Acronym for Nationalsozialistische (National Socialist)

Nord-süd: North-South

Obergruppenführer: The rank of General

Ofenanlage: Oven installation

Offenbach: A German city

Öfen: Ovens

Öffentlich: Public

Österreichisch: Austrian

Planrost: Level grate

Politruk: Soviet political officer charged with a number of functions among the troops, including political supervision and agitation. Commonly translated as "Commissar"

POW: Prisoner of war

Propaganda: Any organized movement to spread particular doctrines, information, etc.

Rauchgasanalyse: Smoke gas analysis

Rauchkanalschieber: Smoke channel slider (control mechanism)

Refractory: Fire retardant or resistant

Reichsarbeitblatt: German government worksheet

Reichsführer: Reich leader. Position occupied by Heinrich Himmler from 1929 to 1945

Reichsmark: The traditional German monetary unit

Revisionists: Those who look back again in order to correct or improve

RM: Reichsmark

RSHA: Reichssicherkeitshauptamt. Reich Main Security Office formed in 1939. Departments: Intelligence, Gestapo, Criminal Police and the SD (Sicherheitsdienst)

Sach-Entlausung: Material delousing

Saugzuganlage: Exhaust installation

Schädlingsbekämpfung: Pest control

Schlachtfeld: Battlefield

Schmiedeeisengebläse: Wrought-iron blower

Schornsteinfutter: Chimney casing

SD: Sicherheitsdienst, Security Service

Sonder: Out of the ordinary routine

Sonderbaumassnahmen: Special construction and building, undertaking and procedures

Sondermassnahmen: Special undertaking and procedures

Sonderveröffentlichung: Special publication

SS: Schutzstaffel, protective echelon

SS-Neubauleitung: SS Office for new construction

SS-Obersturmführer: The rank of Lieutenant

SS-Sturmbannführer: The rank of Major

SS-WVHA: SS Wirtschaftsverwaltungshauptamt, Central Office of the SS Economic Administration

Staatsarchiv: State archives

Stammlager: Original camp (Auschwitz) "Central Camp"

Sterbebücher: Death records

Sturmbannführer: The rank of Major

Tagelohn-Arbeiten: Part-time work

Taschenbuch: Pocket book (notebook)

Tesch: Bruno Tesch, engineer

Testa: Acronym for Tesch und Stabenow, German engineering firm

Topf: A German engineering and manufacturing firm

Topf-Doppelmuffel-Einäscherungs-Ofen: Topf two-chambered crematory oven

Topf-Zugverstärkungs-Anlage: Topf facility for increased circulation

Übergabeverhandlung: Transfer negotiations or proceedings of transfer

Verhältnismässig: In relation to VL-Kremationsofen: A crematory oven type built by the firm Volckmann-Ludwig

Wärmebilanz: Heat balance

Wärmewirtschaft: Heat distribution

WVHA: Wirtschaftsverwaltungshauptamt (Central Office of Economic Administration)

Zyklon B: Cyclon B chemical disinfestant



Back to Table of Contents
Next Contribution
Previous Contribution