The Combined Cipher machine - 1942-1962

Cipher machines were extensively used by both the Axis and the Allies during WWII.

The Germans had the Enigma, the Lorenz SZ 40/42 and the Siemens T-52.

The Japanese had the Type B Cipher Machine (Angooki Taipu B).

The Americans had the Converter M-134-C (SIGABA) and the Hagelin M-209.

The British had the Typex.

Another machine that became increasingly important for the Allies, in the period 1943-45, was the Combined Cipher Machine - CCM. Unfortunately, this machine has not received a lot of attention from historians because there is limited information available on its internal operation and use in the field.

Here I have attempted to present information on the CCM from various sources that are not easy to find (the timelines section has detailed sources): 

Historical overview

The need for a combined US-UK cipher system was apparent by 1941 so the British gave the Americans their Naval Cypher No 3 (a codebook enciphered with additive tables) for use in the N. Atlantic convoy operations.

The Americans wanted to use the M-138-A strip cipher and the M-209 cipher machine for intercommunication but both were rejected by the British. On the other hand, the Typex cipher machine which was considered secure could not be manufactured in large numbers. The British wanted the SIGABA but the Americans were not willing to share their high-level cipher machine or divulge its operating principle to their Allies.

Instead, a compromise solution was reached. A cipher attachment was developed by the US Navy (by Lt. Commander Donald W. Seiler USNR) that could be installed in the SIGABA and the Typex and it turned them into the Combined Cipher Machine – CCM. The CCM cipher attachment was a 5-rotor non reciprocal system issued with 10 reversible rotors.

Three versions were produced:

1). The CSP 1600 (CCM Mk I), which was a modification of an existing SIGABA with the CCM attachment installed.

2). The CSP 1700 (CCM Mk II), which was a SIGABA with a fixed CCM attachment.

3). The CSP 1800 (CCM Mk III), which was an adapter for the British Typex cipher machine (Typex Mk 23).

Production started in December 1942 and by November 1943 enough machines had been distributed to naval units to introduce the system in the North Atlantic. In April ’44 the CCM became operational with all five British and American armed services. By May ’45 the CCM was used extensively by the British Army, RAF and Navy.

Total production was at least 8.631 adapters and self-contained machines.

Operation and crypto security of the CCM (standard version) (a):

5 rotors – Positions 12345

Rotors have fixed cam contours (notches).

Rotor No3 steps with every key depression, it moves rotors No2 and No4, rotor No2 moves rotor No1 and rotor No4 moves rotor No5.

Non reciprocal encipherment (cipher-decipher function)

10 reversible rotors issued - 967.680 possible wheel orders

Certain rotor combinations can cause a machine period of only 338.

Message length limited to 200 groups in a message 

Message procedure (1944):                                                      

Two 5-letter indicators placed at the start and the end of the message. First indicator is the crypto net indicator, the second is the enciphered message wheel order.

The wheel order used to encipher the message (chosen at random) is enciphered on the key list’s initial code wheel alignment (contained in the keylist) before being sent as the second message indicator.

US procedure: Different initial code wheel alignment for SECRET, CONFIDENTIAL and RESTRICTED messages.

US-UK combined cipher procedure:  Two different initial code wheel alignments titled (A) and (B). The key list in use will define the classifications of messages for which (A) and (B) are to be used.

Cryptosecurity problems:

As early as 1943 it was discovered that under certain conditions the device had a dangerously low cipher period of 338.

A detailed study of the CCM’s period was not begun by OP-20-GM until 16 January 1944, six months after the short cycle defect was first discovered. The OP-20-GM investigation showed that, on one CCM circuit alone, seven daily keys between 12 February and 20 April 1944 produced the short 338 period. Once the problem of the short period had been fully identified, it could be avoided by taking various precautions, including the banning of certain rotor combinations and drastically restricting the length of messages. However, problems with lobstering rotors and the resulting short cycles were still being detected as late as March 1945.

A security evaluation in July ’44 came to the conclusion that despite these problems the CCM was a well-designed cipher machine since it had close to a million possible wheel orders, was non reciprocal so cribs could not be placed with certainty and the irregular stepping of the end rotors countered the cyclometric stepping of the fast-moving rotor No3.

Postwar developments:

In 1946 the CCM was acquired by the State Department and given the designation MCB. The modified version used by the State Department was the CSP 2200, with some rotors moving backwards. It continued to be used till July 1959. It was replaced by a cipher machine developed by the State Department called MEC.

In US and UK service the standard CCM was given the codename AJAX and in the early 1950’s the rotor stepping system was modified in order to increase crypto security. This version was called HERMES. It was replaced in the mid 1950’s by the KL-7 ADONIS. Another CCM version was codenamed LUCIFER but there is no information available on whether its stepping system had been modified.

NATO selected the CCM in 1951 to serve as the cipher system for second level communications (Naval, Army and Air Force Commands, down to Divisional H.Q. plus NATO war vessels). It was modified into the HERMES standard in 1952 and replaced by the KL-7 ADONIS in 1956. A number of CCM machines continued to be used for meteorological traffic and as a reserve till 1962.

The CCM HERMES modification (1952) (b)

In the early 1950’s the standard CCM was modified:

The stepping of the rotors was changed so that ‘The purpose of this modification is to change the rotor stepping pattern so that the No. 1 rotor stepping contact (operated by rotor No. 2) will control the stepping of rotor No. 5, and the No. 4 rotor stepping contact (operated by rotor No. 4) will control the stepping of rotor No. 1’.

Thus, it seems that in the HERMES version rotor No3 steps with every key depression, it moves rotors No2 and No4, rotor No2 moves rotor No5 and rotor No4 moves rotor No1.

Non reciprocal encipherment (cipher-decipher function)

20 reversible rotors were issued per machine (instead of the original 10) so that the possible wheel orders were increased to 59.535.360.

The wheels were equipped with rotatable cam contours (notches).

Use of message bisection, padding and variable spacing in order to counter cribbing.

Message length was 300 groups.

Simplex message procedure - HERMES (1952):

Two indicators placed at the start and the end of the message. First indicator is the crypto net indicator, the second is the message wheel order. The second indicator is phoneticized and transmitted in clear as the second to sixth groups of the message.

a). For messages classified Restricted, Confidential and Secret:

Single encipherment.

A 5-letter indicator is selected at random and sent with the message as the second to sixth 5-letter group (phoneticized). The machine is set up according to the daily key, the rotors are aligned to the letters of the message indicator, then the 10-letter Random Group No1 (contained in the keylist, different for each day) is enciphered and the resultant 10 letters are known as the intermediate sequence. The intermediate sequence is formed into five digraphs by pairing the 1st letter with the 6th, the 2nd with the 7th, the 3rd with the 8th, the 4th with the 9th and the 5th with the 10th. The digraphs are then used on the rotor-setting Table No. 1 (contained in the keylist) in order to identify the 5 rotors to be used for the message and their initial position. If a rotor number found in a cell is one that has been found in a cell that has already been recorded, the next cell to the right, which contains an available rotor number, is used. If the end of the row is reached, the first cell in the same row is resorted to. The rotors are rearranged and aligned according to the arrangement and alignment thus recorded from Table No 1. This is the message rotor arrangement and message rotor alignment on which the message is then enciphered.

b). For messages classified Top Secret:

Double encipherment.

A 5-letter indicator is selected at random and sent with the message as the second to sixth 5-letter group (phoneticized). The machine is set up according to the daily key, the rotors are aligned to the letters of the message indicator and the 10-letter Random Group No1 and 10-letter Random Group No2 (contained in the keylist) are enciphered. The resultant 20 letters are the intermediate sequence. The first ten letters of the intermediate sequence are formed into five digraphs and used with the rotor-setting Table No. 1 to identify the 5 rotors to be used for the message and their initial position (as above). The rotors are rearranged and aligned according to the arrangement and alignment thus recorded from Table No 1 and the message is enciphered. The result is known as the intermediate cipher text.

The message is enciphered a second time using the following procedure: The second ten letters of the intermediate sequence are formed into five digraphs by matching the 11th with the 16th, the 12^th with the 17th, the 13th with the 18th, the 14th with the 19^th and the 15th with the 20th. The digraphs are then used on the Table No. 2 (containing letters) and the letters found in the cells are recorded, regardless of repeats. The rotors are realigned to the letters thus recorded but they are not rearranged. The intermediate cipher text is enciphered preceded by the 11th to the 15th letters of the intermediate sequence and followed by the 16th to the 20th. These groups are enciphered in these positions to provide a check to the deciphering operator who, on recognizing them as part of the intermediate sequence, is assured that the first phase of decipherment has been accomplished correctly.

Cryptosecurity problems:

Despite the modifications to the rotor stepping system, the rotatable cam contours and the issuing of more rotors the CCM was only expected to remain secure for a few years and its security depended mainly on the laborious cipher procedures (key lists, simplex indicators, message bisection, padding, variable spacing) not on the underlying cryptosecurity of the device.

Compromise of the CCM by foreign powers

The CCM entered service with the Allied armies in 1943-44 and by that time the German codebreakers had neither the resources nor the time needed to mount a determined effort to analyze and solve this cipher machine. Instead, they focused their limited resources on the Allied systems that they could reliably exploit.

According to the available information the German Navy’s codebreaking department B-Dienst investigated the CCM in 1944. Traffic was first recognized in April ’44. During the summer a study of indicators was carried out. In October ’44 seven messages with identical first and second indicators were discovered. The system was thought to use non reciprocal substitution alphabets on plain text. Research ended in January ’45. No information is available on whether the other German codebreaking agencies (Inspectorate 7/VI, Luftwaffe Chi Stelle, OKW/Chi) evaluated the CCM traffic.

The use of the CCM in the postwar period meant that it became a target for the codebreakers of the Soviet Union. There is information pointing to the exploitation of the State Department’s MCB and of the Army’s CSP-1700/SIGROD (used by military attachés) in the 1950’s.  

The CCM must have been used by US and British forces during the Korean War of 1950-53. At this time there is no information on its use and possible compromise in Korea.

Timelines:

1940 - US Navy Department and British Admiralty agree to use Naval Cypher No3 as the British-American Naval Cypher (1), US ciphers not authorized to be shared with other countries (2)

1941 - Naval Cypher No3 distributed to ships in early 1941 and made effective on 16 June 1941. (3)

1942 – Talks between US and British officials regarding a combined cipher system, British book ciphers rejected by US, US side proposed M-138-A and M-209 and these were rejected by British. Typex production too low to satisfy need for a combined machine. SIGABA not released to the British by the US authorities for reasons of national security. (4)  

CCM attachments for SIGABA and Typex were developed mainly by the US Navy by Lt. Commander Donald W. Seiler USNR (5). By October ’42 a prototype adapter was demonstrated to the British (6) and the CCM was officially approved by the Anglo-American Combined Communications Board in October 1942 (7). Production started in December 1942. (8).

Three versions of the CCM were produced (9):

1). The CSP 1600, which was a modification of an existing SIGABA with the CCM attachment installed. The machine could be converted back into the standard SIGABA.

2). The CSP 1700, which was a SIGABA with a fixed CCM attachment. Thus, it could not be modified back into the standard SIGABA.

3). The CSP 1800, which was an adapter for the British Typex cipher machine. The version with the CCM attachment was called Typex Mk 23.

Production (10):

ECM adapter (CSP 1600) – 3.500 units

CCM (CSP 1700) – 631 units

Typex adapter (CSP 1800/Typex Mk23) – 4.500 units

Alternatively: (10a)

CSP 1600 - 6.500 

CSP 1700 - 2.000 

1943 – CCM Mk II (self-contained version) rushed into service with first deliveries in May ’43, Typex Mk 23 adapters first delivered in July ’43. (11)

By 1^st November 1943 enough machines had been distributed to Atlantic naval units to introduce the system in the North Atlantic (about 250 British and Canadian ships equipped with the cipher machine).  (12)

By 31 December ’43 the system was made effective for Limited Combined Naval Use for areas outside the North Atlantic.  (13)

By the end of 1943, 2,200 Typex adaptors and about 114,000 CCM rotors had been delivered to Britain and 1,525 ECM adaptors distributed to the US Navy in the Atlantic, while distribution in the Pacific was about to begin. In addition, about 500 CCM Mk II/CSP 1700 had been supplied to the British and the Canadians, and about 100 to the US Navy. (14).

Tests carried out during the spring of 1943 at the US Naval Computing Machine Laboratory in Dayton, Ohio revealed that, when using some rotors, its period was only 338.  (15).

1944 – In April ’44 the CCM became operational with all five British and American armed services.  (16)

By 31 December ’44 1.220 naval Typex machines had been modified into CCM versions.  (17)

A detailed study of the CCM’s period was not begun by OP-20-GM until 16 January 1944, six months after the short cycle defect was first discovered. The OP-20-GM investigation showed that, on one CCM circuit alone, seven daily keys between 12 February and 20 April 1944 produced the short 338 period. Once the problem of the short period had been fully identified, it could be avoided by taking various precautions, including the banning of certain rotor combinations and drastically restricting the length of messages. However, problems with lobstering rotors and the resulting short cycles were still being detected as late as March 1945. (18)

A security evaluation in July ’44 came to the conclusion that despite these problems the CCM was a well-designed cipher machine since it had close to a million possible wheel orders, was non reciprocal so cribs could not be placed with certainty and the irregular stepping of the end rotors countered the cyclometric stepping of the fast-moving rotor No3. (19)

New key lists and cipher wheel were issued in the period 1944-45.  (20)

The German Navy’s codebreaking department B-Dienst investigated the CCM in 1944. Traffic first recognized in April ’44. During the summer a study of indicators was carried out. In October ’44 seven messages with identical first and second indicators were discovered. The system was thought to use non reciprocal substitution alphabets on plain text. Research ended in January ’45. (21)

No information is available on whether the other German codebreaking agencies (Inspectorate 7/VI, Luftwaffe Chi Stelle, OKW/Chi) worked on the CCM.

1945 – From the middle of ’44 up to February ’45 the CCM was distributed to the British Pacific Fleet and equipped all major war vessels. (22)

By May ’45 the CCM was used extensively by the British Army, RAF and Navy. (23)

On 1 January 1945 a new indicator system was introduced. (24)

In February ’45 the CCM rotor basket (CSP 1600 – SIGAMUG) and basic rotors (CCBP 0111) are considered compromised (Colmar incident). The CCM material is lost when a truck containing the 28^th Infantry Division’s SIGABA cipher machine was stolen from Colmar, France. Traffic in the compromised systems was initially suspended. Eventually, this basic rotor set continued to be used by the fighting forces because there was no other alternative. The compromise of the CCM rotors revealed a serious deficiency in the emergency plan for the CCM since no provision had been made for the production and storing of reserve rotors. (25)

In March ’45 the crypto material lost in Colmar was discovered submerged in the Greesen River. (25)

From February to August ’45 17.840 new rotors were wired to replace the compromised set CCBP 0111. (25)

CCM rotors and keylists (25)

CCM rotors: initial worldwide rotors CCBP 0111, replacement CCBP 0211 (new worldwide), CCBP 0311 (reserve worldwide), CCBP 0113 (Pacific area), CCBP 0213 (reserve Pacific)

CCM keylists: CCBP 0101 (worldwide), CCBP 0131 (European area combined), 514-ESG (Special limited combined)

Postwar developments (AJAX, SIGROD, HERMES, LUCIFER, MCB):

The standard CCM system in US service is codenamed AJAX. The Army’s CSP 1700 is designated SIGROD. The SIGROD is distributed to military attachés. The HERMES modification is introduced in 1952. A CCM LUCIFER is mentioned in 1953. (26)

State Department

1946-1950’s - In early ’46 the State Department retired the M-325 SIGFOY cipher machine and replaced it with the CCM. The version used by the State Department was the CSP 2200 (HCM Mark 4/SCM Mark 2) (27). In State Department use the CCM was designated MCB (28). The MCB continued to be used till July 1959 (29).

US, UK, NATO

1946-1950’s – Discussions are held between US and British crypto experts on the need to modify the CCM in order to increase its security or design a new cipher machine to replace it. Concern over CCM security problems. Some interesting statements are made in these reports:

1947 – ‘In the opinion of the British, the present Combined Cipher Machine (CCM) is not sufficiently secure to resist increasingly effective cryptanalytic techniques. In addition, the British are anxious to develop a replacement for their present high level cipher machine (Typex)’. (30a)

 ‘Almost one year has gone by since our last conference with the British on cryptographic equipment for use in Combined Communications. At that conference the principal subject was the possible improvement of the CCM to increase its cryptographic security’. (30b)

‘The Army Security Agency has carefully studied the Navy proposal for modifying the CCM and finds the following objections thereto: a. Although the proposed changes would increase the cryptographic security of the machine, this increase would not be of sufficient degree to justify their adoption in view of the practical difficulties that would be encountered: ….’ (30b)

‘As the CCM machines now stand, we know that mechanically and electrically they are satisfactory and stand up well in operation; cryptographically we know where we stand, as a result of several years study of the present arrangements. But we have not had any lengthy period to study the new proposals for motion and reversed stepping; it is impossible to forecast what unforeseen cryptographic weaknesses might be uncovered by further intensive study’. (30b)

‘The ASA believes that while at this moment the incorporation of either the Navy's or the Army's proposed changes in the CCM appear to offer the possibility of currently improving the CCM cryptographically, this improvement does not even now appear to be quite sufficient to eliminate all questions as to what the security of the so-improved CCM might be say five years from now. We know now that a theoretical solution of CCM messages would still be possible and we also agree that granting that there 'were a sufficient desire to make what is theoretically possible a practical reality, by use of many people and analytic machines, daily solution of CCM keys could be achieved. Thus, nothing substantial or of long-term benefit would be gained by modification of the CCM along the lines of the Navy's or the Army's proposals, which involve mere changes in rotor motion, with or without reversed stepping’. (30b)

‘In view of the fact that the Army finds the Navy proposal for CCM modification impractical and that the Army is convinced that any proposals which would involve material physical changes in the CCM which cannot be made in the field are impractical and inadvisable, the Army believes that as an interim measure additional security in Combined Communications can more readily, more economically and more practically be achieved by more frequent changes in rotors’.  (30b)

1949 – ‘a. It is not in the best Interest of the US to agree to the full and complete interchange of cryptographic principles with the British. b. The release of the ECM under prevent circumstances is not warranted’, ‘Although the CCM, if properly used, is a highly secure machine, the experts of both nations agree that the cryptographic principles employed in the machine are not as secure as is considered desirable for highest level United States - United Kingdom communications’ (30c)

‘The British state that the (World War II version of the basic ECM) was disclosed to them in 1942. A limited number of Britishers have seen it. They do not however, have enough information to build a duplicate of our ECM. The British probably could construct a machine resembling our ECM’ (30d).

1950 - ‘The US, as a matter of policy, should not disclose to any foreign country the complete details of the ECM, even though technical considerations might favor the use of the ECM for combined communications’

‘As regards the effects on our own communication security, of the disclosure to the British of the detailed information referred to above, there probably would be no change in the security evaluation of the ECM. Although the US has always emphasized the importance of the physical security of our cipher machines, our evaluation of the ECM is predicated on possible ultimate enemy possession of the ECM machine, and our security is based upon regularly changing the rotors and key lists. The privacy of U.S. communications against cryptanalytic attack by any foreign power, including the British, could be assured’ (30e).

Development of KL-7 and KL-47 cipher machines to replace the CCM (31).  

Modification of CCM to HERMES standard (32).

Possible compromise of the MCB and the SIGROD by the Soviet Union (33).

1951 - Selection of CCM to serve as the NATO Second level communications cipher (Naval, Army and Air Force Commands, down to Divisional H.Q. plus NATO war vessels) in June 1951 (34).

1952 - The number of CCM immediately available to NATO are 943 machines (35). Modification of CCM to HERMES standard. Issuing of 20 rotors per machine and use of rotatable cam contours. Change of the rotor stepping mechanism. Changes made effective for NATO on 1 October 1952. (36)

1955 - Approximately 1.875 CCM machines have been distributed to NATO countries. NATO evaluation of the CCM is the following: ‘Both versions of the CCM have been in use for many years and require considerable expenditure for maintenance and spare parts. In addition, the security provided is largely dependent on operating procedures rather than on a strong cryptoprinciple’. Replacement of the CCM by the KL-7 ADONIS is authorized in June 1955 (37). Starting in 1955 the CCM is retired from US service (38).

1956 – NATO CCM to be replaced on 1 July 1956 by KL-7. A small number continued to be used for meteorological traffic and as a reserve (39).

1959 - The State Department retires the MCB worldwide in July 1959 (40).

1962 – NATO military committee orders the disposal of CCM equipment and associated materials (41): ‘As a result of exhaustive studies by the Standing Group Meteorological and Communications-Electronics Committees, it has been determined that the CCM equipment must now be regarded as totally unsuitable for the encryption of meteorological traffic. In fact, the CCM equipment is in such bad condition due to age, past usage, long term storage, lack of spare parts, and lack of adequately trained maintenance personnel, that it is of no further use’.

Sources:

(a). Operating instructions for ECM Mark 2 (CSP 888/889) and CCM Mark 1 (CSP 1600) - dated May 1944, NARA - RG 457 - Entry 9032 - ‘CCM 1943-45’ (NARA id: 2811488): OP-20-G reports: ‘Cross Tie in CCM from 1-wheel to 4-wheel’, ‘C.C.M., termination of researches on at Arlington Hall’, ‘Short Cycles ln CCM’, ‘The period of the CCM’.

(b). NATO Archives Online: S.G. 171/4 ‘Report by the Communications-Electronics coordination section to the Standing group on plan for the employment of the Combined Cipher Machine (CECS 15/6)’ and S.G. 171/25 ‘Report by the Communications-Electronics coordination section to the Standing group on Provision of an off-line literal cipher machine for general NATO communications (CECS 97)’

(1). UK National archives: ADM 1/27186 ‘Review of security of naval codes and cyphers 1939-1945’, p40-41

(2). NSA’s Friedman collection: ‘Directive to G-2 Covering Interchange of Secret Technical Information with Representatives of British Government’ - A272416 and ‘Background of Policy in Regard to Making SIGABA Available to the British’ - A2435719

(3). UK National archives: ADM 1/27186 ‘Review of security of naval codes and cyphers 1939-1945’, p41

(4). NSA’s Friedman collection: ‘Background of Policy in Regard to Making SIGABA Available to the British’ - A2435719, ‘Establishment of Joint Cryptographic Systems for the United States and Great Britain’ - A2435737, ‘Conference on Joint U.S. and British Cryptographic Systems’ - A2435740

(5). NSA’s Friedman collection: ‘History of Invention and Development of the Mark II ECM-Op-2-- S-5’, p8 - A273704 and ‘U.S. Naval Code and Signal Laboratory-History of’ - A40422

(6). UK National archives: ADM 1/27186 ‘Review of security of naval codes and cyphers 1939-1945’, p37

(7). Journal of Intelligence History: ‘The Admiralty and Cipher Machines During the Second World War: Not So Stupid after All’, p6

(8). Journal of Intelligence History: ‘The Admiralty and Cipher Machines During the Second World War: Not So Stupid after All’, p7

(9). NSA’s Friedman collection: ‘History of Invention and Development of the Mark II ECM-Op-2-- S-5’- A273704 and Center for Cryptologic History: The SIGABA/ECM II Cipher Machine: "A Beautiful Idea"

(10). NSA’s Friedman collection: ‘History of Invention and Development of the Mark II ECM-Op-2-- S-5’, p8 - A273704

(10a). NSA’s Friedman collection: ‘U.S. Naval Code and Signal Laboratory-History of’, p5 - A40422

(11). UK National archives: ADM 1/27186 ‘Review of security of naval codes and cyphers 1939-1945’, p37

(12). UK National archives: ADM 1/27186 ‘Review of security of naval codes and cyphers 1939-1945’, p37

(13). UK National archives: ADM 1/27186 ‘Review of security of naval codes and cyphers 1939-1945’, p38

(14). Journal of Intelligence History: ‘The Admiralty and Cipher Machines During the Second World War: Not So Stupid after All’, p7

(15). Journal of Intelligence History: ‘The Admiralty and Cipher Machines During the Second World War: Not So Stupid after All’, p7

(16). UK National archives: ADM 1/27186 ‘Review of security of naval codes and cyphers 1939-1945’, p38

(17). UK National archives: ADM 1/27186 ‘Review of security of naval codes and cyphers 1939-1945’, p38

(18). Journal of Intelligence History: ‘The Admiralty and Cipher Machines During the Second World War: Not So Stupid after All’, p7-8

(19). OP-20-G report: ‘C.C.M., termination of researches on at Arlington Hall’

(20). UK National archives: ADM 1/27186 ‘Review of security of naval codes and cyphers 1939-1945’, p38

(21). NSA’s Friedman collection: ‘German Naval Communications Intelligence’ Volume If-210-Copy #3, Chapter IV, Pp 135-161 ‘, p8 - A66767

(22). UK National archives: ADM 1/27186 ‘Review of security of naval codes and cyphers 1939-1945’, p39

(23). UK National archives: WO 193/211 ‘Wireless, cable and signal (including cipher) communications: policy and codes: action from report of Godwin-Austen Committee’: Typex and CCM distribution list for operation NESTEGG (1945) (the CCM systems are in columns 5 and 6)

(24). UK National archives: ADM 1/27186 ‘Review of security of naval codes and cyphers 1939-1945’, p38

(25). NSA’s Friedman collection: ‘History of Converter M-134-C Volume 3’ - A523210, chapter XVI ‘The Colmar compromise’.

(26). AJAX: ‘A History of U.S. Communications Security Post World-War II’, p92, SIGROD: Center for Cryptologic History: The SIGABA/ECM II Cipher Machine: "A Beautiful Idea", p26 and p41, HERMES: NATO Archives Online: S.G. 171/4 ‘Report by the Communications-Electronics coordination section to the Standing group on plan for the employment of the Combined Cipher Machine (CECS 15/6)’, LUCIFER: NSA’s Friedman collection: ‘UK/US Communications Security Conference 1953 Report of the Security Sub-Committee to the Executive Committee’ - A522921

(27). San Francisco Maritime National Park Association: CSP and Other Crypto Designators, ‘History of the Signal Security Agency’ - Volume 1-Organization, p113, NSA’s Friedman collection: ‘State Department Use of Cipher Machine Electrical Stepping’ - A272413, ‘Establishment of Joint Crypto Centers’ - A67035,  ‘U.S. Naval Code and Signal Laboratory-History of’ - A40422

(28). Intelligence and National Security: ‘State Department cipher machines and communications security in the early Cold War, 1944–1965’, p3

(29). Intelligence and National Security: ‘State Department cipher machines and communications security in the early Cold War, 1944–1965’, p4

(30a). NSA’s Friedman collection: ‘U.S. and British Collaboration on Combined Cipher Machine Development’ - A2435927

(30b). NSA’s Friedman collection: ‘Army/Navy Conference on CCM Modification’ - A2435934

(30c). NSA’s Friedman collection: ‘Replacement of the Present Combined Cipher Machine’ - A2436098

(30d) ‘Replacement of the "Combined Cipher Machine (CCM)’ - A2436048

(30e). NSA’s Friedman collection: ‘Replacement of the Present Combined Cipher Machine (CCM)’ - A2435986

(31). ‘History of the TSEC/KL-7 ADONIS & POLLUX’, p9 by Dirk Rijmenants

(32). NATO Archives Online: S.G. 171/4 ‘Report by the Communications-Electronics coordination section to the Standing group on plan for the employment of the Combined Cipher Machine (CECS 15/6)’ and S.G. 171/25 ‘Report by the Communications-Electronics coordination section to the Standing group on Provision of an off-line literal cipher machine for general NATO communications (CECS 97)’

(33). Intelligence and National Security: ‘State Department cipher machines and communications security in the early Cold War, 1944–1965’, p5-8 and ‘Spymaster: Startling Cold War Revelations of a Soviet KGB Chief’ by Tennent H. Bagley

(34). NATO Archives Online: S.G. 7/40 ‘Standing group decision on S.G. 7/40 – A report by the Communications-Electronics coordination Section on Provision of a general high grade cyher system for North Atlantic treaty organization powers’.

(35). NATO Archives Online: S.G. 171/3 ‘Report by the Communications-Electronics coordination section to the Standing group on the provision of secure means for second level communications within NATO forces (CECS 15/4)

(36). NATO Archives Online: S.G. 171/4 ‘Report by the Communications-Electronics coordination section to the Standing group on plan for the employment of the Combined Cipher Machine (CECS 15/6)’

(37). NATO Archives Online: S.G. 171/25 ‘Report by the Communications-Electronics coordination section to the Standing group on Provision of an off-line literal cipher machine for general NATO communications (CECS 97)’

(38). ‘A History of U.S. Communications Security Post World-War II’, p92

(39). NATO Archives Online: SGWM-809-56 ‘Memorandum for the Standing group meteorological committee - Subject: Cryptographic Equipment for Meteorological Use’

(40). Intelligence and National Security: ‘State Department cipher machines and communications security in the early Cold War, 1944–1965’, p4

(41). NATO Archives Online: SGM-279-62 ‘North Atlantic military committee standing group – Subject: ‘Disposal of the Combined Cipher Machine (CCM) and associated materials’.

Pics: CSP 1600 and 1700 pics from Crypto museum, Typex Mk 23 pic from National Cryptologic Museum. Reprinted with permission.

Additional information: Frode Weierud has also presented information on the CCM at his site Crypto Cellar Research.

Acknowledgements: It was Ralph Erskine (a legend of cryptologic history) who, years ago, sent me files on the CCM. Without those reports I wouldn’t have thought to research the CCM further. I also have to thank Frode Weierud for his insightful comments.