Free hosting http://www.fortunecity.com shut down at April 30th, 2012.
Here you can read the orignial three ENIGMA M4 messages published by Mr. Ralph Erskine:
LETTERS TO THE EDITOR
6 December 1995
This letter originally appeared in Cryptologia October 1996;Volume XX, Number 4.
In "Cipher-text Only Cryptanalysis of ENIGMA" (Cryptologia. 19(4) (1995): 405-423), James Gillogly stated that "A rough count of English translations of Enigma solutions from the Bletchley Park cryptanalysts for two arbitrarily selected days in 1943 shows means of about 340 letters (107 messages on 6 Aug 43) and 550 letters (89 messages on 13 Nov 43). On both days several translations of over 1000 letters were forwarded from Bletchley Park: four on 6 Aug 43 and ten on 13 Nov 43. The longest were about 1500 and 2300 letters respectively" [pp. 405, 406].
James Gillogly refers only in general terms to the microfilm version of the Enigma messages in class DEFE 3 at the Public Record Office (PRO), London, without giving specific file references. However, it is important to appreciate that the DEFE 3 files contain two completely different types of message.
There are virtually no decrypts of German Army or Air Force signals on the DEFE 3 files. Bletchley's Hut 3 almost invariably summarised the decrypts and sent them, as "CX/MSS" signals to the various commands in the British Army and Royal Air Force authorised to receive Ultra. The CX/MSS label was intended to disguise the signals as containing intelligence obtained by agents. Only the Admiralty received the verbatim text of decrypts. Over 500,000 naval Enigma decrypts are on DEFE 3.
In wartime, the maximum length of a naval Enigma signal was supposed to be 80 groups (320 letters) (para. 36 of Der Schlüssel M - Verfahren M Allegemein (M. Dv. Nr. 32/1)). For German Army or Air Force traffic, the limit was 250 letters (para. 3 of Schlüsselanteitung zur Schhüsselmaschine Enigma (H. Dv. 14 - edition dated 13 January 1940)), although in practice some signals seem to have been up to 280 letters long.
Unfortunately, James Gillogly does not say whether the "translations" to which he refers are of naval signals or not. If they are not, then they are most unlikely to be evidence that the Germans transmitted Enigma signals breaking the 250 or 320 letter barrier. The CX/MSS messages from Hut 3 amalgamated the text of German signals which had been broken down into parts for Enigma purposes. A 1,000 letter Hut 3 signal, for example, probably therefore merely represented a German signal split into parts - or a collection of different signals bearing on the subject in question.
It would be surprising if the 250 (or 320, for the Kriegsmarine) letter limit was often broken. Each Enigma signal had to state how many groups it contained. The count was given in plain language and therefore easily monitored. Serious infractions, such as the figure of 550 letters mentioned in the article, would almost certainly haw been pounced upon and culprits reprimanded or punished.
It is only fair to add that some "raw" (i.e. not decrypted) naval Enigma intercepts give specific counts of up to 89 groups for some messages from the U-boats, possibly because splitting the signals into two would have increased the danger from high-frequency direction-finding (two separate signals would have been taken about 15 seconds longer to transmit than a combined signal). However, long naval Enigma signals are generally divided into two or more parts, which suggests that para. 36 of M. Dv. Nr. 32/1 was generally being followed, as one would expect.
If James Gillogly has indeed found clear evidence of "long" naval Enigma signals (especially those containing as many as 1,000 (or even 500) letters, the Kriegsmarine was repeatedly guilty of serious breaches of cipher discipline, which would be surprising. But if he was relying instead on Hut 3's summaries of German signals, there would appear to be nothing to indicate that any of the German forces broke the relevant prohibition in the various cipher regulations as often, or as seriously, as his article suggests.
Since the point is of considerable historical importance, it would therefore be interesting to know on which specific files in the DEFE 3 series the 1943 signals to which he refers are to be found, and whether their contents relate to the German navy, army, or air force. Moreover, applying his Figure 3, his ingenious cryptanalytical method would have had little or no chance of success against wartime Enigma signals containing a maximum of 320 letters, if at least eight plugboard connections were used (10 were employed in wartime).
James Gillogly's attack involves the Index of Coincidence. Apparently the cryptanalysts at Bletchley Park were unaware of William F. Friedman's pioneering paper on the subject in the Riverside Publications'scries. Alan Turing, while at Bletchley, seems to have discovered the technique underlying the Index independently, but called it "the repeat rate": I. J. Good, "Studies in the History of Probability and Statistics. XXXVII. A. M. Turing's Statistical Work in World War II," Biometrika, 66 (1979), 393.
I enclose a copy of intercepts of what were almost certainly naval Enigma signals. As usual in practice, there are some garbles in the intercepts. There are two obvious ones in the first two signals, where the first two, and the last two, groups should be identical. Those groups are, of course, indicators, and do not contain message text.
You may wish to publish them as a challenge in Cryptologia , so that people may try to succeed where Bletchley Park failed! These signals were intercepted in the North Atlantic during the Triton black-out, and were probably enciphered on four rotor Enigma (M4). Any one trying to break them would therefore need to take into account the wiring of M4, which is given in "Naval Enigma: M4 and its Rotors" (Cryptologia 11(4) (1987), 235).
APPENDIX II (continued), TO THE COMMANDING OFFICER, H.M.S. HURRICANE'S
No. 110/S, DATED 25TH NOVEMBER, 1942.
TEXTS OF SIGNALS READ
CLXP LWRU HCEY ZTCS OPUP PZDI UQRD LWXX FACT TJMB HDVC JJMM ZRPY IKHZ
AWGL YXWT MJPQ UEFS ZBCT VRLA LZXW VXTS LFFF AUDQ FBWR RYAP SBOW JMKL
DUYU PFUQ DOWV HAHC DWAU ARSW TXCF VOYF PUFH VZFD GGPO OVGR MBPX XZCA
NKMO NFHX PCKH JZBU MXJW XKAU OD?Z UCVC XPFT CDXP LWRU VA
T.O.R.1152/19/221 (53 GROUPS).
FCLC QRKN NCZW VUSX PNYM INHZ XMQX SFWX WLKJ AHSH NMCO CCAK UQPM KCSM
HKSE INJU SBLK IOSX CKUB HMLL XCSJ USRR DVKO HULX WCCB GVLI YXEO AHXR
HKKF VDRE WEZL XOBA FGYU JQUK GRTV UKAM EURB VEKS UHHV OYHA BCJW MAKL
FKLM YFVN RIZR VVRT KOFD ANJM OLBG FFLE OPRG TFLV RHOW OPBE KVWM UQFM
PWPA RMFH AGKX IIBG FCLC QRKM VA
T.O.R.1840/19Z (62 GROUPS).
YKRB MGVA TMKF NWZX FFII YXUT IHWM DHXI FZEQ VKDV MQSW BQND YOZF TIWM
JHXH YRPA CZUG RREM VPAN WXGT KTHN RLVH KZPG MNMV SECV CKHO INPL HHPV
PXKM BHOK CCPD PEVX VVHO ZZQB IYIE OUSE ZNHJ KWHY DAGT XDJD JKJP KCSD
SUZT QCXJ DVLP AMGQ KKSH PHVK SVPC BUWZ FIZP FUUP YKRB MGVA VA
T.O.R.0321/21Z (54 GROUPS).
0557Z Corrections to above W 51 FIZP W 54 MGVA.