National Academy of Sciences (NAS)

Report of the Committee on
Ballistic Acoustics






Washington, DC 1982




Two of the most conspicuous features of the Channel I recording are the complete absence of siren sounds for the first two minutes following the BRSW/WA conjectured shots and the clear presence of such siren sounds for the next 36 seconds. Several sirens are heard in succession each in turn rising and falling in intensity as would be the case if a motorcade were rapidly passing an open microphone. The siren sounds provide critical tests of both the BRSW/WA scenario and that of the Committee.

The absence for two minutes of siren sounds. at a time when they should be heard presents a serious difficulty for the BRSW/WA hypotheses. According to that scenario, the motorcycle with the open microphone was located in a precisely known position behind the President's car in the motorcade as it passed through Dealey Plaza when the President was assassinated. Many witnesses agree that sirens were activated shortly after the final shot and as the motorcade speeded up for its dash to Parkland Hospital. The complete absence of siren sounds for two minutes is difficult to explain on this scenario, and the sounds, when they do appear do not seem appropriate for a motorcycle in the motorcade, or even one catching up to the motorcade. If Officer McLain had the open microphone, it is particularly surprising that he picked up no siren sounds while accompanying the motorcade to the hospital but, at the same time, his microphone was so sensitive that it could pick up the Channel II cross talk from a nearby vehicle.

The absence of siren sounds for two minutes is fully compatible with the Committee's scenario, which does not require the open microphone to have been in the procession. James Bowles' hypothesis that the motorcycle was at the Trade Mart can be supported by reasonable arguments but there no firm evidence for that location. Although the two-minute absence of the police sirens is obviously compatible with the Committee's scenario, the timing of the appearance of the sirens requires careful examination. The cross correlation between the "hold everything . . ." phrases on Channels I and II provides a relative timing of events that can be tested for reasonableness with respect to the siren sounds. Absolute times were obtained by running Channels I and II from the original Dictabelt and Audograph disk, with the speeds adjusted to provide the correct frequency for the 60 Hz hum on the original recordings. In this operation it was found that "hold everything . . ." on Channel I begins 123 seconds before the siren sounds and on Channel II there is 64 seconds of continuous recording between ". . . Go to the hospital . . ." and "hold everything . . ." which gives 181 seconds between ". . . Go to the hospital . . ." and the beginning of the sirens. Since the distance from the assassination site to the Trade Mart is 2.273 miles, this corresponds to an average speed of 43.8 miles per hour if the trip began at the time of "Go to the hospital." At first consideration this appears to be surprisingly slow for a trip to the hospital, but there were turns, traffic, a heavy car, Mrs. Kennedy and a Secret Service Agent crawling over the back of the car, and a critically wounded passenger to slow the average speed. The speed we estimate is compatible with the testimony of Agent Greer, the driver of the President's car, in volume II, page 121 of the Warren Commission Report(3): ". . . I was getting through traffic and through streets as fast as I could get through . . . I would estimate that I must have been doing between 40 and 50 at least 50 miles per hour at times. We might have been going as fast as 50 miles an hour I am sure. . . ."

If one assumes that on the streets and access ramps the average speed is 40 miles per hour, that the average acceleration in a turn is 0.2g, and that the Zapruder film gives the time to leave Dealey Plaza, the above 181 seconds would require an average speed on the Stemmons Freeway of 58.5 miles per hour, which seems reasonable in view of Agent Greer's testimony. It should be noted, however, that there is considerable uncertainty as to the speeds attained, the location of the open microphone, and the time following the assassination at which the "Go to the hospital" was broadcast. Although this discussion shows the compatibility of the ". . . hold everything . . ." identification with the known firm data, it should not be misinterpreted as a proof of this interpretation or as a reliable determination of the location of the vehicle with the open microphone, since there is considerable uncertainty as to speed of the vehicle. There is also contradictory evidence about the time interval between the assassination and the "Go to the hospital . . ." As this time is lengthened the average speed is reduced. However, it should be noted that the assumption of a long time interval makes more acute the difficulty with the BRSW/WA scenario discussed in Appendix C.




This Appendix is written in response to the Committee's assignment to recommend the kinds of tests, analyses, and evaluations needed to obtain better information from the recordings. However, the existence of this Appendix should not be misinterpreted as a Committee recommendation that these tests and analyses should be carried out. If there were to be further studies of the Dallas Police Department Channel I recording in the hope of demonstrating the validity of the conjectured shot from the grassy knoll, the information listed below could be sought.

1. The original Dictabelt could be studied more extensively for possible evidence either for or against the possibilities of the Dictabelt being a copy or containing superposed recordings. No evidence favoring either of these possibilities has so far been found in a physical examination of the belt or in studies of the recording. Further studies could include a careful search on the original belt for a second hum at about 60 Hz which would characterize a copy and an examination of the 60 Hz signal for continuity and possible indications of interceptions. Such studies, however, will now be difficult and may require the construction of a special drum playback machine for the shrunken and stiffened Dictabelt which now causes marked flutter when it is played back on the normal machine.

2. With the information on the timing of the Channel I recording provided by the cross correlation between Channels I and II discussed in Section IV, the Channel I recording could be examined more carefully for the existence of possible shots in the portion of the recording that corresponded to the time of the assassination (between 65 and 95 seconds on the BRSW time scale). However, it is unlikely that evidence for shots will be found in that region, since the noise level was much higher there and that portion of the recording has already been examined by BRSW, as described on page 35 of the BRSW report, and no impulse patterns identifiable as gunfire were found; furthermore, there is some evidence that the open microphone was not in the motorcade.

3. There could be an independent analysis following generally the WA procedure but applied to all four of the conjectured shots. In the case of the conjectured grassy knoll shot, it would be of interest to see if the P value for the hypothesis of random locations of impulses cast doubt on that hypothesis. Analysis by the WA method of the impulses attributed to the three Texas School Book Depository shots would be a test both of the method and of this attribution, which contradicts the evidence that the relevant impulses occur approximately one minute after the assassination. If these impulses do fit the hypothesis of three shots, is the open microphone trajectory the same as in the BRSW studies and does it fit with the best limits that can be photographically inferred?

4. The BRSW analysis of the three shots attributed to the Texas School Book Depository could be repeated with a well defined, normalized, and objective selection process for the impulses and echoes to see if the indications of three shots associated with a reasonable microphone trajectory persisted when the unnormalized and subjective selection of impulses and echoes was eliminated.

5. Attempts could be made to see if the reliability of the analyses could be improved by utilizing the availability of amplitude information even though it is recognized that amplitude information can sometimes be misleading. Acoustic spectra and logarithmic Fourier transform studies might help. Unfortunately, one cannot deal with the Dictabelt recording as a faithful reproduction of the sound pressure at the microphone due to the distortion of the radio and recording systems, which include automatic gain control, so it will be difficult to untangle the distortion effect in retrospect.

6. Independent analyses could be made of the probabilistic calculations both by BRSW/WA and by the present committee, with a critical review of the hypotheses on which the calculations are based. The studies could include the investigation of alternative hypotheses such as other sources of non-random impulse locations and studies of prior and posterior probabilities.

7. A study could investigate means for confirming that the open microphone was actually in Dealey Plaza. This study could examine the recording for the presence and absence of sounds of crowds, the lateness of siren sounds, the possibility of detecting a Doppler shift in the siren sounds, study of the motorcycle sounds to determine if they indicated speeds compatible with the course of the motorcycle presumed by BRSW/WA to have the open microphone, identification of the kind of motorcycle from its sound, cepstral analysis, AGC effects, etc. Bowles (2) reports that Officer McLain, after hearing recordings of Channels I and II stated that there was "no way" that his mike could have been the one stuck open. As the present report was about to be printed, Officer Leslie Beilharz (who was not in Dealey Plaza at the time of the assassination) told the Committee chairman by telephone that there was a "good possibility" that his microphone may have been the one stuck open. Additional testimony could be obtained as to the location of the open microphone and attention should be given to the many questions raised in the report of James C. Bowles, including those on the microphone location.

8. A detailed analysis could be made of the interpretation of the more than 200-millisecond time displacement between the conjectured shots of the BRSW and the WA studies.

9. The Zapruder film could be analyzed further to see if the apparent incompatibility between the conjectured shots and the data inferred from the camera's angular accelerations can be removed.

10. Clear objective procedures for the selection of the impulses should be used, and they should be well described. The publication of the results should be sufficiently clear and in sufficient detail to permit an effective outside scientific appraisal.

11. The sources of several extraneous sounds on the Channel I tape could be investigated further to see if their identification could determine the location of the microphone. One such sound is the bell-like tone that occurs about 8 seconds after the conjectured grassy knoll shot. Another is the phrase differently identified by various listeners as "All right Chaney," "All right Jackson" or ''I'll check it" that occurs approximately 10 seconds before the impulses conjectured by BRSW/WA to be due to a grassy knoll shot.

12. An effort could be made to localize the motorcycle accurately to test with the narrowest possible limits the compatibility between the photographic and acoustic evidence on motorcycle location.

13. The consistency of the 60 Hz hum on Channel I with the voice time reports could be tested, provided Channel I is shown to be recorded without interruptions for a sufficiently long time that the implications are not importantly affected by the large possible errors in the time reports, which are based on several different and unsynchronized time indicators. Although the Committee has made no effort to determine the cost of the above investigations, it would be considerable. As discussed in Sections III, IV and VII, the Committee considers the evidence against the BRSW/WA conjectured assassination shot already to be so strong that the expected results of such investigations would not justify their cost.


Back to the top


Table of Contents


Back to archive of acoustical evidence

Back to JFK reports and documents menu

Back to JFK menu

Dave Reitzes home page



1. Appendix to Hearings Before the Select Committee on Assassinations of the House of Representatives Ninety-Fifth Congress, Volume VIII, US Government Printing Office, Washington, DC, 1979.

2. James C. Bowles, The Kennedy Assassination Tapes: A Rebuttal to the Acoustical Evidence Theory (copyrighted and unpublished).

3. 3. Hearings Before the President's Commission on the Assassination of President Kennedy, US Government Printing Office, Washington, DC, 1964.

4. Report released December 1, 1980, by the Federal Bureau of Investigation and prepared by the FBI Technical Services Division, Washington, DC, and dated November 19, 1980.

5. Minitab Manual, by Thomas A. Ryan, Jr., Brian J. Jainer, and Barbara F. Ryan, published by Minitab Project, Statistics Department, 215 Pond Laboratory, Pennsylvania State University, University Park, PA 16802.


Back to the top


Table of Contents


Search this site
    powered by FreeFind

Back to archive of acoustical evidence

Back to JFK reports and documents menu

Back to JFK menu

Dave Reitzes home page