The FBI Laboratory: An Investigation into Laboratory Practices and Alleged Misconduct in Explosives-Related and Other Cases (April,1997) Office of the Inspector General, Department of Justice

PART THREE: ANALYSIS OF PARTICULAR MATTERS

Since Whitehurst joined the Laboratory in 1986, he has repeatedly complained about SSA Terry Rudolph, who preceded Whitehurst as the Laboratory's senior examiner of explosives residue. Whitehurst alleges that Rudolph was incompetent and that the Laboratory sought to ignore or cover up his deficiencies. In this section, we address allegations that Whitehurst and others have made concerning Rudolph, and we evaluate the Laboratory's actions in response to those allegations.

Terry Rudolph worked as an explosives residue examiner in the Laboratory from 1979 until 1988, when he began teaching at the FBI Academy in Quantico, Virginia. After Whitehurst joined the Laboratory in 1986, he worked with Rudolph to become qualified to examine explosives residue. Whitehurst soon began complaining to his unit chiefs that Rudolph was sloppy in that he maintained a messy work area and performed inadequate examinations.

In 1989, Whitehurst voiced his concerns about Rudolph for the first time outside the Laboratory. During the trial in United States v. Psinakis, Rudolph was expected to testify about his identification of the explosive PETN on certain evidence. After the prosecutor learned the defense intended to challenge Rudolph's analyses, Whitehurst was asked to re-examine the evidence. Whitehurst also found PETN in his examinations, and he attended the trial prepared to testify. Without first raising his concerns with the prosecutor or Rudolph, Whitehurst approached a defense expert and said he thought the FBI's identification of PETN may have resulted from contamination of the evidence due to Rudolph's sloppy work habits.

Whitehurst ultimately did not testify at the trial. In Part Four of this Report, we discuss our evaluation of his conduct and his claim that the FBI improperly retaliated against him by suspending him for one week for his actions.

Rudolph did testify in Psinakis. At the end of the trial, the jury acquitted the defendant. In July 1989, the prosecutor, Assistant United States Attorney (AUSA) Charles Ben Burch, wrote to the FBI complaining that Rudolph's analysis was deficient, that the judge had nearly excluded his testimony, and that Rudolph had been seriously impeached by the defense.

In August 1989, the Laboratory completed two internal reviews of Rudolph's casework. MAU Chief Jerry Butler reviewed 200 cases, found numerous administrative shortcomings, and recommended a further in-depth review. CTU Chief Roger Martz reviewed 95 case files, reported that Rudolph's analyses supported the results and that Martz found no technical errors, and recommended there be no further technical review of Rudolph's cases. The Laboratory concluded that no further action concerning Rudolph was necessary.

In 1991, the FBI OPR opened an investigation concerning Rudolph after Whitehurst complained not only about his sloppy work but also that Rudolph had perjured himself, lied to an AUSA, and abused annual leave, and that Rudolph and his technician Edward Bender were racists. As a result of Whitehurst's allegations, the Laboratory also initiated a third review of Rudolph's case files, this one by MAU Chief James Corby.

After reviewing 200 cases, Corby reported that he found 57 lacking adequate documentation or information to support the stated conclusions. CTU examiner Lynn Lasswell also reviewed the 57 cases identified by Corby. In April 1992, SAS Chief Kenneth Nimmich advised Laboratory Director John Hicks that Rudolph would be asked to review the 57 cases and, if possible, reconstruct from his personal recollection, diaries, or other personal notes sufficient documentation for the findings reported. Nimmich stated that a memorandum should be prepared for each file describing any additional information.

Nimmich also recommended in April 1992 that Rudolph be severely reprimanded for his lack of professionalism and inattention to detail. Instead, Hicks admonished Rudolph orally at a meeting in which Hicks also gave Rudolph a cash incentive award. In June 1992, the FBI advised Rudolph that the FBI OPR inquiry had not developed facts warranting administrative action. In March 1993, Nimmich reported to Hicks that Rudolph had reconstructed the 57 files and that Nimmich recommended the matter be closed.

Within the Laboratory, MAU Chief Corby advocated a further review of Rudolph's case work. In May 1994, after investigating Whitehurst's allegations on several matters, the OGC recommended that Corby review all of Rudolph's cases. After reviewing 654 of Rudolph's cases, Corby reported in November 1995 that 24% contained errors or were administratively or technically incomplete. Rudolph disputed these findings. He retired from the FBI in June 1996.

To investigate the Rudolph matter, we conducted sworn and transcribed interviews of Edward Bender, Steven Burmeister, Charles Calfee, James Corby, Terry Rudolph, Roger Martz, Kenneth Nimmich, and Frederic Whitehurst. We also interviewed other witnesses, including Milton Ahlerich, Roger Asbury, Ben Burch, John Dietz, Frank Doyle, John Hicks, James Kearney, Lynn Lasswell, Randy Murch, Robert O'Brien, Ralph Regalbuto, Steven Robinson, John Sylvester, and Don Thompson. We reviewed all available documents produced by the FBI pertaining to the Psinakis case, the reviews of Rudolph's case files, and the relevant FBI OPR investigations.

Based on our investigation, we conclude that, in Psinakis and in numerous other cases, Rudolph did not competently or professionally perform his work as an examiner. As is discussed infra in Part Three, Section H9, we also note similar problems in certain work that Rudolph did in the UNABOM case. We further conclude that the Laboratory did not adequately investigate or resolve the concerns about Rudolph after the Psinakis prosecutor's July 1989 letter, after Butler's 1989 review, or after Corby's 1992 review. We recommend that a notation concerning this Report's findings be included in each of Rudolph's case files. We further recommend that the FBI not employ Rudolph in any capacity in the future.

II. The Psinakis Case

This case involved an American citizen suspected of smuggling explosives to the Philippines. During the investigation, a large quantity of detonating cord that had been stripped, or cut along the side so the explosive inside could be removed, was found in the suspect's garbage. At the Laboratory, Rudolph examined a white powder extracted from the cord and determined, through the use of x-ray powder diffraction (XRD), that it was PETN, an explosive commonly found in detonating cord. The suspect's home was then searched, and FBI agents found tools that were submitted for examination to ascertain if they had been used to strip the cord and extract the powder.

In January 1982, Rudolph conducted a liquid chromatography test on white powder removed from the tools and concluded it was PETN. He issued AE dictation stating that PETN was found on the tools, including pliers and a utility knife. A Laboratory report dated February 18, 1982, similarly stated that the identified tools had been instrumentally examined and determined to contain PETN. Neither Rudolph's dictation nor the final report identified the instrumental analyses performed. This report was given to the prosecutor and turned over to the defense.

In 1989, Rudolph was called to testify at the Psinakis trial. AUSA Burch recalls that Rudolph assured him that the examinations were sufficient, conclusive, and could easily be used at trial. Burch learned through discovery that the defense was prepared to offer expert witnesses to challenge Rudolph's conclusions. As a result, Burch retrieved certain evidence from the court's custody and sent it back to the FBI Laboratory for additional testing. By this time, Rudolph was no longer working in the Laboratory as an examiner. Whitehurst conducted the tests and confirmed the presence of PETN by the use of gas chromatography/mass spectrometry. He then went to San Francisco so he could testify about his results if needed.

In late May 1989, the court in Psinakis held an evidentiary hearing on the admissibility of expert testimony. While waiting to testify, Whitehurst approached a defense expert and told him he had concerns about the reliability of Rudolph's conclusions. Whitehurst told the defense expert that Rudolph's work area had always been very dirty and possibly contaminated, and he suggested that this sloppy behavior could have been the source of the explosive found on the evidence. Whitehurst did not tell the prosecutor or Rudolph about his misgivings before speaking with the defense.

At the May 1989 hearing, the court did not allow Whitehurst to testify about his own test results because the court concluded the government had improperly removed the evidence from the court's custody for further testing. Burch mistakenly thought the court clerk had the judge's approval to release the evidence. Defense counsel, however, asked the court to have Whitehurst held on call because someone at the FBI had doubts about Rudolph's testimony. Whitehurst ultimately did not testify.

Rudolph testified at the evidentiary hearing about his identification of PETN on the tools. He acknowledged that the only instrumental technique he had used was liquid chromatography (LC). He agreed with statements in a treatise that LC most often provides only a tentative confirmation and that a final confirmation requires use of an ancillary method such as mass spectrometry and infrared spectroscopy. Rudolph admitted he had not used any confirmatory techniques in addition to the LC test.

To counter the defense argument that LC alone was insufficient to identify PETN, Burch elicited from Rudolph that his opinion rested on many other factors. Rudolph noted that the cord found in the garbage was found to contain PETN; that the tools were of the type used to strip detonating cord; that he had examined microscopically a known sample of PETN and white powder from the tools and they compared essentially identically ; that he tested the powder with diphenylamine and it gave a blue color in just seconds which is another factor that tells me that I'm dealing with PETN ; that the sample was immediately soluble in a mixture of acetonitrile and water, again, a factor, an indicator that we are dealing with the same material [PETN] ; and, additionally, I took into consideration that in the time that I had done these analysis I have never ever encountered another explosive that interfered with the analysis of PETN on this [liquid chromatography] system.

The court asked Rudolph why the diphenylamine test and other tests he described were not documented in his notes. Rudolph responded, When I examine a case I put in my notes things that are important to me when I . . . give testimony. I don't write my notes for the United States Attorney. I don't write my notes for the defense. I write my notes for myself. Rudolph said he had done thousands of tests since 1982 and could not possibly remember them all. The court asked, Isn't that one of the reasons you keep notes? Rather than respond directly, Rudolph said this case was different because he and his technician remembered it specifically. Rudolph also said he often used this case as an example in teaching classes.

On further examination by the defense, Rudolph was asked the following:

A: Yes.

A: Yes.

Rudolph also stated the following:

Rudolph admitted that liquid chromatography was not used to identify total unknowns. He said that as a chemist he had learned to do things in an expedient way, but yet still efficient. As an example, he said he would confirm the identity of PETN in blasting caps by liquid chromatography, because I could do a liquid chromatography analysis in a few minutes while it would take 45 minutes to do x-ray powder diffraction. Regarding the evidence in Psinakis, Rudolph stated that there is absolutely no doubt that that material was PETN, absolutely none. I felt as strong about that identification that that material on those blades were PETN as I have in any analysis I have ever done.

After the evidentiary hearing, the defense urged the court to exclude the evidence because all the witnesses, including Rudolph, agreed that LC was not an adequate test to identify PETN. Without directly countering this argument, AUSA Burch noted that the FBI had recently tested the powder from the utility knife by the use of mass spectrometry and another test and determined it was PETN. Burch also argued that Rudolph based his opinion on information in addition to the LC test. Burch stated that Rudolph doesn't purport to be somebody who is simply a chemist testifying. He is a forensic examiner of materials. He uses chemistry as one of the bases for his opinion. Burch argued that Rudolph's testimony should be admitted and the jury could assess its weight.

The court ruled:

After this ruling, the defense moved to exclude Rudolph's testimony because it offered an investigative opinion rather than a scientific one. The defense attorney stated, [Rudolph] bolsters his opinion, as I understand it, by saying I was a trained FBI agent and, therefore, I look things over and I see certain things and this helps me in my opinion.' The court responded, [H]e is entitled to tell the jury what he based his conclusion on. Some of these things may be a little strange for a scientist, but he will be testifying as a scientist, not as an FBI agent.

Rudolph later testified at the Psinakis trial. On direct examination he testified that he identified PETN through the use of LC in conjunction with other factors which indicated to him that he was dealing with PETN. On cross-examination, Rudolph admitted that he had other instruments available to confirm the presence of PETN but that he did not use them. Rudolph agreed that what happened in this case is that [he] used one method which is used to separate substances, not to identify them, and [he] didn't use anything else in the whole FBI lab. On redirect AUSA Burch asked Rudolph, [W]as your opinion that the material was PETN based solely upon the liquid chromatography test that you ran? Rudolph answered, No it was not, or I would have not -- If it was just based solely on that, I would have used some confirmatory techniques.

The trial ended in an acquittal.

In a four-page letter dated July 8, 1989, AUSA Burch informed Laboratory Director John Hicks that Rudolph's performance in the Psinakis case was deficient. Burch stated, I believe part of the reason for the acquittal stemmed from some serious questions that arose concerning the handling of exhibits involving trace or residue amounts of explosives and the analysis of these exhibits at the FBI laboratory. Burch complained that Rudolph relied on the hearsay reports of a field agent in rendering an expert opinion that evidence contained PETN. Burch observed:

(Emphasis in original).

Rudolph, Burch stated, used liquid chromatography as the only chemical test to ascertain the presence of PETN, and he failed to perform confirmatory tests. Burch noted that the defense called a world-renowned expert who testified that liquid chromatography was the equivalent of a presumptive test that did not rule out the possibility of compounds other than PETN. Burch noted that the case raised serious questions about the Laboratory's procedures:

In this letter, Burch did not criticize Whitehurst, but instead observed that he appeared sincerely concerned about the integrity of the judicial process.

Hicks responded to Burch on July 28, 1989, by writing, I share your concerns and as a result of this matter, I have instituted an internal audit of the protocols used in the identification of explosive residues.

In reviewing Rudolph's laboratory work and testimony in Psinakis, we identified several significant problems. As noted later in this section, we found similar problems in his work on other cases.

Rudolph acknowledged that his identification of PETN on the tools was based in part on the fact that stripped detonating cord was found in the defendant's garbage. In his interview with the OIG, Rudolph observed that given this information, he presumed the material on the knife was PETN and he used LC simply as a confirmatory test.

Rudolph's approach reflects a fundamental misunderstanding of the role of a forensic scientist. As an investigative matter, the FBI had good reason to suspect that the defendant had used the tools to strip the detonating cord found in his garbage. As a forensic scientist, however, Rudolph could not identify PETN based in whole or in part on the field agent's suspicions. Rather, his conclusions had to be based on a scientific examination.

Rudolph failed to distinguish between the separate and distinct roles of an investigator and a forensic scientist. With his academic training, Rudolph should have known not to state his scientific conclusions more strongly than could be supported by the underlying analytical results. Had he recognized this fact, he would have acknowledged in his Laboratory reports and testimony that the LC tests he performed gave results consistent with, but did not necessarily identify, the presence of PETN on the tools.

Whitehurst has generally alleged that FBI examiners in explosives-related cases have purposefully slanted reports to favor the prosecution. Although he did not make this complaint about Psinakis specifically, the case merits comment on this issue. At best, Rudolph's explanation for his opinion in Psinakis reflects incompetence. Given the tests that Rudolph described, he could only say the results of his 1982 examinations were consistent with the presence of PETN. By opining that PETN had been found on the tools, Rudolph overstated the significance of his analytical results in a way that supported the government's theory of the case. This overstatement partly reflected that Rudolph inappropriately relied on information from the field agent in reaching his forensic conclusions.

3. Inadequate File Documentation

Rudolph failed to adequately document the work he claimed that he had done in Psinakis. At the trial, he testified that he prepared his notes for his own use and not for the defense or the prosecutor. These remarks reflect a basic misunderstanding of the purpose and importance of adequately documenting case files. The notes should allow someone to understand the analyses done and the basis for the conclusions reached by the examiner. The absence of such notes, as Psinakis illustrates, means that an examiner may not be able credibly to defend his or her conclusions at a later date. His supervisors should also be faulted for approving his AE dictation in the absence of adequately documented files.

Because it is well understood in the scientific community, Rudolph should have recognized the need to perform a confirmatory test in addition to the LC before concluding that PETN was found on the tools. His failure to do so reflects not only that he improperly based his opinion on the assumption that the defendant had stripped PETN from the detonating cord found in the garbage, but also that Rudolph did not follow any identified protocol in examining the evidence.

5. Conclusion

We conclude that Rudolph's performance in Psinakis was wholly inadequate and unprofessional. We do not find a factual basis to conclude that he intentionally overstated or biased his conclusions.

III. The Laboratory's 1989 Reviews of Rudolph's Casework

In August 1989, the Laboratory conducted two partial reviews of Rudolph's casework. Based on those reviews, Laboratory management concluded that further action was not required. As explained below, those reviews were not adequate to resolve concerns about Rudolph's work in Psinakis or in other cases.

A. Factual Background

After receiving Burch's letter complaining about Rudolph, Hicks gave it to SAS Chief Kenneth Nimmich and instructed him to review Rudolph's casework. Nimmich in turn asked MAU Chief Jerry Butler to review Rudolph's work in Psinakis and to also review a representative sample of Rudolph's cases to determine if appropriate analytical techniques were applied and properly performed. Butler concluded that the analytical procedures used in Psinakis were weak but laboratory accepted practice in 1982.

Over a period of several weeks, Butler reviewed Rudolph's work in approximately 200 cases and prepared a memorandum dated August 2, 1989, which described the preliminary review. Butler found numerous administrative shortcomings in the files such as insufficient notes, missing charts and weak analytical procedures. In light of the itemized weaknesses found in Rudolph's work and the potential serious impact these types of weaknesses could have on the proper administration of justice, Butler recommended that an examiner from the CTU do an in-depth review of Rudolph's case work. Rudolph told the OIG that Butler also directed him to return to the files any notes and charts that Rudolph had retained himself.

Nimmich agreed with Butler's recommendation and orally asked CTU Chief Roger Martz to conduct the further review. Over approximately two weeks, Martz reviewed 95 cases in which Rudolph had worked as a principal or an auxiliary examiner. In an August 16, 1989, memorandum to Nimmich, Martz summarized his findings and stated:

Martz cross-referenced Butler's August 2, 1989, memorandum and recommended that no further technical reviews be performed on Rudolph's case work.

With regard to the Psinakis case, Martz noted that Rudolph had been criticized for not performing confirmatory analyses. Martz observed that while liquid chromatography (LC) would not be the instrument of choice to identify an unknown powder, it could be sufficient depending on other circumstances to identify an explosive. Martz also noted, It is not unusual for a defense attorney to deliberately ask why a technique, which he knows wasn't employed, wasn't used in the identification of a chemical.

During the OIG investigation, Nimmich and Martz gave conflicting accounts of what Martz was asked to do in his review of Rudolph's cases. Martz recalled that Nimmich asked him to determine if Rudolph had in fact done some analyses to support his reported conclusions. Martz said he did not attempt to determine whether the tests conducted by Rudolph were analytically sufficient to support the reported results, but instead whether there was some work in the file to support the conclusions drawn.

Martz told the OIG that he also informed Nimmich orally in 1989 that Rudolph did the very minimum work to come to a conclusion and he did a very poor job of documenting his work. Martz, however, did not mention these things in his August 16, 1989, memorandum to Nimmich. Martz also informed the OIG that in his 1989 review, he found that approximately 10% of Rudolph's files lacked any notes at all. This information also was not included in Martz's August 1989 memorandum.

Nimmich told the OIG that he expected Martz to review the technical sufficiency of Rudolph's work. Nimmich further said he understood that Martz had done such a review, because Martz stated in his memorandum that chemical, instrumental, or physical analyses were performed in all the cases and that [t]hese analyses were sufficient to base an expert opinion as to the results that were provided. Nimmich said he interpreted these remarks to mean that Martz was satisfied that a sufficient amount of work was done to reach the stated conclusions. Nimmich told the OIG that he relied upon Martz's conclusions in determining that no further review of Rudolph's work was warranted.

Rudolph, on his own initiative, prepared a letter dated August 25, 1989, to Laboratory Director John Hicks. In this letter, Rudolph attempted to respond to criticisms MAU Chief Butler had made in his August 2, 1989, memorandum. Rudolph defended at length his work in Psinakis. With regard to record keeping and note taking, Rudolph said that many files lacked notes because he had retained them himself because the FBI's filing system was unreliable. He stated that his unit chiefs knew of this practice and that in the past five years he had received one exceptional and four superior ratings for case management and control.

In his August 25, 1989, letter to Hicks, Rudolph also said that he had returned almost all the notes and serials to the FBI files and added detailed comments to files where such materials were missing. Rudolph also observed that the quantity of notes an examiner takes is a matter of personal preference. Although Rudolph asked that this letter be made part of the official record, Hicks said he refused to accept it because he thought the issues had been resolved through Butler's review.

Laboratory management failed to assure that concerns about Rudolph's casework were thoroughly investigated in 1989. First, neither Butler in his initial review nor Martz in his subsequent review addressed the concern raised by AUSA Burch that Rudolph in Psinakis had erroneously relied on information from a field agent instead of conducting sufficient confirmatory tests to identify PETN. On a related point, as part of the 1989 reviews of Rudolph's work, Laboratory management failed to obtain and review a transcript of Rudolph's trial testimony in Psinakis. In light of the prosecutor's complaints, the transcript should have been reviewed.

In light of the conclusions stated in Butler's preliminary review, Laboratory management also failed to take appropriate further steps. Butler noted numerous administrative shortcomings, such as insufficient notes, missing charts, and weak analytical procedures in his review of some 200 cases. Rudolph says Butler directed him to return any notes or charts to the files. This directive was insufficient. First, it did not in any way sanction Rudolph for work habits that could, as the Psinakis case illustrates, undermine if not eliminate the value of the Laboratory's results at trial. Second, Rudolph had worked on several hundred cases other than the 200 Butler reviewed. At the least, in 1989 Laboratory management should have directed a more comprehensive review of Rudolph's casework.

We also find fault in the way Nimmich and Martz handled the follow-up review. Given Butler's findings, the Laboratory should have reviewed Rudolph's work to determine whether sufficient analyses were done to support the stated conclusions. Although Nimmich might reasonably have expected Martz, as an experienced examiner and unit chief, to understand the need for a thorough technical review, in retrospect Nimmich should have taken steps, preferably through written instructions, to assure that Martz understood this to be his task. It also would have been desirable for Martz to have clearly stated the object and methodology of his review in his memorandum.

Whatever he understood Nimmich's instructions to be, Martz stated the conclusions of his review in a misleading way. He observed that analyses had been performed that were sufficient, yet he told the OIG that he did not review the sufficiency of Rudolph's work to support the stated conclusions. Martz's August 16, 1989, memorandum shows that he knew of Butler's August 2, 1989, memorandum, which recommended an in depth review of Rudolph's cases. Martz in his memorandum indicated he conducted a technical review and recommended that there be no further review of Rudolph's cases. As a unit chief, Martz should have recognized that this misleadingly suggested that he had completed an in depth review and concluded that further review was not necessary.

Martz also failed to note in his memorandum that, in his review, he found that notes and other documentation were missing. These findings deserved comment even if Nimmich did not ask Martz to conduct an administrative review of the files. Finally, Martz stated in his August 16, 1989, memorandum that, while other tests could have been performed, no changes would be made in the reporting of the 95 cases reviewed. Martz lacked any basis to make this statement if, as he told the OIG, he did not assess whether the analyses identified in the files were sufficient to support the stated conclusions.

Martz's review of the Psinakis case was inadequate to address the concerns raised by AUSA Burch. Martz commented that LC might be sufficient to identify explosives, depending on other circumstances. This begged the relevant question of whether LC was sufficient in Psinakis, which it clearly was not. Martz also noted that it was not uncommon for defense attorneys to question examiners about tests they knew had not been performed. The proclivities of defense counsel were not pertinent to the issues Martz should have been addressing. Martz's comments about Psinakis inappropriately tended to excuse Rudolph.

Nimmich told the OIG that he understood from Martz's memorandum that Martz had concluded that Rudolph had a sufficient basis for his conclusions in Psinakis. Martz confirmed in his interview with the OIG, however, that he did not review Rudolph's work in Psinakis and did not address AUSA Burch's concerns about the lack of confirmatory techniques. Given these facts, Martz should not have included his comments concerning the Psinakis case in his memorandum, because they misleadingly suggested that he had approved Rudolph's work.

Finally, we find that Hicks did not take sufficient steps in response to the concerns raised by AUSA Burch's letter. Given the specific allegations, Hicks should have assured that someone at least reviewed Rudolph's testimony in Psinakis. Hicks told the OIG that he did not remember reviewing the testimony and did not recall hearing that any one else reviewed it; Nimmich did not recall if it had been reviewed; and none of the documents provided to the OIG by the FBI suggests that the Laboratory reviewed Rudolph's testimony.

Moreover, Hicks advised Burch that based on his complaints about Rudolph, the Laboratory would conduct an internal audit of the protocols used in the identification of explosive residues. Hicks told the OIG that he understood that such an audit was done as part of whatever file reviews were ordered by Nimmich. Nimmich, not surprisingly, said he did not consider such reviews to be an audit of the Laboratory's protocols for examining explosives residue. Our investigation did not identify any documents suggesting that a general audit of the protocols was ever done as a result of Burch's letter. If Hicks intended such an audit to occur, he failed to communicate his instructions clearly to others in the Laboratory.

In sum, the Laboratory's 1989 review of Rudolph was inadequate. The allegations that prompted the review came not from Whitehurst but from an Assistant United States Attorney with first-hand knowledge of the alleged deficiencies. The AUSA not only rendered his own low opinion of Rudolph's work, but repeated the similar view of the district court judge who almost excluded Rudolph's testimony. The AUSA further stated that Rudolph's inadequate work contributed to an acquittal. These were serious charges. That the Laboratory did so little in response to these allegations is deplorable. The Laboratory should have recognized Rudolph's incompetence in 1989 and initiated a complete file review and appropriate disciplinary measures. This was not only required by the proper administration of justice, but it also might have obviated the great time and effort expended in later reviews of Rudolph's files that were still continuing seven years later.

IV. The FBI OPR Investigation in 1991-92

In late 1990 and early 1991, Whitehurst again complained within the Laboratory about Rudolph's work habits and also made allegations of other misconduct, including that Rudolph was a racist, had abused annual leave, had perjured himself in a trial, and had lied to an AUSA. After an investigation by the FBI OPR, the FBI Administrative Services Division (ASD) advised Rudolph in June 1992 that the inquiry had not developed facts warranting any administrative action against him and it considered the matter closed. As explained below, we conclude that FBI OPR should have investigated certain of these allegations further, but we do not find facts indicating that FBI OPR or the ASD improperly sought to ignore or cover up the allegations made by Whitehurst.

In December 1990 Whitehurst prepared a draft memorandum detailing various allegations against Rudolph and recommending, among other things, that MAU Chief James Corby review all of Rudolph's cases. Whitehurst repeated his complaints that Rudolph was sloppy, had failed to conduct appropriate tests, and had not properly documented his work. He also alleged that Rudolph and his technician Edward Bender were racists, that Rudolph had perjured himself in a case in the Southwest, that Rudolph had lied to AUSA Burch by telling him the Laboratory lacked equipment in 1982 to do certain tests, and that Rudolph had abused annual leave.

Whitehurst discussed his memorandum with Corby. At Corby's recommendation, Nimmich forwarded the memorandum to FBI OPR in January 1991, and FBI OPR opened an investigation. In March 1991, Nimmich also directed Corby to review a number of Rudolph's cases. That review is discussed in the next section. The FBI OPR investigation was conducted primarily by Special Agent Robert O'Brien, who reported to Special Agent Ralph Regalbuto. During 1991, FBI OPR interviewed several witnesses, including Rudolph, Whitehurst, and others who worked with them in the Laboratory. FBI OPR concluded that the evidence did not support Whitehurst's allegations. Based on FBI OPR's investigation, the ASD advised Rudolph in a letter dated June 22, 1992, that the inquiry was considered closed.

Based on our review, we conclude that FBI OPR should have conducted a more thorough investigation with respect to three of Whitehurst's allegations.

Whitehurst alleged that both Rudolph and Bender were racists and that this affected their work product. None of the witnesses interviewed by FBI OPR substantiated the allegation that Rudolph made racist remarks at work or was a racist. Several witnesses did confirm, however, that Bender regularly made racial jokes or remarks in the Laboratory.

Given Whitehurst's allegations, FBI OPR should have pursued its investigation further by asking witnesses if they knew of any specific case in which Bender's racial remarks or any biases might have affected his work. In response to the OIG investigation, O'Brien of FBI OPR said that a further investigation of Bender was not undertaken because there was no evidence that racial views had influenced his work and he was no longer an FBI employee. Similarly, Ralph Regalbuto, who supervised O'Brien at the time of the investigation, stated that FBI OPR would not have investigated the allegations against Bender because there was no indication of conduct that might be referred to a criminal investigative body and FBI OPR lacked authority to investigate non-FBI employees.

FBI OPR should have pursued its investigation to the point of asking witnesses if they knew of instances where Bender's views on race had affected his work. If the facts suggested they did, it is conceivable that some remedial action would have been appropriate with regard to cases he worked on while at the FBI. Moreover, if there were facts suggesting he slanted conclusions because of racial animus against a suspect, a criminal referral might have been appropriate.

In the course of our investigation, we contacted the individuals interviewed by FBI OPR in its earlier investigation and asked if they knew of any instances where Bender's racial views affected his casework. Several individuals again recalled him making racial comments in the Laboratory, but no one identified any specific instances where they thought his attitudes affected his work. These same persons stated that they did not think Bender would have altered reports or data based on the defendant's race.

The second allegation by Whitehurst that we think merited further investigation by FBI OPR is that of Rudolph's alleged perjury. Whitehurst claimed that in an unidentified case in the southwestern United States, Rudolph falsely testified that his initials were on a piece of evidence. According to Whitehurst, Rudolph told him about this incident to illustrate that [b]efore you embarrass the Bureau, you should be willing to perjure yourself.

In response to the FBI OPR investigation, Rudolph denied ever falsely stating that his initials were on evidence. Both O'Brien and Regalbuto of FBI OPR advised the OIG that because Whitehurst had not provided more specific information about the case in which the alleged perjury occurred, it was not necessary to investigate the allegation further once Rudolph denied it. O'Brien also noted that Whitehurst may have misunderstood remarks that Rudolph intended as teasing or a joke.

We disagree. Whitehurst in his allegations noted that the testimony was in a southwestern court, that Rudolph had only testified six or seven times before this incident, and that form FD-126s used by the Laboratory would list trials in which Rudolph testified. In view of the serious nature of the accusation, and the information identified by Whitehurst, we think FBI OPR should have attempted to identify cases in the Southwest in which Rudolph had testified and to review transcripts of his testimony. If that review disclosed testimony similar to that described by Whitehurst, FBI OPR then could have investigated further by contacting the prosecutor and seeking to examine any evidence that remained available.

Whitehurst also accused Rudolph of falsely telling AUSA Burch in the Psinakis case that he could not have more thoroughly examined the evidence because Rudolph lacked the equipment in 1982 that he had in 1989. Rudolph, Whitehurst noted, had stated in a paper published in 1983 that he used mass spectrometry and infrared spectrophotometry in 1981 and 1982 and that he had used these techniques in hundreds of explosives cases. Rudolph denied ever providing false information to AUSA Burch.

O'Brien stated that FBI OPR did not investigate this allegation further because it did not seem to contain the elements of a lie or concern an important issue. O'Brien observed that Rudolph's published statements that he used certain equipment in 1982 did not mean it was available to him when he did the Psinakis examinations. O'Brien also noted that Rudolph's former unit chief Charles Calfee thought Rudolph would have been accurate in stating that certain equipment was unavailable because it was still in an experimental mode within the Laboratory.

FBI OPR did not interview AUSA Burch about this issue or review Rudolph's 1983 paper. Regalbuto of FBI OPR acknowledged that these might have been reasonable investigative steps, depending on the circumstances, but noted that the investigating agent has some latitude in determining if a sufficient investigation has been done. He also observed that if the investigation was insufficient, the FBI's Administrative Summary Unit (ASU), which makes recommendations based on the investigations, should have asked for more to be done.

Rudolph's alleged lying about the availability of equipment was a serious issue that merited further investigation by FBI OPR. Despite Rudolph's denial, O'Brien should have pursued this matter further by at least questioning Burch about it and reviewing Rudolph's paper.

After FBI OPR completed its investigation of the allegations against Rudolph, the ASU reviewed the matter and recommended it be closed. The ASU is part of the ASD, which later advised Rudolph that the inquiry was considered closed. In the ASU, the agent who reviewed the matter was John Dietz, who had been assigned to the ASU on temporary duty for three months. In an interview with the OIG, Dietz stated that he did not know either Rudolph or Whitehurst at the time of the investigation and he said that he had no reason to think the allegations were discounted because they were made by Whitehurst. Dietz acknowledged that, in hindsight, further investigation might have been helpful, but observed that he must have been convinced at the time that the FBI OPR investigation had been sufficient.

In sum, we conclude that FBI OPR should have investigated further the allegations concerning Bender's racial bias and Rudolph's alleged perjury and the alleged lie to AUSA Burch. Our review of the FBI OPR investigation and the ASU's resulting recommendation to close the matter did not disclose facts indicating that there was a deliberate effort to dismiss or ignore Whitehurst's allegations.

V. The 1992 Corby Review

Because the FBI OPR did not have the technical expertise to review Whitehurst's allegations concerning Rudolph's casework, the Laboratory itself conducted another case review. SAS Chief Kenneth Nimmich directed MAU Chief James Corby to review a representative sample of Rudolph's cases.

Corby reviewed approximately 200 cases and found many serious flaws in Rudolph's work. He described his findings in a handwritten draft memorandum that he gave to Nimmich in the spring of 1992. Corby noted that Rudolph had failed to follow his own explosives residue protocol, had formed conclusions and prepared dictation without a basis, had failed to run standards or confirmatory tests, had offered opinions to fit the case scenario or findings of other units whether or not supported by his own analyses, had failed to label charts properly, and, where data was present in the file, had sometimes made technical errors.

Based upon his review, Corby recommended that appropriate disciplinary measures immediately be administered to SSA Rudolph for unacceptable casework performance and that such disciplinary action include censorship, suspension and probation for a period of time. Corby also recommended that SSA Rudolph immediately be barred from participating in any explosive-related program or research being conducted by the FBI laboratory and that all files be thoroughly reviewed in those cases where SSA Rudolph testified before any judicial proceeding in order to determine if further action is warranted in this matter.

Nimmich returned the draft memorandum to Corby and told him it was not his place to recommend particular disciplinary measures. According to Nimmich, he also asked Roger Martz, then the CTU Chief, and CTU examiner Lynn Lasswell to participate with Corby in a panel review of Rudolph's cases. Nimmich told the OIG that he told the panel members to review the cases to see if there were errors that we needed to address back to a prosecutor, to a defense attorney, or anything of that type in terms of bad casework, if you would, errors that would have been made, misrepresentations of what was actually there. He expected each panel member to review all of the files.

The panel review evidently was not implemented in the manner Nimmich contemplated. Corby believed that Lasswell and Martz became involved only after Corby had given his draft memorandum to Nimmich. Lasswell received 57 cases from Corby and reviewed them for technical and administrative errors. He took detailed notes of his findings and gave them to Corby. Among other things, Lasswell observed that certain of Rudolph's cases lacked sufficient tests to support the stated opinions, that notes and charts were missing for some evidence, and that confirmatory techniques had not been used. Lasswell thought these problems were very serious and could greatly affect the cases if they went to court.

Martz, when interviewed during the OIG investigation, could not recall participating in the 1992 panel review. While he remembered talking with Corby and Lasswell about their review, Martz said he did not remember reviewing 200 cases himself and he had no notes related to any such review. Lasswell and Corby, like Nimmich, each thought that Martz was also reviewing Rudolph's cases. Our investigation, however, did not identify any memorandum, notes, or other documents by Martz reflecting such a review.

Nimmich prepared a memorandum to Hicks dated April 30, 1992, reporting on the results of the panel review of Rudolph's cases. Nimmich's memorandum stated that 200 cases were reviewed by a panel consisting of Corby, Martz and Lasswell. The memorandum observed:

The memorandum reported that [n]o instances of fraud or intentional misrepresentations were found during this review; however, it was evident that the quality of work was severely lacking.

The memorandum recommended that Rudolph receive a severe reprimand based on the lack of professionalism and attention to detail reflected in his casework. Nimmich also recommended that for the 57 files with unacceptable documentation, Rudolph should be asked to bring the working notes up to an acceptable level through the use of personal diaries, notes, or recollection and to prepare a memorandum reflecting the additional information for each file.

Nimmich's memorandum does not indicate that copies of it were sent to any of the panel members. Corby said he did not see the memorandum until several years after it was prepared; Martz and Lasswell said they had not seen it before it was shown to them during the OIG investigation. Nimmich recalled consulting with Corby in preparing the memorandum, but Corby did not remember such a discussion. Nimmich also recalled that he consulted with Corby and Hicks before recommending that Rudolph receive a severe reprimand.

On May 18, 1992, Hicks discussed the file review with Rudolph. Without consulting Nimmich, Hicks decided to verbally admonish Rudolph rather than reprimand him. Rudolph recalled that Hicks gave him a mild chewing out and told him he was not being reprimanded because his unit chiefs had approved his work. Rudolph said that in this meeting, Hicks also gave him a $500 incentive award for something Rudolph had recently done, and Hicks said words to the effect that maybe this would help your day. The verbal admonishment was the only sanction imposed by the FBI on Rudolph for the poor quality of his work. During the OIG investigation, Rudolph said he was surprised by this leniency, as he had expected and even thought he deserved a letter of censure.

Consistent with Nimmich's recommendation, Rudolph was directed to attempt to bring the files up to an acceptable level by adding information to them. In an August 18, 1992, memorandum to Hicks, Rudolph identified changes he made in 40 of the 57 files. Rudolph stated in his memorandum that [n]o attempt was made to alter or change any conclusion or report, only to improve the clarity and understanding of what was done.

Rudolph was directed to place a memorandum in each file documenting that changes were made. In his memorandum to Hicks, Rudolph resisted this action, stating that it would only serve as a red flag in any future defense subpoena and could draw unwarranted attention to the file. Rudolph stated that most of the additions and labeling that was done is something an examiner might do anyway in sprucing up the file' before a court testimony and did not need to be memorialized in the file.

Despite Rudolph's protest, Nimmich required him to prepare a memorandum for each file reflecting that changes had been made. Nimmich said he reviewed the memoranda himself and directed that they be placed in the files. Based on these actions, Nimmich wrote a memorandum to Hicks dated March 12, 1993, advising that the review of Rudolph's cases should be considered closed and no further action be taken.

The 1992 review of Rudolph's cases and the Laboratory's response to that review were insufficient in several respects.

At the outset, Nimmich should have given clear, written directions to those participating in the review as to its objective and the procedures to be used. Had such directions been given, substantial problems of miscommunication or misunderstanding might have been avoided. Nimmich indicated in his April 30, 1992, memorandum to Hicks that a panel of Corby, Lasswell, and Martz had reviewed 200 of Rudolph's cases. In fact, Lasswell only reviewed 57 cases, and it is unclear whether Martz reviewed any at all as part of the 1992 review.

On a related point, Nimmich should have circulated to the panel members drafts of the sections of his April 30, 1992, memorandum which described the panel's findings. This would have assured that the memorandum that later went to Hicks accurately described what each panel member had done in the review and that they agreed with the description of their findings. Moreover, reactions to the drafts by Corby or Lasswell might have been significant to Nimmich as he considered his recommendations for sanctions against Rudolph for the condition of his files.

Given the problems identified in the 1992 case review, we also think that Laboratory management failed to take sufficient remedial steps or to impose adequate sanctions on Rudolph. The 1992 case review identified serious deficiencies in 57 of approximately 200 cases reviewed. But Rudolph had worked on hundreds of cases before leaving the Laboratory in 1988. Once Laboratory management learned that a case review identified deficiencies in more than 25% of the reviewed cases, a comprehensive review of all of Rudolph's case work should have been undertaken.

We do not fault Hicks and Nimmich for directing Rudolph to attempt to bring the 57 files to an acceptable level and to document his actions, but this directive was not followed appropriately by Rudolph or monitored adequately by management. During the OIG investigation, Rudolph stated that he did not recall documenting in the individual memoranda every change he made to the files. Rudolph also admitted that it was not uncommon for him to label charts or otherwise to change files before trial without documenting these actions. This echoes his earlier statements to Hicks in his August 18, 1992, memorandum when Rudolph argued he should not be required to place a memorandum in each file reflecting any changes, because it was common for examiners to spruce up a file without documenting that action.

Rudolph's statements reflect a fundamental misunderstanding of the importance of accurate work notes and adequate case documentation. As noted earlier, the lack of such documentation may mean, as was demonstrated in Psinakis, that it is impossible to determine what was done in earlier analyses. Moreover, work notes are generally understood to have been prepared contemporaneously with the examinations or analyses they concern. Such notes can be misleading if they are created or spruced up at a later date without that fact being indicated in the notes themselves. Their preparation sometime after the work they describe obviously can be relevant to the weight or credibility of any testimony that is based on them. When Rudolph communicated to Hicks the view that it was common and appropriate for examiners to spruce up their files before trial without documenting such action, Laboratory management should have taken appropriate steps to advise Rudolph and others that such a practice is not acceptable for a forensic laboratory and would not be tolerated.

Despite the findings reported in Nimmich's April 30, 1992, memorandum, Rudolph received only an oral admonition, one of the most lenient punishments available. The 1992 file review revealed that Rudolph's cases had extensive problems with inadequate documentation, insufficient confirmatory tests, and conclusions that were not fully supported by the information in the files. Rudolph should have been seriously disciplined for his inadequate work and his failure to return documentation to the files in accord with directions he was given in 1989. We find unpersuasive the suggestion that Rudolph deserved no more than an admonishment because unit chiefs had approved his work. The case files do suggest that his unit chiefs, particularly Charles Calfee, did not adequately review his work to assure that it was appropriately documented and that the stated conclusions were reasonably supported. This fact does not excuse Rudolph's lack of professionalism. He should have recognized the shortcomings in his own work, particularly given his academic credentials in chemistry and experience in the Laboratory.

VI. The 1995 Corby Review

In the spring of 1993, Laboratory Director Hicks named James Kearney to replace Kenneth Nimmich as the chief of the SAS; Nimmich in turn took Kearney's former position as the chief of the FSRTC in Quantico. Shortly after Kearney took his new position, James Corby approached him to complain about Rudolph's work and to argue that a further review should be made because there were serious problems in the files.

After learning of Corby's concerns, Kearney asked Martz to review several of Rudolph's problem files to see if corrective action had been taken. Martz prepared a memorandum for Kearney that described the contents of particular files but did not state any findings or conclusions. During the OIG investigation, Martz explained that he had been unable to find certain notes and charts that Rudolph said had been returned to the files. Martz, however, did not recall discussing this point with Kearney, but remembered only giving him the memorandum.

Corby was so concerned about the condition of Rudolph's files that he asked Kearney to raise the issue with Hicks. That concern led to Corby meeting sometime in the spring of 1993 with Hicks, Martz, Kearney, and Wayne Taylor, who was then Hicks' deputy. Corby argued that technical problems with Rudolph's work merited a further review. According to Hicks, Martz disagreed and represented that Lasswell also disagreed with Corby. Martz denies saying Lasswell disagreed with Corby and told us he thinks Lasswell also found problems in Rudolph's work.

Hicks recalls concluding that the Rudolph matter should be closed in 1993 because he thought the allegations had been reasonably investigated and no technical deficiencies had been found in the several case reviews. During the OIG investigation, Hicks stated that his conclusion was significantly influenced by his belief that Lasswell had not found serious problems in Rudolph's cases, and Hicks said he would have reacted differently had he known that Lasswell in fact thought there were serious errors that would affect Rudolph's ability to testify to the results.

In February 1994, Whitehurst's attorney Stephen Kohn wrote to the FBI describing various allegations regarding the Laboratory, including complaints about Rudolph. During the spring of 1994, the OGC conducted an investigation in response to Kohn's letter. In a May 26, 1994, memorandum describing the results of the investigation, the OGC concluded that Corby should undertake a final, more comprehensive review of all of Rudolph's cases. The FBI memorandum observed that such a review of the files would most likely reveal that they are sloppy and that his [Rudolph's] conclusions are not supported by appropriate documentation.

One of the OGC attorneys involved in the 1994 investigation, John Sylvester, recalls that the Laboratory Division was furious with the recommendation for another review. Kearney, however, said that by May 1994 he had independently concluded that such a review should be done. In any event, in June 1995, about a year after the OGC made its recommendation, Corby was directed to review all cases in which Rudolph had worked as a principal examiner or auxiliary examiner in the MAU.

Corby's instructions were outlined in a June 12, 1995, memorandum from Kearney to Milton Ahlerich, who had recently become the Laboratory Director after Hicks retired. The memorandum asked Corby to categorize his findings as follows:

In a memorandum dated November 30, 1995, Corby reported the results of his review. Corby noted that he characterized Rudolph's conclusions as correct if he found any basis for the reported results in his file review. Applying this standard, Corby found 20 files in category four, 137 in category three, 76 in category two, and 421 in category one. Corby concluded that 24% of Rudolph's cases were in categories 3 or 4 and did not meet the administrative or technical guidelines at the time the cases were worked. (Emphasis in original.) In Corby's opinion, they would not be acceptable under close judicial scrutiny, or past or present peer review. (Emphasis in original.)

Rudolph was allowed to review the cases Corby placed in categories 2, 3, and 4 and to respond in writing. On May 7, 1996, Rudolph submitted a nearly 200-page response in which he defended his work and strongly disputed many of Corby's criticisms. Rudolph made general responses to certain recurring issues and also addressed individual cases. In an interview during the OIG investigation, Corby commented on Rudolph's responses.

In this section, we assess Corby's findings and Rudolph's response. We begin by discussing several recurring problems identified by Corby.

Corby noted in his 1995 review that Rudolph seemed to report a disproportionately large number of examinations relative to the number of specimens. Rudolph responded that it was common for examiners to conduct multiple examinations of the same specimen. Unfortunately, Rudolph's files generally lacked work notes or other documents that would explain the number of examinations conducted.

In several cases, Corby found that Rudolph had failed to follow protocols. Rudolph argued that the FBI Laboratory did not have any official protocol during his tenure there. Even so, as Corby noted, Rudolph had described a protocol in the FBI's 1983 Symposium on Explosives Residue Analysis, and we do not understand why he would disregard that protocol in his own work.

Corby also found that charts or notes were missing in many cases. Rudolph offered several responses: the documents may have been lost during the multiple file reviews, he kept documents himself because of the FBI's inadequate filing system, and his unit chief's approval of his work indicates that adequate documentation was once there. Rudolph's responses are unpersuasive. He was directed by Butler in 1989 to return any notes or other documents to the files, so the files should not have been incomplete in 1995. Moreover, his former Unit Chief Charles Calfee observed that the commonly understood practice was that an examiner would make a notation in the file if he or she had removed or retained documents. No such notations appear in Rudolph's files. Rudolph's case files in general are much more incomplete than others we reviewed, and therefore we find it implausible that the shortcomings in Rudolph's files are primarily due to deficiencies in the FBI's filing system.

One of the main criticisms made by Corby in his 1995 review was that Rudolph's files reflected sloppy note taking and other administrative deficiencies, such as insufficient documentation, charts that did not have specimen or file numbers, and charts without identified peaks. In his response, Rudolph again observed that his unit chiefs had approved his work and stated that he only recalled one, Roger Asbury, asking for more precise notes and that none had asked that charts be completely identified. In fact, in a 1987 Progress Review for Rudolph, Asbury observed that communications of results could improve with more comprehensive and detailed notes in preparing reports. Rudolph signed this Progress Review.

In his OIG interview, Rudolph also defended the condition of his files by stating that he dismissed identifying all of these notes and charts as not important because he would do it if it goes to trial. He said he would spruce up a file if a case went to trial; that is, he would identify peaks on charts, add additional notes if necessary, and clean up the notes already in the file. But Rudolph did not document in the file which material had been added at a later date. As we have noted earlier, Rudolph fails to recognize the importance of accurate, complete work notes and documentation, and his practice of sprucing up files is both unprofessional and unacceptable for any credible forensic laboratory.

Corby's 1995 review also criticized Rudolph for using ion chromatography (IC) as the only identification technique in some cases and for failing to identify all the peaks on charts from x-ray powder diffraction (XRD). Rudolph responded that IC has long been used as an identifying technique and that in some instances other techniques could not be employed. The files, however, generally lack information that would indicate why other techniques were or were not used. With regard to the XRD charts, Rudolph argued that labeling was not necessary because he could recognize what the peaks represented when he later reviewed the charts. He also maintained that once he identified the main component, he compared the remaining peaks against peaks for other known explosives. The lack of adequate documentation is inexcusable, despite Rudolph's claim that he could later recognize the peaks, and makes it impossible to corroborate his assertion that he made comparisons with other unlabeled peaks.

Another recurring problem noted by Corby is that Rudolph failed to report results that might have been significant. In his response, Rudolph asserted that this is mostly a matter of experts differing about what constitutes a significant result. In an interview with the OIG, Corby observed, and we agree, that this is another manifestation of Rudolph's inadequate note-taking practices. If tests or analyses yield results that could affect the examiner's conclusions, those results should be recorded in the examiner's notes. This will assure that potentially useful information is not lost, and the examiner can document his reasons for not utilizing particular results in forming his conclusions.

In responding to Corby's 1995 review, Rudolph also addressed particular cases. As part of the OIG investigation, we reviewed a sample of the cases and concluded that Corby's criticisms were for the most part justified. Our ability to evaluate Corby's findings was limited, however, by the fact that Rudolph had made further changes in the files in responding to Corby's review. During a February 28, 1996, interview with the OIG, Rudolph admitted that he was still sprucing up files after the most recent review and was not documenting the changes. Accordingly, when we reviewed particular cases and could not validate Corby's criticisms, it was unclear whether this reflected particular additions made to the file after Corby's review.

Our limited review of case files convinced us that Corby's findings were generally correct. There is one issue that was not addressed in Corby's most recent review or the earlier reviews, and that is contamination. As illustrated by the Psinakis case, Whitehurst has complained for some time that Rudolph, because of his sloppy work habits, could have reached conclusions based on his own contamination of the evidence. During an OIG interview, Rudolph stated he had never contaminated evidence but admitted that he did not always wear gloves in the Laboratory, place paper down when doing examinations, or take control swabs of his work area. Rudolph also admitted that his work area was unkempt and that a messy laboratory was almost his and Bender's trademark. These remarks suggest that Rudolph did not appreciate the significant problems of contamination in explosive examinations and therefore failed to take appropriate preventive measures.

In a substantial number of his cases, Terry Rudolph did not perform his work as an examiner in a manner that would withstand peer review or judicial scrutiny. In Psinakis, he did not adequately document his case work, he failed to conduct required confirmatory tests, and his stated conclusions lacked a valid scientific basis. The reviews of Rudolph's work conducted by the Laboratory after Psinakis confirm that his lack of competence was not isolated to that case.

Rudolph displayed an attitude towards case documentation that is inconsistent with the presentation of credible scientific conclusions. His belief that notes are only for the examiner's own use and that files may be spruced up on the eve of trial is unacceptable. In our investigation we found no evidence that other examiners had made after-the-fact changes to case documentation without noting such actions in the file.

That Rudolph continued to spruce up his files without documenting the changes even after he had been directed by Nimmich to produce such documentation, and while his files were being reviewed, was insubordinate and constitutes willful misconduct.

Rudolph attempted to justify his conduct by noting that unit chiefs had approved his work. The condition of Rudolph's files suggests that his unit chiefs, particularly Charles Calfee who served as his unit chief from 1979 through 1986, did not sufficiently review his work. That fact, however, does not excuse Rudolph's failure to conduct confirmatory tests, to run appropriate standards or controls, to follow protocols, or to document his work appropriately.

Our investigation did not identify facts suggesting that Rudolph made intentional misrepresentations in his files or attempted to slant his results to favor the prosecution. Indeed, our own review of Rudolph's cases, our interviews with him, and the 1995 review by Corby indicate that Rudolph indiscriminately disregarded appropriate scientific methods and failed to document his work without regard to whether the results favored the prosecution or the defense.

In the Rudolph matter, Laboratory management repeatedly failed to address serious concerns about the very integrity of the Laboratory's forensic results. A complete review of Rudolph's case work should have been conducted in 1989, after AUSA Burch complained about Rudolph's conduct in Psinakis and MAU Chief Butler identified numerous administrative shortcomings in 200 cases and the need for an in-depth review.

The 1989 review by Martz of 95 cases was not sufficient. Nimmich should have given written directions to assure that an in-depth review did occur. That review should have encompassed all of Rudolph's cases. Martz presented his conclusions in a misleading way that incorrectly suggested he had reviewed and approved the technical sufficiency of Rudolph's work and that Rudolph had done nothing wrong in Psinakis. We did not conclude that Martz intentionally sought to mislead in his memorandum, but, whatever he understood his instructions to be, Martz should have stated more clearly what he did to reach his conclusions.

We especially deplore the inadequacies of the Laboratory's 1989 review because (1) it was prompted by an AUSA who stated that Rudolph's shortcomings contributed to an acquittal of a defendant in a federal prosecution, and (2) a proper review in 1989 could have obviated the need for later efforts to evaluate Rudolph's work. Hicks' inadequate response to the AUSA's letter and Martz's misleading memorandum contributed most to the failure of the 1989 review.

Laboratory management also failed adequately to respond to the results of the 1992 review. Again, Nimmich should have provided clear, written instructions concerning the objectives and methodology of that review. He also should have allowed Corby, Lasswell, and Martz to review relevant parts of his memorandum that purportedly described their conclusions. We think Laboratory Director Hicks erred in rejecting Nimmich's recommendation that Rudolph should be severely reprimanded and deciding instead to impose a mere verbal admonition. Moreover, the Laboratory failed to assure that Rudolph had returned materials to the identified files and fully documented any additions or changes he made. Given the findings in Corby's 1992 review, the Laboratory again should have directed a complete review of Rudolph's files, rather than determining that the matter should be closed.

Although we conclude that Laboratory management failed to assure that the allegations about Rudolph were adequately investigated and resolved, we cannot conclude that those allegations were deliberately ignored or that there was an effort to cover up Rudolph's deficiencies as an examiner.

The OGC appropriately recommended in 1994 that Corby undertake a comprehensive review of Rudolph's cases. As noted above, we generally agree with the conclusions reached by Corby when he completed the review in November 1995.

Based on the above findings, we recommend that a notation describing the conclusions of this Report should be included in each case file for which Rudolph prepared AE or PE dictation. Rudolph, as noted above, retired in June 1996. Accordingly, we do not recommend disciplinary action against him. We understand that after his retirement, he did some work for the FBI on a contractual basis. Based on our investigation, we recommend that the FBI not employ him in the future.

Finally, we note that the Rudolph matter illustrates several respects in which the Laboratory policies or procedures could have been improved. During Rudolph's tenure in the Laboratory, there was no formal quality assurance program. The problems exhibited in Rudolph's case work might have been prevented if such a program had been implemented and had provided guidelines for case documentation, adequate case review, and the use of properly validated protocols. We comment on these issues further in Part Six of this Report, which discusses general recommendations to enhance the quality of the Laboratory's forensic work.

#####

The FBI Laboratory: An Investigation into Laboratory Practices and Alleged Misconduct in Explosives-Related and Other Cases (April,1997) Office of the Inspector General, Department of Justice