On May 10, 1977, in Pulaski County, Arkansas, circuit court, William H. Knott was convicted of two counts of capital felony murder and sentenced to two consecutive life terms without parole. Knott filed a petition for a writ of habeas corpus in the federal district court for the Eastern District of Arkansas. He alleged he was deprived of effective assistance of counsel when his trial counsel failed to secure expert assistance in order to attack the trial testimony of the state toxicologist. After evidentiary findings by the magistrate, the district court, the Honorable G. Thomas Eisele presiding, granted the writ. This appeal followed. We reverse.

Facts.

In the state trial court, evidence revealed that in the early morning of July 9, 1976, a fire consumed a house containing Knott’s former wife and three other people. Knott’s former wife and another individual died in the fire. Two witnesses saw a male running from the scene wearing a dark shirt and darker pants. When Knott was arrested, approximately three hours later, his clothing fit these descriptions. Knott’s hair was singed and his ear had several wounds including a blister. One witness testified that, two days prior to the fire, he heard Knott tell his wife that he would burn the house down. The state toxicologist, Berwin Monroe, testified that he found traces of a petroleum product on Knott’s hands and metal traces on Knott’s hands and pants whose composition was consistent with that of a bucket found in the burned house. One of the occupants of the house testified that the bucket did not “belong” in the house. Upon conviction, Knott appealed to the Arkansas Supreme Court. His conviction was affirmed. Knott v. State, No. CR 77-211 (Ark. Apr. 24, 1978).1

In his second petition for a writ of habeas corpus in federal court, Knott alleged for the first time that he had received ineffective assistance of counsel at his trial. Federal Magistrate Henry L. Jones ruled that Knott had not deliberately bypassed his opportunity to raise this issue in state court and that Knott’s failure to do so and filing of the petition at issue were not an abuse of the writ. Although Knott alleged his counsel was ineffective in several respects, the magistrate found that Knott’s counsel was ineffective only because of his failure to secure assistance in attacking the testimony of the state toxicologist. These findings were adopted by Judge Eisele. The State appeals the granting of the writ of habeas corpus. Knott cross-appeals, raising all of the issues raised before the magistrate as alternative grounds for relief.2

*1210 Exhaustion of State Remedies.

Prior to filing his petition for post-conviction relief in federal court, Knott sought state relief under Ark.R.Crim.P. 37. However, he did not allege ineffective assistance of counsel as grounds for relief. The State argues that Knott cannot raise this issue for the first time in federal court.

However, under the Arkansas rule, Knott is barred from filing a new petition stating grounds for relief which could have been raised in his initial petition. Ark.R.Crim.P. 37.2(b); Scott v. State, 592 S.W.2d 122 (Ark. 1980). Under similar circumstances, this court has reasoned that a petitioner had sought all available state remedies and had thus exhausted his state remedies. Witham v. Mabry, 596 F.2d 293, 299 n.7 (8th Cir. 1979).

The Supreme Court has held that a state procedural bar, like the Arkansas rule, can serve to preclude federal relief if petitioner’s failure to assert his or her claim in the original state proceeding was a “deliberate by-pass” under Fay v. Noia, 372 U.S. 391, 83 S.Ct. 822, 9 L.Ed.2d 837 (1963). Murch v. Mottram, 409 U.S. 41, 93 S.Ct. 71, 34 L.Ed.2d 194 (1972). We deem the Fay standard applicable in this case, despite intervening Supreme Court decisions narrowing its applicability, see Wainwright v. Sykes, 433 U.S. 72, 97 S.Ct. 2497, 53 L.Ed.2d 594 (1977), because Knott was not assisted by counsel in filing his state petition. Rinehart v. Brewer, 561 F.2d 126, 130 n.6 (8th Cir. 1977).

The magistrate and trial court found that Knott’s initial failure to raise the ineffective assistance issue did not constitute “deliberate by-pass.” After his initial federal petition was denied, Knott was assisted by another inmate in filing his state petition. This inmate advised Knott to assert the same grounds he had asserted in his federal petition. Knott’s state petition mentioned inadequacy of counsel as a reason why certain issues were not raised on direct appeal. Knott later testified that he believed such constitutional claims were properly raised in federal court. We agree with the district court, this series of events does not evidence the requisite deliberate waiver. We also uphold the magistrate’s finding that the filing of the petition at issue was not an abuse of the writ.

Ineffective Assistance of Counsel.

In order to prevail on an ineffective assistance of counsel claim, a petitioner must prove both (1) that his or her attorney failed to perform with the degree of skill and diligence with which a competent attorney would perform under similar circumstances, United States v. Easter, 539 F.2d 663 (8th Cir. 1976), cert. denied, 434 U.S. 844, 98 S.Ct. 145, 54 L.Ed.2d 109 (1977), and (2) that his or her counsel’s incompetence prejudiced his or her defense. United States v. Johnson, 582 F.2d 1186 (8th Cir. 1978); Reynolds v. Mabry, 574 F.2d 978 (8th Cir. 1978). The magistrate and trial court found both of these elements in Knott’s counsel’s failure to secure assistance in attacking the testimony of the state’s expert.

The state’s expert, Arkansas State Toxicologist Berwin Monroe, testified concerning three general findings from his examination of Knott’s body and clothing. First, he found traces of metal in several “smears” on Knott’s pants. The composition of these traces matched the composition of a metal bucket found on the back "porch of the burned dwelling. Second, Monroe found similar metal traces on the palms of Knott’s hands. Finally, Monroe detected a flammable, petroleum product on the palms of Knott’s hands. These findings were made using three procedures: trace metal detection, energy dispersive x-ray fluorescence analysis, and gas chromatography.

John Achor, Knott’s counsel, challenged this testimony on cross-examination. During cross-examination, Monroe conceded that the metal traces on Knott’s pants could have been acquired up to 10 months before the day of the fire. Achor elicited testimony from Monroe that many inexpensive metal goods, including most buckets and wire products, are composed of the same elements found on Knott’s hands and pants. Achor suggested through examination of Knott that the metal traces could have been acquired while Knott was in jail, exposed to numerous metal objects. Finally, Achor *1211showed that Monroe could have taken photographs of the metal traces on Knott’s hands as they appeared under ultraviolet light during the trace metal test. Counsel adduced through Monroe that this would have allowed the jury to compare such pictures to similar pictures of the bucket or of a hand which had held the bucket or some clothing which had touched the bucket. He argued this would have enabled the jury to form its own conclusion about the similarity of the metals’ composition.

At the beginning of his cross-examination, Achor impeached Monroe using Monroe’s prior inconsistent statement at the preliminary hearing that he had found “no real landmarks” on Knott’s hands. Achor later showed that many petroleum products exist, including some perfumes, and that Monroe could not state that the product found on Knott’s hands was gasoline or diesel fuel. Through cross-examination of Monroe and examination of Knott, Achor suggested that the petroleum residue could have been acquired while Knott was burning railroad waste, a petroleum product, in a “hobo jungle” prior to the morning of the fire. Achor also extracted testimony from Monroe revealing that he found no flammable substance in the bucket.

In the evidentiary hearing before the magistrate, petitioner elicited the testimony of Irving Stone, Chief of the Physical Evidence Section at the Institute of Forensic Sciences in Dallas, Texas. On the basis of this testimony, the magistrate found that petitioner’s counsel did not possess the requisite expertise to effectively cross-examine Monroe. The magistrate concluded that petitioner’s counsel should have sought an expert to advise and counsel him and placed such an expert on the witness stand for the defense.3 The magistrate found that Stone suggested six weaknesses in Monroe’s testimony which petitioner’s counsel should have exploited. First, Stone testified that trace metal detection tests were used in very few laboratories. Monroe contradicted this testimony. Stone did not specifically indicate why the test was not widely used or question its utility for the limited purpose of demonstrating that a person had contact with metal of a general type. Second, Stone testified that he had never heard of trace metal detection being used when the holding of a gun was not the underlying issue. Stone did not, however, indicate why the technique would be ineffective in other situations. Trace metal detection techniques can be used both (1) to examine the dispersion of metal traces in order to compare it with the pattern ordinarily created by handling of a specific type of object, e.g., a gun, and (2) to determine if the composition of the metal traces is consistent with the composition of a metal object. It is perhaps arguable that holding a gun yields a characteristic pattern of metal traces while holding a bucket does not. But Monroe did not testify that the pattern of metal traces indicated that Knott had held a bucket. He simply stated that the trace metal detection tests indicated that traces of metal of a similar composition to that of the metal bucket were found on defendant’s hands and pants. This finding, was also corroborated by the findings of the energy dispersive x-ray fluorescence analysis. At the habeas hearing, Stone testified that the findings of the energy dispersive x-ray test used by Monroe were “positive hard findings” which he could not dispute.

The third problem identified by the magistrate was that Knott’s pants contained a green dye, one element of which is bromine. Stone testified that a finding of bromine is often an indication of the presence of leaded gasoline, but would not so indicate in this case because of the green dye. Monroe testified that he found a petroleum product on Knott’s pants only during redirect examination. During the habeas hearing in federal court, Monroe also indicated that he found more bromine in certain spots on *1212Knott’s pants than in control areas on the back of the pants.

Stone’s fourth criticism which the magistrate adopted suggests that Monroe could have conducted more tests to determine if an accelerant had been in the bucket. Monroe admitted during direct examination that he did not find any traces of a flammable substance inside the bucket and Achor elicited the admission again during cross-examination. Revealing that further tests could have been performed would not have advanced Knott’s defense unless these tests could have conclusively disproven the presence of an accelerant. Stone did not suggest that this was the case.

The magistrate’s fifth criticism was that the tests were done approximately nine hours after the fire. Stone testified it was unlikely that a petroleum product would remain on a person’s hand for that amount of time. During the habeas hearing, Monroe contradicted this testimony. At trial, Achor elicited testimony concerning when the tests were conducted. He also suggested through examination of Monroe and Knott that the traces of a petroleum product could have had many sources. The only step Achor failed to take was to produce evidence that petroleum products would not remain on an individual’s hands for nine hours.

The final element in Stone’s critique of Monroe’s testimony concerns the order of the tests performed on Knott’s hands. Stone stated that the trace metal detection and gas chromatography tests could not both be accurately performed. He urged if the chromatography test (which revealed the presence of a petroleum product) was performed first, the necessary collection of samples with cotton swabs would disturb the pattern of metal traces. If the trace metal test were performed first, the necessary spraying of the hands with an alcohol solution (before viewing them under ultraviolet light) would introduce a flammable substance which would taint the results of the chromatography test. Although there is some confusion in his testimony, Monroe suggests that he took the samples for the chromatography before he sprayed Knott’s hands with the alcohol solution. He testified that the samples were taken from several spots on Knott’s hands, leaving adequate areas on which metal traces could still be examined through the trace metal test. Because Monroe did not testify that these traces formed a characteristic pattern or a pattern consistent with the carrying of a metal bucket, mere disturbance of the pattern caused by the sample collection was not critical. Even if the trace metal test was performed first, Monroe testified that the alcohol spray would not have dissolved the petroleum product and, because alcohol is an organic compound, it would have been easily distinguished from the petroleum product.

Although petitioner’s trial counsel probably should have increased his knowledge of the relevant scientific techniques and principles by consulting an expert such as Irving Stone or by studying literature in the field, we have difficulty in light of the existing record holding that counsel’s representation was constitutionally inadequate. Human nature is such that most people think they have a better understanding of the demands of an event after it has happened. Trial of law suits is peculiarly susceptible to hindsight appraisal of another lawyer’s endeavors. When trial counsel exercise their judgment in making strategic decisions, third party post-trial construction of strategic alternatives cannot be the sole basis for finding constitutional deficiency. United States v. Weir, 657 F.2d 1005 (8th Cir. 1981); Robinson v. United States, 448 F.2d 1255 (8th Cir. 1971).

Of course, each case must be evaluated individually. Serious dereliction in counsel’s representation might well be established where material witnesses are not called to testify. See Eldridge v. Atkins, 665 F.2d 228 (8th Cir.-1981). For example, if an expert witness could readily verify that “blood” was actually “paint,” counsel might be deficient in failing to pursue such a witness. Cf. Miller v. Pate, 386 U.S. 1, 87 S.Ct. 785, 17 L.Ed.2d 690 (1967) (knowing use of false evidence violates due process clause). Unfortunately, in today’s technological society, experts generally can be found to render “scientific” opinions on either side of any question. Where there is *1213substantial contradiction in a given area of expertise, it may be vital in affording effective representation to a defendant in a criminal case for counsel to elicit expert testimony rebutting the state’s expert testimony. Cf. Williams v. Martin, 618 F.2d 1021 (4th Cir. 1980) (denial of funds to secure pathologist to evaluate cause of death). Counsel’s failure to become versed in a technical subject matter in order to conduct effective cross-examination or failure to properly seek and produce witnesses at trial may constitute a constitutional flaw in the representation of a defendant in a particular case. However, in this case, the evidence does not support such a claim.

Trial counsel testified at the habeas hearing that he decided the best method of persuading the jury in the state trial was to reveal the weaknesses of Monroe’s testimony in layman’s terms. He chose to cross-examine the state’s expert concerning the presentation and meaning of his findings rather than engage him in a dialogue concerning the scientific bases of his techniques. This mode of examination may often be more meaningful and persuasive to lay juries than a technical cross-examination. Many trial lawyers also believe there is danger in affording an expert opportunity to further explain and emphasize his or her direct testimony. In the present case, counsel testified that he deliberately based this choice on these considerations.

In the habeas hearing, Stone, a forensic scientist, testified that the problem with Monroe’s testimony, although he found it technically correct, was that it left him with an incorrect impression. He thought Achor could have asked more effective questions. In making the latter statement, Stone assumed the role of an expert lawyer rather than a forensic scientist. It is not a scientific conclusion to suggest that another lawyer could have done a better job. Achor’s decision not to explore the technical bases of Monroe’s testimony was not contrary to what a reasonably competent attorney might have done in the same situation.

Achor effectively challenged Monroe’s equivocal findings by emphasizing: (1) the metal traces could come from many different environs and their presence did not, as Monroe conceded, prove that Knott had handled the metal bucket; (2) the expert stated he found no “real landmarks” on Knott’s hands; (3) Monroe admitted the petroleum base substance detected could come from many sources and he could not testify it was gasoline or diesel fuel; and finally (4) the state’s expert failed to produce any photographs to allow the jury to draw its own conclusions.

Although all the evidence at trial was circumstantial, there was evidence other than scientific proof. Knott’s former wife lived in the house. Another occupant of the house overheard Knott threaten his wife two days before the fire that he would “burn the damn house down around your ears.” Knott took the stand and told an incredible story. He said that he had been arrested on the 8th of July for being drunk and spent the day in the Little Rock municipal jail. He was released at 4:00 p. m. and then met a stranger with whom he rented a car. The stranger wanted to drive to Pompano Beach, Florida and wanted to know how to get to Memphis. Knott went to Memphis with the stranger in order to do some “drinking.” He testified he was in Memphis at the time of the fire, approximately 3:00 a. m. He said the stranger went to sleep in the back of the car and Knott drove the car back to Little Rock, arriving at 5:00 a. m. He then went to the house of a man whose name he did not know to buy a pint of wine. Knott said he then went back to the interstate highway and sat down and drank some of the wine. He then began walking to a mission when he was arrested at 5:45 a. m. At the time of his arrest, his hair was singed and he had a blister on his ear. His clothing was similar to that worn by an individual seen fleeing the burning house. Achor’s critique of Monroe’s testimony could not have been improved sufficiently to alter the jury’s verdict.

We conclude that counsel’s preparation and cross-examination of the state’s expert falls well within the permissible boundaries of constitutionally adequate representation. It may not have been what other lawyers would have done; it may have been different than what a forensic scientist would *1214have done. Achor would have been better prepared if he had studied the literature or consulted an expert. Nonetheless, we find Achor’s cross-examination of Monroe effectively raised nonincriminating inferences which a jury could draw. We do not believe a more technical analysis would have altered the verdict. To hold otherwise, at least in this case, gives impetus to a form of second guessing which is possible in every case and does not comport with the constitutional standards which require defense counsel to exercise only that skill and diligence possessed by competent counsel under like or similar circumstances.

The order granting the writ is vacated; the cause is remanded with directions to enter an order denying the writ.