OPINION

FRANSON, Acting P.J.

Appellant was found guilty by a jury of destroying jail property in violation of Penal Code section 4600, arson in violation of Penal Code section 451 and rioting in violation of Penal Code section 404. The jury also found that appellant intentionally damaged and destroyed property of a value exceeding $100,000 as provided by Penal Code section 12022.6, subdivision (b).

After denying a motion for a new trial based on juror misconduct allegations, the trial court sentenced appellant to prison for a term of eight years.

For the reasons to be explained, we hold that appellant's conviction must be reversed for juror misconduct which prejudiced appellant's right to a fair trial.

THE FACTS

During the evening twilight hours of June 3, 1984, a riot occurred at the minimum security section of the Kern County Lerdo jail facility. The minimum security section was fenced and surrounded by a common area containing about 620 prisoners.

The rioting continued for several hours and involved 100 to 250 inmates, some of whom set fire to a maintenance building and other structures resulting in a loss exceeding $242,000.

Correctional Officers Johnson, Davis and Narvaez were standing in the "fishbowl," a part of the jail office with observation windows, approximately 50 to 100 yards from the maintenance building. Each officer used binoculars in an effort to identify individual inmates participating in the rioting and burning of the buildings. Many of the inmates were wearing towels or T-shirts wrapped around their heads and could not be identified. Officer Davis observed one inmate toss a burning mop inside the maintenance building as well as two other inmates throw burning T-shirts through the window of the maintenance building. Davis was unable to identify any of the inmates who threw burning materials into the maintenance building. Although Davis knew appellant prior to June 3, he did not identify appellant during the evening in question.

Officer Narvaez was unable to identify any of the inmates participating in the rioting.

However, Officer Johnson, who was also using binoculars, testified that he recognized appellant as the individual who tossed a burning mop into the maintenance building. Johnson said he also observed appellant throw a burning shirt on the roof of the maintenance building. Johnson recognized appellant from an encounter approximately two weeks earlier when Johnson had disciplined appellant for violating a facility rule.

Appellant denied that he was involved in the rioting or in the burning of the buildings. Other prisoners corroborated his testimony.

Thus, Officer Johnson's identification of appellant through the binoculars as the inmate who threw the burning materials into and onto the maintenance building was the critical evidence supporting appellant's conviction.

DISCUSSION

Appellant's motion for a new trial based on juror misconduct included a declaration of Juror Dooley that during jury deliberations Dooley "went home and used binoculars to see if a witness could have possibly seen what he [Officer Johnson] said he did. After using the binoculars I took that information back to the deliberations of the jury the next day." (1a) The only reasonable inference to be drawn from Dooley's declaration is that he conducted his own at-home experiment to determine if Officer Johnson could have identified appellant as the culprit who threw the burning mop into the maintenance building at the distances and in the light established by the evidence, and after concluding that the officer could have so identified appellant, reported his findings to the other jurors during deliberations. Thus, the jury received evidence outside of the courtroom which appellant was unable to meet or answer thereby establishing juror misconduct. ( Higgins v. L.A. Gas Electric Co. (1911) 159 Cal. 651, 656-657 [ 115 P. 313]; see 7 Witkin, Cal. Procedure (3d ed. 1985) Trial, §§ 307, 308, pp. 306-308.)

(2) Even if Dooley did not actually advise the jury of his findings, he tainted his own deliberations thereby violating appellant's right to 12 impartial jurors. ( People v. Pierce (1979) 24 Cal.3d 199, 208 [ 155 Cal.Rptr. 657, 595 P.2d 91].)

The standard for evaluating the effect of out-of-court juror experiments was articulated by our Supreme Court 90 years ago in People v. Conkling (1896) 111 Cal. 616 [44 P. 314]. There, two jurors, to satisfy themselves at what distance a rifle discharge would powder mark cloth, procured a rifle out of the courtroom and experimented with it. (3a) "Here was a clear case of the jury's obtaining evidence by unauthorized experiments made without the presence and knowledge of the defendant." ( Higgins v. L.A. Gas Electric Co., supra, 159 Cal. at p. 659.) The Conkling opinion explained: "The distance between the deceased and defendant at the time the fatal shot was fired was a vital issue in the case. The clothing worn by the deceased was in evidence, and when exhibited to the jury showed no powder marks. . . . [The defendant's] affidavits . . . [showed] . . . that during the progress of the [trial] two of the jurors borrowed a rifle similar to that with which the deceased was killed, bought some cotton drilling, retired to the outskirts of the city, and there made experiments by firing the rifle, for the purpose of determining at what distance powder marks would be carried by the fire. . . . [The jurors] were evidently honest, and desirous of getting at the truth of the matter; but they were too zealous, and their misconduct in this particular demands a retrial of the case. Jurors cannot be permitted to investigate the case outside the courtroom. They must decide the guilt or the innocence of the defendant upon the evidence introduced at the trial. It is impossible for this court to say that this outside investigation did not affect the result as to the character of the verdict rendered. For, when misconduct of jurors is shown, it is presumed to be injurious to defendant, unless the contrary appears." (111 Cal. at pp. 627-628, italics added.)

Although there is no showing in the present case that Dooley's binoculars were "similar" to the binoculars used by Officer Johnson (seven powered) or that the light conditions and distances used at the time of Dooley's personal experiment were similar to the conditions at the time Officer Johnson identified appellant, Dooley's experiment nevertheless had the same effect as the Conkling juror's experiment. It enabled Dooley to receive evidence outside the presence and knowledge of appellant going to the crucial element in the prosecution's case, the identity of the appellant. (Cf. People v. Phillips (1981) 122 Cal.App.3d 69 [ 175 Cal.Rptr. 703].) Thus, "[i]t is impossible for this court to say" that Dooley's experiment did not affect the jury verdict. ( People v. Conkling, supra, 111 Cal. at p. 628.)

People v. Cooper (1979) 95 Cal.App.3d 844 [ 157 Cal.Rptr. 348] and Locksley v. Ungureanu (1986) 178 Cal.App.3d 457 [ 223 Cal.Rptr. 737] cited in the dissent are distinguishable on their facts. In Cooper, a juror's affidavit stated that during the trial the juror had driven her car slowly by some pedestrians and "`estimated whether or not a person seated in a moving car could make the observations testified to by the police officers.'" The officers had testified that driving at two to three miles per hour they had observed the defendant, who was walking along the same street, reach toward his waistband and throw a shiny object away. The juror then "concluded" that the officers' testimony was credible. The reviewing court held that the juror's mental processes of "estimating" distances and "concluding" that the officers were credible could not be considered because they were in no way objectively verifiable and therefore did not constitute misconduct. (95 Cal.App.3d at p. 852.)

Locksley v. Ungureanu, supra, 178 Cal.App.3d 457 involved a negligence action by a pedestrian who was struck by defendant's automobile while crossing a street at an intersection with no crosswalk. The defendant driver was blind in one eye. The juror's affidavit stated that on "`[t]he day before we found for the defense, I drove home and closed my left eye once or twice to see what my view would be like. [¶] I [also] drove from my residence . . . to [the] Courthouse, with my left eye closed completely with my left eyelid.'" ( Id., at p. 460.) The reviewing court noted that the juror's acts of twice driving his car while covering one of his eyes was an objectively verifiable act but then observed "it is not at all clear" that the juror's conduct was impermissible since the subject matter did not invade a new field but was merely an experiment on an issue within the evidence, to wit, the ability of a one-eyed individual to drive. Nevertheless, the court then said "assuming arguendo that there was misconduct, it must be noted that not every minor infraction of the rules by a juror mandates a new trial. If the misconduct is of such a non-probative nature that it could not in the nature of things have prevented a party from having a fair trial, the verdict should not be set aside." ( Id., at pp. 461-462.) The reviewing court could not find that the trial court had abused its discretion in finding that the experiment could not have prejudicially affected the outcome of the case particularly after the jury had deliberated for four and one-half hours.

Thus Cooper and Locksley do not aid in the resolution of this case. Riding in a slow moving automobile to determine whether the actions of a nearby pedestrian could be observed and driving an automobile with one eye closed to determine the visual acuity of a one-eyed driver do not produce results which can be objectively corroborated. (4) Further, any verifiable facts produced by the Cooper and Locksley experiments are so commonly understood by anyone who drives an automobile or has two working eyes and the ability to close one eye that it rationally can be said that the experiments produced nothing new by way of evidence for the jury. For this reason, the experiments can be characterized as " nonmisconduct" or "harmless misconduct" whichever phrase best suits the fancy of the reviewing court. On the other hand, a juror's observations through a particular powered set of binoculars at a particular distance under particular light conditions is not a matter of common knowledge. It produces a result which is beyond the ken of the other jurors even though they may have had some general familiarity with binoculars.

The problem of jurors procuring new evidence by their own efforts has long troubled the courts. We note that jurors are instructed in civil cases that they are forbidden to make any independent investigation of the facts of the case. (BAJI No. 1.00.5 (7th ed. 1986).) The instruction tells the jurors that they must decide all questions of fact from the evidence received during trial. "This means . . . that you must not on your own visit the scene, conduct experiments, or consult reference works for additional information." (Italics added.) This instruction ideally should be given at the outset of the trial. A violation of this instruction is described by a learned commentator as "gross" misconduct. (7 Witkin, Cal. Procedure, op. cit. supra, § 308, p. 307.) An instruction similar to BAJI No. 1.00.5 has been adopted for use in criminal cases. (CALJIC No. 1.03 (4th ed. 1986 pocket pt.) p. 2.) Unfortunately, no similar instruction was given in the present case.

(3b) Because of Dooley's misconduct, a rebuttable presumption of prejudice arose. ( People v. Pierce, supra, 24 Cal.3d 199, 207; People v. Honeycutt (1977) 20 Cal.3d 150, 156 [ 141 Cal.Rptr. 698, 570 P.2d 1050].) As stated in Hasson v. Ford Motor Co. (1982) 32 Cal.3d 388, 417 [ 185 Cal.Rptr. 654, 650 P.2d 1171], the presumption may be rebutted by "an affirmative evidentiary showing that prejudice does not exist or by a reviewing court's examination of the entire record to determine whether there is a reasonable probability of actual harm to the complaining party resulting from the misconduct." Because the record before us discloses no evidence whatsoever by the prosecution to rebut the presumption of prejudice arising from Dooley's misconduct, the only way to affirm this case would be to find the misconduct harmless on an examination of the entire record. This we cannot do.

Respondent contends that because the verdict forms finding appellant guilty of arson and rioting were dated September 5, 1984, Dooley's misconduct on the evening of September 5, 1984, could not possibly have affected the deliberations. (5) This contention will not fly. The verdicts were not returned until September 6. Until the verdicts were actually returned, the jury could have reconsidered and changed its verdicts during its deliberations on September 6. As explained in Magee v. Superior Court (1973) 34 Cal.App.3d 201 at page 218 [ 109 Cal.Rptr. 758] (overruled on other grounds in People v. Norris (1985) 40 Cal.3d 51, 56 [ 219 Cal.Rptr. 7, 706 P.2d 1141]): "A verdict is not a verdict unless rendered in open court and until such event occurs even though the jury might in its deliberations have come to a tentative view on what verdict it might bring, until it has ended its deliberations and stated its position in [open] court, there is plenty of opportunity for the jury to change its view. . . ." Thus, the inference that the jury dated its verdicts September 5, 1984 (the verdicts conceivably could have been misdated), does not rebut the presumption of prejudice.

(3c) People v. Martinez (1978) 82 Cal.App.3d 1 [ 147 Cal.Rptr. 208] notes that because a fair trial includes the right to an unbiased jury, the question whether a defendant has been injured by jury misconduct in receiving evidence outside of court necessarily depends upon (1) whether the jury's impartiality has been adversely affected; (2) whether the prosecution's burden of proof has been lightened, and (3) whether any asserted defense has been contradicted. "If the answer to any of these questions is in the affirmative, the defendant has been prejudiced and the conviction must be reversed." (82 Cal.App.3d at p. 22, italics added.)

Officer Johnson's testimony that he was able to identify the appellant as the person who threw the burning mop into the maintenance building at dusk from a distance of 75 to 100 yards by the use of binoculars was the only evidence of appellant's guilt. Officers Davis and Narvaez who also had observed someone toss the burning mop inside the maintenance building were unable to identify appellant as the culprit. Davis also used binoculars at the same time as Officer Johnson in an attempt to identify individual inmates participating in the rioting in the yard. Both Davis and Johnson testified that the inmate who threw the mop did not have anything around his head to conceal his identity yet Officer Johnson identified appellant, and Officer Davis was unable to identify appellant.

(1b) Applying the Martinez standard of review, it becomes obvious that juror Dooley's experiment affected his own impartiality, it lessened the prosecution's burden of proof, and it contradicted appellant's defense that he was not the inmate who threw the burning mop into the maintenance building.

The very integrity of the jury deliberative process is at stake here. A juror conducted his own experiment at home which inferentially affected his verdict thereby creating a rebuttable presumption of prejudice. Because the prosecution failed to rebut the presumption and because the record does not permit a finding of harmless error, the judgment of conviction must be reversed.

The judgment is reversed.

Best, J., concurred.