Christopher Jarecki appeals from the judgment of sentence entered against him in the Court of Common Pleas of Berks County in this criminal action. Appellant raises the following issues for our review: (1) whether the evidence was sufficient to support appellant’s convictions; (2) whether the trial court erred in denying appellant’s motion to suppress identification evidence; and (3) whether the verdicts were against the weight of the evidence. Having thoroughly reviewed the record and considered the arguments of the parties, we vacate and remand for a new trial.
The facts of the case are as follows: On October 3, 1989, a robbery occurred at the Weis Market in the Muhlenberg Shopping Center, Berks County. Linda Young, a store employee, was counting money in the office area of the store when the perpetrator approached her from behind. Ms. Young was ordered, at gunpoint, to give the robber all of the money in the office drawers and safe. A second employee, Elsie Boone, entered the office at that time. The suspect turned, pointed the gun at Ms. Boone, and fled the office. Several other employees observed the robber as he exited the building and made his escape. Three male employees pursued the robber to a nearby Ground Round restaurant parking lot. The three men observed a gray or silver Pontiac Trans-Am speeding away from the area. Approximately six hundred dollars in cash was subsequently discovered in the restaurant’s parking lot.
*288The witnesses described the suspect as a Caucasian male in his late twenties to early thirties, approximately 5'9" to 6' tall, with a medium build and light brown hair. He wore a blue zippered jacket, blue slacks, a white shirt, and a “Giants” baseball cap. In addition, a blue scarf covered his face so that only his eyes and the underside of his chin were visible. The witnesses also noted that the suspect ran with a limp.
One of the witnesses, Joe Giandomenico, who had pursued the robber, claimed that he saw the robber’s face without the scarf. He also claimed that the robber wore dark glasses. No other witnesses, including the two witnesses who pursued the robber with Mr. Giandomenico, saw the suspect’s entire face or claimed that the robber wore glasses. Mr. Giandomenico was also the only witness who saw the suspect enter the Trans-Am.
During the ensuing investigation, Mr. Giandomenico assisted the police in making a composite drawing of the suspect. The composite was published in a local newspaper on October 16, 1989. The newspaper article also described the suspect and the car in which he allegedly escaped. Police received two subsequent anonymous phone calls alleging that appellant fit the newspaper’s description and had access to a gray Trans-Am.
On November 13,1989, approximately six weeks after the robbery, Officer Barry Hadley took a photo display book containing seventy-five photographs to the . Weis Market. Officer Hadley had incorporated appellant’s photograph into the display. The witnesses perused the photo display collectively, and four witnesses, including Joe Giandomenico, selected appellant’s photo from the display.
Appellant was arrested and charged with robbery and related offenses. Appellant filed an Omnibus Pretrial Motion for Relief on February 15, 1990, requesting the suppression of the witnesses' identifications. A hearing was held on April 3, 1990, and the Honorable Calvin E. Smith denied appellant’s suppression motion in an opinion and order dated May 2, 1990. Following a jury trial, appellant *289was convicted of robbery (two counts), theft by unlawful taking, and terroristic threats. Post-verdict motions were filed and denied, and Judge Smith sentenced appellant to a total of five to ten years incarceration, plus fines.1 This timely appeal followed.
Appellant asserts in his first issue that the evidence was insufficient to support his convictions. Our review of appellant’s claim is limited to the following well-established standard:
In reviewing the sufficiency of the evidence, we must review the evidence presented and all reasonable inferences taken therefrom in the light most favorable to the Commonwealth, as verdict winner. The test is whether the evidence, thus viewed, is sufficient to prove guilt beyond a reasonable doubt. The credibility of witnesses and the weight to be accorded the evidence produced are matters within the province of the trier of fact; the fact finder is free to believe all, some, or none of the evidence.
Commonwealth v. Taylor, 324 Pa.Super. 420, 424-425, 471 A.2d 1228, 1229-1230 (1984) (citations omitted). In addition,
the sufficiency of the evidence must be evaluated upon the entire trial record. All of the evidence must be read in the light most favorable to the Commonwealth and it is entitled to all reasonable inferences therefrom. The effect of such a motion is to admit all the facts which the Commonwealth’s evidence tends to prove. Also, in'passing upon such a motion, all evidence actually received must be considered, whether the trial rulings therein were right or wrong.
Commonwealth v. Maldonado, 343 Pa.Super. 154, 158, 494 A.2d 402, 404 (1985) (emphasis in original) (citations omitted).
*290After careful consideration, we conclude that the evidence of record was sufficient to sustain appellant’s convictions. Four eyewitnesses unequivocally identified appellant at trial. In addition, appellant had access to a gray Trans-Am. On the face of the record, the evidence is sufficient and we will not arrest judgment on appellant’s convictions.
Although the evidence of record is sufficient to support appellant’s convictions, a new trial is warranted if the convictions are based on inadmissible evidence. Commonwealth v. Maybee, 429 Pa. 222, 226, 239 A.2d 332, 335 (1968). Appellant, in his second issue, asserts that the trial court erred in denying his motion to suppress the photographic identification evidence. After careful review, we conclude that the trial court erred in denying appellant’s suppression motion, and we hold that a new trial is warranted.
In reaching our determination, we have applied the following standard:
The scope of our review is limited to determining whether the factual findings of the court in the suppression hearing are supported by the record, and whether the conclusions of law drawn therefrom are correct.
Commonwealth v. Scaine, 337 Pa.Super. 72, 77, 486 A.2d 486, 488 (1984) (citing Commonwealth v. Webb, 491 Pa. 329, 332, 421 A.2d 161, 163 (1980)). “We may reverse the denial of a motion to suppress only if the [suppression] court erred in the legal conclusions drawn from its factual findings.” Commonwealth v. Blassingale, 398 Pa.Super. 379, 391, 581 A.2d 183, 188 (1990).
Although the suppression court found that several witnesses made their photographic identifications at the same time, the court determined that the identification procedure was not suggestive because the police officer cautioned the witnesses not to discuss the photographs. A review of the record, however, reveals that the witnesses did not heed the officer’s request. Rather, the witnesses discussed the photographic display in an attempt to select the robber. Given *291the circumstances of this case, and the manner in which the identifications took place, misidentification was highly likely. It is well established that evidence will be suppressed if, under the totality of the circumstances, the identification procedure was “so impermissibly suggestive as to give rise to a very substantial likelihood of irreparable misidentification.” Commonwealth v. Monroe, 373 Pa.Super. 618, 622-23, 542 A.2d 113, 114-15 (1988), appeal denied, 522 Pa. 574, 559 A.2d 36 (1989) (quoting Simmons v. United States, 390 U.S. 377, 384, 88 S.Ct. 967, 971, 19 L.Ed.2d 1247 (1968)). Our review of the record convinces us that the circumstances surrounding the photo identification procedure make the accuracy of the identifications highly suspect. Accordingly, we conclude that the trial court erred in failing to suppress evidence of the photographic identifications at trial.
The suppression court relied on Brown v. Blackburn, 625 F.2d 35 (5th Cir.1980), as support for its conclusion that the photo identification procedure was not suggestive. We note that even if Brown supported the suppression court’s decision, we would not be bound by the decision of another jurisdiction.2 However, a review of Brown shows that Broivn actually supports suppression in this matter. The Fifth Circuit court recognized the potential for error in a situation where two witnesses viewed a photographic display at the same time. The court specifically noted that the witnesses should have viewed the pictures separately to mitigate the potential suggestiveness of the situation, even *292though no discussion took place and the second witness did not see which photograph the first witness selected.
We find that the facts of the instant case are exponentially more egregious than the facts in Brown, as the suggestiveness of the photographic identification procedure is evident from the record. Officer Hadley took the photographic display into the store and showed it to the witnesses approximately six weeks after the robbery, and at least four witnesses viewed the display at the same time. All but one of the four identifying witnesses had seen only the eyes, chin, and hair of the robber, and none of them had described the robber’s eyes as distinctive prior to the photographic identification. Furthermore, the witnesses had access to the published composite sketch prior to the identification proceeding. Joe Giandomenico, the witness who claimed to see the robber’s entire face and on whose description the composite sketch was based, was the first to select a photograph from the display. The other witnesses had the opportunity to observe Mr. Giandomenico select a photograph, thus allowing the other witnesses to infer that the robber was pictured in the display. As Mr. Giandomenico claimed to have seen the robber’s entire face and his description was the basis for the composite drawing, it is likely that the others searched the display for a person who matched the composite sketch and Mr. Giandomenico’s description, instead of relying on their own independent recollections.
In addition, the testimony at the omnibus pretrial hearing shows that the witnesses discussed the photographs as they perused the display. Officer Hadley testified that Joe Giandomenico, Linda Young, Agnes Morris, and Elsie Boone were present at the photographic display, and that the three women watched Mr. Giandomenico page through the book. (Notes of Testimony, 4/3/90, pp. 23-24.) Officer Hadley stated that he asked the witnesses not to discuss the display among themselves, but the record shows that discussion did take place among the witnesses.
*293The testimony of Linda Young and Agnes Morris is especially revealing. Linda Young testified that the three women watched Joe Giandomenico select appellant’s photo. She also stated that they discussed the photographs among themselves prior to selecting appellant’s photograph. One portion of her testimony at the pretrial hearing went as follows:
[Mr. Hracho]. You guys talk about [the photo display] if you saw somebody in there?
[Ms. Young]. A little bit. We looked at the pictures, we just said he’s not the one or he’s not the one and we got— I myself got it down to two and then I looked at it some more and I picked him out.
Q. Fair to say then that you and the rest of the people, you were looking at it at the same time, more or less going through the pages?
A. Yes.
Q. Okay. Anybody say anything like, does that guy look familiar? What about this guy’s hair, or something like that?
A. I would say so. We didn’t talk much. We just went back and forth, you know, just, you know, looking at them.
(N.T., 4/3/90, p. 67.) In addition, Agnes Morris gave the following testimony after she claimed that the parties discussed the photographs among themselves:
[Mr. Stevens]. Who was discussing the pictures?
[Agnes Morris], Linda and Elsie.
Q. What were they saying, do you remember?
A. I don’t remember the words.
Q. Okay. Did you participate in that discussion with them?
A. Yes, cause I was trying to pick out—
Q. Do you remember what you said, if anything, to anybody?
A. No, I don’t remember.
*294Q. Okay. Now, did you at that time or at any time after that identify anybody in those photographs?
A. Yes.
Q. And when was that?
A. Within a couple minutes.
Q. Were they all there together when you did that? A. Yes.
Q. You mean Elsie and Agnes were still there.
A. Elsie and Linda.
Q. Whoever it was.
A. Yes.
Q. And what did you say.
A. I don’t remember.
Q. Well, how—
A. I remember pointing to a picture, saying those are the eyes.
Q. Yes? Who did you say that to?
A. To general [sic], to whoever who was there [sic]. Q. The two women and Officer Hadley.
A. Yes.
(N.T., 4/3/90, pp. 131-132.)3
In light of the above , testimony, and other testimony of record, we are persuaded that the photo identification procedure was highly suggestive and that any reference to the photographic identifications should have been suppressed at trial. Of the witnesses who selected appellant’s photograph from the display, only Joe Giandomenico claimed to have seen the robber’s face. The other three witnesses saw only the robber’s eyes and the underside of his chin. Although all three of these women testified that their identification at *295the photographic display was based on appellant’s distinctive eyes, none of these women claimed that the robber’s eyes were distinctive at the time of their initial description, nor did they know their color or shape. The photo identification took place six weeks after the robbery, the witnesses saw only a small portion of appellant’s face, and the witnesses were not unequivocal in their identifications at the time of the photographic display.4 The women observed Mr. Giandomenico select a photograph, and they discussed the photographs among themselves as they attempted to pick out the robber. Due to the suggestiveness of the photographic procedure, the likelihood of misidentification was extremely high.
In addition to the suggestive nature of the collective photo identification procedure, we also note that the photographic display itself was suggestive. Even a correctly conducted photographic display can lead to misidentification if a photograph is emphasized in some way. Simmons v. United States, supra. In the instant case, we have reviewed the seventy-five photographs that were contained in the display, and we observed that the background in appellant’s photograph varied from the background of the other photographs in the display. Also, the other photographs were primarily head shots, while appellant’s photograph was taken from a further distance and showed his entire torso. Finally, the date was prominently displayed in appellant’s photograph, but not in any of the other photographs. These differences emphasized appellant’s photograph, and the differences increased the risk of misidentification.
The witness testimony at appellant’s trial includes references to the photographic display. As noted, that was error. Furthermore, we cannot conclude that the error is harmless, as the jury verdict may have been affected by evidence of the photographic identifications. Accordingly, *296we hold that any evidence or testimony which refers to the photographic display or identifications must be suppressed on remand.
Although we conclude that the photographic identification procedure was unduly suggestive and that reference to it at trial constituted prejudicial error, that fact does not necessarily undermine the reliability of a subsequent in-court identification at appellant’s retrial. The Pennsylvania Supreme Court discussed the reliability of identifications in Commonwealth v. McGaghey, 510 Pa. 225, 507 A.2d 357 (1986), and noted the factors to consider in determining whether the suggestive procedure taints the reliability of a subsequent in-court identification. The Court stated:
The problem with an impermissible suggestive identification is the potential for misidentification, resulting in a due process violation if that identification is admitted at trial. Suggestiveness alone will not forbid the use of an identification, if the reliability of a subsequent identification can be sustained. To do so, the Commonwealth must establish that the in-court identification resulted from the criminal act and not the suggestive encounter.
Most recently, in Commonwealth v. James, 506 Pa. 526, 486 A.2d 376 (1985), we reiterated the necessary factors in determining whether a victim had an independent basis for an in-court identification:
... the opportunity of the witness to view the criminal at the time of the crime, the witness’ degree of attention, the accuracy of the witness’ prior description of the criminal, the level of certainty demonstrated by the witness at the confrontation, and the length of time between the crime and the confrontation.
Commonwealth v. McGaghey, supra, 510 Pa. at 228, 507 A.2d at 359 (citations omitted) (emphasis added). The trial court can determine that an in-court identification is admissible if, under the totality of the circumstances, the trial court finds “that the in-court identification had an independent origin ‘sufficiently distinguishable to be purged of the *297primary taint’.” Commonwealth v. Buehl, 510 Pa. 363, 385, 508 A.2d 1167, 1178 (1986), quoting Commonwealth v. Ransome, 485 Pa. 490, 497, 402 A.2d 1379, 1383. In the instant case, identifications that the trial court determines have an independent basis, untainted by the suggestive photographic display, may be admissible.
We note that the trial court determined in its findings of fact and conclusions of law that Linda Young formed the basis of her identification at the time of the robbery, and therefore she had an independent basis for her identification. We also note that Linda Young did not refer to the photographic display on direct examination.5 As the trial court did not find that the photographic display was suggestive, the court never reached the issue of whether the suggestive display irreparably tainted any subsequent in-court identifications by Joe Giandomenico, Agnes Morris, or Elsie Boone. Once any reference to the photographic display is removed from the testimony of these three witnesses, the determination of whether any or all of these witnesses have an independent and untainted basis for an in-court identification in any subsequent proceeding is a question for the finder of fact to determine on remand.
For the foregoing reasons, we vacate and remand for a new trial at which any reference to the photographic display shall be suppressed.6
*298Judgment of sentence vacated. Remanded for further proceedings consistent with this determination. Jurisdiction relinquished.
. We note that the record does not contain a trial court opinion in support of the denial of appellant's motions for a new trial and/or in arrest of judgment, as required by Pa.R.A.P.1925. However, we have determined that the record has provided us with adequate information to conduct an effective appellate review of the issues.
. Appellant has directed us to several other federal cases which address the suggestiveness of a mutual photo identification. (Appellant’s brief at 13.) Although we are not bound by these determinations, we do note that these cases are consistent with our determination that the photographic identification procedure in the instant case was suggestive. See U.S. v. Bagley, 765 F.2d 836, (9th Cir.1985), cert. denied, 475 U.S. 1023, 106 S.Ct. 1215, 89 L.Ed.2d 326 (1986) (joint confrontation is a disapproved identification procedure; better procedure is to keep witnesses apart when they view photographic displays); see also Rudd v. State of Florida, 477 F.2d 805 (5th Cir.1973); accord Brown v. Blackburn, 625 F.2d 35 (5th Cir.1980) (collective photographic displays can be mutually reinforcing and must be evaluated on a case by case basis).
. We also note that Mr. Hracho asked her the following question on cross-examination:
Q. But the day of this photo line-up you and Elsie and Linda sure standing around, let’s face it, everyone is saying, what about this guy’s hair, this kind of thing? Is that what was going on? You’re talking about it as you were going through the pages?
(N.T., 4/3/90, p. 134.) Her affirmative answer further supports our finding that the discussions between the witnesses contributed to the suggestiveness of the identification procedure.
. In fact, each of the three female witnesses stated the relative certainty with which she selected appellant’s photograph:
Linda Young— 60% certain
Agnes Morris— 65% certain
Elsie Boone— 70% certain
. In fact, Linda Young was not asked about the photographic display until counsel for appellant questioned her on cross-examination. However, appellant’s earlier motion to suppress the photographic identifications had been denied. In addition, we note that Officer Hadley testified about the photographic display prior to Linda Young’s testimony. Accordingly, it was arguably good defense strategy for counsel to attempt to undermine Linda Young’s credibility by revealing the uncertainty of her photographic identification and the potentially suggestive nature of the collective photographic display.
. Appellant also asserts that the verdicts are against the weight of the evidence. Our review of a challenge to the weight of the evidence is limited to the following standard:
A motion for a new trial on grounds that the verdict is contrary to the weight of the evidence concedes that there is sufficient evidence to sustain the verdict but contends, nevertheless, that the verdict is against the weight of the evidence. Whether a new trial should be granted on grounds that the verdict is against the weight of the *298evidence is addressed to the sound discretion of the trial judge, and his decision will not be reversed on appeal unless there has been an abuse of discretion. The test is not whether the court would have decided the case in the same way but whether the verdict is so contrary to the evidence as to make the award of a new trial imperative so that right may be given another opportunity to prevail.
Commonwealth v. Taylor, 324 Pa.Super. 420, 425, 471 A.2d 1228, 1230 (1984) (citations omitted). Given our determination that appellant’s second issue warrants a new trial, we need not address the merits of appellant’s weight of the evidence claim. However, we note that given our standard of review, we are unable to conclude that the verdict at trial was against the weight of the evidence presented to the jury in light of the record we have before us.