concurring specially:

The two Smiths, who are not brothers, were convicted for armed robbery and sentenced to 25 years in prison based, almost entirely, on the identification testimony of one eyewitness, Patricia Campbell. The only controverted issue at trial was the identification of the perpetrators of the robbery. The eyewitness identification was not corroborated by any physical evidence: There were no fingerprints, no surveillance camera photographs, no traceable cash or money orders, and no items of apparel worn by the robbers which were offered in evidence.

Throughout the trial, the robber who pointed a hand gun and shouted “Don’t anybody move,” was referred to as “robber number one.” His companion, who vaulted the service counter and picked up the cash, wore a mask that covered the bottom half of his face. The companion was called “robber number two” during the trial. Ms. Campbell identified Roderick Smith as robber number one, and she identified the masked robber number two as Henry Smith.

Immediately after the robbery, Ms. Campbell described robber number one to a police officer as a black Negro male in his late twenties, five foot eight to five foot ten, broad shoulders, medium build, dark complexion, medium “natural” and clean shaven. About one month after the robbery, Ms. Campbell was shown some photographs by a postal inspector. She picked out the photograph of Roderick as the person who held the gun. Shortly thereafter, she described the gun-wielding robber as broad and stocky, not fat, and kind of dark skinned. Ms. Campbell also made an in-court identification of Roderick Smith as the man whose photograph she had earlier selected. In court, Ms. Campbell modified somewhat her prior description of robber number one by stating that she could not tell the length of his hair because he was wearing a cap. At the time of his arrest and at trial, Roderick Smith had a shaved head and was wearing a goatee. Her opportunity to observe the robbers was very brief, the conditions were exciting, and Ms. Campbell testified that during these events she was “surprised, more or less shocked.” The only evidence offered to corroborate Ms. Campbell’s identification of the armed robber as Roderick was testimony of Betty Gaither, an eyewitness, who testified that she was “not sure” that Roderick was the first robber, and who could not identify the second robber at all. The second corroborating witness was Mr. Frugoli, who also testified that he could not positively identify the man, because the only resemblance he could see between the robber and Roderick was that his “nose is similar to the one I saw [on robber number one].”

On the same date that Ms. Campbell selected Roderick’s picture from a photographic spread, she also selected Henry’s picture as the second robber. She was more hesitant in her identification of Henry than she was of Roderick, although she likewise made an in-court identification of Henry. Her ability to see and remember robber number two must have been impaired by the fact that he wore a scarf or handkerchief concealing his face from the bridge of his nose down. No corroboration of any kind was made of Ms. Campbell’s identification of Henry.

Ms. Campbell remained firm in her identification testimony of both men, and it is evident, even from the cold transcript, that Ms. Campbell was a convincing witness whose honesty was unquestioned.

I concur in the affirmance of these convictions because the majority opinion’s observation that we “still cling to the notion *1365 that the testimony of one witness, if solidly believed, is sufficient to prove the identity of a perpetrator of a crime,” is an accurate statement of the current law. (See, e. g., United States v. Telfaire (1972) 152 U.S.App.D.C. 146 , 469 F.2d 552 ; United States v. Levi (4th Cir. 1968) 405 F.2d 380 ; 7 Wigmore, Evidence §§ 2030-34.) I concur specially because the premise of the rule is that such eyewitness identifications are generally reliable and that premise is, at best, highly dubious, given the extensive empirical evidence that eyewitness identifications are not reliable. (See, e. g., P. Wall, Eye-Witness Identification in Criminal Cases (1965); E. Borchard, Convicting the Innocent: Errors of Criminal Justice (1932); Note, Did Your Eyes Deceive You? Expert Psychological Testimony on the Unrealistic of Eyewitness Identification, 29 Stan.L.Rev. 969 (1977). See also United States v. Wade (1967) 388 U.S. 218 , 228 n. 6, 87 S.Ct. 1926 , 18 L.Ed.2d 1149 (citing Wall, Borchard and other legal and psychological studies of the identification problem); United States v. Brown (1972) 149 U.S.App.D.C. 43, 54-55 , 461 F.2d 134, 145-46, n. 1 (Bazelon, C. J., dissenting) (same).)

As the Supreme Court observed in United States v. Wade, supra, 388 U.S. 218 , 87 S.Ct. 1926 , 18 L.Ed.2d 1149 :

The firmness of eyewitness identification, as often as not, is an indicium of unreliability. As Professor Borchard has observed, “[t]he positiveness of witnesses is sometimes ... in inverse ratio to their opportunity for knowledge or to their reliability.” (Borchard, supra, at 50.) Even when an initial pretrial identification is tentative and unsure, “Once an opinion is formed, other emotions, such as pride and stubbornness, make for confirmation of the original identification rather than for open-minded reconsideration.” (Id. at 261.)

Misidentifications are rarely the product of conscious falsehood. Rather, these mistakes are the result of “[t]he normal and universal fallibilities of human sense perception and human memory” as well as of the susceptibility of the human mind to suggestive influences. (Wall, supra, at 9.) As Professor Wigmore observed:

Although the present state of the law compels my concurrence in the affirmance of these convictions, I cannot conscientiously concur in the affirmance of Henry Smith’s conviction without suggesting the need for reconsideration of the one witness principle in cases, like his, in which the eyewitness identification is totally uncor *1366 roborated by any Other witness or any other evidence.

. The Court’s observations in Wade are fully supported by the empirical studies about the reliability of eyewitness observations. As Professor Borchard points out in his study of 65 convictions of completely innocent people, “The major source of these tragic errors is an identification of the accused by the victim of a crime of violence. This mistake was practically alone responsible for twenty nine of these convictions.” Borchard, supra, at xiii. See also Wall, supra at 187.