At No. 2 Cannon Street, Sarah Milson kept watch over a commercial warehouse and its keys. On the night of 11 April 1866, she went down to answer the street bell and never returned. The cook found her dead at the foot of the stairs, her skull crushed. Weeks later, a man was brought before the Central Criminal Court, charged with the murder of the woman who had entrusted him with money. The trial would weigh the value of an eyewitness identification when the witness has already seen the suspect paraded before her, and the gravity of an alibi when circumstantial evidence seems to converge.
Arthur Griffiths, who later chronicled the case, termed the evidence against William Smith extremely strong; Montagu Williams, counsel for the defence, left a retrospective account in his Leaves of a Life, published in 1890. Williams had sat through the trial; Griffiths composed his chapter long after the event. Their broad agreement reconstructs the record without converting suspicion into certainty.
Sarah Milson was a widow who had served for several years as housekeeper to Messrs. Bevington and Sons. A man named Kit assisted in securing the premises: he locked the outer doors once the clerks had departed, delivered the keys to Sarah, and kept separate custody of the safe key. In his account of the proceedings, Williams notes that the only nocturnal entrance opened onto Cannon Street. On the evening in question, having completed his rounds, Kit summoned Sarah by means of the speaking tube; she came down, took the keys, and saw him to the street. He observed a light still burning in the vestibule. According to his testimony, the cook and Sarah were the sole persons remaining within the premises.
The bell rang at about ten minutes past nine. The cook, who was in her bedchamber, was preparing to go down when she heard Sarah call out that the visitor was for her. It was the last time she heard her mistress alive. Going down later to fetch a candle, she discovered the body at the foot of the stairs; the head was severely fractured and blood covered the steps. An officer discovered an iron crowbar near the body, bearing no visible bloodstains, which Williams’s narrative nevertheless presents as the lethal instrument. Williams summarizes the depositions without reproducing any medical testimony.
Prior to her death, Sarah had occasionally received visits from a man. The cook had never set eyes upon this caller, though she remembered that Sarah had borrowed two sovereigns shortly before one such visit and later repaid them. A letter discovered in a box belonging to the deceased opened another avenue of inquiry. Signed “George Terry,” it reproached Sarah for having borrowed money under false pretences—the care of an ailing husband—and threatened exposure to her employer or her sister should she refuse the terms proposed. A receipt preserved alongside the letter bore the signature of “W. Denton, for George Terry,” giving an address at 20 Old Change.
Detective John Moss proceeded to Eton with these documents. There, at 6 Eton Square, he found William Smith and his mother. Smith acknowledged the handwriting on the receipt and admitted having drawn up a memorandum for a man. He also confessed to having occasionally used the name of Denton. According to Williams, he stated that he had seen Sarah three times and had delivered receipts to her after receiving payments. On the matter of the sums involved, the testimonies did not entirely agree: George Terry stated that Smith had obtained twelve shillings on one visit, whereas Smith spoke of larger amounts. The paper appeared to link both men to Sarah; it could not, of itself, determine who had rung the bell that night.
The prosecution relied upon a neighbouring witness. Mrs. Robbins, housekeeper of an adjoining warehouse, stated that she had gone out shortly before eight o’clock and returned around ten minutes to ten. As she rang her own bell, the door of Bevington’s slammed with great violence. She saw a man descend the steps, brush past her, and turn his face toward the vestibule lamp. She described a direct exchange of glances—“his left eye and my right eye met at the same moment,” according to the phrase Williams attributes to her deposition—and identified the passer-by as Smith. The identification appeared all the more striking as she claimed to have observed him in ordinary circumstances, in passing, without suspecting the significance her evidence would assume.
Yet the manner in which this identification had been obtained became a central issue at the trial. Inspector Foulger related that he had caused Smith, dressed in ordinary attire and flanked by two plainclothes officers, to walk from Bow Lane to the Mansion House along Cannon Street. Mrs. Robbins had been instructed to observe the party from her doorway; later, she walked through a room where several men were assembled with the prisoner. Williams records that on her second passage through the room, she pointed out the third man. The defence argued on cross-examination that this contrivance had fixed her attention upon Smith prior to the formal identification parade. Proximity to the suspect and the scene of the crime might aid recognition; it might just as easily suggest it.
Other depositions appeared to bolster the Crown’s case. A warehouse porter stated he had seen Smith inquiring after Sarah some months earlier; a woman acquainted with Sarah declared she had likewise seen him upon the premises. Catherine Collins, formerly in service to Mrs. Robbins, affirmed having seen him call repeatedly at the adjoining house. George Terry, then an inmate of the St Olave’s workhouse, testified that he had engaged Smith to recover the debt owed by Sarah. He described how the letter had been drafted in a public house, detailed Smith’s visits to the warehouse, and recounted his return with money that they subsequently divided. Thus, around a debt and a caller, the prosecution forged a chain that purported to lead directly to murder.
Then came the alibi, and with it another journey, far more difficult to reconstruct. Henry Giles, a boatbuilder at Eton, stated, according to Williams, that he had seen Smith in a beerhouse around seven o’clock on either the 11th or 12th of April. Griffiths places this encounter specifically on the night of the 11th, the date of the murder. In Williams’s account, Smith remarked that he had forty miles to travel that night, implying a journey to London and back. Another witness observed him walking toward Slough on the evening of the 11th. The 7:43 p.m. train could, according to the guard cited by Williams, reach Paddington by 8:40; the Metropolitan lines then allowed access to Farringdon Street or Moorgate. A train departed Paddington at 10:45 p.m. for Windsor, arriving at 11:43. The prosecution could thus advance a plausible timeline: Smith might have crossed London, reached Cannon Street, dispatched Sarah, and returned to Eton.
The defence, however, summoned witnesses who placed the prisoner in Berkshire at the crucial hour. A shoemaker for whom he had worked that day, the artisan’s son, a photographer who claimed to have seen him in Windsor between eight and half-past eight, three men with whom he had played cards, the publican’s wife who had drawn him a glass of ale after eleven o’clock, and his two sisters were called to the stand. Williams writes that these depositions, precise in their reckoning of time, threw the prosecution’s theory into utter disarray. He knew the district well, having resided in the very country upon which the alibi rested.
The trial took place before the Central Criminal Court on 13 and 14 June 1866, Baron Bramwell presiding. After the learned judge’s summing-up and the deliberation of the jury, William Smith was found not guilty and discharged. This verdict resolved the question of his guilt as laid before the court; it did not reveal the true author of the crime. Williams stated that he knew of no subsequent discovery that ever shed light upon the deed. Griffiths cited the outcome as a classic instance of an escape from an unjust conviction; yet the prisoner was not declared innocent by virtue of the real culprit being brought to book: he was acquitted, and the murder remained without a known author.
Griffiths records a curious postscript, which must not be confused with substantive evidence against Smith. While serving as governor of the convict prison at Gibraltar in December 1869, two convicts successively confessed to the Cannon Street murder. The first offered a circumstantial narrative; the second adopted the identical particulars. Griffiths writes that an inquiry by the Home Office dismissed both statements as fabrications, as both men were already incarcerated at Gibraltar when the crime was committed. This retrospective anecdote solves nothing: it merely demonstrates how the name of an unsolved murder attracts those who seek, whether for fleeting notoriety or calculated design, to claim it for their own.
Justice had cleared William Smith; it had not illuminated the fate of Sarah Milson. At Cannon Street, a bell had sounded, a door had opened, and the visitor could never be identified with certainty. Between Terry’s letter, Mrs. Robbins’s gaze, the railway timetables, and the witnesses of Eton, every piece of evidence seemed to pull first in one direction, then in another. There remained a categorical verdict and an unbroken mystery: the verdict set Smith free without naming the slayer, while the dead woman was left without the single truth the law had failed to secure.