Mary Castle’s account of bell ringing at Stapleton Grove, Bristol, is notable for a qualification she chose to include. She described wires moving and bells sounding during an afternoon in July 1836, but explicitly declined to vouch for a claim made about the wires overnight. That boundary between her own observation and what she had been told is essential to understanding the case.
The disturbance was brief. Bells were taken down for the night, restored the following morning and, according to Castle’s later letter, did not trouble the household again. The evidence therefore concerns a particular episode, rather than a documented long-running haunting. It does not establish why the bells rang or why the activity ceased.
A letter written in 1841
The surviving first-person account is a letter dated 8 May 1841, written by Mary Castle to Mrs Shawe. Edward Moor included it in Bealings Bells. Castle was the wife of Michael Hinton Castle, associated with Stapleton Grove. Her letter places the incident in July 1836, about five years before she wrote.
That makes it a recollection by a participant, not a daily entry from the afternoon itself. It is still much closer to the household than a modern retelling, and its internal distinctions deserve to be preserved. The incident year is 1836; neither the letter’s date nor the publication of Moor’s book creates a separate 1841 disturbance at Stapleton.
The text also gives the case a clear geographical identity. This is Stapleton on the Bristol side of the historical county geography, not a similarly named village elsewhere. The map uses a broad Stapleton locality reference. The church photograph is a modern landmark view and should not be mistaken for Castle’s home or the room through which the wires passed.
The afternoon sequence
Castle reported that a sitting-room bell began ringing one afternoon. Within about half an hour, nearly all the household bells were involved. They rang singly and in groups of three or four. She described the descending wires moving as though they were being pulled. This is a claim about a domestic bell installation, whose normal operation already involved mechanical movement along wires.
A bellhanger examined the system without discovering a cause, according to the letter. The activity subsided in daylight and resumed around dusk. Castle’s narrative does not provide a clocked series of measurements, but it does distinguish an interruption from a simple assertion of continuous ringing. It also records an attempt to obtain an ordinary tradesman’s assessment.
A sliding part of the installation on a wall attracted attention. Castle went into the room above the passage and watched it shaking and rattling for about five minutes. The arrangement included a simple wire passing through the floor. This observation is one of the most concrete details in her account because it locates movement in a component of the bell system, rather than merely placing the observer somewhere within hearing distance.
The account does not establish a complete view of the wire from one end to the other. Nor does it identify a force acting on the fitting. A component appearing to move without an observed pull is a reason to inspect the installation; it is not, by itself, proof that every ordinary point of access or source of movement was excluded.
Removing and restoring the bells
After the renewed disturbance had continued for about an hour, Castle asked for all the bells to be removed. The household then had a quiet night. When the bells were put back the next morning, there was no recurrence reported. By the time of her letter in May 1841, she knew of no repetition.
Removing the bells prevented their normal audible ringing, so the quiet night cannot be treated as a direct test of whether the wires were still moving. Castle made precisely that distinction in relation to the servants’ account. They said the wires continued to move during the night, but she did not certify this from her own observation and allowed that it might have been imagination.
That is an unusually useful piece of source criticism supplied by the witness herself. It would be misleading to merge the servants’ unverified overnight statement with Castle’s afternoon observation and then attribute both to her. It would be equally misleading to suppress her caution because an overnight continuation makes a more dramatic story.
What the household thought
Castle said the bells had been fitted with new copper wires about twelve months earlier. She also described the servants being together and the house being thoroughly searched. These were reasons she offered for thinking a straightforward prank unlikely. They are statements within her retrospective account, rather than independently recorded search procedures.
The reported age and material of the wires may be useful to a mechanical inquiry, but neither supplies a diagnosis on its own. The letter does not record a fault found in a mounting, a test of electrical conditions or a reproduced failure. The fact that a system had been renewed a year earlier neither proves it faultless nor identifies a defect.
Castle and her husband considered electricity an explanation. That suggestion belongs to their interpretation of the event. The account provides no measured electrical activity or demonstrated way in which it operated this installation. A modern summary should therefore describe their idea as a suggestion, rather than announcing that electricity explained the ringing.
What remains uncertain
There is enough detail to reconstruct the broad sequence: an initial sitting-room bell, more bells joining it, examination, a pause, renewed activity at dusk, removal and quiet, followed by restoration without recurrence. There is not enough to reconstruct the complete physical system or every person’s position during each occurrence.
The brief duration also matters. Once the episode ended, there was no repeated event on which to improve the observations. An unexplained end is no more diagnostic than an unexplained beginning. The absence of recurrence in Castle’s recollection does not establish that taking down and replacing the bells removed a particular fault, although that remains a relevant possibility for the physical history of the installation.
Assessment and location
The classification is Insufficient evidence. Castle’s letter records a specific disturbance and a practical response, while carefully limiting an overnight claim she did not witness. It does not demonstrate a physical cause, deliberate interference or paranormal agency. The household’s electrical suggestion remains untested in the published account.
The map pin marks approximate Stapleton locality near the photographed parish church. It does not identify Stapleton Grove, an exact historical room or a surviving bell system, and it grants no access to private property. No public-place designation is attached to the household setting.
Sources and further reading
The principal source is Mary Castle’s letter of 8 May 1841, reproduced in Edward Moor, Bealings Bells (1841), pp. 95–99. Her account is read here as a first-person recollection of July 1836, with the servants’ overnight statement explicitly kept separate. Other bell cases in the volume concern different houses; their reported features should not be imported into the Stapleton account. Source checked 13 November 2026.