In July 1924, Harry Houdini joined tests of the Boston medium Mina Crandon, publicly known as Margery. An electric bell rang, a cabinet moved and a megaphone fell in a darkened séance room. Houdini said he detected ordinary bodily movements behind the effects. J. Malcolm Bird, an investigator and supporter of the medium, disputed his account and published records which he believed undermined the exposure. The resulting disagreement concerns observations, controls and the difference between a possible trick and a demonstrated one.
The principal sources here are two participants arguing against each other: Houdini’s 1924 pamphlet and Bird’s book, first published in 1925. Neither should be treated as a neutral transcript of the whole controversy. Reading them together reveals important agreements about the sittings as well as sharp disputes about what happened in the darkness. This article concentrates on the July tests rather than combining all of Margery’s later mediumship into one undifferentiated claim.
The prize and the investigators
Houdini’s Houdini Exposes the Tricks Used by the Boston Medium “Margery” describes the Scientific American investigation and its prize for convincing psychic phenomena. He lists William McDougall, Daniel F. Comstock, Walter Franklin Prince, Hereward Carrington and himself as committee members, with Bird acting as secretary. O. D. Munn, the magazine’s proprietor, also attended the Boston sittings. The presence of several investigators did not mean that they reached one conclusion.
Houdini believed Bird’s magazine reports had encouraged the impression that the medium was effectively approved before the full committee had decided. Bird’s book rejects the story of Houdini arriving just in time to prevent an imminent award. He describes a subcommittee still examining the case and only one colleague prepared to give a definite favourable verdict. This is a concrete disagreement about the investigation’s status, not merely about whether an impressive performance occurred.
The bell box on 23 July
In Houdini’s account of 23 July, he sat to Margery’s left, holding her left hand and maintaining ankle contact. Her husband sat on her other side, while Bird controlled the link between their hands. A box containing an electric bell and batteries could be activated by pressure on its top. The explanation offered in the sittings was that Walter, identified as Margery’s deceased brother, operated it. That identification belongs to the mediumistic claim.
Houdini said he prepared by making his lower leg unusually sensitive, then rolled up his trouser leg to detect movement. With the box between his feet, he claimed to feel Margery’s ankle gradually shift while maintaining enough contact to reassure an ordinary observer. He then felt muscular movements which he associated with her foot pressing the apparatus. His conclusion was that contact had concealed movement rather than prevented it.
The claim is more specific than saying the bell could be rung by a foot. Houdini asserted that he felt the manoeuvre during the sitting. It remains his account of what he detected in darkness, however, rather than a film or a direct view of the foot striking the device. That distinction became central when Bird compared the later exposure with the sitting records.
Bird’s records and the disputed control
Chapter XLVIII of Bird’s “Margery” the Medium reproduces records of 23 and 24 July and explains his objections to Houdini. The 23 July record describes interruptions when apparatus was moved, the positions of hands and feet, an illuminated plaque and the bell sounding. Bird says the record was revised before acceptance and presents it with signatures including Houdini’s and Munn’s. This is the record as published by Bird; the original signed sheet has not been independently examined here.
Bird argued that the published photograph of the positions did not support the foot movement Houdini described. He also objected that Houdini’s explanatory drawing changed positions and proportions. Those objections are claims about the quality of an exposure: whether its proposed mechanism actually fits the arrangement under which the original effect occurred. A convincing demonstration under different conditions cannot automatically settle that question.
The two writers also meant different things by control. Houdini described allowing a manoeuvre to continue while he detected it. Bird expected control to prevent the suspected manoeuvre and treated assurances that it was maintained as evidence against fraud. A record saying that control was satisfactory therefore cannot be interpreted without asking what the person making that statement believed he was doing.
The cabinet and megaphone
Another 23 July episode involved Bird leaving his position to obtain an illuminated plaque. Houdini proposed that the interruption freed Margery’s right hand and foot. He said she could prepare the cabinet for a push and put the megaphone on her head, then restore apparent hand and foot control before flicking the megaphone towards him. His pamphlet describes the cabinet falling and the megaphone arriving near his feet.
Bird’s response is particularly informative because it is not a blanket denial of physical possibility. On page 420 he accepts that the cabinet-and-megaphone sequence could have been accomplished as Houdini proposed during that sitting. He disputes presenting the explanation as a directly detected act, rather than a working hypothesis. He also blames the lapse in control and argues that the conditions differed from earlier sittings which other investigators considered stronger.
This creates a narrow area of agreement: the conditions of that particular episode left room for an ordinary mechanism. It does not resolve whether the mechanism was used. Conversely, a claim that earlier sittings had better control does not repair the evidential gap in the episode under discussion. Each performance has to be assessed under its own recorded conditions.
The second sitting and the table
The 24 July sitting took place in Comstock’s hotel apartment. Houdini said Margery used her head and shoulders to move a table while her wrists remained on the chair arms and her hands were held. He described exploring beneath the table with a free hand, finding her head near its edge and quietly telling Munn what he believed he had detected. His pamphlet also describes a repeat of the bell experiment.
Bird printed a timed record of table movements and renewed assurances about the hand and foot contacts. He rejected Houdini’s table explanation, saying his own free hand was on Margery’s shoulder or in her lap, and that Munn’s exploration had found nothing incriminating. He acknowledged that those particular observations were not in the sitting record. The dispute thus includes retrospective testimony on both sides, alongside a contemporary record whose completeness the writers themselves debated.
A photograph of participants posed after a sitting can clarify a seating arrangement, but it is not a photograph of every position throughout a dark performance. Similarly, a signature agreeing that described events occurred does not necessarily endorse another participant’s explanation of them. Both points are necessary when these sources are reproduced as apparently decisive proof for either side.
An award withheld, a controversy continuing
The Society for Psychical Research’s modern source survey records the committee’s verdict on 12 February 1925: insufficient evidence that the phenomena were supernormal, with no prize awarded. It distinguishes that outcome from an explicit collective finding of fraud and notes Carrington’s favourable position. The survey also describes later investigations with differing assessments. Those later episodes are not used here as retroactive confirmation of what either man felt on 23 or 24 July.
The July tests remain useful because the disagreements can be located precisely. The bell’s operation was disputed; a lapse in cabinet control was acknowledged as allowing an ordinary explanation; the table episode rested on conflicting tactile observations. Houdini’s proposed methods challenged the paranormal interpretation, while Bird’s published rebuttal challenged the finality of Houdini’s exposure. Neither a famous magician’s confidence nor a supportive investigator’s assurance substitutes for an independently adequate record.
Sources and related reading
Read Houdini’s original pamphlet and Bird’s book, especially chapter XLVIII, pages 404–424, for the opposing accounts. The archive copy of Bird is a London “Sundial Edition”, printed in the United States; its OCR contains errors. A PBS transcription also reproduces Houdini’s July account. The SPR survey supplies the later verdict and bibliography. The image is a posed historical group, not a captured supernatural manifestation.
Compare the Fox sisters’ rappings and competing explanations for another investigation in which testimony, physical possibilities and later interpretation have to be separated. The Boston tests are presented as historical research. No private residential address or inferred séance-room coordinates are provided.
Map location — approximate area: The pin uses Boston municipal area. It is a rounded locality reference, not the private séance room. Coordinate reference: U.S. Census Bureau 2025 Gazetteer representative area point. It gives no directions to private property.