Vera Renczi appears in almost every encyclopedia of female serial killers, usually with a photograph, a count of thirty-five, and a cellar in Romania lined with zinc coffins in which she sat of an evening with her dead husbands and lovers. This archive carries her with a victim count of none, because Romanian and Serbian researchers have looked for the trial, the arrest, the birth registration and the town, and found no trace of any of them. There is no court file, no police record, and no local press coverage of a mass exhumation of thirty-five men, which is not the sort of event a country's newspapers miss. What can be traced is the story's arrival in English, in the sensational American and British magazine press of the late 1920s and 1930s, in the genre of the exotic foreign murderess, and its passage from there into the true crime encyclopedias of the 1990s, which cite each other.
She is the reason this page is worth rewriting rather than repeating. An international list of female poisoners is the place in this archive where the gap between what a woman was convicted of and what she is credited with runs widest, and where the numbers most often come from somewhere other than a court.
Set out what the record actually holds for these eight cases and the shape is unmistakable.
Helene Jegado worked as a cook in houses, presbyteries and inns across Brittany from 1833, and people around her died. In one three-month period seven died in the household of the priest at Guern, including Jegado's own sister. She moved on and it happened again, at Bubry, at Locmine, at Auray, at Rennes. She nursed the sick devotedly, wept at every funeral, and said more than once in front of witnesses that wherever she went people died. She used arsenic, which was in every farmhouse in France as rat poison and which produced symptoms indistinguishable from the cholera then moving through the region. That is why she lasted eighteen years. She was tried at Rennes in December 1851 on three counts of murder and three of attempted murder, everything before 1841 barred by prescription. Twenty-three deaths were laid out in the indictment as evidence of system and estimates of the total run to thirty-six. She was convicted and guillotined. Three counts, twenty-three deaths described, thirty-six estimated.
Maria Catherina Swanenburg was known to everyone in the Zijloord district of Leiden as Goeie Mie, Good Mie, because she was always ready to sit with the sick and help with a funeral. Between 1880 and 1883 she poisoned more than ninety people with arsenic and at least twenty-seven died, including both her parents. She was the beneficiary of burial insurance policies she had taken out on them, sometimes several on one person, for sums of a few guilders each, which in that quarter was worth killing for. She was caught in December 1883 when a family she had been feeding all fell ill at once and a doctor tested the food. She was convicted of three murders, which was all the evidence would carry, and died in prison in 1915. She is the most prolific killer in Dutch history, on a conviction record of three.
Martha Needle killed her husband and their three daughters with Rough on Rats, a commercial arsenic preparation, between 1885 and 1891. She wept at every funeral, tended every grave, and was widely pitied. In 1894 she was keeping house for two brothers in Melbourne and engaged to one of them; the other objected to the marriage, asked about the deaths in her family, and shortly afterwards fell ill. He recovered when he went away, relapsed when he came back, and died. The surviving brother's doctor arranged for him to drink, under observation, a cup of tea she had prepared. He was violently ill, the tea was analysed, and it was full of arsenic. The bodies of her husband and children were exhumed and all contained arsenic. She was tried for one murder only, convicted, and hanged at the Old Melbourne Gaol in October 1894 at thirty-one. What she gained across nine years was a few hundred pounds of insurance.
Amelia Sach ran a lying-in home in Finchley that took in the babies of unmarried women for a fee. Her advertisements offered accommodation for confinement and adoption of the child for twenty-five to thirty pounds, a large sum and more than the cost of keeping a baby. The infants were passed to Annie Walters, who gave them chlorodyne, a medicine containing morphine, and disposed of the bodies. Walters was caught in November 1902 carrying a dead infant. Both women were convicted of the murder of one child, the one Walters had been observed with, and were hanged side by side at Holloway in February 1903, the first executions carried out there. How many babies passed through the house over two years was never established. Contemporary estimates ran into dozens, and dozens is an estimate, not a finding.
Magdalena Solis was a prostitute from Monterrey, hired in 1962 by two brothers who had told a poor village in the mountains of Nuevo Leon that they were prophets of the Inca gods. Their promised apparitions had failed to appear, so they paid her to supply one; she came out of a cave in smoke and was presented as a goddess reincarnated. She then took over the cult. Villagers who lost faith or tried to leave were denounced and killed in the cave, beaten and burned and cut, and their blood collected and drunk. It ended in May 1963 when a fourteen-year-old boy walked fourteen kilometres to the police and said he had seen people killed. The officer who went back with him was killed there too, along with the boy. Solis and her brother were convicted of those two murders, the two that could be proved, and the earlier ritual deaths were never individually charged. That is why this archive gives her range as two to eight or more, and why she is filed as a described case rather than a confirmed one.
Jeanne Weber is the hardest of them, because in her case the state twice looked at the evidence and said there was nothing there. In March and April 1905 four children died in her care in one Paris building: two nieces, a sister-in-law's daughter, and her own son. She had been alone with each of them and each had marks on the throat. At her trial in 1906 the case turned on Leon Thoinot, one of the most eminent forensic physicians in France, who examined the bodies and testified that the deaths were natural and the marks were not strangulation. She was acquitted to public acclaim, and the press treated her as a woman wronged by suspicion.
She killed again. In 1907 a nine-year-old boy died while she was working for a family in the Indre under an assumed name; Thoinot was consulted and the case collapsed a second time. In May 1908 at Commercy she was found in a room with the strangled body of a seven-year-old boy, with the mother and a doctor at the door. Even then Thoinot did not withdraw. She was declared insane that October and died in an asylum in 1918. So the count attached to her, up to ten, sits against a record in which most of the deaths were formally found to be natural, and no murder conviction stands. The case is taught in the history of forensic medicine as the clearest instance of an expert's authority outweighing the evidence in front of him, and children died in the gap.
Beverley Allitt is the exception, and it is worth saying why. Her count is not disputed. In fifty-nine days on a short-staffed children's ward at Grantham in 1991 she attacked thirteen children; four died and nine survived, several with lasting brain injury. She used insulin, potassium, air injected into the bloodstream and suffocation, she raised the alarm each time, and she was present at every resuscitation. The break was a laboratory result: a surviving twin found to have an impossibly high insulin level with no corresponding C-peptide, which cannot happen naturally and means insulin was injected. A missing ward logbook and an allocation record putting her on duty at every collapse did the rest. She was convicted in May 1993 of four murders, three attempted murders and six counts of causing grievous bodily harm.
What is contested in her case is not the number but the explanation. She was diagnosed with Munchausen syndrome by proxy, a contested category, and has been detained in a high-security hospital rather than a prison ever since. She has never given a full account of why. The Clothier Inquiry reported in 1994 on how she came to be employed at all, given a sickness record that should have been examined, and recommended screening nursing applicants for the pattern; its recommendations were widely described afterwards as unimplemented.
Read together, the eight cases say something the individual entries do not. One of them appears to be fictitious and one, Allitt, has a count nobody disputes. In the remaining six the charge sheet covers a fraction of what is attributed, and in Weber’s case nothing at all now stands. In every instance the reason is evidentiary rather than lenient. Jegado's early deaths were barred by prescription. Swanenburg's twenty-seven were beyond what the evidence would carry. Needle was tried for the one poisoning a doctor had watched happen. Sach was convicted over the single infant a witness had seen in Walters's arms. Solis was convicted of the two killings with a body and an officer attached to them. The higher figures are not lies, but they are descriptions of a pattern rather than findings against a person.
Arsenic runs through the three nineteenth-century poisoning cases and explains their arithmetic. It was sold openly as rat poison, it was in every kitchen, and it killed with the symptoms of the gastric illness and cholera then killing people constantly in districts with no sanitation. Jegado was a cook. Swanenburg nursed the sick and laid out the dead. Needle was a wife and mother. A poisoner in that world did not need to be clever. She needed to be trusted, and to be in the house.
The around-the-world framing is what makes the numbers worse rather than better. A French, Dutch, Australian or Romanian case reaches an English-language list through a chain of retellings, and at each step the figure is easier to carry than the file it came from. Nobody rechecks a Romanian court record before repeating thirty-five. The same archive contains the reverse case, a Mexico City killer whose file is solid and whose profile the police got wrong, so the problem is not that foreign cases are unreliable. It is that distance removes the thing that would settle them.
That is why Vera Renczi is carried here with a count of none rather than dropped. Deleting her would leave the story circulating with no correction attached to it. If a document ever turns up, the entry says plainly that it should be rewritten. Until one does, the honest position is that a woman who appears in almost every encyclopedia of this kind probably never existed at all.