Arsenic, Toxicology, and the Problem of Science in the Courtroom

Contemporary debates about expert witnesses and scientific evidence in courtrooms are nothing new.¹ 150 years ago people—scientists, lawyers, jurists, and journalists—raised questions about when to rely on expert witnesses, how much weight should be given to their testimony, and how to evaluate the results of their scientific tests alongside other forms of evidence. In 1874, an article in the NY Times noted:

The “affaire Moreau,” (see Times Sept. 25) presenting as it does, a vast array of chemical evidence, naturally awakens reflections as to the testimony of experts. Some days ago allusion was made in the columns of this paper to the difficulty attending the reception of this species of evidence, from the juryman’s point of view; but he is far from being the only one who is at times embarrassed when examinations of this kind are entered upon. Although a lover of his profession, satisfied as to the competency of his process, and sure of the accuracy of the statements which he makes, the man of science is sometimes cruelly wrung when he feels that upon that competency and accuracy there hangs a fellow-creature’s life; that a word from him has power to save or kill. This was never more fully exemplified than on the trial of Mm. Lafarge, which in many particulars resembles the recent case in Paris.²

The “affaire Moreau” referred to a recent case in Paris in which careful chemical analysis had helped convict the herbalist, Moreau, of poisoning his two wives. The NY Times article ends by celebrating medical science: “But the medical science of to-day can force to the scaffold the subtlest of criminal that ever handled poison.”³ While the NY Times seemed to endorse the use of science in the recent Moreau case, it was skeptical of the scientific testimony in the Lafarge case.

In 1840 Marie Lafarge was tried for murder, accused of having poisoned her husband with arsenic. The core details of the case were widely accepted:

Marie was unhappy in her marriage to Charles Lafarge. Nevertheless, outwardly, she seem to display affection and care for her husband, writing intimate letters to him and helping him secure funds and assistance in his business dealings. In December 1839 Charles traveled to Paris on business. While there, Marie sent him a letter and some cakes. After eating only a small piece of the cake, he almost immediately became violently ill. Although he didn’t consult a physician, he recovered sufficiently to return home. Shortly after he returned, his condition worsened and within a few days he died. Marie’s mother-in-law accused her of poisoning Charles. When Marie’s room was searched, they found arsenic.

Based on this evidence, Marie was arrested in January 1840. She was tried in September, ultimately found guilty of murder, and sentenced to life in prison. The case hinged on expert testimony and the use of a new scientific technique: the Marsh process to detect arsenic.⁴

The prosecution’s case rested largely on the post-mortem examination performed by four local physicians. They found that some of the food and drink Marie had given her husband contained “arsenious acid,” that the bile vomited up from his stomach did not contain any, but that the liquids in his stomach did. This evidence seemed to indicate that Marie was guilty. Her defense countered by reading a letter sent by Mathieu Orfila, a famous Parisian chemist, who disputed the results of the initial post-mortem examination because the physicians did not carry out sufficiently specific tests. In particular, they stopped when the observed a yellow precipitate normally associated with arsenic. Orfila claimed that this was inconclusive because it could have been produced by “some animal matter normally present in the bile.”

At this point, according to the NY Times, the court called on two local chemists—MM. Dubois and M. Dupuytren—to conduct further tests. They used the recently developed Marsh’s process to test the original substances taken from Charles Lafarge’s body and found no trace of arsenic. The court was not satisfied with these results—owning in part to the contradictory nature of the two tests—and requested that Dubois and Dupuytren perform the same tests on material taken directly from Charles’s body (according to the NY Times):

On the morning of Sept. 8 the operation commenced. After removing the earth (the body had been buried over seven months) the coffin was discovered, which contained a body in a hideous state of decomposition. This human paste was put in earthenware pots and carried into the court-room. Six small stoves were placed in a circle, heated red hot, and incense thrown on them from time to time, but this was insufficient to control the putrid odor which penetrated the room. In the centre of this fiery circle the chemists continued their hideous task, in the presence of the court, and of such of the spectators as had stomachs strong enough to witness the show. On the 9th, Dupuytren, for himself and his colleagues, reported that after the most searching tests, they could not detect a trace of arsenic in the remains.⁵

Although these results seemed to exonerate Marie Lafarge, the court called for yet another round of tests. This time the court looked to Paris for its expert chemists. Of the chemists contacted, only Mathieu Orfila agreed undertake the tests. Orfila came to the court with two assistants—his lab assistant De Bussey and “a M. Ollivier, who was a doctor rather than a chemist. In reality there was only one expert, but even the defense acknowledged that he was the prince of his science—the Hermes Trismegistus of toxicologists.”⁶ When Orfila completed his tests he announced that he had found arsenic in the body. Further, he assured the court that the arsenic came neither from the reagents he used nor the soil in which the body had been buried. The court considered Orfila’s testimony authoritative; the jury deliberated for an hour before pronouncing its verdict and sentencing Marie Lafarge to life in prison.

The NY Times was clearly skeptical of the court’s reliance on Orfila as the sole expert witness and, further, doubted the skills of M. Ollivier. The NY Times also seems to have been concerned by Orfila’s procedure, which took place through the night and in private—in clear contrast to the public tests performed by Dubois and Dupuytren. The NY Times also raised questions about how much arsenic Orfila had detected. In a similar case involving arsenic poisioning, one of Orfila’s colleagues was asked to compare the amount of arsenic found in the later case with that found in Charles Lafarge’s body. The chemist replied that he couldn’t answer because the amount found by Orfila was too small to admit of comparison: “It was so infinitesimal that it could not fulfill the condition of a standard of coparison, when we use the words ‘more’ or ‘less.’” Finally, the NY Times questioned the purity of Orfila’s reagents and his, and by extension the court’s, use of scientific evidence.

Orfila had performed Marsh’s process three times. The first two, using reagents purchased locally, produced no appreciable trace of arsenic. The third time, using reagents reagents brought from Paris by Orfila’s assistant, produced noticeable deposits of arsenic. After the tests, Orfila carefully gathered up all his reagents and apparatus and took them back to Paris with him. At no point were the reagents from Paris tested for purity from arsenic, as the local reagents had been. As the NY Times pointed out, given the unusual nature of Orfila’s single positive test, it should have been rejected as evidence.

Concerns about the evidence and Orfila’s testimony had arisen within days of the trail. The second issue of the British Medical Journal, published 10 October 1840, contained a report on the Lafarge trail. It focused on Orfila’s tests and testimony. The BMJ recounted the initial negative tests and then summarized Orfila’s positive results. Given Orfila’s care in testing the soil around the coffin and non-digestive parts of the corpse, the BMJ concluded that “This report, taken in conjunction with the symptoms observed during the illness of M. Lafarge, will, we think, convince the most cautious medical jurist that the unfortunate man must have been poisoned by arsenic.”⁷ But the BMJ pointed to some uncertainties in medicine—such as the arsenic leeching out of the bones during putrefaction or why Orfila found no arsenic in the stomach but did in the “visceral mass.” Despite these questions, the BMJ supported Orfila’s results and even explained away the negative results of previous test, at one point suggesting that Dubois and Dupuytren were inexperienced and pointing the difficulty in using Marsh’s apparatus: “their want of success depended upon their having employed in their operations insufficient quantities of animal matter, and upon their inexperience in the use of Marsh’s apparatus, which in unpracticed hands may fail to separate the metal from the fluids containing it.”⁸ Having looked closely at the medical evidence, the latter half the article reviewed the non-medical evidence. It concluded by affirming the medical evidence and condemning the French legal system: “That M. Lafarge perished from the effects of arsenic cannot be questioned, but, after having deliberately weighed the arguments for and against conviction, we do not hesitate to say, that the evidence advanced against the prisoner is not, by any means, sufficiently strong to warrant a verdict of guilty.”⁹ The final paragraph rails at the French system: “Really, a French trial may not inaptly be compared to a French fox-hunt, or a Spanish bull-bait, in which the animals pursued (or rather worried) are allowed no chance of escape!”

A few weeks later, again in the pages of the BMJ, another article lambasts the French system, calling it theatrical and indecent. The article raised a series of questions:

first, whether there were any evidence of the existence of arsenic in the body of M. Lafarge; secondly, whether this arsenic was other than the portion which is known to exist naturally in certain parts of the human body; thirdly, whether it could have been derived from any other source; fourthly whether it had been administered during life; and, lastly, whether it might not have been so administered inadvertently, i.e., by some other means than with the intention or knowledge of the accused.¹⁰

Dismissing the first and second concerns, the article dwells on the third: whether the arsenic came from a different source. The article claimed that iron oxide was often contaminated with arsenic, thus casting doubt on Orfila’s results. Further, the article calls for more tests to determine the quantity of arsenic commonly found in iron oxides in order to limit possible errors in future cases. While stopping short of rejecting Orfila’s results, this second article in the BMJ clearly questioned them.

The Lafarge trail attracted attention beyond the medical literature. In July 1842, the Edinburgh Review ran a long article detailing the case and raising numerous objections to the proceedings. The Edinburgh Review pulled no punches in its condemnation of the trial:

Now then we ask, who is there, who being a juryman, would from such evidence as this come to these two distinct affirmative conclusions—

1. That Lafarge did die, poisoned by arsenic.

2. That his wife knowingly administered that arsenic?

In must be recollected … we have been compelled to omit many things which require consideration by any one who would fairly estimate the value of the French system of procedure. … Looking back through the whole evidence, carefully weighing each separate item adduced, trying its worth by every test which the experience of ages has suggested, we are satisfied that there was not sufficient evidence to prove that the deceased came to a violent end; still less to show that his wife was the guilty cause of his death.¹¹

By the turn of the century, the Lafarge trial had assumed a place in the canon of poisoning mysteries. A series of articles in McClure’s Magazine by Marie Belloc Lowndes recounted mysterious French crimes. Her first article reviewed the Larfarge trial: “Great French Mysteries. The Strange Case of Marie Lafarge.” Lowndes had no intention of reviewing the evidence, except insofar as it offered her a chance to make the case into something of romantic mystery.¹²

A slightly more sober recounting of the Lafarge trial was offered in C. J. S. Thompson’s Poison Mysteries in History, Romance and Crime (PHiladelphia, 1924). Thompson reviews the salient aspects of the case, injects just enough mystery and romance, and then concludes: “That Lafarge died from the effects of arsenical poisoning there seems little doubt, but by whom administered has never ben conclusively proved, and the tragedy remains among the poison mysteries still unsolved.”¹³

The early concerns about evidence, expert witnesses, and toxicology seem to have been replaced by romantic stories about a woman wronged. Most accounts seem to have decided that Marie Lafarge was wrongly convicted of a crime she did not commit. However, at least the medical literature defended the use of toxicology in the courtroom, and pointed to the success of Orfila’s use of Marsh’s procedure to detect arsenic. Even later popular literature accepted the validity of the science, the expert witness—in this case, the Hermes Trimegistus of chemistry—while doubting the non-scientific evidence.

Notes—

¹For a recent report on the use of science in court, see the article in Wired: “Brain Scan Evidence Rejected by Brooklyn Court”.⇑

²“Madame Lafarge. Her Trial for Alleged Poisoning of her Husband,” NY Times (8 November 1874).⇑

³“The Moreau Affair,” NY Times (25 September 1874).⇑

⁴Recent articles on toxicology look to this case as the beginning of forensic toxicology. See, for example, Richard A. Pizzi, “Pointing to Poison,” Today’s Chemist at Work (September 2004): 43–45, and B. Mile, “Chemistry in Court,” Chromatographia 62(2005): 3–9.⇑

⁵“Madame Lafarge. Her Trial for Alleged Poisoning of her Husband,” NY Times (8 November 1874).⇑

⁶“Madame Lafarge. Her Trial for Alleged Poisoning of her Husband,” NY Times (8 November 1874).⇑

⁷Maritn H. Lynch, “Analysis of Madame Lafarge’s Trial, with Remarks on the Medical Evidence” British Medical Journal 1, no 2 (10 October 1840), 17.⇑

⁸Maritn H. Lynch, “Analysis of Madame Lafarge’s Trial, with Remarks on the Medical Evidence” British Medical Journal 1, no 2 (10 October 1840), 18.⇑

⁹Maritn H. Lynch, “Analysis of Madame Lafarge’s Trial, with Remarks on the Medical Evidence” British Medical Journal 1, no 2 (10 October 1840), 19.⇑

¹⁰“Poisoning with Arsenic” in “Provincial Medical & Surgical Journal” published in the British Medical Journal (19 December 1840), 202.⇑

¹¹“Trial of Madame Lafarge” Edinburgh Review (July 1842): 359–396, quotation from 396.⇑

²Marie Belloc Lowndes, “Great French Mysteries. The Strange Case of Marie Lafarge” McClure’s Magazine 38(April 1912): 603–618.⇑

¹³C. J. S. Thompson’s Poison Mysteries in History, Romance and Crime (Philadelphia, 1924), 278.⇑