Scientific Correspondence

Nature 393, 747-748 (25 June 1998) | doi:10.1038/31612

Deadly relic of the Great War

Caroline Redmond1, Martin J. Pearce1, Richard J. Manchee1 & Bjorn P. Berdal2


The curator of a police museum in Trondheim, Norway, recently discovered in his archive collection a glass bottle containing two irregularly shaped sugar lumps. A small hole had been bored into each of these lumps and a glass capillary tube, sealed at its tip, was embedded in one of the lumps (Fig. 1). A note attached to the exhibit translated as follows: "A piece of sugar containing anthrax bacilli, found in the luggage of Baron Otto Karl von Rosen, when he was apprehended in Karasjok in January 1917, suspected of espionage and sabotage".

The curator noted the mention of anthrax with some alarm and he despatched the exhibit to the Norwegian Defence Microbiological Institute in Oslo which subsequently passed it to the Defence Evaluation Research Agency, Chemical and Biological Defence, Porton Down, UK.

Although 80 years had elapsed since the discovery of the sugar, the legendary durability of spores of Bacillus anthracis gave us reason to hope that some spores had survived, and we decided to try to revive them. To eliminate the chances of contamination of the sample, it was handled in a class 3 microbiological safety cabinet that had been fumigated with formaldehyde vapour and thoroughly cleaned.

We found no difficulty in removing the glass capillary tube from the sugar lump. The capillary was roughly 20 times 2 mm with tapered ends and contained a brown fluid. We opened it by scoring it with an ampoule file and cracking it over a sterile Petri dish. We transferred the liquid that remained in the capillary to a sterile 1.5 ml plastic tube, and estimated the volume of the liquid to be 10-15 mul. Liquid that spilled from the capillary was transferred to a separate sterile tube. The fragments of the capillary were also transferred to a tube and 1 ml of L-broth (tryptone 10 g, yeast extract 5 g, sodium chloride 5 g and water to 1 litre) was added. We also washed the Petri dish twice with 1 ml of L-broth and placed the washings in sterile tubes.

We incubated the washings and the glass fragments under static conditions at 37 °C for 8 days. After incubation, 200 mul of these cultures were spread on 7% horse-blood agar and L-agar medium (identical to L-broth but solidified by the addition of 2% Difco Bacto agar).

We cultured the liquid from the capillary directly onto agar because originally it must have contained a large number of spores to improve the chances of causing infection. We spread 2 times 0.5 mul and 1 times 1 mul volumes of the liquid onto 7% horse-blood agar and L-agar. If successful this would have allowed us to determine the number of surviving spores. We incubated these plate cultures for 24 h at 37 °C, but none of them showed visible B. anthracis colonies.

We therefore resorted to enrichment, which is routinely used to revive apparently moribund bacteria. We added 2 mul of the original capillary suspension to 10 mul of L-broth and incubated the mixture at 37 °C for 8 days. We then spread 5 mul on 7% horse-blood agar to detect visible colonies. Eventually, therefore, all the liquid samples, washings and capillary fragments were subjected to the same enrichment procedure and culture on 7% horse-blood agar. We summarize the results in Table 1. Even after prolonged incubation in recovery media, the numbers of viable organisms that grew on subculture were extremely small, showing that the contents of the capillary were on the very edge of viability.

We carried out McFadyean's test on all the colonies to confirm that they were B. anthracis. In this test a subculture is made from the colony into a few millilitres of blood and the mixture is incubated for 4 h at 37 °C. A smear is then made on a microscope slide, fixed by heat or alcohol and stained with polychrome methylene blue. The rods of B. anthracis stain dark blue and are surrounded by a pink capsule; this confirmatory test provides incontrovertible identification of B. anthracis.

We also analysed the samples by polymerase chain reaction (PCR) as a back-up to the culture procedure. We placed 2 mul of the capillary contents into 10 mul of water in a 1.5 ml sterile Eppendorf tube and stored it overnight at 4 °C before analysis. We carried out separate PCR reactions on 2 mul aliquots of the resuspended sample, using primers specific for an anthrax-specific genomic marker (Ba813), the lethal factor gene (lef) on plasmid pXO1 and the capsule B gene (capB) on plasmid pXO2. PCR reactions were prepared using Ready-To-Go PCR beads (Pharmacia Biotech) in a final volume of 25 mul. For each primer set, positive and negative control reactions were prepared using 10 ng of B. anthracis DNA or no DNA, respectively. A positive PCR containing Ba813 primers was spiked with 2 mul of sample to determine whether any PCR inhibitors were present.

The PCR thermal cycling conditions were: 95 °C for 5 minutes; 35 cycles at 95 °C for 30 seconds each; 55 °C for 30 seconds; and 72 °C for 60 seconds. The final extension time was increased to 10 minutes. We analysed PCR products by agarose gel electrophoresis at 80 volts for 1 hour on a 2% agarose gel containing 1 g ml-1 of ethidium bromide. The gel was viewed over 302-nm ultraviolet light.

Spiking of a positive PCR with sample did not decrease amplification efficiency, showing that no PCR inhibitors were present. For all primer pairs, amplified DNA of the correct size was observed from the sample. This is consistent with the capillary containing pXO1+/pXO2+ B. anthracis.

We therefore confirmed the presence of B. anthracis in the specimen by both culture and PCR. It proved possible to revive a few surviving organisms from the brink of extinction after they had been stored, without any special precautions, for 80 years.

One of the most intriguing questions of this investigation is what Baron Otto Karl Robert von Rosen, a Swedish/German/Finnish aristocrat, and his companions were doing in this remote area of northeastern Norway in the depths of winter; Karasjok is in Finnmark, about 18 kilometres from the Finnish frontier.

When they were arrested, all without passports, the contents of the baron's luggage were both revealing and compromising. When the Sheriff of Kautokeino, who was present at the group's arrest, derisively suggested that he should prepare soup from the contents of the tin cans labelled "Svea kott" (Swedish meat), the baron felt obliged to admit that each can actually contained between 2 and 4 kilograms of dynamite1. Later, after the baron had been extradited from Norway, bottles of curare, "microbial cultures" and 19 sugar cubes, each containing a tiny glass tube holding anthrax microbes, were also identified. It seems, however, that the single specimen that we tested was the only one to have survived to the present day.

The baron claimed that he was an activist for Finnish independence and that their objective was to destroy lines of communication and transport, especially to Russian-controlled areas. According to his companions, the expeditions to these areas were in fact organized by the Germans and there had been several previous visits to Finland with sabotage equipment. Berlin had approved a proposal to use B. anthracis against reindeer used for sledging British arms through northern Norway2.

The Norwegian newspapers at the time were convinced that the contaminated sugar cubes were to be used to disrupt the transport of merchandise by horses and reindeer between Skibotten and the Finnish border. The grinding of the sugar and its glass insert between the molar teeth of horses would probably result in a lethal infection as the anthrax spores entered the body, eventually facilitated through the small lesions produced in the wall of the alimentary tract by the broken glass.

It is not known whether reindeer eat sugar lumps but presumably the baron never had the chance to carry out this piece of research. Even if all the tubes had been used by the sabotage team it is unlikely that they would have caused much disruption to transportation, as anthrax is not transmissible directly from horse to horse. However, if the meat from a dying animal had been consumed without adequate cooking, it is likely that human fatalities from gastrointestinal anthrax would have followed.

At that time Norway's foreign policy was one of neutrality, but the commercial traffic across northern Norway was advantageous to the countries at war with Germany. This covert breach of neutrality, coupled with negotiations in progress in Kristiana (now Oslo) that were to bring Norway closer to the allies, resulted in the massive torpedoing of Norwegian ships in the Arctic Ocean.

After the baron was apprehended he was sent to Kristiana and held in custody for three weeks. He was then released and expelled to Sweden as a result of diplomatic pressure. Unfortunately the complete story was never officially discovered by the Norwegian authorities. It was clear that the baron was involved in acts of sabotage in Finland but the full nature of the contents of the baron's 'hunting equipment' was not revealed until after his expulsion. The disclosure of the dangerous biological agents in his luggage provoked heavy criticism of the Norwegian Ministry of Justice.

This small but relatively important episode in the history of biological warfare is one of the few instances of where there is confirmation of the intent to use a lethal microorganism as a weapon, albeit 80 years after the event. It did not, however, make any significant difference to the course of the Great War.



  1. Berg, R. Norway on its Own, 1905-1920 (Univ. Oslo Press, 1995).
  2. Hugh-Jones, M. WickhamSteedandGermanbiologicalwarfareresearch. Intell. Nat. Security 7, 379–402 (1992).
  1. Detection Department, CBD, Porton Down, Salisbury SP4 0JQ, UK
  2. Norwegian Defence Microbiology Institute, PO Box 4302 Torshov, N-0401 Oslo, Norway