Issue Number Three: Sentinel Devices

Sentinel Organisms: “they look out for the environment!”

Under what conditions can animals and plants be considered as good sentinels for an environment? Christelle Gramaglia looks at the uses of shells to detect water pollution.

Thanks to the development of new methods at the interface of ecology and chemistry first pioneered in the 1970s, environmental sensors such as sentinel organisms are today used to detect signs of disturbances that remain indiscernible to humans, providing specific data on the noxiousness of pollutants.[1] Many animals and plants have since been tested and “enrolled” (Akrich et al. 2006) in environmental monitoring programs, making it possible to lower the threshold for detecting toxins in air, soil and water, and allowing investigations on the effects of low doses of particular pollutants on the environment. As a result of new funding and attention to the issue, lab and outdoor experiments multiplied in subsequent years.

Ecotoxicologists study the potential environmental impact of pollutants by observing how they affect chosen organisms at the physiological, cellular and molecular levels. They do not aim at producing data about pollutants’ impact on humans directly. However, their findings on the contamination of plants and animals guide authorities who decide whether catching and consuming specific plants and animals should be regulated or not. For instance, in our case study, in the Gironde estuary, downstream the River Lot (Southwestern France), ecotoxicologists’ findings on heavy metal concentrations in aquatic species led to a ban on the consumption of some fish and shellfish.

This paper examines scientific uses of the Corbicula fluminae as a sentinel organism to detect zinc and cadmium pollution in a lab experiment and along the River Lot. It will give us the opportunity to discuss the role of plants and animals in environmental monitoring, and explore how the very concept of “sentinel” takes on new meanings when applied to the surveillance of risks.


Cadmium is naturally present in small quantities in most metalliferous ore. Its physical and chemical properties are numerous. It is highly resistant to corrosion and heat and not easily soluble in water. These characteristics make it an ideal material to add to domestic household items such as batteries, paint and plastics. While its industrial uses expanded in the 20th century, its release into the environment started in the 19th century with the production of zinc. For each ton of zinc that is smelted, about 3 kilograms of cadmium residue is also produced. Unlike zinc, which is necessary for the healthy functioning of human, animal and plant metabolisms, cadmium is highly toxic even at low doses. It can generate bone, kidney, liver and reproductive disorders (Nordberg 2004).

In Aveyron, the aftermath of 150 years of zinc metallurgy activities is not immediately discernible. With its rural surroundings and its 1,400 inhabitants, the town of Viviez has managed to keep a quiet atmosphere despite the presence of several factories and wastelands. The water of the Riou Mort, which runs through the town and then into the River Lot a few kilometers downstream, looks quite limpid. The apparent cleanliness of the waters belies the fact that high concentrations of heavy metals released by the local zinc factory over the past century are still trapped in the riverbed. The toxins can also be found in the River Lot sediments and affect the quality of the whole water system up to the Gironde estuary. The confluence of the rivers is currently being closely monitored by scientists.

Ecotoxicologists study pollution in order to assess its noxiousness on aquatic plants and animals. To this end, they have designed new experiments to obtain data showing how different heavy metals affect aquatic life, irrespective of the quantity or chemical status of these metals. To help them overcome difficulties involved in understanding complex phenomena which the human eye and instruments alone cannot fully grasp, they chose to “enroll” a heterogeneous cohort of life forms based on their specific ability to detect zinc and cadmium at low doses and to survive despite the damage caused by these pollutants.


The scientists I met said they rapidly realized that a mollusk named Corbicula, an invasive species originating from Asia, could help them understanding better the dynamics and effects of pollution, which technical tools had been unable to capture. These animals, whose biology is now well known, live at the interface between river water and sediments. Their breathing and feeding activities involve filtering high quantities of water and ingesting the pollutants present in it.

Working with Corbicula has several practical advantages. They are easy to find in lakes which are not contaminated by cadmium. They are not expensive to breed. They can be transported and kept in the animal house in the lab and require minimum care. They can survive for up to six weeks without food provided they have a supply of oxygenated water, including tap water. Their food, a microscopic alga called Scenedesmus, can be cultivated without difficulty either. Corbicula adapt perfectly to the artificial rivers used in the lab made of ordinary PVC pipes, gravel and water. Their life cycle is not as short as that of Drosophila, but it is still short enough to enable scientists to observe their development and reproduction in a time adapted to the rhythm of laboratory work (Kohler 1993). If all mollusks can detect heavy metals in water at low doses and generally react to their presence, all species are not equally resistant. The distinctive feature of Corbicula is their ability to survive both zinc and cadmium at the same time. Except when the dose is lethal, they can assimilate and concentrate these substances. However, after exposure to these substances they grow smaller and their behaviors are impaired. These signs provide ecotoxicologists with valuable information about the toxic effects of heavy metals. Their distinguishing traits make them rather attractive animals, i.e. “good candidates to act as lab and sentinel organisms”.

The scientists I observed in the lab and in the field call on Corbicula in different ways in their experiments. The organisms are handled and treated carefully because their performance depends directly on their well-being. In the simplified but controlled conditions of the lab, ecotoxicologists put them in artificial rivers in which they can introduce heavy metals progressively to examine their impact. In situ, Corbicula are placed in cages and immersed in rivers such as the Riou Mort. They are carefully transported to the field in a cooler filled with water oxygenated by a pump before being placed in groups of 25 in different locations both upstream and downstream from the zinc factory in Viviez. They are usually picked up 15 days later by the technician and PhD students who brought them there. Each group of mollusks is kept in a labeled bag before being taken back to the lab. In the dissection room, they are cut into three pieces. The gills are separated from the viscera and the soft body because cadmium impacts them differently: the gills are directly in contact with the water and toxins, while the viscera have a high accumulating potential. This preparation facilitates later interpretation as data obtained from each part can be compared (Lynch 1988). Animal pieces are either analyzed immediately or frozen for later analyses.

Physiological, biomolecular or genetic assays can then be performed. The Corbicula are weighed, numbered and crushed. Spectrometric analyses of the soft bodies, viscera and gills enable the measurement of the quantity of cadmium filtered and concentrated by the organisms at one location. Other tests are achieved by mercury saturation to measure the rate of metallothionein, a cell protein produced by the liver to trap heavy metals and thus reduce their toxicity. This protein, being directly correlated with their ingestion, is a biomarker, i.e. an indicator of the presence of heavy metals. Damage to DNA can also be visualized by polymerase chain reaction. The structure of uncontaminated genes can then be compared to those exposed to pollution, making legible new evidence of the toxins’ effects on aquatic life.

Through these experiments, ecotoxicologists showed that Corbicula could not survive very long in the highly polluted zones immediately downstream from Viviez. When moving down the Riou Mort, the bioavailability of heavy metals decreases and the mollusks manage to cope, although they are smaller and their reproductive functions are compromised. They display other signs of serious contamination, including a higher rate of cadmium and metallothionein in the body and genetic abnormalities (Andres et al. 1999, Baudrimont et al. 1999). Experiments demonstrated that the impacts of the heavy metals decreased downstream in the River Lot up to the Gironde estuary. In the estuary, despite the distance from Viviez, the presence of salty marine water increased bioavailability of toxins and risk of contamination for aquatic life.


The Corbicula do things that humans and most machines cannot do. They “look out for the environment continuously” as one ecotoxicologist told me—a task that would otherwise require expensive and difficult to maintain technical equipment. Nonetheless, the mollusks should not be considered as mere tools. They are “unfree partners, whose differences and similarities to human beings, to one another, and to other organisms are crucial to the work of the lab” (Haraway 2008: 72). They are collected in lakes before being installed in rivers like the Riou Mort and the Lot. Their efficiency in detecting heavy metals is related to the fact that they come from similar aquatic ecosystems which they can speak for. The success of the experiments depends on the safe transfer of the mollusks to the experiment sites. All disturbances to the specimens must be minimized to isolate the ones caused by pollutants. For this reason, paying attention to their specific needs, i.e. learning to understand what matters for them, is a crucial part of scientific work (Gramaglia and Sampaio da Silva 2012).

Yet, the notion of model organism should be distinguished from that of sentinel organism. Model animals are manufactured in the lab and for the lab. They are cut off from any kind of environment they could have lived in to be used as proxy. Their point of view on the phenomena at stake is not to be taken into account. Whether sentinel organisms are “enrolled” in monitoring protocols and surveillance tasks which efficiency depends a great deal on their embeddedness in specific places and their ability to express preferences.

The military term for soldiers posted at the outskirts of a given territory applies well to the Corbicula installed in cages at various points of the Riou Mort and the River Lot. The reason mollusks can complete their assigned task is because they are part in a network dedicated to the surveillance of hazards. They stand for their locations. Knowledge about pollution is produced through the gathering of information emanating from the different stations (and also different species which may have different preferences). It is assembled in the lab operating as a “center of calculation” (Latour 1987). The contrast between the upstream situation and the downstream one brings evidence of damage into view. This comparison can also reveal unexpected phenomena, such as the fact that heavy metals like cadmium affects the Gironde estuary 400 km from the contamination point since toxic effects increase when fresh and brackish water meet. Monitoring environmental hazards with the help of sentinel organisms and collective sensing devices provides scientists with insights on the changing geography of pollution and the variable effects of low doses of toxins on a large territory.

However, the figure of the sentinel should be distinguished from that of the whistle blower too. The latter is often understood as a professional breaking the rules of confidentiality applying to his/her domain of expertise to denounce a hazard, therefore speaking out against a collective he/she is related to. By contrast, a sentinel is a part of an existing surveillance device, whether institutionalized or not. Its efficiency depends on the degree of integration in a network allowing coordination, but also the exchange and processing of information. The actions of both the sentinel and whistle blower figures can be regarded as collective achievements involving humans and non-humans, but in the case of the second figure, the alert is an act of dissociation.[2]

Pollution, especially in its chronic and accumulative forms, is difficult to understand. The effects of toxins depend on a plurality of factors: the chemical status of the pollutants as well as the circumstances and the biology of the species affected by them. If a new form of localized/distributed vigilance involving different forms of life is needed to better anticipate and document the negative consequences of our actions, such a systematic vigilance requires not confining ourselves with existing norms of exposure but, instead, building surveillance networks once an alert is confirmed to capture early warnings of hazards. This new localized/distributed system could work, provided we pay close attention to the messages different sentinels carry about themselves, and reflect on what this information means for the environment we share with them.

[1] Wilhelm Nylander, a Finnish botanist working for the Museum of Natural History in Paris was one pioneer. He mapped air pollution based on his interpretations of lichen distribution in the Luxembourg gardens. He called the plants “hygiometers” as they helped him assess the salubrity of a location (1869). However, he worked mostly alone and no other scientist of his time continued his research. Animals like canaries may have been used as sensors for hazards in everyday practice, but their contribution was not investigated scientifically.
[2] Evidence for tension between the two can be found in the work of Cordelia Hesse-Honegger, a Swiss science illustrator who started documenting damages caused by the aftermath of Chernobyl disaster to insects’ morphology in Western Europe. Raffles (2010) tells us that Hesse-Honegger’s first accounts were dismissed on the ground that she had not observed a control group, nor had she collected enough data to allow statistically significant findings matching academic standards. Her aesthetic way of portraying insects’ deformity magnified their singularity too much, according to scientists who refused to consider the information on irradiation her specimens were bearing as relevant. While her later research responded to these criticisms by providing a larger set of data, she regretted that the insects on her pictures were not acknowledged as sentinels capable of monitoring genetic damages affecting potentially many life forms. Her isolation and outsider status made it even more difficult for her to be heard.


This study was funded by the French National Agency for Research (ANR project Re-Syst 08-CES-014). Interviews were conducted by Delaine Sampaio da Silva and I with researchers and lab staff from the GEEMA/AE team at the University of Bordeaux (France) who agreed to be observed and questioned.


Akrich, Madeleine, Michel Callon, and Bruno Latour. 2006. “Sociologie de la traduction. Textes fondateurs. ” Paris: Presses de l’Ecole des mines.

Andres, Sandrine, Magalie Baudrimont, Yvon Lapaquellerie et. al. 1999. “Field transplantation of the freshwater bivalve Corbicula fluminea along a polymetallic contamination gradient (River Lot, France) – Part I: Geochemical characteristics of the sampling sites and cadmium and zinc bioaccumulation kinetics.” Environmental Toxicology and Chemistry, 18(11):2462-2471.

Baudrimont, Magalie, Sandrine Andres, Jacqueline Metivaud, et. al. 1999. “Field transplantation of the freshwater bivalve Corbicula fluminea along a polymetallic contamination gradient (River Lot, France) Part II: Metallothionein response to metal exposure: a field illustration of the metal spillover theory” Environmental Toxicology and Chemistry, 18(11):2472-2477.

Gramaglia, Christelle and Delaine Sampaio da Silva. 2012. “Researching water quality with non-humans. An ANT Account.” In J.H. Passoth, B. Peuker, M. Michael Schillmeier (eds) Agency without Actors?: New Approaches to Collective Action. Routledge: London: 178-195.

Haraway, Donna. J. 2008. When species meet. Minneapolis: University of Minnesota Press.

Kohler, Robert. 1993. “Drosophila: A life in the Laboratory.” Journal of the History of Biology, 26(2):281-310.

Latour, Bruno. 1987. Science in Action: How to Follow Scientists and Engineers through Society. Cambridge, MA: Harvard University Press.

Lynch, Michael. 1988. “Sacrifice and the transformation of the animal body into a scientific object. Laboratory culture and ritual practice in the neurosciences.” Social studies of science, 18(2):265-289.

Nordberg, F. 2004. “Cadmium and health in the 21st Century – historical remarks and trends for the future.” BioMetals, 17(5):485-489.

Nylander, Wilhelm. 1866. “Les Lichens du Jardin du Luxembourg.” Bulletin of the Society of Botany, 13:364-371.

Raffles, Hugh. 2010. Insectopedia. New York: Vintage Books