The Problem of Brown Tree Snakes: Why Trojan horses only work in Troy

Several years ago, while working as an Animal Care Technician for the USDA’s National Wildlife Research Center (NWRC), I became intimately familiar with the case of the Brown Tree Snake (Boiga irregularis). For those of you who are unfamiliar with the story, it goes something like this:

At the end of World War II, war-time ships were busy with activities around Guam, an island about a thousand miles south of Japan. A few Brown Tree Snakes hitched a ride to Guam, an island their species had never been to. The snakes were probably drawn to the mice already stowing-away on the ships. However, once on Guam the species started to flourish. Their numbers, with more abundant prey resources than in their native ranges, started to show exponential growth.

This snake has wiped out several bird and lizard species, including some that were endemic to Guam. This has caused an increase in the abundance of spiders on the island. Determined to regain our beautiful Helen taken by the Trojans (and also to alleviate the infrastructure problems caused by the snakes) Uncle Sam was rallied to war, and set out across the ocean to destroy its enemy.

The first attempts at controlling the snakes focused on using trained dogs, and eliminating individual snakes by hand. These measures were put into effect in 1994, 40 years after the introduction of the snake to the island. These methods were largely unsuccessful in lowering snake populations. By 1999, no headway had been made in crashing the gates of Troy, and a new strategy was needed.

Researchers from the NWRC quickly collected specimens and started testing the most effective control methods. It was found that over-the-counter grade aced into a dead mouse was enough to kill brown tree snakes. Over the next 14 years, different delivery methods were tested and implemented.

As of 2015[1], the newest tree snake eradication efforts involve injecting acetaminophen into dead mice, attaching them to a parachute, and dispersing them, by helicopter, throughout the jungle of Guam[2]. Here’s how it is supposed to work:

The little parachutes get caught in the canopy, dangling their deceptive prize for the snakes to find. Snakes that eat the mice are exposed to the toxin inside, and quickly die. It is more or less the population control equivalent of the Trojan horse, or in this case the “Trojan mouse”. The dead snakes reduce the size of the reproducing population, and (hopefully) the total population can be brought under control. It seems simple and straight-forward. However, remembering some simple concepts from evolution and behavior, the story gets more complicated.

For instance, Charles Darwin noted great variation among individuals within a species, and used this to develop his theory of natural selection. The process of natural selection is one by which the varieties of a species that are most successful in an environment reproduce more, causing an increase in the abundance of that variety, compared to other varieties. Through the many environmental pressures, species change through time to fit the constraints of their surroundings. Modern biologists have recognized that many behaviors act as variable traits that are influenced by heredity. In other words, natural selection acts on behaviors as well.

Given this, consider the snakes on Guam. Prey selection is a variable behavior in brown tree snakes, with snakes “eating a wide variety of vertebrate prey, including reptiles, birds, and small mammals…”3. In relation to the quality and abundance of these food sources, when one food source becomes scarce or dangerous, snakes that don’t rely on this food source will have higher reproductive success than a snake that does.

By dropping our “Trojan mice”, we are simply adding a human selection pressure to snakes. We are not eradicating the brown tree snake, because we cannot possibly get every snake to take the bait.  By killing the snakes that will sometimes eat a dead mouse, we have simply given a leg up to the snakes that avoid mice as a prey item. These include the snakes that eat only birds and lizards. Effectively, we have increased the relative abundance of snakes that prefer “rare meats”, and are punishing snakes that are likely to provide beneficial services, such as rodent control.

Other activities employed against the brown tree snake include “detector dogs, traps, nighttime fence searches, oral toxicants, barriers, prey reduction, and public outreach activities.[3]” The total costs of these programs is about $3,000,000 annually (although last year the budget was increased by $500,000 “for the manufacturing of an automated bait delivery system for the Toxicant Bait Drop Project[4]”). Seems like a lot of money to control some snakes.

Given that there are roughly 2,745,400 brown tree snakes on Guam (50 snakes/hectare x 54908 hectares on Guam), I propose a simpler and more effective solution. Put a bounty of $1 on snakes.

Considering the high density of snakes of the island, to make a salary of $31,000[5], a hunter would have to collect about 124 snakes per working day (31,000 snakes/250 working days in a year). To trap this many snakes, a hunter would need to cover an area of only 6.2 acres per day, or roughly 6 football fields. Given an 8 hour day, traps, trained dogs, and the human technological advantage, it seems reasonable that the activity of snake hunting could become economically profitable. Profitable hunting has been a proven eradication method of any species; look at the history Passenger Pigeon[6] if you need proof that it works.

Therefore, it seems unreasonable that we spend $3,000,000 a year applying a selection pressure, when we could spend $2,745,400 to eliminate all the snakes and provide around 85 islanders with jobs until the snakes are eliminated. If they really are a threat to biodiversity, and could potentially cost us billions of dollars, it would make sense to pick a more proven and effective method of eradication. Let’s leave the elegant deception to the ancient Greeks and focus on methods modern capitalists can support.

–Gabe Buckley
B.S. Zoology with a minor in Business Administration
Profession Science Masters in Zoo and Aquarium Management, class of 2015

[1] See page 73.


[3] See page 75.

[4] See page 73.

[5] the average salary of Fishing and Forestry Occupations in Guam. See “Annual mean wage” in “Farming, Fishing, and Forestry Occupations”.


14 thoughts on “The Problem of Brown Tree Snakes: Why Trojan horses only work in Troy

  1. leorah

    I really like your proposed solution – I think it kills a lot of snakes with one stone, so to speak, by providing income to locals and incentive for removal, as well as theoretically bringing down costs of removal. The one flaw I see here is that potentially those benefiting from the bounty would also realize that once the snakes are gone, so is the income, and so would end up supporting the retention of a viable population in order to continue to profit from it. On the plus side, even if this was the case, hopefully the retained population would be low enough due to the hunting that it could give local species a chance to recover or adjust to the predators or at least slow the rate of ecological damage.
    The only question I have is whether in the case of the “Trojan Mice” we are really likely to see selection working quickly enough to benefit the snakes. I doubt social learning could occur in this case, and based on a quick Google search of the brown tree snake diet, I simply wonder if there would be a significant number of snakes who would actively avoid the mice if they were an easy target…?

    1. mberne

      I never knew that these snakes caused such an issue on the island! I like your suggestion of having people hunt the snakes with a bounty put on them because this helps the community while eliminating a serious problem all while keeping the cost to a minimum. Also, using a variety of methods to remove the snakes would not allow snakes that do not eat the mice to have a higher reproductive success, therefore controlling the diversity of the the populations prey preference, giving a better chance at catching them and not just passing prey preference off to another susceptible species.
      Are there any natural predators to these snakes on the island or any economic value such as food or the use of skins? This could have an even higher incentive for people to go out and hunt this invasive species that is threatening their native species!

  2. mplatt

    I can see your business minor in this solution and I think it is truly brilliant! Why spend 3,000,000$ annually for an unlimited amount of time when you could spend less for a shorter time period until the snakes are gone? This has many benefits including giving profit to the natives. However like Leorah said it could be a dangerous game to give a profit to someone with a time limit. If these people begin to depend on the money, and spend every work day hunting snakes, there could be an economic crisis of sorts once that profit is gone. Often with invasive species a predator is released to hunt the invasive animal. Do you know if they attempted this? Either way I think this is a viable solution with a relatively easy implementation process.

  3. smorton

    I had heard of the issues around the brown tree snake on Guam before but had never been exposed to the whole story, it is really interesting. I completely agree with your assessment of the effects these “Trojan Mice” will have on the snake population. Snakes that take dead prey will be eliminated and snakes that avoid the bait will live on to reproduce. Not only does this select for individuals who will not take the bait and who may prefer “rare meats” but also it could be inferred that those who did consume poisoned mice had lower fitness to begin with. Snakes that took dead mice may have had on average lower fitness than snakes who could afford to avoid these ‘lazy” meals, now you are not only selecting for smarted snakes (ones who avoid dead mice) but also stronger (those who could expend the energy to hunt) ones. Very interesting topic it will be exciting to see how things play out in the next few years.

  4. Lisa Angeloni

    It would be interesting to follow up on these populations to test your hypothesis that snakes will evolve mechanisms to avoid mice as prey items. Another study on this topic comes to mind: “Phillips BL & Shine R. 2006. An invasive species induces rapid adaptive change in a native predator: cane toads and black snakes in Australia. Proceedings of the Royal Society B 273:1545-1550”. In this case, the introduction of toxic cane toads to Australia in 1935 caused red-bellied black snakes to have reduced preferences for toads as prey and greater resistance to their toxins, as well as smaller heads that may prevent eating large doses of toxin. Their findings did not suggest that these changes came about through learning or acquired toxin resistance, but rather that they evolved via natural selection over that time period.

  5. kburrus

    I really enjoyed this article. You had a lot of very good information about the brown tree snakes and backed it up with a lot of support. I can really tell that you did your research on the topic and know a lot about the brown tree snake on Guam. It is very interesting how the eradication process is driving natural selection and that sometime in the future the dropping of the mice wont work. I also really liked your title and how you compared the mice with the Trojan horse. Great job on this article.

  6. ematth

    I really enjoyed your proposed solution. We spend millions of dollars a year on eradicating animal species that have grown in abundance, often acquiring the label of a “pest,” when we also have millions of people living off of $1 a day just to put food on the table, and maybe even a roof over their heads. I had also never thought about how certain eradication processes often drive selection to eliminate individuals who may have desirable traits, such as your example with snakes that may help control an overabundance of house mice.
    Great job!

  7. amajor

    I, for one, would say that your proposal is a brilliant idea and one that should be taken. By placing a money-based bounty on the invasive snakes’ heads in a poor or developing country, it would be a grand incentive that will prove to be very successful. I know this because bounties like this has been involved in wiping out species before. The greatest examples of this is the fur trade and the thylacine. The bounty placed on the furs of beavers and other fur bearers have wiped them out of their former ranges and have kept their populations in control in other places. In the case of the Thylacine (aka the Tasmanian wolf or tiger), a bounty was placed on it’s head in 1890 (a British pound or 30 US dollars for every thylacine) and that was one of the factors that lead to it’s extinction in the wild in 1930. So, I know that for a fact that it will definitely work for the snakes. But I would be a little but cautious because something like this was done for the Burmese Pythons in Florida and very few snakes were brought in. In all though, I love this proposal because it’s turns the local people into a solution and not as a pest. In order for ANY conservation project to work, you must work with the local people and they can become a great asset for the project.

    I love how you’ve brought in the consideration that there are variations within the species and that using poisoned mice would only make it worse. So, in all, great job, my friend, great job.

  8. reneev

    I was impressed by your mathematical skills in coming up with a solution. I feel like today, people will come up with a solution and put it right to work before considering the logistics, like you mentioned. Trojan Mice probably seemed like a great idea at first glance, but whoever put it into action failed to look deeper into the issue and what new problems it might create. In your proposed plan it seems like you’ve thought out how much money it will cost and all of the benefits that would arise from implementing this strategy. I do think there would be a lot of backlash from reptile enthusiasts and animal activists who don’t understand why it’s a necessary evil. I would say it might be a good idea to consider relocating the snakes to their original habitat, rather than just killing them. It will add some cost, but it would make a lot of people happier.

  9. Lacey Humphreys

    I think that your solution is great, both economically feasible and the work is not too strenuous. I do have a couple questions though; first, are these snakes poisonous? Could they cause a potential threat to those dogs and humans hunting it? Also, was there any evidence about other predatory animals eating these poisonous mice? I am really curious to the impact on local birds of prey who predate mainly on small rodents.

  10. ssteele1

    Of course, like the others, I think the idea is great. It’s intuitive and so….. simple. Really makes you wonder what other kinds of things we are spending millions on when there are more sensible solutions. I really liked your explanation of the evolution. Its pretty simple to understand that not all snakes like mice, but I hadn’t thought about the “Trojan mousing” worsening the problem by selecting for snakes who like birds and lizards which was the problem in the first place. Great thought and solutions!

  11. Savimay

    I really think this is a brilliant plan. Your idea would be much more inexpensive than investing the money towards harvesting mice/buying mice, poisoning the mice, and dropping them out of helicopters. Since the government is already willing to spend the money on eradicating the snakes, it would make more sense to try and distribute the funds back into its own economy. There could also be adverse effects to throwing acetaminophen into the environment that we are still unaware of. Your idea just seems much more practical, inexpensive, and more likely to succeed. Great blog Gabe!

  12. cfalvo

    It’s a shame so many dollars have already been spent on a much more extravagant and ineffective removal method, because this is such a simple and cost-effective solution. I wonder about the motivation of the local people in getting on board with this, the monetary incentive is definitely persuasive, but maybe not all citizens can be so easily swayed. I think making sure to educate the public about the ecological benefits of this project would be helpful in possibly incentivizing, but more importantly ensuring that they know what they are doing is important for the island.

  13. amccoy15

    I really like how you purpose an alternative way to eradicating the tree snakes. I think if it was implemented guam would save a lot of money and mice. And the people would be more actively seeking the snakes out acne radiating them faster then waiting for the snakes to find mice and have an affect from the mice.

Comments are closed.