January 2017


Issue Home >>

Have you ever heard of the world-famous local fish called Poecilia reticulata? You may know it as the guppy, canal fish, seven colours or millions. Many know this tiny fish from hours of childhood fishing pleasure, or as an aquarium pet. The Trinidadian guppy is famous because it is sold in many countries as an aquarium fish, it is the focus of research by many scientists abroad, and it has been introduced to many countries to control mosquitoes since it eats mosquito larvae.

Mosquitoes are a nuisance and threaten human health because they are biting insects that feed on human blood, and they are vectors, or carriers, of several life-threatening diseases such as dengue, yellow fever and malaria. The recent spread of these and other mosquito-borne diseases, such as Chikungunya and Zika, spurred renewed efforts to control mosquitoes. One popular method is to introduce larvae eating fishes, such as guppies, to water where mosquitoes breed in the hope that they will eventually reduce the number of adults.

As recently as 2013, guppies were released into fresh water in Pakistan to fight a dengue epidemic. In 2014, school children and volunteers joined the Guppy Campaign, a programme that released guppies into puddles as a means of fighting malaria. And in 2015 and 2016, they were used in Brazil to control the spread of Zika and dengue. Perhaps because of the familiarity and ease of using this approach, there were reports of people independently releasing guppies in disease-affected countries. A search for “mosquito control guppy” in Google produces hundreds of articles and websites promoting this approach.

But among these you will also find an article recently published by my colleagues and I, casting doubt on the effectiveness and wisdom of using this strategy.

Guppy Campaign volunteers use a simple effective strategy to convince others that guppies will eat mosquito larvae – jars of guppies eating mosquito larvae. The problem with this, and similar experiments showing that guppies eat mosquitoes is that the fish had no other food available.

There is strong evidence that guppies prefer other foods. In experiments in which guppies were given mosquito larvae and other foods, the guppies ate more of the other foods. The faeces of guppies caught in the wild showed that guppies in nature ate even fewer mosquito larvae than in the experiments. Here in Trinidad, we observed that guppies feed extensively on mosquitoes when they are in planters of stagnant water, but not when they are in moving waters or their natural settings. They also eat fewer mosquitoes in polluted water, probably because of the wider variety of foods available.

So, it is not that guppies do not eat mosquito larvae – they do – but it seems that they much prefer to eat other things, and will only eat a lot of mosquitoes when other foods are scarce. In other words, guppies may not be as effective in controlling mosquitoes as we think.

So, why may introducing guppies to guppy-free areas not be wise?

Chances are that the introduced guppies will be released into the wild or escape, for example, through flooding, or young guppies can hitch a ride elsewhere on birds or other animals. Guppy entry to any site is highly likely to result in the fish establishing a population that thrives in its new home. Guppies have become established in at least 69 countries outside of its native range, and use for mosquito control is implicated in about 60% of these cases. Escapees can easily establish new populations because of the very characteristics that make them attractive for pets, research, and mosquito control – they are hardy little fishes that easily adapt to new conditions, reproduce often, give birth to young fish, and grow and start reproducing quickly.

Guppies can tolerate and quickly evolve in a wide range of conditions. They are found in small ditches, drains, ponds, streams and lakes, in clear or muddy, fresh or brackish water, in temperate to tropical countries. They can be found in polluted water, for example, the drains along the Priority Bus Route, sewers and oil-contaminated streams. A mated female can store sperm for months, and use them to fertilise several batches of her eggs. Females collected from the wild are almost always pregnant, and this, plus their high adaptability, means that a single pregnant female has more than an 80% chance of starting her own successful population if introduced to a new environment, producing her first batch of young about 28 days after the eggs are fertilised.

Guppies alter the streams into which they are introduced. Guppies introduced to previously guppy-free parts of streams in T&T competed with resident species for resources, eventually reducing the diversity of local fishes to become the most abundant species. They changed the biology of the remaining fish species, by altering their reproduction, growth, survival, and hence density (that is, the number of individuals in an area).

In countries beyond their native range, the effects of guppy introduction may be more pronounced. In Hawaii, poeciliids (guppies and their cousins), introduced since the 1920s, were found to be 10–30 times more numerous than native fishes. Some native fishes have disappeared from these areas. Guppies also increased the amount of benthic biofilms (the slippery coating on rocks, plants and any solid surfaces in water, that are composed of diverse and complex community composed of algae, bacteria, fungi and other microorganisms embedded in a complex organic matrix). They also increased the abundance of benthic invertebrates such as insect larvae and worms, but this was uneven across species so that some actually decreased while others became more abundant. At least one alien invasive insect became more abundant in the presence of guppies.

Because each species affects its environment in different ways, these changes at introduced sites translate into other changes in the way these stream ecosystems function. One important feature is the cycling of nutrients such as carbon, nitrogen and phosphorus, as this helps ensure that these important building blocks for organisms are available in forms that they can use. Guppies have increased available dissolved nitrogen by up to eight times compared to areas without them, and total organic carbon by up to five times.

What does this mean? Consider the biofilm as an example: biofilm communities play critical roles in aquatic environments. They help cycle nutrients, and supplying energy and organic matter to other stream organisms. They are sensitive and respond to changes in nutrient availability, generally increasing in thickness and extent in response to increases in nutrients. In extreme cases, they can develop those unpleasant green or grey-green mats commonly seen in neighbourhood drains.

So what should be done to control mosquitoes? We recommend against using guppies on a large scale, as they may be ineffective, and have significant risks to local ecosystems and their biodiversity. Instead, more effective methods (for example, mosquito nets or window screens) should be used. Where guppies are deployed, this should be done in well-controlled settings, and be carefully monitored. There is a lot of information on biological control that can be used for guidance on best practice, but we strongly recommend that research into the use of guppies to control mosquitoes needs to merge the medical, health and ecology and evolutionary sciences.

Dr Dawn Phillip is a Lecturer in the Department of Life Sciences, Faculty of Science and Technology, UWI St Augustine