By Laura Martin | September 23rd 2009 12:04 PM | Print | E-mail

    Ask the question “are invasive species bad?” to a group of ecologists and you are sure to raise some eyebrows. Invasive species have been charged with causing extinctions, crowding out biodiversity, altering ecosystem processes such as nutrient cycling, and even changing the evolutionary trajectories of native species. Of course they’re bad.

    But truly answering the question of whether invasive species are “bad” requires the additional consideration, “bad in comparison to what?”

    Last month, on a field trip with the Ecological Society of America conference in New Mexico, I took this picture out of the bus window:

    One of the scientists on the trip asked the tour guide about the impacts of “the invasive shrub that’s everywhere.” I had wanted to ask the same question—the shrub seemed to take up every inch of the landscape, extending as far as the eye could see. “Actually, it’s a native species,” was the tour guide’s reply. And with that, the conversation ended: If a species is native, it must be good… a native species is “supposed” to be there, after all.

    The above anecdote highlights our tendency - as ecologists, natural resource managers, policy makers, and nature enthusiasts – to view all non-native species as unnatural, and therefore unwelcome in a healthy ecosystem. The corollary to this is that we end up viewing all native species as natural, and therefore good. 


    Much of the research that addresses the impact of non-native species makes comparisons at the site level, rather than at the species-level.  Take for example, a review of 94 ecological studies by Liao et al., which found that variables associated with a key ecosystem process, carbon and nitrogen cycling, change significantly following the introduction of a non-native plant species to an area.

    [i] In order to measure the impact non-native species, the 94 reviewed studies compared areas dominated by a non-native plant species to areas where the non-native plant species is not present.

    This type of study, a comparison between an invaded and a non-invaded community, can be very useful in documenting ecological changes at a specific site.  This type of study cannot, however, untangle whether non-native species have different ecosystem impacts than native species. Comparing an invaded site to an un-invaded site can result in a comparison, quite literally, of apples and oranges.

    Consider a researcher who would like to know whether an introduced shrub species has negative impacts on a butterfly species that is found in open grasslands.  In order to do this, he or she compares butterfly performance in patches where the shrub has invaded against patches that do not contain the shrub invader.  Using this sort of comparison, we cannot untangle whether the shrub has an impact on the butterfly because it is a non-native species or because it is a shrub, a plant with very different architecture and leaf characteristics than the grass species. Would a native shrub have the same impact? If a native shrub naturally expanded into this grassland, would we consider it a bad thing? 

    In order to really understand whether introduced species are “different,” and whether their impacts are “bad,” species phylogeny needs to be taken into account. In other words, we need to compare introduced species to their native relatives: We need to compare an introduced shrub to a native shrub, or an introduced apple to a native apple. 


    When phylogeny is taken into account, ecological paradigms can be turned on their heads.  One example where this has played out is in tests of the Enemy Release Hypothesis—the idea that introduced plants have an advantage over native plants because, by packing their bags and moving to a new landscape, they have escaped the natural enemies that are best at attacking them (those insects or pathogens that they have co-evolved with in their native range). Data from numerous comparisons of invaded and un-invaded sites support this hypothesis. However, when ecologists Anurag Agrawal and Peter Kotanen grew 15 native plant species alongside their 15 closely related non-native species, they found that the leaves of the non-native plants were actually more attacked than the leaves of their native relatives—the opposite result of what the Enemy Release Hypothesis would suggest.[ii]

    Recent research has also challenged the assumption that invasive species lead to declines in biodiversity. In a paper published in PNAS last fall, Dov Sax and Steven Gaines analyzed the International Union for Conservation of Nature’s extinction database and found that, on average, plant species introductions have increased the plant diversity on islands, actually doubling the number of plant species. For birds, on the other hand, the rate of non-native species introduction and the rate of native species extinction on islands have been approximately equal, leading to no net changes in bird species richness.[iii] It seems that the answer to whether introduced species cause decline and extinction of native species is not a resounding “yes,” but a more nuanced “occasionally.”

    Negative impacts of a non-native species might also be averted by the evolutionary potential of native species. For example, native soapberry bugs have evolved longer beaks to feed on the thick-skinned fruits of a non-native tree introduced to Florida around 1955.[iv] Non-native species also have the potential to evolve in their new ranges; cane toads, introduced to Australia from South America, have evolved longer legs in their new range that has enabled them to spread more rapidly across the landscape.[v] We normally think of evolution as a painstakingly slow process, but evidence of rapid evolution of both the invaders and the invaded challenges our assumptions about the long-term impacts of invasive species.  


    The belief that non-native species have negative ecosystem impacts has led to significant federal, state, and local efforts to eradicate introduced species.  With many thousands of non-native species in the United States, it is imperative that the correct comparisons be used to sort out the “bad” species from the “not so bad” and maybe even the “good” non-native species.

    Could phylogeny be used to predict the impact of non-native species?  Researchers in California have shown that highly invasive grass species are, on average, significantly less related to the native grasses they are invading than less invasive grasses.

    [vi] If phylogeny predicts invasiveness, does it also predict the impacts of invasion?

    In order to answer this question in a serious way, there must first be more studies that compare non-native species to closely-related native species. We ought to know, for example, if a native cattail would shape invertebrate communities differently than an introduced cattail. We ought to know whether non-native species that are phylogenetically novel (that do not have closely-related native species) are more or less harmful to an ecosystem—or even not harmful at all. 

    Until we know these things, the answer to the question “are invasive species bad?” remains “bad in comparison to what?” 

    [i] Liao et al.  2008.  Altered ecosystem carbon and nitrogen cycles by plant invasion: a meta-anlysis.  New Phytologist 177:3, pgs 706-714.
    [ii] Agrawal and Kotanen.  2003.  Herbivores and the success of exotic plants: a phylogenetically controlled experiment.  Ecology Letters 6: 712-715.
    [iii] S
    ax and Gaines.  2008.  Species invasions and extinction: the future of native biodiversity on islands.  PNAS 105: 11490-11497.
    Carrol and Boyd.  1992.  Host race radiation in the soapberry bug: natural history with the history.  Evolution 46: 1052-1069.
    Phillips et al.  2006.  Invasion and the evolution of speed in toads.  Nature 439: 803
    Strauss et al.  2006.  Exotic taxa less related to native species are more invasive.  PNAS 103: