Advanced Search Map Locator
View Settings

Butterflies My Plant Lists Nurseries Planting Guide Contact Calscape About Calscape
Tap map to see plants native to location
Order by Popularity Order by Common Name Order by Scientific Name Order by # of Butterflies Hosted
Show nursery cultivars Hide nursery cultivars
Show plants not in nurseries Hide plants not in nurseries
Grid view Text view

Plant Choice Matters

Excerpted from "Nature's Best Hope" by Doug Tallamy

Understanding that it is the plants you put in your landscape that to a large degree determine its ability to support life is a good start, but there is an important nuance to your use of plants that will determine whether you actually succeed in bringing more life to your yard. To improve your yard’s ability to support life, you have to use the plant species that are good at passing the energy they have harnessed from the sun to animals. Through a twist of ecological fate, plants differ widely in their ability to do that. How, then, do we know which plants are support life the best?

Native vs Introduced

Nearly all of us get our plants from nurseries, but the plants in most nurseries fall into two very distinct categories: they are either native to your area, that is, they share an evolutionary history with the plant and animal communities in your ecoregion or biome, or they developed the traits that make them unique species elsewhere. Depending on where you live, ‘elsewhere’ is typically from East Asia, although nurseries in the Pacific northwest and coastal California sell many plants from the Mediterranean region, and nurseries in the deep south carry a plethora of tropical plants. In the past, we didn’t care much about where a plant came from; we chose our ornamentals for the sole purpose of meeting a specific aesthetic need. However, when choosing plants to increase ecological function in our yards, geographic origin is the first attribute we must consider. Remember, we are choosing plants to fill particular ecological roles, and plants native to your region are almost always far better at performing local ecological roles than plants introduced from somewhere else.

Novel ecosystems

For as long as Homo sapiens has been on the move around the globe he has carried plants, and to a lesser extent animals, with him. Modern modes of transportation, international trade, and a keen desire to display unusual plants in our yards has turned what was a trickle of introductions in past centuries into a torrent of new species entering North America from foreign lands. So great, so sudden, and so disruptive has the influx of new species been that ecologists now call the majority of today’s ecosystems ‘novel’ ecosystems (Hobbs 2013). They are considered novel because many of the species within them are just meeting each other for the first time in evolutionary history. That is, their interactions with each other are ‘novel’ and are occurring without the tempering effect of long periods of co-evolution.

Although some scientists are excited about the evolutionary potential of novel ecosystems, such potential will not be realized for eons, and in the meantime, novel introductions pose serious threats to the interactions that have already evolved. Many such introductions have been devastating to native populations within ecosystems, and thus the structure and function of those ecosystems. When a novel predator is introduced, for example, prey often have no prior adaptations to defend themselves and quickly fall victim (Stolzenburg 2011). The same type of ecological devastation has occurred time and again when diseases such as chestnut blight and white pine blister rust, and non-native insects like the hemlock wooly adelgid, emerald ash borer, and gypsy moth were introduced to North America from other lands. But the most widespread and underappreciated consequences of creating novel ecosystems are those that occur when introduced plants replace native plant communities.

Invasive Species

Novel ecosystems are created by invasive species, and there are more species of invasive plants, over 3300 in the U.S alone (Qian and Ricklefs 2006), than all of the other invasive organisms combined. Invasive plants, defined as­ non-native species that displace native plant communities, should not be confused with fast growing, ‘aggressive’ native plants for one simple reason: native plants, aggressive or otherwise, have been duking it out with each other, competing for space, light, water, and nutrients, for millions of years. Over the eons, native species have evolved ways to cope with each other and the results of their interactions define the highly diverse species composition of most native plant communities all over the world. Invasive plants, in contrast, have just arrived within a community, and “just’ can be defined as within the last several hundred years, a blink of an evolutionary eye. They also have arrived without their suite of natural enemies, the insects, mammals, and diseases that keep them in check in their homeland. And so, interactions between invasive plants and our native species are anything but tried and true; invasives and natives are just starting to negotiate what their future coexistence will look like, and it will take hundreds or thousands of generations for these negotiations to reach a compromise. Unfortunately, native plant communities are not able to negotiate from a position of power because plants that have proven to be invasive have an enormous competitive advantage over most native plant species which enable many of them to run amuck across the landscape.

When invasive plants like autumn olive, buckthorn, barberry, air potato fern, bush honeysuckle, and Phragmites invade a plant community, they replace the local native species at that site either completely, causing local extinction, or partially, causing a decline in the plants upon which that ecosystem has depended for eons. If introduced plants were the ecological equivalents of the native species they replaced, ecosystems would look different after an invasion, but they would be just as productive (though less stable). Introduced plants may, in fact, be equal to natives in their ability to produce some ecosystem services, but they pale in comparison to natives in perhaps the most critical role plants play in nature. That is, introduced plants are poor at providing food for the animal life that runs our ecosystems.

Plants, in essence, enable animals to eat sunlight; by capturing energy from the sun and storing it through photosynthesis in the carbon bonds of simple sugars and carbohydrates, plants are the basis of every terrestrial and most aquatic food webs on the planet. But animals benefit from the energy captured by photosynthesis only if they can eat plants, or eat something that ate plants previously. And there is the rub: the group of animals that is best at transferring energy from plants to other animals is insects. Unfortunately, most insects are very fussy about which plants they eat.

The Curse of Specialization

Ah, how easy conservation would be if all plants delivered the same ecological benefits; that is, if all plants were the ecological equivalents of each other. We could plant Eucalyptus around the world and plant-eaters everywhere would be a happy as koalas. The nectar-filled butterfly bush so many people plant ‘to help the butterflies’ would actually serve as a larval host for all butterflies (instead of only one species in southern California) and deliver up pollen and nectar to all 4000 species of native bees rather than just a few generalist bees. The resplendent quetzal would be common as starlings because it would be able to eat the fruit of all plants instead of just wild avocado. We could rename the evening primrose moth ‘the every-plant’ moth, and the ornamental bamboos that are consuming yards and road shoulders would feed monarch and queen butterflies as well as milkweeds do.

But alas, specialized relationships between plants and animals are the rule rather than the exception in nature and they are far more common than generalized ones. This is particularly true for specialized relationships involving food webs: those interconnected relationships that transfer energy harnessed from the sun by plants to animals that eat plants, and then to animals that eat animals. Many people refer to this transfer of energy as a food chain, but if you were to make a diagram of a plant and then all of the species that eat that plant, as well as all of the species that eat each of those plant-eaters, the result would look far more like a very complex spider web than a linear chain. Thus, the term food web.

By far, the most important and abundant specialized relationships on the planet are the relationships between insects and the plants they eat. Most insect herbivores, some 90%, in fact, are diet specialists, or what we call ‘host plant specialists,’ restricted to eating one or just a few plant lineages (Bernays and Graham, 1988; Forister et al., 2015). Host plant specialization has been known to and well-documented by entomologists since the early 1960s, but scientists have never been very good at talking to each other, so the importance of host plant specialization in gluing ecosystems together is still under-appreciated by many ecologists, restoration biologists, and particularly by conservation biologists. This is why proposals to reforest tropical areas of the world with Eucalyptus have not been met with jaw-dropping outrage (to wit, blue gum Eucalyptus from Australia is now the most abundant tree in Portugal and African tulip tree is the dominant tree in Puerto Rico), and why more and more, shade coffee marketed as ‘good for the birds’ is being grown under the shade of Eucalyptus, Introduced species of pine, citrus and mango, even though these are plants that make little to no food for birds in the coffee regions of the world. The specialized relationships between insects and plants are so important in determining ecosystem function and local carrying capacity that it is worth spending a little time to explain why this is so, and how these relationships have come about.

Plants, of course, don’t want to be eaten; they want to capture the energy from the sun and use it for their own growth and reproduction. So, in an attempt to deter plant-eaters, they manufacture nasty chemicals and store them in vulnerable tissues like leaves. These chemicals are secondary metabolic compounds that do not contribute to the primary metabolism of the plant. That is, they are not a necessary part of the everyday jobs of living and growing. Instead, their job is to make various plant parts distasteful or downright toxic to insect herbivores. Some well-known plant defenses include toxic compounds like cyanide, nicotine, cucurbitacins, and pyrethrins; heart-stoppers like cardiac glycosides; and digestibility inhibitors like tannins.

You might ask, ‘If plants are so well defended, how then can insects eat them without dying?’ This question dominated studies of plant-insect interactions for three decades, but at this point, the answer has been thoroughly delineated. Caterpillars and other immature insects are eating machines; some species increase their mass 72,000-fold by the time they reach their full size (Richards and Davies, 1977). Because caterpillars necessarily ingest chemical deterrents with every bite, there is enormous selection pressure to restrict feeding to plant species they can eat without serious ill effects. Thus, a gravid (read: pregnant) female moth attempts to lay eggs only on plants with chemical defenses their hatchling caterpillars are able to disarm.

There are many physiological mechanisms by which caterpillars can temper plant defenses, but they all involve some combination of sequestering, excreting, and/or detoxifying defensive phytochemicals before they interfere with the caterpillar’s health. Caterpillars typically come by these adaptations through thousands of generations of exposure to the plant lineage in question. In short, by becoming host plant specialists, insect herbivores can circumvent plant defenses of a few plant species well enough to make a living, while ignoring the rest of the plants in their ecosystem. For our purposes, however, a key point regarding host plant specialization is that it does not happen overnight; although every once and awhile an insect species coincidentally possesses enzymes that are able to disarm a plant species that it has never before encountered in its evolutionary history, it usually takes many eons for an insect to adapt to a new host plant, if it can adapt to it at all.

Does this mean insect specialists have won the evolutionary arms race with plants? Somewhat, but only in relation to the plant lineage on which they have specialized. When viewed across all lineages, plant defenses are very effective at deterring most insects. The monarch butterfly provides a great example. This species is a specialist on milkweeds that use various forms of toxic cardiac glycosides to protect their tissues. Very few insects can eat plants containing cardiac glycosides, but over the ages, monarchs (in fact, the entire Danaid lineage to which monarchs belong) have developed the enzymes that can make cardiac glycosides less toxic. They also have found a physiological mechanism for storing these distasteful compounds in their wings and blood, rendering their own bodies unpalatable to predators. And monarchs have gone one step further; tasting bad after eating milkweed plants does not help a monarch if a bird has to eat it in order to discover its unpalatability. But monarchs, like many other distasteful insects, advertise their bad taste with an aposematic orange and black pattern that serves as a universal waring signal to would-be predators - - ‘Don’t eat me. I taste bad.’

Milkweeds are so named because, in addition to cardiac glycosides, they defend their tissues with a milky latex sap that jells on exposure to air. Insects that attempt to eat milkweed leaves soon find their mouthparts glued permanently shut by the sticky sap. Yet monarchs have found a simple but amazing way to defeat this defense; they block the flow of sap to milkweed leaves (Dussourd and Eisner, 1987). This is an example of a behavioral adaptation (as opposed to a physiological adaptation) and you can easily watch it in action right in your yard. When a monarch caterpillar first walks onto a milkweed leaf, it usually moves to the tip of the leaf and starts to eat. If any latex sap starts to ooze from the wound, the caterpillar immediately stops eating, turns around, and crawls 2/3 of the way back up the leaf. There it chews entirely through the large midrib of leaf. That simple act severs the main latex canals that move the sap throughout the leaf. With the canals blocked, all of the leaf tissues below (distad of) the midrib wound become latex-free and the monarch can eat them without gumming its mouthparts. If the monarch decides to chew through most of the leaf petiole instead of the leaf midrib, latex is blocked from the entire leaf. Incidentally, this behavior provides Monarch hunters with a convenient tool for finding monarch caterpillars, for the leaf flags at the point where the monarch weakened the midrib. Any milkweed plant with a flagged leaf is or has been the home of a monarch!

The advantage of these adaptations is obvious for the monarch, but there are also disadvantages to such specialization, especially in today’s world. Unfortunately for the monarch, the ability to detoxify cardiac glycosides and block latex sap in milkweeds does not confer the ability to disarm the chemical defenses found in other plant lineages. This means that of the 2137 native plant genera in the U.S., the monarch can develop (with very minor exceptions) on only one, the milkweed genus Asclepias. The evolutionary history of this butterfly has locked it into a dependent relationship with milkweeds and if milkweeds should disappear from a landscape, so would the monarch. And this is exactly what has happened across the U.S. in recent years. A growing culture that favors neat, lawn-lined agricultural fields combined with an unwillingness to share designed landscapes with milkweeds helped reduced monarch populations 96% from their numbers in the 1970s as of 2013 (Brower et al. 2012). Monarch declines have been even more disastrous in the population west of the Rockies. Can monarchs adapt to other plant species? In theory yes, but in reality, no. The Monarch lineage has been genetically locked into a relationship with milkweeds for millions of years. Adaptation could conceivably modify this relationship very slowly over enormously long periods but asking monarchs to suddenly (within 30 years!) switch their dependence on milkweeds to an entirely different plant lineage, say, for example, crape myrtle, is like asking humans to develop wings. The number of genetic changes required to make such a switch reduces the probability of it happening before monarchs disappear to near zero.

Please note that monarchs are not exceptions, either in their specialized relationship with milkweeds, or in their current plight. They are typical of 90% of the insects that eat plants; their evolutionary history has restricted their development and reproduction to the plant lineage on which they have specialized. And as we homogenize plant diversity around the world by replacing diverse native plant communities with a small palate of ornamental favorites from other lands, the insects that depend on local native species decline. We have caused these declines by the way we have designed landscapes in the past. But we can and must reverse them by the way we design landscapes in the future, for our plant choices will determine how well our ecosystems function.

I will close this section with some numbers, not just because they demonstrate that introduced plants reduce both species and interaction diversity, but because they hammer home how large these reductions are. A few years ago, my students Melissa Richard and Adam Mitchell set out to measure what happened to caterpillars when invasive plants created a ‘novel ecosystem.’ Finding habitats that were thoroughly invaded by introduced plants such as Autumn olive, multiflora rose, Callery pear, porcelainberry, burning bush, and bush honeysuckle was easy. They typify the ‘natural’ areas near the university of Delaware where we did our study. The trick was finding places that were still relatively free of invasive plants. Using a combination of restored sites, and areas not easily accessed by deer which exacerbate the spread of invasive plants, we finally found what we were looking for: 4 invaded sites and 4 primarily native sites of similar size. Using replicated transects, we counted and weighed caterpillars at each site, once in June and again in late July. By every measure the caterpillar community, and by extension, the community of insectivores that relied on caterpillars for food, were seriously diminished when introduced plants replaced native plants. Even though there was more plant biomass along the invaded transects, there were 68% fewer caterpillar species, 91% fewer caterpillars, and 96% less caterpillar biomass than what we recorded in native hedgerows (Richard et al. 2018). To summarize these numbers in terms of the everyday needs of the animals that eat caterpillars, we found nearly 24 times less food available in the invaded habitats!


My students and I did our study in unmanaged hedgerows, what passes for natural areas where I live. But would we have seen the same impact on insect populations if we had conducted the study in a typical suburban neighborhood? The answer would depend entirely on the percentage of the plant life in the study landscapes that was introduced. Unfortunately, in most urban/suburban, and even exurban landscapes the majority of plants are from somewhere else (Mckinney 2002). My students and I have measured this in 25 year old suburban developments in Delaware, northeast Maryland and southeast Pennsylvania. We didn’t have to work too hard, because these landscapes contained very few plants at all. They were 92% lawn! But of the plants that were there, 79% on average were introduced from other continents. Moreover, they were largely the same species we had studied in the invaded hedgerows: Callery pear, bush honeysuckle, privet, burning bush, Oriental bittersweet, barberry, and Norway maple. Unfortunately, homeowners still landscape with invasive species throughout the country.

For years I have speculated about the consequences of such landscaping choices for birds that many of us would like to share our yards with. I had to speculate because no one had directly measured what happens to bird populations in landscapes that favor introduced plants. I was pretty safe in my speculations because logic dictates that if you take away the food birds need, they won’t do well. This, as the saying goes, is not rocket science. Nevertheless, I need speculate no longer. In the first study of its kind, my student Desiree Narango has measured what happens to Carolina chickadee populations and the caterpillars that support them when native plants are replaced by introduced ornamentals in suburban settings (Narango et al. 2017, 2018.). For three years Desiree and a team of field assistants followed breeding chickadees in the suburbs of Washington D.C. during the nesting season. Using video cameras at the nests, radio isotope analyses, territory mapping, vegetation analyses, foraging observations, and citizen scientists, Desiree was able to quantify all of the variables required to model population growth of chickadees as a function of the percentage of introduced plants within the chickadee’s breeding territories.

Desiree found far more than I have space to describe, but here are some of the highlights of her research. Throughout her study, parent birds foraged for food on native plants 86% of the time. Compared to primarily native landscapes in her suburban study sites, yards dominated by introduced plants produced 75% less caterpillar biomass and were 60% less likely to have breeding chickadees at all. Apparently, chickadees were able to assess the quality of the landscape before they decided whether or not to set up house in one of Desiree’s chickadee boxes. If a chickadee did build a nest in a yard with many introduced plants, it contained 1.5 fewer eggs than nests in yards dominated by natives and those nests were 29% less likely to survive. Chickadees that did decide to nest where there were not enough caterpillars fed their young spiders and small Homopterans like aphids but these food items did not compensate nutritionally for the lack of caterpillars in their chick’s diets. Nest in yards dominated by introduced plants produced 1.2 fewer chicks, and delayed chick maturation by 1.5 days compared to nests located in yards with lots of native plants.

Some of these differences may not sound very big to you, but cumulatively they are making a huge and negative impact on suburban chickadee populations. The consequence of these differences was that chickadee populations achieved replacement rate, that is, produced enough chicks each year to replace the adults lost to old age and predation, only in yards with less than 30% introduced plants. Unfortunately for the chickadees in Washington D.C. suburbs, Desiree found that, on average, 56%, of the plants are introduced.

Although these results are not a surprise, they remove the guesswork from understanding how much our plant choices impact the life around us. Here is solid evidence that, at least for Carolina chickadees, introduced plants are not the ecological equivalents of the native plants they replace, but it is hard to imagine why other insectivorous birds would not be similarly affected by introduced plants. Desiree’s research helps us understand that it is the plants we have in our yards that make or break bird reproduction, and not the seeds and suet we so dutifully buy for our feathered friends, although those supplements certainly help our birds after they have successfully reproduced (Marzluff 2014). Her results also give us insight into what is happening beyond our yards in the natural areas that have been invaded by the ornamental plants we have brought from Asia and Europe. We can now better understand one of the factors that has caused 432 species of birds to decline at a perilous rate in North America (State of the Birds Report 2016). Most of all, Desiree’s study and others to come should supply all the motivation we need to landscape with plant function as well as beauty in mind.

Costs vs Benefits

Every now and then a paper appears in the literature that points out the ecological benefits delivered by introduced plants. The conclusion drawn by the authors is always the same; if introduced plants are doing good things in local ecosystems, perhaps we should tolerate, or maybe even encourage their presence. This logic has been used to justify planting more Eucalyptus in California and to discourage municipalities from spending money to fight invasive species almost everywhere. I would agree with this line of thinking on one condition; the net effect of the plant in question must be positive. Benefits cannot be viewed in ecological isolation; they must be compared to the costs or disadvantages associated with a particular plant as well. Demonstrating the benefits that a plant delivers is meaningless unless we also measure the ecological costs that plant brings to the system. Only in this way can we estimate whether the benefits outweigh the costs, or vice versa. If the net effect of an introduced plant improves ecosystem function, then yes indeed, let’s rethink our bias against it.

Here is a typical scenario; Kudzu, the Asian plant that now exclusively occupies over 7 million acres in the southeast U.S. (Forseth and Innis 2004), serves as a host plant to our native silver spotted skipper. This is not just a rumor; I personally have taken images of sliver spotted skipper caterpillars hiding within and eating the curled leaves of kudzu plants in Mississippi. The skipper, a legume specialist, has found the defensive chemicals of kudzu to be within the range of its detoxification abilities, and so it can reach maturity on the otherwise nutritious leaves of this introduced species. We can put the new host association between the sliver spotted skipper and kudzu in the benefits column; kudzu is creating a food web option that did not exist before its introduction. Sticking to the food web theme, though, we now have to consider whether kudzu is having any negative impacts on local food webs. The answer, of course, is yes. When kudzu smothers young oak trees in Camden County, Georgia, for example, a host option for 454 species of caterpillars disappears. Similarly, if black cherry is eliminated from this kudzu patch, 324 caterpillar species are lost. If willows, hickories and maples are covered by kudzu, 247, 229, and 223 species of caterpillars are lost respectively (Native Plant Finder, NWF). Such losses will hold for all of the woody plant genera lost to kudzu at this single site in Georgia. What about herbaceous plants? Well, if kudzu smothers goldenrod, as it surely would, 94 species of caterpillars are lost. If it covers native asters, another 80 species are lost. Sunflowers lost to kudzu host 67 species of caterpillars, horsenettle 67 more species, and so on. By allowing kudzu to invade an area in Camden County, Georgia, we have gained a host plant for silver spotted skipper, but lost host opportunities for the caterpillars of over a thousand other moths and butterflies. And, as with all food webs, removing food from the base of the web reverberates throughout the entire web, impacting all of the species that eat the caterpillars lost to kudzu. No doubt, the net effect of this kudzu invasion is not just slightly negative, it is hugely negative in terms of supporting local biodiversity.


Bernays, E.M. and M. Graham. 1988. On the evolution of host specificity in phytophagous arthropods. Ecology 69: 886-892.

Brower, L. P., O. R. Taylor, E. H.Williams, D. A. Slayback, R. R. Zubieta, and M. I. Ramirez. 2012. Decline of monarch butterflies overwintering in Mexico: Is the migratory

phenomenon at risk? Insect Conservation and Divers. 5:95–100.

Dussourd, D. and T. Eisner. 1987. Vein-cutting behaviour: Insect counterploy to the latex defense of plants. Science 237:898-901.

Forister, M. L., Novotny, V., Panorska, A. K., Baje, L., Basset, Y., Butterill, P. T., Cizek, L., Coley, P. D., Dem, F., Diniz, I. R., Drozd, P., Fox, M., Glassmire, A., Hazen, R., Hrcek, J., Jahner, J. P., Kama, O., Kozubowski, T. J., Kursar, T. A., Lewis, O. T., Lill, J., Marquis, R. J., Miller, S. E., Morais, H. C. Murakami, M., Nickel, H., Pardikes, N., Ricklefs, R. E., Singer, M. S., Smilanich, A. M., Stireman, J. O., Villamarín-Cortez, S., Vodka, S., Volf, M., Wagner, D. L., Walla, T., Weiblen, G. D., and L. A. Dyer. 2015. Global distribution of diet breadth in insect herbivores.  Proceedings of the National Academy of Sciences 112:442-447.

Forseth. Jr., I.N. and A.F. Innis. 2004. Kudzu (Pueraria montana): History, physiology, and ecology combine to make a major ecosystem threat” Critical Reviews in Plant Sciences 23: 401-413.

Hobbs, R. 2013. Novel Ecosystems: Intervening in the New Ecological World Order. Wiley-Blackwell.

Marzluff, J.M. 2014. Welcome to Subirdia: Sharing Our Neighborhoods with Wrens, Robins, Woodpeckers, and other Wildlife. Yale University Press, New Haven, Connecticut. 303pp.

McKinney, M. L. 2002. Urbanization, biodiversity, and conservation. BioScience 52: 883-890.

Narango, D.L., D.W. Tallamy and P.P. Marra. 2017. Native plants improve breeding and foraging habitat for an insectivorous bird. Biological Conservation 213: 42-50.

Narango, D.L., D. W. Tallamy and P. P. Marra. 2018. Nonnative plants reduce population growth of an insectivorous bird. PNAS. 115: 11549-11554.

Qian, H. and R.E. Ricklefs. 2006. The role of exotic species in homogenizing the North American flora. Ecol. Lett., 9, 1293 – 1298.

Richard, M., D.W. Tallamy and A. Mitchell. 2018. Introduced plants reduce species interactions. Biological Invasions 21:983-992.

Richards, O. W., and R. G. Davies. 1977. Imm’s General Textbook of Entomology. 10th edition. Vol. 1: Structure, Physiology, and Development. New York: John Wiley.

Stolzenburg, W. 2009. Where the Wild Things Were: Life, Death, and Ecological Wreckage in a Land of Vanishing Predators. Bloomsbury USA.