Investigating the Soil Seed Bank

Near the top of the world, deep inside a snow-covered mountain located on a Norwegian island, a vault houses nearly a million packets of seeds sent in from around the world. The purpose of the Svalbard Global Seed Vault is to maintain collections of crop seeds to ensure that these important species and varieties are not lost to neglect or catastrophe. In this way, our food supply is made more secure, buffered against the unpredictability of the future. Seed banks like this can be found around the world and are essential resources for plant conservation. While some, like Svalbard, are in the business of preserving crop species, others, like the Millennium Seed Bank, are focused on preserving seeds of plants found in the wild.

Svalbard Global Seed Vault via wikimedida commons

Outside of human-built seed banks, many plants maintain their own seed banks in the soil where they grow. This is the soil seed bank, a term that refers to either a collection of seeds from numerous plant species or, simply, the seeds of a single species. All seed bearing plants pass through a period as a seed waiting for the chance to germinate. Some do this quickly, as soon as the opportunity arises, while others wait, sometimes for many years, before germinating. Plants whose seeds germinate quickly, generally do not maintain a seed bank. However, seeds that don’t germinate right away and become incorporated in the soil make up what is known as a persistent soil seed bank.

A seed is a tiny plant encased in a protective layer. Germination is not the birth of a plant; rather, the plant was born when the seed was formed. The dispersal of seeds is both a spatial and temporal phenomenon. First the seed gets to where its going via wind, water, gravity, animal assistance, or some other means. Then it waits for a good opportunity to sprout. A seed lying in wait in the soil seed bank is an example of dispersal through time. Years can pass before the seed germinates, and when it does, the plant joins the above ground plant community.

Because seeds are living plants, seeds found in the soil seed bank are members of a plant community, even though they are virtually invisible and hard to account for. Often, the above ground plant community does not represent the population of seeds found in the soil below. Conversely, seeds in a seed bank may not be representative of the plants growing above them. This is because, as mentioned earlier, not all plant species maintain soil seed banks, and those that do have differences in how long their seeds remain viable. Depending on which stage of ecological succession the plant community is in, the collection of seeds below and the plants growing above can look quite different.

Soil seed banks are difficult to study. The only way to know what is truly there is to dig up the soil and either extract all the seeds or encourage them to germinate. Thanks to ecologists like Ken Thompson, who have studied seed banks extensively for many years, there is still a lot we can say about them. First, for the seeds of a plant to persist in the soil, they must become incorporated. Few seeds can bury themselves, so those with traits that make it easy for them to slip down through the soil will have a greater chance of being buried. Thompson’s studies have shown that “persistent seeds tend to be small and compact, while short-lived seeds are normally larger and either flattened or elongate.” Persistent seeds generally weigh less than 3 milligrams and tend to lack appendages like awns that can prevent them from working their way into the soil.

The seeds of moth mullein (Verbascum blattaria) are tiny and compact and known to persist in the soil for decades as revealed in Dr. Beal’s seed viability experiment. (photo credit: wikimedia commons)

Slipping into cracks in the soil is a major way seeds move through the soil profile, but it isn’t the only way. In a study published in New Phytologist, Thompson suggests that “the association between small seeds and possession of a seed bank owes much to the activities of earthworms,” who ingest seeds at the surface and deposit them underground. Later, they may even bring them back up the same way. Ants also play a role in seed burial, as well as humans and their various activities. Some seeds, like those of Avena fatua and Erodium spp., have specialized appendages that actually help work the seeds into the soil.

Not remaining on the soil surface keeps seeds from either germinating, being eating, or being transported away to another site. Avoiding these things, they become part of the soil seed bank. But burial is only part of the story. In an article published in Functional Ecology, Thompson et al. state that burial is “an essential prelude to persistence,” but other factors like “germination requirements, dormancy mechanisms, and resistance to pathogens also contribute to persistence.” If a buried seed rots away or germinates too early, its days as a member of the soil seed bank are cut short.

The seeds of redstem filare (Erodium circutarium) have long awns that start out straight, then coil up, straighten out, and coil up again with changes in humidity. This action helps drill the seeds into the soil. (photo credit: wikimedia commons)

Soil seed banks can be found wherever plants are found – from natural areas to agricultural fields, and even in our own backyards. Thompson and others carried out a study of the soil seed banks of backyard gardens in Sheffield, UK. They collected 6 soil cores each (down to 10 centimeters deep) from 56 different gardens, and grew out the seeds found in each core to identify them. Most of the seeds recovered were from species known to have persistent seed banks, and to no surprise, the seed banks were dominated by short-lived, weedy species. The seeds were also found to be fairly evenly distributed throughout the soil cores. On this note, Thompson et al. remarked that due to “the highly disturbed nature of most gardens, regular cultivation probably ensures that seeds rapidly become distributed throughout the top 10 centimeters of soil.”

Like the seed banks we build to preserve plant species for the future, soil seed banks are an essential long-term survival strategy for many plant species. They are also an important consideration when it comes to managing weeds, which is something we will get into in a future post.


Tiny Plants: Idahoa

This is a post I wrote three years ago as a guest writer for a blog called Closet Botanist. That blog has since dissolved, hence the re-post.

This year, we returned to the location in the Boise Foothills where I encountered the plant that inspired this post. I found what might be seedlings of the tiny plant. If that’s the case, the phenology is a bit delayed compared to three years ago. I’ll check again in a week or so. Until then, meet Idahoa.


I have taken a real liking to tiny plants. So many of the plants we regularly interact with are relatively big. Large trees loom above us. Tall shrubs greet us at eye level. Flowering perennials come up around our knees or higher. But how often do we get down low and observe the plants that hug the ground or that reach just a few centimeters above it? Turf grass is ubiquitous and groundcovers are common, but among such low growing plants (or plants kept low), even more diminutive species lurk.

It was a hunt for a tiny plant that sent me down a certain trail in the Boise Foothills earlier this spring. Listening to a talk by a local botanist at an Idaho Native Plant Society meeting, I learned about Idahoa scapigera. A genus named after Idaho!? I was immediately intrigued. Polecat Gulch was the place to see it, so off I went.

Commonly known as oldstem idahoa, flatpod, or Scapose scalepod, Idahoa scapigera is the only species in its genus. It is an annual plant in the mustard family, which means it is related to other small, annual mustard species like Draba verna. It is native to far western North America and is distributed from British Columbia down to California and east into Montana. It occurs in a variety of habitat types found in meadows, mountains, and foothills.

Idahoa scapigera is truly tiny. Before it flowers, it forms a basal rosette of leaves that max out at about 3 centimeters long. Next it sends up several skinny flower stalks that reach maybe 10 centimeters high (some are much shorter). One single flower is born atop each stalk. Its petite petals are white and are cupped by red to purple sepals. Its fruit is a flat round or oblong disk held vertically as though it is ready to give neighboring fruits a high five. Happening upon a patch of these plants in fruit is a real joy.

Which brings me to my hunt. It was the morning of March 20th (the first day of Spring) when I headed down the Polecat Gulch trail in search of Idahoa, among other things. The trail forms a loop around the gulch and is about 6 miles long with options for shortening the loop by taking trails that cut through the middle. I have yet to make it all the way around. Stopping every 10 yards to look at plants, insects, and other things makes for slow hiking.

I was about a half mile – 1 hour or more – into the hike when Idahoa entered my view. A group of them were growing on the upslope side of the trail, greeting me just below waist level. Many of them had already finished flowering and had fresh green fruits topping their thin stalks. At this location they are a late winter/early spring ephemeral. I made a mental note of the site and decided to return when the fruits had matured. Next year, I will head out earlier in hopes of catching more of them in flower.

On the way to Idahoa, I noted numerous other small, green things growing in the sandy soil. It turns out there are countless other tiny plants to see and explore. It got me thinking about all the small things that go unnoticed right underneath our feet or outside of our view. I resolved to move slower and get down lower to observe the wonders I’ve been overlooking all this time.

Further Reading:

Seed Oddities: Vivipary

Seeds house and protect infant plants. When released from their parent plant, they commence a journey that, if successful, will bring them to a suitable location where they can take up residence (upon germination) and carry out a life similar to that of their parents. Their seed coats (and often – in the case of angiosperms – the fruits they were born in) help direct them and protect them in this journey. Physical and chemical factors inhibit them from germinating prematurely – a phenomenon known as dormancy. Agents of dispersal and mechanisms of dormancy allow seeds to travel through time and space — promises of new plants yet to be realized.

There is rarely a need for a seed to germinate immediately upon reaching maturity. In many cases, such as in temperate climates or in times of drought or low light, germinating too soon could be detrimental. The most vulnerable time in a plant’s life comes when it is a young seedling. Thus, finding the right time and space to get a good start is imperative.

The fruits (and accompanying seeds) of doubleclaw (Proboscidea parviflora) are well equipped for long distance dispersal. (via wikimedia commons)

In rare instances, dispersal via seeds offers little advantage; instead, dispersal of live seedlings or propagules is preferable. For this select group of plants, vivipary is part of the reproductive strategy. In vivipary, seeds lack dormancy. Rather than waiting to be dispersed before germinating, viviparous seeds germinate inside of fruits that are still attached to their parent plants.

Occasionally, seeds are observed germinating inside tomatoes, citrus, squash, and other fruits; however, these fruits are usually overripe and often detached from the plant. In these instances, what is referred to as “vivipary” is not a genetic predisposition or part of the reproductive strategy. It’s just happenstance – a fun anomaly. The type of vivipary discussed in this post is actually quite rare, occurring in only a handful of species and prevalent in a select number of environments.

There are three main types of vivipary: true vivipary, cryptovivipary, and pseudovivipary. In true vivipary, a seed germinates inside the fruit and pushes through the fruit wall before the fruit is released. In cryptovivipary, a seed germinates inside the fruit but remains inside until after the fruit drops or splits open. Pseudovivipary is the production of bulbils or plantlets in the flower head. It does not involve seeds and is, instead, a form of asexual reproduction that will be discussed in a future post.

True vivipary is commonly seen among plant communities located in shallow, marine habitats in tropical or subtropical regions, such as mangroves or seagrasses. The term mangrove is used generally to describe a community of plants found in coastal areas growing in saline or brackish water. It also refers more specifically to the small trees and shrubs found in such environments. While not all mangrove species are viviparous, many of them are.

Seedlings of viviparous mangrove species emerge from the fruit and drop from the plant into the salty water below. From there they have the potential to float long or short distances before taking root. They may land in the soil upright, but often, as the tide recedes, they find themselves lying horizontally on the soil. Luckily, they have the remarkable ability to take root and quickly stand themselves up. Doing this allows young plants to keep their “heads” above water as the tides return. It also helps protect the shoot tips from herbivory.

Viviparous seedlings emerging from the fruits of red mangrove (Rhizophora mangle) via wikimedia commons

Another example of vivipary is found in the epiphytic cactus (and close relative of tan hua), Epiphyllum phyllanthus. Commonly known as climbing cactus, this species was studied by researchers in Brazil who harvested fruits at various stages to observe the development of the viviparous seedlings. They then planted the seedlings on three different substrates to evaluate their survival and establishment.

Epiphyllum phyllanthus is cryptoviviparous, so the germinated seeds don’t leave the fruit until after it splits open. In a sense, the mother plant is caring for her offspring before sending them out into the world. The researchers see this as “a form of parental care with subsequent conspecific [belonging to the same species] nursing.” Since the plant is epiphytic – meaning that it grows on the surface of another plant rather than in the soil – local dispersal is important, since there is no guarantee that seeds or propagules dispersed away from the host plant will find another suitable site. That being said, the researchers believe that “vivipary involves adaptation to local dispersal,” since “the greater the dispersal distance is, the higher the risk and the lower the probability of optimal dispersion.”

Epiphyllum phyllanthus via Useful Tropical Plants

While some viviparous seedlings of mangroves can travel long distances from their parent plant and don’t always root into the ground immediately, they maintain their advantage over seeds because they can root in quickly upon reaching a suitable site and lift themselves up above rising tide waters. As the authors of the Epiphyllum study put it, vivipary is “a reproductive advantage that, in addition to allowing propagules to root and grow almost immediately, favors quick establishment whenever seedlings land on suitable substrates.”

There is still much to learn about this unusual and rare botanical feature. The research that does exist is relatively scant, so it will be interesting to see what more we can discover. For now, check out the following resources:

Also, check out this You Tube video of :

Attempts to Avenge the Acts of Cirsium arvense – A Biocontrol Story

Some weeds are so noxious, their crimes so heinous, and their control so challenging that desperation leads us to introduce other non-native organisms to contain them. Alien vs. alien duking it out in a novel environment. It seems counterintuitive – if an introduced species has reached the status of invasive, is it worth the risk of bringing in yet another foreign species in attempt to defeat it? We all know what happened to the old lady who swallowed the spider to catch the fly, yet for decades now we have been doing just this. It’s something we call classical biological control – introducing pathogens, insects, or other organisms to help control the spread of problematic ones.

Such attempts mostly fail, but we keep trying. The attempts made on Cirsium arvense exemplify this. The trouble is that even when such efforts fail, they aren’t always benign, as we shall see.

Canada thistle, a misnomer for Cirsium arvense, is a European native that has been acting in the role of noxious weed for centuries, even in its native land. First introduced accidentally to eastern North America sometime in the 1600’s, it has made its way across the continent and has since become one of our worst weeds in both natural and agricultural settings, as well as in our yards and gardens. Its seeds get around, carried by wind and water, attached to animals or deposited in their dung, stowing away as contaminants in crop seed or passengers in the ballast water of ships. But casual dispersal by seed isn’t quite as troubling as what it does once it takes root.

Several related species of thistle are also pesky weeds, but unlike Cirsium arvense, they are mostly annuals or biennials, spreading only by seed. Cirsium arvense is a perennial plant with roots that spread deep and wide. New shoots form readily along the spreading roots, forming a veritable thicket of stems that can be dozens of feet wide and giving the plant a more appropriate common name, creeping thistle.

The stems of creeping thistle can grow more than four feet tall and are adorned with alternately arranged, prickly, lobed leaves. Groups of small, urn-shaped flowerheads are born at the tops of stems. Flowers are pink to purple, sweet smelling, and favored by pollinators. Individual plants either produce all male flowers or all female flowers, and since individual plants are actually large colonies, an adjacent colony of the opposite sex is necessary in order for the production of viable seeds. Like other plants in the aster family, the seeds come with a feathery pappus, suggesting wind dispersal. However, the pappus is often weakly attached, sloughing off without seeds in tow, leaving them to the fate of gravity.

flowers of creeping thistle (Cirsium arvense) via eol

It comes as no surprise that when plants readily spread by root, stolon, or rhizome, they are well suited to become some of our most bothersome weeds. Eliminating their seed heads does little to reduce their spread. Pulling them out of the ground is futile; you will never get all the roots. Tilling them under only aids in their dispersal since chopped up roots and stems now have the chance to produce new plants. Herbicide treatments can set them back, but they must be repeated on a long-term and exacting schedule in order to thoroughly kill the roots. Considering what we’re up against when it comes to plants like creeping thistle, it makes sense why we would introduce foreign fighters to do our bidding, especially if such fighters are enemies of the plant in their native land.

The list of insects that have been employed (or at least considered) in the fight against creeping thistle is extensive. It includes thistle tortoise beetle (Cassida rubiginosa), seedhead weevil (Rhinocyllus conicus), thistle crown weevil (Trichosirocalus horridus), thistle gall fly (Urophora cardui), thistle stem weevil (Ceutorhynchus litura), thistle bud weevil (Larinus planus), seedhead fly (Orellia ruficauda), thistle flea beetle (Altica carduorum), thistle leaf beetle (Lema cyanella), painted lady (Vanessa cardui), and sluggish weevil (Cleonus piger). Unfortunately, and perhaps not surprisingly, as Bugwood reports, “biocontrol currently provides little or no control of Canada thistle populations, although some agents weaken and kill individual plants.” Despite the fact that there are well over 100 known organisms that consume or attack Cirsium arvense, nothing manages to do long-term damage.

thistle tortoise beetle (Cassida rubiginosa) – a common biocontrol agent of invasive thistle species (via wikimedia commons)

The status of creeping thistle biocontrol efforts on two South Dakota wildlife refuges was reported on in a 2006 issue of Natural Areas Journal. Multiple introductions of at least half a dozen different insect species had occurred beginning in 1986. Nearly 20 years later, they were not found to have had a significant effect on creeping thistle populations. The authors concluded stating they “do not advocate further releases or distribution in the northern Great Plains of the agents” examined in their study. They also advised that “effectiveness be a primary consideration” of any new biocontrol agents and expressed concern that some introduced insects have the potential to attack native thistles.

North America is home to quite a few native thistles, several of which are rare or threatened. A USDA guide to managing creeping thistle in the Southwest highlights the importance of protecting native thistles – “especially rare or endangered species” – from biocontrol agents and gives two examples of endangered thistles in New Mexico that are at risk of such agents.

The federally threatened species, Pitcher’s thistle (Cirsium pitcheri), which is restricted to sand dune shorelines along the upper Great Lakes, has quite a bit working against it. An added blow came a few years ago when it was discovered that the flowerheads of Pitcher’s thistle were being damaged by the thistle bud weevil (Larinus planus), a biocontrol agent employed against creeping thistle in the area. A paper published in Biological Conservation in 2012 examining the extent of weevil damage on the rare thistle cautioned that, “although some biological control agents may benefit some rare plant taxa, the negative impacts of both native insects and introduced herbivores are well documented.”

Pitcher’s thistle (Cirsium pitcheri) via eol

Classical biological control, if and when it works, can be quite valuable, especially if it reduces the need for other management inputs like herbicides and cultivation. Unfortunately, it is rarely successful and can have unintended consequences. Goldson et al. report in a 2014 issue of Biological Conservation that the success rate is only around 10% and that even that 10% is at risk of failing at some point. In his book, Where Do Camels Belong?, ecologist Ken Thompson cites that “only about one in three species introduced as biological controls establish at all, and only half of those that do establish (i.e. about 16% of total attempts) control the intended enemy,” adding that “biological control is just another invasion, albeit one we are trying to encourage rather than prevent, and its frequent failure is another example of how poorly we understand the effects of adding new species to ecosystems.”

Still, while some warn against being too optimistic, others argue that it is an essential tool in the war against invasive species and, while acknowledging that a few introductions have gone awry, assert that “significant non-target impacts” are rare. Clearly, this is a rich topic ripe for healthy debate and one that I will continue to explore. If you have thoughts or resources you’d like to share, please do so in the comment section below.


This post was inspired in part by episode six of The Shape of the World podcast. I highly recommend listening to the entire series.

Death by Crab Spider, part two

Crab spiders that hunt in flowers prey on pollinating insects. Thus, pollinating insects tend to avoid flowers that harbor crab spiders. We established this in part one. Now we ask, what effect, if any, does this interaction have on a crab spider infested plant’s ability to reproduce? More importantly, what are the evolutionary implications of this relationship?

In a study published in Ecological Entomology earlier this year, Gavini, et al. found that pollinating insects avoided the flowers of Peruvian lily (Alstroemeria aurea) when artificial spiders of various colors and sizes were placed in them. Bumblebees and other bees were the most frequent visitors to the flowers and were also the group “most affected by the presence of artificial spiders, decreasing the number of flowers visited and time spent in the inflorescences.” This avoidance had a notable effect on plant reproduction, namely a 25% reduction in seed set and a 15% reduction in fruit weight. The most abundant and effective pollinator, the buff-tailed bumblebee, was deterred by the spiders, leading the researchers to conclude that, “changes in pollinator behavior may translate into changes in plant fitness when ambush predators alter the behavior of the most effective pollinators.”

Peruvian lily (Alstroemeria aurea) via wikimedia commons

But missing from this discussion is the fact that crab spiders don’t only eat pollinators. Any flower visiting insect may become a crab spider’s prey, and that includes florivores. In which case, crab spiders can benefit a plant, saving it from reproduction losses by eating insects that eat flowers.

In April of this year, Nature Communications published a study by Knauer, et al. that examined the trade-off that occurs when crab spiders are preying on both pollinators and florivores. Four populations of buckler-mustard (Biscutella laevigata ssp. laevigata) were selected for this study. Bees are buckler-mustard’s main pollinator, and in concurrence with other studies, they significantly avoided flowers when crab spiders were present.  Knauer, et al. also determined that bees and crab spiders are attracted to the same floral scent compound, β-ocimene. This compound not only attracts pollinators, but is also emitted when plants experience herbivory, possibly to attract predators to come and prey on whatever is eating them.

buckler-mustard (Biscutella laevigata) via wikimedia commons

In this study, the predators called upon were crab spiders. Florivores had a notable impact on plants in this study, and the researchers found that when crab spiders were present, florivores were significantly reduced, thereby reducing their negative impact. They also noted that “crab spiders showed a significant preference for [florivore-infested] plants over control plants.”

And so it is, a plant’s floral scent compound attracts pollinators while simultaneously attracting the pollinator’s enemy, who is also called in to protect the flower from being eaten. Luckily, in this case, buckler-mustard is easily pollinated, so the loss of a few pollinators isn’t likely to have a strong negative effect on reproduction. As the authors write, “pollinators are usually abundant and the low number of ovules per flower makes a few pollen grains sufficient for a full seed set.”

crab spider on zinnia

But none of these studies are one size fits all. Predator-pollinator-plant interactions are still not well understood, and there is much to learn through future research. A meta-analysis published in the Journal of Animal Ecology in 2011 looked at the research that had been done up to that point. Included were a range of studies involving sit-and-wait predators (like crab spiders and lizards) as well as active hunters (like birds and ants) and the effects of predation on both pollinators and plant-eating insects. They concluded that where carnivores “disrupted plant-pollinator interactions, plant fitness was reduced by 17%,” but thanks to predation of herbivores, carnivores helped increase plant fitness by 51%. This suggests that carnivores, overall, have a net positive effect on plant fitness.

Many pollinating insects have an advantage over plant-eating insects because they move quickly from flower to flower and plant to plant, unlike many herbivores which move more slowly. This protects pollinators from predation and helps explain why plant-pollinator interactions are not disrupted as easily by carnivores. Additionally, as the authors note, “plants may be buffered against loss of pollination by attracting different types of pollinators, some of which are inaccessible to carnivores.”

But again, there is still so much to discover about these complex interactions. One way to gain a better understanding is to investigate the effects of predators on both pollinators and herbivores in the same study, since many of the papers included in the meta-analysis focused on only one or the other. As far as crab spiders go, Knauer, et al. highlight their importance in such studies. There are so many different species of crab spiders, and they are commonly found on flowers around the globe, so “their impact on plant evolution may be widespread among angiosperms.”

In other words, while we still have a lot to learn, the impact these tiny but skillful hunters have should not be underestimated.

Death by Crab Spider, part one

When a bee approaches a flower, it is essentially approaching the watering hole. It comes in search of food in the form of pollen and nectar. As is this case with other animals who come to feed at the watering hole, a flower-visiting bee makes itself vulnerable to a variety of predators. Carnivores, like the crab spider, lie in wait to attack.

The flowers of many plants rely on visits from bees and other organisms to assist in transferring pollen from stamens to stigmas, which initiates reproduction; and bees and other flower visitors need floral resources to survive. Crab spiders exploit this otherwise friendly relationship and, in doing so, can leave lasting impacts on both the bees and the flowers they visit.

Species in the family Thomisidae are commonly referred to as crab spiders, a name that comes from their resemblance to crabs. Crab spiders don’t build webs to catch prey; instead they either actively hunt for prey or sit and wait for potential prey to happen by, earning them the name ambush predators. Of the hundreds of species in this family, not all of them hunt for prey in flowers; those that do – species in the genera Misumena and Thomisus, for example – are often called flower crab spiders.

white crab spider (Thomisus spectabilis) on Iris sanguinea — via wikimedia commons

Most crab spiders are tiny – mere millimeters in size – and they have a number of strategies (depending on the species) to obscure their presence from potential prey. They can camouflage themselves by choosing to hunt in a flower that is the same color as they are or, in the case of some species, they can change their color to match the flower they are on. Some species of crab spiders reflect UV light, which bees can see. In doing so, they make themselves look like part of the flower.

Using an Australian species of crab spider, researchers found that honey bees preferred marguerite daisies (Chrysanthemum frutescens) on which UV-reflecting crab spiders were present, even when the scent of the flowers was masked. The spiders’ presence was seen as nectar guides, which “bees have a pre-existing bias towards.” Members of this same research team also determined that both crab spiders and honey bees choose fragrant flowers over non-fragrant flowers, and that, ultimately, “honey bees suffer apparently from responding to the same floral characteristics as crab spiders do.”

Needless to say, crab spiders are crafty. So the question is, when killing machines like crab spiders are picking off a plant’s pollinators, does this affect its ability to reproduce? First let’s consider how pollinators react to finding crab spiders hiding in the flowers they hope to visit.

goldenrod crab spider (Misumena vatia) preying on a pollinator — via wikimedia commons

A study published in Oikos in 2003 observed patches of common milkweed (Asclepias syriaca) – one set was free of crab spiders, the other set was not – and tracked the visitations of four species of bees – the common honey bee and three species of bumble bees. They compared visitation rates between both sets of milkweed patches and found that the smallest of the three bumble bee species decreased its frequency of visitation to the crab spider infested milkweeds. Honey bees also appeared to visit the infested milkweeds less, but the results were not statistically significant. The two larger species of bumble bees continued to forage at the same rate despite the presence of crab spiders.

During the study, crab spiders were seen attacking bees numerous times. Six attacks resulted in successful kills, and of the bees that escaped, 80% left the flower and either moved to a different flower on the same plant, moved to a different plant, or left the patch altogether. These results indicate a potential for the presence of crab spiders to effect plant-pollinator interactions, whether its directly (predation) or indirectly (bees avoiding flowers with crab spiders).

Another study published in Behavioral Ecology in 2006 looked at two species of bees – the honey bee and a species of long-horned bee – and their reactions to the presence of crab spiders on the flowers of three different plant species – lavender (Lavandula stoechas), crimson spot rockrose (Cistus ladanifer), and sage-leaf rockrose (Cistus salvifolius). Honey bees were about half as likely to select inflorescences of lavender when crab spiders were present, and they avoided the crab spider infested flowers of crimson spot rockrose with a similar frequency. On the other hand, the long-horned bee visited the flowers of crimson spot rockrose to the same degree whether or not a crab spider was present.

bee visiting sage-leaf rock rose (Cistus salvifolius) — via wikimedia commons

The researchers then exposed honey bees to the flowers of sage-leaf rockrose that were at the time spider-free but showed signs that crab spiders had recently visited. Some of the flowers featured the scent of crab spiders, others had spider silk attached to them, and others had the corpses of dead bees on them. They found that even when crab spiders were no longer present, the bees could still detect them. Honey bees were particularly deterred by the presence of corpses. The long-horned bees were also exposed to the flowers with corpses on them but didn’t show a significant avoidance of them.

An interesting side note about the presence of silk on flowers. As stated earlier, crab spiders do not spin webs; however, they do spin silk for other reasons, including to tether themselves to flowers while hunting. The authors recount, “on several occasions when an attempted attack was observed during this study, it was only the presence of a silk tether that prevented spiders being carried away from flowers by their much larger prey.”

So, again, if bees are avoiding flowers due to the presence of predators like crab spiders, what effect, if any, is this having on the plants? We will address this question in part two.

When Milkweed Kills

When you think of milkweed, you probably think of the life it supports. The monarch butterfly, for one. As the sole food source for its leaf-eating larvae, monarchs would be a thing of the past if milkweeds disappeared. Numerous other insects feed on its foliage as well, and there are a plethora of organisms that feed on its nectar, including bees, butterflies, beetles, wasps, and other insects, as well as hummingbirds. And speaking of birds, some birds use the silky hairs attached to the seeds to line their nests, while other birds strip stringy fibers from the stems for nest building. And while it is not a major food source for mammals, deer and other animals have been known to sample it. Indeed, milkweed is a veritable life force.

red milkweed beetle (Tetraopes sp.) feeds on milkweed

But it’s also a poisonous plant. The latex sap of milkweed contains cardiac glycosides, among a variety of other toxic chemicals. The plant produces these chemicals to defend itself from herbivory, and so the insects that feed on it have adapted a variety of strategies to avoid being poisoned. Some bite a hole in a leaf vein and wait for the milky sap to drain before proceeding to eat the leaf. Others are able to consume the toxic foliage without being poisoned by it. Some even store the toxic chemicals in their bodies, making themselves poisonous to other organisms that dare consume them.

Aphids on Mexican whorled milkweed (Asclepias fascicularis). One species commonly found on milkweed is the oleander aphid (Aphis nerii), an introduced species that feeds on milkweeds and other plants in the dogbane family.

While milkweed is generally found to be unpalatable to most livestock, those that venture to eat it risk being poisoned and even killed. A guide to milkweed written by the Xerces Society states, “sheep and goats are the most likely to be poisoned because they are browsers and often prefer to feed on weeds over other forages.” Weeds of the West calls Utah milkweed (Asclepias labriformis) “the most poisonous of all western milkweeds,” claiming that “as little as one ounce of green leaf material … can kill an adult sheep.” It also lists swamp milkweed (Asclepias incarnata) as “suspected of causing livestock deaths.” To make matters worse, dead and dried milkweed plants retain their toxicity, which is a problem when they end up in animal feed.

Despite their toxicity, humans have been consuming milkweed for centuries. Young shoots and leaves can be eaten after boiling them several times, refreshing the water each time, and a medicinal tea can be made from the roots. While fatal poisonings of humans haven’t been reported, Nancy Turner and Patrick von Aderkas warn in their book The North American Guide to Common Poisonous Plants and Mushrooms that “uncooked shoots and the mature plants should never be consumed”

But milkweed’s toxic sap is not its only method for killing.

In fact, it may not even be its most deadly. And this is where things get interesting. Last month I arrived at work one morning to find a portion of a dried-up milkweed inflorescence on my desk that had been left there by a friend and co-worker. Stuck to the inflorescence were three, dead, dried-up honey bees, their legs trapped in the slotted hoods of the flowers. Apparently this is a common occurrence; one that is mentioned in nearly every resource about milkweeds that I have read now, and yet I had never heard of it nor seen it until this gift was left for me. I then went out to a patch of milkweed to see this for myself. Sure enough, I found a few dead bees trapped in the flowers of showy milkweed.

dead honey bee stuck in the flowers of showy milkweed (Asclepias speciosa)

Milkweed flowers do not always give up their pollen sacs easily. The slits where the pollinia are found can, on occasion, trap the legs of visiting insects. John Eastman describes this in The Book of Field and Roadside, “insects sometimes become permanently wedged as the fissures trap their feet or the pollinia entangle them, and they die hanging from the flowers.” While milkweed species are native to North America, honey bees are not; they have not evolved alongside the flowers of milkweed, yet they are drawn in, like so many other insects, to the nutritious and abundant nectar.

Native or not, honey bees are not the only insects getting trapped in the flowers. Eastman reports seeing various species of butterflies ensnared as well, and a paper by S.W. Frost lists cluster flies, soldier beetles, and a couple species of moths as unsuspecting victims of these unruly flowers. Frost goes on to observe that, “in spite of the hazards,” bees, wasps, and various other insects “visited the flowers of milkweeds freely.”

In a paper published in 1887, Charles Robertson describes the insect visitors of several different milkweed species. He found an occassional dead insect on the flowers of swamp milkweed, adding that “this occurs only when all or most of the feet are entangled simultaneously, so as to render the insect absolutely helpless.” Observing common milkweed (Asclepias syriaca), Robertson finds that “even when small and short-legged insects succeed in extracting pollinia and inserting them into the stigmatic chambers, they have great difficulty in breaking the retinacula, and often lose their lives in consequence.”

Honey bees were easily the most common victims observed in Robertson’s study, leading him to quip, “it seems that the flowers are better adapted to kill [honey bees] than to produce fruit through their aid.” And a honey bee’s trouble doesn’t always end when she escapes the grasp of the flowers. Pollinia and its connecting tissues can get so tangled around her legs and other body parts that she can no longer forage, subjecting herself to starvation and predation.

To add insult to injury, dead and dying insects stuck to flowers result in another interesting phenomenon. Robertson writes, “many fall prey to predacious insects. I have seen them while still alive, attacked by ants, spiders and [predatory stink bugs].” Eastman adds daddy longlegs to the list of “scavengers” or “cleanup specialists” that come to feed on “flower trapped insects.” As it turns out, visiting the flowers of milkweed can be a dangerous, even deadly, game.

See Also: Idaho’s Native Milkweeds