Wild With Nature

July 27, 2022

She’s fuzzy and yellow, the size of my thumb. The pollen basket on her hind leg holds an orange loaf as she moves methodically from flower to flower. She forces her head down inside a greenish-white Canada milkvetch bloom (Astragalus canadensis), then wiggles onwards to the one above it. We can see at least three bumblebees working this milkvetch patch right now, down here among the sapling willows along the creek. 

I’m here this morning with my friends Greta and Augie Dobrecevich, hoping to learn something from the bumblebees. These large, fuzzy pollinators are easy to notice. They’re also fairly straightforward to identify in the field, without having to collect them. Besides, I’ve talked with a few Helena-area folks recently who have commented on how few bumblebees they’ve seen this spring. All of this adds up to an interesting start for a day in the field. 

How many bumblebees can we find today? What flowers are they visiting? And what can they teach us about this landscape?

It’s already hot this morning. The air is still. The distant mountains are a smoky blue from far-off fires. Once in a great while, a meadowlark sings. Most of our early summer birdsong has already dried up. We’re entering the quiet at the peak of summer. It’s the season of grasshoppers and young birds. It’s the season of flowering sweet clover, nodding thistle, and wild licorice. Is it the season for bumblebees, too? Today we’re hoping to find out.

An intro to bumblebees

Bumblebees are generalist pollinators – which means that just about any type of flower is fair game in our search. But they definitely do have preferences. Some species have long tongues to reach deep inside tubular flowers. Others have short tongues, useful for more-accessible blooms. In the past, I’ve found some species very frequently on plants in the pea family (Fabaceae), such as wild licorice. Others seem to adore thistles. And some flowers, like prostrate vervain (Verbena bracteata), a tiny-flowered native mat-former, seem completely uninteresting to our bumblebees.

Today our field gear is simple. We each have an insect net and a pocket full of plastic vials with snap-on lids. When we find bumblebees on flowers, we’ll try to net them and then transfer each one to a vial. In the shade, we’ve set up a cooler full of ice. We’ll place the not-very-happy bumblebees, in their vials, in the cooler to chill down. Once they stop buzzing around, we’ll be able to remove them and identify them. Then, we’ll let them warm up again and fly off. A cool bumblebee is amazingly docile. They’ll cling gently to a finger as they buzz their wings and raise their body temperature to flight range. I’ve never had a bumblebee try to sting me as it is warming up.

Important note: if you try this at home, make sure to use a cooler with ice or a refrigerator to chill the bees. Freezers are way too cold – they will kill bumblebees, not chill them.

For identification, I am using an excellent Forest Service guide, Bumble Bees of the Western United States. I’m supplementing this guide with an updated identification key created by Montana State University. This key covers female bumblebees of all species known or expected in Montana.

Buzzing in the milkvetch

I’m still watching the same bumblebee in the Canada milkvetch. I raise the net, gauge the distance to her flower, and swing. She tumbles in and begins buzzing ferociously, clearly not happy about this interruption to her breakfast. A pungent smell wafts up from the unhappy bee. To me, it smells exactly like a honey and lemon toddy. I maneuver a vial into the net and ease her in. 

Though it’s early in the day, it’s already clear that the Canada milkvetch is a great bumblebee plant. It’s not long before we have a handful of them in vials, buzzing their displeasure. I carry them up to the cooler, where I check on the first bumblebee of the morning. Augie found this one right as we were starting out, visiting small tumble-mustard (Sisymbrium loeselii) in an area of disturbed soil near where we parked. She’s smaller than the bees on the milkvetch, with a striking band of orange hairs across her abdomen. And she’s already cooled down enough for photos.

Using a 10x lens for magnification, I examine her. She has a moderately long cheek. Funny enough, cheek length is often a critical characteristic to look for when identifying bumblebees. Besides the conspicuous orange hairs on her abdomen, she also has a black stripe across her thorax, sandwiched by yellow. She’s a Hunt’s bumblebee (Bombus huntii), one of the species I see very commonly around Helena. 

She crawls onto my finger and begins to warm up, shaking her wings and buzzing slightly. She moves to my fingertip, preens, and bobs her abdomen up and down. And then she flies off.

Some flowers are tastier

It’s not just interesting to see which flowers the bumblebees are visiting. It’s also interesting to see which ones they aren’t visiting. On my way back to the milkvetch, I stop to check a patch of Canada goldenrod (Solidago canadensis). It’s recently come into bloom, an incredible array of deep, dazzling yellow. And it’s buzzing with pollinators. There are black wasps, yellow-and-black wasps, white-striped wasps, and green wasps. I see a fuzzy orange bee fly on the flowers. But I don’t see a single bumblebee.

I stop to check in with Greta, who is watching a tangle of white clematis (Clematis ligusticifolia). There are wasps visiting these flowers, but no bumblebees here, either.

I’m starting to get a sense of how this landscape might look through bumblebee eyes. It’s not just a pretty tapestry of grasses, shrubs, and flowers. There are lots of resources here for a bumblebee in search of pollen and nectar, but they’re patchy. They’re few and far between. From a bumblebee’s perspective, this landscape must look like a map of flower patches. Canada milkvetch, it seems, is highlighted on this map. Meanwhile, white clematis and goldenrod don’t even show up. And these flower patches exist in a sea of mostly uninteresting grasses, connected by a bumblebee’s memory and flight. 

Flies in the snowberry patch

Now I’m standing in a thick patch of western snowberry (Symphoricarpos occidentalis). I’ve seen bumblebees visiting snowberry flowers before, so I’m hoping I’ll find some here today. Surely they must be here, I think, among all of these sweet-scented, light pink bells. But so far, I’m not seeing any. Instead of the powerful, insistent buzz of the bumblebees, I’m hearing a higher-pitched, soft buzz. The sound is coming from several hairy black flies, as large as a bumblebee, with a dab of orange near their wing bases. These are tachinid flies (family Tachinidae), a group of parasitoids as strange as the ichneumonid wasps I looked at last month. Tachinid flies are very diverse, and many of them are quite colorful. In general, they tend to be medium or large, hairy, and extremely active. These large black ones are likely members of the genus Tachina. Little is known about the biology of this genus, but several Tachina species parasitize cutworms.

I spend several more minutes circling the snowberry patch, watching carefully and listening for the roar of a bumblebee. But none appear. Today, at least, the snowberry is not on the bumblebees’ floral map.

The most popular flowers

I find myself back along the creek in another patch of Canada milkvetch. This one is still in full bloom, and the bumblebee activity here is incredible. Immediately in front of me, five of these large, fuzzy pollinators are going from flower to flower. It’s a full-time job just netting them and putting them in vials. 

I find Greta across the creek; Augie is searching farther downstream. As we cross back over to finish searching the milkvetch patch, we hear a distinctive whistle of wings. It’s a mourning dove, flushing from a chokecherry thicket past our shoulders. We turn around to look. There among the branches, in the cover of a white clematis vine, is a mourning dove nest. Two half-grown nestlings, covered with pin feathers, are looking back at us from the haphazard platform.

We leave the mourning dove chicks in peace and return to the milkvetch patch. The bee diversity is both exciting and overwhelming. There are dozens of bumblebees, with at least three species here. We watch them, swing nets, and transfer massive, grumpy bees into vials. 

It’s time for a break: we’ve filled all of the vials in our pockets with bees. We return to the cooler and compare notes. What Greta and Augie have seen matches with my observations: milkvetch seems to be the most popular flower by far. Augie mentions finding a few bumblebees on Rocky Mountain beeplant (Cleome serrulata), as well.

Identifying bumblebees

Now it’s time to identify the bumblebees from earlier. We take them out one by one. As with the Hunt’s bumblebee I already released, we’re looking at general pattern, coloration, and cheek length. All of these bees are pollen-collecting females. We can see the shiny, concave pollen baskets on their hind legs. 

We have to work fast. Soon after we remove a bumblebee from the cooler, she begins to move, twitching a leg. It doesn’t take long before she stretches her legs and begins to crawl. It’s absolutely endearing – but it means that identification has to happen fast. It’s very hard to examine a bumblebee’s cheek under magnification when she’s crawling around and preparing for takeoff.

This one is a white-shouldered bumblebee (Bombus appositus). She has a strikingly white band of hairs across the front of her thorax, and her abdomen is mostly yellow. After I get done identifying her, I hold her on my finger. She twitches her legs and bends her abdomen down, buzzing it slightly. We transfer her to Greta’s finger and she clambers aboard, allowing us to admire her intricate fuzziness. And then, without warning, she takes off, buzzing heavily away.

Preference and bias

We talk about the patterns we’re seeing. Clearly the bumblebees are loving the milkvetch today. We’ve seen dozens of them visiting it. On other plants, the bumblebee attention has been sparse. So far we’ve just spotted a few on the small tumble-mustard and a few on the Rocky Mountain beeplant.

That’s been it. We’ve been checking other flowers, too: snowberry, goldenrod, clematis. We’ve been trying to check as many different plants as possible. We know that if we assume that milkvetch is attractive and neglect other flowers, its attractiveness may reflect our bias rather than a real pattern. But so far, we haven’t found any other flower patches that can compare in terms of bumblebee interest.

Most bumblebees have a long flight season, much longer than the flowering period of a single plant. So when we think about bumblebee habitat on this landscape, we have to remember that it’s not just about what we see today. It’s very possible that I might have seen bees from these same colonies visiting the two-groove milkvetch (Astragalus bisulcatus) here a month ago. In another month, they may be choosing between Rocky Mountain beeplant, white sweetclover (Melilotus alba), and Nuttall’s sunflower (Helianthus nuttallii) along this stream. If we want to encourage bumblebees, we need to think about a full season of attractive flowers for them.

We’re reaching the end of the morning, and Greta and Augie have to leave. All of the bees we just caught on the milkvetch are cooling down. I decide to do one more foray, going farther afield and specifically looking for floral diversity. Besides the milkvetch, what else are the bumblebees visiting today?

Dryland flowers

I walk out into the dry grassland away from the stream. Here and there, I spot a clump of alfalfa (Medicago sativa), bearing a mixture of deep purple blooms and developing fruits. The hairy goldenaster (Heterotheca villosa) is a patchwork of bright yellow flowers and tawny seed tufts. I do notice a few bees visiting it, but these are smaller species with less hair. The buzzing of bumblebees is nowhere to be heard.

There’s a large patch of Russian knapweed (Acroptilon repens) in a swale running through the grassland. It’s a lumpy, bushy expanse of green, pink, and white in the hot midday breeze. I’m not seeing bumblebees here yet, but this patch is bobbing with activity. There are a few honeybees and the occasional, golden-haired bee fly. But the most noticeable thing is all of the nectaring butterflies. There are checkered whites (Pontia protodice), common wood nymphs (Cercyonis pegala), clouded sulphurs (Colias philodice), and a few blues. They dance nimbly, landing on the flowers and fluttering upwards again. 

Dryland bumblebees

And then I spot a small bumblebee, moving from flower to flower. I inch the net closer. It’s trickier to catch bumblebees in the Russian knapweed than it was in the Canada milkvetch. The stems are stiff and the foliage is dense. I sweep. The bumblebee tumbles into the net. 

The bumblebee abundance here is still nowhere near what we saw on the milkvetch earlier this morning. But still, it’s clear that this patch is on their foraging map. Within a few minutes, I catch five of them – all relatively small, fuzzy worker bees. 

I notice elongate, sausage-like swellings along some of the Russian knapweed stems. These are galls formed by the larvae of the Russian knapweed gall wasp (Aulacidea acroptilonica), a tiny biocontrol insect that reduces the seed production of this non-native plant. These wasps seem to be well-established in this patch.

Farther along the swale, I find another rank, yellow patch of small tumble-mustard (Sisymbrium loeselii). Several bees are visiting these flowers, including another small bumblebee.

A hot breeze is blowing as I walk across the grassland. Thunderheads are building over the mountains. I continue searching, crossing expanses of dry grasses without any flowers at this season. These areas must seem desolate to a bumblebee. I’m probably walking past some bumblebee nests, though. These fuzzy insects are usually ground-nesters, often reusing a rodent burrow for their small colonies.

Knapweed and toadflax

I cross the creek again and find myself on a dry hillside, in an extensive patch of yellow and pale purple. It’s a mixed stand of spotted knapweed (Centaurea stoebe) and dalmatian toadflax (Linaria dalmatica). I know what people think about these plants. They’re among our most hated weeds, both on the Montana noxious weeds list. But today, I’m wondering what the bumblebees think of these plants.

Right away, I find a large female visiting the toadflax. She goes from blossom to blossom, dipping her head inside. I can hear several others in the vicinity. They’re skittish, visiting a few flowers and then departing in a long-distance flight. Nevertheless, I manage to net several of them. From a bumblebee’s perspective, the toadflax appears to be of some interest. It’s not anywhere near as popular as the Canada milkvetch, but it’s definitely more interesting than goldenrod or clematis. 

What about the licorice?

My pocket is full of bumblebees now. I only have one empty vial left. But there’s one more plant I’d like to check today. It’s wild licorice (Glycyrrhiza lepidota), growing down along the stream. Like Canada milkvetch, this robust native plant is in the pea family (Fabaceae). It’s passed peak flowering by now. Many of the once-white flowers have dried up already. In their place, the poky green fruits are swelling up. Once they ripen, they’ll act like peapods with burs, catching on passing animals and dispersing to new locations. 

Even though most of the licorice flowers are fading, right away I find several more small bumblebees, gathering the last bits of the season’s sustenance from this patch. With their rusty-banded abdomens, they all appear to be Hunt’s bumblebees. I watch one of them for a while as she moves diligently from flower to flower. Then I net her.

In other years, when the wild licorice has been at peak bloom, I’ve noticed that it seems almost as popular as the Canada milkvetch. At these times, the licorice has been teeming with bumblebees. Today it’s not quite that exciting. Except for these few rusty-banded females, the bumblebees have turned their attention elsewhere. It’s another illustration of seasonality: even a few weeks can make a big difference in terms of flower popularity.

I know that there must be more bees out here today, but I’m ready to wrap up my observations. We’ve netted 35 bumblebees today. Who are they all?

Patterns in the bumblebees

I return to the cooler. It’s time to identify the remaining 30 bees. I pull them out one at a time, running them through the identification key. Gradually, some patterns emerge. On the Canada milkvetch, three species are common. As we’ve already seen, the white-shouldered bumblebees (Bombus appositus) are mostly yellow, with a striking white band across the thorax. Nevada bumblebees (Bombus nevadensis) often have a black dot between their wings, surrounded by yellow hairs. The third common species on the milkvetch is the golden northern bumblebee (Bombus fervidus), a bright yellow creature with a furry black dash across the thorax.

The strikingly orange-banded Hunt’s bumblebee (Bombus huntii) is a milkvetch visitor, too, but this species seems to be a true generalist. We’ve caught at least one of these on literally every single plant that we’ve found bumblebees on today.

Every one of the 35 bees we’ve caught today has been a female. This makes sense: male bumblebees are most common in the fall, when they’re searching for queens to mate with.

Bumblebees are social insects. This makes them like honeybees but unlike the vast majority of our other bee species. And while honeybees have massive, long-lived colonies with many workers, bumblebees start over again each year. Only the queens overwinter. In the spring, they forage and start new colonies, raising the first generation of worker bees. From then on, the queen stays home and the workers forage.

Queens vs. workers

Bumblebee queens are massive, while the workers tend to be smaller. Most of the bees we’ve found on the Canada milkvetch are huge – so these seem to be queens that are still raising their first-generation workers. In comparison, the Hunt’s bumblebees we’re seeing are diminutive. These are almost certainly workers. The Hunt’s bumblebee queens were visiting flowers earlier in the season. Now they’re staying home, laying more eggs.

The Rocky Mountain beeplant (Cleome serrulata) yields an interesting bee. Like all of our bumblebees today, this is a female, with 12 antennal segments (males have 13). But the hind tibiae of this one are entirely hairy. There’s no shiny, concave pollen basket. Why?

This one looks like a bumblebee – but it’s not! (This one had me fooled. Dr. Casey Delphia, a bee biologist from Montana State University, had to correct me on it.) Instead, this hairy bee is in the genus Anthophora, a large group of bees that generally nest in the ground. Unlike bumblebees, Anthophora females tend their nests on their own, without help from workers – though females of some Anthophora species will share a single entrance hole to their nests.

Different bees on the knapweed

When I switch to the Russian knapweed vials, things really start to get interesting. The first of these bees is a small female with pollen in her baskets. She has orange on her abdomen, like the Hunt’s bumblebees. But the banding isn’t as precise. And looking at her face with my 10x lens, I can see that her cheek is very short. I also notice that the hairs on her face are black, while Hunt’s bumblebees have yellow hairs here. This is a red-belted bumblebee (Bombus rufocinctus). 

And she isn’t the only new species from the Russian knapweed patch. I’ve also caught several two form bumblebees (Bombus bifarius). These females are decked out in black and yellow stripes.

As I move on to the bees from dalmatian toadflax (Linaria dalmatica), I return to some familiar species from earlier. Nevada bumblebees and Hunt’s bumblebees are the common visitors on the toadflax today.

Some of the others aren’t so easy. Sometimes I struggle through the key, debating about whether a cheek is truly long or short. In the end, I’m left with just one bee that remains unknown. I take photos, jot down notes, and hope that some friendly bee expert will be able to help me figure it out later. 

One by one, the bees preen, warm up, and fly off. Finally, I finish identifying the last one. As I had suspected, this rusty-banded female from the wild licorice is another Hunt’s bumblebee.

Counting them up

Among these 35 bees, we’ve identified six species. Plus there’s still that one bee that has me stumped. We’ve learned something today about each of these species: what flowers they like, what flowers they don’t like, and how common they are here.

• Hunt’s bumblebee (Bombus huntii): 8 females. Seen visiting every flower where we found bumblebees today: Russian knapweed (Acroptilon repens), Canada milkvetch (Astragalus canadensis), spotted knapweed (Centaurea stoebe), Rocky Mountain beeplant (Cleome serrulata), wild licorice (Glycyrrhiza lepidota), dalmatian toadflax (Linaria dalmatica), and small tumble-mustard (Sisymbrium loeselii).
• White-shouldered bumblebee (Bombus appositus): 6 females. Only seen on Canada milkvetch (Astragalus canadensis).
• Golden northern bumblebee (Bombus fervidus): 5 females. Only seen on Canada milkvetch (Astragalus canadensis).
• Nevada bumblebee (Bombus nevadensis): 11 females. Seen on Canada milkvetch (Astragalus canadensis) and dalmatian todaflax (Linaria dalmatica).
• Red-belted bumblebee (Bombus rufocinctus): 2 females. Only seen on Russian knapweed (Acroptilon repens).
• Two form bumblebee (Bombus bifarius): 2 females. Only seen on Russian knapweed (Acroptilon repens).

It’s a start

I could spend hours more – no, days more – out here learning about this community. Eventually, I might be able to see this area like a bumblebee does: a four-dimensional patchwork of flowers coming in and out of season, some more interesting than others. There are so many more questions. Are there other bumblebee species here? What other flowers are attractive to them? With more time, I could watch the bees’ behavior and learn the dances they use to collect pollen and nectar.

But for today, I’m content with our foray. We’ve begun to learn which flowers the bumblebees like. We’ve gotten up close and personal with an abundance of bees, far more of them than I would have guessed we might find here. Maybe next summer we can do this again.

Learning from the bees

In the meanwhile, what are our takeaways? First, there are a lot of bumblebees out here. And when we take the time to really look for them, they have so much to teach us. Second, Canada milkvetch is a really popular plant right now for bumblebees. But that won’t last forever: the milkvetch flowers will fade, and the bees will have to look elsewhere for food. The milkvetch isn’t the answer for all of our bees, either. The red-belted bumblebees and the two form bumblebees showed no interest at all in it today. Instead, they visited Russian knapweed, another of those frequently-maligned plants on our state noxious weeds list. 

So if we want to support bumblebees, plants like Canada milkvetch can help. But one plant isn’t going to be enough: we’ll need attractive bumblebee flowers throughout the flight season. We’ll need plants with deep flowers, like the milkvetch, and plants with more-accessible blooms, like the knapweed. If we want bumblebees, we need plant diversity.

In terms of personality, bumblebees remind me of bears. They’re fuzzy, they’re remarkably photogenic, and they go about their business in a charming, bumbling way. After today, I definitely want to see more of them around. So when it comes to planting Canada milkvetch around Helena, sign me up.

Further Reading

Wilson, J.S. & Carril, O.M. (2016). The bees in your backyard: a guide to North America’s bees. Princeton, NJ: Princeton University Press.