What’s Really Going On Inside Those Galls?? (It’s Not Just Fly Larvae)

Many of you already know about the big round galls on the stems of goldenrod plants (there are other kinds of galls on goldenrod too). For those of you who aren’t familiar with the story, the galls are created when a goldenrod gall fly (Eurosta solidaginis) lays an egg on the stem, the larva hatches, and it burrows inside the plant. The goldenrod plant tissue grows rapidly around the larva, creating the gall. The fly then feeds on the plant tissue, shielded from danger by the plant, until it eventually pupates and emerges as an adult fly.

Canada goldenrod (Solidago canadensis) with Maximilian sunflower in the background. The Nature Conservancy’s Platte River Prairies.
A big round gall on a Canada goldenrod plant.
A goldenrod gall fly larva inside a gall. Note the size and shape of it.

It’s a nice story, but that version is way oversimplified. As is almost always true in nature, the more you look into a story, the more complex and fascinating it becomes. For example, that fly larva is not necessarily safe inside that gall.

There are two small chalcid wasp species that attack the little gall fly larva. One wasp species (Eurytoma obtusiventris) lays an egg inside the fly larva before it burrows inside the plant, initiating a sequence of events that turns out poorly for the fly. After the fly larva enters the plant and the gall forms, the wasp larva inside the fly larva hatches (cue creepy music). It starts eating the fly larva from the inside, also stimulating it to create a pupa much earlier than it normally would. That pupa provides a nice shelter, within which the wasp larva completes its consumption of the fly and then creates its own pupa before eventually emerging as an adult wasp.

The second wasp, Eurytoma gigantea, lays its egg inside the gall itself. The female wasp has a long ovipositor that it can insert into the gall, though it can only do so on smaller galls because its ovipositor is only so long. When the wasp egg hatches, the fly larva gains an unexpected and very dangerous roommate. The wasp larva tears the fly larva apart with its mouthparts and eats it. Then it starts eating the same plant tissues the fly would have eaten.

The two wasp species have very different approaches to killing the hapless fly larva, but both are very effective. Rarely, individuals of both wasp species end up sharing the same gall. After dispatching the fly, the two wasps can sometimes both survive and exit the gall after they pupate and become adults. Honor among parasitoids, I guess, but I’m not sure how they both come out of that situation alive. According to researchers, it happens.

I’ve known the basics of the gall fly and wasp stories for years now, though I usually don’t remember the specifics of the wasp tactics. As a result, when I spontaneously decided to collect and cut open a few goldenrod galls I found on a recent hike, I wasn’t surprised that two of the galls contained a creature that definitely wasn’t a fly larva.

This definitely isn’t a goldenrod gall fly larva. The shape is all wrong. If you look to the left of the larva, you can see what might be the remains of the fly larva.

However, as I studied and photographed the non-fly larvae, they didn’t look much like larval wasps. They were too skinny and their bodies were the wrong shape. To the internet I went…

That’s when I learned about Mordellistena unicolor, a tiny beetle that also feeds on the tissue inside galls created by goldenrod and the goldenrod gall fly. A female beetle lays an egg on the outside of the gall in mid-summer and it subsequently hatches and burrows inside. It then tunnels around inside the gall, feeding on whatever it finds – mostly the plant tissue of the gall itself.

As I read about the beetle, I learned a new vocabulary word: inquiline. An inquiline lives in the same home as another creature and (generally) doesn’t cause the host any harm. Some of the papers I read categorized M. unicolor as an inquiline. Others called it a predator. In both cases, the authors acknowledged that the beetle larva was likely to eat the larva of the goldenrod fly if it happened upon it. Does that make it a predator or just a really bad houseguest? Either way, the fly larva loses again, unless the beetle never comes across it.

I’m pretty sure the skinny larvae I found were baby beetles. I also found what looked like the desiccated remains of the gall fly in at least one of the galls inhabited by a beetle larva. I can’t confirm that the fly was killed by the beetle, but it was dead nevertheless.

Between the two wasps and a beetle, there’s a lot of traffic inside those goldenrod galls that our original story that sold galls as a safe haven for the larvae of goldenrod gall flies. “If the wasps don’t getcha, the beetle probably will”, would make a good cross stitch wall-hanging for fatalistic flies. And yet, enough fly larvae survive each year to continue the whole process. That’s lucky for the flies, but also for the wasps and beetle, whose entire life strategy hinges on the flies making galls.

Now, you might be wondering, “What happens if a wasp is already there when the beetle arrives?” It’s a good question. As far as I could find, the beetle is just as happy to munch its way through a wasp larva as it is a fly larva. And, of course, in the case of E. obtusiventris, the wasp larva is INSIDE the fly larva, so it’s a kind of turducken situation for the beetle. (I couldn’t come up with a good portmanteau of wasp, fly, and larva that works as well as ‘turducken’ but you know I tried.)

So, if the first wasp finds the fly larva before it gets in the gall, the fly dies. If the second wasp finds a gall that’s small enough, it’ll lay an egg inside the gall and the fly dies. A beetle might enter the picture and, if so, there’s a good chance that the fly will die then too – or whichever wasp had already dealt with the fly. But not always.

That’s definitely a more complex story than “Fly larva burrows into a stem and a gall forms to protect and feed it.” How about just a little more complexity?

Ok. During the winter, there are at least two bird species that see goldenrod galls as a great place to find a snack. Downy woodpeckers and black-capped chickadees are both known to drill holes into galls and extract the larva from within. In fact, the woodpecker can apparently get the larva out within 30 seconds, which is pretty spectacular. Downy woodpeckers tend to make small neat holes (often slightly conical), where chickadees make bigger, messier holes. Now you have something to look for on your next winter prairie hike!

I’m pretty sure this gall was opened up by a downy woodpecker, given the size and shape of the hole.

When I say the birds extract and eat the larvae, they do, but not if it’s the larva of Eurytoma obtusiventris (the wasp that eats the fly larva from the inside out). At least according to one study from 1974-75, if a woodpecker opened a hole to a gall and found that wasp larva, it tended to just leave it there. There was no report on what happened if a woodpecker found the other wasp or the beetle, and no statement on the preferences of chickadees regarding larval species.

What have we learned, then? The goldenrod gall fly lays an egg on the stem of goldenrod and if its larva successfully hatches, it burrows inside the stem, causing the plant to create a big round gall around it. Inside, the fly larva starts to eat, unaware (we assume) that it might soon die a horrible death from one of many sources. It might get eaten by a wasp larva already inside its body, it might get torn apart and consumed by a different larval wasp, it might be fatally burrowed-through by a beetle larva, or it might be excavated and swallowed by a bird. If, by some miracle, none of those things happen, an adult goldenrod gall fly will emerge after the winter ends and try to keep the whole cycle going. I bet those successful flies have no idea how fortunate they are.

If you want to learn more, there are lots of research on goldenrod galls and the insects you can find inside them. Here are two examples. I wasn’t able to link you directly to the free PDF versions, but if you copy and paste the titles of these articles, you should be able to access them.

Mortality factors affecting Eurosta solidaginis (Diptera: Tephritidae). James H. Cane and Frank Kerczewski. Journal of the New York Entomological Society (1976) 84: 275-282.

Variation in selection pressure on the goldenrod gall fly and the competitive interactions of its natural enemies. Warren G. Abrahamson, Joan F. Sattler, Kenneth D. McCrea, and Aruther E. Weis. Oecologia (1989) 79:15-22

How Restored and Remnant Prairies Each Contribute Resources for Pollinators – Research by Hubbard Fellow Emma Greenlee

This post is written by Emma Greenlee, who recently completed her Hubbard Fellowship year here at The Nature Conservancy in Nebraska. Emma is now moving on to graduate school, where she’ll have the chance to study prairies in even more depth than she did with us. As her independent project during her Fellowship, Emma helped us evaluate the way our restored prairies contribute toward pollinator resources. Specifically, she counted the number of flowers and flowering plant species available in both remnant and restored (planted from seed) grasslands at our Platte River Prairies site. She is sharing some of the highlights from her findings in this post:

During my time as a Hubbard Fellow I conducted an ecology research project comparing flowering species community composition and floral resource abundance in remnant and restored prairies at the Platte River Prairies preserve (PRP). The Nature Conservancy has been doing restoration work along the Platte River since the 1990s. That work was initially done in conjunction Prairie Plains Resource Institute, a restoration-focused NGO based in Aurora, Nebraska. Prairie Plains has immense expertise in constructing restorations from a high-diversity of locally sourced native prairie seeds.

The Nature Conservancy’s goal for restoration work at PRP is to increase habitat connectivity among prairie sites. The philosophy with which TNC has approached prairie restoration in this area is not that restored prairies should be exact replicas of nearby remnant sites. They don’t need to have the same species composition as neighboring remnants, they need to contribute to the area’s prairie habitat connectivity. That, in turn, will benefit the native species persisting in today’s fragmented prairie landscape. In my project, I investigated what floral resources for pollinators (aka flowering plants/forbs) look like across these remnant and restored sites.

Orb weaver spider eating a grasshopper on a foggy morning, September 2022. Photo by Emma Greenlee

Although TNC’s restoration practitioners and land managers do not seek to create prairie restorations that are identical to their neighboring remnants, it is important to understand how both compare in terms of the resources they provide to prairie communities. That way, we can assess our restoration and management approaches and alter our techniques if needed. There are so many aspects of a prairie ecosystem that one could measure to determine this, but because I’m really excited about plant community ecology, I chose to approach my project from that direction. In addition, I wanted to address the role plants play in supporting the prairie ecosystem, and thinking about floral resources for pollinators appealed to me as a way to do this (while still focusing my data collection efforts on the plant community!).

Catsclaw sensitive briar (Mimosa quadrivalvis var. nuttallii), July 2022. Photo by Emma Greenlee

I selected five sites at PRP that contain a remnant and an adjacent restored prairie undergoing similar management. Every two weeks, from early May through mid-October, I collected data within a designated sampling polygon in each of the remnant and restored prairie sites I had selected. Collecting data through time in this way allowed me to see how the prairies changed throughout the season, and this was a highlight of my project.

From seeing the first flowers of the year, like fringed puccoon (Lithospermum incisum), to the new plants I learned about throughout the season like catsclaw sensitive briar (Mimosa nuttallii) and Illinois bundleflower (Desmanthus illinoensis), to the bobolinks I heard singing and the cool invertebrates I saw along the way, I noticed something new every time I went out to collect data. Especially satisfying was the day in June when, as if a switch had flipped, the native tallgrass big bluestem had suddenly overtaken the invasive smooth brome that had been widespread in many sites early in the season. Even though my project primarily involved collecting data on wildflowers, it provided me the opportunity to notice so much more.

Monarda (Monarda fistulosa) was particularly abundant in one of my transects on this sampling day, July 2022. Photo by Emma Greenlee


Results and Discussion

In general, my results suggest that floral resources in remnant and restored sites at the Platte River Prairies were similar this year! The flowering species that were available in a given location, as well as their abundances, varied throughout the season, but I found no evidence that remnants or restorations as a group had more or fewer flowers than the other at any point during the growing season.

In fact, some of my results (see graph below) suggest that remnant and restored sites complement each other in terms of the flowering species they provide. This complementarity indicates that there are different floral resources in adjacent remnant and restored sites at a given time, but that remnants and restorations flip-flop throughout the season in terms of which offer more flowering species for pollinators. Because the paired sites I studied were located geographically near to each other, this suggests that insects are able to move between neighboring remnants and restorations to fulfill their needs throughout the season.

Graphs of the number of flowering species in each of my five sites over time. Red lines correspond to remnants, and blue lines to restorations.

Within the above graph, the red and blue lines zigzag many times, with a remnant (or restoration) containing more flowering species, and then its paired restoration (or remnant) overtaking it to have more flowering species the next month, and then vice versa. This should overall be a positive for pollinators who are able to move back and forth between neighboring sites to fulfill their needs at a given time. (If you’re looking for a clear example of the pattern I’m describing, it’s especially visible in the East Dahms NW site (top middle graph) from August through October.)

Additionally, my results suggest that factors apart from a site’s status as a restoration or a remnant are likely responsible for much of the variation in flowering communities found at PRP. Some sites have sandy soils and hilly topography, while others are largely flat with a shallow water table (underground water is fairly near the surface) and wetland sloughs running through them. Variations like these shape the suite of plant species that is able to establish in an area. In the sites I studied it seems likely that factors like these (soils, topography, hydrology…) are important drivers of plant community composition in the patchwork of prairies along the Platte River, regardless of whether a site is a restoration or a remnant (see graph below).

This type of graph is used by community ecologists to simplify many aspects of the communities they’re comparing down to a two-dimensional plot. In my case, it represents the flowering plant species composition and floral abundance of each of my sites! The axis labels aren’t important, they just represent community similarity: the points that are closer together on the graph are more similar in species composition and flowering abundance, and points that are farther apart are less similar.

On the above graph, circles represent remnants and triangles represent restorations, and each remnant/restoration pair is a different color. Most of the color-coded remnant/restoration duos are found close to each other on the graph, indicating floral community similarities. Interestingly, most of the restorations are also grouped fairly near to each other, as are the remnants! This suggests some community similarities among remnants overall and among restorations overall, but it would take more statistics than I accomplished on this project to find out what those might be!

We can’t draw too many firm conclusions from only one year’s data––factors like that year’s weather, the number of sites I collected data from, and each site’s management all may play a role in the patterns I saw. However, this is still useful information that can be built on in the future, or simply serve as a single-year snapshot of a few aspects of prairie community health on the preserve. In general, what I found this year suggests that restoration work at the Platte River Prairies preserve is largely effective in increasing prairie habitat connectivity by the metric of providing floral resources for pollinators. This is encouraging news, and suggests that the methods used here (and in many other places!) of restoring prairies with a diverse mixture of locally sourced native seeds are sound and contribute to the prairie patchwork landscape of central Nebraska.

I learned so much from this project and from the Fellowship more broadly, and I appreciate everyone who has read my blog posts along the way! It’s been a pretty cool privilege having such an enthusiastic audience to share some of what I’ve learned this year with. Next I’ll be pursuing a Master’s degree at Kansas State University studying plant-pollinator interactions in prairie, and I’m very excited for that! I also look forward to finding new ways to share my prairie enthusiasm with a wide audience.

Caption: A snapshot of the floral diversity at the Platte River Prairies, August 2022. Photo by Emma Greenlee