Per posts on Reddit and Facebook, jars baited with malbec can trap Houdini flies (Cacoxenus indagator), an invasive kleptoparasite of solitary bees in the genus Osmia. The species was first reported in New York in 2011 and has since appeared in Maryland, Massachusetts, New Jersey, Oregon, Pennsylvania, Virginia, Washington, and British Columbia. For anyone with an insect hotel in an affected area, brood losses can be considerable so the news of a control method is welcome news. Currently hotel owners deal with the parasites by squishing them one by one, vacuuming them up, or trapping them on sticky cards situated in the eaves. All of these techniques seem futile given that a single Houdini fly can leave hundreds of eggs.
The trap
The general idea is to put the wine into a container that has small, fly-sized holes, then situate in a nook in your insect hotel. I opted to use 2-oz containers that I had sitting around but several posters on reddit use small mason jars. I then drilled four, 1/8″ holes in the lids, attached segments of cork with a 1 5/8″ screw to either the side of the container or the lid, and then screwed it to the hotel.
Evidence that it works
I set mine up too late in the season to estimate how many Houdini flies can be trapped per day, but I can confirm that some did go inside the traps and drown. Here are two:
Small waterlogged flies are difficult to identify even when you have decent photographs, so I eventually fitted the containers with sticky cards that would immobilize insects before they drowned. Here’s a trap attached via the lid and a Houdini fly that got stuck:
I should emphasize that my traps also caught dozens of small fruit flies that were not Houdini flies. So if you are trying this trap method be cautious in deciding whether it’s actually working or whether it’s just attracting small flies that happen to like red wine. I encourage everyone to use sticky cards and to submit photographs for ID confirmation to either BugGuide (North America only) or iNaturalist. If you opt for the latter you’ll likely get input from me within 24 hours (I’ve ID’d 200 Houdini flies already).
Is it safe for bees?
The holes are too small for many mason bee species but there are plenty of solitary bees that could fit into 1/8″ holes if they wanted to try. So far, however, none has become stuck in any of my nine traps. Some of the bees even use them as landing platforms because the sun hits them in the afternoon.
Design tips
Some thoughts that I’ll update as I continue with testing.
Opt for clear-sided containers so you have a better view of what is inside.
I’m now using a slightly smaller drill bit because 1/8″ allows rather big flies to enter.
Reddit user atrailofdisasters uses a red-pepper shaker, a solution that might work well for those who don’t have a drill and collection of small bits.
What wines work best?
Per reddit, the person who came up with the trap idea, Nicole Kenney, of Portland, Oregon, tested different wines for several years and decided that red wine, especially Argentinian malbec, is best. I have no idea how she did the tests but I’ll link to more information when I can find it. I’ve only tried malbec in mine but will do some comparisons next year during peak Houdini-fly season. Malbec is an old grape grown around the world but apparently thrives in Argentina and much of the acreage is in the mountains. It would be very interesting to know exactly which component is attracting the Houdini flies, of course.
Why does it work?
Houdini flies are members of the vinegar fly family (Drosophilidae) and are known to like fermenting fruit juice (source) so it’s not a surprise they are attracted to wine. And all animals are lazy so if there is food near where females are busy ovipositing, they will happily take a break to a have a drink.
More information
If you’d like to see photographs of Houdini flies I have about 20 on iNaturalist.
Now that my eastern redbuds (Cercis canadensis) are pumping out thousands of pods each fall, I thought I’d start photographing the wasps that show up to parasitize the redbud bruchids (Gibbobruchus mimus) developing inside the seeds.
Eupelmus pulchriceps (Eupelmidae)
I’m starting with this species because it’s the only one I’ve photographed that is currently identified beyond genus on iNaturalist. Here’s the female:
And here, possibly, is the male:
I think this a known parasitoid of the redbud bruchid but it took some digging to figure it out. Kingsolver 2004 lists E. cyaniceps as a parasite but Gibson 2001 asserts, “… all literature recording E. cyaniceps from various Bruchidae appear to be misidentifications of E. cushmani.” And per iNaturalists taxonomy page, Eupelmus cushmani is now known as Eupelmuspulchriceps. That said, BugGuide seems to have a listing for Eupelmus cushmanii but not Eupelmuspulchriceps. Additionally, the 2021 “Checklist of Chalcidoidea and Mymarommatoidea (Hymenoptera) of Canada, Alaska and Greenland” (Huber et al. 2021) contains both species, as does the “Bees and wasps of Central America Extended.” I’m a bit confused and thus I likely made a mistake somewhere.
Kingsolver 2004 mentions that both Heterospilus bruchi and H. prosopidis can be found parasitizing redbud bruchids, so my specimens could one of those (or both), but perhaps even something else in the genus. My ID is because the wasp strongly resembles an observation on BugGuide that is identified as Heterospilus sp.
And here’s the male, which I admit assumes they are the same species. I don’t have any photographs of this species arriving at pods still attached to the tree.
Eurotoma sp.
Klingeman and Carrington 2005 list two unidentified Eurytomidae associated with redbud trees that have bruchid infestations, so this was also not an unexpected find.
A male (below) was one of several that emerged from redbud seed pods. As with the other finds, I’m tentatively assuming that the male/female pairs are the same species.
Dinarmus sp. (Pteromalidae)
Per a suggestion on iNaturalist, this might be in the genus Dinarmus. In (weak) support of that, it does appear similar to observations of that genus on BugGuide and iNaturalist. If the genus is correct, it might possibly be Dinarmus basilis, a species that Klingeman and Carrington 2005 found emerging from eastern redbud seeds in Tennessee. There’s an illustration of a female on page 646 of Sureshan and Narendran 2001, however, that doesn’t look like mine in profile, so I’m hesitant. Similarly, figures 2 and 3 of Pimentel et al. 2024 look much bluer than mine. I also think my specimens have too many antennal segments (9?). Dinarmus basilis is apparently introduced in North America and is popular in IPM. The lectotype (a male from Egypt) is in Florence under the basionym of Entedon basalis (Bouček 1974).
More wasps to come?
I’d be very surprised if my current gallery of wasps is complete, so I’m monitoring several thousand seed pods (below photograph) to see who else might emerge in the months to come. There are two wasps that are mentioned in the literature and that I’m hoping to find. The first is Horismenus missouriensis (Eulophidae; bright metallic green with a “longitudinal median groove on the scutellum“; see fig. 1), mentioned by Burke 1971. And Stenocorse bruchivora (Doryctinae; fig 9b, fig 7) is mentioned by Cushman 2011. Neither BugGuide nor iNaturalist has photographs of these species.
I’m especially interested in finding hyperparasites of some of the wasps mentioned above. And on my to-do list in 2026 is to collect several hundred redbud pods in the spring to see whether any of the redbud bruchid eggs harbor parasites such as fairy wasps (Mymaridae).
Redbud seed bruchid
In case you want to see what the beetle looks like, here’s an adult:
There’s an eastern black walnut (Juglans nigra) on my neighbor’s property that rains down fruit every fall, and I finally got curious about what species might be inside. So far I’ve found four flies, five beetles, two mites, and a parasitic wasp. I’m a little disappointed with the number of wasps but am hopeful that over time I can find more. There are almost always more wasps.
Walnut husk maggots
Almost all of them were filled with fly larvae that are most likely Rhagoletis suavis (Tephritidae). If you collect walnuts for their nutmeat, these are unwanted residents because their activity can make the husks difficult to remove, stain the shell, and even cause the kernel to discolor, become bitter, and shrivel. Commercial walnut groves Juglans regia will even treat spray insecticide to minimize their presence. Below is a typical scene showing a healthy population of the larvae:
And here’s a close-up. It’s interesting that the larvae seem to accumulate dark pigment in each segment. I’m assuming it’s juglone, also known as Natural Brown 7. The presence is probably unavoidable given what it is eating but I wonder whether it confers any protection against predators. It’s certainly allelopathic, antifungal, and can kill fish.
Here’s an adult that I found a few years ago.
There are several species of tephritids that utilize walnuts. Rhagoletis suavis is restricted to the eastern half of North America (iNaturalist distribution map), with the western-most observation being in Colorado. The western species, Rhagoletis completa (BugGuide pics; iNaturalist map), seems to be expanding its range, arriving in Jonesboro, Arkansas in 2007. Another western species, Rhagoletis juglandis (BugGuide pics; iNaturalist map), occurs in Arizona, Utah, New Mexico, Mexico, and apparently British Columbia. If you live in an area with more than one of these species, you’d need to rear larvae to adulthood to get an identification (see fig. 8 here for wing pattern differences). Alternatively, you can try to use the key to larvae in Steyskal 1973.
Fungus gnat
This fly emerged from a pile of rotting walnuts that I’d covered with fine netting to catch whatever hatched when I wasn’t looking. I think it is Leia bivittata (Mycetophilidae). Per Weiss 1919 the larvae move around in a bag of mucous. It’s the only one that appeared. There is nothing in the literature that suggests they eat the fungi in rotting walnuts, so I don’t know how common they are in the fruit.
Vinegar flies
I found two species of vinegar flies (Drosophilidae) hanging out on rotting husks, Drosophila repleta and Drosophila tripunctata, the latter in large numbers. Per Buchner and Symmes 2014, Drosophila are attracted to the rotting flesh caused by Rhagoletis infestation. What’s notable to me is that larvae can tolerate the juglone and related compounds.
Beetles
This scooped scarab (Onthophagus hecate) was wedged inside a crevice in a walnut husk. They eat fungi but will also show up to eat dead animals.
I also found four different kinds of staphylinids. I’ve only been able to identify one, Belonuchus rufipennis, but will update this post if and when an expert on iNaturalist sees them (so probably at least a decade). I would love to know what they are eating. There are certainly plenty of fly larvae present but I don’t have any confirmation that those are their targets.
The rotting walnuts were crawling with mites, often numbering in the dozens. These are extremely small and fast moving so I’m not sure how many different kinds are present. I think there are at least two species, both in the order Mesostigmata. The first image might be in the suborder Gamasina, and possibly in the genus Lasioseius. I have no idea what they are eating but some mesostigs are omnivorous.
Parasitic wasps
Although I didn’t find them inside the walnuts, this wasp in the genus Coptera (Diapriidae) ended up in a yellow pan trap set up near a tray of Rhagoletis suava puparia that I situated atop a bin of several dozen walnuts. Here’s a really low-res video of it walking around on the tray after I rescued it from the pan trap. There was a second individual walking on the tray at the same time, so I don’t think the pan-trap individual was just a rando Coptera attracted to yellow. These wasps apparently dig down into the soil to find pupae. Some articles say they’ll drag the pupae to the surface prior to ovipositing into them.
Members of this genus are rarely observed and are difficult to tell apart, but based on proximity to walnuts it is likely Copterapomonellae, the only species that arrived at pan traps set up under walnut trees in Iowa (two locations) and Indiana in Forbes et al. 2012.
Because I have trays of bait pupae, I’m also monitoring for but haven’t yet seen:
Phygadeuon sp. (Ichneumonidae), a species that attacks pupae of (the closely-related) Rhagoletis juglandis (Andrew Forbes, pers. comm.). I’m not sure whether Phygadeuon attacks Rhagoletis suavis, but there are observations in the eastern USA.
Aganaspis alujai (Figitidae), a parthenogenic wasp that attacks Rhagoletis completa in Mexico (Ovruski et al. 2007, Takata et al. 2025). I have no idea whether there’s a member of the genus in Pennsylvania, so this is another unlikely arrival.
Tetrastichus giffardianus (Eulophidae) “may parasitize Rhagoletis suaviscompleta CRESSON” (Narayanan and Chawla 1962, citing Marlatt 1933, which I cannot locate). Marlatt was writing in a USDA publication so this relates to the United States. But host species is now Rhagoletis completa, so this is probably a western wasp (confirmed: just California).
Absent from the list are hyperparasitoids, which is kind of annoying because I really like them. I will certainly be on the lookout for them, of course. I’ve also read that a fairy wasp attacks the Rhagoletis eggs, and that would also be fun to find.
I’ll post an update in August 2026 with details on what emerges from the several hundred Rhagoletis suava pupae I have. Hopefully there will be at least some Coptera in the mix, and with some luck I’ll get photographs of both sexes. I’m also very curious what types of wasps might seek out Rhagoletis completa that are on eastern black walnuts planted in the western USA, but that hinges on my sister reading this post and setting up some trays in a similar way (simple instructions are in next section). The first record of Rhagoletis completa in Oregon was in 1963 (Berlocher 1984).
Below is a photograph of the trays of pupae. The chair is so I can drink my coffee while scanning for wasps.
Collecting fly puparia
If anyone is curious how I’m getting puparia, I placed infested walnuts in trays filled with sand and vermiculate, and the larvae tunnel down into the medium to pupate. The plexiglass is to keep the rain from flooding the containers, and the whole operation is conducted inside a screened enclosure that keeps the squirrels from pilfering the walnuts. This is making the squirrels angry.
The larvae eventually burrow into the medium to form puparia, but sometimes they simply pupated underneath a walnut (e.g., as in photo below). At some point I’ll sift the media to collect puparia so that I can more closely monitor them. I’m curious whether they can be candled under a microscope to spy on the residents.
Here’s a sampling of the puparia along with some third-instar larvae. I suspect I have 500 or so puparia in the trays. Some pupae apparently stay dormant for two years (Stirret 1936).
Next year I’ll likely repeat all of the above but start monitoring for Coptera much earlier.
