Last summer, Shelly Cox of Missouri posted this photo to the ID Request page on BugGuide.Net:
No one offered a suggestion as to what it might be. She kept it in a container, and several weeks later there was a hole chewed in one end of it but somehow the insect that had emerged was nowhere to be found. I had a suspicion about what this object was, and Shelly ended up sending it to me so I could take it apart and examine it. My suspicion was confirmed: the packet, which was 22 mm long and 8 mm wide, was made mostly of oval leaf pieces, averaging 1 cm long, with several circular end pieces about 5 mm wide. No silk was used in its construction. In the middle I found the remains of a thin, papery cocoon. This could only be the work of a leafcutter bee (Megachilidae: Megachile).
Female leafcutter bees cut pieces from the edges of living leaves and use them to construct nest cells. Each cell contains a mixture of pollen and nectar and a single egg, which hatches into a larva that feeds on the provisions, eventually metamorphosing into an adult bee that chews its way out of the cell. I had seen leaves bearing the distinctive round cuts of leafcutter bees many times, but I had never actually seen a nest cell in person before, so I was excited to have the opportunity to look at this one.
The thing is, leafcutter bees are supposed to make their cells in tubular cavities, like hollow stems or the tunnels excavated by wood-boring beetles. Over the past winter I’ve spotted several cells in round chinks of walls and in the exit holes of mud wasp nests.
It’s hard for me to picture how a bee could form leaf pieces into a neat, tightly packed cylindrical packet without stuffing them into some kind of cavity. Yet browsing through the archive of Megachilidae images on BugGuide.Net revealed that this find was not entirely unprecedented. Five years earlier in Florida, Jeff Hollenbeck had photographed a group of nest cells in a spider web. He reported that “the bee crawled freely around the spider web to get to her chambers.”
I asked around on the Bee Monitoring listserve for thoughts on Shelly and Jeff’s photos, and some responses asserted that these cells must have fallen (or maybe were pulled out by a bird) from whatever cavity they were constructed in. However, if the bee Jeff saw in the web was in fact a leafcutter bee (unfortunately he was unable to get a shot of it before it flew away), it must have still been working on the cells because it would have no reason to visit them after they were provisioned with pollen and nectar and sealed up with an egg laid inside. Also, Shelly’s cell was attached to the milkweed leaf, not simply lying there. A few people mentioned having seen leafcutter bees nesting in larger cavities such as in a cinder block or an electrical switch box–evidence that they don’t require a tight cylindrical space–and Jack Neff noted that many neotropical Megachile species build nests inside folded leaves. Laurence Packer directed me to a paper of his* (illustrated with photos) in which he described having found M. brevis nests in tufts of grass in Florida, and his paper in turn cited a 1953 study from Kansas in which one of 109 nests of this species had been found in a similar situation. So evidently this is not a common phenomenon, but not completely unheard of. As far as I can tell, though, no one has actually watched one being built in the open, so exactly how it is done remains a mystery.
* Packer, Laurence. 1987. The Triangulin Larva of Nemognatha (Pauronemognatha) punctulata LeConte (Coleoptera: Meloidae) with a Description of the Nest of Its Host — Megachile brevis pseudobrevis Say (Hymenoptera: Megachilidae). Journal of the Kansas Entomological Society 60(2):280-287.