The earthquake and subsequent fire that tore through San Francisco in 1906 killed more than 3,000 people, left another 225,000 homeless, and destroyed some 28,000 buildings, leaving around $400 million worth of damage in its wake.

Among the destruction was the building that housed the California Academy of Science—and the academy’s collection of animal holotypes, single specimens used to define a newly discovered species.

“What they did after the earthquake is they made a vault for their holotypes,” said Neal Evenhuis, an entomologist at Hawaii’s Bishop Museum. “If there was an earthquake, it would be sent in an elevator to an earthquake-proof room so nothing could ever happen to those things.”

The point of Evenhuis’ story is this: Holotypes are worth protecting. “Having those specimens is pretty important to knowing what we have on planet Earth. So you don’t want to have anything happen to them,” he said. “Especially if a lot of things go extinct in our time, you have a history of what was there.”

But he has another point, too: The record can exist even when the specimens don’t. In a paper published yesterday in the journal ZooKeys, Evenhuis and Steve Marshall, a professor of biology at the University of Guelph, identified a new species of fly based on photographs alone, the first insect to be named without any supporting physical evidence.

The newly discovered species,
Marleyimyia xylocopae
(Steve Marshall / Zookeys)

Last year, Evenhuis was putting together a chapter about bee flies for a reference book on the flies of Africa, and asked Marshall, a prolific photographer, for images he could use in the text. One in particular stuck out. It had a distinct black-and-yellow pattern and an oddly shaped body. “I’ve worked on bee flies since the 1970s and never seen anything that looked like this,” he said. “I want to see this thing. It’s obviously new. I couldn’t even figure out what genus it belonged to.”

But Marshall hadn’t captured the insect, and so Evenhuis let it go—until, further along in his chapter, he realized the genus: Marleyimyia Hesse, an extremely rare type of fly found in southern Africa. Until that point, there were only three known specimens in the world, of two separate species. The fly in the photo belonged to neither of them.

But with the genus, Evenhuis had enough to satisfy the naming conventions for species laid out by the International Commission on Zoological Nomenclature, the body that governs animal taxonomy. “And so I suggested to Steve, why don’t we just describe this thing and use this photograph as a representation of the holotype?”

The decision was a contentious one. Last May, a group of entomologists wrote a letter to Science arguing that “describing a new species without depositing a holotype when a specimen can be preserved borders on taxonomic malpractice.” There’s always the risk of mistaking a picture for a new species, when in fact it’s an oddly colored, oddly shaped example of one that’s already been defined. And bodies show things that images don’t; a photo can’t reveal new secrets when placed under a microscope.

“We potentially sacrifice the most important things to know about a species when we forego more than superficial evidence of anatomical details,” they wrote. “With millions of species threatened by extinction, it would be tragic were we left with no more than a few photographs and sequences as evidence they were once here.”

Evenhuis agrees, emphasizing that a physical holotype is always better. But sometimes specimens evade collection in the wild, or conservation restrictions forbid scientists from taking them back to the lab. In those cases, an image is better than nothing at all.

If “you do see one and you didn’t get to collect it, and you know it’s absolutely different than anything else in the world, it’s kind of silly to just leave it sitting with a photograph and no name on it,” he said. “We need to know what’s out there, and by publishing these things, it alerts others to go out and look for them.”