This single well-known and widespread butterfly is actually three species in disguise

Key points

• The genus Colobura is widespread in Central and South America and contains some of the most abundant and conspicuous butterflies in the region. For 200 years, it was thought to contain a single species. A study published in 2001 added a second, and a new study describes a third.
• All three species are cryptic, meaning they look nearly identical during parts of their lifecycle, which is why their diversity escaped notice. Biologists only discovered the third species because its caterpillars lack a small but distinctive marking.
• The distribution of Colobura butterflies overlaps in some places, and it’s unclear how individuals looking for mates are able to distinguish members of their own species.

GAINESVILLE, Fla. — The tropical rainforests of Central and South America are one of the most biodiverse places on Earth. Costa Rica alone is home to half a million species, five times more than exist in the entire country of Canada, despite the former having 200 times less land area than the latter.

Yet, just as balsam firs are ubiquitous in parts of Canada and have become synonymous with boreal forests, the neotropics have their own exemplars. The German botanist Andreas Schimper wrote in 1888, “Every naturalist who has been in tropical America will remember Cecropia, belonging to the most conspicuous trees of those countries.”

According to Anisha Sapkota, a doctoral student at the Florida Museum of Natural History, there are emblematic animals as well.

“If you go to the neotropics, Colobura is one of the first butterflies you see,” she said.

The fact that these two groups are considered tropical mainstays is not a coincidence. Cecropia trees are host plants for Colobura caterpillars, and both have a broad distribution that encompasses parts of the Andes Mountains to the west, the Brazilian Amazon to the east, the Atlantic Rainforest in Brazil and Paraguay in the South, and a large swath of Central America and the Caribbean Islands to the North. 

It is interesting but not unusual for a group of organisms to be so widespread. There are 61 species of Cecropia, and most of them have a limited range. Some, for example, are restricted to the Andes while others only grow in the Caribbean. What is strange, however, is that for more than two centuries after it was officially described by Carl Linneaus in 1758, the genus Colobura was thought to contain a single, widespread, highly distinctive species, named C. dirce.

A second species was officially added to the list in 2001. Following an unexpected discovery made in Venezuela, Sapkota recently published the description of a third species in the journal Zootaxa. But rather than clarifying the evolutionary history of Colobura, the existence of three species where there was long thought to be one has resulted in a lot of head scratching.

For starters, Sapkota said, there isn’t much that differentiates them. “It’s crazy to think about. These butterflies look almost exactly the same and are found in the exact same place at the same time.”

Colobura adults appear identical, with the exception of a single gray band on the underside of their forewings that broadens slightly in some individuals, remains uniform in others and either does or does not extend all the way to a yellow perpendicular band beneath it.

The caterpillars are easier to distinguish, though the differences between them remain slight. Colobura larvae have a uniformly black velvet base from which protrude a series of bright yellow stalks tipped with sharp, angled barbs — a universal signal in the plant and animal world that roughly translates to “leave me alone, or there will be unpleasant consequences.” Beneath these barriers to unwanted interactions are more subtle cream-colored markings that come in the shape of tear drops and rings.

In 1852, the Dutch entomologist Jan Sepp decided the different cream-colored markings in caterpillars was enough to recognize two separate species. He was promptly ignored by his peers, and for the next 150 years, naturalists insisted the difference was either due to sexual dimorphism or individual variation within a single species. In 2001, Keith Willmott — now a curator at the Florida Museum’s McGuire Center for Lepidoptera and Biodiversity — revived Sepp’s idea along with several colleagues and formally described a second species, Colobura annulata.

“We assumed that was the problem solved,” Willmott said. Having rectified a case of longstanding taxonomic ambiguity, he called it a day and moved on to other projects. But a growing suspicion that he hadn’t quite finished the job kept surfacing from time to time. “I had a few indications that there was a third species, but I thought it was too implausible.”

Even a second species, he said, felt like it was pushing it at the time.

“It’s a very distinctive butterfly. It’s hard to imagine there being two species in the same area that look alike because they’re so different from anything else.”

A quarter century later, one of Willmott’s colleagues, Andrés Orellana, happened across a group of voracious Colobura caterpillars eagerly defoliating a Cecropia sapling in Venezuela. The black thoracic segments and yellow spines were present, but none of the caterpillars had cream-colored markings of any kind. Orellana shared the photos with Willmott, who decided the new evidence was enough to reopen the case of the cryptic Colobura butterflies.

Anisha Sapkota had just joined the Shirey lab at the Florida Museum and expressed interest in leading the investigation. She and her colleagues developed a rigorous and ramified line of inquiry that would give them a definitive answer. She analyzed highly variable stretches of DNA called barcodes that had been sequenced and curated as part of a long-term survey of Lepidoptera in Costa Rica led by study co-authors Daniel Janzen and Winnie Hallwachs.

Building on these results, Anisha also sent older museum specimens collected throughout the neotropics to Nick Grishin at the University of Texas Southwestern Medical Center, who sequenced their mitochondrial and nuclear genomes.  

She then examined preserved museum specimens to see if there were any subtle differences between individuals that others had missed. Aside from the slightly variable gray band on the bottom of their wings, nothing much as forthcoming. But after staring at them for hours, she noticed a barely perceptible violet glimmer coming off the underside of their wings that seemed to have a slightly different sheen among certain specimens. This immediately put her in mind of ultraviolet radiation, a portion of the light spectrum that many butterflies can see but humans can’t.

“I checked it with a spectrometer and found that even though the butterflies didn’t look violet to me, it reflected UV light,” she said.

The two previously described species and the one she and her colleagues discovered, which they named Colobura cryptica, all reflected UV light at different wavelengths.

There were no other anatomical differences she could find. That included their male genitalia, which in moths and butterflies is often the single most important diagnostic feature used to tell species apart.

Here is a simplified diagnostic key:

• Colobura dirce
  • Adult
    • Gray wing band broadens slightly and does not touch the yellow transverse band
    • Average UV reflectance of 376 nm¹
  • Larvae
    • Cream-colored teardrop on abdominal segments
• Colobura annulata
  • Adult
    • Gray wing band is uniform in width and touches the yellow transverse band
    • Average UV reflectance of 370 nm
  • Larvae
    • Cream-colored rings on abdominal segments
• Colobura cryptica
  • Adult
    • Gray wing band is uniform in width and does not touch the yellow transverse band
    • Average UV reflectance of 344 nm
  • Larvae
    • No cream-colored markings

These slight, hairsplitting anatomical differences might not have been enough to warrant the description of a new species by themselves, but results from the DNA analyses were unequivocal: Colobura contained three separate lineages, and the divisions aligned perfectly with the physical traits Sapkota had observed.

All three species have overlapping distributions throughout parts of Central and South America. Colobura dirce has the largest range, from Mexico to southeastern Brazil, both sides of the Andes and the Caribbean archipelago. Colobura annulata is similarly widespread but is absent in the Greater Antilles. Colobura cryptica has the smallest range, which extends southern Mexico down to the foothills of the Andes.

The Venn diagram of Colobura distributions is responsible for a significant portion of the perplexed head scratching that has attended the description of two new species in this group.

For speciation to occur, there must be some sort of isolation that keeps incipient species from interbreeding. Biologists think that isolation most often comes in the form of physical distance. If members of a species became separated by a mountain range and could no longer interbreed, the two groups would separately accumulate their own unique traits. Eventually, they would diverge to the extent that they would be incapable of interbreeding, assuming they reestablished contact. At that point, biologists would consider them to be separate species.

But physical distance isn’t the only way to create isolation. In the case of Colobura, the authors speculate that it may arise from the way in which butterflies partition their shared space.

“There could be vertical stratification between these butterflies,” Sapkota said. “Colobura dirce is typically found in the undergrowth and C. cryptica and C. annulata are usually found in the canopy, but we do not know for sure because there have been mixed signals coming from different studies.”

The mixed signals might have something to do with the fact that even seasoned butterfly taxonomists have a hard time identifying Colobura species.

It remains unclear, for that matter, how these butterflies are able to recognize each other. They lack the scent glands that some butterflies and many moths use to emit distinctive pheromones. So their choice of mate likely relies on visual cues to some extent. Given that there aren’t many of these to go on, Sapkota wondered whether they might be picking partners based on UV coloration. There’s precedent for this in other groups. Many moths and butterflies — including garden whites, sulfur butterflies and wood tiger moths — have UV-reflecting scales that are thought to function in mate selection. Sapkota is currently working on a study to determine whether this is the case in Colobura, but the results so far have been inconclusive.

However they do it, their ability to recognize members of their own species appears to be foolproof. The nuclear DNA sequenced for the present study showed no signs of hybridization. All three species are on their own, entirely separate trajectories, and they haven’t looked back.

All of which amounts to the fact that even the most basic biological aspects of a familiar and abundant butterfly remain unknown. According to Sapkota, there’s only one way to solve that problem.

“Someone has to go the field to find it out themselves.”

If you’re still eager to learn more about the natural history of Colobura, head over to the Biodiversity Heritage Library for a whimsical description, published in 1952, of the poop perches the caterpillars make during their early instars.

Additional study co-authors are: Andrés Orellana of the Universidad Nacional Experimental del Táchira; Nick Grishin and Leina Song of the University of Texas Southwestern Medical Center; Isidro Chacón of the Museo Nacional Santo Domingo de Heredia; Daniel Janzen and Winnie Hallwachs of the University of Pennsylvania; and Sajan KC of the Florida Museum of Natural History.

Funding for the study was provided in part by the National Institutes of Health, the Welch Foundation, Canada’s New Frontiers in Research Fund (NFRFt-2020- 00073), the Canada Foundation for Innovation’s Major Science Infrastructure program (MSI 42450), BioAlfa and Project UNET (Universidad Nacional Experimental del Táchira)..

Footnotes:

1. To make matters more complicated, Colobura dirce is further divided into two subspecies that have differing UV reflectance values. Colobura dirce dirce reflects light at 362.2 nm, and Colobura dirce wolcotti has a reflectance value of 390.6 nm. The value listed for the species is the average of the two.  

---

---