About the SA Blog Network

Guest Blog

Guest Blog

Commentary invited by editors of Scientific American
Guest Blog HomeAboutContact

When Did Insects Learn to Smell?

The views expressed are those of the author and are not necessarily those of Scientific American.

Email   PrintPrint

Summer beckons in the northern latitudes. As temperatures warm, beachgoers will flood sandy coasts, carrying bags of potato chips and sandwiches. Picnics will sprout in parks supplied with pink watermelon wedges and soda pop. Feasts ripe for people, and undoubtedly bugs too. In mere seconds flies and bees can swarm, but how do they sniff out food so fast? Knowing how to fly could be the reason, according to new research from the Max Planck Institute for Chemical Ecology.

Their findings challenge an old theory on when insects first evolved olfaction. Bugs share an ancestor with crustaceans, which lack the physical machinery typically tied to a sense of smell. Scientists presumed that when this oceanic predecessor moved ashore approximately 480 million years ago it developed the olfactory hardware for sifting aromas that permeated through its new air-based environment.

“Not true. They kept the old way of smelling for a long time after,” says evolutionary neuroethologist Bill Hansson, who led the recent study published in eLife. To draw this conclusion, Hansson’s team examined two of the most primal insects left on the planet: bristletails (Lepismachilis y-signata) and firebrats (Thermobia domestica).

Living primal: A bristletail (top) and firebrat (bottom). Source: Missbach et al. Arthropod Structure & Development (2011) and Tree of Life Web Project/Wikimedia.

Bristletails look like a cross between bite-sized shrimp and spiders. Their origins date back 390 million years to the early days of “insectdom”. Firebrats came 60 million years later; they resemble silverfish, those flat, metallic-shaded bugs that enjoy zipping around bathtubs and basements.

Previous research on insect olfaction has focused primarily on insects with wings, such as fruit flies (Drosophila melanogaster). Bristletails and firebrats are wingless, but close relatives of the fruit fly, so Hansson and his team decided to draw an evolutionary picture of insect olfaction by comparing the three species.

Smelling in fruit flies starts with their antennae, which house smell sensory neurons coated with olfactory receptors (ORs) – protein “satellite receivers” that detect scent molecules. ORs are the present-day gear for smell. They are seen in most terrestrial mammals and insects with a sense of smell. ORs don’t work alone. Another receptor, called Orco, partners with ORs and tunes their response to odors. Orco, however, has never been shown to detect odors on its own.

The researchers scanned the bristletail genome and found no traces of ORs or Orco. Bristletail antennae, however, did respond to odors, albeit mildly relative to fruit flies. Bristletail antennae did accommodate a high density of ionotropic receptors. Thought of as a more ancient form of chemosensation, these proteins were recently implicated in the primitive smell system seen in dragonflies. Using electron microscopy, Hansson’s team looked at the architecture of sensilla, tiny cuticles on antennae that house sensory neurons. They found that the olfactory sensilla of bristletails, in general, lack the structure of modern insects.

Moving down the evolutionary road, firebrats were more bizarre. They had a stronger reaction to the same bouquet of aromas, but still lacked ORs. Firebrats did possess three copies of an Orco-related gene that were only produced by olfactory sensilla.

The results raise a bevy of questions such as: do firebrats represent an intermediate olfaction system where Orco plays a previously undefined role? The findings, however, argue that ORs and modern smell took time to develop.

“It can’t have been the actual movement onto land itself that was the key change in their lifestyle,” says Richard Benton, a sensory evolution scientist at the University of Lusanne, Switzerland, who was not involved with the study.

Bristletail sensilla (left) lack the grooves that often house olfactory neurons, which contrasts with those in firebrats (middle) and leaf insects (right). Missbach et al. eLife (2014).

Hansson theorizes that OR development was urged on by plants growing taller. As food resources moved upwards, insects developed wings and needed a way to find their meals. Supporting his hypothesis, his team examined and detected ORs and Orco in a fourth species, the leaf insect Phyllium siccifolium.

“Walking leafs represent a primitive form of modern, flying insects,” Hansson says. “We show that many tens of millions of years probably passed before the OR-Orco way of smelling developed. Indeed, we do not find it in any non-neopteran [non-flying] insect.”

Benton echoes this sentiment, though with a grain of temperance. “Additional species are needed to really reinforce this point, but it’s a great start,” he says. “It should inspire more research into non-classic insect models, which will help fill the evolutionary picture of this sensory system.

Images courtesy of Bill Hansson.

Nsikan Akpan About the Author: Nsikan Akpan is a nerd, a nomad, and a science reporter. He blogs at and writes for Science Magazine, Scientific American, The Scientist, and elsewhere. See more of his writing at Follow on Twitter @MoNscience.

The views expressed are those of the author and are not necessarily those of Scientific American.

Comments 1 Comment

Add Comment
  1. 1. Asa Kreevich 12:41 pm 05/12/2014

    Interesting article, but a few disturbing misinterpretations. First, describing fruit flies, one of the most highly evolved and specialized of all insects, as “close relatives” of bristletails and firebrats is like writing about humans and their “close relative” the coelocanth. Nor can the leaf insect be described as a “primitive” flying insect. It is, like fruit flies, highly evolved and specialized. I don’t know if these ideas were introduced by the blogger or were things the scientists said, but if the latter then it is another example of the lamentable lack of knowledge about phylogenetics and biodiversity among functional biologists. Second, there is confusion between “smell” and “taste,” two aspects of chemosense. Smell is generally taken to be the detection of airborne molecules, so of course the aquatic crustaceans would not be expected to have sense organs adapted to do this.

    Link to this

Add a Comment
You must sign in or register as a member to submit a comment.

More from Scientific American

Email this Article