How can a moth trace the source of pheromones up to 11 km away?

March 1, 2006 9:20 AM

I read that the "silkworm moth can detect pheromones from other silkworms up to 11km away. The moths immediately fly towards the source." How is this possible?

source:
http://www.youramazingbrain.org.uk/supersenses/pheromones.htm

Pheromones are chemicals, and even assuming that a moth can detect one molecule of it in a small volume of space, how could it then extrapolate the source? It seems it would need to be able to detect a gradient of increasing concentration to know the source, which would imply there being several molecules at minimum. 11 km is given as the max, but let's say 6 km is the average: that would apply a moth could release enough pheromones to fill most of the air within 6 kms to the level that another moth could trace the source. I know a small amount of matter can have an insane amount of molecules, but it still seems amazing that a moth could produces enough pheromones that another moth then had sophistacted enough sensors to trace it 11 km.
Perhaps an explanation of a dog's sense of smell would explain this, as I'm wondering in general how amazing smells like this work on a physical/chemical level, not just for this moth.

posted by Furious Fitness to science & nature (13 comments total)

What if they were hard wired to fly into the wind whenever they smelled a trace? No gradient necessary.
posted by weapons-grade pandemonium at 9:32 AM on March 1, 2006

Maybe the pheromone is a signal to fly against the wind and that would bring the moth in the direction of the pheromone producer. Along the way it would receive more stimuli.
posted by dances_with_sneetches at 9:33 AM on March 1, 2006

They might also have binocular olfaction, as it were.
posted by weapons-grade pandemonium at 9:36 AM on March 1, 2006

Though it doesn't answer your specific question, this book addresses the larger question of how animal sensory organs function.

I just read it last month, and can say that it's not especially well-written, but the subject matter keeps the book pretty interesting.
posted by Dr. Wu at 9:36 AM on March 1, 2006

Some animals do seem to have stereo olfaction. I just saw this the other day:

It has been hypothesized that rats and other mammals can use stereo cues to localize odor sources, but there is limited behavioral evidence to support this hypothesis. We found that rats trained on an odor-localization task can localize odors accurately in one or two sniffs. Bilateral sampling was essential for accurate odor localization, with internasal intensity and timing differences as directional cues. If the stimulus arrived at the correct point of the respiration cycle, internasal timing differences as short as 50 milliseconds sufficed. Neuronal recordings show that bulbar neurons responded differentially to stimuli from the left and stimuli from the right.

They use time delays within their noses!

Also I feel like I need to include this graphic:


posted by jjwiseman at 9:56 AM on March 1, 2006 [1 favorite has favorites]

Interesting about the rat noses.

I'd guess that at an 11km range, the mechanism would be to fly into the wind upon detecting pheremone because I doubt that some sort of stereo sense of smell would be at all useful at that range.
posted by Good Brain at 10:34 AM on March 1, 2006

jjwiseman, that is quite possibly the best graphic ever.
posted by contessa at 10:57 AM on March 1, 2006

For single cells, there are two ways of moving towards a chemical gradient - sampling in space and sampling in time. Eukaryotes (like us, and moths) tend to sample in space, while procaryotes (like bacteria) are too small for spatial gradients to work, and instead sample in time. The key thing is that the signal response is auto-degrading, and basically exhibits habituation. This is characteristic of many signal pathways.

I suspect that on a cellular level the moth is sampling in space, but that on an organism level it's sampling in time. It's flight therefore resembles an oscillation between directed and random motion. When it is subject to a strong signal it flies in a straight direction. When it is within a "no signal" state it flies in a random direction. Within environments of increasing signal, it will tend to fly in a straight direction more than a random direction. It thus exhibits a directed random walk search through the local environment.
posted by meehawl at 11:21 AM on March 1, 2006

Assuming you're not too interested in academic-level discussions of the biology of pheromone molecule detection in the "target" -- although if you are, there a plenty of online references -- and are more interested in how target finds "source", a key phrase is "odor plumes". Tracking odor plumes to their source is an active field of scientific inquiry for standard biological research as well as newer fields like robotics and environmental science.

Anyway, moths (and crabs) have been heavily studied for their ability to track an odor plume to its source. A colorful 1-page PDF with the modest title of A combined approach of simulation and robotic studies to the problem of understanding animal orientation to odor plumes is interesting for its easily-digested ten-cent tour of moth orientation and how it relates to plume-tracking using robotics. Complete with a handy table of success rates for ground-based KnepTurner vs. aerial Counterturner vs. random walk models. Note that a random walk approach ranks very poor in comparison.

If you feel ready for heavier-duty descriptions of how pheromones are tracked to their source, the far more technical 28-page PDF online paper "Tracking of Fluid-Advected Odor Plumes: Strategies Inspired by Insect Orientation to Pheromone" goes into much greater detail of how moths follow plumes, along with general strategies, including the effects of wind direction. Frankly, some of the math is over my head, but the content is not completely inaccessible to the average reader.

Unfortunately nothing near as cool as the rat-poke graphic.
posted by mdevore at 2:27 PM on March 1, 2006

If I'm not mistaken, smell is very poorly understood as far as the senses go. I wouldn't be surprised if the current literature doesn't have a concrete answer for this.
posted by The Wig at 3:46 PM on March 1, 2006

Moths don't smell pheromone plumes at all; they perceive the plumes' characteristic infrared radiation pattern using their spooky moth radio sense.

See all those tiny hairs on the moth antenna? According to Philip Callahan, they operate as actual dielectric radio antennae, tuned to a wavalength somewhere in the ballpark of 50 microns - the wavelength of the infrared radiation emitted by the pheromone molecules.

Given that the sensors are directional and contain multiple tuned elements, the radio sense would presumably operate more like our sense of hearing than our sense of sight - except that the waves being "heard" are infrared electromagnetic radiation, not mechanical air pressure waves.

Because there are so many little hairs, all of slightly different lengths (tunings), the moth can discriminate the "timbre" of the pheromone radiation from the wash of background noise - in much the same way as the tuned mechanical sensors in our own ears allow us to tell the difference between a distant fire siren and the roar of the traffic.
posted by flabdablet at 4:39 AM on March 2, 2006

Feh. s/wavalength/wavelength/
posted by flabdablet at 4:41 AM on March 2, 2006

We really need someone trained in the field commenting here, because we are entering into an area of contention that likely few of us are qualified to discuss on a knowledgeable level.

Callahan's theories are, at a minimum, either controversial or not generally accepted in the scientific community. It appears he came out a popularly perceived second-best after commentary from M. Diesendorf on Callahan's theories in a 1977 copy of the International Journal of Insect Morphology and Embryology. Those nutty bug scientists, they're always fighting over something. The debate sounds like it might have been exciting, but unfortunately occurred long enough ago in a sufficiently specialized field that it's hard to extract most of the juicy details via standard Net searches. But a search shows Callahan still has his supporters to this day.

Callahan isn't helped by all those who surround his work with new-age goofballery (which isn't to say he wasn't a trained PhD in his own field). A quick look around via Google demonstrates that references to his work often live in unfortunate neighborhoods, and possibly it suffers guilt by association. His apparent subsequent 1979 infra-red study of the famous painting Our Lady of Guadelupe at the Basilica of Guadalupe in Mexico City which determines that parts of the image cannot be explained in natural terms, is "inexplicable", and that "the original picture is a miracle" does not bolster his credibility. I say "apparent" because although the name is the same with frequent minor spelling alteration in cites, and the description of him as a university biophysics doctor with infra-red detection background all match, I don't see smoking gun evidence it is the same person.
posted by mdevore at 11:41 AM on March 2, 2006