Primary Scents: Is the iSmell Gadget Real?

[Tsunami]

Not sure whether to place this in this section, or rather in Biology or Chemistry:

there's a new gadget on the market, called the iSmell. According to http://www.villagevoice.com/news/0012,thieme,13436,8.html , this gadget has a collection of more than a 100 'primary' smells, through combination of which you can produce an immense amount of existing smells.

While it would be cool if this actually worked, I have a couple of questions/problems about/with this:

1a) In order to 'reproduce' smells based on a finite number of primary smells, one has to be able to determine how these primary smells are independent from each other, + one has to be able to measure the intensity of these primary smells in each mixture (somewhat similar to the building of all colours based on red green and blue)

1b) I remember reading about a guy called Amoore (sp?) investigating the notion of primary colours in the 90s. His take was that there were 7 primary scents: he investigated this by examining people with scent dysfunctions for certain scents, and discovered 7 different types of cases (ie. each type of dysfunction couldn't function a certain category of scents) and so deduced that there must be 7 primary scents. However, furthering this research, he already came to a number of 32 (I think), which made his idea somewhat dubious. This doesn't say that it is, and maybe his idea has been picked up and improved by others, but my question is then: how?

2) Also, that helps in analyzing the types of independent scents, by how can they reproduce these scents? Ie. how can they address a command to a certain combination of molecules to 'form' a mixture of 10% scent A, 20% scent B, etc. ? These are molecules, don't they interact/annihilate each other?

3) Finally, how can they analyze a certain scent, and determine the gradations they possesses of each primary scent? I can only see this possible through trial & error, but how can they ever make this precise?



In other words, is this marketing crap, or is there something real to it + if it isn't, is it likely that this will be possible in the not so distant future?

 

[Moonbear]

On the chemistry side of things, you could determine the chemical composition of a vapor and replicate it, but that would involve far more than 100 molecules unless you're just replicating a limited number of scents (perhaps for something like an air freshener where you want to change the scent to suit your mood).

On the biology/sensory side of things, there is also a context to scents. The same molecule is found in the scent if buttered microwave popcorn and vomit, but it is pleasant smelling in the context of popcorn, and repulsive in the context of vomit. How or why it works that way, I don't know.

I think you're right to be a bit dubious about the claims. I think it would be wrong to call them "primary" scents, as if you can just mix and match them to get all the scents you would ever be able to detect.

 

[zoobyshoe]

This is something I have never thought about or read about before: what are the primary scents.

If we take the analogy of taste, though, then the number of primary scents would be determined by the number of different scent receptors a given organism has. Our taste is actually pretty limited, I believe, something like: salt, sweet, bitter, and savory (I'm too lazy to google just now).

Pretty much all our senses are limited to the receptors we're eqipped with, rather than what all might be out there to sense.