Blood in the Water

July 12, 2004 8:10 PM

So sharks can smell one part blood in one million parts seawater. How? I understand that the blood would diffuse in water, but wouldn't at least a particle of the blood have to come into contact with the shark? Does diffusion happen that fast?
posted by Coffeemate to Science & Nature (11 answers total)

One part per million might mean one million parts of blood in one trillion parts of water.
posted by caddis at 8:22 PM on July 12, 2004

Yes. Another key, I think, is that one million (or even one billion) parts of water could still constitute a small enough volume to move through a shark's sensing apparatus.
posted by weston at 9:03 PM on July 12, 2004

If my vague memories of high-school chemistry are correct, then one liter of water would contain something like 3 x 10²⁵ molecules of H₂O. So weston's point is very, very true.
posted by kickingtheground at 9:17 PM on July 12, 2004

How?

Well, the leading theory is that certain chemicals fit into olfactory receptor molecules, which activate the associated neurons, but there are others who argue that smell depends on the frequencies of the sensed molecules...

I understand that the blood would diffuse in water, but wouldn't at least a particle of the blood have to come into contact with the shark?

Yes...so what?

Does diffusion happen that fast?

I think there's some kind of fundamental misunderstanding going on here, because the concentration of a solute (something in solution) and its rate of diffusion are two different things. But I'm not quite sure where the disconnect is. Could you clarify your question, Coffeemate?
posted by DevilsAdvocate at 9:24 PM on July 12, 2004

I think he's trying to ask, "How long does it take for my blood to diffuse in the water to attract sharks that aren't close enough to feel the other physical indicators of my presence."
posted by esch at 10:31 PM on July 12, 2004

It has been observed that sharks approach stationary olfactory stimuli from a downstream direction, and are also known to follow scent trails. Nurse sharks in captivity exhibit klinotaxis, or gradient searching, wherein they approach the stimulus in a noticable S-shaped pattern, presumably turning according to which nostrils receives the greatest stimulus.
posted by piskycritter at 5:37 AM on July 13, 2004

Diffusion in water is quite slow, particularly for large molecules found in blood---on the order of fractions of cm per second. The transports that sharks rely on are turbulence and currents. Winds drive surface turbulence, causing mixing down to about 3--4m depth. Surface currents transport the blood cloud, much as wind will blow scents around in the atmosphere. Sharks track the trails of blood upstream, back to its source.
posted by bonehead at 7:41 AM on July 13, 2004

So if you're in a tank that contains 100,000,000 cc's of water (approximately 15 feet long by 15 feet wide by 15 feet deep). There's a shark at the opposite corner and you spill 100 cc's of blood (about 3 fluid ounces). The shark won't instantly smell the blood, even though technically there's one part per million in the tank, because none of it has reached him. Once the blood disperses in the water, however, the shark will be able to smell it. But that won't happen for some time (depending on currents, and how much you and the shark agitate the water).

(Do I have that right?)
posted by jpoulos at 8:27 AM on July 13, 2004

(Pardon the horrible punctuation above. There should be more commas and fewer periods.)
posted by jpoulos at 8:32 AM on July 13, 2004

JPoulos, you're correct. In a completely static environment, like a swimming pool with the filters off, a shark relying on diffusion alone would starve to death (well...). Transport only happens by turbulence (caused by wind/waves and animal motion) and currents (and tides).

The 1 ppm figure has to be for that mililitre of water right at the shark's sense organs. Scent requires contact. Just as you cannot feel a fuzzy towel across a room, you can't smell a compound you don't have contact with. In other words, the blood trail would have to be at least that concentration where the shark crosses it. Seems obvious when I state it that way, but I think that's what you're struggling with.

One part per million isn't that remarkable, by the way. We can smell somethings (like rotten eggs or skunk odours) well below that level. Lethal concentrations for H₂S are on that level---being able to smell that is a survival skill.

Where is a shark's nose anyway? On the gills?.
posted by bonehead at 8:59 AM on July 13, 2004

In sharks, the nostrils do not connect with the pharynx like ours do. They are external pits, usually located in front of the mouth. The nostrils are irrigated either by forward movement of the fish, ciliary movement and/or muscular manipulation of the nostrils.

Read more here.
posted by piskycritter at 12:16 PM on July 13, 2004