A dialog concerning one nose and two eyes
adopted from http://shkrobius.livejournal.com/256863.html?thread=1961311#t1961311
K. Shkrobius, why do we have two eyes and one nose rather than one eye and two noses?
S. Not only will I answer this vexing question in the most satisfactory manner, but I assert that this answering will not require knowing the exact nature of vision!
V. Your boasting is unbecoming in a natural philosopher, S. If you don't know that eyes are used to create a representation of 3D space in the brain, you'd never guess that two eyes are needed for stereoscopic vision and/or for widening the field of view!
S. And I counter that all you need to know is that multiple sensory organs provide additional information conferring some benefit to the animal. It does help to know that it is directional information, but that is not absolutely necessary. The 3D picture can be generated with seven eyes on three heads; it has little to do with having two eyes as such. The reason we have two eyes is tangential to vision itself.
V. That's not good enough. You have to understand why two eyes is better than one, but three eyes are not much better than two. Seven eyes would give as much stereo information as two. Some fish have four eyes, but for very specific reasons. Lizards have experimented with having three, but it didn't go anywhere. I can't think of any animal with one eye, except Euglena and some tiny crustaceans, even in groups where eyes are not particularly prone to trauma, like insects. It might be that the 2-eye design inherited from the common ancestor of mollusks, vertebrates and arthropods is just too difficult to change without significant simplification (like in scallops) because it's too complex. But, again, to evaluate any of these theories you have to know exactly what eyes are for (and we don't always know that, like in case of additional eyes in arthropods).
S. If you argue along these lines, you do need to know the nature of vision. But you can argue along different lines, and then you don't.
I will assume that we are ignorant as to what nose, eyes, and ears are for, yet we are informed on the general constitution of vertebrate animals. All you need to know is that having more than one identical sensory organ will provide extra quality (like directionality) for some sensory inputs (light, pressure) but not the others.
We are bilaterians, and the bilaterians have their organs either in one (the gastric tube) or two (limbs), by their basic design. Now, look at the nose. The nose is always as close to the mouth as possible. So it obviously has something to do with assuring the quality of food. This property is such that directionality does not matter when the food is close to the mouth: you already know where your food is. The organ is unique because you have only one mouth. Furthermore, the sensory input is such that you cannot detect the direction from two locations on the head. You move randomly, sample randomly, and then follow the direction of the greatest gradient. This means chemotaxis. This is all consistent with the idea that nose has some chemotactic or chemosensoric role. Conversely, the nature of this input means that noses will come in ones in the vertebrates. There could be two nostrils, of course - as we are bilaterians, after all.
The eyes and the ears are two. This means that the sensory input travels in straight lines and directionality can be established. Why not four eyes? Four eyes would be necessary if you live at the border between two media that have different laws for propagation of this sensory input, so you need specialized sensory organs for each media. If you know this is not the case, this means other types of specializations, like in snakes, where you have VIS and IR eyes. The third possibility is panoramic input for alerting to predators. Here my evolutionary/ecological argument fully comes into play. Predators do not need panoramic vision. A social group can establish panoramic vision by communication. So the prey will form social groups to alert each other. This solution will be generally preferred to four eyes, because it provides not only alertness but also a new type of intraspecies competition. The need for more than two specialized sensory organs would rarely present itself. This is why you have two eyes and one nose.
Nowhere in this deduction did I use any specialized knowledge about vision or visual processing.
V. So, why are there two nostrils? No explanation at all. May be we have two eyes for the same unknown reason that we have two nostrils? Not all prey is social. Rabbits and mice would benefit greatly from having 3 or 4 eyes... but they don't. In fact, the great majority of herbivorous mammals is not social. Some lizards have 3, so it is technically possible. Mantis shrimps have 2x2 eyes, and apparently some unknown, more advanced kind of stereoscopic vision.
S. The air flow through the nose slowly alternates between the nostrils, because of the turbinate swelling, so the molecules of high and low volatility can both be sensed in a single whiff. http://www.nature.com/nature/journal/v402/n6757/full/402035a0.html
It has nothing to do with stereoscopic olfaction and your analogy is incorrect. Every little kid knows that the two nostrils do not intake air in the same way, and the difference is anatomically obvious. There is no difference between the left and right eyes. It is also obvious from the anatomy that olfaction is central (there is only one olfaction bulb), whereas vision is hemispheric. You do not need to know what vision and olfaction are to notice these significant differences. So I disagree with your criticism. It is all about the basic properties of bilaterian organization.
Bunnies have nearly panoramic vision for the price of having no depth of perception and a 10 deg blind spot in front of them. If you want stereoscopic panoramic vision, then your best solution is indeed becoming social. Again, all you need to know is that in bilaterians mirrored organs are specialized to their side. Having a sensory organ on each side is natural and economic. But the mouth is a singular organ and closely spaced chemosensors cannot be directional; that is why you have only one nose.
The only valid point of criticism I would agree with (but you did not raise this objection) is that, in principle, you can have two widely separated chemosensory organs, one at the mouth and another at the anal pore. As far as I know, only nematoda have such sensilla (amphids and phasmids). See Current Biology 12 (2002) R386 for exposition. I cannot quite explain why this is not the general solution, because it makes a lot of technical sense. Observe that it is not known whether these two noses provide directionality; this is just a hypothesis. Rather, it could be that one nose specializes in chemotaxis, while the tail nose specializes in chemorepulsion.
In any case, you can easily recognize this uncommon situation: to sense the chemical gradients in the most efficient way, the two noses will be separated by the entire length of the body. Having two noses close to each other on one’s face makes little sense. There is another reason, why in land vertebrate animals there is only one nose: there is only one intake hole for respiratory air. We can't have the second nose on our toes because it would be impossible to ventilate it in a controlled way. It is much less obvious why marine vertebrates have only one olfactory organ. But I was not supposed to answer such a question.
V. OK, so you do need to understand the precise mechanism, after all :-)
Virtually all people have visual differences between their left and right eye. Can you be sure that these differences are less important than differences between the nostrils?
Becoming social is not always an option - in fact, very few small mammals are social. To me it seems like four eyes are a better solution. Social mammals occur mostly in open plains, parklands, or in the canopy, where a sentinel can spot a predator from far away, give an alarm call and everybody has time to escape. Mammals living in dense grass or shrub lands, where every fraction of a second counts, almost never become social.
Stereoscopic vision is not the most common reason to have two eyes. Widening the field of view is more important, and it can be achieved by having 3-4 eyes more effectively than by having two.
S. But we are social and the question was about us. Then, what is the merit of stereoscopic panoramic vision to the bunnies living in dense bush? They already have 300 deg vision and their retinal contrast perception is awful. They can detect motion and that's the main thing they need to see. If you do not live in the open spaces or the canopies, why would you need depth perception? You know the sizes of things you expect to see and the distances are determined by the landscape. I do not see great need for these four-eyed superbunnies. It is a lot of botheration for little gain.
V. But we have two eyes simply because our ancestors (small, non-social, nocturnal primates) had two eyes. Having 300 deg vision instead of 360 deg vision in dense bush means that you miss one snake out of six. To primates, who have to choose between binocular and wide-angle vision, lacking a third eye is a grave disadvantage, commonly exploited by, for example, large forest eagles.
S. Missing one snake out of six is a very good deal for a little bunny. It is not supposed to live forever. It is supposed to live long enough to produce fifty more little bunnies.
It is hard to believe that every single ancestor in our lineage had solitary lifestyle. The same goes for any other terrestrial vertebrate. The odds are stuck against four eyes. You have the formidable technical difficulty vs. a much simpler solution that allows going back and forth to solitary lifestyle, back and forth to carnivory and insectivory, etc. Nature is not interested in making superbunnies, it is interested in making the most adaptable lineages. Since your large forest eagles have not been able to decimate our primate ancestors in millions of years, this suggests to me that they must be doing fine with what they have.
V. OK, what if it's a long-living solitary species, like a moose or a tapir? And NO, the bunnies ARE not doing fine. Life span of a mammalian species is estimated to be 1/10 of a life span of an insect species. The rate of extinct/extant forms in mammals is probably 100 times higher than in insects. Can you be absolutely sure that lack of extra eyes is not the reason? Remember, insects have lots of eyes, and most of them have 360 deg. view.
S. Insects do not have 360 deg view. Each compound eye is about 280 deg vs. 220 deg in a bunny. The logomorphs have been around for 45 Myr, which is longer than many modern insects. I do not know what are you concerned about.
OK, you have your moose. It needs to develop an extra pair of eyes at its back. Where would you start? I do not see any way of doing it, frankly. Perhaps the best approach would be having a couple of mirrors on the antlers, like the side mirrors attached to the bicycle helmet. The problem is how to move leucophores (light-reflecting cells) out of the eye onto the antlers, as the mammals do not have iridophores in their skin. That's quite a task. You need an evolutionary scenario like the moose developing metallic looking antlers to impress their females and inadvertently developing a side mirror along the way, as preadaptation. Good luck with this panoramic moose, but that's not four eyes. Developing extra two eyes in the back of one’s head or two eyes on each side of the face would require something extraordinary - for a relatively little gain.
V. Nice idea. But many lizards do have a third eye, so it might be a matter of gene reactivation ;-) Moving this eye to the back of the head is no more difficult than moving one of the nostrils on top of dolphin's skull.
S. This third eye is for circadian rhythm; what good is it for panoramic vision? I recall it was even said that it might be used to concentrate UV light for vitamin D production. Then it is not even an eye proper. In my book, parietal eyes simply do not count. That's just convergent evolution. Also, dolphin’s blowhole can be moved continuously and still be useful to the animal. I can't see how could the eye be moved towards the back of the head continuously being useful all the way. You also need to innervate this eye. There are only so many openings in the skull.
I think you grossly underestimate the delicacy of the job.
V. Woodcocks already have eyes looking almost backwards, so all you have to do is duplicate them...
S. What would be the point: better seeing one's own bill? Perhaps the bird already knows where its beak is. I believe stereoscopic vision does not provide strong survival advantages except in some niches (like in leaping animals).
Of course, all of the above relates to complex eyes. We have not discussed simple eyes. Come to think of it, the birds could've used something like dorsal ocelli that insects use for stabilization of flight. But you need to perceive polarization of light for this function, which means the type of [rhabdomeric] photoreceptor that is missing in the vertebrate animals. This may be the main reason for not having additional simple eyes. The benefit of doubling the two complex eyes is too small, whereas the merit of having extra simple eyes for non-visual functions (timing, orientation) is negligible because the brain is sufficiently large to deduce this information from the visual information supplied by the complex eye. So there is no need for specialization of the eyes into visual (complex) and non-visual (simple) eyes. QED.
Why do we have two eyes and one nose?
K. Shkrobius, why do we have two eyes and one nose rather than one eye and two noses?
S. Not only will I answer this vexing question in the most satisfactory manner, but I assert that this answering will not require knowing the exact nature of vision!
V. Your boasting is unbecoming in a natural philosopher, S. If you don't know that eyes are used to create a representation of 3D space in the brain, you'd never guess that two eyes are needed for stereoscopic vision and/or for widening the field of view!
S. And I counter that all you need to know is that multiple sensory organs provide additional information conferring some benefit to the animal. It does help to know that it is directional information, but that is not absolutely necessary. The 3D picture can be generated with seven eyes on three heads; it has little to do with having two eyes as such. The reason we have two eyes is tangential to vision itself.
V. That's not good enough. You have to understand why two eyes is better than one, but three eyes are not much better than two. Seven eyes would give as much stereo information as two. Some fish have four eyes, but for very specific reasons. Lizards have experimented with having three, but it didn't go anywhere. I can't think of any animal with one eye, except Euglena and some tiny crustaceans, even in groups where eyes are not particularly prone to trauma, like insects. It might be that the 2-eye design inherited from the common ancestor of mollusks, vertebrates and arthropods is just too difficult to change without significant simplification (like in scallops) because it's too complex. But, again, to evaluate any of these theories you have to know exactly what eyes are for (and we don't always know that, like in case of additional eyes in arthropods).
S. If you argue along these lines, you do need to know the nature of vision. But you can argue along different lines, and then you don't.
I will assume that we are ignorant as to what nose, eyes, and ears are for, yet we are informed on the general constitution of vertebrate animals. All you need to know is that having more than one identical sensory organ will provide extra quality (like directionality) for some sensory inputs (light, pressure) but not the others.
We are bilaterians, and the bilaterians have their organs either in one (the gastric tube) or two (limbs), by their basic design. Now, look at the nose. The nose is always as close to the mouth as possible. So it obviously has something to do with assuring the quality of food. This property is such that directionality does not matter when the food is close to the mouth: you already know where your food is. The organ is unique because you have only one mouth. Furthermore, the sensory input is such that you cannot detect the direction from two locations on the head. You move randomly, sample randomly, and then follow the direction of the greatest gradient. This means chemotaxis. This is all consistent with the idea that nose has some chemotactic or chemosensoric role. Conversely, the nature of this input means that noses will come in ones in the vertebrates. There could be two nostrils, of course - as we are bilaterians, after all.
The eyes and the ears are two. This means that the sensory input travels in straight lines and directionality can be established. Why not four eyes? Four eyes would be necessary if you live at the border between two media that have different laws for propagation of this sensory input, so you need specialized sensory organs for each media. If you know this is not the case, this means other types of specializations, like in snakes, where you have VIS and IR eyes. The third possibility is panoramic input for alerting to predators. Here my evolutionary/ecological argument fully comes into play. Predators do not need panoramic vision. A social group can establish panoramic vision by communication. So the prey will form social groups to alert each other. This solution will be generally preferred to four eyes, because it provides not only alertness but also a new type of intraspecies competition. The need for more than two specialized sensory organs would rarely present itself. This is why you have two eyes and one nose.
Nowhere in this deduction did I use any specialized knowledge about vision or visual processing.
V. So, why are there two nostrils? No explanation at all. May be we have two eyes for the same unknown reason that we have two nostrils? Not all prey is social. Rabbits and mice would benefit greatly from having 3 or 4 eyes... but they don't. In fact, the great majority of herbivorous mammals is not social. Some lizards have 3, so it is technically possible. Mantis shrimps have 2x2 eyes, and apparently some unknown, more advanced kind of stereoscopic vision.
S. The air flow through the nose slowly alternates between the nostrils, because of the turbinate swelling, so the molecules of high and low volatility can both be sensed in a single whiff. http://www.nature.com/nature/journal/v402/n6757/full/402035a0.html
It has nothing to do with stereoscopic olfaction and your analogy is incorrect. Every little kid knows that the two nostrils do not intake air in the same way, and the difference is anatomically obvious. There is no difference between the left and right eyes. It is also obvious from the anatomy that olfaction is central (there is only one olfaction bulb), whereas vision is hemispheric. You do not need to know what vision and olfaction are to notice these significant differences. So I disagree with your criticism. It is all about the basic properties of bilaterian organization.
Bunnies have nearly panoramic vision for the price of having no depth of perception and a 10 deg blind spot in front of them. If you want stereoscopic panoramic vision, then your best solution is indeed becoming social. Again, all you need to know is that in bilaterians mirrored organs are specialized to their side. Having a sensory organ on each side is natural and economic. But the mouth is a singular organ and closely spaced chemosensors cannot be directional; that is why you have only one nose.
The only valid point of criticism I would agree with (but you did not raise this objection) is that, in principle, you can have two widely separated chemosensory organs, one at the mouth and another at the anal pore. As far as I know, only nematoda have such sensilla (amphids and phasmids). See Current Biology 12 (2002) R386 for exposition. I cannot quite explain why this is not the general solution, because it makes a lot of technical sense. Observe that it is not known whether these two noses provide directionality; this is just a hypothesis. Rather, it could be that one nose specializes in chemotaxis, while the tail nose specializes in chemorepulsion.
In any case, you can easily recognize this uncommon situation: to sense the chemical gradients in the most efficient way, the two noses will be separated by the entire length of the body. Having two noses close to each other on one’s face makes little sense. There is another reason, why in land vertebrate animals there is only one nose: there is only one intake hole for respiratory air. We can't have the second nose on our toes because it would be impossible to ventilate it in a controlled way. It is much less obvious why marine vertebrates have only one olfactory organ. But I was not supposed to answer such a question.
V. OK, so you do need to understand the precise mechanism, after all :-)
Virtually all people have visual differences between their left and right eye. Can you be sure that these differences are less important than differences between the nostrils?
Becoming social is not always an option - in fact, very few small mammals are social. To me it seems like four eyes are a better solution. Social mammals occur mostly in open plains, parklands, or in the canopy, where a sentinel can spot a predator from far away, give an alarm call and everybody has time to escape. Mammals living in dense grass or shrub lands, where every fraction of a second counts, almost never become social.
Stereoscopic vision is not the most common reason to have two eyes. Widening the field of view is more important, and it can be achieved by having 3-4 eyes more effectively than by having two.
S. But we are social and the question was about us. Then, what is the merit of stereoscopic panoramic vision to the bunnies living in dense bush? They already have 300 deg vision and their retinal contrast perception is awful. They can detect motion and that's the main thing they need to see. If you do not live in the open spaces or the canopies, why would you need depth perception? You know the sizes of things you expect to see and the distances are determined by the landscape. I do not see great need for these four-eyed superbunnies. It is a lot of botheration for little gain.
V. But we have two eyes simply because our ancestors (small, non-social, nocturnal primates) had two eyes. Having 300 deg vision instead of 360 deg vision in dense bush means that you miss one snake out of six. To primates, who have to choose between binocular and wide-angle vision, lacking a third eye is a grave disadvantage, commonly exploited by, for example, large forest eagles.
S. Missing one snake out of six is a very good deal for a little bunny. It is not supposed to live forever. It is supposed to live long enough to produce fifty more little bunnies.
It is hard to believe that every single ancestor in our lineage had solitary lifestyle. The same goes for any other terrestrial vertebrate. The odds are stuck against four eyes. You have the formidable technical difficulty vs. a much simpler solution that allows going back and forth to solitary lifestyle, back and forth to carnivory and insectivory, etc. Nature is not interested in making superbunnies, it is interested in making the most adaptable lineages. Since your large forest eagles have not been able to decimate our primate ancestors in millions of years, this suggests to me that they must be doing fine with what they have.
V. OK, what if it's a long-living solitary species, like a moose or a tapir? And NO, the bunnies ARE not doing fine. Life span of a mammalian species is estimated to be 1/10 of a life span of an insect species. The rate of extinct/extant forms in mammals is probably 100 times higher than in insects. Can you be absolutely sure that lack of extra eyes is not the reason? Remember, insects have lots of eyes, and most of them have 360 deg. view.
S. Insects do not have 360 deg view. Each compound eye is about 280 deg vs. 220 deg in a bunny. The logomorphs have been around for 45 Myr, which is longer than many modern insects. I do not know what are you concerned about.
OK, you have your moose. It needs to develop an extra pair of eyes at its back. Where would you start? I do not see any way of doing it, frankly. Perhaps the best approach would be having a couple of mirrors on the antlers, like the side mirrors attached to the bicycle helmet. The problem is how to move leucophores (light-reflecting cells) out of the eye onto the antlers, as the mammals do not have iridophores in their skin. That's quite a task. You need an evolutionary scenario like the moose developing metallic looking antlers to impress their females and inadvertently developing a side mirror along the way, as preadaptation. Good luck with this panoramic moose, but that's not four eyes. Developing extra two eyes in the back of one’s head or two eyes on each side of the face would require something extraordinary - for a relatively little gain.
V. Nice idea. But many lizards do have a third eye, so it might be a matter of gene reactivation ;-) Moving this eye to the back of the head is no more difficult than moving one of the nostrils on top of dolphin's skull.
S. This third eye is for circadian rhythm; what good is it for panoramic vision? I recall it was even said that it might be used to concentrate UV light for vitamin D production. Then it is not even an eye proper. In my book, parietal eyes simply do not count. That's just convergent evolution. Also, dolphin’s blowhole can be moved continuously and still be useful to the animal. I can't see how could the eye be moved towards the back of the head continuously being useful all the way. You also need to innervate this eye. There are only so many openings in the skull.
I think you grossly underestimate the delicacy of the job.
V. Woodcocks already have eyes looking almost backwards, so all you have to do is duplicate them...
S. What would be the point: better seeing one's own bill? Perhaps the bird already knows where its beak is. I believe stereoscopic vision does not provide strong survival advantages except in some niches (like in leaping animals).
Of course, all of the above relates to complex eyes. We have not discussed simple eyes. Come to think of it, the birds could've used something like dorsal ocelli that insects use for stabilization of flight. But you need to perceive polarization of light for this function, which means the type of [rhabdomeric] photoreceptor that is missing in the vertebrate animals. This may be the main reason for not having additional simple eyes. The benefit of doubling the two complex eyes is too small, whereas the merit of having extra simple eyes for non-visual functions (timing, orientation) is negligible because the brain is sufficiently large to deduce this information from the visual information supplied by the complex eye. So there is no need for specialization of the eyes into visual (complex) and non-visual (simple) eyes. QED.
Why do we have two eyes and one nose?