Wikipedia:Reference desk/Archives/Science/2007 December 1

Welcome to the Wikipedia Science Reference Desk Archives
The page you are currently viewing is an archive page. While you can leave answers for any questions shown below, please ask new questions on one of the current reference desk pages.

December 1

TV set circuit mystery

What circuit in a TV set goes at (approx) line frequency (and creates bursts of hf radiated noise) but is not locked to it?--TreeSmiler 02:19, 1 December 2007 (UTC)

If you run from a VCR or DVD player the frequency is not locked in precisely and can drift around iwth local oscilators or mechanical speed in the play back. Graeme Bartlett 02:54, 1 December 2007 (UTC)

(edit conflict) Vertical deflection. It goes at exactly line frequency, as I recall, but it syncs on the signal, not the line. I wasn't able to find a Wikilink for it. There doesn't seem to be an article on the innards of a television. --Milkbreath 03:11, 1 December 2007 (UTC)

Ah!. But isnt the vertical sync signal synced to the line freq at the transmitter? Otherwise you would have some bars rolling up or down the screen would you not?--TreeSmiler 03:21, 1 December 2007 (UTC)

I don't think so. You'll see clocks that sync off the line to keep time. The power company tries their best to keep the frequency steady, and they do a darned good job of it, but not perfect. Why would the TV station rely on them when it's a simple matter to generate your own, rock-steady 60 cycles? The vertical oscillator in the set is synchronized by the sync pulses in the signal. Hum bars are a symptom of trouble in the set, like a dry filter cap in the power supply. --Milkbreath 03:55, 1 December 2007 (UTC)

Hmmm! (sorry) That would explain it -- as this article I just found does. Thanks. [1] .Do we have an article on Hum bar Apparently not.--TreeSmiler 04:26, 1 December 2007 (UTC)

I think such a system couldn't have worked in western Europe, because they were using the PAL system which goes 60 Hz, while the power supply goes on 50 Hz. – b_jonas 08:44, 2 December 2007 (UTC)

Wait a minute now. Doesn't PAL use 50 Hz? —Bromskloss 15:50, 4 December 2007 (UTC)

Also the picture is self is strongly modulated by the line frequency as it tends to be similar from line scan to line scan. Then as the whole picture switches scene the noise suddenly stops (or starts). Graeme Bartlett 07:07, 1 December 2007 (UTC)

That seems to be similar what Im picking up on my scope. ie the amplitude of the bursts seems to vary with time (presumably when the scene changes- altho the tv is in another romm so i cant be sure). Do you have any further info on that phenomenon?--TreeSmiler 14:06, 1 December 2007 (UTC)

I expect that "modern" (1941 and after) TV transmitters use a crystal controlled oscillator rather than the utility frequency for the deflection. Early TV systems, such as that of E.F. W. Alexanderson in the 1920s, used mechanical scanning and had a motor synchronized to the power system. The receiver was likely powered by a line locked in synchrony with the power line at the transmitter, so there would be only a slight phase difference between them which was easily adjusted and fairly stable, to prevent the picture rolling. Edison 03:00, 2 December 2007 (UTC)

CD4+CD25bright

Is CD4+CD25bright another way of saying CD4+CD25high? --Seans Potato Business 03:08, 1 December 2007 (UTC)

CD25 bright is CD25 high, and CD4+ is just CD4 positive, in case that wasn't clear also. Someguy1221 00:15, 2 December 2007 (UTC)

Strong Interaction Exploitation for Macro-Engineering Hobby Project

Having nothing else scheduled for the day (the day in question being sometime late in the 100th century AD), I'd like to send a von Neumann machine off into space to have a family and then spend an epoch or two transforming some uninhabited planets into a hybrid Dyson Sphere/Ringworld (which I imagine as an oblong terraced sphere rotating on its long axis to generate artificial gravity on the interior) and am deeply concerned that scrith may not be available on eBay at that time, so I may be at a loss for a material strong enough to make it big enough to also be interesting. I'd like to carry as much of the load as possible under tension, so I'm imagining an innumerable number (which is to say that I have decided not to count them) of spokes that cross the span from one side to another. I need strong wires- much stronger than carbon nanotubes. I guess I have a little while to work out some of the details but here's the best I've come up with. (Here, have a complimentary grain of salt before continuing). By manipulating infinitesimal black holes in ways yet unspecified, confine a stream of protons (or, as long as we're playing around with confined spcaes, super-protons made from decay-proof top and bottom quarks ;-) into the extremely confined space-time geometry of the tiny black hole's interior, and then, taking advantage of very clever timing, trajectories, and an Acme(r) Maximum Luck Field Generator, the black hole's own tendency to evaporate due to Hawking Radiation as a way of opening up an end in another place (white hole? Aaaargh), hope (against hope) to hold the geometry open with the string of protons held inside. Then, inspire the string of protons to behave as a single macro nucleus- only with all its strength focused on a single one-dimensional axis- by providing an appropriately charged negative charge at the ends (simulating the presence of electrons), which also, by the by, provides the means for attaching the ends of the "wire" to the things it needs to hold together. Being particularly naive from this particular technological vantage point, I'm hoping to cheat on the distances required- make the strings short, but connect points that are very distant. Essentially, I think I'm imagining a narrow-throated Einstein-Rosen Bridge held open, not by some exotic anti-energy, but by the presence of the protons themselves. At the end of the day (when it is appropriately dark), my question is singular. How strong could a "confined geometry strong force tension member" be? I'd be happy with anyone's best guess, considering. DeepSkyFrontier 05:51, 1 December 2007 (UTC)

Ringworlds and Dyson spheres have one exceedingly difficult problem. They are unstable. If one part of a ringworld gets a little bit closer to the star than it should be - and (consequently) the part on the opposite side is a bit too far away then because gravity gets stronger as you get closer to the star, the side closest to the star is attracted more strongly than the part on the far side - which serves to accentuate the problem. Hence you have a system that just won't stay put. SteveBaker 06:29, 1 December 2007 (UTC)

I'm sure you could manage some kind of stabilizing propulsion with all that energy you're collecting from the Dyson Sphere ;-) Someguy1221 11:14, 1 December 2007 (UTC)

What are you using for reaction mass? Remember, your sphere weighs at least as much as all of the planets, moons, asteroids and other debris in your solar system. SteveBaker 23:38, 1 December 2007 (UTC)

The step-terrace scheme will fail, I think. How would you prevent each latitude ring from falling toward the equatorial plane? For your strong-force cables, I think we have a little problem with electrical charge. For your proposed schedule, forget it. The universe will have converted itself to pure computronium long before that time. -Arch dude 13:36, 1 December 2007 (UTC)

To actually answer the question, I think you're confusing the nature of these geometric (actually topological) structures, as they do not extend in any spatial dimension. They are, from our perspective, 0-dimensional, and have another dimension in hyperspace. What this means is that you can't use it as a tension or compression member because it doesn't extend out in any real direction - it is just a hole is space. This is if I understand your question correctly.

Furthermore, these holes cannot be stabilized with photons. I'm afraid you need exotic matter - no photon is able to penetrate the hole before it collapses.

And to answer Steve, lots of things are unstable, like Maglevs, LaGrange points, and modern fighter jets. That's why we have corrective feeback loops. SamuelRiv 14:07, 1 December 2007 (UTC)

Yes - but it takes energy (which I'll grant you have plenty of) and reaction mass. As you throw off mass, you have no way to recover it - so gradually, your system 'evaporates'. The mass of a Dyson sphere is immense - it's estimated that you'd need all of the mass of our solar system to form even a skinny ringworld. You aren't talking about pushing a spaceship around - you're talking about pushing the combined masses of Jupiter and Saturn around. For a full 1 AU sphere - thousands of them! A 1 AU sphere would rapidly cook it's inhabitants without some kind of unobtainium heat radiators...so it's gonna be a lot bigger than 1 AU. The power involved in moving this thing is decidedly non-trivial! It's pretty well established that these things are impossible for a bunch of practical reasons...but I find instability to be the biggest. SteveBaker 23:38, 1 December 2007 (UTC)

I answered the question, but asked for clarification. You cannot use such bridges as tension or compression members - the new space that you create will just stay the same as you move your holes closer or farther apart. There is a possibility that you'll lose effective mass due to gravity waves if they are accelerating relative to each other, but I don't see a way for them to act as a tension member in the way you're describing. SamuelRiv 16:29, 2 December 2007 (UTC)

Protons, for one thing. Yes, thank you. You essentially answered the question by saying, "You can't ask this question this way." What I'm asking has, at its root, nothing to do with black hole geometry. I'm asking about how strong the nuclear strong force is. If it could be exploited- by whatever means- what would be the result. Thank you for addressing the question I asked. Your edit interfered with my own restatement of my question that replaced the comment that you responded to. DeepSkyFrontier 16:45, 2 December 2007 (UTC)

Most desirable place to live

Aside from the fact the individuals may have different preferences, where is the most desirable place to live, strictly as an environment and in total if all Friends, possessions, etc. could be relocated there, or is this more a matter of personal history such as where you were born, where you grew up, where you had the most rewarding experiences? Second part of the question is where can I find an online questionnaire that will ask me different questions like "What kind of climate do you prefer: warm, mild, cool, varied...?" and then narrow down the places that meet the criteria I entered? 71.100.1.143 07:02, 1 December 2007 (UTC)

It depends entirely on the criteria you choose to rate. There are plenty of attempts to rank the best places to live, both globally and in specific countries. Here are links to some of them: [2] [3] [4] [5] [6] As for an online questionnaire, try this if you are interested in the USA. Rockpocket 08:11, 1 December 2007 (UTC)

Perhaps it would help to work backward somewhat and decide where the least desirable places to live are and why. For example beyond mere climate considerations, living near an active volcano might not be a good idea. (Also consider places that experience lots of earthquakes, hurricanes, floods, etc.) Besides natural issues there are also man made hazards to consider like areas of high crime, or other armed activity. Along man made lines, also important factors would be local health care and whether or not one could drink the local water without getting sick. Anynobody 08:52, 2 December 2007 (UTC)

The irony of this is that if there is a place that is considered by a large number of people to be "the most desireable place to live", then fairly quickly everyone moves there, and then it becomes not such a desireable place to live. An example of this is Los Angeles, which 100 years ago was pretty close to what most people would consider to be paradise: great climate, layed back attitudes, ocean front, mountains, orange groves stretching as far as the eye could see... But now look at it: traffic congestion, smog, you can't see the mountains most days of the year because of the pollution, very high crime rates, and just too many people. So if your idea of the best place to live includes any of these last things, perhaps Los Angeles WOULD be the perfect place for you to live, otherwise you're going to have to have a compromise. You can't have it all... You should prioritize which things you feel are more important about where you'd want to live, and then look for a place that has those things the most. FOr myself, I highly value low crime, no gangs, few people, lots of trees and wilderness, and lots of water. So I ended up living in a place where there's lots of snow and cold winters, which I don't really like, but that was my trade-off... I just use myself as an illustration. Where I live people never lock their homes or cars, and the biggest crime one hears about is when someone gets a DUI because they've been drinking too much due to all the snow and depressing weather! Saukkomies 05:08, 4 December 2007 (UTC)

Clanks

I just read the Girl Genius series, and there's something I can't help wondering. Just how far can clockwork robotics be pushed? I know some remarkable things were invented in the old days using nothing more than gears and pulleys, up to and including small programmable robots, but just how much is possible? Black Carrot 11:17, 1 December 2007 (UTC)

The non-computer parts of a robot are mostly "clockwork", so you are actually asking about the computer parts. It is theoretically possible to perform any computing task using a mechanical computer, because you can build a mechanical Universal Turing Machine. In practice and using mechanical compones built with today's technologies, an elaborate purely mechanical robot is not feasible. A "mechanical" computer becomes possible with nanotechnology. At that size scale, a mechanical computer may be smaller than an equivalent electronic computer because electrons are hard to confine to small volumes. -Arch dude 13:20, 1 December 2007 (UTC)

If you want to be amazed, look up examples of medieval and Renaissance automata. It's really quite amazing what you can do with the technology; it really just comes down to how many gears you can fit into a given space and how much precision you can have with the gears. I've seen some amazing videos of automatons that could write, play complicated instruments, etc. Obviously they aren't intelligent in any sense but they can fake some pretty fine motor skills. --24.147.86.187 18:08, 1 December 2007 (UTC)

I was too hasty earlier. Sorry. There is a class of non-digital control mechanisms that can be used to create analog robots. You can get really complex behavior from relatively simple analog electronics. A lot of very interesting devices of this type were developed prior to the development of the first digital computers. A famous example was the Nordon bombsight. The mechanical equivalent of an analog circuit is perfectly feasible and easy to implement. By contrast to digital, the analog approach is much more compatible with a Victorian/steampunk worldview. -Arch dude 22:22, 1 December 2007 (UTC)

There is a mathematical theorem called "The Church–Turing thesis". It says that in principle, any minimally capable computer can do anything that any other computer can do - providing it has enough time and memory (and of course the right peripherals - printers, robot motors, whatever). This minimal computer is called a Turing machine. So if you could build a clockwork Turing machine - then it could do anything the PC in front of you right now could do - if there were enough memory and time. Looking at the architecture of the Turing machine, it's quite plausible that someone could make a clockwork machine like that. Here, for example is a "Turing machine" made from toy trains. Well, those could be clockwork trains...so there you are.

OK - so that's the theory...but in practice? Well, no. Turing machines are S-L-O-W - and a clockwork machine would be HIDEOUSLY slow. It would need winding up A LOT! If you look at something like Babbages' Analytical engine, it's full of gears and such - and it's equivelent to a Turing machine - so the Church-Turing thesis says it's a proper computer. It's still pretty slow - and I think a literal clockwork motor might be a little impractical as a power source - but a good sized steam engine would work. It's huge - and still pretty slow. If it had ever been completed (it never was) it would have been able to store 50,000 decimal digits - which is about 200 kbits of memory. That's more than the first computer I used in my first job (an Intel 8008 with 4kbytes of memory). It would have been able to multiply to twenty digit numbers, in three minutes. That would take about a billionth of a second on a modern PC - so this machine would take thousands of years to something as (seemingly) simple as displaying the page you are reading right now.

So - yes, in theory, mechanical computers could work - but they absolutely could not be practically useful.

There is a caveat to that - which is that if nanotechnology lives up to some of it's more extreme promises, we'll be able to make very small mechanical systems that would actually be faster than modern electronics. If we get to that point, then mechanical ("clockwork") computers will be back in vogue.

SteveBaker 22:47, 1 December 2007 (UTC)

(What was I thinking?! No discussion of mechanical computers could be considered complete without mention of the one I happen to own!) Mechanical computers do actually exist - you can buy one called the Digi-Comp I (I own one - it's the one in the photo in that article). This thing is made of plastic (if you have an original one) or cardboard (if you have the modern version). You put it together yourself in about an evening. It's pretty minimal - it has a three bit memory and it's power source is you wiggling a lever back and fourth. It does one 3-bit calculation each time you move the lever across and back - maybe one per second (if you go too fast it jams up)...but it can play the game 'Nim', count and add a 1 bit number to a 2 bit number to produce a 2 or 3 bit result! Yes, it's horribly limited and pretty complicated for a cardboard computer. You program it by moving little bits of drinking straw around - you input data by directly switching it's 3 bit memory - and the output "display" is three little windows that show '1' or '0' depending on ths state of the memory. There was once a Digi-Comp II which used marbles rolling down a pinball-like machine with little plastic levers. It had a dozen or so bits of memory and could do multiplication (eventually) - there was a little electric motor driving the mechanism that recycled the balls to the top of the track. SteveBaker 23:11, 1 December 2007 (UTC)

Ah, yes, the Digi-Comps. The Digi-Comp II my uncle gave me for Christmas when I was 10 years old or so was certainly a formative experience. And I do still have it, somewhere; I'll have to haul it out and take a picture of it for our article.

It's quite an amazing machine. It's got a 7-bit accumulator, a 4-bit addend register, and a 3-bit multiplier register. There was even a way to rig it up to do long division, but it required about a hundred marbles to filter through, all without error, and it was virtually impossible, but fun to try. (And mine had this great bug where a marble would occasionally bounce out of the track it was supposed to be in, roll all the way down to the bottom of the board, bounce back up, fall into an unrelated hole, and start a calculation over again that was supposed to be terminating. Loads of laughs.)

I never heard of an electric motor, though; that may have been some sort of field upgrade. Mine, you had to recycle marbles back up to the top yourself if a long multiplication or division required more marbles than you had. —Steve Summit (talk) 02:01, 2 December 2007 (UTC)

Are you interested in selling it? I've always wanted one and I'm no too fussed about the condition it might be in. I have frequently considered building one out of Lego - but without complete details, it's hard to figure out exactly how it works. Contact me at my talk page if you are interested. You're obviously right about the lack of a motor - I have never actually seen one of these gizmo's actually working. I assumed there must be a motor or something because the description of them always talks about an 'Automatic' mode of operation versus 'Manual' - which I took to mean that there must be some kind of automatic ball feeder...but there was certainly only one version of the Digi-Comp II so if your's has no motor then that's that.

The Digi-Comp I (which I have) was made in polystyrene a year or two before the II came out (around 1963 I think). However, they were flimsy and broke easily - and hence they are much rarer than the Digi-Comp II (in working condition at least). A few years ago an enthusiast made a replica of the Digi-Comp I out of heavy cardboard and is now selling these replicas in kit form. However, so far, nobody has ever built a replica of the Digi-Comp II - and whilst it's mechanically easier to understand than the Digi-Comp I, it's actually the more computationally powerful of the two machines. Of course, neither of them are 'Turing-complete' - so in a sense, they are not true computers. However, they are both programmable and can perform arithmetic, so in another sense, they are computers.

For the technically minded, the Digi-Comp I is essentially just three flip-flops which are clocked synchronously. The "wiring" between them is handled by 'programming' the machine by putting short sections of drinking straws onto the front of the machine where they can block parts of the machine from moving. You can also insert mechanical analogs of "AND" and "OR" gates in front of the inputs to the flip-flops (with certain constraints). Because there is only a three bit memory, if you want a program to actually step through multiple instructions, you have to use one or two of the bits as your "program counter" - which leaves you with very little left for doing actual work! But you can set it up to do things like count from zero to seven (which is an impressive sight actually!) - and it can play a minimal version of the game of Nim using seven objects in one heap...which is actually the only program for it that's actually any use for anything! Programming the thing to do ANYTHING is absolutely mind-bending - you need a very good knowledge of what flip-flops can do. The idea that it might ever have been a kids toy is ridiculous!

SteveBaker 05:52, 2 December 2007 (UTC)

Do you know of any good books on low-tech robotics? Spring-powered, and such? Black Carrot (talk) 03:31, 5 December 2007 (UTC)

Hypothalamus and mammillary body

Reading Hypothalamus#Nuclei gives me the impression that the mammillary nuclei (and thus the mammillary body) are a part of the hypothalamus. Is this true? Lova Falk 14:15, 1 December 2007 (UTC)

Neuroanatomists have often categorized the mammillary bodies as part of the hypothalamus. One anatomy text that does so says, "synthetic descriptions may suggest that all hypothalamic nuclei are well delimited structures" and then it describes how hypothalamic nuclei transition gradually into surrounding brain regions. --JWSchmidt 15:14, 1 December 2007 (UTC)

Thank you! As I usual do with your explanations, I quoted you in mammillary bodies. Could you change the reference to a proper one? Lova Falk 16:34, 1 December 2007 (UTC)

You want me to admit how old my neuroanatomy textbook is? Ow. The quote I gave above is from Human Neuroanatomy (8th edition) by Malcolm B. Carpenter and Jerome Sutin (1983) ISBN 0-683-01461-7.--JWSchmidt 19:50, 1 December 2007 (UTC)

That's almost prehistoric. But it's fine for an uncontroversial question like this one. Maybe I should never have thrown my 1977 book called Psychophysiology... —Preceding unsigned comment added by Lova Falk (talk • contribs) 08:40, 2 December 2007 (UTC)

McDonalds Fries

"because there's too much moisture in the potatoes for them to crisp during baking, I needed to release some of the water first. I ran home at record speed (for me), went directly to the kitchen, and tried parboiling the fries before baking them" - am I not correct in thinking that boiling (par or otherwise) will increase the amount of water in the potato?--Seans Potato Business 19:30, 1 December 2007 (UTC)

What this refers to is commonly called blanching - and is required to make crispy fries. It is the reason that most home-cooked fries aren't as good as fast-food fries. Boiling them makes the water in them expand. They are pulled out and quickly cooled to stop the cooking process. Since the outer layer is mostly cooked - it is no longer very permeable and the cold water has great difficulty soaking back into the potato. The end result is a reduction in water inside the potato and a pre-crisped outer layer. This is highly effective in most "wet" foods that you wish to deep fry. The trick is timing the boiling process so that the cold water won't soak back in. -- kainaw™ 21:14, 1 December 2007 (UTC)

Of course the classical way to make fries is to fry them twice - once to a light brown, then take them out of the oil, reheat it, and do the second stage frying at a somewhat higher temperature until they are golden brown. This should have a similar effect. The result also depends on the kind and age of potato, of course. Very fresh ones are not as good for frying as potatoes that have been stored for a while (in a proper cool, dry environment). --Stephan Schulz 22:35, 1 December 2007 (UTC)

• For a wonderful survey of these sorts of nitty gritty food details, check out On Food and Cooking, by Harold McGee. --Sean 00:03, 2 December 2007 (UTC)

I read a recipe decades ago, by Julia Child, in which she said to cook the fries at the high temperature first rather than second. A home deep fryer usually does this rather automatically, because the initial fat temperature drops shortly after the fries are added, probably because the moisture flashing to steam takes heat from the oil. A restuarant fryer has so much heat capacity in the larger amount of hot oil, as well as a more powerful heating element that the temp stays more constant as things are lowered into the oil. Edison 02:55, 2 December 2007 (UTC)

Some people salt the chopped potato a while before cooking and then dry them with a paper towel just before tossing them into the oil. I presume that the purpose of this is to draw water from the fries via osmosis. SteveBaker 14:49, 2 December 2007 (UTC)

Jesus illusion

What's the general name for optical illusions of the kind of the famous "Jesus illusion"? Icek 20:19, 1 December 2007 (UTC)

I'm still looking, but if you like that, you'll love the lilac chaser. --Milkbreath 20:44, 1 December 2007 (UTC)

It's called an afterimage, explained by Ewald Hering. Sadly, it is not called the Hering illusion, though. This site calls it a "stare negative". --Milkbreath 21:08, 1 December 2007 (UTC)

With a dose of pareidolia thrown in for fun. Matt Deres 02:07, 2 December 2007 (UTC)

Goethe was the first to write about this. That Jesus looked more like Che Guevara to me! Saudade7 03:02, 3 December 2007 (UTC)

Thanks for the answers. Saudade7: Maybe Jesus looked like Che Guevara, I don't know ;-) Icek 02:11, 4 December 2007 (UTC)

All I know is that the Jesus Illusion works in some way similar to that of "photograph negatives"; you'll find that if you stare at anything which contrasts greatly between bright and dark (such as a bright sunny window) for 20 seconds or so and then close your eyes, the same will happen; you'll see a sort of negative photograph effect. Sometimes you don't even need to stare; if you have your eyes shut, and as quickly as you can open and close them, then yet again, an illusion occurs. It seems similar to me as screen burn which can happen with LCD screens, like the light receptors on your retina have been temporarily influenced. Is "temporarily damaged" a good way of explaining it? Adamd1008 18:46, 4 December 2007 (UTC)

snowing

hello im shima ghadiri im from iran and its my first time that i meet your site and now im a new member of your site and i have really strong confidence to be areally active member im 15 years old and i study math in grade two in high school i have alot of question in my mind and i thouth this is a best way to find some one in your sitemaby some abroad iranian at the end my english is not raelly good but i can express my idea thank u alot. my first question is:why when is snowing we dont have any i dont know exact word for example light in sky but when rainig we have alot of light in sky that in persion we say radvabargh?

Do you mean a rainbow? --Reuben 20:54, 1 December 2007 (UTC)

For language questions, try our Wikipedia:Reference desk/Language. Of course we do have a Wikipedia that's written and edited entirely in Persian/Farsi here - it has over 28,000 articles (including one about rainbows [7] !) - but that's only a small number compared to the number we have in English. Perhaps a good way to practice your English would be to translate some articles from the English Wikipedia into Farsi - there are plenty to choose from! SteveBaker 22:12, 1 December 2007 (UTC)

Well, if the question is "why do we see rainbows when it rains, but not when it snows?" that's certainly a science question. The answer is that raindrops are close to spherical in shape while snowflakes are not. This matters because a sphere is very symmetrical and therefore when light from the Sun hits a raindrop at a certain angle it will always emerge at a particular angle (for the same color of light) -- the same angle from one raindrop to the next. What makes a rainbow is the rays of light emerging from different raindrops in the same direction. Snowflakes have complicated shapes and the flakes in a mass of snow are all facing different ways. This means that the light from different snowflakes does not point in the same direction, so it does not form a rainbow.

Another possible form of water in the sky is ice crystals. These have shapes more complicated than raindrops but simpler than snowflakes. In some conditions they become aligned (facing the same way) and then you can get an effect somewhat like a rainbow. There are different types; two of them are called sun dogs and halos.

--Anonymous, edited 22:29 UTC, December 1, 2007.

I just wanted to say that that is a delightful and very informative answer... - Nunh-huh 22:51, 1 December 2007 (UTC)

Nice to meet you Shima! I too am working on the assumption that you are refering to a rainbow. Although I am no meteorologist, I can say that from personal experience I have seen a rainbow while it was snowing, not raining. It was very remarkable and everyone I was with was amazed by it. It occurred during a small snow storm in the state of Idaho in the US, and the sun was low in the sky - it being in the afternoon. --Saukkomies 19:32, 2 Dec 2007 (EST) —Preceding unsigned comment added by Saukkomies (talk • contribs)

Is this the word in Persian: رنگين‌ كمان‌، قوس‌ و قزح‌، بصورت‌ رنگين‌ كمان‌ ? That is the word for Rainbow according to this online dictionary I looked on. I just want to make sure we are answering your question correctly! Saudade7 03:09, 3 December 2007 (UTC)