Wikipedia:Reference desk/Archives/Science/2007 May 26

= May 26 =

The gulls have started laying their eggs...
As usual, I can see a black-back nest from my window and I was watching the mother gull today. When not sat on the eggs and keeping lookout, she was stood up so as to allow her eggs to cool slightly, or turning the eggs with her beak and feet to ensure even temperature distribution. I had a moment of realization that the poor hen will be up and down, doing this day and night for the next month or so, with little opportunity for sleep (the male bird brings her food but I don't know where he goes at night). I do feel kinda sorry for her.

Now I'm wondering, which requires the most energy expenditure - constantly tending and protecting an egg until hatching, or nurturing an embryo inside the body for the same period of time? --Kurt Shaped Box 09:30, 26 May 2007 (UTC)


 * You also have to consider not only energy use, but convineance. It's a lot easier to feed yourself when you aren't carrying around a large weight. Actually, come to think of it,this might be why birds lay eggs, it would make sense, since they would struggle to fly with the weight of a foetus, wouldn't they? -- Phoeba WrightOBJECTION! 11:24, 26 May 2007 (UTC)


 * I'm pretty sure that's the accepted thinking - someone correct me if I'm wrong... --Kurt Shaped Box 15:07, 26 May 2007 (UTC)


 * Do pregnant bats fly? I think the egg thing is just the way they are. After all, those eggs still have to be formed inside the hen, and she has to fly around carrying the preformed eggs inside her, or at least the materials required. --0rrAvenger 17:40, 26 May 2007 (UTC)

Perhaps marsupials have the easiest job for the mothers. They don't even seem to notice when they give birth, as the offspring are so small at the time. They don't have to tend any eggs, either, as the offspring find their own way to the pouch and attach to a nipple. StuRat 18:02, 26 May 2007 (UTC)


 * I wonder if laying eggs is as painful as childbirth in humans? It's very hard to tell if a bird is in pain. When my budgies lay, the hen gets a sort of 'intense' look in her eyes - but that could just be the result of her concentrating and putting all her energy into pushing... --Kurt Shaped Box 09:48, 27 May 2007 (UTC)

Nuclear forces
I'm trying to come up with some simple diagrams explaining why nuclear fusion is difficult. The main reason, as I understand it, is that at most levels of interaction the hydrogen ions are positive and thus will repel themselves due to electrostatic repulsion. However if you can get them close enough, the nuclear force should kick in and make them fuse. My questions are: 1. what approximately are the radii that the electrostatic repulsion and nuclear force are active for hydrogen ions? How close do they have to be, in other words, before they fuse? Is this the Coulomb barrier? and 2. why does it release energy? I know that you can calculate the binding energy etc. by looking at the missing mass etc. but in a physical sense what is the mechanism of the energy release? The example given in nuclear fusion seems to say that most of the energy released in a D-T reaction comes from the ejection of a neutron and the subsequent recoil — is this a good way to think about it? (I find fission easier to conceptualize if instead of thinking of it as "converting matter into energy" I think of it as two positively charged nuclei repelling one another suddenly; is this an analog to that?)

Thanks for the help. This is not homework, this is just a non-scientist trying to make sense of this so I can convey the information to other non-scientists. I think I get most of it though I'm having trouble making use of some of the pages (i.e. Coulomb barrier) which seems to require a lot more knowledge of quantum mechanics than I have to even use it (i.e. knowing what "permittivity of free space" means, which even when I click on it is written in 100% technical language). --24.147.86.187 00:35, 26 May 2007 (UTC)


 * The coulomb barrier is how hard you have to throw the 2 atoms to get them to stick together - that's why fusion requires so much heat to work. That is, you give it lots of energy to get over the barrier; then it gives that energy back plus a bit more.
 * The non quantum introduction to nuclear is the semi-empirical mass formula or liquid drop model, which largely balances strong force and coulomb repulsion.
 * These balance out to something like this curve: binding energy per nucleon graph. Between H and O the binding energy rises sharply because the number of nucleons on the surface (which are less tightly bound) compared to the total number of nucleons goes down. Above Fe, the coulomb force dominates.
 * The D-T reaction (H2+H3) produces He4 (+n). He4 is quite special because 2 protons and 2 neutrons both match the magic number 2 - this allows for a larger than normal binding energy (this is a quantum effect from the idea that both neutrons and protons exist in shells like electrons in atomic orbitals).
 * --h2g2bob (talk) 06:39, 26 May 2007 (UTC)
 * Energy needs to be kept by the atom to overcome coulomb forces, which are repulsive. Energy is given away by the binding force which is attractive. It sounds dumb, but it's the best I can come up with. --h2g2bob (talk) 06:57, 26 May 2007 (UTC)


 * I have searched for a picture of the potential, this is the best one I found. The source of the energy liberated in a fusion is the acceleration caused by the potential difference between the outside region and the center of the well. You can see that the question whether energy is liberated or captured depends on the height difference. http://upload.wikimedia.org/wikibooks/en/7/7a/Fhsst_atomnucl6.png


 * How close do they have to be, in other words, before they fuse? It is roughly 1fm, the size of the resulting nucleus. However if you really use the Standard Model on this, you will only get the probability of a fusion and this probability will depend at least on the relative speed and the distance at the point of nearest approach. I have no idea how the probability will vary exactly with these parameters, it depends a lot on the details and is way over my head.

I was told by a chemistry teacher I once had that if the Hydrogen before it is fused is wieghed then it would wiegh more than the fused Helium, the lost mass is turned into energy.67.125.159.224 23:26, 26 May 2007 (UTC)


 * The sum of the mass of a core of deuterum and a core tritium is more than the sum of the mass of a core of helium and a neutron. The rest of the sentence is problematic. Mass cannot be turned into energy because mass is already energy. The energy of an object can be measured from different inertial frames. If the frame moves fast relative to the object the energy measured will be higher than when it moves slow. If the frame of reference does not move at all relative to the object, the measured energy is called the rest energy, rest mass or simply "mass".

Equivalent Resistance
Hello. The formula for equivalent resistance for a parallel circuit is Requivalent = 1/R1 + 1/R2 + 1/R3. How was this proved? Thanks. --Mayfare 00:54, 26 May 2007 (UTC)
 * It's 1/Requivalent = 1/R1 + 1/R2 + 1/R3. Proving it is quite simple.  If you take a very simple circuit consisting of a perfect voltage source of voltage V, and three resistors in parallel, firstly, you know that the each resistor has no effect on the current running through the others.  The current through each resistor is V/Ri.  The sum of the currents through the resistors is V/R1+V/R2+V/R3, or V*(1/R1+1/R2+1/R3).  Ohmic resistance is defined as R=V/I, or 1/R=I/V, so 1/Requiv=V*(1/R1+1/R2+1/R3)/V, and therefore 1/Requiv=1/R1 + 1/R2 + 1/R3.  Someguy1221 01:03, 26 May 2007 (UTC)

If R = V/I, then does that mean Requivalent = V/Isum? Since Vs = V1 = V2 = V3, can Requivalent = V/V(1/R1 + 1/R2 + 1/R3) be correct yielding to Requivalent = R1 + R2 + R3, the formula for equivalent resistance for a series circuit? I am confused. --Mayfare 01:38, 26 May 2007 (UTC)
 * Everything in there is correct except the last step. 1/(1/R1 + 1/R2 + 1/R3) is not the same as R1 + R2 + R3.  For ease's sake, let's change the resistances to a, b, and c.  If 1/(1/a+1/b+1/c) equaled a+b+c, this would imply that 1/(a+b+c) equals 1/a+1/b+1/c.  This is logically impossible for positive values, as the former gets smaller as terms are added to the denominator, and the latter gets larger.  Someguy1221 02:44, 26 May 2007 (UTC)


 * One way to think about (and remember) this is that for resistors in parallel, the conductances add, just as the resistances add for resistors in series. The conductance, (measured in Siemens (unit), ℧, sometimes referred to as "mho") is defined as 1/resistance.  —Steve Summit (talk) 13:09, 26 May 2007 (UTC)


 * For completeness's sake let us also mention that the starting point for all this is Kirchhoff's circuit laws. They tell us that in the case of the resistors in parallel, each one gets the same voltage, and in the case of serial connection, we all have the same current flowing through them. Simon A. 17:45, 26 May 2007 (UTC)

Elephants mammary glands
Why do elephants have their mammary glands at the front of their bodies unlike other mammals which have them towards the end of their bodies? 144.138.100.49 03:31, 26 May 2007 (UTC)


 * Thats not actually so. Most mammals have a number of mammary glands distributed along the milk lines, parallel with the anterior/posterior axis of the ventrum. For example, mice have 5 pairs of mammary glands, the most anterior being between their forelegs, which is the same relative position as human and elephant mammary glands. The unusual things about elephants is that that have a single pair, not the position of that pair . Of course, the most likely reason the have a single pair is because of their small average litter size of just over 1, and pretty much all mammals that have a single pair have them anteriorly positioned. If you are interested in how the glands form in the position they do, offers some insight into the transcription factors involved in regulating their development.  Rockpock  e  t  07:15, 26 May 2007 (UTC) (amended due to Steve Summit's excellent examples, of a few that don't, leading to further research, below  Rockpock  e  t  18:20, 26 May 2007 (UTC))


 * I think the Original Poster can be forgiven, though, given that our Elephant article states that "Unlike most mammals, female elephants have a single pair of mammary glands located just behind the front legs." And while you maybe right about "pretty much all mammals that have a single pair have them anteriorly positioned", many of the examples we're all most familiar with (goats, horses, deer) have them in a posterior position. —Steve Summit (talk) 13:00, 26 May 2007 (UTC)


 * It occurs to me that we're overlooking the most familiar exception to that quoted rule of thumb: us. On the human female, a single pair of mammary glands are located just behind the front legs.  (It's just that we call our front legs 'arms', and 'just behind' becomes 'just below' now that we're walking erect.) TenOfAllTrades(talk) 13:35, 26 May 2007 (UTC)


 * It's interesting to note that people who have a third nipple do indeed have it along the milk lines. --TotoBaggins 18:07, 26 May 2007 (UTC)


 * Good points, Rockpocket and Steve. So, our domesticated milk-giving mammals are the exception: they have several glands, and they are all quite posterior. So, why is that? If these milk lines that Rockpocket mentioned spread all the way along the belly, why are they so concentrated to the rear? Don't the arrangement of a mouse looks much more useful than such a bulky udder? Actually: are there udders in non domesticated mammels? Simon A. 17:32, 26 May 2007 (UTC)


 * So I have a done a little more reading on this and it turns out that my original comments are a little misleading (so I've struck them). There are actually a range of positions of mammary glands across mammalia, for example:


 * However, there is much variation within species, both in number and absolute position, and the reason the positions are so defined and pronounced in domestic animals is due to inbreeding and human interference. For example 50% of the dairy calves are born with more than 4, but the extra teats are "removed" when the calf is only a few days old, so they can better fit into the milking machine (think of that next time you enjoy a glass of milk!). I'm afraid that's not true.  Very rarely are calves born with extra teats.  Healthy, well formed udders are genetically selected for, and cows are not inbred.  Genetics in the dairy industry is quite advanced.  A healthy, happy cow is also good for the farmer.  Sometimes a cow will have an extra teat, but it's usually not functional and not really an issue at all.  The other thing to consider is spatial distribution, there are no more than six in any given region, so if you wish to have more than six, you have to spread them out along the AP axis. However, I can't find any information as to why some species have only two at the anterior and others only two at the posterior, beyond mechanistic explanations. This is mere speculation, but it could be for reasons of access. The image of the goat shows how the position to the posterior allows the kid to suckle with ease. Obviously it makes sense for human and primates to have anterior glands, since we are bipedal and hold our young. Perhaps the posture of elephants make it easy for the young to suckle under their forelegs. Of course it could also simply be one of those accidents of evolution. I got most of this from Animal Science and Industry, Merle Cunningham, ISBN 9780130462565.  Rockpock  e  t  19:33, 26 May 2007 (UTC)


 * Well, at any rate, I think that Rockpocket  deserves the Wikipedia Reference Desk Above-And-Beyond-The-Call-Of-Duty award for conspicuously thoroughly researched and presented data! —Steve Summit (talk) 20:18, 26 May 2007 (UTC)


 * The cat/dog/rat totals don't add up correctly. As for being an accident in evolution.... I'm not so convinced myself. Two things come to mind when talking about elephants. One is their size, obviously, and the second is the amount they travel in any given day. Maybe the location of the glands has to do with not staying still? --Wirbelwind ヴィルヴェルヴィント (talk) 20:59, 26 May 2007 (UTC)


 * Oops. Thats embarrassing - I have fixed them now, thanks. My guess would be you are probably right. The image seems to demonstrate that the position of the elephant gland allows the calf to suckle from a position at the side of its mother. This would be most helpful to suckle on the move, avoiding the risk of being trampled on. The human and primate anterior positioning permits a similar thing. In contrast, grazing animals will stand still most of the time the young suckle. The size of the animal is also an issue, and whether the mother lies down to suckle her young. There are also issues about directing the young to the nipple. Some animals use pheromone type cues, meaning the young have to use their own olfactory function to feed. In other species, the mother plays a much more active role, physically guiding the pup to the nipple. All of these factors would impact on the evolutionary pressures driving mammary gland positioning. BTW, Steve, I have recently acquired a professional interest in another aspect of suckling, so this research was not entirely altrustic. Rockpock  e  t  21:18, 26 May 2007 (UTC)

Bird identification
I'm in the Pacific Northwest, and I saw a bird that I thought was a juvenile crow, but although it looked black from a distance, its feathers (on its body anyway) were really dark variegated metallic tones. It looked pretty much like a pigeon-sized crow, but its head (or maybe its crown feathers) seemed smaller. Anchoress 03:33, 26 May 2007 (UTC)


 * maybe some kind of starling. -- Diletante 14:59, 26 May 2007 (UTC)
 * Thanks for the reply, but I kinda doubt it. This bird looked amost exactly like a crow, except it had a slightly sleeker head and up close its feathers were a really dark - almost black - metallic pattern. And it was much smaller, of course. But still pretty big for a bird. The pic provided was a completely different shape. Anchoress 18:44, 26 May 2007 (UTC)


 * Could be a Jackdaw? They're sometimes found in North America, although it's on the wrong coast if it is (!). More likely, it could be a Common Grackle, Rusty Blackbird, Brewer's Blackbird or Brown-headed Cowbird, all of which are native to or visitors of your area. Any good? --YFB ¿  19:25, 26 May 2007 (UTC)
 * Wow, thanks for those links! The Grackle was the closest, although the distinguishing feature of the bird I saw was the almost invisible metallic markings over its whole body, but the shape of the Grackle is definitely very close (the bird I saw had less of a ruff tho). Thanks again! Anchoress 00:53, 27 May 2007 (UTC)


 * No problem :-) I think, based on your location, that it's likely to be the Bronzed subspecies (Quiscalus quiscula versicolor) which looks like the one at the top of this page. The amount of colouration you see will depend heavily on the lighting conditions and the state of the bird's feathers, but it's hard to say with certainty without a photo. There don't seem to be any other species in my National Geographic Field Guide to the Birds of North America which match your description, so I'm reasonably sure it's one of the species I linked. Try to get a snap if you see it again! --YFB ¿  02:07, 27 May 2007 (UTC)

Exor gate
Any idea what is the table for 3 bit EXor gate?


 * It's not well-defined. I tend to imagine a parity generator.
 * It's very well defined and it is a parity generator. It's also the same as 2 two input xor gates.  An even number of 1's generates a 0 and an odd number of 1's generate a 1.  This is true for any xor gate of any size input.  --Tbeatty 06:19, 26 May 2007 (UTC)


 * That's well-defined as a relation which (a) is useful and (b) the two-input XOR is a natural limiting case of, but it's not clear that the 3- or more input forms are meaningfully "A or B but not both", which is what I think of as the definition of "exclusive OR". —Steve Summit (talk) 12:15, 26 May 2007 (UTC)


 * I didn't get that with the "2 two input xor gates", but I suspect this is what you meant as the answer to the question:


 * Is that correct? —Bromskloss 09:55, 26 May 2007 (UTC)


 * Two XOR gates where A and B feed the first gate - then C and the output of the first gate feed the second gate - would form a function which is of some use. But XOR is defined as a two-valued function and the term simply isn't used by circuit designers or programmers for devices or functions with more than two inputs.  The definition can be expanded in more than one way - you could have a function like a 'Parity generator' which produces a '1' if there are an odd number of '1' bits on the input - or you could make a function 'true if exactly one input is true' - or a function 'true if all but one input is true' - any of those things would behave like XOR for two inputs - which one should be the definition of XOR for more than two inputs?  I'm a programmer - and I used to be a circuit designer - and I've never heard of anyone using the term XOR for more than three two inputs.

Hence any truth table you might come up with for such a device would be entirely arbitary and ought to have a different name that more clearly explains what the function is - there is no one clear generalisation of XOR that everyone would agree on. SteveBaker 14:54, 26 May 2007 (UTC)


 * Steve: An engineer might agree with you but I'd imagin that a mathematician might find the parity the only consequent choice. For AND and OR gates, we generalization is obvious, and this is mathematically because they are associative and commutative operators. This allows you to arrange all the operands (ingoing signals) in an arbitrary order and always pair off neighboring ones by ANDing them and then replacing the pair by its result. In other words: If we have only binary AND gates with fan-out 1, and we arrange them in an arbitrary way (barring feed-back, of course) to get a single output, we always end up with the same man-inputs AND. The same holds for OR, and if you try it for XOR, you are bound to end up with parity. Simon A. 17:38, 26 May 2007 (UTC)


 * I do circuit design today and multiple input XOR gates are used all the time. It's provable that (A XOR B XOR C) => (A XOR B) XOR C => A XOR (B XOR C) => (A XOR C) XOR B.  There is no ambiguity at all in logic.  You can use the truth table above (which is correct).  Since the two input XOR seems intuitive, start there.  A XOR B = (A AND B*) OR (A* AND B). A XOR C = (A AND C*) OR (A* AND C)).  B XOR C = (B AND C*) OR (B* AND C).  Once you drop those identities into all the equations, you will find there is no alternative definition.  A XOR B XOR C is very well defined and suffers no logical ambiguity.  --Tbeatty 05:34, 27 May 2007 (UTC)


 * Or in short, "XOR is also associative and commutative, and therefore the gneralization to three operands is just as obvious as with AND and OR." I'm surprised that Steve Baker didn't see it that way.  Oh, but now I notice Steve Summit's point: if you think of the English language meaning of "or", you get a different generalization.  I think my comment as to that is, "who cares about English?" --Anonymous, May 27, 2007, 05:38, edited 06:02 (UTC) after counting Steves in the thread.


 * It's clear that the generalized, three-or-more-input operator is useful. My only quibble (and it is indeed a minor one) is that I think the useful name for the generalized operator is something like "parity generator".  But if you want to call it "XOR", or "Wakalixes", I can't stop you. :-) —Steve Summit (talk) 06:08, 27 May 2007 (UTC)


 * I agree with Steve's first sentence above, but I think that WP should set an example of the correct use of "XOR". We have to differentiate between what is verifiable and what we assume to be common sense.  Here are two references:
 * "exclusive OR n. a logical operation working on two variables..." (The Concise Oxford English Dictionary and The New Oxford American Dictionary)
 * and:
 * "exclusive disjunction. If p and q are statements, then..." (The Concise Oxford Dictionary of Mathematics)
 * In other words, XOR is defined only for two inputs. There is no formal justification for the term "3-input XOR gate", even though everyone thinks they know what it is. --Heron 12:01, 27 May 2007 (UTC)


 * How about the logic that Xor gives 0 for same value( 0 or 1) of n input( ex:-0000 or 1111) and 1 for different value( 0 & 1) of n input(ex:-0101 or 1010)??59.92.240.228


 * That's not an interpretation I've ever heard. Do you have a reference for that?  It does, however, raise the issue of the XNOR gate, which is straightforward if you define it (correctly) as NOT (A XOR B), but confusing if you define it as an 'equivalence gate', which implies a different truth table from the correct definition for > 2 inputs. --Heron 22:18, 27 May 2007 (UTC)

Catalysis and equilibrium
An interesting point has come up in relation to chemical equilibrium. How does a catalyst reduce the time needed for a reaction mixture to attain equilibrium, reactants $$\rightleftharpoons$$ products? I think that it speeds up either the forward reaction or the backward reaction, but not both as seems to be suggested in some places. Do anyone know of a good primary source that can be cited to support this idea?

My understanding is that a catalyst reduces the activation energy for a specific reaction by providing an alternative mechanism by which that reaction can proceed. The activation energies for the forward and backward reactions are not related to each other because the mechanisms are different, so I see no reason why the catalyst should affect both reaction rates. Is this right in general? Petergans 17:57, 26 May 2007 (UTC)


 * Nope. A catalyst does indeed speed both the forward and the reverse reaction&mdash;remember, the alternative reaction mechanism that the catalyst provides will work in both the forward and reverse directions.


 * See also Catalysis. TenOfAllTrades(talk) 18:24, 26 May 2007 (UTC)


 * Looking at it mathematically, consider the Arrhenius equation. For the forward reaction, we have $$k_f = A_f e^{{-E_{af}}/{RT}}$$, and for the reverse, $$k_b = A_b e^{{-E_{ab}}/{RT}}$$. At equilibrium, the forward rate of reaction equals the reverse rate. So, we'd have $$k_f \cdot [A][B] = k_b \cdot [C][D]$$, where A, B, C, and D are reactants and products. The position of equilibrium, that is, $$ \frac {[C][D]}{[A][B]} = \frac {k_f}{k_b} $$, equals $$ \frac {k_f}{k_b} = \frac {A_f e^{{-E_{af}}/{RT}}}{A_b e^{{-E_{ab}}/{RT}}} = \frac {A_f}{A_b} e^{{-(E_{af}-E_{ab})}/{RT}} $$. Everything in the last expression is a constant (at constant temperature) even when a catalyst is added, since it reduces activation energy of both the forward and reverse processes equally, so $$ (E_{af}-E_{ab}) = const$$. Borbrav 20:28, 26 May 2007 (UTC)

There is nothing wrong with this argument for reactions for which it can be assumed that $$k_f \cdot [A][B] = k_b \cdot [C][D]$$ or that similar expressions will apply. However, the law of mass action is not universally valid because rate expressions do not universally follow stoichiometry. Secondly, for a multi-step reaction there may be more than one maximum in the energy vs. reaction coordinate plot. Will the catalyst affect both of them? Thirdly, in a multi-step reaction the rate-determining step may be different in forward and backward reactions. Will the catalyst affect both of them? Fourthly, an enzyme may initiate a chain of reactions of which it can possibly reverse only the first one but not the rest.

The key word in my question is general. Petergans 22:52, 26 May 2007 (UTC)

About Contacts
1) Why is it that after wearing my contacts for about the whole day, around the pupil of my eye (supposed to be gray due to the contact) gets brown (my natural color)?

2) What's protein buildup?

3) Can someone link me to a site showing all the decoration contacts available?

I would look myself, but I need to leave somewhere (I'm in a hurry). I didn't want to forget to ask. PitchBlack 20:43, 26 May 2007 (UTC)


 * How about you look things up yourself when you have time, then come back and ask about what you can't figure out yourself? --Tugbug 23:18, 26 May 2007 (UTC)


 * Yeah, youre partially right. Can someone still help me on 1 and 2? 74.161.71.66 00:48, 27 May 2007 (UTC)

More nipples
Why don't birds have nipples? --84.67.135.166 22:48, 26 May 2007 (UTC)


 * Because they never evolved them. They're not mammals. &mdash; Kieff | Talk 22:57, 26 May 2007 (UTC)

What's with the evolution? Birds are fed by the yolk in their eggs. When they are born, they still have some of that stuff in their tummies, so there's no need. .

Plus, while mammal babies need milk because they don't have good teeth yet, chickees already have a gizzard with teeth to take care of seeds and stuff. If they don't have teeth, they swallow stones to act as teeth. Besides many birds can care for themselves right after they're born, according to our article, bird. --JDitto 01:32, 27 May 2007 (UTC)

Birds developed neither nipples nor breastfeeding. Indeed, as said above, they evolved different strategies of caring for their young. However, some birds did evolve something remotely resembling breast milk. Please see our Crop milk article. Cheers, Dr_Dima.

H3
I was reading about the Lunar Regolith and read about the massive amount of H3 that is in it. I also was reading the Fusion article and it talked about H3 + H3 reactions which would produce no radiation. If there is plenty of H3 on the Moon and we get there safe and sound how much energy does it take to fuse together H3?67.125.159.224 23:39, 26 May 2007 (UTC)


 * I think the idea of "plenty" that you have right now is a lot different from what the article means. Like there are "plenty" of diamonds in namibia, but that does not really mean namibians can burn diamonds to heat their houses. Tritium is also highly radioactive all by itself, fusion or not.