Wikipedia:Reference desk/Archives/Science/2009 September 23

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.

September 23

British Weights and Measures Act

I am trying to locate the text of a certain British Weights and Measures Act (I think that's what it was called). Here are some things I think I remember from it:

• The document was from sometime in the 19th century.
• The document defined the yard as the length of a certain metal rod, and other measures of length in terms of the yard. As a sort of backup should this metal rod be lost or destroyed, the text also gave the length of a seconds pendulum in inches (as "so many Inches, and so-and-so many ten thousandths of an Inch", rather than using Arabic numerals like a normal person would do).
• The document defined a gallon in terms of the volume of so many (ten?) pounds of water.
• There was another backup, so that should the pound standard be destroyed, the weight of so many cubic inches of water would then be taken as the pound (or some similar rule of that sort).

There is a list of such acts at Weights and Measures Act, but there were eight of them in the C19. This EB article may be the sort of thing you were after - it would point to the 1824 act. More EB here. Good luck. --Tagishsimon (talk) 01:15, 23 September 2009 (UTC)

Possible to sustain adult human entirely intravenously?

As far as I know, people who are in a vegetative state are given nutrients via tubes to the stomach and I guess occasionally IV. Would it be possible to sustain an adult human entirely via IV? Essentially bypassing the entire digestive system? The kidneys & bladder would still be operational, but the esophagus, stomach, and intestines would all be unused.

It certainly wouldn't make for an active lifestyle! But I'm just interested in the feasibility of providing the necessary vitamins, amino acids, sugars, etc. over the long-term.

Can we do that yet? 218.25.32.210 (talk) 02:23, 23 September 2009 (UTC)

Yes. Total parenteral nutrition. - Nunh-huh 02:26, 23 September 2009 (UTC)

Yup. Not much of a life though. By the way, that stuff's insanely expensive. Shadowjams (talk) 07:21, 23 September 2009 (UTC)

I don't know how you all come to the conclusion that the intestines would be unused. Bilirubin, for example, from decay of erythrocytes, is the chemical that is dumped in the highest amount into the intestines, through the gallbladder. It's what makes the color of shit. So, eventually, your intestines have to move from all the stuff that comes through the gallbladder. --Ayacop (talk) 09:19, 23 September 2009 (UTC)

Hernia support

Resolved

What type of doctor(s) would diagnose and treat an abdominal hernia ? (I truss that you will answer promptly.) StuRat (talk) 03:26, 23 September 2009 (UTC)

The Internist or General Practitioner is a good starting point. A general surgeon (or pediatric surgeon if the patient is young) could do the repair. Hernia good jokes lately? Edison (talk) 04:59, 23 September 2009 (UTC)

It's best to be serious, and not engage in disrupturing. →Baseball Bugs ^{What's up, Doc?} carrots 09:23, 23 September 2009 (UTC)

Any physician can diagnose an abdominal hernia. They can then refer you to a general surgeon for treatment since abdominal hernias rarely heal on their own. Livewireo (talk) 15:07, 23 September 2009 (UTC)

OK, thanks for the answers ! StuRat (talk) 14:29, 27 September 2009 (UTC)

most oftenly mutated genes?

Hello, what are the, say ten genes the mutations of which are the most common ones? Note, I'm not asking for the most common single mutations but cumulatively the genes. That is, mutation r123 on gene X may be the most common mutation around but mutations on gene Y may be more common because there are so many different ones. ----Ayacop (talk) 08:34, 23 September 2009 (UTC)

It sounds like you are asking for the gene with the most diverse alleles. In that case, V(D)J recombination is worth noting although it may not result in the most diverse gene.152.16.15.144 (talk) 16:58, 23 September 2009 (UTC)

You need to specify whether you're asking about germ line mutations or somatic mutations. I'm pretty sure that the most frequent class of somatic mutations that occur in (higher?) vertebrates, is the affinity maturation that occurs in B lymphocytes. If you're asking about germ line mutations, I don't know the answer, but there are so called hot spots, as described in our article, where the mutation rate is up to 100 times higher than at other sites in the genome. The major histocompatibility complex has such hot spots. Whether VDJ recombination is within the scope of your question is a matter of definition (I'd say it isn't), but if it were, the variability introduced in our antigen receptors by this process is so huge — a limited number of genes produces a vast number of receptors; 10^{(pick a number, 14 perhaps)} — that it would would dwarf anything else. VDJ recombination is, AFAIK, the only physiological somatic DNA recombination that occurs in vertebrate cells (or, for that matter, eucaryote cells). If anyone knows about other examples, I'd love to hear about it. --NorwegianBlue ^talk 18:24, 23 September 2009 (UTC)

I'm not really sure that this question can be answered but I'll echo the other two answers above and suggest that the OP clarify the question. For example: "which 10 genes have the highest number of different alleles?" or "what are the 10 most frequently mutated genes in cancer cells?" or "what are the 10 genes that are most frequently involved in a new (de novo) disease-causing mutation?" or "which single nucleotide polymorphisms have the highest population prevalence?" or "what are the 10 most common inherited Mendelian disorders?" --- Medical geneticist (talk) 21:14, 23 September 2009 (UTC)

Thanks for steering me to the right words. I would like to see a list that answers the question

• what are the genes that are most frequently involved in a new (de novo) disease-causing mutation? (say, in Europeans)

OTOH, lists that answer the other questions User:Medical geneticist gave would be interesting as well. --Ayacop (talk) 06:54, 24 September 2009 (UTC)

Since I steered you to the question, I guess I'm obligated to try a response...  ;-) I still don't think I can provide a complete answer, but in general the answer is going to be dictated by two variables: (1) the incidence or prevalence of the disease associated with mutations of the gene in question, and (2) the proportion of cases of that disease that are caused by new mutations (as opposed to inherited from an affected or carrier parent).

• Most autosomal recessive genetic disorders are caused by ancient mutations that are present in the general population at a certain carrier frequency and are not typically caused by new mutations, so we can eliminate that category.
• Autosomal dominant genetic disorders are often later onset conditions that don't necessarily impair reproduction and therefore can be inherited, although there is quite a range in terms of severity and reproductive fitness (for example, Huntington's disease, achondroplasia, some forms of osteogenesis imperfecta, marfan syndrome, polycystic kidney disease). Each of these conditions will therefore have a different new mutation rate. Most cases of achondroplasia are due to new mutations on the paternal copy of the gene and new mutations are not uncommon in marfan syndrome and neurofibromatosis, but new mutations are uncommon (by definition) in familial hypercholesterolemia.
• New mutations make up a significant fraction of isolated cases of X-linked disorders (for example, Duchenne muscular dystrophy).

Ok, so in order to answer the question what you really want is a list of genetic disorders by frequency. I found this one but I can't vouch for the list being comprehensive or the numbers being correct (and there are probably differences among various world populations). Then you want a list of genetic disorders by new mutation rate, which I'm having a hard time finding with Google (it could be that such a list just doesn't exist due to limited knowledge-base). You would then be able to intersect the lists to find the disorders having a combination of relatively high incidence and relatively high new mutation rate. I'll do some OR and see if I can come up with some candidates... --- Medical geneticist (talk) 14:22, 24 September 2009 (UTC)

This database looks like a promising source for data on prevalence of individual disorders. I don't think it can answer the question about new mutation rate, though. --- Medical geneticist (talk) 15:19, 24 September 2009 (UTC)

A pity that mining the literature is still so difficult. Many thanks for your efforts. --Ayacop (talk) 17:06, 24 September 2009 (UTC)

Brain cysts from cat parasites in humans

I remember hearing that these are common in humans, and that they change peoples behaviour to make then more relaxed and more outgoing. Just how true is that? 89.242.104.32 (talk) 09:59, 23 September 2009 (UTC)

Are you talking about toxoplasmosis? --Mark PEA (talk) 13:04, 23 September 2009 (UTC)

I hope you're not hoping to give out cats to all of your uptight and introvert friends -- sort of conjured up the tapeworm diet. DRosenbach ^{(Talk | Contribs)} 14:37, 23 September 2009 (UTC)

Toxoplasmosis may also increase traffic accidents,[1] and Rhesus factor may be protective against this.[2] Fences&Windows 19:52, 23 September 2009 (UTC)

'99% accurate'

I've often seen various positive/negative tests advertised, usually in the media, as being '99% accurate' and suchlike. What is usually meant by this in terms of false positives/false negatives? 94.168.184.16 (talk) 12:54, 23 September 2009 (UTC)

Check out sensitivity and specificity. Presumably giving a single number as in your question depends on the context. For example, let's say you have a cancer test that achieves 99% sensitivity, but only 80% specificity. They will probably quote the 99% figure, since the consequences for a false negative are much more serious than for a false positive: a false negative could cause a patient to dangerously delay treatment, while a false positive will probably just cause some unnecessary followup testing. An employment drug screening test might place the emphasis on specificity, since it might be better to let a few stoners in than to risk a lawsuit over a false positive. --Sean 13:39, 23 September 2009 (UTC)

I think a false positive is pretty serious for a cancer test too - the increased stress on someone who thinks they may die anytime soon is not insignificant - especially if it's multiplied by 20% of the people who are tested. This is why some tests which sound (on the face of it) to be amazingly good are not carried out in practice. SteveBaker (talk) 15:41, 23 September 2009 (UTC)

Tests with so many false positives are usually used for initial screening and there would be a follow up test (which is more expensive or unpleasant or time consuming or whatever than the initial one) to dramatically reduce the number of false positives. This is true, for example, for breast cancer - a large majority of positive results on a standard mammary scan are false, and the people doing the scans are careful to make it clear that they are just testing to see if you need the more thorough test, rather than to find out if you actually have breast cancer. --Tango (talk) 16:19, 23 September 2009 (UTC)

In public advertisements for self-tests, it is just as much a legal issue. For example, a pregnancy test if handled and used correctly is 100% accurate. However, there could be some damage to the test in shipping. The test item could have a shelf-life that expired before it was used. The user could be an idiot and assume that the male is supposed to pee on the stick instead of the female. Further, the user could be dumbfounded by the complexity of what + and - means on the result. So, with all those uncontrollable factors in the test, it cannot claim 100% accuracy. -- kainaw™ 13:46, 23 September 2009 (UTC)

And while we're at it... it's worth noting that 99% is pretty good for some applications, pretty bad for others. 1% multiplied by a lot of people is still a lot, and depending on the consequences, can be worse than not testing at all. --98.217.14.211 (talk) 15:29, 23 September 2009 (UTC)

Indeed. Consider a hypothetical 'terrorist sensor' deployed to airports. Let's assume that it can read minds and judge, with 99% accuracy, which passengers are planning to hijack an aircraft and fly it into a building. Is that good enough? On the one hand, if you have a hundred terrorists, you only lose one building. That sounds pretty good. On the other hand, you have roughly two billion air passengers worldwide, each year — so with 99% accuracy, you will incorrectly detain and torture forty million people. Hm. And those numbers assume that the device performs as well in the field as the manufacturer claims; reality is usually much worse than sales brochures. Meanwhile, the cost of installing those devices in every airport in the world gets deducted from budgets which could have been used to do old-fashioned police work, or deliver aid to war-torn countries that breed terrorists, or just to build schools and hospitals and provide other societal benefits. TenOfAllTrades(talk) 17:09, 23 September 2009 (UTC)

You are blurring two types of error. If a terrorist sensor failed to pick up one out of every 100 terrorists, it would be labeled 99% effective. That does not imply that it will consider one out of every 100 non-terrorists a terrorist. It could never consider a non-terrorist a terrorist. It may consider 50 out of 100 non-terrorists a terrorist. The 99% isn't measuring that in this case. On the other hand, if a terrorist sensor were to consider one out of every 100 non-terrorists a terrorist, calling it 99% effective would be a lie. The effectiveness is based on the purpose. The purpose is to identify terrorists and the measure of 1/100 non-terrorists doesn't imply anything about the effectiveness of identifying terrorists. I feel it is important to clarify this because this sort of incorrect argument is commonly used to sway opinion against things by creating a false fear. Terrorism is one of those areas where people attempt to create false fear in order to argue against anything and everything. -- kainaw™ 17:20, 23 September 2009 (UTC)

Yes, I realize that the positive and negative error rates are distinct, and often confused or conflated. I should have been more specific in my definitions. My point was to amplify on the grandparent poster's comment — that is, a superficially low false-positive rate can give a surprisingly high error rate if one is screening for low-probability events. On the point of false fears and terrorism, in general my experience is that sloppy fear-driven arguments are usually made to encourage (not to discourage) doing anything and everything. (Cover-your-ass and 'gotcha' approaches to politics and government bureaucracy play a role here.) The No Fly List in the United States is a perfect example. As far as I know, there's still been no evidence that it has ever prevented a terrorist attack, yet it inconveniences thousands of people and has reached a cost of roughly one billion dollars. And the No Fly List is screening on something that ought to be quite specific (though not necessarily unique): people's names. TenOfAllTrades(talk) 17:37, 23 September 2009 (UTC)

Well literally speaking it should mean that 99% of results are correct, so that the total of false positives plus false negatives should be only 1% of all tests. To see how this might go in practice, a test for a rare disease might be expected to be positive one in one hundred times it is given. In that case, consider applying the test to 200 people if it has a false negative rate of 0.5% plus a false positive rate of 50%. Then 197 people would test negative (with one false negative) and 3 people would test positive (with one false positive), for a combined error rate of 1%. As the example illustrates a test that gives the right result 99% of the time could actually be very inaccurate in the direction of results that rarely occur. (It should mentioned that sometimes accuracy statements are referring to the accuracy in both directions, i.e. both positive and negative, in which case the situation illustrated by this hypothetical would not apply.) Dragons flight (talk) 16:25, 23 September 2009 (UTC)

There's an interesting side effect to calculating the success rate that way. Suppose only 1% of people have our dread disease. If the black box always lights up and says the patient doesn't have the disease, then its overall results are still 99% accurate: 100% correct on the uninfected 99% of the population, and 0% accurate for the 1% who are infected. Such a magical "everything is okay" box would have an absolutely brilliant success rate (99.9999%+) detecting terrorists at the airport, too. TenOfAllTrades(talk) 17:44, 23 September 2009 (UTC)

Hmmm... I smell a multi-million dollar government contract there. Quick! To the patent office! — Lomn 18:23, 23 September 2009 (UTC)

Speaky of false predicting, apparently James Randi used to write a note every morning and put it into his pocket. It said: "I, James Randi, will die today!" and was signed with that day's date. If he died that day, people would wonder how he predicted it.[3] Fences&Windows 19:58, 23 September 2009 (UTC)

Psst! Read Bayesian statistics. Fences&Windows 19:58, 23 September 2009 (UTC)

Genetics

How can I prove the law of segregation for waxy gene in maize?

I think it was proven when meiosis was discovered. DRosenbach ^{(Talk | Contribs)} 14:08, 23 September 2009 (UTC)

I'm not sure I completely understand the OP's question, but the article waxy corn would be a good place to start, and it gives some references about the genetics of the waxy gene in question. Alternatively, is the OP asking for more practical advice on how to set up an experiment that would demonstrate the Mendelian inheritance of the waxy gene? --- Medical geneticist (talk) 20:51, 23 September 2009 (UTC)

Bird with no legs ?

Resolved

In the movie The Fugitive Kind, Marlon Brando describes "a bird with no legs, with a blue belly so it's hidden from below, which spends it's entire life flying". Is there any truth to this at all ? I know birds can spend days flying, and could eat insects in the air, and maybe drink water when it rains, and some can even mate in the air and sleep in the air. I can't see how they could lay eggs and raise chicks there, though. StuRat (talk) 15:29, 23 September 2009 (UTC)

Fiction != Reality. SteveBaker (talk) 15:35, 23 September 2009 (UTC)

That's not a very useful comment. There are many works of fiction based at least partially of reality. That's why I'm asking if any part of this is true or not. StuRat (talk) 15:43, 24 September 2009 (UTC)

It's the Huma bird. A myth, an impossible animal, though its existence was taken as a possible reality for a few hundred years (like the Sycthian Lamb). As for giving birth in the air, I imagine the babies would have to be pretty well-grown by the time they hatched! Egg comes out, hatches in mid-air, bird instinctively flies out! A silly idea, but a fun one. --98.217.14.211 (talk) 15:41, 23 September 2009 (UTC)

A real-life bird which was at one time believed to have no legs was the bird of paradise. This was not believed in places where they actually lived, though; see the lead section of the article. --Anonymous, 20:33 UTC, September 23, 2009.

There are birds with no legs, though, due to injury or disease. They can sometimes get around quite successfully on the ground by sliding or hopping around on the bellies propelled by their wings. My grandparents have fed the birds in their garden for decades and they have blackbirds with a congenital defect that makes them kind of club footed and their feet often fall off later in life. They survive just fine, although that gene would probably have been removed from the gene pool had by grandparents not made it so easy for them to get food. --Tango (talk) 16:23, 23 September 2009 (UTC)

- In order to achieve flight, birds run-and-jump or hop-and-jump. Without legs, they couldn't take off.

- A white underbelly would grant greater camouflage than a blue one.

- If the bird never lands, does it lay scrambled eggs?

B00P (talk) 17:06, 23 September 2009 (UTC)

If it never lands, it never needs to take off! QI says pink is actually the best disguise, that is apparently what they used to paint some military planes. --Tango (talk) 18:00, 23 September 2009 (UTC)

While I agree that the bird-with-no-legs-giving-birth-in-the-air thing is purely mythical, a run- or hop-and-jump isn't necessary under all conditions. If the bird made a habit of only landing in breezy areas, it could simply face into the wind and start flapping — the wind could in principle give it an airspeed close to, or even in excess of, its stall speed. The very clever bird would only ever land at the top of trees (difficult to do without legs, I grant you) or near the edge of sheer cliffs. A brisk shove sidways with the wing and he's falling...then flying. TenOfAllTrades(talk) 18:48, 23 September 2009 (UTC)

I've seen gulls with no legs taking off successfully. It looks really clumsy - but they can do it. --Kurt Shaped Box (talk) 20:04, 23 September 2009 (UTC)

The common swift comes pretty close. 95.112.148.48 (talk) 19:00, 23 September 2009 (UTC)

Swallows feed on the wing (on airborne insects), drink by dipping their beaks while skimming low over ponds, and may remain airborne for extended periods. In Mediaeval Europe there were various widespread erroneous beliefs about swallows, including one that they never landed/perched and had no legs (for why would God have given them useless appendages?), hence in heraldic art the swallow (under the name hirondelle) was/is traditionally depicted without legs (and also, for some reason, without a beak). 87.81.230.195 (talk) 19:59, 23 September 2009 (UTC)

There is some Monty Python joke to be made here, but I can't figure it out. John Riemann Soong (talk) 18:06, 24 September 2009 (UTC)

Does 'fully laden' in this context sound less silly to a Commonwealth person? Sometimes with Monty Python sketches one just doesn't know which anachronistic language sillying is standard unremarkable British communication and which is not. Sagittarian Milky Way (talk) 20:59, 24 September 2009 (UTC)

"Fully laden" sounds perfectly cromulent to this middle-aged Brit, but it tends to be used in formal or industrial contexts, such as in official notices or data panels referring to the weights of railway wagons and the like, rather than in casual speech. 87.81.230.195 (talk) 02:10, 25 September 2009 (UTC)

OK, thanks all. StuRat (talk) 14:27, 27 September 2009 (UTC)

NIGHT VISION GOOGLES

If a man standing in front of a mirror is a pitch black room, is wearing a night vision goggles. Can he see his image on the mirror? Explain. —Preceding unsigned comment added by Great gamer (talk • contribs)

What kind of night vision goggles? Are they infrared goggles or low-light goggles? That makes a huge difference. -- kainaw™ 17:32, 23 September 2009 (UTC)

Modern night vision goggles use an image intensifier to amplify the effect in ambient low light. This requires a minimum level of light. In pitch darkness, there is no light to amplify, so no image is seen. Older generations of devices use active infra-red, which would be visible, partly because the device itself has its own IR source, and partly because the man himself projects IR body heat. Axl ¤ [Talk] 17:40, 23 September 2009 (UTC)

Modern night vision goggles need SOME light - not much, but there has to be some. Most of them are sensitive to near-IR and red light, they don't really see blue light or heat very well - although people sometimes confuse NVG's with FLIR (Forward-looking-infrared) cameras which most definitely CAN see heat, but are unsuitable to wear as "goggles". So, in this case, if there is some tiny amount of light in the room, the man will be able to see his reflection in the mirror. If the room is utterly black - then the goggles will have boosted their amplification levels to the maximum possible and all you'll see is "snow" generated by electrical noise from the electronics (like you get when you disconnect the video from your TV). However, most people underestimate the amount of light amplification these gadgets can generate. In practice, it's almost impossible to get things so dark that you can't see anything through the goggles. Light leaking from the NVG's green display around the edges of the goggles provides enough light to see by. When we had a pair of US military goggles on loan, we tried to get a really, REALLY dark room to observe the 'noise' phenomenon first-hand, we found that light leaking under the wood/sheet-rock walls, shining up through the edge of the carpet from the adjacent corridor was quite enough to see by! An LED on a hard disk drive INSIDE a computer in the room cast enough light that even after bouncing off internal structures and out of some ventilation slots, the light was so bright that you could use it to make shadow puppets on the wall opposite. SteveBaker (talk) 23:13, 23 September 2009 (UTC)

Surely, there could be something in that room you could cover them with that would work? Have a guy lie over the doorjamb. Sagittarian Milky Way (talk) 04:49, 24 September 2009 (UTC)

The doorjamb was something we'd figured out in advance - we turned off all of the lights and other equipment in the adjacent room. But light leaking under the walls was quite unexpected. Certainly, you can force the goggles into complete darkness by putting the lens-caps over them - but we wanted to see how things looked just before things got too dark to see. (At the time, I was working on an accurate computer graphics simulation of night vision goggles and infra-red camaras so we did a lot of messing around with NVG's. Driving at night with your headlights turned off and wearing NVG's is a dangerous and exciting thing!) SteveBaker (talk) 12:34, 24 September 2009 (UTC)

Ha ha. Not in the city it isn't. You could see fingerprints by cloudlight alone. (or maybe there was snow on the ground, don't remember) What kind of floor-wall joint was that? That's very unusual. Sagittarian Milky Way (talk) 16:13, 24 September 2009 (UTC)

As I recall, it was a concrete floor with wooden 'stud' walls sheeted with sheet-rock (aka plaster-board) with skirting boards on both sides. Both the room - and the corridor outside were carpeted. I happened to be the one wearing the goggles at the time - and as we knocked out more and more sources of light by turning off computers, disconnecting the phone, etc, this weird sparkling glow appeared around the bottoms of three walls of the room (it was an interior computer room with no windows and only one door) - the room on the other side of the fourth wall had the lights turned off to prevent light from leaking around the door frame. For a while, I couldn't figure out what the glow was - until someone walked along the corridor outside and their shadow dimmed down the glow as they walked along. We eventually pulled up the carpet and ran a strip of duct-tape between the skirting board and the concrete floor - which got the room pretty amazingly dark. But light leaking around the edges of the goggles themselves was hard to eliminate. Even if you pushed the rubber eye-pieces hard up against your face, the light refracted through your skin and appeared as a dim glow around your eyes to anyone else wearing goggles. Getting rid of every last photon is very hard! SteveBaker (talk) 17:48, 24 September 2009 (UTC)

"What do you see when you turn out the light I can't tell you, but I know it's mine" ~~ from With a Little Help from My Friends Cuddlyable3 (talk) 17:00, 24 September 2009 (UTC)

Time Dialation

When Einstein did his thought experiment showing time dialation (with the photon bouncing between two mirrors), why can we assume that the length between the two mirrors doesn't contract?

{{homework}} Fences&Windows 20:04, 23 September 2009 (UTC)

It's not. I had a homework problem along the lines of calculating the new angle a stick makes with the horizontal when it starts moving. The solution was pretty easy, but to show that the component of the length of the stick perpendicular to the velocity doesn't change I argued that a beam of light travelling the length of this component would take the same amount of time in both reference frames, because while an obvserver moving with the stick will have her/his clock run slower, the path that the light takes is longer by an identical factor. But the time dialation equation stems from Einstein's thought experiment mentioned above, which assumes that the length between two mirrors doesn't change. So the logic seems circular, hence the question.

I'll direct you to Special relativity, Introduction to special relativity, Time dilation#Simple inference of time dilation due to relative velocity and Inertial frames of reference#Special relativity while I wait for the physics cavalry to arrive. p.s. Sign your posts using ~~~~. Fences&Windows 23:34, 23 September 2009 (UTC)

The thought experiments are just meant to demonstrate problems with the classical theory that special relativity addresses. Time dilation isn't the unique way to resolve the specific problem brought up by this specific thought experiment (this is the one with the mirrors on the top and bottom of a moving train car, right?), but it is a solution that's part of a larger self-consistent framework that solves a lot of other problems as well. It's not quite correct to say that the equations were derived from certain thought experiments, although I think a lot times it's taught that way because it makes it easier to understand what some of the rationale is. Rckrone (talk) 02:38, 24 September 2009 (UTC)

Alright, so then why isn't there any length contraction? I see that the Lorentz transformation is y'=y, but where does that come from (it could be y'=ay).

Wait, are you referring to Time dilation#Simple inference of time dilation due to relative velocity? The length between the two mirrors doesn't contract because length contraction only occurs in the direction of motion. Or, put another way, contract with respect to what? Traditional length contraction works by comparing distances in a two frames of reference that are moving with respect to each other. Both the reference frame of the ground and of the train have no movement in the y direction relative to each other, so they will not see any differences in y lengths. — Knowledge Seeker দ 06:21, 24 September 2009 (UTC)

Well in physics it all comes down to whether or not we observe the effect. As far as we can tell, y'=y for inertial frames moving relative to each other in the x direction. You could probably come up with a mathematically consistent transformation that involved y'=ay, but we don't see it, so it likely wouldn't have any meaning in the physical world. Of course, hardly anyone thought that time dilation was real at first, since nobody had observed it yet, but careful measurements showed it was really there. So Einstein's assumption that we can ignore the possibility that the length contracted between the mirrors is ok because the result that follows is what we observe. Also, it is extremely likely that you can show that by starting with the postulates of relativity, and some other facts about physics and math, that y'=y and z'=z, but ultimately, we accept it as true because it fits our observations. Jkasd 07:12, 24 September 2009 (UTC)

There can’t be a contraction perpendicular to the direction of an object’s motion, given that the laws of physics are the same in all inertial frames of reference.

We’ll basically use a proof by contradiction. To contradict observed reality, make the hypothesis that moving objects shrink perpendicular to their direction of travel.

Now consider a bullet which has exactly the same diameter as the barrel of a particular gun. Then if the bullet travels at high speed down the barrel of the gun, the bullet’s diameter shrinks, and the bullet easily passes down through the barrel.

However, a frame of reference in which the gun is at rest isn’t the only possible inertial frame of reference. You could just as legitimately use a frame of reference in which the bullet is at rest, and the gun is moving. In that case, the hypothesis that moving objects shrink perpendicular to their direction of travel would require that the gun barrel would have to become narrower than the bullet, which means that either the bullet would jam in the barrel, or perhaps the barrel would explode.

Thus, the bullet must both pass easily through the gun’s barrel, and the bullet must get jammed in the barrel due to the barrel being too narrow, which is a contradiction. Therefore, one must reject the hypothesis that moving objects shrink perpendicular to their direction of travel. A similar argument can be used to show that a moving object doesn’t expand perpendicular to its direction of travel, either.

The argument above ultimately boils down to experimental evidence. The above argument assumes that the laws of physics are the same in all inertial frames of reference, but that’s only a valid assumption because it’s consistent with the experimental evidence. Red Act (talk) 14:31, 24 September 2009 (UTC)

Okay, but let's say we wanted to put a ruler into a desk (there is a high relative velocity between the two). Normally when both are at rest, the ruler and the desk drawer have the same length. But if they were moving relative with each other, then, from the reference frame of the desk, the ruler would contract in the direction of motion and fit easily...but from the ruler's frame of reference, the desk would contract, making the ruler impossible to fit into the drawer. So there seems to be a contradiction even if length only contracts in the direction of motion.

From the drawer's reference frame there is no problem getting the ruler in, just a rather violent crash afterward as the ruler cuts through its side. From the ruler's frame, the front side goes into the drawer, then it crashes through, and then the back end of the ruler enters the drawer. This comes out of the Lorentz transformations because time is transformed as well as the x coordinate. So there is no paradox, because things that happen at the same time in one reference frame don't necessarily happen at the same time in another reference frame. For more information, see Relativity of simultaneity. Jkasd 04:50, 25 September 2009 (UTC)

Also, proof by contradiction isn't much use in physics, except maybe to help understand some concepts. Actually, nothing is "proved" at all in physics; a theory just accumulates lots of evidence in favor of it, until it is widely accepted. Jkasd 05:07, 25 September 2009 (UTC)

You can't ultimately prove that a physical theory is correct, but you can prove that a physical theory is incorrect, if the theory isn't internally consistent. You can also prove that a physical theory must be true, given the assumption that some other set of physical theories are correct. Of course, the set of physical theories that you're taking as axiomatic in your proof might later be shown by experiment to be incorrect.

Yeah, a ruler going into a drawer at a relativistic speed isn't a good example, because all observers will simply agree that the ruler smashes into (and probably through) the back of the drawer. A better example that would illustrate the point you're trying to make would be something like this:

A super-fast jet is normally stored in a hangar that's exactly the same length as the jet. The hangar has doors on both its front and back. Now fly the jet through the hangar, flying in through the front door, and out the back door. From the hangar's frame of reference, the jet while flying is shorter than when it's at rest, so it's possible to close both hangar doors simultaneously while the jet is flying through the hangar. But from the jet's frame of reference, the hangar is shorter when the jet is flying, so it is impossible to close both hangar doors simultaneously while the jet is flying through the hangar. That's the kind of contradiction that I think you were trying to point out.

However, there's a way out of that apparent contradiction, that doesn't violate the assumption that the laws of physics are the same in all inertial frames of reference. Namely, one of the laws of physics can be that a clock attached to the front of a moving object will show a time that's later than a clock attached to the back of that object. It also happens to be true that both moving clocks appear to run slower than a non-moving clock, but that's not what we really care about here. All we care about here is that two different inertial frames of reference will disagree about whether two events are simultaneous or not, if the two events occur at different positions along the direction of relative motion.

Suppose the hangar doors are closed "simultaneously", according to clocks attached to the front and back doors, that have been "synchronized" according to the hangar's frame of reference. The experiment, as described from the perspective of the hangar's frame of reference, is that once the jet is entirely within the hangar, the two doors are closed simultaneously, and then the two doors are opened simultaneously before the jet leaves the hangar. An alternate description of the experiment, as described from the perspective of the jet's frame of reference, is that the hangar's back door closes and opens while just the front of the jet is in the hangar, and then later the hangar's front door closes and opens while just the back of the jet remains in the hangar. Note that although there are two different descriptions of the timing of events during the fly-through, both the hangar and the jet agree that the fly-through was successful. This is qualitatively different from the case with a bullet going through the barrel of a gun, in which case the contradiction requires the bullet traversal to be both successful and unsuccessful.

In a different experiment, supposed the hangar doors are pulled shut onto the jet "simultaneously", according to clocks attached to the jet, that have been "synchronized" according to the jet's frame of reference. For simplicity, assume that the hangar doors are rather flimsy, such that the jet can crash right through them without changing speed much. Otherwise, we'd have to deal with more than two different inertial frames of reference, which would make the problem much more complicated. In this experiment, both doors can crash into the jet. The process, as described from the perspective of the jet's frame of reference, is that at some time when the jet's front is sticking out the back of the hangar, and the jet's back is sticking out the front of the hangar, both doors are closed simultaneously, causing the jet to collide with both doors. Alternatively, as described from the perspective of the hangar's frame of reference, the process is that the front door is closed on the back of the jet, causing one collision, and then later the back door is closed on the front of the jet, causing another collision. Once again, there are two different descriptions of the timing of the events involved, but everybody agrees that both doors collided with the jet, and agrees with where on the jet the collisions took place.

Note that the two experiments are two distinct experiments, and all reference frames agree about which experiment is which. In the first experiment, all observers agree that the hangar doors are triggered by clocks that are attached to the hangar. In the second experiment, all observers agree that the doors are triggered by clocks that are attached to the jet. And the two experiments are not equivalent. If L is the length of the hangar or jet at rest, in the first experiment, the doors are controlled by clocks which are a distance L apart according to the inertial frame in which the clocks are syncronized (the hangar's). In the second experiment, the doors are controlled by clocks close to the middle of the jet, which are less than a distance L apart according to the inertial frame in which the clocks are synchronized (the jet's).

It doesn't work to try to use the same kind of agument as above to wiggle out of the contradiction with a bullet going through a gun barrel. The contradiction with the jet and hangar was avoided by there being a law of physics that says that that a clock attached to the front of a moving object will show a time that's later than a clock attached to the back of that object. But that difference in clock times doesn't help with a bullet going through a gun barrel, since it doesn't matter how you label the time, for example, at which the middle of the bullet is in the middle of the barrel. Red Act (talk) 18:08, 25 September 2009 (UTC)

What is the sun fusing?

I learned a little while ago that you can't fuse normal hydrogen because it would produce a helium atom with no neutrons, and those don't exist because they are too unstable. So is it true that the Sun does not fuse any normal hydrogen? It only fuses deuterium and tritium? I'm only talking about hydrogen btw, not other elements. ScienceApe (talk) 19:00, 23 September 2009 (UTC)

There are various reactions going on in the sun, the ones that dominate in the sun are the ones in the proton-proton chain reaction. It is true that at no point are two hydrogen nuclei fused to make helium-2 (which does not exist). For interests sake the other cycle is the CNO cycle which dominates in larger stars (greater than x1.5 solar mass). Elocute (talk) 19:09, 23 September 2009 (UTC)

But according to that article, you can fuse two hydrogen nuclei into deuterium which can then be fused into other elements. ScienceApe (talk) 20:17, 23 September 2009 (UTC)

That's right (as per the article) - there are intermediate reactions in the overal set of fusion reactions which convert Hs into He

Note that a proton can be converted into a neutron plus an antielectron (positive charge) (and of course some energy either way)

See the neutron article for examples of neutrons converting into protons and vice versa.83.100.251.196 (talk) 21:11, 23 September 2009 (UTC)

Vaporizer gun???

i was reading an ardicle and it mentioned something called a vaporizer gun for alternatice smoking. but there isnt a article on it on wikipedia, anyone know about it?

Perhaps the Vaporizer article is what you're after? -- Finlay McWalter • Talk 19:10, 23 September 2009 (UTC)

Are you thinking of an electronic cigarette, or possibly a metered-dose inhaler? TenOfAllTrades(talk) 19:13, 23 September 2009 (UTC)

im not sure, this was something you put the tabacco in. like a pipe, but you dont burn it? Talk

Rocket Problem

So let's imagine that sand is being shot into a car by a truck behind it. The truck maintains a constant distance from the car. The sand is being shot at a rate of 10kg/s, and at a speed of 5m/s. Initially the car is at rest, with a mass of 2000kg. What is the final velocity of the car after 100s?

I managed to find two "solutions", both which appear to be plausible but give different answers. I started by saying that dp/dt of the car = (10kg/s)*(5m/s) = 50N. This seems to make sense, as the sand will always have a 5m/s greater velocity than the car (the distance bewteen the truck and the car can be discounted). Integrating both sides, I get that p = (2000kg + 10kg/s*100s)*v = 5000,v=1.6m/s.

Second solution: As before, but with the equation adjusted a bit. m*dv/dt = u*dm/dt, dv=u*dm/m...integrating both sides, v=5m/s*ln(3/2) = 2.something. Which answer is right, and why? I suspect the second one, but don't know why it would be better than the first. —Preceding unsigned comment added by 76.68.245.124 (talk • contribs) 19:19, 23 September 2009 (UTC)

So at the end of your second paragraph, you say that p_final = 5000 kg*m/s, but you said in the first paragraph that the car has a mass of 2000kg. By my math, p = mv and v_final is 2.5 m/s. Am I missing something? Can you give some more decimal places for your second solution? 66.178.138.125 (talk) 19:42, 23 September 2009 (UTC)

Never mind, see below. 66.178.138.125 (talk) 19:56, 23 September 2009 (UTC)

The first solution has a problem - it assumes that you have to accelerate all of the sand loaded on to the car from rest. What actually happens is that the sand added later on starts out with some 'free' momentum, as the truck has been accelerating to maintain the same velocity as the car. The second solution looks better:

Acceleration as a function of time t, given a constant force F (of 50 N, per your calculation) and total mass of the car plus sand m(t):

• a(t) = F/m(t)
• a(t) = 50 N / (2000 + 10*t) kg

Integrate (velocity at time t is the integral of acceleration), and you're off to the races. A bit more than 2 is what I get, too (though less than the 2.5 of the above editor — he he didn't account for the mass of the sand on board the car). TenOfAllTrades(talk) 19:52, 23 September 2009 (UTC)

You and the original poster both made the mistake of writing F = ma = m dv/dt in a situation where m is a function of time. The correct formula is F = m dv/dt + v dm/dt. I'm pretty sure 1.6 m/s is the correct answer. -- BenRG (talk) 20:06, 23 September 2009 (UTC)

No, I didn't make that mistake; I explicitly noted that mass is a function of time. 1.6 m/s is the result that you get if you assume (incorrectly) that the mass of the car remains constant at its maximum of 3000 kg. TenOfAllTrades(talk) 20:20, 23 September 2009 (UTC)

Just to clarify everyone agrees that F=50N, and that m=2000+10t (kg) ?- the rest should be simple from there.83.100.251.196 (talk) 20:24, 23 September 2009 (UTC)

Thus I would have written dv/dt = 5/(200+t) - getting the second answer.83.100.251.196 (talk) 20:30, 23 September 2009 (UTC)

it assumes that the sand transfers its extra velocity relative to the car as momentum (and so force), and thus once that is done is travelling at the same speed as the car.83.100.251.196 (talk) 20:35, 23 September 2009 (UTC)

Yes, the first solution is correct, assuming that the speed of 5 m/s is relative to the ground. The formula m*dv/dt = u*dm/dt is incorrect. Ehrenkater (talk) 20:40, 23 September 2009 (UTC)

u*dm/dt=dp/dt=d(mv)/dt=m*dv/dt+v*dm/dt Ehrenkater (talk) 20:44, 23 September 2009 (UTC)

I meant to say 5m/s relative to the car, sorry. What difference would that make?

Actually, I would have to disagree with you. If it were 5m/s relative to the ground, then the momentum transfer of the sand wouldn't always be 50N, because the sand would have to decelerate less later on. So force=10kg/s*(5m/s-v).

There are some ambiguous details about the problem that lead to different answers depending on how they're interpreted.

• Does the sand stick to the car, does it collide inelastically and then fall to the road, or does it collide elastically? It's unclear from the way the problem is worded, but I'm assuming it's supposed to be the first case, since the OP's two solutions seem to make that assumption.
• Is the sand being fired at 5 m/s relative to the road, or 5 m/s relative to the car and truck? In the first case the OP's first solution is correct, and in the second case the OP's second solution is correct (both assuming the sticky sand).

Rckrone (talk) 20:54, 23 September 2009 (UTC)

For the first comment, yes the sand sticks...in retrospect calling it a car was a bad idea. And the sand was supposed to be fired at 5m/s relative to the car at all times...but wouldn't this mean that the force = 10*(5-v), and not 50?

Only the change in momentum of the sand matters, so if the sand is moving at 5m/s relative to the car, when it hits Δv = 5 m/s. So Δp/t is 50N. If it was 5 m/s relative to the ground then it would be (10 kg/s)*(5 m/s - v), but in that case it's easier to just sum up the total momentum and divide by the total mass as in the OP's first solution than to worry about forces. Rckrone (talk) 21:10, 23 September 2009 (UTC)

Sorry, what I meant to say was: the sand was supposed to be fired at 5m/s relative to the car at all times, but if it were fired at 5m/s relative to the ground, wouldn't F=10*(5-v)? You said that if the sand were fired at 5m/s relative to the ground, then my first solution would be corrent. But my first solution used F=50N.

This problem represents a tricky case where the change in momentum per time of the car is not equal to the force that the sand is exerting. That's because in addition to exerting a force on the car, the sand also contributes its own final momentum to the new car+sand system, which is evidenced by the increase in mass of the car. In the case where the sand is going at 5 m/s relative to the road, the cars momentum increases at a constant rate of 50 (kgm/s)/s, however the force the sand exerts is only (10 kg/s)*(5 m/s - v), and the additional (10 kg/s)*v is contributed through the sand (now traveling at a speed of v) staying with the car. In the case where the sand is at 5 m/s relative to the car, the force that the sand exerts is a constant 50 N, but the contribution to the car's momentum per time is more, since it contributes an additional (10 kg/s)*v. The problem can be solved using either change in mometum of the car, or force of the sand on the car, but they shouldn't be confused since they're not equal. The first solution treats the change in momentum per time of the car as 50 N rather than the force. Rckrone (talk) 21:59, 23 September 2009 (UTC)

Okay great, that makes a lot of sense...thanks a lot!

2X solution from 1X solution?

If it says the 1X solution contains some wt/vol of substance A, vol/vol of substance B, and x mM of substance C, how do I make a 2X solution? Do I just multiply every substance by two and put them in the same volume of water? Sorry I have no idea here. 199.76.176.159 (talk) 23:31, 23 September 2009 (UTC)

If what says that? --Tango (talk) 23:40, 23 September 2009 (UTC)

This is homeopathic terminology: a 1X solution is 1 part substance to 10 parts water, a 2X solution is 1:100. So to a close approximation, you make a 2X solution my mixing one part 1X with 9 equal-sized quantities of water. If you keep diluting like this until you reach 24X, you have created a standard homeopathic preparation -- i.e., pure water. Looie496 (talk) 00:32, 24 September 2009 (UTC)

Another usage of this term (probably what the OP is referring to) is the process of making a concentrated stock solution. This is a common lab procedure for storing buffers, etc. It's also a convenient way to make a more complex solution by adding different stock components and then bringing the solution up to the final volume with water. In this usage, 2X is a concentrated solution (either add twice the amount of solute per unit volume, or dissolve in half the water). It's not uncommon to have a 10X, 100X or even 1000X concentrated stock solution, depending on the substance. --- Medical geneticist (talk) 01:17, 24 September 2009 (UTC)

Yep, for example, my lab sometimes uses 10X DMEM. We just dilute it 10-fold. Just double the amount of solute, or use half the water. Tim Song (talk) 02:49, 24 September 2009 (UTC)

DNA Structure

Hi, I'm extremely interested in the field of Cancer Research; I was wondering if anyone out there is a specialist in the structural form of DNA. If so, please leave your e-mail below, for I have a word document with a DNA Model I have drawn which I would like to have looked over and checked. Please Help!

Thanks, 74.184.100.154 (talk) 23:48, 23 September 2009 (UTC)

The ref desk staff only offer help right here on the ref desk pages. Also, not many people are willing to post their email addresses. If you can place your document on a web site someplace - it's likely that someone would be able to take a look at it and reply here. SteveBaker (talk) 23:57, 23 September 2009 (UTC)

Then e-mail me at <redacted> if you have knowledge about DNA. Thanks, 74.184.100.154 (talk) 00:00, 24 September 2009 (UTC)

I assume you have already looked at our DNA article. Noting your e-mail name, can I ask whether your model agrees with our article/s. Kaiwhakahaere (talk) 00:05, 24 September 2009 (UTC)

I think, I would just like to make sure. Thanks, 74.184.100.154 (talk) 00:07, 24 September 2009 (UTC)

You might also want to look at DNA structure, DNA nanotechnology and Molecular models of DNA. If you have read all of those, you should be able to have a good idea if your DNA model is correct. You might also like to read Cancer research. As SteveBaker says, if you can place your document on a website somewhere then someone could perhaps look at it. Incidently, I have removed your email address - responses will be put on this page, rather than e-mailed to you. -- PhantomSteve (Contact Me, My Contribs) 07:36, 24 September 2009 (UTC)

If the OP wishes to allow e-mail from other Wikipedia users, the correct procedure is 1) Register an account with Wikipedia, and 2) tick in "My preferences" at "Enable e-mail from other users". Then others can e-mail you from your User or User talk page by using the "E-mail this user" feature. They will not see your e-mail address but you will see theirs.Cuddlyable3 (talk) 15:51, 24 September 2009 (UTC)

True - but irrelevant. The RefDesk does not provide answers by email - period. The idea is that both question and answers must be available for everyone to read. This is vitally important because that's the only way that errors in answers are caught and fixed by subsequent contributors. It also ensures that once someone has carefully researched the answer and provded a decent reply - nobody else wastes their time doing the work all over again! Furthermore (it is hoped!) people will search the archives to see if their question has already been answered before they ask it...although that's something that doesn't happen half as often as we'd like! If OP's get answers via email, they could be horribly wrong and nobody would ever know about it and maybe 30 different people might end up researching and providing almost identical answers. Both would be a serious subversion of the way we work. So we don't do that...ever...as a matter of policy. SteveBaker (talk) 17:36, 24 September 2009 (UTC)

With respect SteveBaker my post is both true and relevant. It is relevant because the OP has every right to wish for e-mail contact and needed to be told that Wikipedia offers that as a built in optional function. It is less relevant though also true that not everyone wants to release their e-mail address while others, including the OP, don't mind. If and when any user chooses to send the OP an e-mail it will be neither your, mine, the RefDesk's or Wikipedia's business whether that private correspondence is "horribly wrong". PhantomSteve already made the point that the OP's address should not appear here, without making a doom warning about it. The subject is properly covered here.Cuddlyable3 (talk) 19:31, 24 September 2009 (UTC)

Well, I'm glad you all could help.................... 74.184.100.154 (talk) 20:35, 24 September 2009 (UTC)