Wikipedia:Reference desk/Archives/Science/2013 August 18

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.

August 18

what substance is this

hello, we all know these electronic devices, mostly for computers, with a fancy casing that feels rubber-like or velvety to the touch. I have a DVB-T adapter like this. The thing is, its casing got sticky and somewhat gooey over time, I think because of this coating, whatever it is. What is it and is it dangerous (notorious plasticizers evaporating and whatnot?) Уга-уга12 (talk) 01:25, 18 August 2013 (UTC)
If I'm worried, is there a way to get checked for this stuff (absent concrete medical complaints)? Will one's GP do it? (I think not), and if not, who will? Thank you everyone Уга-уга12 (talk) 01:37, 18 August 2013 (UTC)

We won't be able to give you medical advice, but without even knowing the make and model of the device, it's unlikely that anybody will be able to give you any information at all. Looie496 (talk) 20:03, 18 August 2013 (UTC)

Is it possible the material was exposed to some type of oil, such as peppermint oil ? This has the effect of making many plastics tacky (or completely dissolve in sufficient quantities of the oil). StuRat (talk) 07:10, 20 August 2013 (UTC)

How accurate is long-range DNA analysis?

Consider this scenario:

There are two men, who are non-identical brothers, and I know that one of them is my father, but not which one. Today, DNA analysis can tell me with certainty which one is my dad.

Suppose now that I am not the son of one of those men, but the great great grandson of one of them, and that we have their DNA profiles on record. Is it just as easy to say which one of them was my great great grandfather? And if not, for how many generations is the verdict’s certainty beyond doubt, and for how many before it has no value at all?

And is the scenario different if the siblings are two women, or one women and one man? Myles325a (talk) 01:58, 18 August 2013 (UTC)

Straight line male ancestors generally are identifiable a long way back. Red Y-chromosomal Adam for some info on this. ←Baseball Bugs ^{What's up, Doc?} carrots→ 03:15, 18 August 2013 (UTC)

Wouldn't Y-Chromosome testing be almost useless in determining which of two siblings is the father? If Y-Chromosome is useful in identifying patriarcal lineage because it changes very little, it also means two male siblings would have almost the same Y-Chromosome as their own father. The test can be useful over several generations to determine if you are patrilineally related to a person, but I'm not sure it would be useful in determining which of two brothers is the father (assuming the brothers have the same father) Effovex (talk) 18:18, 18 August 2013 (UTC)

Ah, I think you're right that the Y-Chromosome wouldn't be enough to fit your scenario. It should be enough to prove that the great-great-great-grandfather was (or was not) in your line. But not enough to prove which of his sons was your great-great-grandfather. Additional information would be needed. As with the blood test, it can be a lot easier to disprove paternity than to prove it. ←Baseball Bugs ^{What's up, Doc?} carrots→ 23:58, 18 August 2013 (UTC)

• The most powerful method is whole-genome sequencing. Given that the body of Richard III, who died in 1485, was recently identified by comparison with the DNA of a known modern descendant, it's clear that the approach can go back a long away. Not indefinitely, because each extra generation cuts the signal in half, but quite a long way. Looie496 (talk) 19:59, 18 August 2013 (UTC)
  • Richard III was identified by matrilineal descent -- mitochondrial DNA passed down almost unchanged from Richard's mother down a line of females to the mothers of two test subjects today, which all matched.

  For the scenario raised, you need to be looking at autosomal DNA -- ie DNA from chomosomes other than X and Y. The simplest example of that is the CODIS test used in forensics. But the amount of autosomal DNA inherited from a particular ancestor halves with each generation you go back -- so whereas 13 CODIS markers is very likely to be enough to determine paternity, it may or may not be enough even at the level of grandparentage. (You would need to be very sure of all three other grandparents' CODIS markers, and hope that the two to four markers inherited from the ambiguous fourth grandparent, (i) are different between the two candidates, and (ii) are different from what might have come from any of the other three grandparents).

  You can get round that to some extent by going to more markers, eg an Affymetrix 500,000 SNP marker chip as used by eg 23andMe, though those are SNP markers, which have less variation (both intrinsically, and in the population) than the STR markers used for forensic DNA, so you need a lot more SNP markers even to be as informative as the CODIS STR markers.

  But there certainly is some current interest from people really into genealogical DNA in trying to see whether they can find blocks of DNA that remote cousins may each have inherited from a common ancestor. But it's hit-and-miss, because different cousins will for the most part have inherited quite different blocks of DNA from their common ancestor, and the size of blocks also can be fairly random, so there's a real possibility that there may turn out to be no DNA that the distant cousins may in fact have inherited in common.

  You would also need to establish that the brother ancestor contender had different DNA in that block (on both chromosomes), so you would need to find yet more cousins that happened to pinpoint (both copies of) his DNA for that exact same block. No doubt somebody here can give more of an estimate for the exact numbers, but I suspect at even the level of only a few generations back, you would need to be finding a lot of different cousins to try to establish the brother's likely DNA marker values; and there's also a very good the relevant sections of the brother's DNA might not have successfully been passed down to anybody -- see Galton-Watson process for how really quite likely it is that a particular DNA section (originally somebody's Y-chromosome, but a similar analysis actually applies to any distinctive section of DNA) may end up not being represented in the population at all.

  So probably you may by chance find particular blocks of DNA that can be traced back to particular ancestors. But probably it quite rapidly becomes really quite unlikely for you to be able to expect to be able to find a block of DNA that can be used to identify any particular given individual ancestor. Jheald (talk) 20:56, 18 August 2013 (UTC)

  • Here's some information from ISOGG about 23andMe's "Relative Finder" feature, which attempts to do this; though it looks like the page was created in 2010, and may not have been updated that much since. Jheald (talk) 21:16, 18 August 2013 (UTC)

OP myles325a back live. Thanks to all, and especially Jheald for his detailed explanation. I imagine that there is, somewhere, an algorithm that tells you exactly how much chance you have of determining the great great grandfather question, but it is prob. v technical. Myles325a (talk) 06:08, 20 August 2013 (UTC)

The reason why this is hard to answer is that it relies entirely on the additional data of other people somewhere in the chain of ancestry. If you have DNA from every relative all the way up and down the chain you can know with certainty. If you have DNA from only you and your potential great-N-grandfathers, it is completely unknowable. So there's no relevant algorithm. Wnt (talk) 14:18, 20 August 2013 (UTC)

OP myles325a back live. Are you sure you aren't just making this all up, Wnt? If what you say is true, then I cannot know the truth even when one of those men is my grandfather and the other his brother. And you assert that it is not as if there might be some doubt concerning the result there, the truth "...is completely unknowable". I don't believe it. Myles325a (talk) 06:01, 21 August 2013 (UTC)

Basics of blisters

After doing a bunch of insufficiently-gloved shovelling today, I read the blister article and was left wondering on a few things:

1. "Causes" says that blisters cushion tissues under damaged skin, "protecting it from further damage and allowing it to heal". Does this mean that friction-caused blisters only form when all layers of the skin have been damaged?
2. Why do blistered areas feel "hot"? Does the serum exert some weird kind of pressure on the underlying tissues, or is this an effect of the damage that provokes the formation of the blister?
3. "Treatment" says that the fluid comes from the damaged cells, while the intro says that it's Serum (blood). Is one wrong, or is blood serum somehow inside the damaged cells? If so, how does it get in the cells, while if not, how does it leave blood vessels (without visible bleeding happening) to collect and form a blister?

Perhaps the answers to #1 and #2 are in "Treatment", but I wasn't able to understand it well. Clarification of that section would be appreciated. Nyttend (talk) 02:19, 18 August 2013 (UTC)

The fluid in blisters comes from interstitial fluid, which in turn comes from serum. Fluid in suction blisters is even described by some sources as being interstitial fluid, or at least it is used as a stand-in for it for biomarker studies of the interstitial fluid.^[1][2][3] That said, blister fluid is not truly the same as interstitial fluid because of course the act of blistering itself involves things secreted to prevent infection or promote healing, such as lysozyme and thymosin beta 4.^[4] (As the latter factor can assist in scar-free healing of cardiac tissue,^[5] I think that a careful and systematic scientific reevaluation of cupping therapy as practiced by ancient Greek physicians and their intellectual heirs (e.g. hijama) may be advisable...) Wnt (talk) 21:35, 18 August 2013 (UTC)

2) I think the heat is real, and is a result of the healing process. Any rapid cellular changes, including healing, generate heat. For comparison, bacteria growing in a pile of manure can make it hot, too. StuRat (talk) 07:07, 20 August 2013 (UTC)

I don't think this is the major factor - see inflammation. As described by Aulus Cornelius Celsus, "calor" (heat) is one of the four classical signs, and as the article mentions, it is largely due to increased blood flow. Remember the skin is at much less than the classic 37 C core body temperature. My thinking is that this isn't directly related to blistering - for example, I never noticed any particular heat with the painless (for me - YMMV) blisters raised by phenol. Wnt (talk) 14:15, 20 August 2013 (UTC)

audio recording

when one records audio from speakers with a microphone, why is the resulting sound different from the original? and how does one make it more similar? (i know the basics of acoustics, digital audio, how speakers and mics work, ect...) thank you, 70.114.248.114 (talk) 03:04, 18 August 2013 (UTC)

You're picking up sound through your ears in a somewhat different way than a microphone does. ←Baseball Bugs ^{What's up, Doc?} carrots→ 03:20, 18 August 2013 (UTC)

True, but I don't think that's quite what the OP is talking about. Let's say you're listening to music through a speaker, while holding a mic. If you play the sound recorded from that mic through the speaker it won't sound the same as it did the first time. There are any number of reasons for this, including the introduction of background noise (also known as "air"), and the reduction of stereophonic sound to mono because the mic cannot capture a binaural recording. The mic you're using is also unlikely to be as high quality as the one the original artist used. You can compare what's happening with path loss as some of the concepts are similar. There's also information at high fidelity. As for increasing the fidelity, the easiest way is always to get the best possible initial recording rather than to try to fix it afterwards (for example, dubbing rather than using a mic). A sound engineer (even an amateur one) is the kind of person you'd need to talk to for advice, but I suspect that a lot of it comes down to drudge work, as is the case with most restoration projects. Matt Deres (talk) 14:44, 18 August 2013 (UTC)

Thanks for the detailed info. He's basically asking the same question as below, only for audio. ←Baseball Bugs ^{What's up, Doc?} carrots→ 03:20, 19 August 2013 (UTC)

Consider what happens to the sound (vibrations of the air) between speaker and microphone. The sound spreads outwards like ripples in a pond - then some of them will hit other objects and bounce back. The reflected ripples then interfere with the original sound and add echoes and other effects that were not present in the original recording. One can reduce this effect using something like an anechoic chamber - but it's impossible to completely eliminate it because (for example) the microphone and speaker themselves reflect sound - and no fancy coverings of wall, floor and ceiling can ever completely remove 100% of reflections.

Then of course there is the problem that the loudspeaker can't perfectly match the electrical signals being fed to it and the microphone can't perfectly convert air pressure measurements into electrical signals. Add to that stray vibrations coming from other sources - traffic in the street outside, that kind of thing...and again, you can add sound insulation - but it won't ever be perfect. There are other effects such as resonance in parts of the equipment such as the microphone stand and any cables leading to it that add more layers of change to what reaches the microphone.

Since both speaker and microphone are analog devices, any source of electrical noise will affect them - and magnetic interference is also possible if they use electromagnetic technology to do their jobs.

If the input signal is digital and so is the output - then resampling that digital waveform introduces errors and (as I said below) the Nyquist limit becomes an issue.

With increasing levels of cost and sophistication, all of these effects can be greatly reduced - but it's fundamentally impossible to totally eliminate any one of them. So, no - your re-recording cannot possibly ever be "perfect".

SteveBaker (talk) 13:06, 19 August 2013 (UTC)

recording video

i'm not asking cause i want to record copyrighted stuff, but is it possible to get nearly perfect quality by placing a video camera in front of a screen playing a video file? and if so, how? thanks, 70.114.248.114 (talk) 03:38, 18 August 2013 (UTC)

At the very least, I expect the camera would need to be totally in sync with the TV, e.g. by having the same megapixel dimensions and exactly synching with the scanning process. Check out Kinescope and see how they did it in the old days. ←Baseball Bugs ^{What's up, Doc?} carrots→ 03:53, 18 August 2013 (UTC)

At the very least, the recording camera would have to have twice the linear resolution (ie four times the number of megapixels) of the original picture because unless the alignment between camera and TV is 100% perfect, you're resampling the image and the Nyquist limit kicks in. But realistically - no, you'll never get a perfect quality image - there is sure to be some degradation no matter what. SteveBaker (talk) 23:45, 18 August 2013 (UTC)

Is it really twice the linear resolution? I'd say it has to be higher than the original, and the higher the better, but twice? The minimum wavelength is two pixels, so I'd say the "2" factors cancel out. - ¡Ouch! (^{hurt me} / _{more pain}) 06:00, 22 August 2013 (UTC)

Concur with others comment here. The quality of video from a camera pointed at a TV is going to be much less than the original, unless specialised equipment is used. Even still pictures of TV turn out rather poorly (wp:OR). See also Scan conversion. The ref I just added to that page maybe of interest "Digital scan converter" which also mentions the use of vacuum tube converter technology for applications such as converting radar & sonar displays. This indicates the lengths needed to got to to get good results say when converting film to video. --220 of ^Borg 12:02, 19 August 2013 (UTC)

Contradicting specification on the density of Sulfur hexafluoride ?

In the body text of the article "Sulfur hexafluoride". The density is said to be 6.12 g/L at sea level conditions. But the right side property column says 6.17 g/L. And a yahoo answer says 6.52 g/L using (density = 146.06 g / 22.4 L = 6.52 g/L). So which is it? Electron9 (talk) 15:51, 18 August 2013 (UTC)

Which reliable source is connected to which numbers? Find the original reference work, and you can find out. It should be noted that based on differences in measuring techniques, different reliable sources may have different reported values. Gases, being compressible and highly dependent on temperature, have very variable density, and depending on which set of "standard" conditions (be it STP, or Standard state, or Room temperature, any of a number of other conditions) will greatly affect the value. So, 1) Check the original sources and 2) careful observe what conditions those sources used to obtain their results. That's how you're going to get your values. --Jayron32 19:48, 18 August 2013 (UTC)

Where to find a reliable (online) source for physical data on gases? Electron9 (talk) 21:06, 18 August 2013 (UTC)

Personally, I'd go with the appropriate MSDS. 24.23.196.85 (talk) 05:34, 20 August 2013 (UTC)

Stephen Hawking and The "Multiverse" - Does he really believe it?

I was watching Into the Universe with Stephen Hawking - The Story of Everything. I couldn't really tell if the narrator was supposed to pretend that he was Stephen Hawking or a different voice, but the narrator gave his alternative explanation of a perfectly well-tuned universe - the hypothesis of a multiverse. So, I did a Google search on the topic of the multiverse, and as usual, Wikipedia appeared at the top of the search results. I went straight to the criticisms, which seemed to be fairly plausible. Does Stephen Hawking really believe in the multiverse speculation, or is that the opinion of the documentary's narrator? If it were me, then I'd probably say that our universe is just a very special place, carefully constructed and assembled by mother nature. Sneazy (talk) 20:24, 18 August 2013 (UTC)

One question to consider is whether you believe this is the only possible way the universe could have turned out. For example, a couple of estimates for the entropy of the entire observable universe are 2^305 bits (Seth Lloyd, 2002) or 2^350 bits (Charles Egan, 2010). Naively, that suggests that the present state of our universe is one out of 2^(2^305) or 2^(2^350) possibilities.

Now if those other states actually exist, then physics can leave it down to pure blind chance that we find ourselves in 1 out of the 2^(2^305) or 2^(2^350) possibilities. But if only this universe exists, then how did the information that selects it from 2^(2^350) other possibilities come about?

And beyond that, if the universe really could have turned out differently, then in some sense that in itself makes them "real" possible worlds -- if they are (or were) truly possible, then they are (or were) real alternatives.

If they are irretrievably separated, and can never ever interact, then arguably that makes it irrelevant whether they "really" exist in the present or not -- or only as alternatives that could have existed.

But positing that they do "really" exist is fun for physicists, because that opens the door that down the line maybe some form of interaction really is possible. It also saves physicists from having to explain how one possibility really turns out to happen (or to have happened), whereas all the others apparently didn't (or haven't). It probably helps sell books and tv documentaries too. 22:03, 18 August 2013 (UTC)

This is why Hawking believes in parallel universes. Count Iblis (talk) 22:43, 18 August 2013 (UTC)

I am not sure how one is going to test this. Maybe this is one of those unfalsifiable mysteries of the universe. Even if other universes exist, I cannot imagine how their existence would ever affect our universe, given the vast and accelerating expanse of our universe. So, I would agree with you that if they are irretrievably separated and can never interact, then that makes it irrelevant whether they really exist in the first place. It may be fun for a physicist, but practically speaking, I am not sure how you can test something without proposing any assumptions about it. If an alternative universe has a completely different set of physical laws, then everything else would be left to the imagination! For practical experimental purposes, maybe it's best to stay in this world, in this universe. When someone asks about the reason for why our universe is the way it is, we may simply answer, "Because that is just how our universe works, and we just happen to live in it. In any case, this is no longer a scientific question. Science can't answer everything; this is why you have the Humanities." Sneazy (talk) 23:20, 18 August 2013 (UTC)

It's not unfalsifiable, you can test if time evolution is unitary; to get rid of the "many worlds" of the Many Worlds interpretation of quantum mechanics requires a violation of unitary time evolution. Count Iblis (talk) 23:30, 18 August 2013 (UTC)

I don't think he "believes" in it in the sense that you beleive in God. I think like many rational people who have looked into this possibility, he thinks it is a very interesting hypothesis which deserves us attempting to test it, if we can. And if we can't, well it is still intellectually interesting, but he won't start making decisions about his own life based on a dogma that says that this must be true, or give 10% of his income to a church that says that this IS the truth. --Lgriot (talk) 11:10, 19 August 2013 (UTC)