Wikipedia:Reference desk/Archives/Science/2013 November 13

= November 13 =

Messages artifically coded into DNA?
Can this be taken seriously? It was accepted by Icarus. Bubba73 You talkin' to me? 00:47, 13 November 2013 (UTC)


 * If I'm reading that right, they want to check the DNA of Earth organisms for signs that a message is encoded in it, presumably from aliens who first seeded the Earth with life. The problem I see is that if this happened when life first appeared on Earth, billions of years ago, the "signal" would be totally lost to the "noise" of random mutations, by now.


 * On the other hand, if they visited recently, say a few thousand years ago, and placed the signal DNA where we can find it, then we might still be able to decipher it from the descendents of that organism. (Heck, if they used a long-lived organism, it might still be the same one they implanted.) StuRat (talk) 01:41, 13 November 2013 (UTC)


 * I believe that they are talking about a long time ago because the abstract says "might remain unchanged over cosmological time scales". Bubba73 You talkin' to me? 02:32, 13 November 2013 (UTC)


 * Certainly any message would be horribly corrupted by now - but if there was a deliberate effort to retain the message (and if it's simple enough) then they could use advanced error-correcting codes to make the message retrievable. We know that 35% of our useful genes are shared with algae and plants - so if the message was implanted before the split between animals and plants - then a third of it would still be there...and it's really quite easy to come up with error-correcting codes that can survive total destruction of two thirds of the message.  The tricky part is that generating a useful message in the useful parts of the DNA would be difficult - you couldn't choose just any old string of A's, G's, C's and T's without making proteins that wouldn't work to keep creatures alive and reproducing.  The likelyhood is that any message would have to be stored in "non-coding" or "junk" DNA...but the problem with that is that parts of the molecule that aren't important get mutated more quickly because changes don't kill the mutated organism and there is no evolutionary pressure to stop them from changing too quickly.


 * Perhaps we're not thinking big enough though. In the book (but not the movie) "Contact" (by Carl Sagan), the aliens somehow engineer a digital image of a perfect circle far down the expansion of PI.   Now *that* is "godlike power"!


 * What blows your mind is that since PI doesn't repeat and is infinitely long, it is certain that there are many digital pictures of circles embedded in it! (Sadly, there are also full color photos of Aardvarks and an infinite number of photos of Angelina Jolie - so this has little to teach us of the mind of God!)  After I read "Contact", I wrote software to see if there was an image of a circle in the binary expansion of PI to the first few billion digits - and there isn't anything large enough to be convincing in base 2.  :-(


 * Perhaps the message the aliens left was "Let there be life!"...and that message has survived through the 4 trillion generations since the first cell was formed. SteveBaker (talk) 04:08, 13 November 2013 (UTC)
 * Hmm. I suspect you're using a bit more than that (the decimal expansion) of pi doesn't repeat (that is, never settles into an infinite repeating pattern).  Probably you're assuming that pi is a normal number, or at least normal to base 10?  That's almost certainly true, but no one knows how to prove it.
 * For example, 0.1101001000100001000001000000100000001... (hope the pattern is clear) also "doesn't repeat and is infinitely long", but it doesn't have any photos of Angelina as contiguous binary strings corresponding to JPEGs. I suppose you could argue that it necessarily encodes such photos in some other way; any JPEG can be coded as a natural number, and every natural number shows up as the number of 0s between two 1s.  Is that what you meant? --Trovatore (talk) 04:40, 13 November 2013 (UTC)
 * Thank you. As far as the image in the bits of pi, nothing or no one can change what pi is.  Bubba73 You talkin' to me? 04:32, 13 November 2013 (UTC)
 * When I see so-called documentaries discussing "ancient aliens" who supposedly built all of our ancient wonders and perhaps also re-engineered our DNA to be smarter or whatever, they never address the obvious question: If that superior species had to tinker with our DNA, who had to tinker with their DNA? And so on... kind of a "turtles all the way down" issue. ←Baseball Bugs What's up, Doc? carrots→ 05:09, 13 November 2013 (UTC)
 * It doesn't have to be turtles all the way down. For example, if human scientists made intelligent organisms, it wouldn't entail that we were made. Perhaps hyper advanced aliens got really bored and decided to monkey around and make Earth life, nothing about that requires that they were made. Of course, though, I'm not saying I think that this happened, but that being engineered doesn't imply the engineers were too.Phoenixia1177 (talk) 11:47, 13 November 2013 (UTC)
 * Having considered any number of crazy ideas of my own devising, I recognize a kindred spirit in these Kazakhstanis. But despite the effort they take, this bioqabbala doesn't actually prove much: there's no significance, no estimate of probability that I can see.  Note they are claiming that the message was embedded into the genetic code.  However, from a paper I enjoyed in PLOS just recently  it is fairly convincingly argued that the genetic code was evolving in known evolutionary time: single celled organisms didn't just show up with a finished product ribosome, but spent close to a billion years around 3 billion years ago figuring out how to do it better and better.  (That paper itself is way out there, but IMHO it's on the part of the branch that ought to hold)


 * I do not exclude the possibility of aliens tinkering with DNA of early life - I don't think that the "turtles all the way down" idea is disproved, because it seems like the pace of life (and physics) was asymptotically faster the closer you come to the Big Bang, so that any number, perhaps even an infinite number, of epochs of civilization, each shaping the next, are possible. But it'll take more proof than this. Wnt (talk) 07:23, 13 November 2013 (UTC)

It's worth noting that the DNA content we share with algae is shared because it's so useful. Functional DNA appears to mutate at a slower rate, because organisms with deleterious mutations in useful DNA tend not to reproduce. It would be quite a feat to encode a message that is also indispensable into an existing organism. Someguy1221 (talk) 11:35, 13 November 2013 (UTC)


 * I'd just like to note that with the exception of Wnt's comment, all of the comments above show a misunderstanding of what the cited paper says. It does not argue that messages can be found in the DNA, it argues that messages can be found in the genetic code -- that is, in the code that defines the relationship between DNA and the proteins that it produces.  If you don't have a reasonably deep understanding of genetics, you will need to read the genetic code article to understand what that means.  I personally think it's a bizarre idea, but even so, we should at least be clear on what idea we are discussing. Looie496 (talk) 15:46, 13 November 2013 (UTC)


 * That's ridiculous. The genetic code is really simple.  It is a set of rules (kinda like a computer programming language) for making proteins from a description stored in DNA.  It's an incredibly simple mechanism: For each three letter combination of A's, G's, C's and T's, a different amino acid chunk is added to the protein being constructed.  A couple of those combinations mean "STOP" and a few are unused.  The code is just about the most simple language imaginable - there really isn't anything there to imply a deep mystery.  DNA codon table explains the whole thing in a tiny table that fits on half a page of text.  There simply isn't room in that encoding to carry any sort of a message!  That's flat out impossible - and if they are even discussing that, they are complete idiots.


 * They must be talking about a message that's embedded in DNA strands themselves - and some misreading or mistranslation of what this research group are doing has occurred.


 * It's entirely possible to embed binary (well, technically, base-4) messages in DNA. There are actually companies out there that will do that for you for a fee (http://www.genscript.com, for example).  You can go to them and for just 28 cents per base-pair, they'll make DNA for you with any sequence of A/G/C/T you like!  There are papers being written about standardizing the "BioCode" used to store messages - which A's, G's, C's and T's correspond to which binary digits - how ASCII text may be converted to BioCode, what error-correction scheme should be employed and so forth.  Two strains of common bacteria are routinely used for this: E. coli and D. radiodurans.


 * This is well-established science and it's done for things like DNA watermarking. You can buy a pen with specially marked DNA in the ink which can be used to sign important documents in such a way that would allow future investigators to prove beyond reasonable doubt that that particular pen was used to sign that document.


 * Civilisations of the future may very well find themselves able to read messages that this generation of humans embeds in the DNA of all manner of living things. It's only a matter of time until someone decides to leave such a message.


 * Hence, if you believe that aliens ever came to earth (which seems entirely unlikely), then it's entirely plausible that we might find a message left by them in the DNA of some or all living things (depending on how long ago they came here). However, because of the need to employ a highly redundant code in order for the message to survive natural selection and mutation - it might be quite hard for us to detect and decode it.


 * If you truly did want to leave a message for the future - that's a great way to do it because unlike physical objects like stone tablets or CD-ROMS, it's self-renewing and would be more or less guaranteed to still be here millions of years into the future.


 * But there is simply no possibility to embed any kind of a message in the "genetic code" mechanism that transcribes it - there simply isn't enough information in the code to say anything more than "Hello"...and certainly not enough information to allow some future civilization to figure out the encoding and translate the language!
 * SteveBaker (talk) 16:25, 13 November 2013 (UTC)


 * Steve, Bubba linked to the full paper. All you have to do is read the abstract to see that I am telling the truth.  I never said it made sense.  I did say one thing a bit misleadingly though -- they aren't exactly talking about complex messages embedded in the genetic code, more like information that provides evidence of intelligent design. Looie496 (talk) 16:36, 13 November 2013 (UTC)
 * Sorry - I didn't mean to imply that the posting here was nonsense - to the contrary, it was a valuable contribution. I only meant to imply that the paper seems to be either nonsensical - or misinterpreted in some way. SteveBaker (talk) 21:34, 13 November 2013 (UTC)
 * In theory, assuming a three letter code and 22 amino acids [ sic] + stop, you have 64 positions which can express 23 outcomes, i.e. 64 x 23 options = 5 bytes + 6 bits.  Perhaps the Gods do not speak Unicode, or perhaps they are Chinese. :) Wnt (talk) 19:50, 13 November 2013 (UTC)
 * While it's true that there are 64 possible outcomes - of which only 23 (-ish) are used, this is not particularly wasteful. With a base-4 number system (A/G/C/T), you get 16 possible outcomes from a two letter code and 64 possible outcomes from a three letter code - so any less than we have now is clearly not enough. So godlike or evolved, you need a three letter code...and that's what we have.  I don't know enough biochemistry to say whether life could have been "designed" to get by with only 15 amino acids...but I suspect not. Another theory about this is that periodically, DNA will lose one base-pair.  This causes the whole code to slip one digit.  Having unused codons forces the DNA transcription to purposefully fail under such situations rather than to generate a bunch of junk proteins that don't do anything useful...which I believe to be important to the mechanisms of DNA repair and such like. If there had been exactly 15 amino acids plus one stop code...or if there were 63 amino acids - then that would be impossible. SteveBaker (talk) 21:34, 13 November 2013 (UTC)
 * Yes a three-position codon unit is the minimum needed to be able to handle 23 distinct amino acids. However the specific mapping is more or less arbitrary. In theory any of 64-factorial/23-factorial different codes could have become established. If you postulate aliens who could tinker with early life of a very fundamental level (quite unlikely, I think) then they could have made that choice, leaving essentially a single number, the index of the actual coding table sued out of all the possible tables. But that gives us a value without an encoding scheme. If I tell you that 100110100100011101010000111101010111001001010111111100001010111001001111001100011 is an encoded message, well maybe it is. but unless you have some idea what encoding scheme it uses, there is no meaningful message there. (By the way, I understand that the Genetic code isn't completely random. It seems to be so arranged that one-base errors will often result in an amino acid with similar chemical properties to the one originally coded for, thus reducing the chance that mutations are completely destructive, as I understand it. I don't know how many distinct genetic mappings ("codes") would have this property.) DES (talk) 22:44, 13 November 2013 (UTC)
 * Note that above I was taking 64 x 23, not min(64, 23), and considering all possible genetic codes, not all possible things encoded with a genetic code. The paper I cited makes a good argument that life initially got by on a much smaller set of amino acids, including simple repetitions of a single pair of amino acids, and that, oddly enough, our proteins still show traces of some of this ancestry!  I should add that there is quite some interest in adding new amino acids to the code for purposes of synthetic biology, and there's been considerable progress. Wnt (talk) 22:46, 13 November 2013 (UTC)
 * If you want to consider all possible genetic codes, you may actually have to go beyond 23. As our expanded genetic code article explains, modified organisms have been created that use altered genetic codes involving at least 40 different nonstandard amino acids.  It's not clear what the full range of possibilities is. Looie496 (talk) 00:43, 14 November 2013 (UTC)
 * Thanks, I should have linked that article. For past alien designers these aren't an option, unless we suppose life lost much of the code (and with it, the message).  For us creating new life, that's another matter.  Actually, there's one thing I'm interested in which I haven't seen - is there a way to make a basic amino acid closely homologous with lysine/arginine that doesn't have any nitrogen in it? (Except the backbone, and that also could use replacement...)  I'm sure you can guess where I'm thinking that would be useful! Wnt (talk) 01:31, 14 November 2013 (UTC)


 * Personally, if anything found outside of earth with A/G/C/T as the genetic code, I'm going for at least common origin. Human's share 99+% with Chimpanzees.  And more than 70% with fruit flies.  With all the variety of life flowing from arrangements of 4 molecules, it seems non-random that the exact same 4 would be found elsewhere.  --DHeyward (talk) 23:15, 13 November 2013 (UTC)

Amount of information
How did the authors get to 384 bits that are supposedly encoded in the genetic code? It's of course log2(6464), but there aren't 64 possibilities for each of the 64 codons. The way I would calculate the number: There are 20 standard amino acids (for simplicity forget the 2 nonstandard ones, as they are encoded in a nonstandard way), and a stop codon, i.e. k=21 items that have to be encoded. And there are n=64 codes. We assign one of the k items to each code; that would make kn possibilities. But that number includes genetic codes that do not encode all k items, so we have to subtract the number of possibilities to encode less than k items, which is (k-1)n. The amount of information in bits is then

log2(kn - (k-1)n)

For k=21 and n=64 this is approximately 281 bits. Do I make a mistake? Icek (talk) 04:12, 14 November 2013 (UTC)


 * There's no reason to consider the multiple encodings of an amino acid to be equivalent when you're talking about embedding a message for genetic sequencers to discover. Then it's $$4^3$$ for 4 nucleotides and 3 slots per codon.  --Tardis (talk) 14:05, 14 November 2013 (UTC)


 * The genetic code table is (at most) a list of 64 things in some specific order. So there are indeed 64-factorial possible ways that the table could have been 'designed'...so there are about 1.27 x 1089 different possibilities.  If you numbered all of them in some meaningful manner, you could represent the table as an 89 decimal digit number...that's about 32 bytes.  Sadly, not quite enough to say "We're aliens from the planet Zarg!" in ASCII - but you could just about do it in something like radix-50 encoding.  But the problem with using the genetic code table like this is that there is no obviously unique way to number the possible combinations of the table.  We might decide that AAA, AAC, AAG, AAT, ACA, ACC, ACG...TTT is the logical order - but that depends on how some biochemist back 100 years ago randomly chose to name the four bases - and it depends on having an alphabet that has 'A' at the beginning and C-before-G.  This ordering would be weird to the Greeks, for example, because the order of the letters in their alphabet is different.  Our aliens might find some entirely different order to be more logical.  Hence, the numbering of the entries is entirely random - so you can't "store" data in the code without having either enough message to allow for decryption technologies to figure out the message (and then enough language to allow linguists to translate it!) - unless you  have some kind of manual and 'rosetta stone' telling you how to read the message.  Sadly, the length of message needed for decryption (let alone translation) would be a hundred times longer than the genetic code table - and any description of how the table is encoded would probably require about the same amount of storage.  So what we'd need to look for is a MUCH larger piece of alien information, describing how do decode that paltry 32 bytes of information.  That's just stupid!  Who writes, encodes and somehow stores, thousands of words of description to allow someone else to decode just 32 bytes of data?  Why wouldn't they just store the core message in an easier encoding scheme in the first place?


 * So, no...there is absolutely no conceivable way that we can 'discover' a message embedded in the genetic code...if there is a retrievable message, it's stored in the junk DNA in a highly redundant error-correcting code. That's why this is all bullshit...anyone with the slightest background in information theory would know this.  Sadly, a bunch of idiots trying to sneak "intelligent design" into evolution via the back door don't know that.


 * SteveBaker (talk) 18:21, 14 November 2013 (UTC)
 * Nitpicking: The number of bytes of information in a permutation of 64 things is log2(64!)/8 = approximately 37. Icek (talk) 01:05, 15 November 2013 (UTC)


 * The 64^64 part seems quite dubious, if that's accurate (I didn't check that bit). Because if you use that, you're assuming the aliens had 64 possibilities to pick from for each of the 64 values, i.e. 63 amino acids and a stop, but somehow it just worked out all 64 positions only needed to draw from 20-odd.  Or else they encoded the message so poorly that it's degenerated to that.  (I actually liked the notion that they engineered life with 63 AA's to begin with, but the data is looking stronger for a gradual evolution even in early history, not a kit from the gods) Wnt (talk) 00:54, 15 November 2013 (UTC)


 * (edit conflict)
 * Now I am less sleepy than when asking the question, and my formula still seems correct to me. It would be more than 64! if the number of amino acids plus stop codon were higher than 24. To make that obvious to everybody, let's see it with smaller numbers: Suppose there are only 4 different codons, and 2 different amino acids (so 3 items to be coded for, including the stop codon):
 * If we encode 3 or fewer items (if we would not care whether there are fewer), then there are 3 possibilities for each codon, giving us 34 = 81 different genetic codes.
 * But we have to subtract the number of genetic codes that code for 2 or fewer items, and there are 24 = 16 of them.
 * So we have 65 possible genetic codes that encode exactly 3 items, more than 4! = 24.


 * Back to the real genetic code: The rationale behind the 384 bits of information seems to be that there could be up to 63 amino acids, and maybe a minimum of 20 or 10 or whatever number of amino acids. The number of possible genetic codes (that encode at least 10 amino acids) is 6464 - 1064 - 964 - ... - 164, and the logarithm of that number is still very very close to the logarithm of 6464.


 * But why would one actually stop at 63 amino acids? One could imagine many more possible amino acids. That's not the path the authors followed, but one would enumerate the different molecules in a systematic way, there would be a lot more information. The authors see the nucleon number of the side chain of an amino acid as significant part of the encoding scheme. Tryptophan has the highest side-chain nucleon number, 130. How many thinkable side chains are there with 130 or fewer nucleons? There are alone 872 saturated hydrocarbon side chains up to nonyl (see http://oeis.org/A000625 ). There are many more possible side chains if you include those that have double bonds, nitrogen, oxygen and sulfur atoms... one could look at it from this point of view, and find much more information in the genetic code (even if it's only noise).
 * Of course there would be biochemical constraints, e.g. a set of amino acids that doesn't include any hydrophilic side chain won't work.


 * Returning to the case of just 20 amino acids and a stop codon on 64 possible codons, I don't believe the numerology of the authors of the article that started this discussion (and I didn't really study the article to see where they get their probabilities from), but there could be information in it. There are many different coding schemes, but not too many reasonably simple ones, compared to the possible number of genetic code. If I take the order suggested in the article (by nucleon count), assigning Stop->0, Gly->1, Ala->2 ... Trp->20, and also order the 4 bases by nucleon count (CTAG), I can write numbers in base 21 in that way. Just for fun, encoding the value of the elementary charge in natural units (in other words: the square root of the fine structure constant) in the codons starting with C: CCC->Gly, CCT->His, CCA->Glu, CCG->Ala, CTC->Ile, CTT->Ile, CTA->Met, CTG->Stop (and maybe a few more Stops, the Stop codon means 0, and the few more may indicate end of number). And with 21 items to be coded for, one needs in principle only 21 reserved codons (in order to make sure that there is at least one code for each amino acid and the stop codon) and one then has 43 base-21 digits freely available (about 7e+56 possible numbers); and I bet there are not enough reasonably simple coding schemes and not enough "really important" numbers to make a hypothetical discovery of such a number insignificant (I don't see the authors claiming something like such a discovery).
 * Icek (talk) 01:05, 15 November 2013 (UTC)

Volcanism (Earth versus Mars)
(moved here from the Humanities desk)

Approximately when volcanism will stop on Earth as on March for example. tx4urtm --YB ✍ 00:31, 13 November 2013 (UTC)


 * If and when the earth's internal heat goes cold, and the planet is dead. ←Baseball Bugs What's up, Doc? carrots→ 00:45, 13 November 2013 (UTC)


 * I'm guessing that the OP means 'Mars'. Volcanism on Mars confirms that we have never observed any active volcanism on Mars or any other phenomena which imply volcanism there more recently than a couple of million years ago, so it may indeed be the case that Mars is no longer actively volcanic; in which case this is a reasonable question and Bugs' reply rather unfriendly. Having said that, I suspect that we simply don't know an answer. --ColinFine (talk) 01:09, 13 November 2013 (UTC)


 * As a practical matter, volcanism would probably slack off or stop well before the earth became totally a cold stone. ←Baseball Bugs What's up, Doc? carrots→ 03:50, 13 November 2013 (UTC)


 * (ec) A couple million years is short for geological time scales, so I wouldn't call that "volcanically dead", but merely dormant. The main difference is the size of the two planets, with Earth considerably larger than Mars.  Thus, it takes much longer to cool from it's formation.  Another big factor is that the Earth has a large moon, which causes significant tidal heating, while Mars only has smaller moons.  A minor factor is that the Earth is closer to the Sun, so gets more solar heating, but of course that's mostly at the surface, with little of it making it's way deep underground.  It's entirely possible that the Earth will remain volcanic until the Sun becomes a red giant. StuRat (talk) 01:31, 13 November 2013 (UTC)


 * You are wrong about the heating by lunar tides. The heating produced by them is minuscule as compared to radioactive and gravitational sources. 09:22, 15 November 2013 (UTC)


 * Do you have a source for that ? Also, by "gravitational sources" do you mean the residual heat from the formation of Earth, or something else ? StuRat (talk) 02:58, 16 November 2013 (UTC)


 * The Earth is denser than and has 10 times the mass of Mars, but less than 4 times the surface area. The dense radioactive elements in the Earth's core give off a lot of energy, and the energy is retained much longer than it was on Mars.   We are nowhere near cooling off as far as I am aware (I am sure someone will find an estimate, it may be after the sun goes red giant).  In fact, there's evidence there was a naturally occurring nuclear reaction in the crust of Africa at some point. μηδείς (talk) 01:28, 13 November 2013 (UTC)


 * YB seems to be a Francophone Quebecker, per his user page. In French, the words for "Mars" and "March" are identical, which probably explains the error in the question.  I wonder whether the song The Waters of March gave its name to the Dr Who episode The Waters of Mars. --Trovatore (talk) 02:10, 13 November 2013 (UTC)


 * Yes, I had to go through my list of the 12 joueurs de hockey de l'année québécoise to make sure I'd understood: Gardiner, Trottier, Messier, Lafleur, Lemieux, Dionne, Joliat, Chabot, Primeau, Pilote, Perrault, Parent. I hope our Canadian friend understands this was très amusant, et tout en plaisantant. μηδείς (talk) 02:43, 13 November 2013 (UTC)

As Medis points out, the earth is 10x more massive - so if it started out with the same overall temperature as Mars, then it contained 10 times as much heat energy. So if they both radiated heat at the same rate, you'd expect it to take 10 times longer to cool off to the point where there would be no more vulcanism. However, there are a ton of variables here. With four times the surface area, you'd expect heat loss at four times the rate - and with ten times the mass, you'd wind up with maybe two or three times the amount of time. But there is more to it than that. There is the tidal effect of the moon - but also the core is heated by the radioactivity there - most of the heating comes from Uranium238 - and with a half-life of 4.5 billion years - we've already used up half of it. A NASA site I found says that over the past three billion years the core has probably cooled by a few hundred degrees...I'm not sure how many degrees it would need to reduce to eliminate vulcanism though. The core is somewhere between 5,000 and 6,000 degrees - so I'd guess that our vulcanism would easily outlast the life of the sun.

Remember, Mars and Earth are about the same age - 4.5 billion years. If we think that Mars was volcanic as recently as a couple of million years ago - and if the sun will swallow us up after 7.5 billion years - then volcanism only has to survive for three times what it managed on Mars. Between the larger mass and the tidal effects of the moon (plus closer proximity to the sun, insulating effect of the atmosphere, etc) - this doesn't seem unlikely.

Another way to estimate this is that the inner core is solidifying at a rate of about a half millimeter per year. A meter every 500 years - a two kilometers every million years - so if that rate were maintained, then it'll take 3 billion years for the whole planet to become solid. Of course that depends on that rate not changing - and there are many complicating factors that might speed it up or slow it down.

Bottom line is that we don't know - but it seems that the odds are very good that the sun will kill us before we run out of volcanoes.

SteveBaker (talk) 03:50, 13 November 2013 (UTC)


 * Again, the Earth is also denser than Mars. This may be a result of the collision event that created the moon, and threw off the less dense layers to create the moon and into space.  The denser parts are proportionally more radioactive, so it is likely an equal amount of Mars's mass was less heat producing than the Earth's mass from the beginning. μηδείς (talk) 04:01, 13 November 2013 (UTC)
 * The excess density is not due to any collision. It is a result of compression by the high internal pressure inside the Earth. The uncompressed density of the Earth is about 4.5 g/cm3. For Mars compression is negligible. Ruslik_ Zero 09:22, 15 November 2013 (UTC)


 * Note that Mars' core is still under investigation but is believed to be liquid .  I've even seen speculation that it is liquid, but at risk of freezing in the future .  This seems of great significance for several reasons, foremost being that it appears that Providence has apparently reserved an opportunity for the planet to form a magnetic field just as its water and carbon dioxide are warmed up into an atmosphere by the brightening sun, so that rather than being rapidly stripped by the solar wind, its future oxygen atmosphere might remain for billions of years.  But of course, that is very speculative indeed! Wnt (talk) 07:08, 13 November 2013 (UTC)


 * I suspect that Earth (core) cooling is not primarily limited by radiation, but by heat transport from the core to the surface. Notice how easily the surface changes temperature by 20 K or so between day and night, or even 40 K between summer noon and winter morning - still a geologically trivial time. Part of the internal heat transport is done by convection (which is reasonably efficient), and part by conduction (which is not). Since the diameter of the Earth is (slightly less than) 2 times that of Mars, its core is much better isolated - even if he material is not exactly styrofoam, 3000000m thick rock walls should meet any building standard for energy efficiency ;-). --Stephan Schulz (talk) 07:37, 13 November 2013 (UTC)


 * Our article Future of the Earth says (with a source) that plate tectonics will cease on Earth about 1.1 billion years from now, so I'd imagine that vulcanism would cease in roughly the same time frame. Of course, all such estimates are extremely speculative. Deor (talk) 19:46, 13 November 2013 (UTC)


 * I doubt that assumption. The plates can fuse while the Earth still has a substantial molten core, which continues to produce volcanoes, black smokers, etc. StuRat (talk) 20:13, 13 November 2013 (UTC)


 * Something like this seems to have happened with the planet Venus, which, if I remember from PBS in the 90's, is believed not to have any major tectonic or volcanic activity for 600 million years--but when it last did, a good third or two thirds of the planet was resurfaced in one giant burst. μηδείς (talk) 02:46, 14 November 2013 (UTC)


 * Are you aware that the molten lava released in volcanic eruptions doesn't come from Earth's core? It comes from shallower sources in the mantle. — Preceding unsigned comment added by Dauto (talk • contribs)


 * One should sign one's corrections. μηδείς (talk) 23:12, 14 November 2013 (UTC)

Is it theoretically possible for machines/vehicles to float like they do in sci-fi stories?
I'm thinking like the squids or ships from The Matrix, where they seem to employ some poorly-defined electromagnetic way of levitating. Is this within the realm of possibility? And if so, how could it actually be utilized, and what kind of energy output would it require? Goodbye Galaxy (talk) 15:21, 13 November 2013 (UTC)
 * Yes, our article on levitation describes the methods that have been used. Magnetic levitation is probably the most useful method. Looie496 (talk) 15:35, 13 November 2013 (UTC)


 * (ec) No.
 * Blocking, cancelling or negating gravity doesn't seem to be possible without resorting to some fairly ridiculous tricks involving entirely impractical things like neutron star material and such. So we're left with something that can exert a force equal in magnitude and opposite in direction.  This is done routinely with "magnetic levitation" - such as in the Transrapid train system in Germany which "hovers" 15 centimeters above the track.  The problem with that is that you need some magnetic (possibly electromagnetic) surface beneath the vehicle - so you're limited to things like trains that run on tracks of some kind...although perhaps you could consider cars on a road.
 * Another possibility is to use Newton's third law (every action has an equal and opposite reaction) - which implies some kind of downward thrust - such as in a hovercraft or a Jet pack or (my personal favorite) water-powered jet packs. But you're not going to get the kind of quiet, passive floating of a Star-Wars Landspeeder - and it takes continuous use of energy to keep the machine off the ground.
 * One fun thing that a lot of people talk about in this context is an Ionocraft - which supposedly uses an electric field to levitate - but which is in fact just a kind of hovercraft with a rather interesting way of creating an airflow. They are impractical beyond a very small scale device.
 * Our article Levitation explains a few other mechanisms - but mostly they only work with incredibly tiny objects and with massive expenditures of energy.
 * Short answer: No.
 * SteveBaker (talk) 15:46, 13 November 2013 (UTC)


 * I'm guessing it's far too weak, but could the earth's magnetic interior not function as a global "track" for (electro)magnetic levitation? Goodbye Galaxy (talk) 16:00, 13 November 2013 (UTC)


 * Yes, too weak. And if we could create a magnetic field strong enough to levitate a large vehicle using the Earth to oppose it, it would rip apart any metal ferromagnetic objects nearby. StuRat (talk) 19:28, 13 November 2013 (UTC)


 * The problems mentioned by SteveBaker can be dealt with by some advanced civilization as follows. The spacecraft uses photon rocket for propulsion, when it lands on Earth, the aliens don't want the emmitted gamma rays to sterilize the local environment. A way to do that is to use an extremely strong magnetic field to convert the gamma rays to axions via the Primakoff effect. Count Iblis (talk) 16:05, 13 November 2013 (UTC)


 * I think that some fairly mundane mechanisms should not be written off out of hand. A means of applying force to surrounding air or the ground is needed, and the array of electromagnetically mediated mechanisms available is already surprisingly large.  One should also not assume that it should look exactly as portrayed in the stories.  Otherwise, someone a couple of centuries ago might similarly have answered the question of whether self-propelled horseless carriages were improbable sci-fi with a categorical "no". You need to allow for what might yet be discovered. Plausible mechanisms might include somehow maintaining a higher pressure under the craft: essentially a hovercraft with a perfect skirt.  With mechanisms such as plasma windows, ionic thrusters and electrohydrodynamic thrusters, it is easy to imagine systems that could plausibly confine such a pressure without physical skirts, albeit at the cost of some air movement and perhaps some pyrotechnics.  Another mechanism might involve properties of a prepared surface (say, in a city, like we already do) with suitable dielectric, conductive or magnetic properties.  If we can already levitate graphite and frogs using a static magnetic field against full gravity, it seems we should be saying "maybe" rather than "no".  —Quondum 16:32, 13 November 2013 (UTC)


 * Also, a tethered vehicle provides some additional options for levitation, since it can now have unlimited electricity delivered via the tether, and can be made more stable than a free-floating vehicle. This could work well with an ion engine, for example, as mentioned previously.  The tether could also be connected to a ground vehicle, so the entire apparatus can move together.  So, you could have a nice floating observation car above a train, for example. Just watch out for those tunnels ! :-) StuRat (talk) 19:48, 13 November 2013 (UTC)


 * I would not say impossible, because theoretically all you need to levitate a vehicle by action-reaction is to have some particle which is capable of passing through air, water, and perhaps solid matter (terahertz light, X-rays, neutrinos, etc.) but which can be bent around and sent back the way they came by human agency (like many other particles). For example, you could have a couple of focussed neutrino mirrors (I mean corner reflectors) on the bottom of your car which communicate (via neutrino, no doubt) with broadcast stations in several cities on the far side of the planet, which send tremendous beams of neutrinos back at them.  The neutrinos, of course, are caught by mirrors at the station and reflect back to you again.  Only trouble with those flying chariots is that they get dicey if they fly too fast and can't hold to their projected trajectories, on account of the slow speed of light. Wnt (talk) 19:58, 13 November 2013 (UTC)


 * Do you have a link to such a particle, Wnt, or are you just using sounds with no relation to reality and pretending they are words? μηδείς (talk) 22:50, 13 November 2013 (UTC)
 * Really no need for that tone. I was amused by his whimsical idea. 86.134.51.230 (talk) 23:34, 13 November 2013 (UTC)
 * He said theoretically, a word which should be used carefully. :) Besides, I did describe a particle, a neutrino, and a design; we just need to build some of those pesky mirrors to reflect them.  I'm not an expert, but I think that's pretty basic subatomic engineering.  If you simply make a macroscopic quark-gluon plasma, would that be enough to start refracting neutrinos?  Or if you can get some fine control over the structural changes in nuclei so that you can try to shift them directly to some sort of extended unified planar structure that might be in an island of stability?  Admittedly, there I'm speculating, but what I know for sure is that if you can reflect neutrinos you ought to be able to levitate stuff.  (Note this is essentially a variant of beam-powered propulsion, and if the device fails to focus and recover essentially all of the neutrinos, it is very expensive.  The consequences of irradiating stray peasants on the ground with the neutrino beam are also not very clear to me, though surely better than with the laser system ... call it "research".) Wnt (talk) 01:17, 14 November 2013 (UTC)


 * Thanks, Wnt. My point, IP 86, was nothing personal with Wnt.  But when you start positing imagined entities with no evidential basis whatsoever, like string theory, there's no way to evaluate it.  This being the ref desk, we should at least identify when we have no evidence for something.  I post all sorts of things like my reference to ice-fishing arctic tribes.  But I wouldn't be insulted if someone challenged me on that. μηδείς (talk) 02:43, 14 November 2013 (UTC)


 * One problem with levitating vehicles that's usually ignored in sci-fi is their tendency to glide down any sloped surface, no matter what kind of levitation is being used. I could imagine some compensating computer-controlled thruster technology, for example if the levitation is handwaved as gravitational cancelling, the authors might explain the compensating thrusters as graviton emitters, or with electromagnetic levitation, as tractor beam emitters which mainly keep the emitting vehicle in place rather than pulling foreign objects around. But the number of authors who actually address this issue, which has been amply observed with real-life hovercraft, is very low. Most sci-fi just pretends the problem didn't exist. - ¡Ouch! (hurt me / more pain) 06:31, 15 November 2013 (UTC)


 * How can a problem with non-existent technology be ignored? Helicopters seem to be the closest thing and they figured out how to compensate for ground effect.  Also, unless the terrain varied in mass density, the gravitational field of the earth can be modeled as a point.  The force necesary to counteract is a radial square law from the center of mass.  The terrain contour matters little to gravity. Anything "anti-gravity" shouldn't slide.  Quite the opposite.  Sudden changes in countour while moving would cause the constant force anti-grav levitation to crash into it, not follow it.  --DHeyward (talk) 08:53, 15 November 2013 (UTC)

Aqueous cream as a moisturiser?
For many years I've used aqueous cream as a moisturiser. I don't have any medical condition that requires me to use it so this isn't a medical question. I started using it around age 19 when a friend said that I already had lines around my eyes when I smile (I don't know but I'm guessing that would still have been the case if I had used Oil of Olay since I was six; I think it's just the arrangement of my face). For reducing the rate at which my skin appears to age, does aqueous cream make any sense? — Preceding unsigned comment added by 129.215.47.59 (talk) 18:42, 13 November 2013 (UTC)


 * Moisturizers alone don't actually reduce skin aging. In the US, all the advertisers can say is it "reduces the appearance of fine lines".  This effect only lasts as long as the moisturizer is on the skin.


 * To reduce skin aging, the most important factor is avoiding UV light damage. This means avoiding sunlight, wearing a hat, or using sunblock.  Now, some sunblocks also contain moisturizers, so, in that sense, a moisturizer could reduce skin aging. StuRat (talk) 19:21, 13 November 2013 (UTC)


 * According to this story the answer is 'no' as the scientist said that this was the first time that it had been scientifically proven that a cosmetic face cream could effect a "clinically discernable improvement" in wrinkles. Richerman    (talk) 23:56, 13 November 2013 (UTC)


 * Aqueous cream is about as effective at prevening signs of ageing as any other face cream in the market - and a darn sign cheaper. Props to you!