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

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.

April 13

Minimum population required for the survival of a race.

The article on Angam Day currently states:

Upon eclipsing a population of 1,500, a number considered to be the minimum required for the survival of a race, Angam Day was declared.

Does a race really need a minimum population of 1,500 to survive? If true, why? If false, how many people does a race need to avoid extinction? Thank you in advance. --190.19.69.254 (talk) 04:33, 13 April 2013 (UTC)

Hmmmm. That's a pretty ugly part of that article. I would much prefer that the term ethnic group had been used throughout. Race is an unclear word with unfortunate connotations. The article Nauruan people, linked from Angam Day, avoids the term race completely. I note that the claim of 1,500 being the minimum number for a race to survive is unsourced. That's not good, and only makes things worse. I went hunting and couldn't find a source myself. You've got me thinking about what we should do with the article. HiLo48 (talk) 04:54, 13 April 2013 (UTC)

Reads like unmitigated bullshit to me. What makes an ethnic group (or race) a distinct unit is a shared culture. I'm not sure there's any arbitrary lower or upper limit on that, merely that the group has a certain level of cultural cohesion and distinctiveness, and I certainly can't find any literature that indicates that there's some official or "scientific" or whatever reason that would indicate that 1500 is some magic number. --Jayron32 05:00, 13 April 2013 (UTC)

The article clearly states that the number 1500 came from Brigadier General Griffith, the Australian administrator. It does not claim that the number has any deeper validity. It doesn't cite a source for that, but I don't see anything implausible about it. Looie496 (talk) 05:17, 13 April 2013 (UTC)

The article makes that assertion, but that assertion (nor the quote from Griffith) has any sources to support it. The big issue is if Griffith said directly, "if the Nauruans were to survive as a race, the population should be no less than 1,500.", then we would need to put quotes around that, as I have done, and provide a footnote to make it clear where the quote comes from. Without sources, the statement really shouldn't stand, however. --Jayron32 05:22, 13 April 2013 (UTC)

For what it's worth, I've managed to work out that the person referred to here is Thomas Griffiths (general). I'll add a wikilink to our article (and fix the spelling). Looie496 (talk) 16:46, 13 April 2013 (UTC)

The problem is extremely difficult for human populations because "survive" can mean "not all die" or "survive as a distinct culture". But if a group of people were cut off from the rest of humanity for some considerable time (either physically cut off - or culturally unwilling to take mates from outside of their cultural group) - then genetics and inbreeding would be the problem. The considerations in our Minimum viable population article would take effect. For large vertebrates, the accepted number is between 500 and 1000 individuals if the population is carefully managed (ie scientists decide who breeds with who!) - and more like 4,000 if not carefully managed. On that basis, you might think that 1500 people isn't enough - but it's in the right ballpark, and this isn't an exact science. But that kind of complete elimination of mating outside of that cultural group seems unlikely.

A classic example of how this can happen is the Pitcairn Islands - which was uninhabited until six men, eleven women and a baby arrived there after the infamous "Mutiny on the Bounty" incident in 1790. Since the island was hundreds of miles from any other land - and not of much interest to anyone, the hapless mutineers were left alone - and (inevitably) the population steadily grew to around 600 people. This is widely accepted as being the most inbred group of humans in the world. But even with such extreme in-breeding, the occasional outsider has managed to add enough genetic variation to keep the population viable - and according to most studies, they aren't suffering too badly.

Who says they aren't suffering too badly? Law enforcement in the Pitcairn Islands. Seems like they were raping children with impunity there till 1999. Sagittarian Milky Way (talk) 02:57, 15 April 2013 (UTC)

So we can conclude that there isn't some magical cutoff at 1500 people...at least not genetically. SteveBaker (talk) 13:16, 13 April 2013 (UTC)

Since the entire human race has an effective population size of probably around 5,000 (if not less) then no. Culture does weird things though. ~ Amory (u • t • c) 17:26, 15 April 2013 (UTC)

someone wrote a program to find that out. i just can't remember who, or the program name, or when. 70.114.248.114 (talk) 03:48, 14 April 2013 (UTC)

Is 2 centuries a long enough test to falsify the 1,500/4,000 individuals inbreeding rule? Sagittarian Milky Way (talk) 02:57, 15 April 2013 (UTC)

Seasonality of human hair growth

My friend, in her early seventies, is convinced that her hair grows more quickly in the Spring and more slowly in the Winter. Her evidence for this is the much shorter length of time it takes for white roots to appear after she has dyed her hair. (I have no reason to doubt her eyesight.) This sounds like folklore to me but I can find nothing to indicate that anyone has ever studied the matter. Thank you for your help. — Preceding unsigned comment added by 109.12.63.61 (talk) 07:24, 13 April 2013 (UTC)

This is an interesting question, as it is one of the few topics on which I think the written word does not match reality. If you google (google "rate of hair growth") or search in textbooks, you find almost all estimates lie in two categories: Those which say hair grows at a constant rate of 0.5 inch (13 mm) per month no matter what (the Wiki article human hair growth is one example), and those that say it varies depending on age, health, intake of certain vitamins, race, and a few other minor factors and is between 7 mm and 20 mm per month, which seems much more likely. What seems difficult to find is any refrence that says it varies depending on where on the head it grows from, and there's no reference that says it is seasonal, depite the fact that it most definitely IS seasonal for other mammals. Hairdressors are taught that it grows at a constant rate.

However, in my experience, it grows considerably faster on the back of my head and neck, and a lot slower on the top of my head. It also grows quite a bit faster in autum and slowly during winter. This is reflected in when I need to go to the hairdressor. I'm also in my seventies. I'm sure that it grew faster when I was younger, but suspect that the variation in growth rate was hidden when I had a younger thicker hair.

In my experience, cutting hair very short and cutting frequently casues it to react by growing faster. However hairdressors are taught that that is a fallacy. I had a friendly argument about it once with a hairdressor friend, so I experimented over a two year period, keeping records. I was right. However, I live in a hot climate, and cutting hair short probably raises skin temperature. If I lived in a cold climate, it may well be that cutting hair short would lower the skin temperature and therefore the growth rate. This suggests a conflict between two factors: skin able to grow hair faster when it is warm during warm weather, and the evolved capability of mammals to grow a thick winter coat in time for winter.

Wickwack 121.215.67.60 (talk) 08:41, 13 April 2013 (UTC)

You're in your seventies and it still grows on the top of your head? I'm jealous. HiLo48 (talk) 08:45, 13 April 2013 (UTC)

Don't fret too much. It grows so slowly on top, and so thinly, it might as well not bother. But I still need to have it cut from time to time. Wickwack 121.215.67.60 (talk) 08:54, 13 April 2013 (UTC)

Sorry Wickwack, although I defer to your superiority in age, I'm afraid hairdressers are taught that hair grows at different rates according to both position on scalp and time of year. Now if only I could find a reference for it. (I know they're taught it as I sat in on some of the sessions for the hair students when I was a therapy mature student about 5 years ago.) --TammyMoet (talk) 13:45, 13 April 2013 (UTC) I found this study which looks like a pretty neat exemplar of how to do citizen science! --TammyMoet (talk) 13:51, 13 April 2013 (UTC)

Very good Tammy! I only wish that age did bring superiority! Perhaps hairdressers are better taught in your country (the UK if I remember right), or perhaps knowlege has been updated since the hairdressors I've known did their training. I've found them adamant that it grows at a constant rate. The study you found has some limitations though - for instance they assert that temperature has little to do with it, but haven't proved it. One reasonably expects that hair growth rate is dependent on skin temperature, but the skin temperature relationship to climate would be confounded by the human practice of adjusting clothing to suit the season. They tested only one person. Did they check that she had the same diet/nutrition/calorific intake throughout? Wickwack 121.221.31.213 (talk) 14:34, 13 April 2013 (UTC)

You're indeed right to highlight the role played by diet and nutrition in hair growth: the study also failed to take blood samples to check on the endocrine status of the subject as I've found elsewhere that thyroid status affects hair growth as does oestrogen/progesterone status in women. It's not a complete study by any means, but an example of what can be achieved by someone wishing to follow the scientific method and make a difference. --TammyMoet (talk) 15:37, 13 April 2013 (UTC)

ques on tetrahedral and octahedral voids

A compound formed by two element a&b the anion b are located at the corner of the cube and face centres whereas cation a occupy all the tetrahedral void what is the simplest formula of the compound? — Preceding unsigned comment added by 70.39.184.248 (talk) 09:59, 13 April 2013 (UTC)

Please do your own homework.

Welcome to Wikipedia. Your question appears to be a homework question. I apologize if this is a misinterpretation, but it is our aim here not to do people's homework for them, but to merely aid them in doing it themselves. Letting someone else do your homework does not help you learn nearly as much as doing it yourself. Please attempt to solve the problem or answer the question yourself first. If you need help with a specific part of your homework, feel free to tell us where you are stuck and ask for help. If you need help grasping the concept of a problem, by all means let us know. Roger (Dodger67) (talk) 11:59, 13 April 2013 (UTC)

What is a microstate?

Entropy is by definition $-k\sum _{i=1}^{\Omega }P_{i}\ln P_{i}$ where k is Boltzmann's Constant and $P_{i}$ is the probability of a particular state given our limited knowledge of the system. But, how is state defined? What fully characterizes a physical state?

150.203.115.98 (talk) 10:26, 13 April 2013 (UTC)

It is the complete information about the system. Suppose you have two systems in different physical states, then that means that there exists an experiment that you can do on the systems which will have a different outcomes, at least statistically. If on the other hand the two systems are in the same physical state, then no experiement can have (statistically) different outcomes.

You can then try to define the state of a system by specifying a list of experiments and what the outcomes of each of these experiments should be. In quantum mechanics, the outcome of a measurement is, in general, not pre-determined. Now, you do one experiment and then immediately repeat that same experiment you will get the same outcome (in the limit that the time between them is zero). If you do experiment A and then experiment B, you can compare that with doing this the other way around. If there is no difference, then measuring B will give you independent information than you get from measuring A. You can then add another such experiment C that doesn't interfere with A and B. If you try to make this list of experiments larger and larger, you will find that some point, you can no longer make this list larger. You will then have what is called a "complete set of observables" for the system. Then the state of a system is completely defined by specifying a list of the outcomes for each of the obsevables contained in such a list. The outcome of any other experiment can be predicted (in the form of a probability ditribution over the possible outcomes) when the state is specified. Count Iblis (talk) 12:27, 13 April 2013 (UTC)

That's really interesting, and I can see the article Complete set of commuting observables now. A state is basically characterized by its measurable properties, which makes sense, especially from a pragmatic point of view. It doesn't really make sense to call two states "different" if there doesn't exist any experiment, even in principle, to distinguish between them.

Now, suppose you have a physical system, and you want to change its state. How would you do this, in general?

150.203.115.98 (talk) 15:09, 13 April 2013 (UTC)

Changing state is already described in "general" terms. In various branches of physics, state change is described mathematically; for example, in quantum mechanics, physicists talk about "applying an operator" to the system. This is about the most precise and generalized way we can describe changing state for one variable, subject to physical constraint. Unfortunately, such generalization tends to be a bit obtuse, and some people find these decriptions difficult to intuitively connect to experimental physical systems. This is because the specific way you change a particular state for a particular system depends on what physical process corresponds to the change of that state variable. In other words, we write a very clean mathematical formalism to describe an operator; but it is not always immediately evident how that operator corresponds to an experimental procedure.

For example, in atomic physics, one state variable models the energy-level of the electron; you can change that energy level by (in general) applying an energy transform operator to the electron. In specific cases, that operation manifests as compton-scattering; or shooting laser photons at the atom; or colliding the atom with other warm atoms (thermal excitation); and so on; in each case, practical details arise; it may be impractical to change one state-variable without affecting thousands of other parameters. Nimur (talk) 16:57, 13 April 2013 (UTC)

An example of a particular state is an associated energy level, such as the ground state of a partice. Plasmic Physics (talk) 12:28, 13 April 2013 (UTC)

In thermodynamics, a microstate describes the position and momentum of every particle. It's the full description of the system. For example, with N particles, "particle 1 is at (3,2,5) and moving at 3/ms north, particle 2 is at (4,5,1) and moving at 9 m/s east, etc" counts as a microstate. A macrostate is something you care to measure on a macroscopic scale. For example, what's the density of the gas? How uniform is the gas? What's the velocity distribution of its particles?

There is an inherent arbitrariness in the definition of a microstate. If particle 1 is moved by 1 nm, does that count as the same microstate, or a different one? How about 0.0001 nm? First, this doesn't actually matter--entropy is defined as the logarithm of the number of microstates, so changing the precision with which you distinguish microstates only changes entropy by a constant. Differences in entropy would stay the same. Second, quantum mechanics sets a fundamental limit on how accurately you can measure phase space. dx*dp cannot be smaller than Planck's constant, because the uncertainty principle says you can't simultaneously measure position and momentum more accurately than that.

Over time, systems tend towards the macrostate with the most microstates. For example, their velocity distributions tend to become the Maxwell distribution. "A uniform gas" accurately describes many more microstates than "a gas in a container where half the container is empty". This isn't surprising--of course if every microstate is equally likely, and the particles randomly choose one, that microstate would most likely correspond to the most likely macrostate. This tendency towards the most likely macrostate is the second law of thermodynamics. --140.180.240.67 (talk) 18:34, 13 April 2013 (UTC)

Does that agree with the definition Count Iblis gave? And, don't you need to know other properties of the particles such as their mass, charge etc.? And

150.203.115.98 (talk) 20:10, 13 April 2013 (UTC)

You do. I was giving a simplified classical view of an ideal monoatomic gas, which I think is helpful for intuition. For such a gas, all the molecules can be considered point sources with no charge and the same mass. Sorry for not stating this explicitly. --140.180.240.67 (talk) 20:20, 13 April 2013 (UTC)

Eyesight

Can a teenagers social development be affected by poor eyesight? — Preceding unsigned comment added by 176.250.139.80 (talk) 11:15, 13 April 2013 (UTC)

Yes. (Strange question.) HiLo48 (talk) 11:21, 13 April 2013 (UTC)

Obviously anything that affects communication affects social development. However, what do you mean by "poor eysight"? Deficiencies that can be corrected with spectacles should not affect it, except to the extent that wearing spectacles may temporarily affect self esteem or may limit playing in team sports. Wickwack 121.221.31.213 (talk) 11:30, 13 April 2013 (UTC)

If you hadn't just admitted to being in your 70's, calling glasses "spectacles" would have told me. :-) StuRat (talk) 04:04, 14 April 2013 (UTC)

One tries to write for an international reader here. "Glasses" has multiple meanings, (e.g., magnifying glass) while "spectacles" has only two, and it's obvious which of the two meanings applies here. Wickwack 121.221.215.92 (talk) 09:49, 14 April 2013 (UTC)

...And of course it's a matter of degree. If the person's eyesight was sufficiently good that glasses were only needed for some specific task like driving or reading - then the effect should be very small indeed. On the other hand, I know an adult who's corrected eyesight is so bad that he can only read things a few inches from his nose - even with thick, heavy glasses. He can't drive, play video games or watch TV at all. Since those are all major teen social activities I'm sure that must cause significant social impairment in teenagers. On the other extreme, my son is mildly red/green color blind and sailed through most of his teenage years without even knowing it - which implies that there was no social impairment whatever. So the answer to our OP is obviously: Yes, it's possible for extremely poor eyesight to cause serious social impairment - but mildly poor eyesight does not cause impairment. It's not a very useful answer without knowing details about a specific case - but in that case we'd be unable to offer advice because it would constitute "giving a medical prognosis" - which we're strictly prohibited from doing. SteveBaker (talk) 12:48, 13 April 2013 (UTC)

While I don't disagree it's a matter of degree and quite variable depending on the imparity, I don't think it's as simple as you suggest particularly since the question was about 'social development be affected' which can be very broad. For example, depending on the person's interaction with peers, teachers and parents, I imagine it's entirely possible they will go through their teenage years with a mild myopia without really realising it, yet it will still affect them (e.g. they will avoid certain activities and be less good at them then they would be if it were corrected). Of course the same could happen for a whole host of other reasons. As Wickwack mentioned, even with perfect correction with glasses it can still have effects since it can affect self esteem, interaction with peers and in some cases the presence of glasses makes it more difficult to participate even without considering those effects. (Again these can also arise for a whole host of other reasons.) Contact lenses can help, but have their own issues. And depending on factors like whether the person is going to need glasses in addition (for times when they don't want to or can't where contacts), the level of the correction required (particularly astigmatism), how frequency the prescription changes and the price in the locale, contact lenses may not be financially possible. Of course for some people and in some places, corrective lenses of any kind may not be possible for financial reasons. Nil Einne (talk) 14:21, 13 April 2013 (UTC)

http://www.youtube.com/watch?v=ls2lC7DQFMI --Digrpat (talk) 23:06, 13 April 2013 (UTC)

In the case of uncorrected poor vision, there can be an interesting split in school. Some will move to the front of the classrooms, in order to see what the teacher is writing on the board, and tend to become "teacher's pets", while others will give up and fall behind in school. Either path could potentially negatively affect their social development. StuRat (talk) 04:12, 14 April 2013 (UTC)

Last spacecraft launch to orbit with *one* astronaut?

When is the last time that a spacecraft launch occurred which made at least one orbit that only had *one* astronaut in the capsule. If this was a private company, when is the last time that a National space agency did so. In either event, was the last launch in the Mercury program the last time the United States did so?Naraht (talk) 11:35, 13 April 2013 (UTC)

The most recent manned space flight with a single crew member seems to have been Shenzhou 5, the first manned space flight of the People's Republic of China, in 2003. Gandalf61 (talk) 12:23, 13 April 2013 (UTC)

The last U.S. single astronaut mission was Mercury-Atlas 9 in 1963. The last Soviet one may have been Soyuz 3 in 1968. No private company has yet done orbital flight. The company SpaceX is trying to do the first private manned orbital spaceflight in mid-2015 but it will carry a crew of three. Rmhermen (talk) 15:30, 13 April 2013 (UTC)

Cost of RNA vs protein

Hi Ref Desk,

I'm wondering what the energy cost to cell is for making an RNA molecule vs the protein for which it codes. Intuitively, I would say that the protein is much more expensive, but a study ([1]) suggests that RNA costs 5x more energy than a protein (49.3 phosphate bonds per nucleotide (x3!) vs 30.3 per amino acid)! I can't find any other estimates to compare this to - any tips?

Cheers,

Aaadddaaammm (talk) 12:16, 13 April 2013 (UTC)

It's not common for someone to come to us with a good reference in hand - thanks! You probably are aware already, but I should say a key thing to bear in mind here is the difference between total synthesis vs. polymerization of precursors. RNA and protein are both going to get degraded sooner or later back to amino acids and XMPs, so the total synthesis doesn't really apply except in growing populations of cells.

So far as I recall, the basic ratio for polymerization should be just one pyrophosphate released per charged amino acid in aminoacyl tRNA synthetase, with that energy being sufficient to power the ribosome, but EF-Tu/eEF-1 contributes another phosphate bond to improve accuracy. (Note that pyrophosphatase makes the PPi effectively count as two bonds split) Other elongation factors exist, for example EF-G at the end of the protein. I can't rightly recall at the moment where the other ~P the author of your reference comes from during elongation. So we're talking about 3 ~P I can count (ignoring the one extra at the end), 4 according to the author, plus parts. By comparison, I would count the "cost of polymerization" in the RNA as being a PPi per bond created (i.e. XTP to XMP). So to me the ratio that would seem to matter in steady-state metabolism would seem more like 2:3 with the amino acid being only a bit cheaper than a codon.

Anyway, what's interesting to me about this is that the selective pressure against having a high level of mRNA and protein per cell seems like it is going to be much less for an organism that remains in a stable state for a long period of time than one which is is exponentially growing. If an organism's whole ecology is geared toward putting on weight and splitting, that seems like it ought to lead to different overall gene expression level than if it is trying to hide for a long time and avoid getting eaten. Hmmm... Wnt (talk) 19:48, 13 April 2013 (UTC)

It's worth bearing in mind that a single mRNA molecule, once transcribed, can be translated multiple times. (Heck, multiple ribosomal complexes can attach to the same mRNA simultaneously, allowing translation of multiple copies of a protein at the same time from a single mRNA: [2].) What this means is that the energy 'overhead' cost of one mRNA molecule can – potentially – be 'amortized' over many, many translated protein molecules. In other words, even if transcription is 'expensive' on a per-base basis, it still accounts for a relatively small part of the entire protein biosynthesis energy budget because mRNA is reusable. TenOfAllTrades(talk) 03:23, 14 April 2013 (UTC)

Spatial Encoding

Since my question has been removed I post it here.

What are the typical computer video compressions algorithms names which use the human retina approach?2A02:8422:1191:6E00:56E6:FCFF:FEDB:2BBA (talk) 21:09, 13 April 2013 (UTC)

(I see where this was removed from Talk:Retina) - unfortunately, it is possible that this still isn't the best place, that Wikipedia:Reference desk/Computing would get you better answers. I'm certainly not aware of any video compression that attempts to follow the retina's approach literally. Figuratively... I can't tell. For example, if I do a Google search for "video compression" "edge finding" I get [3], but I can't really say whether that method of finding dominant edges to align stereo pairs has any relation to the biological scheme at all; certainly it is not a direct copy.

My gut feeling is that the two circumstances aren't comparable because we accept a huge loss of resolution from our retina. We aren't aware of individual rod and cone inputs on their own. So any computer algorithm that worked just like the retina would be seen as a really crummy way to store an image, until such time as the images are sooo high res that we really don't care. I think... Wnt (talk) 22:01, 13 April 2013 (UTC)

Why do you say there's a huge loss of resolution? As far as I know the eye's resolving power is pretty close to the cone spacing. -- BenRG 00:15, 14 April 2013 (UTC)

Our vision isn't nearly as good as it seems. We have lots of blind spots where we looked at something bright and burnt out a few cons and rods, or a blood vessel covers them, etc. Why don't we see black spots all over our field of view ? Out brain does an amazing job at filling in the missing spots. Unfortunately, this also means we aren't always seeing reality, but sometimes just what our brain makes up to fill in the gaps. StuRat (talk) 07:09, 14 April 2013 (UTC)

Perception is reality. 202.158.66.204 (talk) 09:56, 14 April 2013 (UTC)

Thanks for the quick answer I had thought to this. There are projects for getting 60TB per hard drive for the end of the decade. As it did for The GB step,there are reason to full the space. there will exist camera that would produce 25GB images with actuals compression techniques.

I can't imagine diffuser aren't planning the result for video download.I'm sure the algorithms are already written.

By the way I was thinking about the retina side only (no real care about the second eye) and not the decompression methods.

The method of "decompression" is always varying from person to person and time (optical nerves and human brain ). 2A02:8422:1191:6E00:56E6:FCFF:FEDB:2BBA (talk) 22:28, 13 April 2013 (UTC)

My knowledge of the spatial encoding in the optic nerve is based entirely on having read the article section just now, but it looks very similar to the wavelet transform, which is used in JPEG 2000, for example. The most popular video compression formats right now (e.g. h.264 and all versions of MPEG) don't use wavelets, but do use the discrete cosine transform, which is closely related. All image and video compression is based in one way or another on human vision. Even "uncompressed" images represent colors as RGB triples, a representation that's closely tied to human standard color perception. -- BenRG 00:15, 14 April 2013 (UTC)

• I don't really see how the retina can be compared to any image compression algorithm. The huge difference is that the retina has quite a small high-resolution zone, called the fovea -- it is roughly the size of your fist held at arm's length. The most important type of compression the retina uses is to represent only this very small region at high resolution. Nothing like that could work for a computer representation, because there is no way of knowing what part of an image will be the focus of attention. (I'm oversimplifying a bit, but I don't want to flood the board with verbiage.) Looie496 (talk) 03:15, 14 April 2013 (UTC)

Plaster of Paris as a structural material yet it is relatively water soluble, what gives?

Calcium sulfate dihydrate, the main component of plaster of Paris, has a solubility of 2 g/L. Yet this material is used to make all sorts of structural materials. So what I am trying to figure out is why doesnt this stuff dissolve? Perhaps the plaster components are relatively crystalline and have a low surface area, but still I would think that after a few years or decades (see Borujerdis House as one of many antiquities made of P of P) of rainy weather that it would be corroded away. Or, maybe this material is not used for exteriors very much except in dry climates.--Smokefoot (talk) 23:53, 13 April 2013 (UTC)

Paster of Paris is never used as a structural material, not in my country anyway. The term "structural" denotes a material that is load bearing. Plaster of paris is only used in non-load-bearing and decorative applications - someting else underneath (brick, concrete, framework, etc) takes the load - load being either static & dynamic downward forces due to gravity and/or wind loading. Actually, the ability of moisture to wreck plasterwork can be considered a virtue. It lets you know you have rising damp, or a leak in the roof, well before real structural damage is done, and it is easily repaired at mimimal cost. Where used as suspended ceilings where it must sustain its own weight, it is re-inforced with cellulose fibres or other semi-woven material.

I know nothing about the Persian building you cited, but none of the inticate work shown is structural. There's nothing wrong with using plaster work to decorate ceilings and other inside surfaces - this is extremely common in modern western buildings. Despite the labelling on the images, I expect the exterior decoration (which is applied, not structural) is all water-resistant stucco, not plaster. Wickwack 121.215.41.163 (talk) 02:56, 14 April 2013 (UTC)

Oh, good points. I used the wrong wording (structural). My question is simpler - why would anyone decorate the exterior of any building with a material that is fairly water soluble. --Smokefoot (talk) 03:04, 14 April 2013 (UTC)

Read Whitewash. Also note that Persia had (and still has) a rather dry climate, so water solubility wouldn't have been a big problem there. 24.23.196.85 (talk) 03:38, 14 April 2013 (UTC)

I thought so too, but fortunately I checked before writing my first post. It turns out that their annual precipitation (680 to 1700 mm) is somewhat greater than the precipitation where I live (Western Australia), hardly dry, and you certainly would not use plaster on an external surface here. That's why I think that the external decoration is not plaster but stucco. One of the characteristics of whitewash is that is a non-durable paint (once popular in the US) requiring regular re-application. If you used it for the sort of decoration depicted in the OP's link, you'd soon end up with a smooth surface and no decoration. Wickwack 60.230.245.70 (talk) 06:29, 14 April 2013 (UTC)

I think that stucco and plaster of Paris are the same thing (despite what is stated in stucco), calcium sulfate, which is fairly water soluble, so now what? --Smokefoot (talk) 12:32, 14 April 2013 (UTC)

There are various recipes for stucco. Stucco can indeed be made with plaster of paris. It can also be made with portland cement, which makes it completely waterproof, but with a lower quality finish. It also gives an ugly grey colour, but that can be fixed with pigments. There are a number of other water resistant mixes, eg lime-based (as in mortar used to stick bricks together), and you can also use a combination of plaster of paris with cement to give a degree of water resistance. Wickwack 124.182.15.108 (talk) 12:50, 14 April 2013 (UTC)

The answer might just be Relative humidity. Homes which use plaster are usually heated (which lowers the average RH). In older homes that have old fashioned larder rooms, the plaster in them (if they are plastered at all) often become crumbly. Plaster needs to be fairly damp in order to disassociate like this. In the tropics, were relative humidity is high and a artificial heat is not required, then plaster is seldom seen. If it is used, then a lime based wall coating is probably leaning towards hydraulic lime and that is not plaster of paris.Aspro (talk) 20:02, 14 April 2013 (UTC)

• Could plaster of Paris be used in Lath and plaster construction? --Jayron32 04:26, 15 April 2013 (UTC)
  • Certainly. By the 20th century, that's what it mostly was -- it dries much faster than the older lime-based plaster, and has fire protection qualities. --jpgordon^{::==( o )} 14:53, 15 April 2013 (UTC)

In common usage it seems that term Plaster of Paris (CaSO4·2H2O) is used mostly for its pure form but it is also known as gypsum. Its pure form could be used for plastering Lath but its setting time might be a bit too quick for a plasterer to get a good finish over a large surface. Builders gypsum is almost the same thing but it is modified to give a longer working time (i.e., Plaster of Paris multi-phased by the addition of lime etc). Pure lime plaster on the other-hand takes very much longer to cure. The latter, is only needed today for re-plastering some old buildings (especially stone and mortar dwellings) that where built without a damp-proof course. So yes. the white stuff one can buy from art & craft shops called Plaster of Paris will do the same same job. --Aspro (talk) 15:32, 15 April 2013 (UTC)

Lath and plaster was, of course, only used for indoor work, mostly in the USA. Wickwack 120.145.63.92 (talk) 00:50, 16 April 2013 (UTC)