Wikipedia:Reference desk/Archives/Science/2015 November 23

Welcome to the Wikipedia Science Reference Desk Archives
The page you are currently viewing is an archive page. While you can leave answers for any questions shown below, please ask new questions on one of the current reference desk pages.

November 23

The difference between cavity and sinuses

Someone wrote on Facebook that the difference is that "the 4 sinuses contain only blood and limph but the cavity is empty space which organs occupy". Is that true? I tried to find any source for that claim and I couldn't. I would like to get some help to verify it. 92.249.70.153 (talk) 05:55, 23 November 2015 (UTC)

Maybe they were referring to the dural venous sinuses, which apparently are lined with lymph vessels [1]. However, our article lists more than four. --NorwegianBlue ^talk 08:59, 23 November 2015 (UTC)

Thank you. They're refereeing to the paranasal sinuses rather than to those sinuses that you mentioned that are not exist in the bones. 92.249.70.153 (talk) 09:29, 23 November 2015 (UTC)

Is a bone considered as an "organ"?

Is it every bone considered as an "organ"? according to what, we can determinate what is organ or what is not? Can it be that we have 206 bones that each one of them considered an organ? 92.249.70.153 (talk) 08:51, 23 November 2015 (UTC)

Well, organ (anatomy) gives a broad enough definition to make a bone a type of organ, and bone describes a bone as an organ. On the other hand, on this page a medical professional gives a more restrictive definition by which bones are not organs. --70.49.170.168 (talk) 09:14, 23 November 2015 (UTC)

bone marrow has a good claim to be an organ (although bone marrow is not totally separable from the bone itself), and it forms part of the lymphatic system. Smurrayinchester 09:26, 23 November 2015 (UTC)

But bon marrow is found in all the bones? 09:30, 23 November 2015 (UTC) — Preceding unsigned comment added by 92.249.70.153 (talk)

Does rust accelerate corrosion?

If an iron surface had some rust and then was coated with an effective isolating paste without removing the previous rust, will this rust continue even in the absence of humidity?--82.114.170.222 (talk) 09:08, 23 November 2015 (UTC)

Humidity doesn't cause rust, oxygen does. Water would accelerate the reaction by acting as a sort of catalyst. --Jayron32 15:30, 23 November 2015 (UTC)

Jayron is correct that water by itself doesn't cause rust because rust requires oxygen. But that would be in the more theoretical case of pure water. Most water found in the world has a bunch of dissolved oxygen in it - so water does cause rust as a practical matter. Notice how, for example, the brand new, rust-free Titanic, has rusted away, despite being kept a bazillion feet away from the air, deep under the Atlantic ocean.

The presence of rust does tend to accelerate additional rusting because it's a very rough material and it traps moisture better than smooth, bare metal. If you first remove all of the moisture, THEN coat it with something waterproof, you'll certainly slow down the rate of corrosion. But just painting over rust may not help much. That's why people generally try to remove all existing rust before painting - because doing so removes any traces of trapped moisture.

SteveBaker (talk) 19:40, 23 November 2015 (UTC)

Also, painting over thick rust doesn't work. The rust breaks off in chunks, and takes the paint with it. StuRat (talk) 20:38, 23 November 2015 (UTC)

There is a commercial brand (Hammerite, if I remember correctly) of paint designed to be painted over rust. It does require that the rust layer be thin and not flaky - something easily done by a metal brush mounted on a drill. I have used it and I am quite satisfied. I am not affiliated (regretably?) with the manufacturer in any way אילן שמעוני (talk) 18:41, 27 November 2015 (UTC)

Physical equivalent of asymmetric key algorithms

Is there an actual physical equivalent of asymmetric key algorithms ? Like how (only) the public key can decrypt something which was encrypted with a private key, are there physical locks which can be locked with one key but can be opened with only another key (with a different set of cuts or serrations)? - WikiCheng | Talk 11:02, 23 November 2015 (UTC)

The only physical analogy I can think of is a yale-type lock on a door, where someone already inside can open the door and it will lock when they close it. I have been able to say to a visitor who needed to leave while I was at work "close the door behind you, but make sure you have everything as you won't be able to get back in", after he'd left only someone with a key could get in. I know its not exactly the same. -- Q Chris (talk) 11:35, 23 November 2015 (UTC)

Yes. Also similiar are the pad locks which can be locked by pressing. This amounts to using one key (no key) to lock and another key (the actual key) to open. But I was wondering if there are any locks which require one key to lock and another to open :-) -- WikiCheng | Talk 11:58, 23 November 2015 (UTC)

Ha Ha - in googling I found a lovely poetic analogy "only the right one can unlock the key to my heart". (I know that it is not what you were looking for, but felt it was worth sharing) - Q Chris (talk) 12:05, 23 November 2015 (UTC)

(Not answering the question but..) Simon Singh's The Code Book uses the analogy of padlocks (the kind that can be clicked shut without a key): Alice sends Bob an open padlock (her public key) to her friends, and he can send her a secret message by putting it in a box and locking it with the padlock. Alice (and no one else) can open the box with her (private) key and read the message. AndrewWTaylor (talk) 12:17, 23 November 2015 (UTC)

A physical mailbox? Anyone who has physical access to it can throw messages into it. Only the owner of the private key can get the messages.

The problem with the pad lock analogy is that once someone uses it, further uses cannot lock it again. It forces a different use. 13:44, 23 November 2015 (UTC)Denidi (talk)

I think that's just because physical objects can't be duplicated the way data can. Probably any physical analogy for data will fail on that count. -- BenRG (talk) 18:10, 23 November 2015 (UTC)

A close analogy would be a combination lock with a fixed public unlocking state and a "dial" that only supports irreversible moves, like this cube. For any sequence of moves that returns the cube to its original state, the first half and second half of the sequence could function as a public and a private key respectively. You could plausibly construct a puzzle like this where it's easier to come up with complete move sequences of that type than to find the second half of one given the first half. Also, if the puzzle has enough symmetry that flipped moves in a reversed order also return it to its original state (as this cube probably does), then the same key pair could also be used for "signing", as is possible with RSA. -- BenRG (talk) 18:10, 23 November 2015 (UTC)

This is a loopy idea, but I keep wondering how complicated magnetic field lines can get. Sometimes the Sun seems to get them so tangled up and then spew the lot out into space, but I don't really understand it; I don't understand if ball lightning has a really tangled layout either. "Braided magnetic fields" are a thing [2] - in a conductive plasma, is it possible though to somehow braid a magnetic field in such a way that untangling it takes a prohibitively long time, sort of like decrypting from a public key by brute force, and to "solve" that field by some other complex manipulation? Wnt (talk) 20:00, 23 November 2015 (UTC)

@Wnt: Maybe not so terribly loopy. Or rather loopy in a way that others have been very interested in. The braid groups can be used in cryptography, see e.g here [3] for a recent-ish work, with lots of previous refs. I don't think the authors have a physical instantiation in mind, but at a skim I don't see any reason why it couldn't work. (It would be very very slow, and need some sort of adamantium wire, but the concepts don't demand non-physical braids) SemanticMantis (talk) 22:20, 23 November 2015 (UTC)

@SemanticMantis: I was thinking if you have a perfect conductor (not a superconductor because that pushes magnetic field lines out of itself, but apparently that's not strictly a requirement; maybe stanene? - see [4]) then you can have magnetic field lines inside it that cannot cross one another (thus the adamantium) and which are generated by controlling twists in the magnetic field at the surface of the perfect conductor, then you might be able to add operations per the braid group diagram from either surface, and keep them stably stored somewhere in the middle. This would depend only on how fast you can twist a magnetic field at the surface, I think. (But then again, I have a really poor sense for how twisting magnetic fields really works) Wnt (talk) 12:52, 24 November 2015 (UTC)

+-+ |A| c +-+ | / a |A| |/| s | / / | e |/| |B| ==/ |== ==\ |== p |B| |\| l \ | | \ u |\| |C| g | \ +-+ |C| +-+ +-+ |K| +-+ | | |K| | | +-+ +-+

I wonder if a standard pin tumbler lock could do it by cutting the shear points (the places where the key pins and driver pins meet) on on an angle tilted "up or down" compared to the shear line instead of flat. It would mean the key could only turn in one direction when the pins are engaged. For example, if the top of the key pin were cut "lower-left to upper-right" as facing the lock, it could only turn clockwise after being inserted and only for a key cut to the "upper-right" height. Turning this way, shear line aligns at the start (the key is correct), and turning pushes the driver pin higher (shear pushes against its angled cut). This key can then turn back to start and be removed, but not overall counterclockwise because the shear line does not align with the side of the pin on the left side (the key is not correct). A key cut to the height of the "lower-left" height would not work (the key is "correct" but does not turn) because that motion would make the shear try to push the key pin further down onto the key. A second shear line could be cut on the opposite angle (upper-left to lower-right)...a different key that would only turn in the opposite direction after being inserted. Result: two different keys, one for "lock" other for "unlock". DMacks (talk) 23:13, 23 November 2015 (UTC)

For example, the left diagram has the key (K) that pushes the pin (ABC, with a spring beyond A) up to the extent that the left edge of AB is at the shear line (==). The plug cannot turn clockwise because the right AB edge does not align. Even though the shear line aligns on the left, the key cannot turn left because the shear makes the case push down on B, a motion that is blocked by the key. Only when the key is correct for the correct angle direction can the it turn. In the right diagram, the plug cannot turn clockwise (shear line does not align on right edge of pin) but can turn counterclockwise because doing so makes the plug push up on B, which is just pushing up A and the spring. The pin can always and only slide up, so the plug motion is only allowed in the direction such that the angle of the pin/pin surface transfers a component of the turning motion in towards the spring. DMacks (talk) 23:44, 23 November 2015‎ (UTC)

Another real-world analogy is the drop safe (which we do not seem to have an article on!). The ones in retail establishments, any cashier can drop money into. The ones on the outside of banks for night deposits, the night depositors need an (ordinary, low-security) key to open, But only an Authorized Person with a higher-security key can get the deposited money back out. —Steve Summit (talk) 00:41, 24 November 2015 (UTC)

I would call that a night safe, but that link only goes to a disambiguation page with no onward links. AndrewWTaylor (talk) 08:55, 24 November 2015 (UTC)

There are locks like this where you can push the bolt in and lock it in place with the small key. When you release the bolt (with the small key) and close the door it's locked and you can only open it with the larger key. Sjö (talk) 06:16, 24 November 2015 (UTC)

Thanks everyone -- WikiCheng | Talk 07:08, 24 November 2015 (UTC)

Another example would be a shaft that has two ratchets that only allow motion in opposing directions. As it stands, it cannot turn because one direction is blocked by one pawl and the other by the other (each prohibits turning in the only direction the other allows). Have a key-lock that pulls the pawl back from its gear, a separate such mechanism for each pawl. Now each of the two keys unlocks one but not the other direction of motion of the shaft, and separate keys unlock the two different directions. DMacks (talk) 07:18, 24 November 2015 (UTC)

Obtained electricity or heat as result of a chemical reaction

What kind of reactions would produce heat, and what kind would produce an electrical current? Does burning also generate electricity? Or could burning generate electricity too? At least electrons are moving around. --Scicurious (talk) 15:12, 23 November 2015 (UTC)

To address the first question, see Exothermic reaction. {The poster formerly known as 87.81.230.195} 185.74.232.130 (talk) 15:22, 23 November 2015 (UTC)

If I got it correctly both reaction types, generating heat or electricity, are exothermic. It is not necessary to generate heat in the narrow sense. Just releasing energy makes the reaction exothermic.--Scicurious (talk) 16:05, 23 November 2015 (UTC)

To address the second question, see galvanic cell, electrochemistry, electrochemical cell and redox. --Jayron32 15:27, 23 November 2015 (UTC)

And could we use fossil fuel to implement a kind of gasoline/diesel-air battery? Could we put a membrane between both elements? fuel would be the anode, and air the cathode. --Scicurious (talk) 16:05, 23 November 2015 (UTC)

See Fuel cell and Direct methanol fuel cell. Yes, in other words. Tevildo (talk) 18:54, 23 November 2015 (UTC)

It's all energy, when you get down to it. You can convert heat to electric current and vice versa; electricity generation is mostly done by harnessing heat to heat water that then turns turbines. The reason why we choose one method over another is a question of engineering. A hypothetical perfect fuel cell would be more efficient for electrical generation than burning fuel to heat water, but real-world fuel cells have numerous issues that limit their usefulness. --71.119.131.184 (talk) 19:24, 24 November 2015 (UTC)

Would a hypothetical perfect fuel cell be better than hypothetically perfect conventional combustion, where 100% of the energy released goes into heating the water ? StuRat (talk) 08:20, 26 November 2015 (UTC)

That's irrelevant in the case cited: 71.119 said "for electrical generation than burning fuel to heat water". The question is: how efficiently can hot water be made to generate electricity, and the answer is "not very".--Phil Holmes (talk) 11:57, 26 November 2015 (UTC)

It's relevant to his last sentence, and the OP's "What kind of reactions would produce heat". StuRat (talk) 17:08, 26 November 2015 (UTC)

What is this electronic component?

It came off the power supply board of an old piece of test equipment. It's the same size as a standard fuse. My guess is something like this. The resistance between the terminals is 10K.

— Preceding unsigned comment added by Cash4alex (talk • contribs) 15:12, 23 November 2015 (UTC)

Applying the duck test, I'd say it's a Fuse (electrical).--Denidi (talk) 15:27, 23 November 2015 (UTC)

The picture is a bit out of focus. What does the red line in the middle appear to be made of? Smurrayinchester 15:33, 23 November 2015 (UTC)

I don't know... there are clearly markings from 0 to 10 - this is a 'fuse' with a scale of some sort, and a heavy-duty resistance. My guess is that it has to be some kind of thermometer, measuring the electrical current by the heat generated??? An issue is that the power dissipation (VI = I²R) isn't linear ... something like mercury is ... that isn't consistent with what I see, but then again, I'm not taking how the heat is dissipated into account. Wnt (talk) 15:47, 23 November 2015 (UTC)

A "temperature fuse" is commonly called a "thermal cutoff" and it doesn't look like that. When using inline leads, they tend to look like fat resistors - no glass - no temperature markings. Therefore, I doubt this is related to both "temperature" and acting as a "fuse." Further, temperature of 0 to 10 is a very small range. 209.149.114.197 (talk) 16:18, 23 November 2015 (UTC)

Here's a better picture (not taken with a mobile phone this time...).

It looks like there is a fat glass tube inside, with a small bore with the metal (mercury?) running through. The red part looks like it might be coloured plastic behind the glass. Cash4alex (talk) 17:08, 23 November 2015 (UTC)

An electrochemical hour meter - full explanation here: http://www.vintage-radio.net/forum/showthread.php?t=116566 — Preceding unsigned comment added by 86.130.12.61 (talk) 18:46, 23 November 2015 (UTC)

^^^ That's the answer. I found another ref here [5] - they're also called "elapsed time indicator". The principle is that there is a small blob of electrolyte somewhere in the middle; a small amount of electric current causes mercury to dissolve at one end and electroplate onto the other. As noted above, the high resistance of the device limits its power consumption to what is barely necessary - this is of course going to be wired in parallel (like a voltage meter) rather than in series (like a fuse) to the main circuit. They are still made [6] but now in a mercury-free variant - I didn't see the details of the new device operation. Others are digital and look more different. Wnt (talk) 19:55, 23 November 2015 (UTC)

It could also be a mercury-based coulombmeter. They can be used as run-hour meters. 62.56.60.78 (talk) 18:54, 24 November 2015 (UTC)

Maybe it's the missing Mercury link from the Tardis (very old reference). {The poster formerly known as 87.81.230.195} 185.74.232.130 (talk) 18:12, 25 November 2015 (UTC)

much later...

@Cash4alex:, @Wnt: - this video from the Applied Science YouTube channel goes into a lot of detail, shows one in operation (speeded up), and does some calculations on the coulombmeter's operation. -- Finlay McWalterᚠTalk 22:48, 9 December 2015 (UTC)

The note about Hg+ appears to be relevant. Calomel (Hg2Cl2) is spontaneously produced from mercury (II) chloride and mercury, and of course here mercury is present in excess - I don't know the counterion is Cl- but it seems like it would work. He estimates a 0.35mm tube (0.096 mm2 area). Mercury is 200.59 g/mol, and 13.33 g/cm3, so it is 15.048 cm3/mol = 15048 mm3/mol. Divide by 0.096 mm2 and that's 157000 mm of tube per mol of electrons, assuming Hg+ carriers. This is the point where I abruptly realize that SI units aren't all that rational, because a coulomb, oddly enough, is 1.036×10⁻⁵ mol of electrons! So you have to multiply that conversion factor to get 1.62 mm per coulomb of charge transferred. (I'm getting slightly over double the amount he had, mostly due to the Hg+ but also due to some roundoff early on) Now the resistor network splits the current by 20:1 (I'll take his word for it - notice he has radically different values for the resistors in the two pages of calculations!) and reduces the current to what he measured as 62 microamps, so that's maybe 3.1 coulombs = 5.03 mm per 1000000 seconds. He says that each 500 hours mark is 8 mm, so that's (5.03/8 = 0.63) x 500 = 314.375 hours of marked time expected per 1000000 seconds/60/60 = 278 hours of actual time. Hmmm, my calculation is 12% off whereas his was 5% off...; either way the bottom line is somebody at the company draws out the capillary tubes a little finer or a little narrower or changes the scale until they get the calibration right. Wnt (talk) 06:22, 11 December 2015 (UTC)

Our article: Mercury coulometer. -- ToE 03:09, 4 January 2016 (UTC) Thanks DMacks.

Lack of large marine browsers

The largest land animals (elephants, giraffes, rhinos, hippos, moose) eat leaves or are at least herbivores. So why isn't this the case for marine animals ? Baleen whales do eat plankton, but that includes zooplankton, so they aren't herbivores. The largest marine herbivore I can think of is the dugong. So, what limits the size of marine herbivores ? I wouldn't think it would be the food supply, as the Sargasso Sea has plenty of seaweed, for example. StuRat (talk) 20:48, 23 November 2015 (UTC)

Temperature is a likely limiting factor, see scholarly articles here [7] and here [8]. The former discusses latitudinal gradients in herbivore size and temperature as an important driver. The latter also discusses temperature, and also engages the Metabolic_theory_of_ecology, which relates biophysical and biochemical stoichiometry to ecology and evolution of organisms' life history traits. I think it's no fluke that the manatees and whales are so big - they are endotherms (while most of the ocean's critters are ectotherms), and so their growth rates are not so constrained by ambient temperatures. The great Simon Levin thinks that dimensionality is additionally important, see page 158 onward here [9]. The basic idea is that a pelagic herbivore living in a 3D world is constrained to foraging on macrophytes that predominantly grow on a 2D surface. SemanticMantis (talk) 22:13, 23 November 2015 (UTC)

But land herbivores are also constrained to foraging on plants that predominantly grow on a 2D surface. StuRat (talk) 04:53, 24 November 2015 (UTC)

Right, an elephant is basically living on a 2D surface, most of which can provide food. Pelagic ocean critters live in a 3D world, only a small portion of which can grow macrophytes. This plays into the gross metabolic processes of the individuals and populations, and offers an explanation of why body size of oceanic herbivores is often smaller than their terrestrial counterparts. Also it gives a rationale for why baleen whales and other eat lots of pelagic plankton instead of looking for macrophytes, but then they end up eating animals too, and you don't want to consider them herbivorous. SemanticMantis (talk) 15:19, 24 November 2015 (UTC)

(Also - perhaps User:Obsidian Soul has some additional comments or refs? SemanticMantis (talk) 22:23, 23 November 2015 (UTC))

• Steller's sea cow μηδείς (talk) 23:13, 23 November 2015 (UTC)
• See Manatee#Diet. --Jayron32 02:00, 24 November 2015 (UTC)
• See Dugong#Feeding. --Jayron32 02:01, 24 November 2015 (UTC)
• See Green_sea_turtle#Diet. --Jayron32 02:04, 24 November 2015 (UTC)

Probably the largest marine animals eat plankton because plankton is the most abundant food source in the oceans, and the largest land animals eat plants because plants are the most abundant food source on land. It might not be any more complicated than that.--Wikimedes (talk) 08:05, 24 November 2015 (UTC)

• Perhaps a silly question, but how exactly is a whale supposed to filter the phytoplankton from the zooplankton? The plankton mix freely and I can't see any obvious evolutionary advantage to not digesting zooplankton. Phytoplankton are the main photosynthesizers in the ocean, and I'd argue that anything that eats them counts as a grazer even if they also consume zooplankton. Incidentally, seaweed doesn't have leaves - it's a colony of algae (a group to which most phytoplankton also belong). Smurrayinchester 09:05, 24 November 2015 (UTC)

Sargasso Sea gets to Sargassum, which points to mostly unhelpful Herbivorous fishes, but this can be unravelled to Algae eater. Sargassum is brown microalgae, so I suppose that makes anything that eats it an algivore. They are not notably huge animals, but the pictures don't make this seem that surprising. Large herbivores live on endless plains with grass up to your ass (until they eat it all, that is); these fish have to deal with little clumps that occasionally get big enough to impress - though I'd speculate the contrasting demands of cold-blooded versus warm-blooded metabolism and the modes of locomotion might have something to do with it also. Truly large herbivores like elephants go after more than just grass, but can take down trees. My impression, perhaps false, is that large dinosaurs had even more impressive foliage to munch on. Wnt (talk) 13:49, 24 November 2015 (UTC)

It seems to me that this question could be asked in the opposite way and get to the same answer... Why don't elephants hunt and eat other animals? Could an elephant catch a gazelle? Could a giraffe hunt effectively? It should be rather obvious that it is difficult for large animals to be hunters. It isn't impossible - just difficult to not notice the huge elephant trying to sneak up on you. As for whales... I specifically remember an article from last year discussing how whales hunt giant squid and that it was an unknown practice because it takes place in the deep ocean where whales are not monitored. 209.149.113.52 (talk) 20:50, 24 November 2015 (UTC)

Toothed whales are pack hunters, basically lions of the sea. The article on pack hunters specifically notes both dolphins and orcas in this ecological role. --Jayron32 01:23, 25 November 2015 (UTC)

Another electrical device needing identification

• 
  Array of them at upper left
• 
  Front detail view
• 
  1/4 side view
• 
  Broader 1/4 view

This is in the old pump room of the Ballard Locks. Maybe some sort of heat sink? Something that can be cut in and out of the circuit so that the shift isn't too sudden when power to the pump is turned on and off? The otherwise knowledgeable person giving a tour of the locks wasn't sure. - Jmabel | Talk 23:42, 23 November 2015 (UTC)

Looks like a high-power resistor bank. Googling around, they are (or at least were) a component of pumping stations and other facilities for motor control. DMacks (talk) 00:08, 24 November 2015 (UTC)

Thanks! - Jmabel | Talk 00:23, 24 November 2015 (UTC)