Wikipedia:Reference desk/Archives/Science/2022 February 11

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

February 11

Thermostatically controlled shower

Our shower is supplied from the cold water supply only and it heats the water electrically. The temperature is thermostatically controlled so it does not matter if the water pressure varies. The knob setting the water temperature, although it has radial marker lines, turns smoothly with no sign of discrete positions. There is another knob for "high", "medium" and "cold" which we leave on high. How is the heater likely to be controlled? Will the power be being switched on and off fairly rapidly or will there be some sort of "analogue" control? Thincat (talk) 09:15, 11 February 2022 (UTC)

One possibility, of many, is Infinite switch Greglocock (talk) 09:38, 11 February 2022 (UTC)

Yes, that looks the sort of thing. Thank you. Thincat (talk) 09:52, 11 February 2022 (UTC)

A device like that made today might instead use a potentiometer and sensor setup, maybe as part of a PID controller, rather than deal with the mechanical unreliability of the infinite bimetallic switch. 2602:24A:DE47:B8E0:1B43:29FD:A863:33CA (talk) 09:53, 13 February 2022 (UTC)

Speciation in coral view

A coral diagram seems to be a good way to depict speciation. In the illustration here, created by Lajos.rozsa, the first detail view appears to be at a moment just before the species are separating. But the magnified picture (second from top) looks pretty homogenous. Shouldn't there be some signs of the impending separation, such as a reduced reproduction rate in the middle of the top border? ◅ Sebastian 09:57, 11 February 2022 (UTC)

I have a hard time interpreting the diagram. Its caption in Coral of life says: "Zooming out from a network/tree of individuals to a coral diagram in 4 steps." Usually, in a stepwise zooming-out presentation, the full rectangle of the image of step N shrinks to a smaller rectangle embedded in the image of step N+1. So I guess the "zooming out" is bottom-to-top, which (I think) is unusual. Then, the image in the shrunk rectangle embedded in the image of step N+1 is usually still recognizable as a scaled version of the image of step N. Here, I don't see a correspondence. Also, the branches of what is supposedly a single tree overlap in the middle two images, making it difficult to see this as a tree diagram. As to your specific question: (a) is the reproduction rate in a population about to speciate known to be reduced? and (b) how is the rate supposed to be represented visually in such diagrams?  --Lambiam 10:57, 11 February 2022 (UTC)

Thanks for pointing out those two problems. I agree with your interpretation of the composite image being described bottom-to-top, which is also expressed by the direction of the arrows, and I agree that that's rather unusual. Maybe that could be added explicitly to the caption. The second problem is tougher, since it requires a thorough change of the image file. (Most obviously it can be seen in the last two sub-images: Whereas the smaller rectangle in the second last sub-image has only 3 lines entering the bottom edge, the last sub-image shows 7, and a much more even distribution.)

But the overlapping branches are not a problem: That is the whole point of a coral diagram as opposed to a tree diagram. That said, it is a weakness of the image file that it doesn't show any such case in the last sub-image.

As for the specific questions regarding the OP:

(a) I don't know. But: In the scale of the topmost sub-image, which we should call “sub-image 4”, the two branches diverge right above the edge of rectangle for sub-image 3. That means that, If we moved sub-image 3 up just a millimeter or two, we should see no nodes in a wide stretch of the top edge. I can see two ways for that to happen: All the lines need to either suddenly die out or abruptly change course, deciding for one of the two branches. I avoided the second possibility in my question, since it's harder to explain. But let me try: While the diagram type has, to my knowledge, no explicit rule for the x-coordinate of any node, I would say that expediency demands that it represents the closeness (by phenotype or genotype) to any of the two resulting species, at least in the vicinity of a speciation event. Thus, all offspring of that generation would exhibit a sudden change, which violates my understanding of evolution, which is gradual under normal circumstances.

(b) As I understand it, the rate should be the average of lines pointing upward from each node (divided by the number of lines coming in from below, but that's mostly 1 here, so we're safe to neglect the deviations from 1). ◅ Sebastian 13:18, 11 February 2022 (UTC)

Re. While the diagram type has, to my knowledge, no explicit rule for the x-coordinate of any node, I would say that expediency demands that it represents the closeness (by phenotype or genotype) to any of the two resulting species, at least in the vicinity of a speciation event. There's no such thing as "the vicinity of a speciation event". For speciation to occur two groups of the same species need to be geographically separated for a large number of generations. There's no 'split' in the family tree caused by a slow dwindling of reproduction rates in that 'area'. The x-coordinate in these diagrams (which are not graphs) is just an abstract way to visualize separateness. The way the lines waiver and go off at various angles is misleading. nagualdesign 15:44, 12 February 2022 (UTC)

Sorry, the term “vicinity” wasn't as clear as I thought. Since none of the dimensions of the coral diagram are spatial, it didn't occur to me that anyone could read it as geographic vicinity. What I meant was: “any point within the same contiguous area of the branch whose time is much closer to the speciation event than to any other event”. I hope this clears up the misunderstanding. ◅ Sebastian 23:09, 12 February 2022 (UTC)

Hello, I am the author of the figure. Please do not take the minor details of the picture too seriously, in fact mine is a botanical version of the figure by Maddison and Maddison 1992 which suffers from the same imprecision. Since the system does not allow me to include a figure, I give a link: https://www.cs.ubc.ca/~tmm/talks/birs04/mgp00010.html In my Systematic Biology paper a figure with similar meaning is more fortunate, I can send you a pdf if you email me. In this, a math definition of a coral is also given. Again, apologies for the confusion. Cheers J Podani, podani at ludens dot elte dot hu — Preceding unsigned comment added by 2001:4C4D:19C4:6600:E57C:8BDF:675F:484A (talk) 15:11, 16 February 2022 (UTC)

How many electrons from an atom form the electron gas?

Been wondering which electrons of a metal atom contribute to the electron gas in metals and which ones remain attached to nuclei. I was wondering specifically for lithium. Jo-Jo Eumerus (talk) 13:41, 11 February 2022 (UTC)

Articles like electron gas go way above my head but electrical resistivity and conductivity#In metals is a bit easier to understand. This suggests that it is the outermost shell of electrons that contribute to the conduction and for lithium it's pretty certain just to be its single 2s electron that matters. Note that no electron "attaches to the nucleus", they are all in atomic orbitals around it. Mike Turnbull (talk) 13:51, 11 February 2022 (UTC)

It's also important to note that electrons are not individually identifiable; neither is their location knowable. In terms of modeling metals, the Valence and conduction bands matches up pretty well; in the sea of electrons model, the valence band is those electrons that are capable of participating in electrical conductivity, and gaining energy to reach the conduction band. --Jayron32 16:14, 11 February 2022 (UTC)

Yeah, my question is aimed at determining whether any of these electrons don't participate in conduction [and in superconductors, if any of the electrons don't form Cooper pairs] Jo-Jo Eumerus (talk) 16:59, 11 February 2022 (UTC)

biohacking

Is there a forum about biohacking that is reputable?--2A02:908:426:D280:E489:4BB:F7DD:7E43 (talk) 21:35, 11 February 2022 (UTC)

I don't know, but if such a thing exists, gwern.net or erowid.org would likely have links to it someplace. 2602:24A:DE47:B8E0:1B43:29FD:A863:33CA (talk) 09:55, 13 February 2022 (UTC)

Biohacking means different things to different people, as you'll see from that disambiguation page. If you mean specifically Do-it-yourself biology, that article has plenty of reliable sources and some links to organisations that should have discussion forum via their websites. Mike Turnbull (talk) 12:13, 13 February 2022 (UTC)