Science desk
< October 19	<< Sep \| October \| Nov >>	October 21 >

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.

October 20[edit]

Various types of honey[edit]

Living in the Western United States I purchase my honey at Amazon. It is collected in a Utah farm, It is a thick honey but this honey never crystallizes. Once upon a time I happened to travel to rural Utah for a reason different than collecting honey and in a small grocery shop bought a jar of honey. "Closer to nature" was my motivation. This honey crystallized two month after staying home. What's the difference between the two honeys? Is crystallized honey somehow not as good as the fluid one? Perhaps the one I purchase on Amazon is specially diluted with water to prevent crystallization? I feel there is a lot of mystery in my honey. Thanks AboutFace 22 (talk) 00:56, 20 October 2021 (UTC)[reply]

Our honey article talk about several factors that can affect crystallization, including the nature of the honey itself, temperature/handling, and addition of other ingredients. DMacks (talk) 05:34, 20 October 2021 (UTC)[reply]

[Edit Conflict] Honey (a natural and inherently variable substance to start with) can be processed in various ways, resulting in greater or lesser tendencies to crystallize (which is in itself not a problem: gentle heating will re-liquify it). Such variations in honeys do not make one or another inferior, just different.

Our article Honey discusses this at Honey#Classification by packaging and processing and elsewhere. {The poster formerly known as 87.81.230.195} 90.193.128.151 (talk) 05:38, 20 October 2021 (UTC)[reply]

It may also be worth noting that an enormous amount of honey is fake/adulterated/diluted (ref, ref). Matt Deres (talk) 15:03, 21 October 2021 (UTC)[reply]

How to know if a certain food contans caffeine (or any similar substance?[edit]

Dry food / fresh food / liquid food,

Is there any way to find if the food contains any "detectable" amount of caffeine (preferably doable at home)? Thanks 2001:44C8:4200:4FF2:16EA:EFA7:4BD4:A829 (talk) 03:44, 20 October 2021 (UTC)[reply]

Relatively inexpensive (say $10-20) Caffeine Test Strip Kits (easily websearchable) can be purchased, but these are intended to check that "decaffeinated" coffees and teas do not have excessive caffeine levels, or to roughly measure caffeine levels in drinks that are supposed to contain it. They would probably not be capable of detecting traces of caffeine in other comestibles that ordinarily do not contain caffeine in significant quantities.

Kits for detecting caffeine in breast milk (which may reach levels of around 1% of the level in the mother's ~~milk~~ blood) are or have been also available, and might be usable on other liquids.

More sophisticated and sensitive tests seem to require laboratory-level equipment such as spectrometers. Perhaps other editors may know of other means. {The poster fomerly known as 87.81.230.195} 90.193.128.151 (talk) 05:26, 20 October 2021 (UTC)[reply]

"Detectable" is a question of how you choose to try to detect it--different methods have different lower-limit of detection. So an HPLC system might find a very low level (ppm), whereas a test-strip might only find a few tens of mg per serving-size. It's pretty easy to extract the caffeine from a few tea-bags or coffee-grounds using kitchen chemistry and materials from a local hardware store, and it's sensitive enough to find that different brands/styles of coffee have different content. But it is not easily generalized to "any food" (coffee is a fairly simple combination of chemicals) and obviously depends on how good a balance or scale you have. But if your goal is just "does it contain?" then you don't need to quantify, just observe. DMacks (talk) 05:49, 20 October 2021 (UTC)[reply]

Can HPLC have selectivity to detect a compound as large as caffeine? I thought the instrumentation in biology labs are more suitable for this? Eliza? 67.165.185.178 (talk) 11:10, 21 October 2021 (UTC).[reply]

Depends on the matrix. Caffeine is a pretty small molecule even in the simple-structures organic world, let alone in the world complex and large biomolecules. There's lots of literature for using it to detect caffeine in blood, quantitative validation down to tenths or hundredths of a microgram per mL even without resorting to a mass-selective detector (LC-MS). DMacks (talk) 11:51, 21 October 2021 (UTC)[reply]

If all else fails, consume a significant quantity late in the evening and see if it keeps you up that night. ←Baseball Bugs ^{What's up, Doc?} carrots→ 05:32, 21 October 2021 (UTC)[reply]

Article Cited by Anti-Vaxxers as Evidence COVID-19 Vaccines Are Not Effective.[edit]

This article, Increases in COVID-19 are unrelated to levels of vaccination across 68 countries and 2947 counties in the United States seems to go against the scientific mainstream that COVID-19 vaccines are effective against the spread of COVID-19. Am I misunderstanding something? Has this article passed peer-review? A Quest For Knowledge (talk) 15:52, 20 October 2021 (UTC)[reply]

It doesn't go against anything. It's a data point, and a true one from what I can tell. However, it doesn't mean what you say it means. It doesn't mean that vaccination is not an effective strategy for stopping the spread of disease; rather it likely shows that no place has reached a high enough vaccination level yet. The highest vaccination levels it quotes are 75%... That's not high enough to see the effects they are looking for. What we have seen is that at the same time that vaccination levels are rising, other methods have been entirely abandoned (distancing, reducing travel, masking, etc.) so that any positive effect we would be getting among vaccinated people reducing transmission has been more than offset by the negative effect caused by all of the unvaccinated people abandoning prophylactic measures designed to stop transmission. If we want to know if vaccination is effective, we need to compare transmission rates among fully vaccinated people with those who are not vaccinated at all. This study throws all of these people together in one giant pot and doesn't isolate vaccination status as a single variable, since other factors (all of the prophylactic measures I noted above) are also changing at the same time. Rule #1 of conducting a valid experiment is to isolate your variables. --Jayron 32 16:03, 20 October 2021 (UTC)[reply]

Published in a reputable journal. The header says "correspondence", but I do not know where it fits within the "types of article" journal page. Whichever peer-review occur must have been light, considering the authors used data from September 3 and the article was published on September 30 (the delay between data collection and submission to journal, and the delay between submission of the first draft and publication, is usually several months across most scientific disciplines). This being said, even with proper peer review, the most outlandish things have been published in the best journals (example).

I can add a few methodology remarks to Jayron32’s. Taking a single point in time is not serious - possibly the authors have been sitting on the analysis script for three months until the daily data agreed with their proposed narrative.

Furthermore, they assert that their selection criteria were that the dataset on ourworldindata.org had second dose vaccine data available; had COVID-19 case data available; had population data available; and the last update of data was within 3 days prior to or on September 3, 2021.) If you go and look at the country list in their supplementary file (btw, great job using a word document to hold a data table, guys) it looks quite weird. China is not included (but Russia is, so it is not a case of "potentially fabricated data so we exclude it"), France is not included (I supposed based on the "last update" criterion because looking at the page right now all the rest are met).

Finally, I see no honest reason to present the vaccination rate vs. recent cases differently in figure 1 (world data) and figure 2 (US counties). I clearly see a dishonest reason: even on figure 2, looking at the center lines, you can see a downward slope, i.e. a correlation that goes against their narrative; the bins and whiskers are just here to obscure it. (This is IMO the biggest point that a reviewer should definitely have caught if peer review had been done properly.) I also suppose (but cannot know because the supplementary information does not contain the actual excel file) that their correlation line in figure 1 was drawn with assigning a weight of 1 to every country, though obviously a weighting by population would be more appropriate. Tigraan^{Click here for my talk page ("private" contact)} 10:19, 22 October 2021 (UTC)[reply]

I agree with both Jayron32 and Tigraan above: too many hidden variables (eg. population density, political alignment, anti-infection measures, level of social cohesion and public compliance, population age distribution, income distribution, latitude, temperature and weather, you could go on and on...), far too little statistical analysis, and I don't think it really proves anything one way or the other. It would be interesting to see some more detailed research into the matter. -- The Anome (talk) 10:33, 22 October 2021 (UTC)[reply]