This is what I figured, but then why are 'columns' always referred to when talking about convection and as the source of the granules? What columns?

http://csep10.phys.utk.edu/astr162/l.../interior.html
The convective zone starts at about 85% of the solar radius and extends to just below the surface. It is a region in which the change in temperature with increasing radius is so rapid that the Sun becomes unstable to convection

This agrees with the idea that the convection is turbulent and 'unstable', but then:

http://csep10.phys.utk.edu/astr162/l...anulation.html
Granulation is due to the convection operating below the photosphere that we already mentioned in the section on the solar interior. This convection produces columns of rising gas just below the photosphere that are about 700 to 1000 km in diameter.

Doesn't this contradict the previous statement? How can columns of convection ~1000km in diameter and ~100000km long exist in this turbulent and unstable region?

http://cassfos02.ucsd.edu/public/tutorial/Sun.html
Granuales are small structures approximately 1000 km across that cover the entire solar surface, except for the sunspot regions. They are the tops of deep gas columns where energy is transported by convection ... Individual granuales last for about 20 minutes.

So what exactly is the story about the columns? 1000km diameter, 100000km length, and they only remain stable for 20 minutes at a time, in a region where they can't form at all?

__________________
there is no governor anywhere

Let us consult the online Merriam-Webster dictionary, and seek out the definition of "column". We observe defintion 1C: an accumulation arranged vertically. This is the elusive "column" you are looking for. Turbulent, disorganized convection, occurs in the form of "bubbles" or "blobs" or "Globs", which are loosely arranged in columnar form. The "columns" are elusive, they move horizontally, and can be broken vertically, but they are definitely there. We can see them come & go, in the variable behavior of granules at the visible "surface" of the sun. And that is as expected.

In short, our knowledge of the physics of convection in the sun is not now brought into serious doubt or question, simply because somebody wrote a webpage using the word "column".

__________________
The point of philosophy is to start with something so simple as not to seem worth stating, and to end with something so paradoxical that no one will believe it. -- Bertrand Russell

Unless he is specific to the contrary, I think that is the right assumption. The title is, after all, supposed to be informative as regards the content of the paper.

The numbers are logarithms, which remain the standard way of expressing abundances. In a modern paper, all of them would be normalized to a value of 12 for hydrogen. But that was not the case in 1929. Look right at the bottom of page 65, going onto page 66, and see Russell's discussion of how the logarithms for Earth & meteorites have been adjusted to "reduce them approximately to the scale of log Q" (Q is defined on page 57 as the "total mass of the atoms or molecules of the substance per unit area of the sun's surface").

So, in table XVIII, the first entry is Ba for Barium, and the 5.4 is a logarithm, meaning that the abundance of Barium is 10^5.4, on the same scale as table XVII, where the abundance of Hydrogen is given as 10^11.5. Solar abundances are still tabulated that way in papers today.

Well, there's no doubt that the paper is outdated, it was published in 1929! Some of its conclusions remain correct, others doubtless not. Its real value now is historical, it was one of several papers which demonstrated that the sun was really made mostly of hydrogen, a conclusion which remains true to this day. It is neither necessary, nor expected, that everything in it must be true (as is the case for modern papers as well).

But pay careful attention to table XIX. Each element is associated with a specific wavelength in the next column. It is not Helium which is unobserved on the sun, it is the Helium spectral line at wavelength 5875 Angstrom units which is not observed. Remember that the root for "Helium" is "Helios", meaning "sun"! Pierre-Jules-César Janssen discovered the unexplained 5875 Angstrom unit spectral line, in the solar spectrum in 1868, and the new element was named Helium, since it was unknown on Earth. It was isolated in a terrestrial laboratory in 1895, by Sir William Ramsay in London and independently by N. A. Langley and P. T. Cleve in Uppsala, Sweden.

Helium was well known to be visible in the sun in 1929, so one can only assume that Russell did not see it for whatever reason. He does say that his data suffers from calibration problems, so he might have had instrumental issues that prevented him from seeing Helium when he would have expected to. I don't see a specific mention of it in the paper, so I don't know.

That may well be. Look at page 70, right under the heading "IV ASTROPHYSICAL CONSIDERATIONS", and find "a) The calculated abundance of hydrogen in the sun's atmosphere is almost incredibly great". Clearly the large inferred abundance of hydrogen concerned him. But then, skip forward to the middle of page 71 and find this: "It does not seem to have been noticed that the theoretical difficulties (b) and (c) can be greatly alleviated, if not removed, by the assumption that the difficulty (a) does not exist - in other words, the solar atmosphere really does consist mostly of hydrogen." On the one hand, the inferred large abundance of hydrogen is contrary to the standard assumption of the early 1900's that the sun was made mostly of heavy elements (Eddington's book, The Internal Constitution of the Stars, which features the opacity argument for a mostly hydrogen sun, was published only in 1926, and was the first big step in that direction). On the other hand, Russell notes that a mostly hydrogen atmosphere, which is inferred from observation, solves some big theoretical problems for the atmosphere. Add that to Eddington's earlier discussion of the solar interior, and you have a pretty strong argument in favor of a mostly hydrogen sun.

But Russell knew, as any good scientist should, that the alternatives need to be explored. He came up with an alternative explanation, and put it forward so it would be available, if anyone wanted to followup. I don't know what ever became of the suggestion. But I do know that the science of stellar atmospheres, both observation & theory, are far advanced today compared to 1929, and I suspect one could run that speculation down in a text book like Mihalas' Stellar Atmospheres, given the motivation to do so.

__________________
The point of philosophy is to start with something so simple as not to seem worth stating, and to end with something so paradoxical that no one will believe it. -- Bertrand Russell

Not so fast, big guy. Here is how the authors accomplished this nefarious deed (emphasis is mine).
In that paragraph there are 8 references. In ascending order, they are dated 1964, 1966, 1969, 1981, 1987, 1992, 1998, and 1999. Only 1 from 1964, and covering the years up to 1999. So I would say that "why hasn't anyone in all this time ever considered this part of the standard model before, ..." seems a mighty peculiar question to ask, if you are reading the same paper I am (The Eclipsing Binary V1061 Cygni: Confronting Stellar Evolution Models for Active and Inactive Solar-Type Stars, Guillermo Torres, et al., accepted to The Astrophysical Journal; 5th paragraph of section 5.2).

__________________
The point of philosophy is to start with something so simple as not to seem worth stating, and to end with something so paradoxical that no one will believe it. -- Bertrand Russell

A more accurate term is turbulent plume.

__________________
there is no governor anywhere

Excellent response to my questions about the Russell paper, thanks Tim Thompson. I was looking around for some more info on stellar convection and stumbled upon this paper, A few things we do not know about the Sun and F stars and G stars (postscript only, viewer can be found here). What is your take on the concerns of the author? Here are some of them:

* We do not know how to make realistic model atmospheres - We do not know the energy distribution or the photospheric spectrum of a single star, even the sun. We do not know what spectrum corresponds to a given effective temperature, gravity, or abundances, The uncertainties in solar abundances are greater than 10%, except for hydrogen, and solar abundances are the best known.

* We do not have good spectra of the sun or any other star - Low resolution, low-signal-to-noise spectra do not contain enough information in themselves to allow interpretation. Spectra cannot be properly interpreted without signal-to-noise and resolution high enough to give us all the information the star is broadcasting about itself.

* We do not have energy distributions for the sun or any other star - I get requests from people who want to know the solar irradiance spectrum, the spectrum above the atmosphere, that illuminates all solar system bodies ... I say, "Sorry, it has never been observed. NASA and ESA are not interested."

* We do not know how to determine abundances - One of the curiosities of astronomy is the quantity [Fe] ... What makes it peculiar is that we do not yet know the solar abundance of Fe and our guesses change every year ... Another curiosity of astronomy is that Grevesse and Sauval have decided a priori that the solar Fe abundance must equal the meteoritic abundance of 7.50 and that a determination is good if it produces that answer. If the solar abundance is not meteoritic, how could they ever determine it?

* One half the lines in the solar spectrum are not identified - It is imperative that laboratory spectrum analyses be improved and extended, and that NASA and ESA pay for it. Some of the analyses currently in use date from the 1930s and produce line positions uncertain by 0.01 or 0.02A.

This was written near the end of 1999, has the situation changed dramatically since then?

__________________
there is no governor anywhere