Posted on

Paul Drude: The Theory of Optics (1900)

In 1900 Drude brought out his book on optics.  Though comprehensive, the aim of the book was to introduce the reader to recent developments in the field and to enable them to follow the new work in the area.  Drude’s book was quickly translated into English by R. A. Millikan and Charles Riborg Mann.  The earliest edition of the translation seems to be from around 1902 and has a foreword by A. A. Michelson dated February 1902.  Michelson notes that the translation of Drude’s book makes the progress of the last ten years (1892-1902) in optics accessible to the English-speaking public and gives the first complete description available in English of “the development of the electromagnetic theory” and “the relation between the laws of radiation and the principles of thermodynamics.”

So, this book and its translation bring together a lot of elements – a kind of time capsule or snapshot of a highly transitional moment.  For example, in 1900, the electron, which had around as electrons, “ions” or “corpuscles” during the 1890s, had already even been officially discovered.  One of Drude’s translators, R. A. Millikan, would go on to measure the charge of the electron most laboriously from 1908 to 1913 with apparatuses like this:

X-rays were being created and misinterpreted sometime as longitudinal vibrations of the aether (says Pais Inward Bound page 41):

But Drude seems to have avoided that particular problem and kept all his electromagnetic waves transverse – at least partly because he usually keeps polarization in mind when discussing waves.  Another angle on the Drude 1900 snapshot is the foreword by A. A. Michelson, who is probably familiar to most people as the Michelson of the 1887 Michelson-Morley experiment that suggested the effects of the earth’s motion through the aether were very hard to detect at all in terms of its direct effect on light (assuming no aether dragging).  Of course, Michelson is also apparently the only physicist to have figured in an episode of Bonanza! (1962).  Drude’s book describes the Michelson interferometer but says little about Michelson’s youth in the Wild West.  Here is Michelson in the 1920s measuring the non-detectability of the aether at a site far from what might one day be the set for Bonanza!

A chess game in the aether:

Posted on

Larmor’s Rococo Electron

By February 1894 no one felt that Larmor’s theory with its thermodynamically-driven, charged (but otherwise hydrodynamic), core-powered vortices was doing a good job of grappling with the aether, notes Buchwald on page 162 of From Maxwell to Microphysics.  On April 2, Larmor wrote to Fitzgerald:

“My interest in the whole thing is not so keen as it was; but I still claim that as a dynamical analogue of the aether-action it is far wider than any other I ever heard of…”

So Larmor was not too keen on the whole model.  Moreover on March 30, Fitzgerald had pointed out that if you were going to have matter at the core of the vortices, you could just let the matter there do the electromagnetic work that was making the aether stress, pulsate, lose elasticity and rupture or just get whirled around the fizzy edges of the vortices.  You could even have poor old simple, unstressed, unruptured, un-fizzy, non-vorticidal matter just move around and have charges and so on.

By June 14 Larmor had abandoned the “vortex atom – the source of all his problems” notes Buchwald on page 167.  By August, he was working out how electrons worked (mind you, this is three years before J. J. Thomson officially discovered something that we now think of as electrons – though you can see why – given what Larmor was replacing with “electrons” – Thomson wanted to call them something else) which resulted in a much simpler and less “baroque” set of models.  Buchwald calls the final form of Larmor’s “vortex atom” the “baroque mechanism” and it does seem bizarrely convoluted.  The electron that Larmor used to replace the baroque mechanism of the vortex atom is more Rococo and it echoes in some mechanisms that Olafur Eliasson installed as “Baroque Baroque” works in what looks like an incipiently Rococo palace in Vienna in 2015:

So, as Buchwald notes (page 127 of From Maxwell to Microphysics), Larmor was one of the first Maxwellians to give up Maxwellian theory.  The fact that Maxwellian theory was disappearing rapidly in the late 1890s will no doubt come as something of a shock since Maxwell’s basic field equations are correct – at least as they are now understood – what had to be given up was essentially the aether as an overall dynamic engine.  Something aetherial could be retained as microfields and things such as vaguely aetherial “virtual oscillators” and so on were retained as late as 1926 or so even in quantum mechanics, but the aether as a sort of universal “imponderable” luminiferous fizzy medium powered by vortices was disappearing very quickly in the late 1890s.  As Buchwald says on page 127, in 1893, Larmor was a Maxwellian.  By 1895, he was not.  Larmor’s critique of a magneto-optic theory using an elaborated aether in 1895 concludes the theory “retains the dynamical equations and surface conditions which belong to the luminiferous aether under ordinary circumstances, merely adding on to the electric force a new part of magneto-optic origin.  This would hardly be open to objection if it worked, but it is admitted that it does not work…”

Posted on

Joseph Larmor: Aether Rupture

So, sometime before August 1894, Larmor decided he needed some built-in discontinuities to drive energy localization in the aether.  Poynting and J.J. Thomson (the Thomson who measured the energy of what we now think of as electrons in 1897) had already been using imaginary “tubes” for this kind of discontinuous energy motion and localization in the aether, but Larmor preferred to use an unknown microphysical entity that could drop the elasticity of the aether to zero in the intermolecular gaps.  This (to quote Larmor as quoted in Buchwald From Maxwell, page 147) allowed for the case when:

“A charged body…causing rotational strain in the aether…{so that} the aether inside another surface…{loses} its rotational elasticity as the result of instability due to the presence of molecules of matter;  the strain of the aether all around the surface must readjust itself to a new condition of equilibrium; the vortical lines of strain will be altered…”

And there you have it:  the aether “ruptures” (i.e. its elasticity goes to zero in some region) allowing electromagnetic events to zip along the zero elasticity ruptures as equilibria are restored.

Buchwald points out some more things about all this that are far from obvious.  Lord Kelvin turned up and said he was giving up on vortices (yikes!  And he sort of invented vortex aether theory himself! In the 1860s!)

and aether rupture didn’t work well for some electromagnetic integrals going back to Faraday in the 1840s.  So (Buchwald, page 158), Larmor needed to have a more explicit mechanism in the microphysical workings of his energy localization theory and so he invented something to solve Lord Kelvin’s dislike of vortices as commonly imagined and at the same time keep all the Maxwellian aether machinery running as well: the thermodynamically-driven, charged (but otherwise hydrodynamic), slippery, core-powered aetherial vortex.  Quite a feat, (as Buchwald remarks) “without even a wink at demonstration.”  The aetherial outer fizz of these vortices constantly rotated, driven by “heat” while the “charge” in the magically lubricated non-rotating vortex “core” stressed the surrounding aether.  Pretty slick!  And it uses some kind of charged particle to keep the whole thing running and hooked up to the aether since those cores keep stressing that aether. Why rupture when you can just stress? 

But would that really work very well?  Fitzgerald wondered in his letters to Larmor (as Buchwald reports on page 160 or so), for while the aetherial vortices go on churning, it’s the little electrical “monads” that are doing a lot of the work by supplying positive and negative charges inside the cores of the vortices.  This did not seem very satisfactory to Fitzgerald or Larmor and by February even J. J Thomson was getting worried and noted that the draft paper of Larmor’s that he had been reviewing did not seem to “grapple” very well with the connection between ether and matter.   

Posted on

Joseph Larmor: Stressing the Aether

We left Classical Physics in flux at the end of its tether, last post.  How was it going to reach a new (if short-term) stability?  After all, it’s 1893 and Classical Physics has to be coherent enough to be considered replaceable or modifiable at some point for some theoretical concerns sometime in the early 20th century.

Just down the street from the Cavendish lab in 1893 (seen above more recently), Joseph Larmor was working out the details of how the aether “stores energy reversibly in differential rotation.” (Buchwald, page 136, From Maxwell to Microphysics).  To describe this period (1893), Buchwald quotes Eddington writing in the early 1940s, when Eddington said:

“…when Larmor started his main work…on the ever-urgent problem of the ultimate relation of matter and electricity and aether…Classical Physics was indeed at the end of its tether.”  As usual, I might add.

Also, of course, the tethering problem was the question: what happens when magnetic fields interact with things so as to cause a shift in what is happening with light?  And what was jerking the chain on the tether was the usual Maxwellian one: what happens at the boundary between media, as when the reflection of light by a substance is changed by a magnetic field?  Whereupon, Larmor started from the usual Maxwellian standpoint: the energy involved is all in the field – which is to say, somehow in the aether, where apparently it needs to be stored reversibly in differential rotation.  It all makes sense, right?  Or does it?  Which, of course, was exactly what Larmor was wondering.

Having the energy all in the field did not really answer Larmor’s own question since for the methods he wanted to use to analyze magneto-optic events, he wanted to localize the energies in different parts of the continuum.  He was so sure that the Helmholtzian polarization theory of energy localization was wrong (and that the Maxwellian localization in aether stress was correct) that he misinterpreted the results of an experiment with a charged capacitor partially immersed at the interface of an insulating fluid and the atmospheric air as we know it (Buchwald p.139) in 1893.  By 1895 he had accepted the implications of polarization within the media: the energy follows different configurations in different media.  Larmor eventually changed his mind by another route and then came back to looking at the energy distribution.  Still, in 1893, he was sure that the energy was only in the aether, though the discontinuities in the aetherial energy caused by the non-energetic presence of ordinary matter could cause other forces to become manifest.

According to Buchwald, Larmor changed his mind around April 18, 1895 “nearly 8 months after Larmor introduced the electron” to account for what seemed to be happening in his models: aetherial vortex stress and aetherial energy ruptures.  For a while, the electron (for Larmor, a rather aetherial entity that could be embedded in positive or negative charges) filled in the gaps, but, says Buchwald (p. 141) by around April 18, Larmor had abandoned Maxwellian theory as we will see in the next posts as the full implications of the electron (not as yet officially discovered as “corpuscular”) start to come into play.