On this page I have listed some very interesting papers about experiments on the one-way speed of light. The topic is discussed in depth in my book, Unified Revolution: New Fundamental Physics, (2014).
Some (many?) physicists do not seem to be aware of that only the round-trip speed of light has been very well-tested. The round-trip speed of light is indeed isotropic and is represented by the constant c. To measure the round-trip speed of light, we only need one clock and because of this we are not dependent on any clock synchronization procedures. In general, this is not the case when setting up experiments for measuring the one-way speed of light.
To measure the one-way speed of light, one generally needs two clocks. To synchronize two clocks over distance, one in general needs to know the one-way speed of light. Poincare, who believed that light propagates through the ether, thought it would be impossible to detect motion against the ether and thereby not the true one-way speed of light. For convenience he suggested that clocks should be synchronized based on "assuming" that the one-way time of light was half of the well-tested round-trip time of light. Einstein simply abandoned the ether and assumed the one-way speed of light was identical to the quite well-tested isotropic round-trip speed of light.
Also, for practical purposes, slow clock synchronization transportation basically gives the same synchronization as the Einstein-Poincare synchronization. Even if the clocks are moved through air, this is equivalent to synchronization in a vacuum with light signals. This is actually a very important point in relation to some experiments where the light signal is sent through a material medium and measured with clocks moved through air (for example by the earth rotating the stationary clocks), see Unified Revolution for further details and mathematical proof why slow clock transportation is identical to Einstein synchonization with light also under atomism.).
The Michelson-Morley experiment only excluded the view on the one-way speed of light based on old ether theories; it did not exclude the ether theories after FitzGerald introduced length contraction and Larmor added time dilation to it. However, FitzGerald and Larmor did not come up with any method to detect motion against the ether, even if they proved that the Michelson-Morley experiment was also consistent with the "modern" ether view of the time. Einstein´s theory was the simplest to explain experiments as long as no one had come up with experiments to detect motion against the ether. The big question is: Is it really true that we cannot detect motion against the modern ether?
To detect motion against the ether (void) requires quite clever set-ups and a deep understanding of space and time. Poincare, Lorentz, Larmor, and FitzGerald could not come up with experiment set-ups to do this. However, researchers have later likely come up with experiments to do so and a series of these experiments have been performed. They claimm to have detected motion against the void (ether) and thereby have also been able to estimate the true one-way speed of light. The true one-way speed of light is anisotropic, as suggested by Joseph Larmor, among others, in the year 1900.
These experiments should have been repeated and discussed in much more detail. Instead, the physics establishment has ignored these experiments (and even tried to ridicule some of the researchers that have done these experiments). The fact is most physicists have only studied a few chapters of special relativity theory while at university and have taken SR for granted because they were told by their professor that it was the most well-tested theory in physics. Sadly, most physics students have not even heard about the existence of a series of experiments strongly indicating that the true one-way speed of light is anisotropic. Their professors typically (not always) decided on focus on experiments that supported the mainstream view.
The one-way speed of light measured with Einstein-synchronized clocks must, by definition, be isotropic and represented by the constant c. However if the true one-way speed of light is anisotropic, then the Einstein one-way speed of light contains a synchronization error.
This is not the first time that the scientific establishment has slowed down science for decades due to ignoring observations not fitting the conventional and accepted views. In finance, for example, researchers had found out in early 1900 that empirical stock and commodity price returns were not Gaussian-distributed, but rather they had high peaks and fat-tails compared to the Gaussian distribution. This observation was basically ignored by the academic establishment for the next 70 to 80 years. In cosmology, the observed distribution of galaxy velocities was ignored even in the early 1990's. One assumed that the theoretical models were correct and the data were wrong. Years later it became more generally accepted that the observed distribution had a higher peak than the model distribution and that the model was either wrong or at least incomplete.
It is high time for physicists to stop ignoring the one-way speed of light experiments that show we likely (this is still open to debate) can detect motion against the void (ether) and that the true one-way speed of light is anisotropic. Any great physicist should, at a minimum, push forward thorough discussions and repetition of these experiments!
Below are links and references to some very interesting papers with respect to the one-way speed of light experiments (PS I am not endorsing all the claims in these papers. Still a series of independent researchers claim to have detected motion against the ether and ansitropy in the one-way speed of light. I think as a minimum some of these experiments should be repeated and looked at more closely):
Marinov, S. (1979): Measurment of Labratory's Absolute Velocity Conclusion: the one-way speed of light is anisotropic!
Sent light through a 500 meter cable (5 meter below surface, temperature control +- 1K. Rubidium clocks.) Conclusion: the one-way speed of light is likely anisotropic! Their experiment has basically been ignored!
The Roland De Witte 1991 Detection of Absolute Motion by Reginald T Cahill. Research project with Belgacom. 1.5 km coxical cable. Three ceesium atomic clocks on each side. Conclusion: the one-way speed of light is anisotropic! Criticized by Spavieri et al (2012).
Marinov, S. 2007: New Measurement of the Earth’s Absolute Velocity with the Help of the “Coupled Shutters” Experiment, Progress in Physics, Volume 1, (pdf downloade) Conclusion: the one-way speed of light is anisotropic!
Cahill, R. 2012: One-Way Speed of Light Measurements Without Clock Synchronisation, Progress in Physics, Volume 3. (pdf) Conclusion: the one-way speed of light is anisotropic!
Monstein, C. and Wesley, J.P., 1996: Solar System Velocity from Muon Flux Anisotropy, Aperion Volume 3This paper is extremely interesting; it refers to a series of studies using different methods to detect the solar systems velocity against such things as the cosmic microwave background. What is particularly noteworthy is that a series of clever one-way speed of light experiments (some of them mentioned above) gives approximately the same speed of the earth against the void (the velocity against the void can be approximated by such things as the velocity against the CMB) as other completely independent methods, such as muon flux anisotropy (that is an approximation measurement for velocity against void).
Researchers claiming that the one-way speed of light cannot be measured should ask themselves what is the probability that one will find approximately the same velocity of the earth against the ether (void) as the speed of the solar system against the CMB? Poincare was right that light had to propagate through something; however, he was possibly wrong on his point that velocity against the ether (void) not could be detected.
Spavieri et. al. 2012: Can the one-way speed of light be used for detection of violations of the relativity principle?. The authors claim that it is not possible to detect motion against the ether if one takes into account clock synchronization. This question is still open to debate, and only experiments combined with in depth study of space and time will be able to give the answer. They seem to not have discussed the possibility that the slowdown of light in material mediums could be proportional anisotropic, as explained in Unified Revolution. Slow clock transportation synchonization will not canel out the Einstein synchonization error (ansitropy) effect in a material medium if the slow down of light is proprotional anisotropic. Cahill has commented on the Spavieri critics of the DeWitte experiment
2010 (MIT Technology Review) The One-Way Speed of Light Conundrum
In my book, Unified Revolution, several new experiments are suggested that likely can detect the motion against the void. These experiments can likely distinguish between mathematical atomism and SR. Mathematical atomism shows that all of the SR equations are correct for Einstein synchronized clocks, but that the one-way speed of light, as well as length contraction, one-way time dilation, and the relativity of simultaneity under SR all contain Einstein synchronization errors. Mathematical atomism bascially leads to all the same mathematical end results as SR when using Einstein synchronized clocks, but with a deeper insight.