Author  Method  Value  Time of first publication 
Experimental value  1/a = 137.03599911(46) a = 0.007297352568(24) 

Armand Wyler  a=(9/16p^{3})�(p/5!)^{1/4}  1/a=137.0360824 a=0.007297348 
C.R.Acad. Sci. , A269 , 743(1969) and A272, 186(1971) D. PALLE.

Anastass Anastassov  a_{0}^{1}=8p^{2}3^{1/2} a^{1}=a_{0}^{1} + g; g^{1 }= p+p^{1/2}/4 a^{1}=a_{0}^{1} + 4/(4p+p^{1/2})  a_{0}/(p^{3}+p+p^{3}) 
a_{0}^{1}=136.75741 a^{1}=137.03621 a^{1}=137.03599915 
The theory of relativity and quantum of action. The book In Bulgarian.
Sofia, 2003. 
Hans de Vries  a_{0}^{1} =
G^{2} / exp(p^{2}/2). G = 1+a(2p)^{0}(1+a(2p)^{1}(1+a(2p)^{2}(1+... 
After term 0 0.0071918833558268 After term 1 0.0072972279174862 After term 2 0.0072973525456204 After term 3 0.0072973525686533 CODATA2002: 0.007297352568(+/24) 
October 4, 2004.

Dr. M. Geilhaupt  Unified force equation yielding mass and charge of the electron. a(t)=ln(f(t)) a =3/4*(1a(t))^{2} 
a=ln(3) a=1/137.112 
1984 (internal information university of bielefeld) 1998 
Dr. M. Geilhaupt  Unified force equation. Fg/Fc=24/N_{M}^{2} other constants. All values in full correspondence with experimental ones. 
N_{M}=10^{22} a=1/137.03603 
1984 (private information) 1998 (internet) 2000 
M. Wales  a = m_{0}ce^{2}/(2h) every physics text book, but a is dimensionless and because h is an action therefore ha is an action ah = m_{0}ce^{2}/2 = w = Wales quantum of internal action of the electron = 4.8352812012D36, i.e., if we use CODATA 1986 values e and h therefore a = w/h = 0.00729735301 (CODATA(1986) is 0.00729735308) i.e., alpha is the ratio of the quantum of action of the electron to the quantum of action for the Compton photon. And so on. For details look his homepage.  1/a = 137.035989621263 = 1/0.007297353073...  Internet, Feb, 1998 
I. Gorelik  n+s=1/a n=137 s(n+s)=p^{2}/2 n  number of rotations in zt per one rotation in xy, s  shift per one rotation. Solution for this system gives: 1/a=137.036010988... Error: 0.000000157 
1/a=137.036010988 Error: 0.000000157 
Internet, June, 25,1999 The message "Alpha collection" to newsgroup "sci.physics" 7l0rt2$rf4$2@news.sovam.com 
Steven B. Harris  Development of upper method: From the form sn=p^{2}/2 to the form: 1/a = 137 (1 + k^{2}), where k = p/(sqr(2)�137). And further to form: 1/a = 137 [1 + k^{2}  k^{4} + k^{6}  k^{8} ....]. From the letter to newsgroup: ...It might be fun to fool around with some k terms which involve pi and 137, or p and 1/alpha, and see if you can get any that get closer to 1/alpha. Power series like this are found in QED, and alpha is a dimentionless number, so numberology here is not as crazy as in the rest of physics. We might even discover something empirical (like the Balmer formula) which later physics might "explain."... 
1/a= 137.036020454 for factor (1+k^2) 137.03472 for factor (1+k^2k^4) 137.03477 for factor (1+k^2k^4+k^6) 
Internet, June, 28, 1999 The message "Re: Alpha collection" to newsgroup "sci.physics" 7l6mhf$5ma@dfwixnews11.ix.netcom.com 
I. Gorelik, Steven B. Harris  Development of two upper methods s(n+ms) = p^{2}/2 From the letter to newsgroup: ...our result: s(n+ps) = p^{2}/2 with 1/Alpha = 137.035990750 is very interesting, and is deep inside of experimental values. It is very possible that expression s(n+ps) = p^{2}/2 can be explained geometrically. This is very near to surface of 4d cylinder made by rotating electron in atom. pi/sqr(2) corresponds to base of cylinder in xy, and the same pi/sqr(2) corresponds to "height" of cylinder in zt. Cylinder is inclined to all axes x, y, z, t, so the exact expression for the left side of equation s(n+?*s) we'll find and prove something further, after studding 4d pictures... 
m=0, 1/a=137.036020454 m=1, 1/a=137.036010988 m=2, 1/a=137.036001533 m=3, 1/a=137.035992087 m=p, 1/a=137.035990750 
Internet, June, 29, 1999 The message "Re: Alpha collection" to newsgroup "sci.physics" 
Ivan Makarchenko  1/a^{2} = 137^{2} + p^{2}  1/a = 137.0360157  Newsgroup fido7.su.science, 2000, in Russian. 
Jerry Iuliano  He gives a lot of unperceivable numerology

???  I didn't study these pages yet. 
James G. Gilson  a(n_{1}, n_{2}) =
a(n_{1}, infinity)p(n_{1}n_{2})/p Feynman's conjecture of a relation between a, the fine structure constant, and p. 
a =a(137,29)= 0.0072973525318... 

Ivan Gorelik  Gravity fine structure constant, or gravity constant in Normalized Units.  G' = 1/Exp(a+1/a) or 2G' = 1/sh(a+1/a) 
November 1999. Look: Normalized Units, and page: The Earth's Core Is Overheated Emptiness 
Here is my letter Golden Section and Alpha numbers send to the newsgroup sci.physics 01.16.2001.
Hi!
Several days ago I tried to find the fine structure constant, which is ~ 1/137.036, but suddenly received the value of Golden Section, exactly. Does anybody know the relation between these numbers, as much as precise? If the relation would be simple and the precision is great, then we could find the precise value of fine structure constant.
Here is the way of my erroneous search of Alpha.
Alpha is the velocity of an electron in the hydrogen atom; let's denote it "a".
Velocity of light is 1.
When light made one revolve around the nucleus, the electron will cover only "a" fraction of circumference.
But modern physics must be a little modified. :)
Accordingly to Space Genetics particles do not exist constantly, but periodically flash in a set of dots, drawing so the trajectory of a particle.
Consequently, we don't know; is the electron rotated clockwise or vise versa?
Consequently, it can cover the "a" fraction of circumference or the "1a" fraction of circumference.
If we take into account the SR!!!, we'll receive the equation:
2a^3  3a^2  a + 1=0
which gives no Alpha, but generally known Golden Section number:
(sqr(5) +/ 1) / 2.
It is quite strange that in order to receive it I took into account the SR.

Paul Lutus wrote:
...
>The finestructure constant is empirical, Phi is not (it is exactly
>(sqrt(5)1)/2 ). There might be a relation, but it is very doubtful.
Me: I hope it exists.
> * Actually, and not surprising given its degree,
>2a^3  3a^2  a + 1 = 0 has three solutions:
...
> a = 1/2
> a = (1+sqrt(5))/2 (0.618034...)
> a = (1sqrt(5))/2 (1.618034...)
Me: I would take:
 a_1 = c/2, as clockwise.
 a_2 = 0.618034...*c, as visa versa.
 And I would though out a_3 = 1.618034...*c, as speculative tachion result.
Thank you and regards.
Helge Kragh. The finestructure constant before quantum mechanics. Eur. J. Phys.
24 (2003) 169–173. Online at stacks.iop.org/EJP/24/169,
M. J. Duff, L. B. Okun and G. Veneziano.
N.V. Makhaldiani and Z.K. Silagadze.
D. PALLE.
Jacob D. Bekenstein.
E. E. Krieckhaus.
To the page Alpha method.
To index of Space Genetics.