
2 1 Introduction
x of M, such that the time like vector v is colinear to a probability current j = ρv ∈
M(ρ > 0) of a particle.
The gauges U(1) and SU(2) are the groups of the rotations upon themselves
of the plane generated by {n
1
, n
2
}, or “spin plane” [2], and the three-space E
3
( j)
orthogonal to j generated by {n
1
, n
2
, n
3
} respectively.
A momentum-energy tensor T = ρT
0
associated with one of these two gauges, is
defined by a linear application n ∈ M → T
0
(n) ∈ M which implies the product of a
suitable physical constant by the infinitesimal rotation upon itself of the three-spaces
generated by {v, n
1
, n
2
} or {n
1
, n
2
, n
3
} respectively.
The trace of this energy-momentum tensor appears in the first term L
I
of the
lagrangian of the chosen theory.
Let us denote a · b (written a
μ
b
μ
when a galilean frame {e
μ
},μ = 0, 1, 2, 3, is
used) the scalar product of two vectors a, b ∈ M.
The second term L
II
contains terms in the form A · j, B · j, and also sums
W
k
· j
k
( j
k
= ρn
k
, k = 1, 2, 3) where A is an electromagnetic potential, B, W
k
∈
M are vectors of space-time (bosons). Note that a term in the form B · j
i
where j
i
is
isotropic appears in the electroweak theory.
The invariance in a gauge transformation implies a change in the expression of
A and the bosons, related to the rotation of the spin plane or the three-space E
3
( j),
and, in this last case, a change in the field associated with the bosons. These changes
are well known. What is less or not at all known is the fact that these changes are
related to the rotation of the spin plane and the three-space E
3
( j) in the case of
U(1) and SU(2).
It does not seem possible to treat the SU(3) gauge, as it is used in chromodynamics,
with a complete interpretation in the geometry of M. But it is possible, without
changing the standard l agrangian of the this theory, to replace this gauge by the
direct product of three SU(2) × U(1) gauges, simply by the change of the eighth
boson G
8
into G
8
/
√
3, associated with the multiplication by
√
3 of the eighth Gell-
Mann matrice followed by a suitable decomposition of this matrice into the sum of
two matrices each one similar to the third isospin matrice.
If the chromodymanics theory, as it is, at present constructed, were confirmed
by experiments, and if the possibility of the replacement of G
8
by G
8
/
√
3were
infirm, one would face the following paradox. Two theories (Dirac electron, Glashow-
Weinberg-Salam electroweak), widely confirmed by experiment, would be entirely
enclosed in the geometry of space-time and a third one would not be enclosed.
On the contrary if this replacement were experimentally confirmed, the real space-
time algebra would appear not only as a tool allowing noticeable simplifications in
the calculations of confirmed theories (for the hydrogenic atoms see for example
[5]), but also as a way to put in evidence fundamental properties of these theories,
and at least like the indispensable language of quantum mechanics.
In an Addendum, we have given a geometrical construction of electromagnetism,
which may be applied as well to the electromagnetic properties of charges endowed
with a trajectory, like the ones of the charged particles in quantum mechanics whose
the presence is based on a probability.