Reciprocal Symmetry and Correspondence between Relativistic and Quantum Mechanical Concepts

Mushfiq Ahmada

Department of Physics, Rajshahi University, Rajshahi-6205, Bangladesh

Abstract

We recall the principle of objectivity that physics should be independent of the quantities we define. We
have postulated that the study of motion in terms of slowness (reciprocal of velocity) is as valid as the study
in terms of velocity. We require that a kinematical theory should be reciprocal symmetric. We have given a
mathematical definition of reciprocal symmetry. It follows from this definition that velocities must have an
upper bound and slowness must have a lower bound. From group property requirement that the relative
slowness (which is a difference between two slowness) is also a slowness, we have shown that slowness are
discrete. The same motion may be represented by continuous quantities (like velocity) and by discrete
quantities (like slowness). A hyperbolic transform relates a Galilean velocity to a “Lorentz algebraic
velocity”, which obeys Lorentz algebra. Isomorphism between real numbers and imaginary numbers allow
us to replace all real numbers by corresponding imaginary numbers. This changes all hyperbolic functions
to circular functions. Circular functions are periodic and represent wave motion. Hyperbolic functions are
monotonic and represent particle motion. We have shown that real-imaginary symmetry establishes the
equivalence between particle representation and wave representation. We have also shown that application
of reciprocal symmetry on Planck’s distribution law gives Fermi-Dirac distribution law. For this we have
not invoked exclusion principle.

Key words: Reciprocity, Reciprocal Symmetry, Galilean velocity, Hyperbolic transformation, Lorentz
algebraic velocity, Discreteness, Reciprocal velocity and Slowness, Reciprocal distance and Proximity,
Discrete-Continuous Duality, Wave-Particle Duality, Einstein’s postulate, Lorentz transformation, Finite
difference equation, Boltzmann’s function, Planck’s distribution, Fermi-Dirac distribution

1. Introduction

Velocity is defined as the distance covered in unit time, V = x/t. Reciprocally, we may define slowness V’
as the reciprocal of V i.e. V’ = t/x. Physics does not depend on the quantities we define (objectivity).
Therefore, it should be possible to describe motion in terms of slowness just as it is possible in terms of
velocities1. If velocity has an upper bound, the corresponding slowness has a lower bound. The difference
between two slowness is also a slowness (group property). Therefore, the difference between two slowness

a
mushfiqahmad@ru.ac.bd
has a lower bound i.e. slowness are discrete. If a motion is described in terms of velocity, it will be
continuous. The same motion, if described in terms of slowness, will be discrete. Therefore, “discrete-
continuous duality” is another duality like wave-particle duality.

Imaginary numbers are no less real than real numbers. A quantity, which can be represented by a real
number may also be represented (with equal validity) by an imaginary number. If the real arguments of a
hyperbolic function are replaced by corresponding imaginary numbers, the function goes over to the
corresponding circular function. Therefore, we expect that a physical event which can be described in
terms of hyperbolic functions should also be describable in terms of circular functions and vise versa.
Circular functions are periodic and have wave aspects. Hyperbolic functions are monotonic and have
particle aspect. Quantum mechanics deals with “wave functions” and addresses itself primarily to the
discrete-wave aspects of motion. Galilean and relativistic kinematics look at the continuous-particle aspect.
A proper understanding of these dualities will, we hope, help us understand better the relation between
quantum mechanics and relativity. In this paper, we intend to study the relation between wave and particle
aspects of motion. We shall try to represent the same motion as “wave motion” represented by circular
functions and as “particle motion” represented by hyperbolic functions. We shall study, in some detail,
motion in one space dimension only. In the last section we shall briefly introduce the extension to the
general case. This will be in Part II below. The conceptual part will be in Part I.

In Part III we shall study the relation between reciprocity and quantum srtatistics.

Part I. Reciprocal Symmetry

2. Reciprocal Symmetry

Postulate 1: Description of a physical event in terms of a physical quantity v is equivalent to the

description in terms of the reciprocal quantity v’= 1/v

We need to give a mathematical form to the above postulate.

Definition 1. A function u ⊕ v will be called reciprocal symmetric if it remains invariant when the
arguments u and v are replaced by their reciprocals.
u ⊕ v = 1/u ⊕ 1 / v (2.1)

Postulate 2: The law of composition of velocities should be reciprocal symmetric.

Lorentz transformation (2.2) below is an example of a reciprocal symmetric function.
u/c − v/c
u/c ⊕ (-v/c ) = (2.2)
1 − u.v/c 2
(2.2) fulfills reciprocal symmetry condition (2.1).

c / u ⊕ (− c / v ) = u/c ⊕ (-v/c ) (2.3)

Another example of a reciprocal symmetric function is the reciprocal of (2.2).

1 1 − u.v / c 2
u/c ⊕ *(-v/c ) = = (2.4)
u / c ⊕ (−v / c) u / c − v / c

3. Reciprocal Symmetry and an Invariant under Addition

Theorem: u = c is an invariant under reciprocal symmetric laws of composition ⊕ or ⊕ * (3.1)

Proof: Let
c/u ⊕ (-c/v) = w/c (3.2)
Using (3.1)

u/c ⊕ (-v/c ) = w/c (3.3)

Therefore, using (3.2)

{c/u ⊕ (-c/v)} ⊕ (c/v ) = w/c ⊕ c/v (3.4)

Assuming ⊕ is associative, (-c/v) ⊕ (c/v ) = 0 and c/u ⊕ 0 = c/u , (3.4) gives

c/u = w/c ⊕ c/v (3.5)
Similarly using (3.3) we have
u/c = w/c ⊕ v/c (3.6)
Comparison between (3.5) and (3.6) shows
1
w/c ⊕ v/c = (3.7)
w/c ⊕ c/v
If v = c (3.7) gives
1
w/c ⊕ 1 = (3.8)
w/c ⊕ 1
Therefore,
w/c ⊕ 1 = 1 (3.9)
Therefore, v = c is an invariant under addition ⊕ . The same arguments will apply in case of ⊕ * .
Therefore, the theorem is proved.
4. Reciprocal Symmetry and Upper and Lower Bounds

With (2.2) as the law of composition and the values of u, v satisfying the condition − c ≤ u , v ≤ c we
define a set R (of relativistic numbers) so that the “sum” or “difference” of any two elements of R is also an
element of R. c and − c are the upper and lower bounds of R.
If we equate c to the speed of light, the set R above, together with theorem (3.2) fulfills Einstein’s
postulate.

We rewrite (2.4) as below

1 − u '.v' / c'2
u ' /c'⊕ * (-v' /c') = (4.1)
u ' / c'−v' / c'
Using (4.1) as the law of composition and the values of u ' , v' satisfying the condition c' ≤| u '|, | v'| . We
define a set Q (of quantum mechanical numbers) so that the “sum” or “difference” of any two elements of
Q is also an element of Q. c is the lower bounds of magnitudes of the elements Q.

5. Discreteness of the Reciprocal Set

Definition: A set of numbers will be called discrete if the difference between two elements of the set is
never 0.

Theorem: Q is a discrete set of numbers

Proof: Let c' ≠ 0 . According to the statement following (4.1) the difference between any two elements of
Q is never 0. Therefore, Q is a set of discrete numbers.

Part II. Kinematics

6. Galilean Velocity and Lorentz Algebraic Velocity or Slowness

To exploit this reciprocal symmetry, it will be convenient to look for a representation in which the velocity
V does not vary from 0 to ∞ , but (the transform v) varies from 0 to c, where c is a finite velocity. In this
representation v’ will vary from ∞ to c’=1/c. The transformations v/c = tanh(V/c) and v’/c’ = 1/tanh(V/c)
are suitable. We shall call V the Galilean velocity and v the corresponding Lorentz algebraic velocity.
Consider the transforms
 1 1+ v /c
V / c = ln (6.1)
2 1− v /c
and

1  1 + v ' / c' 
V / c = ln −  (6.2)
2  1 − v' / c' 
where v'= 1 / v and c'= 1 / c . (6.1) and (6.2) are identities (see (11.6) below) if
v / c = tanh (V / c ) and v' / c' = 1 / tanh (V / c ) (6.3)
The bounds are
− 1 ≤ v / c ≤ 1 and | v' / c'| ≥ 1 (6.4)
The relative Galilean velocity W = U – V has transforms
1 1+ w/c
W / c = ln (6.4)
2 1− w/ c
and

1  1 + w' / c' 
W /c = ln −  (6.6)
2  1 − w' / c' 
where
u/c −v/c
w / c = (u / c) ⊕ (−v / c) = (6.7)
1 − u.v / c 2
and

1 − u '.v' / c'2
w' / c' = (u ' / c') ⊕ *(− v' / c') = (6.8)
u ' / c'−v' / c'
(6.7) and (6.8) are Reciprocal Symmetric i.e. they remains invariant under inversion v / c → c / v etc.
Reciprocal symmetry (2.3) guarantees consistency between (6.7) and (6.8) so that w' / c ' = c / w . (6.7) and
(6.8) also ensure that
| w' / c'| ≥1 if | u ' / c'|, | v' / c'| ≥1 (6.9)
(6.9) translates the statement that the difference between two slowness (relative slowness) is a slowness.
(6.9) also shows that slowness are discrete.

7. Distance, Proximity and Discrete Time

The distance covered in time t is2

t /T
1 1+ v / c  1 1+ x / L
Vt / L = ln  = ln (7.1)
2 1− v /c  2 1− x / L
where T is the quantum of time and L = cT and
 t /T
1+ v / c 
  −1
x / L = (t / T ) ⊗ (v / c ) =  1− v /c 
(7.2)
t /T
1+ v / c 
  +1
1− v / c 
Also
t /T
1  1 + v' / c'  1  1 + x' / L' 
Vt / L = Ln −  = ln −  (7.3)
2  1 − v ' / c'  2  1 − x' / l ' 
where x'= 1 / x , L'= 1 / L . x' is the “proximity” corresponding to x . And
t /T
 1 + v' / c' 
−  +1
x ' / L' =  1 − v' / c' 
(7.4)
t /T
 1 + v' / c' 
 −  −1
 1 − v' / c' 
x' / L' is complex if t / T is fractional. When t / T is an even/odd integer
t /T
 1 + v' / c' 
  +1
x' / L' = (t / T )⊗' (v' / c') =  1 − v ' / c' 
. x ' / L' ≥1 for t / T = even integer (7.5)
t /T
 1 + v' / c' 
  −1
 1 − v' / c' 

t /T
 1 + v' / c' 
  −1
x' / L' = (t / T ) ⊗ (v' / c') =  1 − v' / c' 
. x ' / L ' ≤1 for t / T = odd integer (7.6)
t /T
 1 + v' / c' 
  +1
 1 − v' / c' 
The distance covered (or proximity lost) by relative slowness w' is

z ' / L' = (t / T )⊗' (w' / c') . z ' / L' ≥1 for t / T an even integer (7.7)

z ' / L' = (t / T ) ⊗ (w' / c') . z ' / L' ≤1 for t / T an odd integer (7.8)

Reciprocal symmetric multiplication “ ⊗ ” is distributive with respect to reciprocal symmetric addition

“⊕”

(t / T )⊗'{(u ' / c')⊕' (− v' / c')} = {(t / T )⊗' (u ' / c')}⊕'{(t / T )⊗' (− v' / c')} (7.9)

(t / T ) ⊗ {(u ' / c') ⊕ (− v' / c')} = {(t / T ) ⊗ (u ' / c')} ⊕ {(t / T ) ⊗ (− v' / c')} (7.10)
The above relations show that the distance covered (or proximity lost) by relative slowness w' is the same
as the relative distance covered (or proximity lost) by velocities/slowness u and v (or u ' and v' ).
Therefore, proximity z ' is discrete.

8. Circular and Hyperbolic Transforms of Velocity

There exist a one-one correspondence between hyperbolic functions and circular functions; they have
similar properties. Therefore, we expect that a physical event which can be described in terms of hyperbolic
functions should also be describable in terms of circular functions and vise versa. We shall explore this in
this Section and in the following Sections. We re-write (2.1) as
1 + (v / c )
V / c = ln (8.1)
1 − (v / c )
2

and as
1 + i.(v~ / c )
V / c = −i. ln (8.2)
1 + (v~ / c )
2

Like (2.1) and (8.1), (8.2) is also an identity if

v~ / c = tan (V / c ) (8.3)

9. Wave and Particle Representations of Distance Traveled

Corresponding to (7.1) and using (8.2)

t /T
 1 + iv~ / c  1 + i~
x/L ~
Vt / L = −i. ln  = −i ln − i ln d (~
x) (9.1)
 ~ 2 
 1 + (v / c ) 1 + (~
x / L)
2

~
Therefore, in this representation the distance function, d (~
x ) , is given by
~
d (~
x ) = cos(Vt / L ) + iSin(Vt / L ) (9.2)
In particle (monotonic) representation the same motion is represented by
t /T
 1+ v / c  1+ x / L
Vt / L = ln  = ln = ln d (x ) (9.3)
 
 1 − (v / c ) 1 − (x / L )
2 2

where

d ( x ) = exp(Vt / L ) (9.4)

10. Galilean Transformation and Wave and Particle Representations of Lorentz Transformation
We introduce another representation of the and distance covered by
1 + i.(~
x / L) 1 − i.v~t / L
X = −i.L/ . ln and − Vt = −i.L. ln (10.1)
1 + (~x / L) 1 + (vt / L )
2 2

(10.1) differs from (9.1) by the introduction of L = ct in place of L = cT of Section 3.

The Galilean relative distance X’ as observed by an observer moving with velocity V in time t is X − Vt .
Using (10.1) we have

1 + i.(~
x / L ) 1 − i.v~t / L c~t '+i.(~x ')
X ' = X − Vt = −i.L. ln . = −i (ct ). ln (10.2)
1 + (~x / L) 1 + (vt / L ) (ct ) 2 + (~
x)
2 2 2

where
~~ 2(
~ t + x .v / c
t '=
)
and ~
x'=
~
x − v~t
(10.3)
1 + (v~ / c ) 1 + (v~ / c )
2 2

For hyperbolic transforms the relations corresponding to (10.1) are

1 + (x / L ) 1 − vt / L
X = L/ . ln and − Vt = L. ln (10.4)
1 − (x / L ) 1 − (vt / L )
2 2

And the relations corresponding to (10.2) and (10.3) are

ct '+ ( x')
X ' = X − Vt = (ct ). ln (10.5)
(ct ) 2 − ( x )
2

where

t' =
(
t − x.v / c 2 ) and x' = x − vt
(10.6)
1 − (v / c ) 1 − (v / c )
2 2

(10.6) is Lorentz transformation. (10.3) is the analogues of Lorentz transformations in wave representation.

11. Galilean, Wave and Particle Representations of Lorentz Invariance

(10.3) gives the invariance

(c ~
t ' ) 2 + (~
x ') = (ct ) + (~
x)
2 2 2
(11.1)
(10.6) gives Lorentz invariance3

(ct ' ) 2 − ( x') = (ct ) − ( x )

2 2 2
(11.2)

To see the images of the right hand sides of (11.1) in Galilean space, consider using (10.1) and L = ct
1 + i.(~
x / L ) 1 − i.(~
x / L) (ct )2 + ~x 2 . 1
X − X = −i.L. ln = −i.(ct ). ln = 0 (11.3)
1 + (~x / L ) 1 + (~ x / L) (ct )2 + (~x )2 (ct )2 + (~x )2
2 2
Similarly using (10.4)

X − X = (ct )ln
(ct )2 − x 2 . 1
=0 (11.4)
(ct )2 − (x )2 (ct )2 − (x )2
Therefore, the right hand sides of (11.1) and (11.2) correspond to the right hand side of (11.5) below. Using
(10.1) to (10.6), we observe that the left hand sides of (11.1) and (11.2) correspond to the left hand side of
(11.5).

( X − Vt ) − ( X − Vt ) = X − X (11.5)
Both of (11.1) and (11.2) are images of (11.5).

12. General Reciprocal Symmetric Transformation

we have seen in (2.9) that Lorentz transformation (2.7) is reciprocal symmetric. But the general Lorentz
transformation

{
U 1 − (V/c ) + 1 − 1 − (V/c )
2 2
} U.V
V2
V−V
U +ˆ (−V) = 2
(12.1)
1 − U.V/c
is not reciprocal symmetric. We need a reciprocal symmetric general transformation to replace Lorentz
transformation. The following transformation has the properties required of Lorentz transformation and is
reciprocal symmetric4.
r U + (−V ) − i.UxV / c
U + (−V) = (12.2)
1 − U.V/c 2
The reciprocals of U and V are given (for arbitrary G) by

UxG VxG
G +i G+i
U* = c and V* =
c (12.3)
U.G/c 2 V.G/c 2
so that

(U / c ).(U * / c) = 1 (12.4)
and
r r
U * +(−V*) = U + (−V ) (12.5)

Part III. Quantum Statistics

13. Abstract of Part III

Discreteness of energy has been exploited to replace Boltzmann's differential equation by the
corresponding finite difference equation. The difference equation is, then, symmetrized so that the equation
remains invariant when step d is replaced by –d. This symmetry adds an extra degree of internal freedom
and the solutions of this equation come in pairs of reciprocals. One of the solutions gives the classical
Boltzmann’s function and gives Planck’s law of distribution. The other solution is the reciprocal of the
classical solution and gives the corresponding Fermi-Dirac relation. We have not invoked exclusion
principle to derive Fermi-Dirac relation. Symmetrization of Boltzmann’s equation is enough. We have also
shown the correspondence between reciprocity and phase; and that change of reciprocity is equivalent to a
change of phase by π /2.

14. Introduction to Part III

Fermi-Dirac distribution is based on Pauli’s exclusion principle obeyed by half-integral spin particles5. It is
not clear why half-integral spin particles obey exclusion principle. A correspondence between reciprocity
and discreteness has already been observed. We shall also see below (Section 15) the relation between
reciprocity and phase. Discreteness is a feature of quantum mechanics, and the phase of the function
contains vital information of the state. It is, then, reasonable to expect that there exists a relation between
reciprocity, reciprocal symmetry (Section I) and quantum statistics. We want to study this below.
Boltzmann’s relation is an exponential function which satisfies a differential equation. To explore its
symmetry properties we shall replace the differential equation by the corresponding finite difference
equation and symmetrized it. We intend to study the properties of the solutions of this symmetric equation,
and the implications of these properties in quantum statistics.

15. Reciprocity and Odd Integral Phase

A rotation in reciprocity space may be generated by the operation below.

v(θ ) + i tan ϕ
v(θ + iϕ ) = (15.1)
1 + i.v(θ ). tan ϕ
If v(θ ) = tanh (θ ) (24.5) gives v(θ + iϕ ) = tanh(θ + i.ϕ ) .
If v(θ ) = i tan (θ ) (24.5) gives v(θ + iϕ ) = i. tan(θ + ϕ ) .
For s an odd integer
1 −1
tanh(θ + i.π .s / 2) = and tan(θ + π .s / 2) = (15.2)
tanh(θ ) tan(θ )

16. Differential Equation and Classical Boltzmann’s Function

Classical Boltzmann's function g satisfies the differential equation6

 dg
= − wg = − w. exp(− w.H ) (16.1)
dH
where

g = e − wH (16.2)

The number N (H ) of particles in energy level H is given by Boltzmann’s probability distribution

7
relation

N ( H ) = N (0)e − wH . (16.3)

where N (0) is the number of particles when H=0

17. Planck’s Radiation Formula

We invoke Planck’s hypothesis8 to write H as an integral multiple of δ

H = n.δ where n is an integer (17.1)
In case of black body radiation the total number of particles is
∞
1
N = N 0 ∑ e − pwδ = N 0 (17.2)
p =0 1 − e − wδ
The total energy is
∞
e − wδ
E = N 0δ ∑ pe − pwδ = N 0δ (17.3)
p =1 (1 − e ) − wδ 2

Therefore, the average energy of a particle is

δ
E/N = wδ
(17.4)
e −1

18. Discrete Energy and Boltzmann's Symmetric Finite Difference Equation

In (16.1), which gives Boltzmann’s function, energy H has been treated as a continuous variable. To take
into account discreteness of H (17.1) we replace the differential coefficient in (16.1) by the corresponding
difference quotient
g (H + δ ,δ ) − g (H ,δ )
(18.1)
δ
(18.1) is not symmetric. To make it symmetric we replace the difference quotient in (18.1) by the
corresponding symmetric difference quotient

Dg ± ( H , δ ) g ± ( H + δ , δ ) − g ± ( H − δ , δ )
= (18.2)
D( H , δ ) 2δ
We now impose the symmetry (under negation) condition

Dg ± ( H , δ ) Dg ± ( H ,−δ )
= (18.3)
D( H , δ ) D( H ,−δ )

Using (18.2) and (18.3) we find the requirement

g ± ( H ,−δ ) = g ± ( H , δ ) (18.4)
(18.4) is symmetry under negation. We shall see below that it is also symmetry under reciprocation since

δ in g ± ( H , δ ) comes as an exponent. Therefore, combining both negation and reciprocation (18.4) may
be called reflection symmetry.

We now replace (16.1) by9. [We drop the δ in g ± ( H , δ ) and write simply g ± ( H ) ]
Dg ± ( H )
= −W .g ± ( H ) (18.5)
D( H , δ )
We require that (18.5) should go over to (16.1) in the limit δ →0

19. Solution of Difference Equation

One of the solutions of (18.5) is

Sinh(wδ )
g + ( H ) = e − wH with W = (19.1)
δ
To verify (19.1), using (18.2) we write

Dg ± ( H ) e wδ − e − wδ − wH Sinh( wδ ) − wH
=− e =− e = −Wg + ( H ) (19.2)
D( H , δ ) 2δ δ
Agreement between (18.5) and (16.1) is achieved since

Dg ± ( H ) dg
δ
→
→0
g and W δ
→ w
→0
(19.3)
D( H , δ ) dH

20. Boltzmann's Reciprocal Symmetric Functions

wδ
We observer the symmetry in W under the reflection e → − e − wδ
e wδ − e − wδ
W (e wδ
)=
2δ
= W − e − wδ ( ) (20.1)

Symmetry (20.1) gives an extra degree of freedom. We exploit this to find a second solution of (18.5). This
is

{
g − ( H ) = − e wδ } H /δ
= exp{(2π .s − i + wδ )(H / δ )} (20.2)
Where s − is a half-integer. Both the solutions (19.1) and (20.2) may be written together
g ± ( H ) = exp{(2π s ± i m wδ )( H / δ )} (20.3)

Where s + is an integer and s − is a half-integer. g + (H ) and g − (H ) are related through reciprocity

relation

g + ( H ).g − ( H ) = (−1) H / δ = (−1) n (20.4)

21. Symmetry Properties of Boltzmann’s Functions

Splitting of Boltzmann’s function g of (16.2) into g + (H ) and g − (H ) is a consequence of reflection

symmetry (20.1) which, in turn is a consequence of (18.3) and (18.4).

22. Hidden Parameter

The solution of (16.1) is independent of δ , but the solutions (20.3) of (18.5) are δ dependent. The hidden
dependence has come out in course of the transition from (16.1) to (18.5). The dependence δ becomes
more visible if we write in place of (20.3)
H /δ
 1 m tanh (wδ / 2 ) 
g ± ( H ) = ±  (22.1)
 1 ± tanh (wδ / 2 ) 

23. Normalization and Negation Energy

In case of g − ( H ) the number of particles corresponding to (17.2) is

∞
N − = N 0 ∑ − e wδ ( ) p
(23.1)
p =0

This sum diverges for wδ ≥ 0 . Therefore, we impose the condition wδ ≤ 0 and write wδ '≥ 0
where δ ' = −δ . This gives
∞
1
(
N − = N 0 ∑ − e − wδ ' ) p
= N0
1 + e − wδ '
with δ ' = −δ ≥ 0 (23.2)
p =0

Using (23.2) and following the same procedure as in Section 5, the average energy is
δ'
E− / N − = wδ '
(23.3)
e +1

24. Symmetric Average Energy

We shall call < H ± > the average energy corresponding to g ± . Using results of Sections 5 and 11
δ
< H + >= wδ
with δ ≥0 (24.1)
e −1
δ'
< H − >= wδ '
with δ '≥ 0 (24.2)
e +1
If we write δ = hω and w = 1 / kT the above relations become

hω
< H + >= hω / kT
with ω≥0 (24.3)
e −1
And
hω '
< H − >= hω ' / kT
with ω ' = −ω ≥ 0 (24.4)
e +1

25. Boltzmann’s Reciprocal Function and Exclusion Principle

− wH
Boltzmann’s function g + ( H ) = e forbids particles from occupying states with high positive values
wH
of H . Similarly | g − ( H ) |= e forbids particles from occupying states with high negative values of H .

With this reciprocal Boltzmann’s function we do not need Dirac’s Hole Theory and its ‘filled up’ states;
wH
function | g − ( H ) |= e will stop particles from dropping into negative energy states.

26. Conclusion to Part III

We have observed equivalence between phase and reciprocity. We have seen that when Boltzmann’s
differential equation is replaced by the corresponding symmetric finite difference equation, the solution
splits into two functions which are images (negative reciprocal) of each other. One of them gives Planck’s
distribution relation. The other solution contains the equivalent of Exclusion Principle, involves negative
energy and gives Fermi-Dirac relation. We have observed a relation between reciprocity and quantum
statistics.

27. General Conclusion

We have defined Reciprocal Symmetry and we have applied it to find an invariant under addition. This
invariant quantity is either an upper bound of a set of quantities or a lower bound of the reciprocals of this
set. The set bounded above together with this invariant quantity corresponds to Einstein’s postulate. The
reciprocal set discrete. Therefore, we find that discreteness is directly related Einstein’s postulate or
Lorentz transformation. The distance covered by a moving body is continuous if it is represented in terms
of velocity and time. If it is represented in terms of slowness and time, the distance covered becomes
complex. It becomes real and agrees with the familiar value for integral values of time. In Part III we have
seen that the usual Boltzmann’s function gives Plank’s law of distribution while the reciprocal function
gives Fermi-Dirac law of distribution. Reciprocals and reciprocal symmetry relate Einstein’s postulate and
Lorentz transformation to discreteness, wave motion and Fermi-Dirac statistics. Reciprocal symmetry,
therefore, relates quantum mechanics to relativity.

28. References

1
Mushfiq Ahmad. Discrete and Continuous Representation of the Same Motion. Physics Essays. March
2009; Vol. 22, No. 1.
2
Mushfiq Ahmad. Reciprocal Symmetry and its Relation to Einstein’s Postulate, Lorentz Transformation
and Discreteness . Journal of Scientific Research. I(2), 270-274 (2009). DOI: 10.3329/jsr.v1i2.1875
3
C. Moller. The theory of relativity. Oxford: Clarendon Press, 1952. 1952
4
Mushfiq Ahmad and Mohammad Shah Alam. Relativistic Requirement and Comparison between
Reciprocal Symmetric Transformation and Lorentz Transformation. Physics Essays. 2009; Vol. 22, No. 2.
5
Wikipedia, the free encyclopedia. Fermi-Dirac Statistics.
http://en.wikipedia.org/wiki/Fermi%E2%80%93Dirac_statistics
6
Rudolf Kurth. Axiomatics of Classical Statistical Mechanics. Pergamon Press. 1960.
7
http://www.answers.com/topic/boltzmann-distribution
8
Feynman, Leighton, Sands. Feynman Lectures on Physics. Addison-Wesley Pub. Co.
9
Mushfiq AhmadReciprocal Symmetric and Origin of Quantum Statistics.
http://www.arxiv.org/abs/physics/0703194