n
=
3
{\displaystyle n=3}
,
p
=
2
{\displaystyle p=2}
. Theorem of Gauss.[ edit ]
By (1):
∫
∫
d
u
(
1
)
d
u
(
2
)
[
∂
(
x
(
2
)
x
(
3
)
)
∂
(
u
(
1
)
u
(
2
)
)
A
23
+
∂
(
x
(
3
)
x
(
1
)
)
∂
(
u
(
1
)
u
(
2
)
)
A
31
+
∂
(
x
(
1
)
x
(
2
)
)
∂
(
u
(
1
)
u
(
2
)
)
A
12
]
=
=
∫
∫
∫
d
v
(
1
)
d
v
(
2
)
d
v
(
3
)
∂
(
x
(
1
)
x
(
2
)
x
(
3
)
)
∂
(
v
(
1
)
v
(
2
)
v
(
3
)
)
A
123
,
{\displaystyle {\begin{aligned}\int \int du^{(1)}du^{(2)}\left[{\frac {\partial \left(x^{(2)}x^{(3)}\right)}{\partial \left(u^{(1)}u^{(2)}\right)}}A_{23}+{\frac {\partial \left(x^{(3)}x^{(1)}\right)}{\partial \left(u^{(1)}u^{(2)}\right)}}A_{31}+{\frac {\partial \left(x^{(1)}x^{(2)}\right)}{\partial \left(u^{(1)}u^{(2)}\right)}}A_{12}\right]=\\=\int \int \int dv^{(1)}dv^{(2)}dv^{(3)}{\frac {\partial \left(x^{(1)}x^{(2)}x^{(3)}\right)}{\partial \left(v^{(1)}v^{(2)}v^{(3)}\right)}}A_{123},\end{aligned}}}
,
A
123
=
∂
∂
x
(
3
)
A
23
−
∂
∂
x
(
2
)
A
13
+
∂
∂
x
(
1
)
A
23
,
A
α
β
=
−
A
β
α
{\displaystyle A_{123}={\frac {\partial }{\partial x^{(3)}}}A_{23}-{\frac {\partial }{\partial x^{(2)}}}A_{13}+{\frac {\partial }{\partial x^{(1)}}}A_{23},\quad A_{\alpha \beta }=-A_{\beta \alpha }}
Introducing the supplements
1
c
A
23
=
B
(
1
)
,
1
c
A
31
=
B
(
2
)
,
1
c
A
12
=
B
(
3
)
,
{\displaystyle {\sqrt {\frac {1}{c}}}A_{23}=B^{(1)},\quad {\sqrt {\frac {1}{c}}}A_{31}=B^{(2)},\quad {\sqrt {\frac {1}{c}}}A_{12}=B^{(3)},}
c
b
∂
(
x
(
2
)
x
(
3
)
)
∂
(
u
(
1
)
u
(
2
)
)
=
N
1
,
c
b
∂
(
x
(
3
)
x
(
1
)
)
∂
(
u
(
1
)
u
(
2
)
)
=
N
2
,
c
b
∂
(
x
(
1
)
x
(
2
)
)
∂
(
u
(
1
)
u
(
2
)
)
=
N
3
,
{\displaystyle {\sqrt {\frac {c}{b}}}{\frac {\partial \left(x^{(2)}x^{(3)}\right)}{\partial \left(u^{(1)}u^{(2)}\right)}}=N_{1},\quad {\sqrt {\frac {c}{b}}}{\frac {\partial \left(x^{(3)}x^{(1)}\right)}{\partial \left(u^{(1)}u^{(2)}\right)}}=N_{2},\quad {\sqrt {\frac {c}{b}}}{\frac {\partial \left(x^{(1)}x^{(2)}\right)}{\partial \left(u^{(1)}u^{(2)}\right)}}=N_{3},}
where
b
=
|
b
11
b
12
b
21
b
22
|
=
|
∑
c
α
β
∂
x
(
α
)
∂
u
(
1
)
∂
x
(
β
)
∂
u
(
1
)
∑
c
α
β
∂
x
(
α
)
∂
u
(
1
)
∂
x
(
β
)
∂
u
(
1
)
∑
c
α
β
∂
x
(
α
)
∂
u
(
2
)
∂
x
(
β
)
∂
u
(
1
)
∑
c
α
β
∂
x
(
α
)
∂
u
(
2
)
∂
x
(
β
)
∂
u
(
2
)
|
{\displaystyle b=\left|{\begin{matrix}b_{11}&b_{12}\\b_{21}&b_{22}\end{matrix}}\right|=\left|{\begin{matrix}\sum c_{\alpha \beta }{\frac {\partial x^{(\alpha )}}{\partial u^{(1)}}}{\frac {\partial x^{(\beta )}}{\partial u^{(1)}}}&\sum c_{\alpha \beta }{\frac {\partial x^{(\alpha )}}{\partial u^{(1)}}}{\frac {\partial x^{(\beta )}}{\partial u^{(1)}}}\\\sum c_{\alpha \beta }{\frac {\partial x^{(\alpha )}}{\partial u^{(2)}}}{\frac {\partial x^{(\beta )}}{\partial u^{(1)}}}&\sum c_{\alpha \beta }{\frac {\partial x^{(\alpha )}}{\partial u^{(2)}}}{\frac {\partial x^{(\beta )}}{\partial u^{(2)}}}\end{matrix}}\right|}
is the discriminant of the arc on
M
2
{\displaystyle M_{2}}
,[ 1] thus because of
d
S
2
=
b
d
u
(
1
)
d
u
(
2
)
,
d
S
3
=
c
d
x
(
1
)
d
x
(
2
)
d
x
(
3
)
{\displaystyle dS_{2}={\sqrt {b}}du^{(1)}du^{(2)},\quad dS_{3}={\sqrt {c}}dx^{(1)}dx^{(2)}dx^{(3)}}
:
it follows
∫
∫
d
S
2
[
N
1
B
(
1
)
+
N
2
B
(
2
)
+
N
3
B
(
3
)
]
=
=
∫
∫
∫
d
S
3
1
c
[
∂
∂
x
(
1
)
(
c
B
(
1
)
)
+
∂
∂
x
(
2
)
(
c
B
(
2
)
)
+
∂
∂
x
(
3
)
(
c
B
(
3
)
)
]
{\displaystyle {\begin{aligned}&\int \int dS_{2}\left[N_{1}B^{(1)}+N_{2}B^{(2)}+N_{3}B^{(3)}\right]=\\&\quad =\int \int \int dS_{3}{\frac {1}{\sqrt {c}}}\left[{\frac {\partial }{\partial x^{(1)}}}\left({\sqrt {c}}B^{(1)}\right)+{\frac {\partial }{\partial x^{(2)}}}\left({\sqrt {c}}B^{(2)}\right)+{\frac {\partial }{\partial x^{(3)}}}\left({\sqrt {c}}B^{(3)}\right)\right]\end{aligned}}}
Here, the normal
N
{\displaystyle N}
goes to the exterior. For the generalized divergence:
d
i
v
(
B
(
1
)
B
(
2
)
B
(
3
)
)
=
1
c
∑
α
∂
∂
x
(
α
)
(
c
B
(
α
)
)
{\displaystyle \mathrm {div} \left(B^{(1)}B^{(2)}B^{(3)}\right)={\frac {1}{\sqrt {c}}}\sum _{\alpha }{\frac {\partial }{\partial x^{(\alpha )}}}\left({\sqrt {c}}B^{(\alpha )}\right)}
or introducing the system reciprocal to
B
(
α
)
{\displaystyle B^{(\alpha )}}
B
(
α
)
=
∑
β
c
(
α
β
)
B
β
:
{\displaystyle B^{(\alpha )}=\sum _{\beta }c^{(\alpha \beta )}B_{\beta }:}
:
it follows
d
i
v
(
B
1
B
2
B
3
)
=
1
c
∑
α
∂
∂
x
(
α
)
(
c
∑
β
c
(
α
β
)
B
β
)
{\displaystyle \mathrm {div} \left(B_{1}B_{2}B_{3}\right)={\frac {1}{\sqrt {c}}}\sum _{\alpha }{\frac {\partial }{\partial x^{(\alpha )}}}\left({\sqrt {c}}\sum _{\beta }c^{(\alpha \beta )}B_{\beta }\right)}
or eventually, if
B
β
=
B
β
=
∂
B
∂
x
(
β
)
{\displaystyle B_{\beta }=\mathrm {B_{\beta }} ={\frac {\partial B}{\partial x^{(\beta )}}}}
can be represented as gradient of a scalar (invariant) quantity
B
{\displaystyle B}
:
d
i
v
(
B
1
B
2
B
3
)
=
Δ
B
=
1
c
∑
α
∂
∂
x
(
α
)
(
c
∑
β
c
(
α
β
)
∂
B
∂
x
(
β
)
)
{\displaystyle \mathrm {div} \left(\mathrm {B} _{1}\mathrm {B} _{2}\mathrm {B} _{3}\right)=\Delta B={\frac {1}{\sqrt {c}}}\sum _{\alpha }{\frac {\partial }{\partial x^{(\alpha )}}}\left({\sqrt {c}}\sum _{\beta }c^{(\alpha \beta )}{\frac {\partial B}{\partial x^{(\beta )}}}\right)}
[ 2]
n
=
3
{\displaystyle n=3}
,
p
=
1
{\displaystyle p=1}
. Theorem of Stokes.[ edit ]
By (1):
∫
d
u
[
∂
x
(
1
)
∂
u
A
1
+
∂
x
(
2
)
∂
u
A
2
+
∂
x
(
3
)
∂
u
A
3
]
=
=
∫
∫
d
v
(
1
)
d
v
(
2
)
[
∂
(
x
(
1
)
x
(
3
)
)
∂
(
v
(
1
)
v
(
2
)
)
A
23
+
∂
(
x
(
3
)
x
(
1
)
)
∂
(
v
(
1
)
v
(
2
)
)
A
31
+
∂
(
x
(
1
)
x
(
2
)
)
∂
(
v
(
1
)
v
(
2
)
)
A
12
]
{\displaystyle {\begin{aligned}&\int du\left[{\frac {\partial x^{(1)}}{\partial u}}A_{1}+{\frac {\partial x^{(2)}}{\partial u}}A_{2}+{\frac {\partial x^{(3)}}{\partial u}}A_{3}\right]=\\&\quad =\int \int dv^{(1)}dv^{(2)}\left[{\frac {\partial \left(x^{(1)}x^{(3)}\right)}{\partial \left(v^{(1)}v^{(2)}\right)}}\mathrm {A} _{23}+{\frac {\partial \left(x^{(3)}x^{(1)}\right)}{\partial \left(v^{(1)}v^{(2)}\right)}}\mathrm {A} _{31}+{\frac {\partial \left(x^{(1)}x^{(2)}\right)}{\partial \left(v^{(1)}v^{(2)}\right)}}\mathrm {A} _{12}\right]\end{aligned}}}
A
23
=
∂
A
2
∂
x
(
3
)
−
∂
A
3
∂
x
(
2
)
,
A
31
=
∂
A
3
∂
x
(
1
)
−
∂
A
1
∂
x
(
3
)
,
A
12
=
∂
A
1
∂
x
(
2
)
−
∂
A
2
∂
x
(
1
)
{\displaystyle \mathrm {A} _{23}={\frac {\partial A_{2}}{\partial x^{(3)}}}-{\frac {\partial A_{3}}{\partial x^{(2)}}},\quad \mathrm {A} _{31}={\frac {\partial A_{3}}{\partial x^{(1)}}}-{\frac {\partial A_{1}}{\partial x^{(3)}}},\quad \mathrm {A} _{12}={\frac {\partial A_{1}}{\partial x^{(2)}}}-{\frac {\partial A_{2}}{\partial x^{(1)}}}}
or
∫
d
s
[
d
x
(
1
)
d
s
A
1
+
d
x
(
2
)
d
s
A
2
+
d
x
(
3
)
d
s
A
3
]
=
=
∫
∫
d
S
2
[
N
1
1
c
(
∂
A
2
∂
x
(
3
)
−
∂
A
3
∂
x
(
2
)
)
+
N
2
1
c
(
∂
A
3
∂
x
(
1
)
−
∂
A
1
∂
x
(
3
)
)
+
N
3
1
c
(
∂
A
1
∂
x
(
2
)
−
∂
A
2
∂
x
(
1
)
)
]
{\displaystyle {\begin{aligned}&\int ds\left[{\frac {dx^{(1)}}{ds}}A_{1}+{\frac {dx^{(2)}}{ds}}A_{2}+{\frac {dx^{(3)}}{ds}}A_{3}\right]=\\&\quad =\int \int dS_{2}\left[N_{1}{\frac {1}{\sqrt {c}}}\left({\frac {\partial A_{2}}{\partial x^{(3)}}}-{\frac {\partial A_{3}}{\partial x^{(2)}}}\right)+N_{2}{\frac {1}{\sqrt {c}}}\left({\frac {\partial A_{3}}{\partial x^{(1)}}}-{\frac {\partial A_{1}}{\partial x^{(3)}}}\right)+N_{3}{\frac {1}{\sqrt {c}}}\left({\frac {\partial A_{1}}{\partial x^{(2)}}}-{\frac {\partial A_{2}}{\partial x^{(1)}}}\right)\right]\end{aligned}}}
Here, the line integral is orbiting around the normal
N
{\displaystyle N}
in the negative sense, thus clock-wise if the coordinate system is a right-system; because by § 1 the directions
d
x
d
s
{\displaystyle {\frac {dx}{ds}}}
,
N
′
{\displaystyle N'}
,
N
{\displaystyle N}
are following each other like the coordinate axes, where
N
′
{\displaystyle N'}
is the normal (which is directed outwards of the area framed on
M
2
{\displaystyle M_{2}}
) of the framing-
M
1
{\displaystyle M_{1}}
. Ordinarily, one prefers a positive sense of circulation and therefore the generalized rotation:
r
o
t
(
1
)
(
A
1
A
2
A
3
)
=
1
c
(
∂
A
3
∂
x
(
2
)
−
∂
A
2
∂
x
(
3
)
)
{\displaystyle \mathrm {rot} ^{(1)}\left(A_{1}A_{2}A_{3}\right)={\frac {1}{\sqrt {c}}}\left({\frac {\partial A_{3}}{\partial x^{(2)}}}-{\frac {\partial A_{2}}{\partial x^{(3)}}}\right)}
etc.
r
o
t
(
1
)
(
A
(
1
)
A
(
2
)
A
(
3
)
)
=
1
c
(
∂
∂
x
(
2
)
(
∑
β
c
3
β
A
(
β
)
)
−
∂
∂
x
(
3
)
(
∑
β
c
2
β
A
(
β
)
)
)
{\displaystyle \mathrm {rot} ^{(1)}\left(A^{(1)}A^{(2)}A^{(3)}\right)={\frac {1}{\sqrt {c}}}\left({\frac {\partial }{\partial x^{(2)}}}\left(\sum _{\beta }c_{3\beta }A^{(\beta )}\right)-{\frac {\partial }{\partial x^{(3)}}}\left(\sum _{\beta }c_{2\beta }A^{(\beta )}\right)\right)}
etc.
n
=
4
{\displaystyle n=4}
,
p
=
3
{\displaystyle p=3}
.[ edit ]
∫
∫
∫
d
u
(
1
)
d
u
(
2
)
d
u
(
3
)
[
∂
(
x
(
2
)
x
(
3
)
x
(
4
)
)
∂
(
u
(
1
)
u
(
2
)
u
(
3
)
)
A
234
+
∂
(
x
(
1
)
x
(
3
)
x
(
4
)
)
∂
(
u
(
1
)
u
(
2
)
u
(
3
)
)
A
134
+
+
∂
(
x
(
1
)
x
(
2
)
x
(
4
)
)
∂
(
u
(
1
)
u
(
2
)
u
(
3
)
)
A
124
+
∂
(
x
(
1
)
x
(
2
)
x
(
3
)
)
∂
(
u
(
1
)
u
(
2
)
u
(
3
)
)
A
123
]
=
=
∫
∫
∫
d
v
(
1
)
d
v
(
2
)
d
v
(
3
)
d
v
(
4
)
∂
(
x
(
1
)
x
(
2
)
x
(
3
)
x
(
4
)
)
∂
(
v
(
1
)
v
(
2
)
v
(
3
)
v
(
4
)
)
A
1234
{\displaystyle {\begin{aligned}&\int \int \int du^{(1)}du^{(2)}du^{(3)}\left[{\frac {\partial \left(x^{(2)}x^{(3)}x^{(4)}\right)}{\partial \left(u^{(1)}u^{(2)}u^{(3)}\right)}}A_{234}+{\frac {\partial \left(x^{(1)}x^{(3)}x^{(4)}\right)}{\partial \left(u^{(1)}u^{(2)}u^{(3)}\right)}}A_{134}+\right.\\&\left.+{\frac {\partial \left(x^{(1)}x^{(2)}x^{(4)}\right)}{\partial \left(u^{(1)}u^{(2)}u^{(3)}\right)}}A_{124}+{\frac {\partial \left(x^{(1)}x^{(2)}x^{(3)}\right)}{\partial \left(u^{(1)}u^{(2)}u^{(3)}\right)}}A_{123}\right]=\\&\quad =\int \int \int dv^{(1)}dv^{(2)}dv^{(3)}dv^{(4)}{\frac {\partial \left(x^{(1)}x^{(2)}x^{(3)}x^{(4)}\right)}{\partial \left(v^{(1)}v^{(2)}v^{(3)}v^{(4)}\right)}}\mathrm {A} _{1234}\end{aligned}}}
A
1234
=
∂
∂
x
(
4
)
A
123
−
∂
∂
x
(
3
)
A
124
+
∂
∂
x
(
2
)
A
134
−
∂
∂
x
(
1
)
A
234
,
{\displaystyle \mathrm {A} _{1234}={\frac {\partial }{\partial x^{(4)}}}A_{123}-{\frac {\partial }{\partial x^{(3)}}}A_{124}+{\frac {\partial }{\partial x^{(2)}}}A_{134}-{\frac {\partial }{\partial x^{(1)}}}A_{234},}
A
123
=
−
A
131
=
−
A
321
=
A
231
=
A
312
=
−
A
213
{\displaystyle A_{123}=-A_{131}=-A_{321}=A_{231}=A_{312}=-A_{213}}
etc.
Introducing the supplements
−
1
c
A
234
=
B
(
1
)
,
1
c
A
134
=
B
(
2
)
,
−
1
c
A
124
=
B
(
3
)
,
1
c
A
123
=
B
(
4
)
,
{\displaystyle -{\sqrt {\frac {1}{c}}}A_{234}=B^{(1)},\quad {\sqrt {\frac {1}{c}}}A_{134}=B^{(2)},\quad -{\sqrt {\frac {1}{c}}}A_{124}=B^{(3)},\quad {\sqrt {\frac {1}{c}}}A_{123}=B^{(4)},}
−
c
b
∂
(
x
(
2
)
x
(
3
)
x
(
4
)
)
∂
(
u
(
1
)
u
(
2
)
u
(
3
)
)
=
N
1
,
c
b
∂
(
x
(
1
)
x
(
3
)
x
(
4
)
)
∂
(
u
(
1
)
u
(
2
)
u
(
3
)
)
=
N
2
,
−
c
b
∂
(
x
(
1
)
x
(
2
)
x
(
4
)
)
∂
(
u
(
1
)
u
(
2
)
u
(
3
)
)
=
N
3
,
c
b
∂
(
x
(
1
)
x
(
2
)
x
(
3
)
)
∂
(
u
(
1
)
u
(
2
)
u
(
3
)
)
=
N
4
,
{\displaystyle {\begin{aligned}-{\sqrt {\frac {c}{b}}}{\frac {\partial \left(x^{(2)}x^{(3)}x^{(4)}\right)}{\partial \left(u^{(1)}u^{(2)}u^{(3)}\right)}}&=N_{1},&{\sqrt {\frac {c}{b}}}{\frac {\partial \left(x^{(1)}x^{(3)}x^{(4)}\right)}{\partial \left(u^{(1)}u^{(2)}u^{(3)}\right)}}&=N_{2},\\-{\sqrt {\frac {c}{b}}}{\frac {\partial \left(x^{(1)}x^{(2)}x^{(4)}\right)}{\partial \left(u^{(1)}u^{(2)}u^{(3)}\right)}}&=N_{3},&{\sqrt {\frac {c}{b}}}{\frac {\partial \left(x^{(1)}x^{(2)}x^{(3)}\right)}{\partial \left(u^{(1)}u^{(2)}u^{(3)}\right)}}&=N_{4},\end{aligned}}}
where
b
{\displaystyle b}
is the discriminant of the arc-element on
M
3
{\displaystyle M_{3}}
, so because of
d
S
3
=
b
d
u
(
1
)
d
u
(
2
)
d
u
(
3
)
,
d
S
4
=
c
d
x
(
1
)
d
x
(
2
)
d
x
(
3
)
d
x
(
4
)
{\displaystyle dS_{3}={\sqrt {b}}du^{(1)}du^{(2)}du^{(3)},\quad dS_{4}={\sqrt {c}}dx^{(1)}dx^{(2)}dx^{(3)}dx^{(4)}}
:
it is given
∫
∫
d
S
3
[
N
1
B
(
1
)
+
N
2
B
(
2
)
+
N
3
B
(
3
)
+
N
4
B
(
4
)
]
=
=
∫
∫
∫
d
S
4
1
c
[
∂
∂
x
(
1
)
(
c
B
(
1
)
)
+
∂
∂
x
(
2
)
(
c
B
(
2
)
)
+
∂
∂
x
(
3
)
(
c
B
(
3
)
)
+
∂
∂
x
(
4
)
(
c
B
(
4
)
)
]
{\displaystyle {\begin{aligned}&\int \int dS_{3}\left[N_{1}B^{(1)}+N_{2}B^{(2)}+N_{3}B^{(3)}+N_{4}B^{(4)}\right]=\\&\quad =\int \int \int dS_{4}{\frac {1}{\sqrt {c}}}\left[{\frac {\partial }{\partial x^{(1)}}}\left({\sqrt {c}}B^{(1)}\right)+{\frac {\partial }{\partial x^{(2)}}}\left({\sqrt {c}}B^{(2)}\right)+{\frac {\partial }{\partial x^{(3)}}}\left({\sqrt {c}}B^{(3)}\right)+{\frac {\partial }{\partial x^{(4)}}}\left({\sqrt {c}}B^{(4)}\right)\right]\end{aligned}}}
where
N
{\displaystyle N}
goes to the exterior. Therefore it is given for the generalized four-dimensional divergence:[ 3]
D
i
v
(
B
(
1
)
B
(
2
)
B
(
3
)
B
(
4
)
)
=
1
c
∑
α
∂
∂
x
(
α
)
(
c
B
(
α
)
)
,
{\displaystyle \mathrm {Div} \left(B^{(1)}B^{(2)}B^{(3)}B^{(4)}\right)={\frac {1}{\sqrt {c}}}\sum _{\alpha }{\frac {\partial }{\partial x^{(\alpha )}}}\left({\sqrt {c}}B^{(\alpha )}\right),}
d
i
v
(
B
1
B
2
B
3
B
4
)
=
1
c
∑
α
∂
∂
x
(
α
)
(
c
∑
β
c
(
α
β
)
B
β
)
,
{\displaystyle \mathrm {div} \left(B_{1}B_{2}B_{3}B_{4}\right)={\frac {1}{\sqrt {c}}}\sum _{\alpha }{\frac {\partial }{\partial x^{(\alpha )}}}\left({\sqrt {c}}\sum _{\beta }c^{(\alpha \beta )}B_{\beta }\right),}
,
d
i
v
(
B
1
B
2
B
3
B
4
)
=
◻
B
=
1
c
∑
α
∂
∂
x
(
α
)
(
c
∑
β
c
(
α
β
)
∂
B
∂
x
(
β
)
)
.
{\displaystyle \mathrm {div} \left(\mathrm {B} _{1}\mathrm {B} _{2}\mathrm {B} _{3}\mathrm {B} _{4}\right)=\square B={\frac {1}{\sqrt {c}}}\sum _{\alpha }{\frac {\partial }{\partial x^{(\alpha )}}}\left({\sqrt {c}}\sum _{\beta }c^{(\alpha \beta )}{\frac {\partial B}{\partial x^{(\beta )}}}\right).}
n
=
4
{\displaystyle n=4}
,
p
=
2
{\displaystyle p=2}
.[ edit ]
∫
∫
d
u
(
1
)
d
u
(
2
)
[
∂
(
x
(
1
)
x
(
2
)
)
∂
(
u
(
1
)
u
(
2
)
)
A
12
+
∂
(
x
(
1
)
x
(
3
)
)
∂
(
u
(
1
)
u
(
2
)
)
A
13
+
∂
(
x
(
1
)
x
(
4
)
)
∂
(
u
(
1
)
u
(
2
)
)
A
14
+
+
∂
(
x
(
2
)
x
(
3
)
)
∂
(
u
(
1
)
u
(
2
)
)
A
23
+
∂
(
x
(
2
)
x
(
4
)
)
∂
(
u
(
1
)
u
(
2
)
)
A
24
+
∂
(
x
(
3
)
x
(
4
)
)
∂
(
u
(
1
)
u
(
2
)
)
A
34
]
=
=
∫
∫
∫
d
v
(
1
)
d
v
(
2
)
d
v
(
3
)
[
∂
(
x
(
2
)
x
(
3
)
x
(
4
)
)
∂
(
v
(
1
)
v
(
2
)
v
(
3
)
)
A
234
+
∂
(
x
(
1
)
x
(
3
)
x
(
4
)
)
∂
(
v
(
1
)
v
(
2
)
v
(
3
)
)
A
134
+
+
∂
(
x
(
1
)
x
(
2
)
x
(
4
)
)
∂
(
v
(
1
)
v
(
2
)
v
(
3
)
)
A
124
+
∂
(
x
(
1
)
x
(
2
)
x
(
3
)
)
∂
(
v
(
1
)
v
(
2
)
v
(
3
)
)
A
123
]
{\displaystyle {\begin{aligned}&\int \int du^{(1)}du^{(2)}\left[{\frac {\partial \left(x^{(1)}x^{(2)}\right)}{\partial \left(u^{(1)}u^{(2)}\right)}}A_{12}+{\frac {\partial \left(x^{(1)}x^{(3)}\right)}{\partial \left(u^{(1)}u^{(2)}\right)}}A_{13}+{\frac {\partial \left(x^{(1)}x^{(4)}\right)}{\partial \left(u^{(1)}u^{(2)}\right)}}A_{14}+\right.\\&\quad \left.+{\frac {\partial \left(x^{(2)}x^{(3)}\right)}{\partial \left(u^{(1)}u^{(2)}\right)}}A_{23}+{\frac {\partial \left(x^{(2)}x^{(4)}\right)}{\partial \left(u^{(1)}u^{(2)}\right)}}A_{24}+{\frac {\partial \left(x^{(3)}x^{(4)}\right)}{\partial \left(u^{(1)}u^{(2)}\right)}}A_{34}\right]=\\&=\int \int \int dv^{(1)}dv^{(2)}dv^{(3)}\left[{\frac {\partial \left(x^{(2)}x^{(3)}x^{(4)}\right)}{\partial \left(v^{(1)}v^{(2)}v^{(3)}\right)}}\mathrm {A} _{234}+{\frac {\partial \left(x^{(1)}x^{(3)}x^{(4)}\right)}{\partial \left(v^{(1)}v^{(2)}v^{(3)}\right)}}\mathrm {A} _{134}+\right.\\&\quad \left.+{\frac {\partial \left(x^{(1)}x^{(2)}x^{(4)}\right)}{\partial \left(v^{(1)}v^{(2)}v^{(3)}\right)}}\mathrm {A} _{124}+{\frac {\partial \left(x^{(1)}x^{(2)}x^{(3)}\right)}{\partial \left(v^{(1)}v^{(2)}v^{(3)}\right)}}\mathrm {A} _{123}\right]\end{aligned}}}
Introducing the supplements
1
c
A
12
=
B
(
34
)
,
1
c
A
13
=
B
(
42
)
,
1
c
A
14
=
B
(
23
)
,
1
c
A
23
=
B
(
14
)
,
1
c
A
24
=
B
(
31
)
,
1
c
A
34
=
B
(
12
)
,
{\displaystyle {\begin{aligned}&{\sqrt {\frac {1}{c}}}A_{12}=B^{(34)},\quad {\sqrt {\frac {1}{c}}}A_{13}=B^{(42)},\quad {\sqrt {\frac {1}{c}}}A_{14}=B^{(23)},\\&{\sqrt {\frac {1}{c}}}A_{23}=B^{(14)},\quad {\sqrt {\frac {1}{c}}}A_{24}=B^{(31)},\quad {\sqrt {\frac {1}{c}}}A_{34}=B^{(12)},\end{aligned}}}
c
a
∂
(
x
(
1
)
x
(
2
)
)
∂
(
u
(
1
)
u
(
2
)
)
=
N
34
,
{\displaystyle {\sqrt {\frac {c}{a}}}{\frac {\partial \left(x^{(1)}x^{(2)}\right)}{\partial \left(u^{(1)}u^{(2)}\right)}}=N_{34},}
etc.
−
c
b
∂
(
x
(
2
)
x
(
3
)
x
(
4
)
)
∂
(
v
(
1
)
v
(
2
)
v
(
3
)
)
=
N
1
,
{\displaystyle -{\sqrt {\frac {c}{b}}}{\frac {\partial \left(x^{(2)}x^{(3)}x^{(4)}\right)}{\partial \left(v^{(1)}v^{(2)}v^{(3)}\right)}}=N_{1},}
etc.
where
a
{\displaystyle a}
or
b
{\displaystyle b}
are the discriminants of the arc-element of
M
2
{\displaystyle M_{2}}
or
M
3
{\displaystyle M_{3}}
:
∫
∫
d
S
2
[
N
12
B
(
11
)
+
N
13
B
(
13
)
+
N
14
B
(
14
)
+
N
23
B
(
23
)
+
N
24
B
(
24
)
+
N
34
B
(
34
)
]
=
=
−
∫
∫
∫
d
S
3
{
N
1
1
c
[
∗
+
∂
∂
x
(
2
)
(
c
B
(
12
)
)
+
∂
∂
x
(
3
)
(
c
B
(
13
)
)
+
∂
∂
x
(
4
)
(
c
B
(
14
)
)
]
N
2
1
c
[
∂
∂
x
(
1
)
(
c
B
(
21
)
)
+
∗
+
∂
∂
x
(
3
)
(
c
B
(
23
)
)
+
∂
∂
x
(
4
)
(
c
B
(
24
)
)
]
N
3
1
c
[
∂
∂
x
(
1
)
(
c
B
(
31
)
)
+
∂
∂
x
(
2
)
(
c
B
(
32
)
)
+
∗
+
∂
∂
x
(
4
)
(
c
B
(
34
)
)
]
N
4
1
c
[
∂
∂
x
(
1
)
(
c
B
(
41
)
)
+
∂
∂
x
(
2
)
(
c
B
(
42
)
)
+
∂
∂
x
(
3
)
(
c
B
(
43
)
)
+
∗
]
}
{\displaystyle {\begin{aligned}\int \int dS_{2}\left[N_{12}B^{(11)}+N_{13}B^{(13)}+N_{14}B^{(14)}+N_{23}B^{(23)}+N_{24}B^{(24)}+N_{34}B^{(34)}\right]=\\=-\int \int \int dS_{3}\left\{N_{1}{\frac {1}{\sqrt {c}}}\left[^{\ast }+{\frac {\partial }{\partial x^{(2)}}}\left({\sqrt {c}}B^{(12)}\right)+{\frac {\partial }{\partial x^{(3)}}}\left({\sqrt {c}}B^{(13)}\right)+{\frac {\partial }{\partial x^{(4)}}}\left({\sqrt {c}}B^{(14)}\right)\right]\right.\\N_{2}{\frac {1}{\sqrt {c}}}\left[{\frac {\partial }{\partial x^{(1)}}}\left({\sqrt {c}}B^{(21)}\right)+{}^{\ast }+{\frac {\partial }{\partial x^{(3)}}}\left({\sqrt {c}}B^{(23)}\right)+{\frac {\partial }{\partial x^{(4)}}}\left({\sqrt {c}}B^{(24)}\right)\right]\\N_{3}{\frac {1}{\sqrt {c}}}\left[{\frac {\partial }{\partial x^{(1)}}}\left({\sqrt {c}}B^{(31)}\right)+{\frac {\partial }{\partial x^{(2)}}}\left({\sqrt {c}}B^{(32)}\right)+{}^{\ast }+{\frac {\partial }{\partial x^{(4)}}}\left({\sqrt {c}}B^{(34)}\right)\right]\\\left.N_{4}{\frac {1}{\sqrt {c}}}\left[{\frac {\partial }{\partial x^{(1)}}}\left({\sqrt {c}}B^{(41)}\right)+{\frac {\partial }{\partial x^{(2)}}}\left({\sqrt {c}}B^{(42)}\right)+{\frac {\partial }{\partial x^{(3)}}}\left({\sqrt {c}}B^{(43)}\right)+{}^{\ast }\right]\right\}\end{aligned}}}
where the normal plane
N
α
β
{\displaystyle N_{\alpha \beta }}
is given by
[
N
′
N
]
{\displaystyle [N'N]}
[ 4] and
N
′
{\displaystyle N'}
, which is the normal of
M
2
{\displaystyle M_{2}}
directed outwards of the area limited on
M
3
{\displaystyle M_{3}}
as mentioned in § 1. By that, the generalized vector divergence becomes:[ 5]
D
i
v
(
1
)
(
B
(
12
)
B
(
13
)
B
(
14
)
B
(
23
)
B
(
24
)
B
(
34
)
)
=
1
c
∑
∂
∂
x
(
α
)
(
c
B
(
1
α
)
)
{\displaystyle {\mathfrak {Div}}^{(1)}\left(B^{(12)}B^{(13)}B^{(14)}B^{(23)}B^{(24)}B^{(34)}\right)={\frac {1}{\sqrt {c}}}\sum {\frac {\partial }{\partial x^{(\alpha )}}}\left({\sqrt {c}}B^{(1\alpha )}\right)}
etc.
D
i
v
(
1
)
(
B
12
B
13
B
14
B
23
B
24
B
34
)
=
1
c
∑
α
∂
∂
x
(
α
)
(
c
∑
β
,
γ
c
(
1
β
)
c
(
α
γ
)
B
β
γ
)
{\displaystyle {\mathfrak {Div}}^{(1)}\left(B_{12}B_{13}B_{14}B_{23}B_{24}B_{34}\right)={\frac {1}{\sqrt {c}}}\sum _{\alpha }{\frac {\partial }{\partial x^{(\alpha )}}}\left({\sqrt {c}}\sum _{\beta ,\gamma }c^{(1\beta )}c^{(\alpha \gamma )}B_{\beta \gamma }\right)}
etc.
The system
B
α
3
α
4
=
∑
β
3
,
β
4
c
α
3
β
3
c
α
4
β
4
B
(
β
3
β
4
)
=
1
c
∑
c
α
3
β
3
c
α
4
β
4
A
β
1
β
2
{\displaystyle B_{\alpha _{3}\alpha _{4}}=\sum _{\beta _{3},\beta _{4}}c_{\alpha _{3}\beta _{3}}c_{\alpha _{4}\beta _{4}}B^{\left(\beta _{3}\beta _{4}\right)}={\frac {1}{\sqrt {c}}}\sum c_{\alpha _{3}\beta _{3}}c_{\alpha _{4}\beta _{4}}A_{\beta _{1}\beta _{2}}}
shall be called the system dual to
A
α
β
{\displaystyle A_{\alpha \beta }}
and be denoted as
A
α
3
α
4
∗
{\displaystyle A_{\alpha _{3}\alpha _{4}}^{\ast }}
. So it follows
D
i
v
(
1
)
(
A
12
∗
A
13
∗
A
14
∗
A
23
∗
A
24
∗
A
34
∗
)
=
1
c
A
234
,
{\displaystyle {\mathfrak {Div}}^{(1)}\left(A_{12}^{\ast }A_{13}^{\ast }A_{14}^{\ast }A_{23}^{\ast }A_{24}^{\ast }A_{34}^{\ast }\right)={\frac {1}{\sqrt {c}}}\mathrm {A} _{234},}
D
i
v
(
2
)
(
A
12
∗
A
13
∗
A
14
∗
A
23
∗
A
24
∗
A
34
∗
)
=
−
1
c
A
134
{\displaystyle {\mathfrak {Div}}^{(2)}\left(A_{12}^{\ast }A_{13}^{\ast }A_{14}^{\ast }A_{23}^{\ast }A_{24}^{\ast }A_{34}^{\ast }\right)=-{\frac {1}{\sqrt {c}}}\mathrm {A} _{134}}
etc.
and
D
i
v
(
1
)
(
A
12
A
13
A
14
A
23
A
24
A
34
)
=
1
c
A
234
∗
,
{\displaystyle {\mathfrak {Div}}^{(1)}\left(A_{12}^{\text{ }}A_{13}A_{14}A_{23}A_{24}A_{34}\right)={\frac {1}{\sqrt {c}}}\mathrm {A} _{234}^{\ast },}
D
i
v
(
2
)
(
A
12
A
13
A
14
A
23
A
24
A
34
)
=
−
1
c
A
134
∗
{\displaystyle {\mathfrak {Div}}^{(2)}\left(A_{12}^{\text{ }}A_{13}A_{14}A_{23}A_{24}A_{34}\right)=-{\frac {1}{\sqrt {c}}}\mathrm {A} _{134}^{\ast }}
etc.
where
A
234
∗
{\displaystyle \mathrm {A} _{234}^{\ast }}
etc. are to be formed from
A
α
β
∗
{\displaystyle A_{\alpha \beta }^{\ast }}
, as
A
234
{\displaystyle \mathrm {A} _{234}}
etc. from
A
α
β
{\displaystyle A_{\alpha \beta }}
.
In the case
D
i
v
(
A
α
β
∗
)
≡
0
{\displaystyle \mathrm {Div} \left(A_{\alpha \beta }^{\ast }\right)\equiv 0}
it therefore follows
A
α
β
=
A
α
β
=
∂
A
α
∂
x
(
β
)
−
∂
A
β
∂
x
(
α
)
{\displaystyle A_{\alpha \beta }=\mathrm {A} _{\alpha \beta }={\frac {\partial A_{\alpha }}{\partial x^{(\beta )}}}-{\frac {\partial A_{\beta }}{\partial x^{(\alpha )}}}}
.
n
=
4
{\displaystyle n=4}
,
p
=
1
{\displaystyle p=1}
.[ edit ]
∫
d
u
(
∂
x
(
1
)
∂
u
A
1
+
∂
x
(
2
)
∂
u
A
2
+
∂
x
(
3
)
∂
u
A
3
+
∂
x
(
4
)
∂
u
A
4
)
=
=
∫
∫
d
v
(
1
)
d
v
(
2
)
(
∂
(
x
(
1
)
x
(
2
)
)
∂
(
v
(
1
)
v
(
2
)
)
A
12
+
∂
(
x
(
1
)
x
(
3
)
)
∂
(
v
(
1
)
v
(
2
)
)
A
13
+
∂
(
x
(
1
)
x
(
4
)
)
∂
(
v
(
1
)
v
(
2
)
)
A
14
+
∂
(
x
(
2
)
x
(
3
)
)
∂
(
v
(
1
)
v
(
2
)
)
A
23
+
∂
(
x
(
2
)
x
(
4
)
)
∂
(
v
(
1
)
v
(
2
)
)
A
24
+
∂
(
x
(
3
)
x
(
4
)
)
∂
(
v
(
1
)
v
(
2
)
)
A
34
)
{\displaystyle {\begin{aligned}&\int du\left({\frac {\partial x^{(1)}}{\partial u}}A_{1}+{\frac {\partial x^{(2)}}{\partial u}}A_{2}+{\frac {\partial x^{(3)}}{\partial u}}A_{3}+{\frac {\partial x^{(4)}}{\partial u}}A_{4}\right)=\\&\quad =\int \int dv^{(1)}dv^{(2)}\left({\frac {\partial \left(x^{(1)}x^{(2)}\right)}{\partial \left(v^{(1)}v^{(2)}\right)}}A_{12}+{\frac {\partial \left(x^{(1)}x^{(3)}\right)}{\partial \left(v^{(1)}v^{(2)}\right)}}A_{13}+\right.\\&\qquad \left.{\frac {\partial \left(x^{(1)}x^{(4)}\right)}{\partial \left(v^{(1)}v^{(2)}\right)}}A_{14}+{\frac {\partial \left(x^{(2)}x^{(3)}\right)}{\partial \left(v^{(1)}v^{(2)}\right)}}A_{23}+{\frac {\partial \left(x^{(2)}x^{(4)}\right)}{\partial \left(v^{(1)}v^{(2)}\right)}}A_{24}+{\frac {\partial \left(x^{(3)}x^{(4)}\right)}{\partial \left(v^{(1)}v^{(2)}\right)}}A_{34}\right)\end{aligned}}}
or
∫
d
s
∑
d
x
(
α
)
d
s
A
α
=
∫
∫
d
S
2
∑
N
α
3
α
4
1
c
A
α
1
α
2
{\displaystyle \int ds\sum {\frac {dx^{(\alpha )}}{ds}}A_{\alpha }=\int \int dS_{2}\sum N_{\alpha _{3}\alpha _{4}}{\frac {1}{\sqrt {c}}}A_{\alpha _{1}\alpha _{2}}}
with the corresponding orientation (by § 1). Therefore, it is given for the generalized rotation[ 6] with the common signs:
R
o
t
(
12
)
(
A
1
A
2
A
3
A
4
)
=
1
c
(
∂
A
4
∂
x
(
3
)
−
∂
A
3
∂
x
(
4
)
)
{\displaystyle \mathrm {Rot} ^{(12)}\left(A_{1}A_{2}A_{3}A_{4}\right)={\frac {1}{\sqrt {c}}}\left({\frac {\partial A_{4}}{\partial x^{(3)}}}-{\frac {\partial A_{3}}{\partial x^{(4)}}}\right)}
etc.
R
o
t
12
(
A
1
A
2
A
3
A
4
)
=
∂
A
2
∂
x
(
1
)
−
∂
A
1
∂
x
(
2
)
{\displaystyle \mathrm {Rot} _{12}\left(A_{1}A_{2}A_{3}A_{4}\right)={\frac {\partial A_{2}}{\partial x^{(1)}}}-{\frac {\partial A_{1}}{\partial x^{(2)}}}}
etc.
The integral forms
n
=
4
{\displaystyle n=4}
in the notation of the absolute differential calculus.[ edit ]
As appendix, the methods of the already mentioned absolute differential calculus shall be demonstrated, because it will be applied later; while it is less suited for the transformation of the actual integral form, it can hardly be avoided in connection with other vectorial formations which are more combined. In the mentioned work,[ 7] Christoffel shows, based on the differential equations for the second derivative
∂
2
x
(
α
)
∂
x
¯
(
λ
)
∂
x
¯
(
μ
)
{\displaystyle {\frac {\partial ^{2}x^{(\alpha )}}{\partial {\bar {x}}^{(\lambda )}\partial {\bar {x}}^{(\mu )}}}}
, that from a covariant system
A
α
1
α
2
…
α
p
{\displaystyle A_{\alpha _{1}\alpha _{2}\dots \alpha _{p}}}
of
p
{\displaystyle p}
-the order , a system of
p
+
1
{\displaystyle p+1}
-th order emerges as follows:
A
α
1
α
2
…
α
p
/
α
p
+
1
=
∂
∂
x
(
α
p
+
1
)
A
α
1
α
2
…
α
p
−
∑
h
=
1
p
∑
β
=
1
n
{
α
p
+
1
α
h
β
}
A
α
1
…
α
h
−
1
β
α
h
+
1
…
α
p
,
{\displaystyle A_{\alpha _{1}\alpha _{2}\dots \alpha _{p}/\alpha _{p+1}}={\frac {\partial }{\partial x^{\left(\alpha _{p+1}\right)}}}A_{\alpha _{1}\alpha _{2}\dots \alpha _{p}}-\sum _{h=1}^{p}\sum _{\beta =1}^{n}\left\{{\begin{matrix}\alpha _{p+1}\alpha _{h}\\\beta \end{matrix}}\right\}A_{\alpha _{1}\dots \alpha _{h-1}\beta \alpha _{h+1}\dots \alpha _{p}},}
where the Christoffel symbols of second order with triple-indices arise, which are defined as follows:
{
α
β
γ
}
=
∑
δ
=
1
n
c
(
γ
δ
)
[
α
β
δ
]
=
∑
δ
=
1
n
c
(
γ
δ
)
⋅
1
2
(
∂
c
α
δ
∂
x
(
β
)
+
∂
c
β
δ
∂
x
(
α
)
−
∂
c
α
β
∂
x
(
δ
)
)
.
{\displaystyle \left\{{\begin{matrix}\alpha \beta \\\gamma \end{matrix}}\right\}=\sum _{\delta =1}^{n}c^{(\gamma \delta )}{\begin{bmatrix}\alpha \beta \\\delta \end{bmatrix}}=\sum _{\delta =1}^{n}c^{(\gamma \delta )}\cdot {\frac {1}{2}}\left({\frac {\partial c_{\alpha \delta }}{\partial x^{(\beta )}}}+{\frac {\partial c_{\beta \delta }}{\partial x^{(\alpha )}}}-{\frac {\partial c_{\alpha \beta }}{\partial x^{(\delta )}}}\right).}
Ricci and Levi-Cività denote this as the covariant differential quotient of
A
α
1
α
2
…
α
p
{\displaystyle A_{\alpha _{1}\alpha _{2}\dots \alpha _{p}}}
with respect to
x
(
α
p
+
1
)
{\displaystyle x^{\left(\alpha _{p+1}\right)}}
. The prime separates the indices added by differentiation from the others. For the contravariant differential quotient it is given:
A
(
α
1
α
2
…
α
p
/
α
p
+
1
)
=
∑
β
=
1
n
c
(
α
p
+
1
β
)
[
∂
∂
x
(
β
)
A
(
α
1
α
2
…
α
p
)
+
∑
h
=
1
p
∑
γ
=
1
n
{
β
γ
α
h
}
A
(
α
1
α
2
…
α
h
−
1
γ
α
h
+
1
…
α
p
)
]
.
{\displaystyle A^{\left(\alpha _{1}\alpha _{2}\dots \alpha _{p}/\alpha _{p+1}\right)}=\sum _{\beta =1}^{n}c^{\left(\alpha _{p+1}\beta \right)}\left[{\frac {\partial }{\partial x^{(\beta )}}}A^{\left(\alpha _{1}\alpha _{2}\dots \alpha _{p}\right)}+\sum _{h=1}^{p}\sum _{\gamma =1}^{n}\left\{{\begin{matrix}\beta \gamma \\\alpha _{h}\end{matrix}}\right\}A^{\left(\alpha _{1}\alpha _{2}\dots \alpha _{h-1}\gamma \alpha _{h+1}\dots \alpha _{p}\right)}\right].}
Then we have, as it can be easily shown:
A
1234
=
A
123
/
4
−
A
124
/
3
+
A
134
/
2
−
A
234
/
1
=
c
D
i
v
(
B
(
1
)
B
(
2
)
B
(
3
)
B
(
4
)
)
=
c
∑
α
,
β
c
α
β
B
(
α
/
β
)
=
∑
α
∂
∂
x
(
α
)
(
c
B
(
α
)
)
,
=
d
i
v
(
B
1
B
2
B
3
B
4
)
=
c
∑
α
,
β
c
(
α
β
)
B
α
/
β
=
∑
α
∂
∂
x
(
α
)
(
c
∑
β
c
(
α
β
)
B
β
)
{\displaystyle {\begin{aligned}\mathrm {A} _{1234}&=A_{123/4}-A_{124/3}+A_{134/2}-A_{234/1}\\&={\sqrt {c}}\mathrm {Div} \left(B^{(1)}B^{(2)}B^{(3)}B^{(4)}\right)={\sqrt {c}}{\underset {\alpha ,\beta }{\sum }}c_{\alpha \beta }B^{(\alpha /\beta )}={\underset {\alpha }{\sum }}{\frac {\partial }{\partial x^{(\alpha )}}}\left({\sqrt {c}}B^{(\alpha )}\right),\\&=\mathrm {div} \left(B_{1}B_{2}B_{3}B_{4}\right)={\sqrt {c}}{\underset {\alpha ,\beta }{\sum }}c^{(\alpha \beta )}B_{\alpha /\beta }={\underset {\alpha }{\sum }}{\frac {\partial }{\partial x^{(\alpha )}}}\left({\sqrt {c}}{\underset {\beta }{\sum }}c^{(\alpha \beta )}B_{\beta }\right)\end{aligned}}}
with the connection
−
1
c
A
234
=
B
(
1
)
=
∑
β
c
(
1
β
)
B
β
{\displaystyle -{\frac {1}{\sqrt {c}}}A_{234}=B^{(1)}={\underset {\beta }{\sum }}c^{(1\beta )}B_{\beta }}
,
1
c
A
134
=
B
(
2
)
=
∑
β
c
(
2
β
)
B
β
{\displaystyle {\frac {1}{\sqrt {c}}}A_{134}=B^{(2)}={\underset {\beta }{\sum }}c^{(2\beta )}B_{\beta }}
etc.
A
234
=
A
23
/
4
−
A
24
/
3
+
A
34
/
2
=
c
D
i
v
(
1
)
(
B
(
12
)
B
(
13
)
B
(
14
)
B
(
23
)
B
(
24
)
B
(
34
)
)
=
c
∑
α
,
β
c
α
β
B
(
1
α
/
β
)
=
∑
α
∂
∂
x
(
α
)
(
c
B
(
1
α
)
)
,
=
c
D
i
v
(
1
)
(
B
12
B
13
B
14
B
23
B
24
B
34
)
=
c
∑
α
,
β
,
γ
c
(
1
α
)
c
(
β
γ
)
B
α
β
/
γ
=
∑
α
∂
∂
x
(
α
)
(
c
∑
β
,
γ
c
(
1
β
)
B
β
γ
)
{\displaystyle {\begin{aligned}\mathrm {A} _{234}&=A_{23/4}-A_{24/3}+A_{34/2}\\&={\sqrt {c}}{\mathfrak {Div}}^{(1)}\left(B^{(12)}B^{(13)}B^{(14)}B^{(23)}B^{(24)}B^{(34)}\right)={\sqrt {c}}{\underset {\alpha ,\beta }{\sum }}c_{\alpha \beta }B^{(1\alpha /\beta )}={\underset {\alpha }{\sum }}{\frac {\partial }{\partial x^{(\alpha )}}}\left({\sqrt {c}}B^{(1\alpha )}\right),\\&={\sqrt {c}}{\mathfrak {Div}}^{(1)}\left(B_{12}B_{13}B_{14}B_{23}B_{24}B_{34}\right)={\sqrt {c}}{\underset {\alpha ,\beta ,\gamma }{\sum }}c^{(1\alpha )}c^{(\beta \gamma )}B_{\alpha \beta /\gamma }={\underset {\alpha }{\sum }}{\frac {\partial }{\partial x^{(\alpha )}}}\left({\sqrt {c}}{\underset {\beta ,\gamma }{\sum }}c^{(1\beta )}B_{\beta \gamma }\right)\end{aligned}}}
with the connection
1
c
A
12
=
B
(
34
)
=
∑
α
,
β
c
(
3
α
)
c
(
4
β
)
B
α
β
{\displaystyle {\frac {1}{\sqrt {c}}}A_{12}=B^{(34)}={\underset {\alpha ,\beta }{\sum }}c^{(3\alpha )}c^{(4\beta )}B_{\alpha \beta }}
etc.,
where we could write, following the things stated above, also
A
α
β
∗
{\displaystyle A_{\alpha \beta }^{\ast }}
instead of
B
α
β
{\displaystyle B_{\alpha \beta }}
.
Thus
1
c
A
234
=
D
i
v
(
1
)
(
A
α
β
∗
)
=
∑
α
c
(
1
α
)
∑
β
,
γ
c
(
β
γ
)
A
α
β
/
γ
∗
{\displaystyle {\frac {1}{\sqrt {c}}}\mathrm {A} _{234}={\mathfrak {Div}}^{(1)}\left(A_{\alpha \beta }^{\ast }\right)={\underset {\alpha }{\sum }}c^{(1\alpha )}{\underset {\beta ,\gamma }{\sum }}c^{(\beta \gamma )}A_{\alpha \beta /\gamma }^{\ast }}
etc.
1
c
A
234
∗
=
D
i
v
(
1
)
(
A
α
β
)
=
∑
α
c
(
1
α
)
∑
β
,
γ
c
(
β
γ
)
A
α
β
/
γ
{\displaystyle {\frac {1}{\sqrt {c}}}\mathrm {A} _{234}^{\ast }={\mathfrak {Div}}^{(1)}\left(A_{\alpha \beta }\right)={\underset {\alpha }{\sum }}c^{(1\alpha )}{\underset {\beta ,\gamma }{\sum }}c^{(\beta \gamma )}A_{\alpha \beta /\gamma }}
For
D
i
v
(
A
α
β
∗
)
≡
0
{\displaystyle {\mathfrak {Div}}\left(A_{\alpha \beta }^{\ast }\right)\equiv 0}
we have, as already mentioned above,
A
α
β
=
A
α
β
=
A
α
/
β
−
A
β
/
α
{\displaystyle A_{\alpha \beta }=\mathrm {A} _{\alpha \beta }=A_{\alpha /\beta }-A_{\beta /\alpha }}
and
D
i
v
(
A
α
β
)
=
∑
α
c
(
1
α
)
∑
β
,
γ
c
(
β
γ
)
A
α
/
β
γ
−
∑
α
c
(
1
α
)
∑
β
,
γ
c
(
β
γ
)
A
β
/
α
γ
=
=
∑
α
c
(
1
α
)
[
∑
β
,
γ
c
(
β
γ
)
A
α
/
β
γ
−
∂
∂
x
(
α
)
(
∑
β
,
γ
c
(
β
γ
)
A
β
/
γ
)
]
.
{\displaystyle {\begin{aligned}{\mathfrak {Div}}\left(A_{\alpha \beta }\right)&={\underset {\alpha }{\sum }}c^{(1\alpha )}{\underset {\beta ,\gamma }{\sum }}c^{(\beta \gamma )}A_{\alpha /\beta \gamma }-{\underset {\alpha }{\sum }}c^{(1\alpha )}{\underset {\beta ,\gamma }{\sum }}c^{(\beta \gamma )}A_{\beta /\alpha \gamma }=\\&={\underset {\alpha }{\sum }}c^{(1\alpha )}\left[{\underset {\beta ,\gamma }{\sum }}c^{(\beta \gamma )}A_{\alpha /\beta \gamma }-{\frac {\partial }{\partial x^{(\alpha )}}}\left({\underset {\beta ,\gamma }{\sum }}c^{(\beta \gamma )}A_{\beta /\gamma }\right)\right].\end{aligned}}}
since in Euclidean space (vanishing of the Riemann symbols) the permutation of the differentiation order
A
β
/
α
γ
=
A
β
/
γ
α
{\displaystyle A_{\beta /\alpha \gamma }=A_{\beta /\gamma \alpha }}
is allowed, and
∑
c
(
1
α
)
∑
c
(
β
γ
)
A
β
/
γ
α
{\displaystyle \sum c^{(1\alpha )}\sum c^{(\beta \gamma )}A_{\beta /\gamma \alpha }}
represents the contravariant differential quotient of
∑
β
,
γ
c
(
β
γ
)
A
β
/
γ
{\displaystyle {\underset {\beta ,\gamma }{\sum }}c^{(\beta \gamma )}A_{\beta /\gamma }}
with respect to
x
(
1
)
{\displaystyle x^{(1)}}
.[ 8]
↑ The factor
1
b
{\displaystyle {\sqrt {\frac {1}{b}}}}
makes
N
{\displaystyle N}
invariant in
u
{\displaystyle u}
.
↑ Herein one recognizes Beltrami's second differential operator. Wright , l.c., p. 56, for
n
=
2
{\displaystyle n=2}
.
↑ Sommerfeld , Ann. d. Physik, 33, p. 650 (1910) .
↑ Vector product of two four-vectors
[
A
B
]
(
α
β
)
=
A
(
α
)
B
(
β
)
−
A
(
β
)
B
(
α
)
{\displaystyle [AB]^{(\alpha \beta )}=A^{(\alpha )}B^{(\beta )}-A^{(\beta )}B^{(\alpha )}}
. Sommerfeld , Ann. d. Phys., 32, p. 765 (1910) .
↑ Sommerfeld , l. c., 33, p. 651; as with Stokes ' theorem, the minus sign is not included in the definition.
↑ Sommerfeld , l. c., 33, p. 653 und 654.
↑ See Wright , l. c., p. 13 und 22.
↑ Wright , l. c., p. 23.