10.4. SPHERICAL COORDINATES IN p DIMENSIONS 251

where Φ is a continuous function of φ⃗ and θ .1 Then if f is nonnegative and Lebesguemeasurable,∫

Rpf (y)dmp =

∫hp(A)

f (y)dmp =∫

Af(

hp

(⃗φ ,θ ,ρ

))ρ

p−1Φ

(⃗φ ,θ

)dmp (10.9)

Furthermore whenever f is Borel measurable and nonnegative, one can apply Fubini’stheorem and write∫

Rpf (y)dy =

∫∞

p−1∫

Af(

h(⃗

φ ,θ ,ρ))

Φ

(⃗φ ,θ

)dφ⃗dθdρ (10.10)

where here dφ⃗dθ denotes dmp−1 on A. The same formulas hold if f ∈ L1 (Rp) .

Proof: Formula 10.8 is obvious from the definition of the spherical coordinates becausein the matrix of the derivative, there will be a ρ in p− 1 columns. The first claim is alsoclear from the definition and math induction or from the geometry of the above description.It remains to verify 10.9 and 10.10. It is clear hp maps Ā× [0,∞) onto Rp. Since hp isdifferentiable, it maps sets of measure zero to sets of measure zero. Then

Rp = hp (N∪A× (0,∞)) = hp (N)∪hp (A× (0,∞)) ,

the union of a set of measure zero with hp (A× (0,∞)) . Therefore, from the change ofvariables formula,∫

Rpf (y)dmp =

∫hp(A×(0,∞))

f (y)dmp

=∫

A×(0,∞)f(

hp

(⃗φ ,θ ,ρ

))ρ

p−1Φ

(⃗φ ,θ

)dmp

which proves 10.9. This formula continues to hold if f is in L1 (Rp) by consideration ofpositive and negative parts of real and imaginary parts.

Finally, if f ≥ 0 or in L1 (Rn) and is Borel measurable, the Borel sets denoted as B (Rp)then one can write the following. From the definition of mp∫

A×(0,∞)f(

hp

(⃗φ ,θ ,ρ

))ρ

p−1Φ

(⃗φ ,θ

)dmp

=∫(0,∞)

∫A

f(

hp

(⃗φ ,θ ,ρ

))ρ

p−1Φ

(⃗φ ,θ

)dmp−1dm

=∫(0,∞)

ρp−1

∫A

f(

hp

(⃗φ ,θ ,ρ

))Φ

(⃗φ ,θ

)dmp−1dm

Now the claim about f ∈ L1 follows routinely from considering the positive and negativeparts of the real and imaginary parts of f in the usual way. ■

Note that the above equals∫

Ā×[0,∞) f(

hp

(⃗φ ,θ ,ρ

))ρ p−1Φ

(⃗φ ,θ

)dmp and the iter-

ated integral is also equal to∫[0,∞)

ρp−1

∫Ā

f(

hp

(⃗φ ,θ ,ρ

))Φ

(⃗φ ,θ

)dmp−1dm

because the difference is just a set of measure zero.1Actually it is only a function of the first but this is not important in what follows.

10.4. SPHERICAL COORDINATES IN p DIMENSIONS 251where ® is a continuous function of d and @.' Then if f is nonnegative and Lebesguemeasurable,[fe Jdmp= ff tly )dimp = | £ (hp (9.0,p))p” '& (4,0) dmp (10.9)Furthermore whenever f is Borel measurable and nonnegative, one can apply Fubini’stheorem and writept W)dy= [ pr '/ r(n( 6,0, )) ® (4,0) dbdedp (10.10)where here dodo denotes dmp_\ on A. The same formulas hold if f € L!(R?).Proof: Formula 10.8 is obvious from the definition of the spherical coordinates becausein the matrix of the derivative, there will be a p in p—1 columns. The first claim is alsoclear from the definition and math induction or from the geometry of the above description.It remains to verify 10.9 and 10.10. It is clear h, maps A x [0,ce) onto R?. Since h, isdifferentiable, it maps sets of measure zero to sets of measure zero. ThenRP = hy (NUA x (0,09) = hy (N) Uhy (A x (0,0°)),the union of a set of measure zero with h, (A x (0,°°)). Therefore, from the change ofvariables formula,rept (y)dmp, I wax om? (y)dmpDoo! (h, ($.8.p)) pr '® (3.0) dmpwhich proves 10.9. This formula continues to hold if f is in L' (R?) by consideration ofpositive and negative parts of real and imaginary parts.Finally, if f > 0 or in L! (IR”) and is Borel measurable, the Borel sets denoted as Z (R”)then one can write the following. From the definition of mpDow! (h, (9.6.0) p?'® (5.0) dmpI... Lt (ee (3.6 eae dmp_\dm= I pe Lr (mp (6.6.p)) (4,6) dip idmNow the claim about f € L! follows routinely from considering the positive and negativeparts of the real and imaginary parts of f in the usual way.Note that the above equals JAx{0,e0) f (hy (6. 6.p)) p?'® (6, 6) dmpy and the iter-ated integral is also equal toi p?- LF (hp (6, 6 p)) (4,8) dm,_\dmbecause the difference is just a set of measure zero.' Actually it is only a function of the first but this is not important in what follows.