A Generalization of The Spherical Inversion

See discussions, stats, and author profiles for this publication at: https://www.researchgate.
net/publication/305313075
A generalization of the spherical inversion
Article in International Journal of Mathematical Education In Science & Technology · July 2016
DOI: 10.1080/0020739X.2016.1201598
CITATIONS READS
6 874
2 authors:
José Luis Ramírez Gustavo Rubiano

National University of Colombia National University of Colombia
109 PUBLICATIONS 700 CITATIONS 50 PUBLICATIONS 103 CITATIONS
SEE PROFILE SEE PROFILE
All content following this page was uploaded by Gustavo Rubiano on 10 September 2021.
The user has requested enhancement of the downloaded file.

International Journal of Mathematical Education in
Science and Technology
ISSN: 0020-739X (Print) 1464-5211 (Online) Journal homepage: http://www.tandfonline.com/loi/tmes20
José L. Ramírez & Gustavo N. Rubiano
To cite this article: José L. Ramírez & Gustavo N. Rubiano (2016): A generalization of the
spherical inversion, International Journal of Mathematical Education in Science and Technology
To link to this article: http://dx.doi.org/10.1080/0020739X.2016.1201598
Published online: 13 Jul 2016.
Submit your article to this journal
View related articles
View Crossmark data
Full Terms & Conditions of access and use can be found at

http://www.tandfonline.com/action/journalInformation?journalCode=tmes20
Download by: [190.85.246.2] Date: 13 July 2016, At: 12:23

INTERNATIONAL JOURNAL OF MATHEMATICAL EDUCATION IN SCIENCE AND TECHNOLOGY, 
http://dx.doi.org/./X..

1
José L. Ramírez and Gustavo N. Rubiano2
a
Departamento de Matemáticas, Universidad Sergio Arboleda, Bogotá, Colombia; b Departamento de
Matemáticas, Universidad Nacional de Colombia, Bogotá, Colombia
ABSTRACT ARTICLE HISTORY

In the present article, we introduce a generalization of the spherical Received  April 
inversion. In particular, we define an inversion with respect to an ellip-
KEYWORDS
soid, and prove several properties of this new transformation. The
Geometry transformations;
inversion in an ellipsoid is the generalization of the elliptic inversion spherical inversion; inversion
Downloaded by [190.85.246.2] at 12:23 13 July 2016
to the three-dimensional space. We also study the inverse images in an ellipsoid; Pappus chain
of planes, spheres, ellipsoids and other curves. We use the software
Mathematica to generate these graphics. Finally, we generalize the 2000 MATHEMATICS
Pappus chain theorem to ellipsoids. SUBJECT CLASSIFICATION
Primary G; Secondary
N; C
1. Introduction
The circle inversion is one of the most important and interesting transformations in the
geometry. This transformation can be used to study several well-known problems and the-
orems in geometry such as Feuerbach’s theorem, Steiner porism, the problem of Apollo-
nius, the Pappus chain theorem, among others. The interested reader is referred to [1–6]
for further information on inversive geometry. The inversion was introduced by Apollonius
of Perga in his last book Plane Loci,[7] and systematically studied and applied by Steiner
about 1830.[8]
The inversion of a point P in a given circle C is a point P taken from the ray OP such
that OP · OP = r2 , where r is the radius of C. The circle inversion preserves angles and
maps generalized circles into generalized circles, where a generalized circle means either
a circle or a line (a circle with infinite radius). One of the main properties of this method
is the transformation of a straight line to a circle. Many challenging problems in geometry
become much more tractable when an inversion is applied.
Many kinds of generalizations of inversion transform have been presented in the litera-
ture. In particular, we can change the circle of inversion C by other objects, such as parallel
lines,[9] central conics [10–13] or star-shape sets.[14] We can also use a different distance,
for example, the taxicab distance [15,16] or in general a p-distance.[17]
The circle inversion can also be generalized in the three-dimensional space by using a
sphere as the circle of inversion.[2] In [18], the author used this transformation to gen-
erate three-dimensional fractals. Inspired by these works, in this article, we introduce the
inversion in an ellipsoid. Then we extend several well-known properties. In particular, we
CONTACT José L. Ramírez jolura@gmail.com; josel.ramirez@ima.usergioarboleda.edu.co

©  Informa UK Limited, trading as Taylor & Francis Group
2 J. L. RAMÍREZ AND G. N. RUBIANO
Figure . Inversion of a point in a sphere.

prove a Pappus chain theorem in ellipsoids. We also apply this new transformation to sev-
eral geometry objects such as planes, ellipsoids, three-dimensional fractals, etc. We use the
software Mathematica for computing and displaying the images of the three-dimensional
objects studied in this paper.
2. Inversion in a sphere
In this section, we introduce the inversion in a sphere and we show several known
properties.
Definition 2.1: Let C be a sphere centred at a point O in R3 , the inversion in C is the map-
−→
ping μ : R3 \ {O} −→ R3 \ {O} defined by μ(P) = P , where P lies on the ray OP and
OP · OP = r2 , where r is the radius of C.
The point P is said to be the inverse of P in the sphere C. We call C the sphere of inversion,
O the centre of inversion and r the radius of inversion (see Figure 1). It is clear that μ is an
involution, i.e. μ(μ(P)) = P for every point P ࣔ O. The fixed points are the points on the
sphere C. Moreover, P is in the exterior of C if and only if P is in the interior of C. After
inversion, the part of the space inside the sphere is sent off the sphere and vice versa. Any
object that was near the point of inversion is sent far away from the sphere. Any surface
which is invariant under inversion is called anallagmatic.
Some basic properties about spherical inversion are summarized in the following items;
see, for example, Chapter 5 of [2] for more details. Note that it is possible to extend every
property of the circle inversion to the spherical inversion.
r The angle at which two surfaces intersect is equal to that made by their inverses.
r A plane will always invert to either another plane or a sphere. A plane which pass
through the centre of inversion O is anallagmatic. The inversion of a plane that does
not contain the centre of inversion O is a sphere that contains O and whose tangent
plane at O is , and vice versa (see Figure 2).
INTERNATIONAL JOURNAL OF MATHEMATICAL EDUCATION IN SCIENCE AND TECHNOLOGY 3
Figure . A plane is inverted in a sphere inside the inversion sphere, and a sphere touching the centre of
inversion O is inverted to a tangent plane.
Figure . Sphere inverted to sphere and an orthogonal sphere inverted to itself.
r The classical stereographic projection is just a special case of spherical inversion. Con-
sider a sphere C of radius 1 and a plane touching C at the south pole S of C. Then
is the stereographic projection of C with respect to the north pole N of C (cf. [19]).
In particular, if we√take the sphere S 2 of radius 1 centred at (0, 0, 0) and consider a
sphere C of radius 2 centred at the north pole N = (0, 0, 1), then the inversion of S 2
with respect to C transforms S 2 in its plane stereographic projection (see Figure 2
(right)).
r The spherical inversion of a sphere not passing through the centre of inversion is
another sphere that does not contain the centre of inversion. In particular, every sphere
orthogonal to C is anallagmatic (see Figure 3).
r Let A, B, C and D be four distinct points on a line . We define the cross ratio as
AC · BD
(AB, CD) := ,
AD · BC
where AB denotes the signed distance from A to B. The cross ratio is invariant under
inversion in a sphere whose centre is not any of the four points A, B, C or D, i.e. (AB,
CD) = (A B , C D ).
r A sphere through any pair of inverse points is anallagmatic. In particular, a sphere that
is orthogonal to the sphere of inversion is the sphere itself (see Figure 3 (right)).
Figure . Cyclides obtained from cylinders inverted when the centre of inversion is inside of the cylinder
(cuspidal cyclide) and when it lies outside the cylinder (needle cyclide), respectively.
Figure . Cyclides obtained (horn cyclide and ring cyclide) from a double cone and a torus.
r Circular cylinders, double cones or tori invert to a family of curves called Dupin
cyclides (cf. [20,21]), which were originally introduced by the French geometer
Charles Dupin in 1822. In Figures 4 and 5, we show some examples of Dupin cyclides.
r As the centre of the inverted sphere is not, in general, the same as the inversion of
the centre of the initial sphere, the following fact indicate how to find this centre. The
inverse of the centre of a sphere not through the centre of inversion is the inverse of the
centre of inversion with regard to the inverse of the given sphere (see Figure 6). Thus,
if the spheres C and C with centres OC and OC , respectively, are mutually inverse with
respect to the sphere S with centre O, then
OC = μS (μS (OC )) = μS (μC ((O))).

Figure . Inversion of the centre of inversion.

r The radius for the inverted sphere. Let C be a sphere centred at the point O and radius
r, and T be another sphere centred at the point O1 and radius R. If d = OO1 is the
distance between the centres, the inversion of sphere T is another sphere T with ratio
R given by
Rr2
R = .
d2 − R2
In the following theorem, we show the Cartesian coordinates of a point inverted by a

sphere. Afterwards, we show some applications of this theorem.
Theorem 2.1: Let C be a sphere centred at the point O = (a, b, c) and radius r. Then the
inverse of the point P = (x, y, z) in C is the point

r2 (x − a) r2 (y − b)
P = + a,
(x − a)2 + (y − b)2 + (z − c)2 (x − a)2 + (y − b)2 + (z − c)2

r2 (z − c)
+ b, +c . (1)
(x − a)2 + (y − b)2 + (z − c)2
In particular, if C is centred at the origin, then the spheric inversion is the point

r2 x r2 y r2 z
, , . (2)
x 2 + y2 + z 2 x 2 + y2 + z 2 x 2 + y2 + z 2
Proof: Suppose that P = (x, y, z) and P = (x , y , z ) are inverse points in the sphere C
centred at the point O = (a, b, c) and radius r. Since P, P and O are collinear and the rays
−→ −→
OP and OP are the same direction, then
(x , y , z ) = k(x, y, z), k ∈ R+ .

Figure . Circle and spherical inversion of a square and a sphere.
Moreover, from definition of inversion, we have
OP · OP = r2
k · OP · OP = r2 .
Then
r2 r2
k= = .
OP2 (x − a)2 + (y − b)2 + (z − c)2
Therefore,
r2
P = O + (P − O).
OP2
Comparing each coordinate, we obtain Equation (1).
Using the above theorem it is not difficult to find the spherical inversion of three-
dimensional objects.
Example 2.1: In Figure 7, we show the inversion of a square with respect to a circle and the
inversion of a cube with respect to a sphere.
Example 2.2: In Figure 8, we show the spherical inversion of a given ellipsoid.
Example 2.3: In Figure 9, we show the spherical inversion of the three-dimensional Hilbert
curve.
Figure . Spherical inversion of an ellipsoid.

Figure . Three-dimensional Hilbert curve under spherical inversion.
3. Inversion in an ellipsoid
In this section, we introduce a generalization of the spherical inversion. In particular, we
define an inversion with respect to an ellipsoid. The two-dimensional analogous of this
transformation was recently studied in [11–13].
Definition 3.1: Let E be an ellipsoid centred at a point O in R3 , the inversion in E is the
−→
mapping ψ : R3 \ {O} −→ R3 \ {O} defined by ψ(P) = P , where P lies on the ray OP and
OP · OP = (OQ)2 , where Q is the intersection of the ray OP with the ellipsoid.
The point P is said to be the inverse of P in the ellipsoid E. We call E the ellipsoid of
inversion, O the centre of inversion and the number w ࣎ OQ the radius of inversion (see
Figure 10). Note that w is not constant. In particular, if E is a sphere, we obtain the spherical
inversion.
It is clear that the inversion in an ellipsoid is an involution, i.e. ψ(ψ(P)) = P. The fixed
points are the points on the ellipsoid E. Indeed, if F is a fixed point, ψ(F) = F, then OF ·
−→
OF = (OF)2 = (OQ)2 , then OF = OQ; since Q lies on the ray OF, then F = Q. Moreover, P
is in the exterior of E if and only if P is in the interior of E.
Figure . Inversion in an ellipsoid.
3.1. Basic properties

Theorem 3.1: Let P and T be distinct points with inversion radii w and u with respect to E.
If P and T are the inverses of P and T in E, then
⎧ 2
⎪
⎪ w PT
⎪
⎨ , if O, P, T are collinear;
OP · OT
PT = √
⎪ (w2 − u2 ) (w2 (OT )2 − u2 (OP)2 ) + w2 u2 (PT )2
⎪
⎪
⎩ , otherwise.
OP · OT
Proof: If P, Q and O are collinear, the line containing them also contains Q, P and T . There-
fore,
OQ2 OQ2 w2 (OP − OT w2 · T P

P T = OT − OP = − = = .
OT OP OT · OP OT · OP
Now suppose P, Q and O are not collinear. Then neither are P , T and O (see Figure 11).
Let α be the measure of angle P OT , then by law of cosines we obtain in triangle POT,
OP2 + OT 2 − PT 2
cos α = . (3)
2 · OP · OT
Also, in triangle P OT ,
(P T )2 = (OP )2 + (OT )2 − 2 · OP · OT · cos α

w4 u4 w 2 u2 (OP)2 + (OT )2 − (PT )2
= + −2·
(OP)2 (OT )2 OP · OT OP · OT
2
w (OT ) w − u − u (OP) w − u + w2 u2 (PT )2
2 2 2 2 2 2 2
=
(OP)2 (OT )2
2 2
w − u w (OT ) − u2 (OP)2 + w2 u2 (PT )2
2 2
= .
(OP)2 (OT )2
From this the result follows.

Figure . Distance of inverse points in an ellipsoid.
Figure . Inversion of a circle and a sphere.

Note that if E is a sphere, then OQ = w = u = OS, hence

√ √ 4
(w2 − w2 ) (w2 (OT )2 − w2 (OP)2 ) + w2 w2 (PT )2 w (PT )2 w2 · PT
PT = = = ,
(OP)(OT ) OP · OT OP · OT
where w is the radius of the sphere.

In the following theorem, we obtain the Cartesian coordinates of an inverse point in an
ellipsoid.
2
+ (y−b)
2 2
Theorem 3.2: If E is the ellipsoid (x−a)
u2 v2
+ (z−c)
w2
= 1, where the semi-axes are of
lengths u, v and w, then the inverse of the point P = (x, y, z) in E is the point

u2 v 2 w2 (x − a)
P = + a,
v 2 w2 (x − a)2 + u2 w2 (y − b)2 + u2 v 2 (z − c)2
u2 v 2 w2 (y − b)
+ b,
v 2 w2 (x − a)2 + u2 w2 (y − b)2 + u2 v 2 (z − c)2

u2 v 2 w2 (z − c)
+c .
v 2 w2 (x − a)2 + u2 w2 (y − b)2 + u2 v 2 (z − c)2
Proof: The proof runs like in Theorem 2.1.

In particular, if u = v = w = 1, we recover Theorem 2.1.
4. Inversion of curves and surfaces in an ellipsoid

If a point P moves on a curve C, and P , the inverse of P in an ellipsoid E moves on a curve
C , the curve C is called the inverse of C in the ellipsoid E.
Example 4.1: In Figure 12 (up), we show the inversion of a circumference in an ellipse,
while in Figure 12 (down), we show the inversion of a sphere (blue) in an ellipsoid.
Example 4.2: In Figure 13, we show the inversion of a square with respect to an ellipse and
the inversion of a cube with respect to an ellipsoid. Compare this with Example 2.1.
Example 4.3: When the inversion is applied to fractal objects (fractals beyond its rigorous
mathematical definition, present forms full of complexity and beautiful patterns), it is pos-
sible to obtain new and rich structures. As an example we have selected the inversion for the
Sierpinski pyramid as a tetrahedron which is undoubtedly one of the most popular fractals
Figure . Inversion of a square and a cube.

Figure . Sierpinski pyramid under inversion.
Figure . Inversion of a double cone and a cylinder.
in the three-dimensional space. Figure 14 shows the inversion in a sphere and an ellipsoid
of the Sierpinski pyramid, where the sphere of inversion is circumsphere of the pyramid.
Example 4.4: In Figure 15, we show cyclides or Horns obtained from elliptic cones or ellip-
tic cylinders (a cylinder whose cross section is an ellipse) under inversion in an ellipsoid.
Figure . Inversion of an elliptic cylinder.
Example 4.5: If we take an elliptic cylinder (a cylinder whose cross section is an ellipse)
under inversion in an ellipsoid, it produces a special case of cyclides called elliptical, i.e.
cyclides with elliptic lines of curvature (see Figure 16). [22]
4.1. Inversion of planes and ellipsoids
Theorem 4.1: Consider the inversion ψ in an ellipsoid E with centre O.
(a) Every plane containing O is invariant under the inversion.

(b) The image of a plane not containing O is an ellipsoid containing O and homothetic to
E.
Proof:
x2 y2 y2
(a) Let E be an ellipsoid of inversion with equation u2
+ v2
+ w2
= 1 and the plane
Mx + Ny + Tz = 0. Applying ψ to gives

u2 v 2 w 2 x u2 v 2 w 2 y
M +N
v 2 w 2 x 2 + u2 w 2 y 2 + u2 v 2 z 2 v 2 w 2 x 2 + u2 w 2 y 2 + u2 v 2 z 2

u2 v 2 w 2 z
+T =0
v 2 w 2 x 2 + u2 w 2 y 2 + u2 v 2 z 2

× M u2 v 2 w2 x + N u2 v 2 w2 y + T u2 v 2 w2 z = 0Mx + Yy + Tz = 0.
2 2 y2
(b) Let E be an ellipsoid of inversion with equation ux2 + vy2 + w2
= 1 and the plane
Mx + Ny + Tz + S = 0 with S ࣔ 0. Applying ψ to gives

u2 v 2 w 2 x u2 v 2 w 2 y
M + N

u2 v 2 w 2 z
+T
v 2 w 2 x 2 + u2 w 2 y 2 + u2 v 2 z 2

+ S = 0 u2 v 2 w2 Mx + Ny + Tz + S(v 2 w2 x2 + u2 w2 y2 + u2 v 2 z2 ) = 0.
Figure . Inversion of a system of parallel lines.
By rearranging the above equation, we get
2 2 2
Mu2 Nv 2 T w2
x+ 2S
y+ 2S
z+ 2S Mu2 + Nv 2 + T w2
+ + = .
u2 v2 w2 4S2
Note that this is an ellipsoid with centre

Mu2 Nv 2 T w2
− ,− ,− ,
2S 2S 2S
and homothetic to E with ratio
2S
√ .
Mu2 + Nv 2 + T w2

Corollary 4.2: Let 1 and 2 be perpendicular planes intersecting at a line .
(a) If O ࢠ , then ψ(1 ) and ψ(2 ) are perpendicular planes.

(b) If 1 does not contain O but 2 does, then ψ(1 ) is an ellipsoid through O orthogonal
to ψ(2 ) = 2 at O (see Figure 17).
(c) If none of the planes contains O, then ψ(1 ) and ψ(2 ) are ellipsoids orthogonal at
O.
Proof:
(a) It is clear by Theorem 4.1.

Figure . Inversion of a system of parallel lines.

2 2 2
(b) Let E be an ellipsoid of inversion with equation ux2 + vy2 + wz 2 = 1. Let 1 be the
plane Mx + Ny + Tz + S = 0 with S ࣔ 0. Its image in E is the ellipsoid given
by
2 2 2
Mu2 Nv 2 T w2
x+ 2S
y+ 2S
z+ 2S Mu2 + Nv 2 + T w2
+ + = .
u2 v2 w2 4S2
The tangent at O is the plane Mx + Ny + Tz = 0, parallel to 1 and orthogonal to

2 at O.
(c) The proof runs like in (b).

Corollary 4.3: If O ࢡ , the inverse images of the sheaf of planes through are coaxial ellip-
soids through O and P . The inverse images of a system of planes parallel to 0 through O are
ellipsoids homothetic to E tangent to 0 at O (see Figure 18).
Theorem 4.4: Let E be the ellipsoid of inversion with centre O, and E an ellipsoid homothetic
to E. The image of E is
(a) an ellipsoid homothetic to E if E does not passes through O,

(b) a plane if E passes through O.
Proof: Note that an ellipsoid E homothetic to E has equation
x2 y2 z2
+ + + Dx + Ey + Fz + G = 0.
u2 v 2 w2
(a) If E does not passes through O, then G ࣔ 0. The inverse image consists of points
P(x, y, z) for which P lies on the ellipsoid, i.e.
2 2
u2 v 2 w 2 x u2 v 2 w 2 y
+
u2 v2
2
uv wz
2 2 2

v 2 w 2 x 2 + u2 w 2 y 2 + u2 v 2 z 2 u2 v 2 w 2 x
+ +D 2 2 2
w2 v w x + u2 w 2 y 2 + u2 v 2 z 2

u2 v 2 w 2 y u2 v 2 w 2 z
+E +F +G=0
Simplifying, we have

x2 y2 z2 x2 y2 z2 D E F 1
+ + + + + x+ y+ z+ = 0.
u2 v2 w2 u 2 v 2 w 2 G G G G
Since
x2 y2 z2
+ + = 0,
u2 v 2 w2
then
x2 y2 z2 D E F 1
+ + + x + y + z + = 0.
u 2 v 2 w 2 G G G G
This is an ellipsoid homothetic to E.
(b) If E passes through O, then G = 0. Applying ψ to this equation and simplifying
gives the plane Dx + Ey + Fz + 1 = 0, then the result is clear.

5. Generalized Pappus chain

The inversion in a circumference has several interesting applications (cf. [1,3–5]). One of
them is the Pappus chain theorem. A Pappus chain [23] is an infinite series of circles con-
structed starting from the Archimedean figure named arbelos so that the generic circle Ci
(i = 1, 2, …) of the chain is tangent to the circles Ci−1 and Ci+1 and to two of the three
semicircles Ca , Cb and Cr forming the arbelos. Let rn be the radius of Cn , then the Pappus
chain theorem states that the distance of the centre of Cn from AB is equal to 2nrn . It is not
difficult to generalize this theorem to spheres (see Figure 19 ).
In [11], the author generalized the Pappus chain theorem to ellipses. In fact, let E be a
semi-ellipse with principal diameter AB, and E and E0 semi-ellipses on the same side of
AB with principal diameters AC and CB, respectively, both homothetic to E. Let E1 , E2 , …
be a sequence of ellipses tangent to E and E , such that En is tangent to En−1 and En+1 for all
n ࣙ 1. Let rn be the semi-minor axis of En and hn the distance of the centre of En from AB,
then hn = 2nrn . In Figure 20 (left), we show a complete elliptic Pappus chain.
Figure . Pappus chains of circles and spheres.

Figure . Pappus chain of ellipses and ellipsoids.
Figure . Pappus chain of ellipsoids.
We can generalize the Pappus chain theorem in ellipsoids. A Pappus chain of ellipsoids
is an infinite series of ellipsoids constructed starting from an ellipsoid E with principal
diameter AB, and E , E0 ellipsoids with principal diameters AC and CB, respectively, both
homothetic to E, so that the ellipsoid Ei , (i = 1, 2, …) of the chain is tangent to the ellipsoids
Ei−1 and Ei+1 and to two of the three semi-ellipsoids E , E and E0 (see Figures 20 (right)
and 21).
Theorem 5.1: Let (Ei )i≥0 be a Pappus chain of ellipsoids. Then the distance of the centre of
En from AB is equal to 2nρ n , where ρ n is the semi-minor axis of En .
Proof: Let ψ i be the inversion in an ellipsoid centred at B such that ψi (Ei ) = Ei . From
Theorem 4.4, ψi (E ) and ψi (E0 ) are planes perpendicular to AB and tangent to the ellip-
soid Ei . Therefore, the ellipsoids ψi (E1 ), ψi (E2 ), . . . will be inverted to tangent ellipsoids
to parallel planes ψi (E ) and ψi (E0 ). Hence, hi = 2iρ i .
6. Concluding remarks
The study of the inversion in an ellipsoid suggests interesting and challenging problems. For
example, is it possible to generalize the Steiner Porism or Apollonius problems with respect
to an ellipsoid? Moreover, we think that this topic could provoke further development by
interested readers or their students.
Disclosure statement
No potential conflict of interest was reported by the authors.
ORCID
José L. Ramírez http://orcid.org/0000-0002-8028-9312
References
[1] Blair D. Inversion theory and conformal mapping. Vol. 9, Student mathematical library. Prov-
idence (RI): American Mathematical Society; 2000.
[2] Coolidge JL. A treatise on the circle and the sphere. London: Oxford University Press; 1916.
[3] Coxeter HS, Greitzer SL. Geometry revisited. Vol. 19, New mathematical library.Washington
(DC): Mathematical Association of America; 1967.
[4] Ogilvy S. Excursions in geometry. Mineola (NY): Dover Publications Inc.; 1991.
[5] Pedoe D. Geometry: a comprehensive course. New York (NY): Dover Publications Inc.; 1988.
[6] Stankova Z, Rike T. A decade of the Berkely math circle: the American experience, MSRI Math-
ematical Circle Library. Berkeley (CA): American Mathematical Society; 2008.
[7] Frame M, Mandelbrot B, Neger N. Fractal geometry; 2015. Available from:
http://classes.yale.edu/fractals/
[8] Patterson BC. The origins of the geometric principle of inversion. Isis. 1933; 19(1): 154–180.
[9] Hummel F. A method of inversion. Math Gaz. 1942; 26(269): 73–80.
[10] Childress N. Inversion with respect to the central conics. Math Mag. 1965; 38(3): 147–149.
[11] Ramírez J. Inversions in an ellipse. Forum Geom. 2014; 14: 105–117.
[12] Ramírez J, Rubiano G. A geometrical construction of inverse points with respect to an ellipse.
Int J Math Ed Sci Tech. 2014; 45(8): 1254–1259.
[13] Ramírez J, Rubiano G. Elliptic inversion of two-dimensional objects: a graphical point of view
with Mathematica. Int J Geom. 2014; 3(1): 12–27.
[14] Gdawiec K. Star-shaped set inversion fractals. Fractals. 2014; 22(4): 1–7.
[15] Bayar A, Ekmekçi S. On circular inversions in taxicab plane. J Adv Res Pure Math. 2014; 6(4):
33–39.
[16] Nickel JA. A budget of inversion. Math Comput Model. 1995; 21(6): 87–93.
[17] Ramírez JL, Rubiano G, Jurčič-Zlobec B. Generating fractal patterns by using p-circle inver-
sion. Fractals. 2015; 23(4): 1–13.
[18] Leys J. Sphere inversion fractals. Comput Graph. 2005; 29: 463–466.
[19] Tristan N. Visual complex analysis. Oxford: Oxford University Press; 1999.
[20] Chandru V, Dutta D, Hoffmann CM. On the geometry of Dupin cyclides. Visual Comput.
1989; 5(5): 277–290.
[21] Jia X. Role of moving planes and moving spheres following Dupin cyclides. Comput Aided
Geom Des. 2014; 31: 168–181.
[22] Foufou S, Garnier L. Obtainment of implicit equations of supercyclides and definition of elliptic
supercyclides. Mach Graphics Vis. 1993; 2(4): 123–144.
[23] van Lamoen FM, Weisstein EW. Pappus chain, MathWorld-A Wolfram Web Resource. Avail-
able from: http://mathworld.wolfram.com/PappusChain.html
View publication stats

A Generalization of The Spherical Inversion

Uploaded by

Copyright:

Available Formats

A Generalization of The Spherical Inversion

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

A Generalization of The Spherical Inversion

Uploaded by

Copyright:

Available Formats

See discussions, stats, and author profiles for this publication at: https://www.researchgate.

A generalization of the spherical inversion

José Luis Ramírez Gustavo Rubiano

SEE PROFILE SEE PROFILE

The user has requested enhancement of the downloaded file.

ISSN: 0020-739X (Print) 1464-5211 (Online) Journal homepage: http://www.tandfonline.com/loi/tmes20

A generalization of the spherical inversion

José L. Ramírez & Gustavo N. Rubiano

To link to this article: http://dx.doi.org/10.1080/0020739X.2016.1201598

Published online: 13 Jul 2016.

Submit your article to this journal

View related articles

View Crossmark data

Full Terms & Conditions of access and use can be found at

Download by: [190.85.246.2] Date: 13 July 2016, At: 12:23

A generalization of the spherical inversion

ABSTRACT ARTICLE HISTORY

CONTACT José L. Ramírez jolura@gmail.com; josel.ramirez@ima.usergioarboleda.edu.co

Figure . Inversion of a point in a sphere.

Figure . Sphere inverted to sphere and an orthogonal sphere inverted to itself.

OC = μS (μS (OC )) = μS (μC ((O))).

Figure . Inversion of the centre of inversion.

In the following theorem, we show the Cartesian coordinates of a point inverted by a

(x , y , z ) = k(x, y, z), k ∈ R+ .

Figure . Circle and spherical inversion of a square and a sphere.

Moreover, from definition of inversion, we have

Comparing each coordinate, we obtain Equation (1).

Figure . Spherical inversion of an ellipsoid.

Figure . Three-dimensional Hilbert curve under spherical inversion.

Figure . Inversion in an ellipsoid.

3.1. Basic properties

OQ2 OQ2 w2 (OP − OT w2 · T P

(P T )2 = (OP )2 + (OT )2 − 2 · OP · OT · cos α

From this the result follows.

Figure . Distance of inverse points in an ellipsoid.

Figure . Inversion of a circle and a sphere.

Note that if E is a sphere, then OQ = w = u = OS, hence

where w is the radius of the sphere.

v 2 w2 (x − a)2 + u2 w2 (y − b)2 + u2 v 2 (z − c)2

Proof: The proof runs like in Theorem 2.1.

4. Inversion of curves and surfaces in an ellipsoid

Figure . Inversion of a square and a cube.

Figure . Sierpinski pyramid under inversion.

Figure . Inversion of a double cone and a cylinder.

Figure . Inversion of an elliptic cylinder.

4.1. Inversion of planes and ellipsoids

Theorem 4.1: Consider the inversion ψ in an ellipsoid E with centre O.

(a) Every plane containing O is invariant under the inversion.

Figure . Inversion of a system of parallel lines.

By rearranging the above equation, we get

Note that this is an ellipsoid with centre

and homothetic to E with ratio

(a) If O ࢠ , then ψ(1 ) and ψ(2 ) are perpendicular planes.

(a) It is clear by Theorem 4.1.

Figure . Inversion of a system of parallel lines.

The tangent at O is the plane Mx + Ny + Tz = 0, parallel to 1 and orthogonal to

(a) an ellipsoid homothetic to E if E does not passes through O,

Proof: Note that an ellipsoid E homothetic to E has equation

5. Generalized Pappus chain