Namespaces
Variants
Actions

Difference between revisions of "Elliptic geometry"

From Encyclopedia of Mathematics
Jump to: navigation, search
(Importing text file)
 
m (tex encoded by computer)
 
Line 1: Line 1:
A geometry in a space with a Riemannian curvature that is constant and positive in any two-dimensional direction. Elliptic geometry is a higher-dimensional generalization of the [[Riemann geometry|Riemann geometry]].
+
<!--
 +
e0354801.png
 +
$#A+1 = 43 n = 0
 +
$#C+1 = 43 : ~/encyclopedia/old_files/data/E035/E.0305480 Elliptic geometry
 +
Automatically converted into TeX, above some diagnostics.
 +
Please remove this comment and the {{TEX|auto}} line below,
 +
if TeX found to be correct.
 +
-->
  
 +
{{TEX|auto}}
 +
{{TEX|done}}
  
 +
A geometry in a space with a Riemannian curvature that is constant and positive in any two-dimensional direction. Elliptic geometry is a higher-dimensional generalization of the [[Riemann geometry|Riemann geometry]].
  
 
====Comments====
 
====Comments====
Thus, elliptic geometry is the geometry of real [[Projective space|projective space]] endowed with positive [[Sectional curvature|sectional curvature]] (i.e. the geometry of the sphere in <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/e/e035/e035480/e0354801.png" /> with antipodal points, or [[Antipodes|antipodes]], identified). An exposition of it is given in [[#References|[a1]]], Chapt. 19; generalizations are given in [[#References|[a2]]]. Some details follow.
+
Thus, elliptic geometry is the geometry of real [[Projective space|projective space]] endowed with positive [[Sectional curvature|sectional curvature]] (i.e. the geometry of the sphere in $  \mathbf R  ^ {n} $
 +
with antipodal points, or [[Antipodes|antipodes]], identified). An exposition of it is given in [[#References|[a1]]], Chapt. 19; generalizations are given in [[#References|[a2]]]. Some details follow.
  
Let <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/e/e035/e035480/e0354802.png" /> be an <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/e/e035/e035480/e0354803.png" />-dimensional Euclidean space and <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/e/e035/e035480/e0354804.png" /> the associated projective space of all straight lines through the origin. For <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/e/e035/e035480/e0354805.png" /> let <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/e/e035/e035480/e0354806.png" /> be the angle (in the Euclidean sense) between the lines <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/e/e035/e035480/e0354807.png" /> and <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/e/e035/e035480/e0354808.png" /> in <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/e/e035/e035480/e0354809.png" />. If <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/e/e035/e035480/e03548010.png" /> and <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/e/e035/e035480/e03548011.png" /> are two lines in <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/e/e035/e035480/e03548012.png" /> intersecting in <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/e/e035/e035480/e03548013.png" />, then the angle between <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/e/e035/e035480/e03548014.png" /> and <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/e/e035/e035480/e03548015.png" /> is the angle in <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/e/e035/e035480/e03548016.png" /> between the corresponding planes <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/e/e035/e035480/e03548017.png" /> and <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/e/e035/e035480/e03548018.png" /> in <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/e/e035/e035480/e03548019.png" /> (which intersect in the line <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/e/e035/e035480/e03548020.png" />). The space <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/e/e035/e035480/e03548021.png" /> with this metric (and this notion of angle) is called the elliptic space associated with <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/e/e035/e035480/e03548022.png" />. It is of course closely related to the [[Spherical geometry|spherical geometry]] of <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/e/e035/e035480/e03548023.png" />, being in fact a quotient. The topology induced by the metric is the usual one.
+
Let $  E $
 +
be an $  ( n+ 1 ) $-
 +
dimensional Euclidean space and $  P = \mathbf P ( E) $
 +
the associated projective space of all straight lines through the origin. For $  L , L  ^  \prime  \in P $
 +
let $  d ( L , L  ^  \prime  ) \in [ 0 , \pi / 2 ] $
 +
be the angle (in the Euclidean sense) between the lines $  L $
 +
and $  L  ^  \prime  $
 +
in $  E $.  
 +
If $  l $
 +
and $  l  ^  \prime  $
 +
are two lines in $  P $
 +
intersecting in $  L $,  
 +
then the angle between $  l $
 +
and $  l  ^  \prime  $
 +
is the angle in $  [ 0 , \pi /2 ] $
 +
between the corresponding planes $  l $
 +
and $  l  ^  \prime  $
 +
in $  E $(
 +
which intersect in the line $  L $).  
 +
The space $  P $
 +
with this metric (and this notion of angle) is called the elliptic space associated with $  E $.  
 +
It is of course closely related to the [[Spherical geometry|spherical geometry]] of $  S ( E) = \{ {x \in E } : {\| x \| = 1 } \} $,  
 +
being in fact a quotient. The topology induced by the metric is the usual one.
  
Consider for the moment the spherical geometry of <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/e/e035/e035480/e03548024.png" />, i.e. the lines are great circles. Take e.g. the equator. Then all lines in <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/e/e035/e035480/e03548025.png" /> perpendicular to the equator meet in the North and South poles, the polar points of the equator. Identifying antipodal points one obtains <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/e/e035/e035480/e03548026.png" />, in which therefore for every line <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/e/e035/e035480/e03548027.png" /> there is unique point point <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/e/e035/e035480/e03548028.png" />, the (absolute) polar of <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/e/e035/e035480/e03548029.png" /> through which every line perpendicular to <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/e/e035/e035480/e03548030.png" /> passes. Conversely, to every point <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/e/e035/e035480/e03548031.png" /> of <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/e/e035/e035480/e03548032.png" /> there corresponds an (absolute) polar line.
+
Consider for the moment the spherical geometry of $  S  ^ {2} $,  
 +
i.e. the lines are great circles. Take e.g. the equator. Then all lines in $  S  ^ {2} $
 +
perpendicular to the equator meet in the North and South poles, the polar points of the equator. Identifying antipodal points one obtains $  \mathbf P ( \mathbf R  ^ {3} ) $,  
 +
in which therefore for every line $  l $
 +
there is unique point point $  A $,  
 +
the (absolute) polar of $  l $
 +
through which every line perpendicular to $  l $
 +
passes. Conversely, to every point $  A $
 +
of $  \mathbf P ( \mathbf R  ^ {3} ) $
 +
there corresponds an (absolute) polar line.
  
This generalizes. Let <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/e/e035/e035480/e03548033.png" /> be an <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/e/e035/e035480/e03548034.png" />-dimensional plane in <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/e/e035/e035480/e03548035.png" />, then the (absolute) polar of <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/e/e035/e035480/e03548036.png" /> in <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/e/e035/e035480/e03548037.png" /> is the plane <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/e/e035/e035480/e03548038.png" /> of dimension <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/e/e035/e035480/e03548039.png" /> consisting of all points <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/e/e035/e035480/e03548040.png" /> such that for all <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/e/e035/e035480/e03548041.png" />, <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/e/e035/e035480/e03548042.png" />. Thus, for <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/e/e035/e035480/e03548043.png" /> the polar of a line is a line.
+
This generalizes. Let $  d \subset  P $
 +
be an $  r $-
 +
dimensional plane in $  P $,  
 +
then the (absolute) polar of $  d $
 +
in $  P $
 +
is the plane e $
 +
of dimension $  s = n - r - 1 $
 +
consisting of all points $  x = ( x _ {0} :  x _ {1} : \dots : x _ {n} ) $
 +
such that for all $  y = ( y _ {0} :  y _ {1} : \dots : y _ {n} ) \in d $,
 +
$  \langle  x , y \rangle = \sum x _ {i} y _ {i} = 0 $.  
 +
Thus, for $  \mathbf P ( \mathbf R  ^ {4} ) $
 +
the polar of a line is a line.
  
 
====References====
 
====References====
 
<table><TR><TD valign="top">[a1]</TD> <TD valign="top">  M. Berger,  "Geometry" , '''II''' , Springer  (1987)</TD></TR><TR><TD valign="top">[a2]</TD> <TD valign="top">  H. Busemann,  "Recent synthetic differential geometry" , Springer  (1970)</TD></TR></table>
 
<table><TR><TD valign="top">[a1]</TD> <TD valign="top">  M. Berger,  "Geometry" , '''II''' , Springer  (1987)</TD></TR><TR><TD valign="top">[a2]</TD> <TD valign="top">  H. Busemann,  "Recent synthetic differential geometry" , Springer  (1970)</TD></TR></table>

Latest revision as of 19:37, 5 June 2020


A geometry in a space with a Riemannian curvature that is constant and positive in any two-dimensional direction. Elliptic geometry is a higher-dimensional generalization of the Riemann geometry.

Comments

Thus, elliptic geometry is the geometry of real projective space endowed with positive sectional curvature (i.e. the geometry of the sphere in $ \mathbf R ^ {n} $ with antipodal points, or antipodes, identified). An exposition of it is given in [a1], Chapt. 19; generalizations are given in [a2]. Some details follow.

Let $ E $ be an $ ( n+ 1 ) $- dimensional Euclidean space and $ P = \mathbf P ( E) $ the associated projective space of all straight lines through the origin. For $ L , L ^ \prime \in P $ let $ d ( L , L ^ \prime ) \in [ 0 , \pi / 2 ] $ be the angle (in the Euclidean sense) between the lines $ L $ and $ L ^ \prime $ in $ E $. If $ l $ and $ l ^ \prime $ are two lines in $ P $ intersecting in $ L $, then the angle between $ l $ and $ l ^ \prime $ is the angle in $ [ 0 , \pi /2 ] $ between the corresponding planes $ l $ and $ l ^ \prime $ in $ E $( which intersect in the line $ L $). The space $ P $ with this metric (and this notion of angle) is called the elliptic space associated with $ E $. It is of course closely related to the spherical geometry of $ S ( E) = \{ {x \in E } : {\| x \| = 1 } \} $, being in fact a quotient. The topology induced by the metric is the usual one.

Consider for the moment the spherical geometry of $ S ^ {2} $, i.e. the lines are great circles. Take e.g. the equator. Then all lines in $ S ^ {2} $ perpendicular to the equator meet in the North and South poles, the polar points of the equator. Identifying antipodal points one obtains $ \mathbf P ( \mathbf R ^ {3} ) $, in which therefore for every line $ l $ there is unique point point $ A $, the (absolute) polar of $ l $ through which every line perpendicular to $ l $ passes. Conversely, to every point $ A $ of $ \mathbf P ( \mathbf R ^ {3} ) $ there corresponds an (absolute) polar line.

This generalizes. Let $ d \subset P $ be an $ r $- dimensional plane in $ P $, then the (absolute) polar of $ d $ in $ P $ is the plane $ e $ of dimension $ s = n - r - 1 $ consisting of all points $ x = ( x _ {0} : x _ {1} : \dots : x _ {n} ) $ such that for all $ y = ( y _ {0} : y _ {1} : \dots : y _ {n} ) \in d $, $ \langle x , y \rangle = \sum x _ {i} y _ {i} = 0 $. Thus, for $ \mathbf P ( \mathbf R ^ {4} ) $ the polar of a line is a line.

References

[a1] M. Berger, "Geometry" , II , Springer (1987)
[a2] H. Busemann, "Recent synthetic differential geometry" , Springer (1970)
How to Cite This Entry:
Elliptic geometry. Encyclopedia of Mathematics. URL: http://encyclopediaofmath.org/index.php?title=Elliptic_geometry&oldid=15989
This article was adapted from an original article by A.B. Ivanov (originator), which appeared in Encyclopedia of Mathematics - ISBN 1402006098. See original article