Professional Documents
Culture Documents
ptarg
SAGE
ENDOWMENT FUND
THE GIFT OF
Hcnrg W. Sage
A,\\:^XX
I4.1j4^ JSl5
3081
QA
608.H88
Kummer's
quartic surface.
Cornell University
Library
The
original of this
book
is in
restrictions in
text.
http://www.archive.org/details/cu31924001544406
F.
C.
ILonJon:
ffilaBBofo:
50,
WELLINGTON STREET.
F. A. BROCKHAUS.
THE MACMILLAN COMPANY.
(Mcutta: MACMILLAN AND CO., Ltd.
Iclpjifl:
#ei lord:
BombsE ani
E.C.
BY
R?
W: H: IT HUDSON,
M.A., D.Sc.
Cambridge
at the
University Press
J
9 5
1ty
ffiam&rtoge:
PEEFATOEY NOTE.
Ronald William Henry Tuenbull Hudson would have
been twenty-nine years old in July of
this year;
educated at
College in 1900
became a Lecturer
in
Mathematics at University
and
in the early
London
autumn
of 1904.
This book was then in course of printing, and the writer had
himself corrected proofs of the earlier sheets, assisted in this
work by
Mr
T. J.
I' A.
Mr H.
Bateman, of Trinity
Mr Bateman and
by the University
Press.
went
to the
making
knew
the writer
it
will
study.
and breadth;
To those who
PREFATORY NOTE
VI
brilliance
could
is
generally to be
is
as that
of
many
well spent
be spared.
H. F.
27 March 1905.
BAKER.
CONTENTS.
CHAPTER
I.
KUMMERS CONFIGURATION.
PAGE
SECT.
1.
2.
Desmic tetrahedra
The group of reflexions
5.
The 16 6 configuration
The group of sixteen operations
The incidence diagram
6.
7.
3.
4.
...
.......
CHAPTER
12
II.
The Quartic
14
9.
16
17
10.
11.
19
CHAPTER
III.
13.
Orthogonal matrices
Connection between matrices and quaternions
14.
...
...
24
26
27
CONTENTS
viii
FAGE
SECT.
28
15.
The
16.
Quadratic relations
17.
18.
19.
30
32
surface
CHAPTER
33
....
34
IV.
LINE GEOMETRY.
37
38
20.
Polar lines
21.
23.
Apolar complexes
Groups of three and four apolar complexes
Six apolar complexes
24.
41
25.
Klein's 60 16 configuration
26.
Kummer's 16
42
44
27.
Line coordinates
45
28.
Fundamental quadrics
Fundamental tetrahedra
47
22.
29.
....
39
40
configuration
48
CHAPTER
V.
30.
31.
53
32.
Elliptic coordinates
55
33.
Conjugate sets
56
34.
Klein's tetrahedra
35.
Eelations of lines to
36.
Asymptotic curves
37.
38.
The congruence
39.
40.
Relation between
41.
"
57
of second order
4>
58
and
class
....
60
62
63
63
and A
65
66
Confocal congruences
CHAPTER
VI.
43.
Tetrahedral complexes
Equations of the complex and the complex surface
44.
71
45.
The
72
46.
42.
polar line
68
,
69
73
CONTENTS.
IX
CHAPTER
VII.
SETS OF NODES.
SECT>
PAGE
47.
Group-sets
75
48.
Comparison of notations
Pairs and ootads
Eighty Eosenhain odd tetrads
Sixty Gbpel even tetrads
Odd and even hexads
76
49.
50.
51.
5~2.
CHAPTER
77
78
79
80
VIII.
CHAPTER
81
83
86
IX.
The
57.
Multiple tetrahedroids
91
58.
Battaglini's
harmonic complex
Limiting forms
94
98
59.
tetrahedroid
89
CHAPTER
X.
60.
61.
Apsidal surfaces
62.
Singularities of the
63.
Parametric representation
64.
Tangent planes
65.
66.
Curvature
67.
Asymptotic lines
Painvin's complex
68.
101
Wave
102
Surface
...
104
106
108
109
110
112
CHAPTER
XI.
70.
71.
72.
Four
73.
Two
74.
real
real
115
118
121
122
125
126
CONTENTS
CHAPTER
XII.
PAGE
75.
Linear manifolds
76.
77.
Analytical methods
130
78.
The 16e
131
127
dimensions
129
configuration
79.
80.
132
CHAPTER
134
XIII.
Geometry on a surface
82.
Algebraic curves on
83.
The e-equation
84.
142
85.
145
86.
146
137
Kummer's
138
surface
of a curve
141
CHAPTER
XIV.
Quartic curves
149
88.
151
89.
153
90.
154
91.
157
92.
158
93.
159
CHAPTER
XV.
WEDDLE'S SURFACE.
94.
95.
96.
formed
165
97.
Weddle's surface
166
98.
169
Kummer's
160
162
surface can be trans-
CONTENTS
CHAPTER
XI
XVI.
THETA FUNCTIONS.
SECT.
99.
PAGE
...
173
100.
101.
Characteristics
102.
103.
104.
105.
175
and periods
176
179
180
CHAPTER
182
184
XVII.
Tangent sections
107.
Collinear points
190
108.
Asymptotic curves
194
109.
Inscribed configurations
196
188
CHAPTER
XVIII.
Elliptic surfaces
111.
112.
The
113.
Parametric curves
114.
Unicursal curves
115.
invariant
116.
Intermediary functions
117.
Singular curves
118.
119.
120.
Birational transformations of
Kummer
221
Index
Plate (Rummer's
Frontispiece
;;
CHAPTER
I.
RUMMER'S CONFIGURATION.
1.
DESM1C TETRAHEDRA.
of
all
the projective
property.
*,
-y, -z,
),
Q,
(-,
-z,
),
p.
(-, -y,
z,t
),
z,
).
P,
S,
Then
PQRS
and
XYZT
x,
y,
y,
PS and QR
XT
meet both
and YZ, and so on
they are the most general pair, for the preceding harmonic construction is deduced from the fact that, by hypothesis, any face of
one tetrahedron cuts the other in a complete quadrilateral whose,
above property, for
is
system.
h.
KUMMERS CONFIGURATION
[CH.
DESMIC TETRAHEDRA
1]
centre
infinity.
It
is
formed with the same eighteen edges in the figure the three new
tetrahedra are infinite and wedge-shaped, each being formed by
two opposite faces of the cube and the planes containing their
;
parallel diagonals.
and
by intersecting
defined
is
lines, it is self-
a corresponding
plane-theorem in particular, the faces of any two tetrahedra can
be paired in four ways so that the lines of intersection lie in a
face of the third tetrahedron.
The geometrical properties of the
figure are deducible from the identity
reciprocal,
is
may be regarded
as current coordinates of
The two
The
following results
may be taken
as examples.
desmic system.
If three tetrahedra belong by pairs to different desmic systems, the
maining tetrahedra of the three systems belong to another system.
re-
where
are
(y
-f2)('2 -^ ) = 0,
is
given
(z2
- 2 )(.r2 -y s )=0,
metry
by Study .
2,
m,
424.
(1893), p. 382.
12
rummer's configuration
[ch.
2.
of times
The
number
is
member
equivalent to some
fact that
by successive
of the group.
shows
Considered
of the coordinates.
y, z,
two
back into
changes
expressed by the symbolic equation
The
and
this is
again,
A*=l.
Here
it is
applied
we
alter the
infer that it
must
is
the same as
this is expressed
BG = GB=A.
In other words, B and C are permutable and their product is A.
These equations, with others deduced by symmetry, are sufficient
to show that the four operations
1,
A, B,
The
multiplication table
2-3]
3.
In space of three dimensions a point or a plane may be represented by four symbols a, ft, y, S used homogeneously. The
condition of incidence of two elements (a, ft, y, 8) and (a, ft", y, 8')
of different kinds may be taken to be
On
theorem
it
will not
kummer's configuration
[ch. I
4.
and
C)
By combining
in all possible
as subgroups.
with
the members of the group, and the six points lying in each plane
are obtained by operating on the set given in 3 with the corre-
for,
is
unaffected
plane.
on
we have
/3,
7, 8)
(a, 0, 7, 8)
by that
7,
(a,
/3,
operation, thus
/3,
a), and so
7, 8)
contains the
(8,
points
(AB'),
(8, 7,
/?,
(CB')
a),
contains the
points
(B%
and so
on.
The group
many subgroups. Any two operaand their product form, with the identical operation, a subgroup: two examples are (1, AB', BG', CA') and (1, AG', BA', CB').
Further the group can be arranged in many ways as the product
simply as the group, contains
tions
4-5]
by the multiplication
1
tables
kummer's configuration
dd
(1,
[ch.
5-6]
points
LINEAR CONSTRUCTION
;
in
it,
if
they
xxx-
Let the first diagram represent any six planes; the positions of
the crosses make no suppositions as to the linear dependence of the
planes, for the diagram does not indicate that more than three
planes pass through the same point. It is required to fill in the
remaining ten places, if possible, so as to complete the incidence
by the
Thus
corre-
12
there
is
23
34
45
51
and corners
123
234
345
451
512
P.
10
rummer's configuration
[ch.
Each
making the
set
124
and
these,
235
when arranged
452
513
Q,
in the order
352
135
341
524
241
F,
413
the cyclical
of points
and
are
collinear,
341
123
513
512
452
235
Therefore,
by
are collinear,
and similarly
the theorem
is
Hence
proved.
Two skew pentagons, which are so related that the five lines
from any point to meet pairs of corresponding sides are coplanar,
are said to be in lineal position we have now proved that the
twelve pentagons formed by the intersections of five planes taken
in different orders can be arranged in six pairs such that if the
corners of one pentagon are taken alternately a new pentagon
is formed which is in lineal position with the other member of the
In this way we get twelve pairs of pentagons in lineal
pair.
;
position.
LINEAR CONSTRUCTION
6]
11
II,
III,
IV,
V,
six
results
Planes
12
5
II
Of the twenty
Points
134
523
245
IV
III
351
651
623
612
412
645
634
all six
planes.
Each plane occurs five times and the scheme is based on isolating
one, 6, and arranging the rest in a certain cyclical order 12345.
Any of the other planes may be isolated and the rest arranged
in appropriate order so as to lead to the same configuration, thus
1,
35624
2,
41635
3,
52641
4,
13652
5,
24613
it
in the
cyclical order.
Six points can be chosen out of the sixteen so that no three lie in a plane
many ways ; the diagrams are of two types
of the configuration in
kummer's configuration
12
(1)
lie
[ch.
on
different
rows and
columns,
three points
(2)
lie
in a
different rows
columns.
The
was
first
six points in
treated synthetically
The subject
by Reyet and Schroeter J who introduces pentagons
is
in
lineal position.
7.
lie
BC
point
current in
Similarly
in R.
6-7]
13
on a conic.
Further, the twisted cubic curve determined by the six points
ABCDEF
is
projected from
passing through
ABGQR.
The
pencil
A[BGQR]
is
ABG
the same as
BCFE
transformation of coordinates.
CHAPTEE
II.
8.
a cone.
The
surface usually
Kummer*,
coordinates
The
quency,
is
taken as origin.
reciprocal singularity,
is
which
will occur
with equal
fre-
It is cut
* Ernst
died 1893.
8]
15
elementary
is,
all
along a conic,
We
are situated
by
if
When
the singularities
conies on
it
are given.
are'
is
16
[CH.
9.
II
AJJD TROPES.
5 are
and
for
it
is
This
is
done in
One node
The
4, 5, 6.
is
called
and the
six tropes
through
it
are
1, 2,
3,
When
on a
conic,
the six
surface at 0,
it
35,
51,
24,
46,
62
vertices at
Now
a 16 6 configuration
two at a time
named
:
and
are
named by
either
single figures
and the remaining ten by the partitions of these six into two sets
of three.' The two nomenclatures may be used simultaneously,
one for the nodes, and the other for the tropes of the same configuration, as above, and then we have the following types of
incidences
is
12
incident with
1,
2,
2,
1,
3,
4,
5,
6,
,.
0, 12,
13,
14,
15,
16,
123
23, 31,
12,
56,
64,
45.
9-10]
17
two
f
i
3
5
m m m
m m m
Ml m m
With homogeneous
coordinates
is
24
35
i4
ie
51
32
34
36
13
52
54
56
y, z, t let
x,
the node
be
where #
62
12
10.
(0, 0, 0, 1),
46
+& =
at
(),
s in x, y, z.
is
& = 0,
and the enveloping
4>3
is
</>2<>4
is
= 0,
Both
six planes.
taken to be
of these equations
are given.
t
= 0,
the equation
&=
represents the repeated conic passing through the nodes in that
trope,
and hence
<f>4
is
known when
to a numerical factor.
fa^yO-xz.
It is convenient to represent
in
x = y/u = z/u?.
Let
ky,
xs =
h.
of
u giving the
lines of
- 2ks y + z=0,
(s
l, 2, 3, 4, 5,
6)
18
[CH.
II
fafa = ^ia>
fa*
where \
aw?5#6)
an undetermined constant.
We shall take t = to be the plane (123.456) and then
= fa" where fa = is the conic in t = passing through the
fa
nodes 23, 31, 12, 56, 64, 45. It is convenient to represent any
point in the plane t = by the parameters u, v of the two tangents
from it to the conic y* = xz. Thus we write
is
then
is
now
2kt y + z = x (u k e )
given by u = k u v
=k
ke
(v
or
),
by u = k2
= ki.
the equation
- k ) (v - k ) (v - k ) (v - k )
(' - h) (u - k
) (u
(v h) (v & ) (v k ) (u kt ) (u k ) (u k ) = 0.
division by u v it becomes symmetric in u and v,
After
the
expressed in
satisfied
by the
(, v)
= {h,
and hence
is
six points
h), (ks
fa =
where
fj,
is
kt ), (kit kt ),
the conic fa
Ug
then
k2 ), (k ke ), (ke
(ky,
h),
2/&B2
an undetermined constant
further
4fa = x (u-v)
=
fa*
fafa ~Kx u u u ui u ui v
i
and
vi vi vi v 6 v i
so that
os~fa
and
W Ve,
function of
fa
On
(U
- Vft/x = +
+
= fa
in the equation of
= fa J fa* fa fa,
ft,
(ihUzUsViVtVe
2/t
fljtf2
V8 4 WB W6 )
JUyUvUtUiUiUsVyVtViViVs v
10-11]
19
/tt
= u + v,
z/x = uv,
t/x = (u
2y/x
-2
tf)
and the
(v 1 M2 W s fl 4 V6 ?;6
+ Vtfi0
3 tfs
M4 M6 W6)2
we may
0,
etc. ;
= x^ 0,
etc.,
xs = ue vt
(w v) */x13S
and
a^jB
so on.
(k.2
Then
it is
(s
= Ju ui u3 vi v^ + */v
t
V2 V s Ui U
!i
1, 2, 3, 4, 5,
6)
'Us,
a;si4
+ (^ &2) Va^a^ = 0,
and
(h - h) (h - h) ^X^X.^ + (h h) (h ~ h) ^x3ls xm
which are two of the many
irrational forms in
11.
It has been shown that the six tropes through any node cut
any other trope in two triangles whose corners are nodes. Hence
if the six nodes in any plane are partitioned into three pairs,
there is another trope through each pair, and these three tropes
meet in a point which is not a node. The four planes are the
faces of a tetrahedron f and contain singular conies intersecting
by pairs on its edges. The quadric which can be drawn through
two of these conies and one of the remaining two nodes cuts each
* Darboux, Comptes Rendus (1881), xcu, H93.
t Named
after Gopel.
22
20
[CH. II
Thus
entirely.
them
xyzt
k<f>
When
where the value of k depends on the coefficients in
k=
we have the four faces of the tetrahedron, and when k is
small, which is the case most easily realised, the surface lies near
the faces. There is a distinction, important from a metrical point
of view, between the cases when k is positive and negative.
In
the former case the product xyzt is positive and the point (x, y, z, t)
lies either within the tetrahedron or in one of the wedge-shaped
regions opposite the edges.
Fig. 2 shows the region for which
<f>.
Fig.
x, y, z, t
it
Fig. 3.
2.
are
all positive,
and
fig.
Fig. 4.
by the substitution
x'/x
= y'/y=-z'/z=-t'/t,
and gives a
"
tetrahedron
= y'/y=z>/z = t'/t,
" of
11]
21
The
first
by
six lunes.
0.
determine
an orthogonal pro-
<
Fig. 5.
jection and shows one of these circles completely and parts of the
other three. The twelve remaining conies are hyperbolas and six
Only
of these are drawn, of which three appear as straight lines.
three nodes of the inner tetrahedron are shown.
Two nodes which are adjacent on one conic are adjacent on one
other and belong to the same piece of the surface.
* If
may
radius r of
0=0,
+4 (/i+ 1)
(/*
- 2) 2 xyzt=0.
//.
are connected by
frontis-
<f>
The
+ xy),
The
22
piece, of
which
fig.
is
[CH.
II
portions of
by the planes
is
divided
of the tetrahedron,
first.
Fig. 6 shows that the four circles divide the surface of the
sphere into four smaller triangles (three are shown), four larger
Fia. 6.
* Models of the sixteen conies passing by sixes through sixteen points are easily
out of coiled steel wire, beginning with the four circles. They have the
advantage over plaster models that the surface is transparent and does not hide
alternate arcs of conies.
made
11]
and
six quadrangles.
23
On
each of the
an
set.
CHAPTER
III.
(a re ).
matrices,
is
its
conjugate matrix a,
a = (ars)
The
implies
corresponding elements
The product
ars = am
same order is
+b=c
implies ars
+ brs = crs
effected
by adding
thus
.
same order n
is
by
crt
=2
8
a rs bst
(r,t
= l,2,
...
n),
=1
b.
= ab.
* For a completer account see Baker's Abelian Functions, p. 666, where numerous
references are given.
12]
The
second
is
the definition
it
associative, for if ab
so that
it is
important to notice
first
From
factor.
tion of matrices
(CCLjpq
25
=c
and bd
= e,
but ab and ha are in general different matrices and the multiplication is not commutative.
are zeros
is arbitrary; accordingly all the elements of
except those in the leading diagonal, which are unities. It then
= E. For these reasons may often
follows that aE = a and
in which a
EE
be replaced by 1, or omitted.
Let A,.s be the minor or cofactor of ars in
of
a,
the determinant
then
= %r A^atr,
if s
=j= t.
Hence A rs / a is the srth element of a matrix which, when mulIt is called the inverse matrix and is
tiplied into a, gives E.
-1
have
written a
so that we
\
a~ a =
1
When
=f=
0,
E = aa~
1
.
of the
first
determinant of coefficients
vanishes has no inverse.
The
may
by n
may
be taken as the
any row
which the
matrix whose determinant
n elements
of
.
\
easily
If
= ab,
c = ba,
_1
= &_1 a_1
c
c = a \b\.
c
26
[CH. Ill
Accordingly,
if
is
xay = 22 xr ayg =
The
22 y a^xr = yax.
K
whereby
suffixes
first
instance a device
may
be omitted
in
becomes a
almost
calculus,
single
like
algebraical
An
orthogonal matrix
;
thus a
is
chief
use
of
to express clearly
algebraic' identities.
is
orthogonal
aa
This condition
The
and
ORTHOGONAL MATRICES.
13.
conjugate
manipulate matrices
to
quantities.
to obtain easily
is
its
But,
the
is
the inverse of
if
= E.
Since
aa
2t t
Ct,pg
=1,
Z g(lpa aqa =
l
we have deduced
2 a gp2 =1,
Further,
a sp a eq
= 0.
aa = E,
whence
it
bb
= E,
and
c
c is their
= ab,
product we have
= ba,
follows that
cc = abba = aEa = ad = E.
Therefore the product of two orthogonal matrices
is
an orthogonal
ORTHOGONAL MATRICES
12-14]
matrix.
|
= a
|
= E, therefore \a\ a = 2 = 1
= + 1. Hence the determinant of
Further, since aa
|,
therefore
a
+
27
but
an
orthogonal matrix is
1.
The theorem that complementary tworowed subdeterminants of a four-rowed orthogonal matrix are
numerically equal will be useful, and can easily be extended
an
28
[CH.
Ill
because qq"
is scalar.
theorem in matrices.
Let
p = ia + j/9 + ky + 8,
q = ix+jy + kz
r=pq = i(
+j(
t,
@,a\x,y,z,t)
8,-7,
B,-a,^^x,y,z,t)
7.
+ Jc(-/3, a, h,y\x,y,z,t)
+ (-a,-0,-y,B~^x,y,z,t).
Now
rr'
= pp'.qq\
linear forms is (a 2
true for
all
the matrix
8
/8
-/3
a"
/3
. a /3 7
is
8_
orthogonal.
= i (a, /3,
7, 8
z,-y,x)
*,
+j(u,P,y,S]i-z,
+ k(a,
/3,
^, 7,
(a,
7,
8$
t,
x,y)
,
0)
SJ-a.-y,-*,
t),
y,-#,
is
orthogonal
if a?
+y
15.
yx'
z t x
yx t
_x y z
+ z + = 1.
2
y
z
t_
(a, /S, 7, 8)
to denote a
and
(dd)
= (a,
/3,
(6c)
=(7,
8, a,
7, 8
x,
y, z, t
),
l3^-x,-y,z,t),
14-15]
29
= (a,
0, 7, 8
5 z, t,-x,- y).
it is
(ab)
(ac)
(ad)
30
[CH. Ill
by operating on
then
for
(x, y, z, t)
all
QUADRATIC RELATIONS.
16.
we
find
/3
87
7
8-
7
fi-
51
a-
and
8- YJ
first
t-
y x
t x y z.
is
a, /3, 7,
orthogonal.
Now
8 by the operations
x, y, z, t
is
written
(dd)
(ac)
(ba)
-(ab)
(da)-(cc)
-(be)
(cd)
-(ca)-(bb)
(cb)~
(bd)
(db)-(aa)
(ad)
(do),
and the linear forms occupy the same positions as the corresponding
elements in the incidence diagram (p. 8).
Hence in order to find
a set of six, between any four of which a linear identity exists, we
have to exclude from a row and a column the element common to
both.
Many interesting geometrical theorems of fundamental importance for the configuration follow from the algebraic identities
QUADRATIC RELATIONS
1,5-16]
injplied
X X
Zg
Ti
Y*
7*
Z
T
Zi
_TX
then, taking any column,
X + Y +Z + T
s
Write
is orthogonal.
Z4
4 _
we have
= (a + /3s + 72 +
31
8 2 ) (a2
+ y + z + t%
2
showing that the four planes X, Y, Z, T are the faces of a tetrahedron self-polar with respect to the quadric x2 + y2 + z2 + 2 = 0,
and
which
self-polar tetrahedron
is,
X,
Y, Z,
Since a similar
column we get
+ X + X - Y - Z? - 2V = 0,
2
3
2
4
lie
on a
x,x
+y y + z^z + 2\ r, s
%
0,
(XjY&TJ
is
inscribed in
so also the
tetrahedra
(X 7,Z,T ) and
{X2 Y^Z T,) and
1
(X.F^Zy
are similarly related.
{X.Y.Z.T,),
(Zj 7,^2*0.
and (X.Y^T,),
X X +Y Y =
Z Z +T T =
1
0,
Named
t Paul
after Bosenhain.
:
:
32
[CH. Ill
and eighteen of the latter, making thirty sets in all, and of the
quadrics through any one set, four are plane-pairs, being planes of
the configuration.
17.
On substituting x =
a,
= /3, z = 7, t = 8,
of quadric functions
"
'[dd]
which
is also
[da]-[cc]
[bd]
[cd]
[db]
- [ad]
0-[bb]
[ad]
orthogonal.
[dc]_
2a/3-2 7 S,
+ 8,
+ 2 7 8, - a2 + /3 - 7 + S
-a- S , -<y*
j
2a/3
.
27a -2/38,
2
are divided by a
2/3 7
+ @? + y* + S
2aS,
is
obtained
27 +
2
,
-a
2/3S
2/3 7 - 2aS
-/3 + 7 + S
2
when
"
16-18]
33
(e.g.)
[aa]
is
the quadric
= 2(aS-l3y) =
the plane
(aa)
and
(8,
- 7, - /3, a $
so,
y, z, t)
= 0,
so on.
parts
in the configuration.
is
multiplied
by
The ten quadrics are the only invariants of the second degree under the
group of sixteen linear transformations.
Each of the ten quadrics corresponds to a partition of the operations
two sets of three. The product of each set is the
same, and gives the symbol for the quadric.
If the product of the operations p x q x and p 2 q 2 is pg3 the point (p^i) and
the plane (p%q^) are pole and polar with respect to the quadric [p^], and the
four points (p^i), (Pill)) (ft?3)> (^) are *^ e corners of a tetrahedron selfab, ac, be, ba, ca, eb into
18.
The
six linear
/a
/3
P 7
S\
\<c
y
y
tJ'
34
[CH. Ill
line joining
the
31
Then
= (8,
(ac) = (8,
(be) = (7,
(ab)
y, ft a
$ - x, y,-z,t) = p -p
li
S3
z, t)=p u +p
$ x, y, z,t) = pM -p
x,-y,-z,t)=pM + p
(ba) = (7, 8, a, /3 $
x,-y,-z, t ) = p -p^,
(ca) = (ft a, 8, 7 $
=
y,-z,t)=p +p
(cb)
(ft a, S.yQ-x,
7,
ft
a~&-x,-y,
!ss ,
S, a, /3
31
31 ,
3i
Si
12 .
On
equating these to zero we get six fundamental linear complexes, and the null-planes of (a, ft 7, 8) are those six planes of
the configuration which pass through
it.
In the next chapter will
be found a detailed account of this system of complexes. As their
name implies, they are of fundamental importance in the theory of
the 16 8 configuration, and are unchanged when the same operation
of the group is performed upon (a, ft 7, 8) and (x, y, z, t).
It is to be noticed that the ten quadrics and six complexes are
determined by the coordinates alone, and that then the points of
the configuration are the ten poles and six null-points of an
arbitrary plane, and the planes of the configuration are the polars
and Dull-planes of any one of its points.
The
a, |3, y,
we
[dd] (be)
19.
right,
when
divided
by
[dd],
Z.T, + Z,T,+ Z3 T3
+ Zt Tt =
18-19]
X,
fi,
35
v satisfying
\ + fJt.+ V=0,
the four equations
Zj =
the
T=
. .
/j,
Suppose that
(a, /3, 7, 8) is
(00,
y, z, t)
through
the result
= [2FJ.
Then the
surface
(a, /3, 7, S) if
Xl[Zi Ts]
agreeing with
let
be indicated by square
\ + fi + v =
among
the
fundamental quadrics.
Hence Kummer's
and seven other forms can be deduced from this by the group of
This
linear transformations which leave the surface unaltered.
groups
two
complementary
with
connected
is
equations
eight
set of
of eight associated points, and so the total number of equations is
15
= 120.
One equation
of fifteen equations
of each set
is
rule.
Multiply each
32
36
are satisfied
when
is
it
[CH.
Ill
Rummer's
surface.
10,
we arrange the
suitable constants, in
gonal matrix, and then compare the elements with the linear forms.
an orthoLet
further let
-,
c
123
-2
cr456-2
= (*! - kt (kt - k6
)
(*!
then
S136
orthogonal,
and
if
S4
Se
S3
- S6
is
C2
all
real.
Further
the matrix
Vc
136 ! 136
_0_
_0_
Vc14sl46 Vc162^162
Vc362 362
VcM2 |662
V e34ol346
VC546&46
is
also
orthogonal,
showing that
c136 ,
etc.
_0
Vc124 m
VC324 324
Vc624^624.
correspond to
fundamental
quadrics*.
CHAPTEE
IV.
LINE GEOMETRY.
20.
POLAR
LINES.
is
configuration receives
lie
From
these facts
all
and
all
its null-point.
be deduced.
The
is
We
tween polar lines. Introducing a symbol S to denote the correspondence we may write
S(P) =
*
ir,
(tt)
= P,
lines (Minchin).
2=1,
lines (Eouth),
and reciprocal
LINE GEOMETRY
38
[CH. IV
null-plane,
write
S(x) = y,
and
x,
S(y) = x
lines.
APOLAK COMPLEXES.
21.
when they
are, that is
T are not
commutative
when
ST=TS,
be said to be apolar* and ST or TS may be
the complexes
of an involutory point-point and planesymbol
the
taken to be
or collineation f for any plane ir has
transformation
linear
plane
will
two
null-points
P = S(ir)
TS(P)=TS*(Tr) = T(Tr)=Q,
ST(Q) =
ST>(Tr)
= S(Tr) = P,
so that
involutory.
it.
first
complex, then
S(x)
= y,
TS(x)=T(y),
S{T(x)}=T(y),
T does
first
complex.
Further,
if
a;
is
a ray of the
= x,
T(y) = S(x) = y,
T(x)
and y
*
is
also
reciprocal (Ball)
Two
ray.
the former
The
relation
when
latter suggests
(Klein)
is
and
false analogy.
geometrie,
i,
70, 115.
20-22]
APOLAR COMPLEXES
39
therefore such that the polars with respect to one complex of the
rays of the other are also rays of the other.
The assemblage
oo
= T(x),
TS (x) = x,
S(x)
whence
and the common polar
transformation
ST
lines correspond
to
themselves in the
common
ray in
number
if
22.
Su S2 Ss which
,
= S S y = /S (Si, T = SiS
(1, Tu T T ) whose multiplication
Ti
forming a group
2.
2,
Ti
T,
Ti
T,
T,
T,
Ti
Ti
similar to that of
table is
(P),
(P) =
is
3 (P)
SM
Since
(P)
LINE GEOMETRY
40
We
[CH. IV
points and four planes forming a tetrahedron such that any three
faces are the null-planes of their
common
them
in the
three complexes.
polar.
Four
mutually
Slt $ S St and
2,
3,
planes which
may
transformations.
Slt S2 Ss S4
S 8i S SS ...,
$S S
four planes
six points
four planes
. .
making
If
we arrange
SiS^SgSi
the
first
S2 S3
S S4
1
SSS
S
$!
&,$!
S-lSz
o 2 o4
Si
of a tetrahedron
whose
faces are
Ss
and each of these planes contains one of the points in the second
row, which is its null-point in the complex S4
Thus the configuration can be regarded in four ways as a pair of circumscribed
and inscribed tetrahedra.
.
| 23.
The
SIX
APOLAR COMPLEXES.
Koenigs,
22-24]
41
apolar to the
means
Any
rays
since,
The
directrices of (12)
24.
may
The quadrics
common
to
(123, 456)
1, 2, 3, 4,
follows that the directrices (34), being the diagonals of the quadria pair of polar lines with respect to both quadrics.
lateral, are
[CH. IV
LINE GEOMETRY
42
The tetrahedron
quadric (123, 456).
Hence
of the
fifteen
six
are self-polar, and of the ten quadrics six are circumscribed about
any one tetrahedron which is self-polar with respect to the re-
maining
Each
four.
(123, 456)
(126, 453)
(153, 246)
(156, 243)
with respect to which (14, 25, 36) is self-polar, is its own reciprocal
with respect to each of the other three for the first is completely
determined by the four directrices (12) and (13) whose polars with
respect to the second quadric are the same four lines.
;
25.
The
edge
is
common
fifteen
KLEIN'S 60 16 CONFIGURATION.
common
Of the
sixty
same edge.
faces pass
and similarly
six
fifteen
We may
the symbols 12 and 21, and make the convention that sets of
three lines obtained by an even number of interchanges of figures
46
25
13
35
61
24
An
even
24-25]
KLEIN'S 60 1B CONFIGURATION
43
number
21
43
56
64
52
13
35
61
24
Hence
and (34, 25, 16), are collinear with the point 56, 13, 24,
and the same is true for the other corners of (56, 13, 42) so that
tbe two former tetrahedra are in fourfold perspective. Hence the
three tetrahedra represented by the columns of the table belong
to a desmic system ( 1) and the rows represent the other desmic
system which is formed with the same edges.
Again the following three sets of lines are similarly related,
(12, 46, 35)
12
34
56
12
34
56
12
34
56
46
15
32
54
16
32
54
26
13
35
26
14
36
52
14
36
15
24
showing that there are four desmic systems containing the same
one tetrahedron (12, 34, 56). Hence through any one corner of
the configuration pass sixteen lines containing two other corners,
and in each face lie sixteen lines through which pass two other
From the desmic properties it follows that the assemblage
faces.
of these lines is the same in each case, and this number is
60 16/3 = 320.
.
their edges
common
= (12,
to
edge.
following table*:
(aft)
6),
among
34, 56)
[CH. IV
LINE GEOMETRY
44
we have
26.
We
now turn
summer's 16 6 configuration.
1, 2, 3, 4, 5,
correlations determined
the points
Now
1235P...
1234P,
ISP...
12P,
12Q = 3456P
123456P.
P,
12P,
13P...
56P,
Q,
12Q,
13Q...
56Q.
quadric (123) similarly the tetrahedron Q, 56Q, 64Q, 45Q is selfBut these quadrics are
polar with respect to the quadric (456).
;
it
with the pole of this face with respect to the quadric (123, 456).
We have therefore derived from an arbitrary point a configuration consisting of sixteen points
fact,
Kummer's 16 6
null-planes
configuration.
The planes
That the
lie
is,
in
the fact that the triangles 12, 23, 31 and 45, 56, 64 are self-polar
with respect to the section by their common plane.
9.
common
The null-points
of these planes
on their
lines,
rummer's 16 6 configuration
25-27]
45
the same names. The nomenclature is thus interpreted by regarding the name of each element as the symbol of operation
deriving it from an arbitrary point 0.
The symbol
itself must
therefore denote the identical operation and the other operations
1, 2, 3, 4, 5,
=
The diagram
11
22
= 33 = 44 =
55
= 66 =
by
123456.
(p. 17)
46
62
24
35
12
14
16
51
32
34
36
13
52
54
56
now seen
to
135
146
162
124
346
362
324
546
562
524
27.
LINE COORDINATES.
The
ideas
may
La
GSometrie R&glee.
LINE GEOMETRY
46
quadratic relation
complex
~Zas x,
m and
and 26s *s
let
[CH. IV
fl (a)
may be
special
The com-
il are inverse*.
are apolar if
0.
complexes
term
in OjOj
squares.
We
is
the same
sum
shall suppose
Hence
XI (a).
of six
then the
quadratic form
six squares,
can make
&>
By using
(x)
= H (x) = x? + x? + xi + x? + xi + x \
e
and
a>
fl.
much
simplified,
though
is liable
to
is
Xi=Pu-pw
ix2
=p u +p
Xs=Pu-psi,
ixt
!ia ,
=pM + p
sl
<i=Psi-Pn,
ixe
=p +p
si
1!i ,
where it will be noticed that all the coordinate complexes are real,
but three of the coordinates are imaginary. On referring to 16
we see that the coordinate complexes are the fundamental complexes employed in the construction of the 16 configuration.
The lines (xu 2 x3 xit xB xe ) and ( xlt xit xa x4 x5 x6 ) are
polar with respect to the complex x1 = 0.
Hence the transformation of lines associated with each complex consists in changing
the sign of the corresponding coordinate.
Taking only the first
three complexes, a line (x) gives rise to a set of eight lines
In order to deal with points and
(+ xi> + #2> 3> xt> x5> xe)planes we must suppose (x) to describe a sheaf of lines through a
point P, then {xx xit xs xit xs xs ) describes a plane field 8l (P),
the null-plane of P in the first complexf and ( xlt x2 x xt x5 xe )
describes a sheaf whose vertex is S2 S1 (P) or TS (P), and so on.
This method expresses the operations of the group (1, Tu T2 Tz )
in a form which brings out clearly the comparison with the group
,
of reflexions ( 2).
*
t Compare
22.
FUNDAMENTAL QUADRICS
27-28]
28.
The condition
FUNDAMENTAL QUADRICS.
+ 2/2*2 + yz + y z + y
2/i*i
it is
47
zs
za =
+y
0. 0, 0),
(0, 0, 0,
z4 , z 5 ze )
,
intersect
by two
is
intersecting generators
tyi
where
yf
either of which
is
+ yi + y =
s
>
= z + z + zg.
+ xf + ooi = 0,
x?
+ x +
= 0,
The equation x? + xg +
(123, 456)
x?
+ xi + # + x? + x? + x =
2
six poles of
any plane
lie
on a
which
is
a necessary and
sufficient condition
(p. 31).
may be
=p u pw> Z2=Pu+p&,
etc.,
the
written
zf
+ zf + zf^zi + zf + zi,
showing that the triangle of poles 246 is obtained from the triangle
135 by a transformation which may be regarded as a rigid rotation
From this we infer that the points 135
in an " elliptic " plane.
points 246 on the conic.
If the quadric
with
the
alternately
occur
"absolute"
of
an
elliptic space,
the
regarded
as
is
(135, 246)
48
(!,
LINE GEOMETRY
[CH. IV
by eliminating one
It is easily seen,
such relation be
common
to
VlSS^
(^286^1 "T
"T"
Vl23*^5
== ">
arbitrary,
2
s
relations,
of rays
FUNDAMENTAL TETRAHEDRA.
29.
Rays common
xx = 0, x2 = 0, xs = 0, *4 =
satisfy
i,
0, 0, 0,
0) and
(1,
- i,
0, 0, 0, 0),
which are seen to be polar lines with respect to both of the complexes #! = 0, #2 = 0, and are therefore the directrices of the
congruence
Similarly
(12).
may be
directrices
The edges
is
= 0, x
ix2
ixk
= 0, xB
is
By taking
we may arrange that
= Pu-p23,
3=P2ipn.,
ixz =pu+Pas,
ixi = p u + p
cance.
= 0.
real
coordinates
ixe
If this tetrahedron
any edge
the other
all
xx
of
the coordinates of
found.
example of
p.
31
46
is
Xn=pSi -p 2,
1
ix e =p3i
+ pn
all
the directrices
FUNDAMENTAL TETRAHEDRA
28-29]
may
49
of Klein's configuration.
There are two tetrahedra (12, 35, 46) and (12, 36, 45) which
have the edges (12) in common with the tetrahedron (12, 34, 56)
of reference, and it has been proved ( 23) that the corners on a
common edge form three harmonic ranges. It is easily found
that the corners of (12, 35, 46) are (1, 0,
those of (12, 36, 45) are (1, 0, 0, 1), (0,
0,
1,
i), (0, 1,
*,
0),
and
In this way,
1, 0).
by taking different pairs of opposite edges of reference, six tetrahedra are found. The remaining eight form with (12, 34, 56) four
desmic systems, and it is therefore sufficient to give one corner in
each system ( 1). These points are (1, 1, 1, 1), (1, i, i, 1), (i, 1, i, 1),
(i, i, 1, 1), and the corners of any tetrahedron are obtained from
these by changing the signs of an odd or even number of co-
For example
ordinates.
(i,
1,-i.l),
(i, 1,
i,
1, i, 1), (i,
1,
i,
1),
i,-l).
hedron as
follows,
arbitrary
constants
namely
+3+2+
15.
u=
We
\2 + fi? = 0.
Take one tetrahedron
pair
is
stants).
are in the
first
it
instance
(a,
0,
0,
(0,
a',
1,
1))
)J
b,
0,
1, 0,
V,
0,
1)|
)}
0,
(0,
(c',
c,
1,
0,
1)
)
and the new constants must be chosen so that the edges intersect.
The points (\a, \'a', V, X) and (fi, /J.b,jj,'b', p) can by proper choice of
\, \', (i, ix be made the same if aa! = b/b', and so on thus a' = b/c,
By taking new multiples of the coordinates and
b' = c/a, c' = a/b.
thereby absorbing three arbitrary constants we may put a=b=c=l,
;
and then
a'
(1, 0,
(0,
h.
1,
= V = c' = 1. The
0,
1,
1)1
)}
0,
1,
0,
1))
(0,
0,
1,
)]
(1,
1,
1)
1, 0,
0,
now
LINE GEOMETRY
50
[CH. IV
pairs of edges in
common
with
The
(1,
0,
(0,
1,
0,
i,
t))
)|
0,
i,
1,
0,
0,
1,
i)\
(0,
)j
(1,
0,
i,
1,
i)\
0,
)j
may be
derived.
p.
43
CHAPTEE
V.
We
character, which
ordinates.
The rays
At a singular point
a double line and therefore breaks into two plane pencils. Correlatively, the rays which lie in any given plane envelope a conic
aud
Each of these points, being
the vertex of a pencil of rays, is a singular point, and similarly
each of the planes at a singular point is a singular plane and
which
vi
and
xvii;
42
1.
52
Thus
[CH.
P the
at a singular point
77-,,
tj- 2
-rr
The proof
surface.
and
used,
of this
is
instantaneous
when
coordinates are
same
cross ratio
it.
We
class;
tangent lines at
are rays.
G
may
of
we
and planes
for
by the
the other.
*
The
quartic section
complex
is
suitably modified by
means
if
it
among
linear
30-31]
null-points
P P P P P
which
is
lt
2,
4,
The
Pr
P as well
53
PPr is
a bitangent, touching
<I>
31.
canonical forms*.
x? + x* + x? + x? + x? + xi =
+ ks x^ + k x + \x? + k x + ke xe =
coordinate complexes xs =
are the six
kx x-?
Then the
2
s,
and
(2).
self-polar
is
them are
(1),
fundamental
by changing the
obvious that (2) is unaltered by
it is
effected
this procedure.
special
then
(3).
When
(1), (2), (3) are satisfied the lines (x) and (kx) determine a
they also
singular point and a singular plane of the complex
;
ys
= k x fix
s
2
Hence
(y) is
(k,
- /*)- y. = 0,
singular plane
We
(*,
/*)- 2
yf = 0.
replacing ks by (ks
therefore the
are
same
/*)_1y and
are
as before.
infinite
2(*,-/*)r *f
1
system of complexes
(4),
which have the same singular points and planes, and are therefore
* First adopted by Klein, in his Inaugwraldusertation (Bonn, 1868)
Math.
54
[CH.
ft)~\
and
family in such a
of the
way
to the cosingular
same complex.
To determine the
on any line
members
order of 4>
(y).
and suppose
tains (y)
we
we may
its coefficients to
all
the
select
be ks
2#g ys =
0,
1ks xt ys =
0,
Xks xs =
2
0,
The
2xa ys = 0,
jectively.
meet any
and
Whence
Xks xs ye =
common
0,
2xs zs =
0,
on any line
is
it.
theorem is independent
and it may be used to prove that the locus
the same as the envelope of singular planes.
<E>,
of singular points is
For, if two out of four elements coincide, their cross ratio vanishes,
and conversely if the cross ratio vanishes, at least two elements
Hence the locus and the envelope have the same
coincide.
tangent lines and therefore coincide. It is important to notice
that
when the
among
the
may
ELLIPTIC COORDINATES
31-32]
55
ELLIPTIC COORDINATES.
32.
/*,
(x).
This equation is of great importance in
determining the relation of (x) to 4>; its roots are called the elliptic
Let (x) cut <& in Plt P2 P3 P4 and let the tangent planes
through (x) touch <J> at A, B, C, D, respectively.
The lines
pencil
belong
to
the
tangent
at
A
and are
lt
2
S
t
respectively singular rays of four cosingular complexes whose
parameters are fi u fi 2 /as
The rays of the first of these
4) say.
complexes which lie in the singular plane PX AP4 form two pencils
one of which has its vertex at Px and therefore contains (x) hence
(x) is a ray of each of the complexes and fi u
jjls
2
^4 are its
elliptic coordinates.
We have seen that the lines
S of the
tangeut pencil at A are protectively related to the parameters
hence the same is true of the points Ps and the lines
fjbg and
Now four points can be proprojecting them from B, G, D.
jectively related to themselves in only four ways, and further
APi and BPX cannot be singular rays of the same complex if (x) is
not a tangent to <1> at P,; thus if we suppose that
GP3
lt BP2
DPi are corresponding tangent lines and singular rays of the
same complex, the four sets of singular rays must be
,
AP AP AP AP
,
/u.
/j,
AP
AP
APU BP GP PP
2
3,
BP GP DP,
AP PP CP DP
AP BP CP DP,
AP>,
lt
3,
4,
4,
3,
2,
in complex
fi lt
/i,,
/*,.
tn,
By
P P P
,
any one
Since
/^,
we must have
O - h)-
/i
of
56
[CH.
where
f(ji)
(/j,
- h) (p - k) (/i -k
and
)(ji- k4)
(7=2A;8
(i
On
multiplying by
line coordinates in
ft
((jl
terms of the
- k ) (fi-k
= ks we
elliptic coordinates in
obtain the
the form
CONJUGATE
33.
SETS.
P\"i, "l"l
we
m"(i
PP
"l"it -tl-fg
AD, BG
and the elements of each
ratio remain unchanged.
BD, GA
set
CD, AB,
the cross
32-34]
CONJUGATE SETS
57
The four points in which any line cuts <E> are determined by a
biquadratic equation, and the four tangent planes through the
same line by another equation. The relation between these
equations is that the cross ratios of the roots are equal. This
condition implies that the ratios of the roots of the reducing
cubic are the same for both equations, and hence that if one
biquadratic
two quadratic
factors,
separated into
two
is
expressed as the product of
the other biquadratic can be similarly
factors
of a
34.
With
family
KLEIN'S TETRAHEDRA.
let
<E>
in
see that
it is
possible to
AiBzCsDi and
We may
name the
A^G^ are
1;
and so
on.
A
A
Br
C,
XJ 2
C2
X/2
p*
is,
p,
is
58
A^B^D^
and
is
A^GiDi
[CH.
common
it
lines of intersec-
transversal
P^PsPi.
PAPB
follows that
is
of points
Now
PA
1
is
Pj [.AiPJ
of conjugacy
it
same complex
is
(A.)
PA
follows that
(ji^)
and
[Pi-B2 ]
is
in particular
Pj^ljSjPa is a quadrilateral
1
is a quadrilateral of rays of (/j^), and every edge of the tetrahedron
is a common ray of two of the three complexes (X), (/*,), (/j?).
Such a tetrahedron may be constructed by taking any three
points on a tangent plane section and completing the three conjugate sets determined by this plane and pairs of points; the
three new tangent planes meet on the surface.
We see in this
way that a given surface has oo 5 inscribed and circumscribed
!i
tetrahedra *.
RELATIONS OF LINES TO *.
35.
We
shall
among
now
relations to
The
line,
and
its
corresponding
$.
fi
2 (*.-A*)^ *.*=<>
and are uniquely determined by the line (so) but conversely, an
arbitrary set of roots determines thirty-two lines, polars of each
;
In what
follows,
t(k,-\)and
Put
x,i
and
(kg \)ye,
then (y)
is
Z (k, - \)-2 */ =
is
a singular ray of
0,
a tangent to
2is 82 = 0.
<1>.
The
34-35]
TANGENT LINES
elliptic coordinates of
59
=
f =
(fi -^(p- fa) =
'2,(k.-\Y(kt -ii.Y*!f.*
0*-X)2(*f
or
or
(jj,
- X)
-^y
l(kg -k
^ = 0,
This shows that the line
is
)- 1
oc-'
= 0.
property of
bitangents
this
way
six
it is
all
if
X=
//. 1
class.
or fa.
Now we
have
seen that fa and fa are the same for all the lines of a tangent
pencil, and hence the whole pencil belongs to the cosingular complexes whose parameters are fa and fa but this can be the case
;
we have proved that (kx)fa (x) and (kx)fa (x) are the inflexional
tangents at the point of contact, and the elliptic coordinates of
these lines are
and
= fa
xs = (h - th) (h
- fa)
{/' (&,)}"*
(oft,*
+ 6*. + c)
<.
Hence
if
{/'(*.)}"*
and
common
point
60
[CH.
tions of signs for the radicals, showing that 4> possesses sixteen
the case of a proper bitangent two pairs of points and two pairs of
planes coincide while only two elliptic coordinates are equal.
If four roots are equal, then
*.
Either (x) passes through a node and three points and four
tangent planes coincide, so that (x) is a generator of the tangent
Hence the
tangent lines at a node and the tangents to a singular conic are
projectively related to the tangents at an ordinary point, correplanes coincide, so that (x) touches the singular conic.
ASYMPTOTIC CURVES.
36.
We
if (x) is
quadratic in
fi
(x) are
the
given by the
/a
S(*,-/*)"1
= 0.
by a
single relation
We
shall
An
by
= const,
inflexional tangent at
or
/i 2
= const.
P can
Hence
35-36]
ASYMPTOTIC CURVES
61
we want
tangent
is (y),
where ys
The
condition
is
=
- x8 d/i}* =
2 (k - n)* dec* = 0,
(k,
to) (k, - p*) //' (kg ) for
2cfy,
or
or
{(ks
/j,)
dxs
and on substituting
ai this
becomes
+ 2d d ^ + ^d ,A=o,
T
%tr
f (h) \h^rd^
^^^h^to
k
to
fh
""*
or
We
have seen that every bitangent has two of its elliptic cothe remaining two are parameters
Hence the
As
= (k - h) (k - to)h {K - Ma)*
s
/' (&)}"*
scroll of
bitangents circumscribing
<I>
is
at
number
of correlations.
62
When ^=fii,
is
[CH.
cone or a tangent of a singular conic. The cusp locus for asymptotic curves reduces to the sixteen nodes, and their envelope is
Each trope
is
by
The parameter
same
is
the
37.
Among
curves corresponding to
/j. 1
=k
i.
The
ki,
hi,
fi
^ = {h-kif{k
-iM)lf\ka ),
of degree
8 in
/*.
line
The
lowering of degree*
On
putting two elliptic coordinates equal to k x and /c2 respecthe line coordinates x 1 and <v2 vanish, and so the various
combinations of sign give only eight lines. Hence two principal
asymptotic curves cut in eight points, besides the nodes, where
tively,
they touch singular conies. At any common point both inflexional tangents have four-point contact, and the whole pencil of
tangents belongs to the congruence (12). Hence the eight common points lie by fours on the directrices of (12), which are so
related that the tangent planes through each touch at the points
*
conic,
of a given
It.
'
36-39]
on the
63
other.
property*.
38.
We
CLASS.
belongs to one of
4>
six congruences, of
#i
0, (ft2
one
elliptic coordinate
/i x is
equal to
ftj
is
a bitangent of
<J>,
for
satisfy
(ft,
or
= 0,
is
ftx
congruences, which
Kummer
may
we
may be taken
is
to
is
be cosingular.
sums of squares.
Thus the theory
is
made
to
surface of a complex.
independent account.
39.
is
given a short
= 0,
0.
2\,# a = 0,
through every point in space pass two distinct rays, the intersections of the plane
The equations
of the
congruence being
X^
0,
+ X^ + \4 # + X,*, +
a
4
\6 a;6a
may
64
The
[CH.
We
rays coincide.
the congruence.
when
it,
(x) of
if
(y) is
yx
so that (x)
and
(y) are
= 0,
t~ks xs y s
common
rays of
= 0,
the complexes whose
all
coefficients are
A3&3,
A, 2 372 ,
^fc,
Aqd/4,
A>gd?g
A^tf/g,
/x.
Now two rays of a linear congruence
do not intersect except on the directrices; these are the lines
whose coordinates are the preceding set of coefficients when
/*
+ 2\V = 0.
Let them be called (f) and ('); then the two consecutive rays
through the point (x, ) pass through consecutive points of (f ')
and therefore
of the
(as
+ das)
lie
complex xx =
be any consecutive ray; then
point
(x,
in the
')
= 0,
"Z^s dxg = 0,
Let
= 0,
X^s'dxs = 0,
'%\
dxx
whence
contains ().
xs dxs
which prove that the pencils (x, %), (x, %) have envelopes. These
the focal surface <J>, and we have seen that (x) touches it at two
Hence the surface is of the
points and meets it at no others.
fourth order, and reciprocal reasoning shows that it is also of the
.are
fourth class.
(x) is
2 Xe
Let
A be
W=
0.
X2 #2a +
then the pencil
(x,
?W + ?W + X.aV + ?W =
\x) of tangent lines of
0,
belongs to the
39-40]
complex
65
= 0,
and the singular ray (as) counts as two intercomplex cone at the point
Hence A and <J> touch at this point. Since 2Xs2#g2 = 0,
(), (') coincide and (x) has four-point contact with <&.
!
the lines
In order that case (3) may arise, every line of the pencil (x, X#)
of the quadratic complex and hence (x) must be an
inflexional tangent of A.
These rays are given by the equations
must be a ray
^ = 0,
lx,a
27W = 0,
= 0,
in
~Z\W = 0, X\
The
number.
fact
W=
0,
surface
established.
is
RELATION BETWEEN
40.
4>
AND
A.
By comparing
S\,a:, a
we
see that
<1>
is
~kt x*
x
Since
is
other.
the equations
and 2
(k,
- k,)' x? = 0,
1
X2 22 + Xs x32 + X4 #2 + X6 a;62 + X6 62 =
we
is
0,
mutual.
The addition
of the
of the complexes
Xj*! 2
+\
x2* + Xs *32
through
h.
all
<3>
+ X4 04 + X
2
6 a?6
+X
6 a?6
their nodes.
5
66
[CH.
CONFOCAL CONGRUENCES.
41.
surface
F of
zx =
it.
Now F is
the singular
we had
ratic
it
is
all,
CONFOCAL CONGRUENCES
41]
67
zt
= 0,
and by symmetry
0,
x{ =
and
it is
t,(\,-k y- zt' =
l
0,
2&s #s
= 0,
(i
= i,
is
the focal
6)
0,
= 0,
two
2( kl -k )-*w/ = 0,
t
sets of coordinates
where
a rs (kr
-X)
s
/' (kr )
We
ZigdfgXg,
</>'
k )(0 2
= -/(X ) (kr)
k )(0 - h){6 - k )(d - h),
(Xs )
</,
e ).
" ways.
At any point P of F draw the tangent
draw any conic through P cutting the section
again in six points.
A partition of these into two triangles
determines the bases of two of Klein's tetrahedra having a common
vertex in this way ten more points and fifteen more planes are
found which complete the configuration of nodes and tropes of S.
geometrically in
plane and in
oo
it
For
different values of
touch the focal surface of the congruence given by this equation and #j=0
along the same octavic curve, and for ten values of ax the discriminating
sextic has a pair of equal roots, and the corresponding singular surface has a
pair of coincident nodes in each trope and therefore a nodal line.
52
CHAPTER
VI.
42.
Raumes
tention of
complex.
named
is
It is a degenerate form of
Kummer
plexes.
It
is
To prove
this
we
is
that every
recall that
Sturm,
in, 355.
42-43]
69
0, 23, 31, 12, has its edges belonging by fours to three fundamental complexes 1, 2, 3 so that by rejecting those whose edges
belong to a particular one the number of available tetrahedra
is reduced to forty.
We therefore start with a tetrahedral complex and select the
rays cutting a given line.
We shall take the singular surface of
the complex for tetrahedron of reference and use current point
coordinates x x x it xa x4 and plane coordinates u lt u 2 us u4 and line
,
coordinates
The
etc.
so that
&=
reference,
()
and
of reference
they
43,
bpuPn.
+ CpuPn = 0,
gives
where
= b c,
and therefore
= c a, y = ab,
a + /3 + y = 0.
/3
+ UtPu + UsPsi = 0,
aih/p^ + fiu /p + yu /p =
UiPu
whence
sl
lss
0,
xa
= 0,
u x1 + M2 ie2 + u3 x3 + u4 x4 =
r
mixed
For the
in
=
plucker's complex surface
70
[ch. VI
touch the section of this cone. The last two equations give the
complex curve enveloped by the rays in any plane (w). If on
the other hand we regard (x) as a fixed point and the u s as current
plane coordinates the equations give the complex cone of rays
(a;), as may be proved by exactly correlative reasoning.
Hence the two equations completely represent the complex the
former can be replaced by the alternative forms
through
+ ^aUiSCi = 0,
/yujc^+ */au x + V^w 4 a;4 = 0,
vftihXi + vau^x^ + V7M 4 = 0,
+
*/yua x2
*Jftu a xa
3
(in
of
which the signs of the radicals are ambiguous), so that the want
symmetry
To
we
is
only apparent.
u x1
1
+ V/3 + vyu x = 0,
+ u^x^ + W3# + u4 x = 0,
2 a?2
As we
need
we
shall not
have further
be (p)
it
(#') is
= 0,
?>i + f>2 + f>s + f
Hence if (x') is any point on the
'
where / = X^x,'.
fi'A*!
&'/"
= fs'M
is
Vaf,^ + V/3& x2 + V7 fs *3 =
an equation which
is
conic section,
Z*'/ u*>
0,
ft
+7=
and %%s xe = 0.
Va^Mj +
\//Sa;2
M2 +
^<yxs us
= 0, %x u =
s
0,
VawjM-!
+ V/3u
w2 +
"i/yviUs
= 0,
From
these equations
#i = 0, &=0
it is
= 0,
is
a nodal
Vi =
line,
that
are singular
43-44]
71
44.
We
line.
+ V/Sw^jj + Viyitg&Vj =
0.
a repeated
u3 = q13 ui = qli
,
we
all
PqiA - yq x =
13
This
surface
line,
which
it
0.
is
the singularity
is
the fact that the tangent plane does not change as the point of
contact moves along the
plane,
is
line, as in
The plane
= 0,
torse or developable.
lines,
is
touched by
Ui(PquUs + vqls u2 )
0,'
and consists of the two points (1, 0, 0, 0) and (0, y/q^, ftjqu, afau)
which are both nodes. These points are the vertices of pencils of
and correspond to
rays of the complex in the singular plane %x =
AA
the points
Thus
we have
of 34.
in addition to
eight
they are
A
B
(1,
0,
0,
0)
A,
(0,
1,
0,
0)
B, ( 7 / ?21
(0,
0,
1,
0)
A (0,0,
0,1)
7/Sm
0,
08/ ?B
A (/?,
0,
a/q
/3/ ?13 ,
a/q u )
a/q,
0/qJ
0,
y/qj
/8/g,
7 /?.
Correlatively there are four points on (p) for which the complex
cone degenerates. Regarded as an envelope, the cone with vertex (ai)
Vowjjtt!
* Pliicker
names
+ V/Sa^ + ^yx3 u3 =
Cayley,
vi,
334.
72
is
<c,
then x2
[ch. VI
= pm x = p
,
ls
xt
= pu
and
Wi
This
which
line,
^12^2 - VPu>us = 0,
= p12 2 + Pi S u3 + pu ut
is
= 0.
through
Vj
= 0,
it
is
v^
...
vt
maybe
P P P P
lt
2,
s,
of p. 55.
They
are
Regarded as a
cone at
vj
12
xx
= 0,
+ rpis^a) = 0, #4 =
and
inter-
yx2/p12 +
+ <Wi/Pu = 0.
y&Bs/Pm
their
coordinates
are
to
at
An
zj 2
@i
45.
THE POLAR
LINE.
The locus of the poles of the nodal line (p) with respect to
complex curves in planes through (p) is a straight line, for it must
lie on the polar plane of (p) with respect to every complex cone
whose vertex is on (p). This polar line% cuts each torsal line and
* Pliioker
names
See Cayley,
ti,
Math. Ann.
iv, 1.
123
Sturm, Linien-
44-46]
73
is
it; reciprocally,
tropes through
it.
The complex
complex
surface
is
current coordinates,
(Pu/Pu + Palp*)
The
*3~*6 ^ g Kj k$ ^ 8_l~6 s.
x? TT
v? TJ
-<1 T
+ ?-J? -"S
+ t-^t
T--"5
? =
+ r-^r
T -"4
1*8
1
the tetrahedron
The
X,
'"2
i.
r.
surface
parameters
is
(b-c)(c-a)(a-b)
ui + iivi
and three nodes, no two of which
g-6)( +X)
U^
+ ll.%
lie
on the
singularities.
46.
J.
is
by Cayley,
the nodal
line,
which
is
In
fig.
7 four
of the conies through the nodes are ellipses, touching each other
by pairs
at the
four are hyperbolas, touching each other by pairs at the other two
pinch points.
74
Now
[CH. VI
which the two tangent planes coincide they divide it into segments
through which real and imaginary sheets of the surface pass alternately.
Since the torsal lines are real both of the segments with
Such a segment is the common edge
real sheets are here finite.
of two finite wedge-shaped pieces of the surface, the angle of the
Fig.
7.
torsal lines.
As
pieces.
it
CHAPTER
VII.
SETS OF NODES.
GROUP-SETS.
47.
On
be studied in
detail.
in
subsequent applications,
Kummer's
configuration
must
Kummer
The 16 6
configuration
collineations
geometrically each
is
effected
its
To each
collinea-
same subgroup. This set of different sets is here called a groupand contains the whole configuration. The sets of a group-set
are equivalent in the sense that they have the same projective
set
( 5) is
is
The
effect of
any collineation
[CH. VII
SETS OF NODES
76
to interchange
make
diagram, the symbol which after these changes is in the first row
and column is the name of the corresponding collineation.
48.
COMPARISON OF NOTATIONS.
47-49]
AND OCTADS
PAIRS
77
A set
called
odd
number
the
is
number
parity
significance.
PAIRS
49.
AND OCTADS.
XX
XX
ITTT
I
One
The
pair
identity
is
and the
symbols.
collineation represented
odd and four even, and, further, the partition into two
sets of four pairs is invariant under the group. The four odd pairs
form an odd octad and the four even pairs form an even octad;
these two octads together make up a group-set, and are said to be
set four are
associated.
An
octad
is
among the
It
SETS OF NODES
78
[CH. VII
50.
These are tetrahedra whose corners are nodes and whose faces
Two of the nodes cap be chosen arbitrarily, the third
must lie in one of the two tropes containing them both, and then
the fourth is determined. There are two types of diagram, in (1)
the points lie in a line and represent the corners and faces of the
same tetrahedron in (2) the points lie in two lines and represent
the corners of one tetrahedron and the faces of another the faces
of the first and the corners of the second are represented by the
other points in the same two lines. The rows and columns give
are tropes.
(1)
(2)
'
...
eight tetrads of type (1) and each of the six pairs of rows and
columns give six tetrads of type (2), making eighty
six pairs of
in
all.
is
1, 3,
represented by
5
or
by
0, 35, 51,
13
The nodes
1, 3,
face.
is
the
We
common
and
all
corner of
common
face.
All the tetrahedra of type (1) are self-polar with respect to one
Since there are ten quadrics
of the fundamental quadrics (p. 31).
we obtain in this
way ten sets of eight tetrahedra, making, once more, eighty in all.
The Rosenhain tetrads are sufficiently characterised by the
From the
of a tetrad
is
identity
inter-
49-51]
TETRADS
79
changes any two corners interchanges the other two at the same
time.
Hence each tetrad is invariant for a subgroup of four
members and there are four tetrads in a group-set. Each set of
eight which are self-polar with respect to the same quadric contains two group-sets
for example the four rows represent the
tetrads of one group-set and the four columns those of another.
:
The
which
faces of a
all
follows that the four singular conies in the faces do not lie on a
quadric surface.
it
51.
lie
*..
2
<
'.'.'.'.
.
'.
>
'.
The Gopel
all.
and
is 0, 12,
34, 56
by three
for
lines
tetrahedron.
The tetrahedron 0, 12, 34, 56 of either nodes or tropes and the fundamental
tetrahedron (12, 34, 56) belong to a desmic system. By taking the latter for
reference it is easily seen that the third member of the system together with
those obtained in a similar way from the other tetrahedra of the same groupFurther the faces of a group-set of Gopel
set form a 16 configuration.
tetrahedra of nodes form another 16 6 configuration.
fl
The
lie
on a quadric.
80
SETS OF NODES
52.
[CH. VII
derived, in
constructed
( 6),
and
is
odd tetrad
A Weber
hexad
is
named
after
even tetrad
odd hexad
Weber.
The
total
number
of hexads
192.
is
CHAPTEK
VIII.
53.
two coordinates
4,
that
we
for reference*
write
unchanged by
by changing the signs of
is
is
+ y i + zi + t + Wxyzt
i
+A
(oft*
Now make
ditions
the point
-" -54
_ /3 + 7
a S -/3V
a/378(S +a -^ -7
4
n=
(a S
and on eliminating
among
'
node
D in the
G,
-S
B= 7 + -/3
-7 a
/3 S
8) a
(a, /3, y,
which determine A, B,
or
+ ff4__,y4_g4
'
- 7 -a )(S + 7 -a -/3 ) (a
2
7 ) (/S ^ - 7 a ) (rffr - a /3 )
(S 2 +/3 2
- yS
forms
a, y8, 7,
S2
-a /3
+/3
'
+ 7 +S )
2
'
the coefficients
Making use
- A - 52 2
(7
+ ABC + D = 0.
2
we can
write
'
_(
n_
g2
+ - & ~ 7s) (S + + /3 + 7 )
~
(a8 - 7) (a8 + 7)
2
64a/9y8 [do]
[eft]
[dc] [dd]
The equation
is
[do] [dd]
[aa] [ad]
_
~ /37 S (2 .A)
(2
eK
'
- B) (2 - 0)
[dd]*
h.
'
82
[CH. VIII
tetrahedron of reference
is
numbers
Aug
rCi
"*4
"'S
""B
(ac)
(6a)
(be)
+ J[bb]
(cb)
J[da]
= 0,
(dd)
to the values of
->
(ca)
~[da][cc](dc)
(bd)'
:
(cb)
35)
(p.
by the plane
at the nodes
(ca)
_[ad]
[66]
[dd]
and
[cc]
'
'
[da]
[aa]
'
'
their cross ratios are equal to the cross ratios of the corre-
sponding
kg.
this is only
t)
(k6
k )
2-A
= [cta3[cfa] _
~
[aa] [ad]
'
A _ (k + h) (h + k)-2 (k k + kX)
2
(k -k )(k -k
w nM
.
|2 = (h
similarly
J
+ h)
^ -+k
s)
+ Ua)
~2(
^
f
k
2)
) (h
6
.*?
(kB
C ^ (K + h) (h +
(k
kt )-2 (k x k,
/c2
)
(kg
+ hh)
ki)
coefficients is the
Lastly
jy^
l,
when
it
k1 ka (k
ls
may
easily
be
verified that
is
the
53-54]
83
Thus
tion.
all
(kjc^), (k3 kt ), (&6 &) to form an involuthe coefficients in the quartic equation are expressed
surface
is
The equation may be written as the sum of the squares of the fundamental
quadrics multiplied by coefficients of which a typical one is
JifC-tlCafCa
~p AtvAlGeJCa
"T" 2i
"
fC prC (jtC'7
is
taken
if
a three-figure symbol t.
54.
Take the
new
orthogonal matrix
(p. 30),
x1 = (dd),
xi = {ab),
x3
= (bc), xi = (ca).
o?i
CLO
Cg
OC
*3?3j
xit
ca
fl?2
>
""
""
***1 1
^4>
3,
~~~
***$
>
**4
*^1j
x2
***4t
^"8
"""
^2
#i
showing that the terms xfx? and xfxf have the same
and that the term x?x^x% gives rise to the expression
respectively,
coefficient,
(x\x3
+ a^as4)
we
the form
v?
(sW + a')
)(
+ 2uv{
)(
+ 2sx^xi x3 x = 0,
),
or
2
3
(!
- 2 ) + SXxXiOCt)
2
X Cayley,
Coll.
62
84
This equation
surface
is
[CH. VIII
Rummer's
common
to a
Chap,
vi,
and the
linear
complex
<h*Pu +
<i*.Pv.
+ /3+7 = 0,
be
to
where now the qrs are any constants, not the coordinates of a line
Further we use the abbreviaas in the case of Plucker's surface.
Then the complex curve in the plane '2v,iXi =
tions ^i = 2g a;g.
,
is its
\//3w2 :2
+ "Jyu
= 0.
x3
is
two rays of the congruence through (a/) are the two tangents from
the complex curve. If (x') is on the focal surface-, these two
rays coincide and (x) must lie on the conic.
Hence the equation
(of) to
is
Vaf^ + V/8&E, +
Before expanding
Vyf.aj,
= 0.
We
<Zi4=<?28=
therefore write
qu
2,
%x
= q<u=m,
qm
nx2 mx3 +
nXi +
2 =
lx3
= qns = n,
lxit
+ mxit
+ nxit
Vaf^+
gives,
V/3^*2 +
*/yi; 3 x3
u=la,
s
= tfcr
The same
(y-/3) + v'13-1
surface
w = ny,
= m[3,
obtained
(a
- 7) + w'y-' (0 -
when
a).
is
a + /S + 7
= 0.
54]
85
ota
/82
72,
and a3
/33
73
and
it
is
w =
v2
v?
+ oo"+
is
P1P2
71Y2
v?
1?
aia a a 3
ftftft
71 727s'
ratios
metrical form
3
= (A-7i)G8.-7)(ft-%)
+ (!-&) (*-&)(* -A).
of
numbers
(l lt
nh, r^),
(l2
m^, n^,
( 3 ,to3 ,w3 )
may be
regarded as the direction cosines of three mutually orthogonal lines, and Rummer's surface can be written in three equivalent
yli li x1
(1-lXi
is
+ Wi m^j) + Vm
2
+ Jn
t
tetrahedron
is
n t x l (n^i
ro3 a;2
+ m x l^) =
1
a degenerate form of
0.
Kummer
surface in
+n
(pst
+ P12) =
The
others are
Pli=Pv>,
Pm = Psi,
Psi=Pll,
is
x12 + xi + x3 + xii = 0,
2
!>
same form
as the
point equation.
The equation
(cf.
p. 21 footnote)
where
sp^+q^+r^-Zpqr-l.
86
55.
The equation
of
[CH. VIII
Rummer's
where
<v
= 0,
...z'
= 0,
are
+ 'Jyy' + \/zz' = 0,
When
six planes.
(xyz), (x'yz),
= x'\
a; =
yy'
= zst
)'
y = Q = y'\
zz' = xx
'
'
namely
(x'y'z),
(x'y'z'),
z=0 = z'
xx' = yy
3, 1
2 2
1,
11,
where ^ denotes a cubic cone having one double line, and so on.
Let to be the number of nodes of the first kind then since each
;
number
We take x = 0, y = 0, z =
or 7.
node of the
first
if
f=
is
is
(m +
28)/6, that
meeting in a
kind, and the quadric cone of tangents there to be
A = a? + y + z
Then
of tropes
to be tropes
0.
B*-AC = fixyz0,
a constant and 8 =
along three generators.
where ^
A=
is
The nodal
on 8
line
is
may be taken
arbitrarily, say
7=0, Z = 0,
then 8
arbitrary constants.
= (x-y-z) YZ-yZ*-zY\
functions of them.
T and Z may
Introduce
X+Y+Z=0,
then
= xYZ + yZX+zXY.
55]
The
=0 = 2,
x=
= X,
x=y= z =
lie
x=
= = x,
87
on the
lines
= y,
X=Y=Z=0,
x=
= yZ + zY,
z=
y=
= Z,
z = = xY+yX,
0=Y,
y = = zX + xZ,
zX + xZ=0, xY+yX = 0.
yZ + zY=0,
Now
the
first
first
of these cones
C=\(yZ+zY)\
and we may put \ =
2
leading
un-
of the surface
is
By
is
{(a
The equation
(of
since
Then
determined.
introducing
new
0.
linear expressions
*i
this equation
+ (yZ + z F) =
= -Z,
Z=+Y,
Jx%
+ *Jyt) +
*fzt;=0,
tropes
-q
= 0, f= 0.
cubic cone
= = f,
77
when
+ mT) + n= 0,
of a thirteen-nodal surface
(1866), p. 114,
and Cayley,
Coll. Papers,
made
to vanish see
vn, 293.
88
where
+ m+n =
[CH. VIII
0,
Hence an
+ nly +
Imz =
0.
form
Now
exist.
unaltered
is
if x, y, z, f,
nj,
are replaced
arbitrary.
it
ax
+ by + cz + ag + fir) + y = 0,
There
is
six planes
and the
coefficients in this
new
be
= bfi = cy.
verified directly,
and
tropes
is
may be
arbitrary.
ax + by + cz
= 0,
as
new
among the
By what
may easily
tropes are
+ by + cz = 0,
vanishing for
the
to
cubic
a'x
b'y
where
and, as before,
four
form*
new
tropes
appear,
etc.
+m
ls Xi
+ m8 a;3 + ns x + lf
x2
+n
xs + lfx Xi
t
x(
- n^a?, + l^)
ljs (,,
+ m -^x2 + nf
2
+ mf x + w
l
x^
-1
8
a;s
'
= 0,
= 0.
CHAPTEK
IX.
THE TETEAHEDEOID.
We
in particular positions.
With
taken
If
(a, /3, 7, 8)
of one node.
We
them from
(ab)
(fia)
(fib)
(ca)
(cb)
(da)
(db)
(ac)
(p.
29)
90
[CH. IX
a Tetrahedroid*;
is
two
conies, intersecting
in four nodes.
The
lie
in the faces.
of the form
F (x
and
after equating
factors.
z\
t )
+ Xxyzt = 0,
any coordinate
Under these
to zero,
The
the equation
may be
h
h
to
2
x2
f m
'
By
conditions for
F show
that
=0.
z2
2
i,
302
new
coordi-
THE TETRAHEDROID
56-57]
nates x, y,
the form
z, t
and putting a
= (fl)l,
rf
/3
= (gm)l, y = (hn)i,
a2
of
/3
f
7
z2
7*
this takes
=0,
x*
a?2
r
/3
/3
91
or on expansion
tfjSV (xi + yi + z* +
+
which
is
(/3
- </ - a
+a
(y
2
i!
In order
(a 4
t')
- /34 - 74) (a + y z )
+ z x ) + 7 ( 7 - a4 - /3 (< + a?f) = 0,
equation (p. 81) becomes when S = 0.
(a 4
- /34 - y)/^
= (ji + v)j'iiJiv,
7
2
2
2
y vvX, z by z "/X[i
put
etc.
yu.
ft
i/
+ y + z ) (a
- {a
2
The two
x2 +
(b
+c z)
+ c ) x + b (c + a ) y2 + & (a + b ) z + a
b2 y2
b 2 c2
= 0.
common
points of contact of a
the variable quadric touches eight fixed lines and the tetrahedroid
special,
The
inscribed in a
When
common
is
(a2
is
MULTIPLE TETRAHEDROIDS.
57.
Double tetrahedroid.
The
which
trope, from
all
tetrahedra.
92
loss of generality
t0 be
(0, 0, 0, 1)
This gives 8
and
= 7,
ft
and
[CH. IX
(0, 1,
1,
corners
0).
becomes*
{a?
As
- 1*)* +
{y
J {a?# + fz*) - S (*
- zj +
a fundamental corner
it
* )
(^ +
**)
spective for two triangles formed by the six nodes in the trope.
In the present case the two corners lie one on each of the common
edges of the two tetrahedra, and the line joining them lies in a
Hence in the case of
face of each and therefore joins two nodes.
a double tetrahedroid every trope passes through two fundamental
corners which are collinear with two nodes, and a second trope
passes through the same two corners.
The arrangement of nodes on any
conic
that
is
projectively equivalent to
shown in
mental
fig. 8,
sextic
II (k
Ak + Bk* +
s
the
Ck.
arbitrary
form
Corresponding to
numbering of the
figure
Triple tetrahedroid.
it is sufficient to
make the
corners.
and
(1, 1,
1, 1)
so that
Then
and a + /3~7=0.
is
If
(0, 0, 0, 1)
MULTIPLE TETRAHEDROIDS
57]
93
Quadruple tetrahedroid.
Any
are not
collinear
are
necessarily
It is sufficiently general
and then
= 0,
Fio.
= /3 = 7,
yitf
10.
_ # & _ ^yl. = Q
The
formed into
its
simplest form
Sextuple tetrahedroid*
is
ke
1.
does not
arise,
+y+z+t=
94
[CH. IX
0, 0, 1),
(0, 1,
-1,
0),
-1),
(0, 1, 0,
0, 1, 0),
(*,
(0, 0, 1,
-1),
(i, 1,
0, 0)
of the tetrahedra
(12, 35, 46),
we
so
take a
surface
p = <y== 8 = 1.
i,
of
the
is
l
os
The nodes
+ yi + zl + t + 4<ixyzt = 0.
i
+y+z+t=
in the trope ix
are
l,i,l,
-l),(l,-i, 1,-1),
and
if
of the tetrahedra
quadrilateral.
By
replacing ix by x
x*
we obtain a
+y
+ zi +
ti
real equation
kxyzt
= 0,
may
and
The whole
configuration of nodes
58.
The
is
by fours on twelve
lines *.
tetrahedroid
is
Rummer's
a special form of
surface
due
to
k^ + hat* + hx
belong to an involution f.
is
+ hx? + k x,? + k
xe2
If
for involution is
/Tl/i'o
Ki ~r
A/g
n/gn/4
/C3 "f~ i^
fC^fCg
tCB
ftg
^ U.
t Sturm, Liniengeometrie,
vii ; Schur,
57-58]
95
h Ox - + h Os - oot) + h (*e 2
2
3 )
O=
0,
+P12 ) = 0,
distinguished, in this notation, by the absence of product
and
is
(Ps4
terms.
all
for,
a^ + a xf + a x? + a x? =
0,
+ \xl =
0,
b x x?
+ \x* +
b % x
the condition for a line (p) to cut them in two pairs of harmonically
conjugate points is
l,(ar b 8
+ ag br )prs =0,
>
is
it follows
(ar + \b r ) (as
touch any given line. If their parameters are X and X' and the
line is a ray, the equation of the complex expresses the condition
X+V=
0.
and
it is
when
polar lines of a singular ray (p) with respect to the two quadrics
which touch it pass through the points of contact and lie in the
tangent planes, which must' coincide since the two polar lines
Hence the congruence of singular rays consists of all
intersect.
the generators of the developables circumscribing the pairs of
quadrics
Xas xs2
for different values of X.
ktbgXn 2 =
96
When
the complex
given, there
is
may be used
an
is
to define
infinite
[CH. IX
number of
it.
be written
%Ap' =
0,
A = a^bg + a br
A=a a a a = A2& /a
= O3a4.Au + <ho>z -Am.
A (c^a^f (aia ) + A A = 0.
where
Let
then
and
4,
Si
Write 4A
= a-
2
,
s,
Xi
then
2A M a
a2
= (1 + ^/l-aAA M
2A li as ai =
Similarly
rs
(1
).
Vl aA^A^
= 0, and
complex
cone
B,
Again
at any point of it breaks up into the two tangent planes.
the intersection of A and the reciprocal of A with respect to B
lies on the singular surfape, for the polar plane of any point of it
with respect to B cuts A in two straight lines and the complex
cone consists of the planes through them. The quadric surface
G, 1c s <vs * = 0, where
2
c8 = as + ftbg /as
quartic intersection of the quadrics A, 2aa:s2
The
"ZbgXg' = 0,
lies
on the singular
value of
fi
exists for
G is one
If '= b b
which
/' and we
all
values of
fi,
and one
find
.
Cl=
2A Si c
whence
!i
AWA AU
1S
a,'
etC
'
= ~^~2A u a
a4
= "(l-^l-aA w A )
2A u ci ci = " (1 + Vl <rA A u),
Si
say,
and
so that
12
cuts
B in a quartic
etc.
a but
to
4
n
58]
97
defined in a reciprocal
manner
ft h,
the complex
a2 w 2 2
+au +
2
3
+ &,' + /3 w + /3
2
3
= 0,
a4 M42
4
2
4
= 0,
is
complex tangentially.
The harmonic complex includes among its rays all the generators of
the quadrics in terms of which it is denned and conversely if a quadratic
complex contains both sets of generators of a quadric it must be
:
harmonic.
The
as,n+l = 0, n a
l,
a % a Z, n ai,
^Og, n/a a, n
6,
s,
= 0,
e+^-en
',
surface,,
u i, n
^ aa, n/"a, n
say,
"a,n
members
if
%A 2i A zl A u A 12 - 2A 3i A 12 A ti A 23 - 2A u A 23 A 2i A 31 =0.
Express this result in terms of the invariants of the
first
two
surfaces.
!i
The
is
Xxf +
A li x12 +
12
^2 2 + A 13 x3 * +
ixi 2
+A lix^+A
2
u xt = 0,
2i
xii =0,
relation.
38
When
one node
lies
IX
LIMITING FORMS.
[CH.
for,
(cf. p.
lie
35), for
when a =
\/yz=
0,
O8yi/-7*0*=O.
Since the polar plane of a node
is
We may suppose
the quartic
latter separ-
ating into
or conjugate
imaginaries.
It
is
Fig. 11.
LIMITING FORMS
59]
99
is
attached to
passing through any other degenerate form for the nodes can be
varied so as to avoid the edges of the fundamental tetrahedra.
;
For
xv,
46
62
24
35
12
14
16
51
32
34
36
13
52
54
56.
72
CHAPTER
X.
60.
The Wave
surface
so
is
by a general
latter is derived
addition, the
Wave
The
are given on
p.
in the form
(x 2
which
is
(a 2 + b 2 ) z2
+a
b2 c2
= 0,
equivalent to
b y
cV
Zk + 2
Zi + Zh
Zh= >
r*
r
b
c r
2
r" = a? 4- y" + z
2
a?a?
a'
where
Ti.
the cone
(ar2
APSIDAL SURFACES
60-61]
101
+ y* + z* = r
+ &_2 y + c-2 ^ = 1.
2
a?
and the
Hence
ellipsoid
-2
Wave
points on the
by taking any
and measuring lengths equal to the
we suppose
that a
finite distances
>b>
c,
all
c to a.
APSIDAL SURFACES.
OM
Let
QT
Fig. 12.
cone having
ROQ
for
method of generating $' by means of spherical curves corresponding to spherical sections of 8, and the relation between the
From this it follows that Q' is an apse of
surfaces is mutual.
the section of S' by Q'OR, and hence the tangent plane at Q' is
perpendicular to Q'OQ.
102
The
may
also
From
near Q'.
is
near
OQ
is
common
touches S' at
QOQ
Q'.
Corresponding to a
in the plane
this it is
[CH.
'
there
is
Q through
a right
x'x
(y'z
+ f + z'" = a? + f + z>,
2
+ y'y + z'z = 0,
0M
is
whose axis
The
is
OM
OM
MQT is
of length
whence
it
a cylinder
Q'.
relation
reciprocation. In
and 0m is
on the reciprocal surface it is evident that the triangle qOm may
be displaced into the position q'Om' by rotation through a right
;
angle in
its plane.
SINGULARITIES OF THE
62.
Wave
Since the
surface
many
is
WAVE SURFACE.
may be deduced in an
elementary manner from those of the ellipsoid. In fig. 12, Q and
It are apses of a central section and the perpendicular at
meets
the wave surface at points Q' and R' such that 0Q'= OQ and
respect to its centre,
QOR
of its properties
As the plane
Q' and R' describe different sheets which
OR, that is, when the central section of the
0Q=
ellipsoid is a circle.
* Lie-Scheffers, Beriihrungstram/ormationen.
which
WAVE SURFACE
SINGULARITIES OF THE
61-62]
therefore a node.
is
103
the ellipsoid
c(a 2
- c )-*,
6 2 )* (a 2
+ a
0,
(b
- c )* (a - c )-*.
Reciprocally, the curve of contact with the ellipsoid of a circumscribing circular cylinder gives on the apsidal surface a curve
at every point of which the tangent plane is the same.
The reciprocal singularities may be obtained by considering
the reciprocal ellipsoid and we infer that there are four real tropes
;
(a2 - 6 2 )* x
It
-c
(b 2
)*
= b (a - c
2
is
The
section
by the plane x =
2
(y
+ z - a ) (b
2
(z 2
+ cV - 6
+ x - ) (&z + a
2
by z =
(x 2
a circle of radius
circles.
c )
x2 - c 2 a2 )
two conies
0,
ellipse of
semi-axes
= 0,
ellipse, intersecting in
is
+ f- c ) {a?x + b
2
-a
&2 )
+ y + z ) (a
2
= 0,
surrounded by an
(x 2
)K
consists of the
The
x2 +
b2 y2
+ c2 z*) =
0.
may be
obtained.
Fig. 13
shows one
plane
z =
is
a2 -c2
2X
a 2 -r
which
is
upon the
b2
b
-c _
C
-r2y ~
2
2
'
ellipse
104
for the
> c).
for
of ellipses
-6
62
2
3
-c
^rr^+^-c
having
[CH.
*2
_,
"''
+ (a2 - b*)i x
(6
- c )* z = b (a - c
2
)*.
Fig. 13.
63.
PARAMETRIC REPRESENTATION.
a2 by
P,
a, 6 2
by
and
by c. Let
diameter
ellipsoid
conjugate
to
the
end
of
the
of
the
be
(, 77, f
QOR (p. 101), then if OQ" = X and OR1 = fi, \ and /* are the parameters of the confocals through P and we have
It is convenient* to replace
b,
),
b) (a c) = a (a X) (a /*),
(b-c)(b-a)^ = b(b-\)(b-fi),
(c a) (c b) = c (c X) (c /*).
a
(a
ff
If P.
at
P to
i'x. 2>2
X (\-fj,)p
2
1
H- it* ~ V) Pi
= abc,
= (a-X)(b-X)(c-X),
= ( - P) Q> - H) (c - /*)
xiii,
238.
X
62-63]
PARAMETRIC REPRESENTATION
and
therefore given
is
105
obtained by measuring a
is
p/c,
prj/b,
by
(a b)(a
c) X* = bcfjr (a X) (a /*),
(b
a) if = ca/x- (b - X) (b - p),
c) (b
-1
(c a)(c b) z* =
(c X.) (c /*),
1
afc//.
X and
fi.
coordinates by
a?
ax 2
Tn order to express
replace
X and p by
f + z = X,
l
+ by + cs = abc/i-1
2
x,
elliptic functions of
a>\>b> /m>c,
so
we take
t
&2
o 6
ac
,,
,,
sn 2 (p, )
=a
abT
*=F^F - ^) =
2
^,
= 6*sn(p, /<;)dn(g, k
y = a* en (p, A) cii (g, A^),
then
aj
a*
dn
),
a> /i>b>\>c,
and
so, in
p', q',
we
functions by
^ = d>
sn2
(?'.^')=
and then
#=
y
c*
dn
(p', A;')
c* en (/,
A')
sn
en
(<?',
A/),
(g', &/),
5 = 6*sn(p',A;')dn(g',^').
* Appell et Laoour, Fonctions Elliptiques, p. 167; Weber, Vierteljahrsschrift der
106
OM* = v
so that v*
Wave
plane at
is
parallel to
is
surface at (x, y,
POR and
abcv^fjr 1
the plane
Now
the tangent
is parallel to
the
(X v) (X /a) = (a X) (b X) (c X),
2
The
z).
QOR
similar triangles
(X
or
TANGENT PLANES.
64.
Let
[CH.
direction cosines of
OQ
X and
/*..
P to the
p^/(a-\),
p^/(c-X),
Pivl(b-\),
OM are
#X^ 7y/6(6-X),
vi\ip^/a(a-\),
Those of
OR
are
ptvl(b-p),
P2?/(a-/*),
OM'
and since
v^p^/c(c-X).
is
p^Kc-fi),
OR
at right angles to
and OM,
direction
its
cosines are
V
PlP* Vi b
(6
- X) (b - /) c (c - X) (c - /i)
aw* (X^
6+
6c (X
The equation
c/i
etC ''
+ 6c)
n)
+ my + nz = 1,
P + m? + n? = v~\
lx
and then
and
bc(fi-X)
ab(/j,
>
X,
and
fi.
'
GQ'.0M'=0A\
TANGENT PLANES
64]
107
Fig. 14.
also
through the
spheres pass through the same two points and the circles in the
other planes of reference. This theorem* gives a method of constructing the tangent plane at any point and the point of contact
of a given plane.
If the equation of the tangent plane
Ix + my + nz =
and
(Iv,
1$ is
mv,
spheres
nv).
of
the
first
the
of
is
three
is
Iv
and
is
(#
2
2/
22
a) = (v a) x,
v a
mv
X a'
v b
nv
X
m, n
b'
v c
Xc
terms of X and
:
in
from which the expressions for I,
fi may be
deduced by means of the formulae already obtained. On substitution for x, y, z in the tangent plane, the tangential equation of
the surface is obtained in the form
-
+
=
vb vc
va -^-r+
0,
(v-^=P + m* + rf).
108
v,
[CH.
we have
plane and
b)(a c) P = (a u) (a v) v~\
(b -c)(b- a) rn? = (b- u) (b - v) tr
(c a) (c b) v? = (c u) (c v) ir
ir = p + m3 4- 2
w = 6cP + cam + aim
(a
2
.
65.
xyzabcXfiuv
m
by
We
respectively.
a-1
-1
-1
ir1
-1
u~*
/*
\_1
abcv^fj,- 1 (X
v)(\
fj,)
= (a X) (b X) (c X),
whence we deduce
v (v
X) (v u) = (v a)
(v
b)(v c),
(0- a)
for all
(0
and
and
-a)(/j,-b){fj.-c)
(u
fj.),
//,
as also do
(j.
b>
and
(i
>c,
a> >b;
CURVATURE
64-66]
109
CURVATURE.
66.
Let L, M,
be the direction cosines of the normal at (x, y,
then a principal radius and direction of curvature are given by
_ dx
P
Now
dy
dz
~dL~ M~dF
= Xxdx = Xaxdx
%xdL
XaxdL
z),
'
%dX
^abcfi~
is
whence by
differentiation
\v~ldv
(a
b)(a- c) L =
2
2dL
du
-^- =
L
ua
whence
Again
(a
Now
meters.
(a
u) (a v),
dv
a'
it is
b)(a
Hence
(a
v).
v*du
dv
form
= abed/judv.
v/i?dXdu
When
'
dX*
expressed in terms of
[X a
\v a
Xb
v
or
X and
Xc
X\
vl
(v
*
The
curves.
first
expression for p
is
dXdv
-a)(v b)
(v
c)
110
and
that
\ may be
symmetrical in
is
\ and u, it follows
The equation
obtained by substituting
is
= v*p.
dX/dv
The
[CH.
for if
is
the
Hence
If
d\2 -
{(X
1;)
/' ()//() + 3
is
f is
cubic*.
c-a
'
showing that the intersections of the normal with the planes of reference
and perpendicular central plane form a range of constant cross ratio.
The
line
element
is
given by
d\ 2
abcdfi 2
Xv
{ufi)'
ASYMPTOTIC LINES.
67.
The
3
fi
dxdl
^,
or
dX
2,lx
\\
further
+ dydm + dzdn = 0,
adu
^
u(/M
du
a)J \u
=^
U = ^"g^""^
(a o)(a c)
coefficients
of
aj =0;
fr-g(-)
(a b)(a c)
fiv
The
dv
aH
are 0,
reduces to
/j,(fi
This equation
is
u) dXdv + v (v X) d/jidu = 0.
unaltered
if X, p, u, v
v~x u~\
,
fir
1
,
are replaced
X~
by
l
,
'
ASYMPTOTIC LINES
66-67]
When
111
the form
d\?
dtf_
(u
To
a)(u b) (u c)~(X a) (X b) (X c)
integrate
(^2/3))
we
if
of the curve
y*=f () = (x-a)(x- b) (x - c)
are collinear, then
+ dxi/y* + dx /y = 0.
d#i/2/i
du/\/f{uj + cZX/V/(X)
expresses that the points (u, V/(w)) and (X, V/(\)) are collinear
line is
= mx + n,
then
(x
(x x ) (x u)(x \);
= (x^ a) (a u) (a X).
and
The
required integral
is
obtained by eliminating
instead of
it
we
m and n;
xx
is
introduce
= (6 c) (#! a),
P = (c-a)(x -b),
y = (a-b)(w -c),
a + /3 + 7 = 0,
a
so that
/(o- tt)(o-X)
This
is
oi
which, again,
69).
(c-it)(c-X )_
the same as
'Jct.lx
(p.
/(b-u)(b-\)
+ VySmy 4- VyriF = 0,
is
The
In terms of
(a
/a
and
v the equation is
- fi) (b - /*) (c - /)
(a
v) (b
v)
1,
(e-v)'
143.
112
[CH.
68.
of which
the
is
is
which
(cf.
the
m,
I,
n,
I',
m',
n', is
x2/X
where
point.
X,
fi,
+ y'/fi + z \v = 0,
2
v are the
v)
x2
is
+ (v + X) f + (X + n) z =
2
(ji
Now
if (x, y, z) is
this
+ v)(v + X) (\ + fi) = 0.
/a,
equation in \,
x2
w2
z2
=
a V+-a^+
X p X y TX
whence
!>
+ fi + v = a + /3 + y a? y2 z
11,
67-68]
painvin's complex
113
The
root
a?
tf
+ tf + z*-
0- y
a?
In order
surface
Thus we
y,
ic
+ y* + 2*-a-/3
is
=1
'
= y + a,
wave
z*
+ y* + sP-y-a
surface
= a + /3.
may be
generated by lines
we have
= (p-\*)(0-p),
-)*.=
a) ( 7
(7
(t* -V) y -/*).
(/3-v)(/3-a)tf
It
is
fi
The
is
/j,
-X) a? + (X + /t) 2* =
Hence it
X = const.
/j,
line
is
the normal
at the origin
is
and whose
radius
PQ
Hence
as a plane
wave
surface
moves
it.
It is a singular plane
it
when
complex curves
touches the
it
foci.
When
it contains one
which is singular. When the plane touches the inner
sheet the major axis is the singular ray and cuts the outer sheet
minor
axis,
at the
foci.
* Boklen, Zeitschriftfur
H.
Math, xxvn,
160.
114
[CH.
Prove that the plane joining PQ to the pole of n touches a confocal whose
parameter is minus that of the confocal touched by n.
we
by
cfix\=al+il'\
c^x6 =cn+m'}
ioFx^al il')
b^xi = bm im'\
The
The
coefficients
we take
(p.
58)
by
2
P x*={kt - X) (A, - h ) (k, - ,^lf (ka )
where
^ and
of contact
ji
this gives
p {al+il')
= {ia - X) V(& 2 - c2
(ia
(ia -
n 2 ),
etc.
From
point coordinates,
it is
and
(I,
c 2 ) (c 2
where
n, V, m', n')
with
them as
- a2)
)(
(a 2 - 6 2 )
)(
=0)
=xi +y2 + z\
v =a?x
rn,
easy to deduce
X, p,
fi
+b2y 2 + c*z2
p, 2
CHAPTER
XI.
69.
Rummer's
The equation
We
real
fundamental tetrahedron 12 34
.
reference.
Then by taking
coordinates
it
X 1 =p u
-p
i3
II.
III.
IV.
* Eohn, Math. Ann. xvm, 99, may be consulted for a more detailed account,
t See an exhaustive paper by Weiler, Math. Ann. vn, 145.
82
116
so that in all cases
%x* = ^PuP-n =
0.
When
be reduced to the
first
by the substitutions
1.
[CH. XI
the multiples of
real, all
69]
be conjugate imagiDaries.
have
is
ie
e ;
= pr^e^9
r,e^>.
where p
=r
Put ks
117
= 1,
(s
'
2, 3, 4, 5, G)
n = ri = ri = r4 = r =
2
2
5
= p,
r<?
showing that p
is real and positive, and that the points representing kg in the plane of the complex variable re ie lie on a circle.
In case II a\ and x2 are conjugate and we must have
/or e~"
r2 e~* e = pr^e^ 9
rH e-^> = prr <r*>
r 1 e~i
ifl
>
',
(s
3, 4, 5,
6)
r^2 = r = r4 = r = r = p
whence
0i =
and
showing that the points
on which ks kt k5 ke lie.
n =
2
=r
" 2
03
= 6*,
r4
= r r = p,
5
Ss=06,
negative;
if
is
positive
nn = r r = r r = p,
4
and the
six points
2,
=6
#5
if
is
negative
0!
nr = rs r = r r = -p,
= 4 = 0a =
is
ir
plexes.
possible.
is
118
70.
The
SIX KEAL
[CH. XI
FUNDAMENTAL COMPLEXES.
These
guished by the order of magnitude of the six coefficients ks
quantities are the parameters of the six nodes on any singular
.
conic,
cyclical order
r>
I a.
We
The
given
pencil of tangents at
(p.
any point
(fa, fa) of
the surface
is
58) by
where
/ O) = O - ^i)
and
c is
(/"
a factor of proportionality.
/j,
SIX REAL
70]
forty-eight in
all
FUNDAMENTAL COMPLEXES
119
same para-
meters
by
fa
= fa.
of fa
give the points where the curve crosses the two principal asym-
//.
nodes on two conies with the three points whose parameters are
ks fa ks+l indicated on each of them. As /a varies from ks to Jcs+1
the asymptotic curve
sweeps out the whole segment twice and
the two branches of the curve coincide in the case of the principal
,
/x.
The
asymptotic curves.
where both
The
inflexional
elliptic region of
fa
fi
and
ft
being
real,
/j,
fj.
+ ifi'
'
is
fa =
given by
fi
ifi,
the surface
= 0,
00
<
/i
<+
00
fi
line passes
Hence
three arcs of conies joining three nodes and forming the boundary
of an elliptic triangle at the same time the points of contact of
the lines obtained from (x) by changing the signs of the coordinates
;
120
one
[CH. XI
interchanged either by one of the colby one of 23, 45, 61. Hence the nodes
and
six hyperbolic
hyperbolic segments
group-set containing
is
its
clearly
visible
attached at
Two
b.
each
is
actually
hyperbolic sheets.
As
This
A/2
fC$
n/4
>
is
iCq
rCt),
given by
left sides
- X)2
+,+,+,+,,, when
are
the
The
must
by
so that
it is
k u k2 ks kt
,
impossible to have fa
fa,
ke k6 fa
,
the n odes
and tropes
are imaginary.
Along any one curve fa varies from k5 through ink 1} giving the intersections with another family of asymptotic curves covering the same sheet.
By continuously varying
of one sheet.
finity to
k
70-71]
k a
SIX
121
by the
directrices (14), (16), (54), (56) and the corresponding collineations transform each sheet into itself. The re-
maining
points
Any
itself.
(23) cuts each sheet in two points forming a harmonic range with
the points where the ray meets the directrices.
Hence these direc-
trices
separated from
is
its
cut
it.
Imaginary
c.
surface.
>
fCi
fC<%
71.
C.
The equation
etc. referred to
a?
+ yi + z + t + A (xH + y
l
(a,l3,y,B)
+ B (y t + z*x )
+ C {zU* + x y-) +Dxyzt =
+ 7 ) = Xk -h)(k -k )
2
z2)
Where
~A =
B
(a
+8 )
2 2
(ff
2 8
)
2
S2
/,
\k -k )(k -k
i
- (7 +
-7
2
3 2
'
5)
(h - k) (k - k)
*(h-K){K-hy
(k s
e)
(&i
- k ) (k - k ) (k - k )
2
<y
/3
0,
and
,
^-/3V
/3
_ h
-(/3
4)
122
[CH. XI
JD
a, /3,
equation I
a.
If
x*
is
a, ft, 7,
C.
For to
every real point (x, y, z, t) on 16 correspond a pair of imaginary
points (ix, iy, z, t) and ( ix, iy, z, t) on la lying on a real ray of
the congruence (56). Now certain rays of this congruence cut I a
in four real points (for example lines joining two nodes), and the
passage to imaginary intersections can take place only through
positions in which two coincide, that is, intersections of the surface
with the directrices (56). Now the line x = 0, y = cuts I a where
is
typical of both I b
and
zi
and
all
C>
2,
+ ti +GzH^ = Q
if
G<
2 and
all
distinguish I b and I
are imaginary
and
sufficient
if
to
c.
72.
II
a.
Consider
first
the case
when
k3
*
(1903), p. 92, of
sixteen, eight,
and four
Kummer
surfaces with
71-72]
Any
tangent line
is
given by
and
fa) (ks
- fit) (k - X)
for
while, for s
be
123
3, 4, 5, 6,
x82f'(ks )
positive.
is
number
is
asymptotic curves fi = ks k it ks or k6
Since for a real line xx and x2 are conjugate imaginaries, only
sixteen out of a group of thirty-two lines are real. Hence there
.
are only eight real nodes and eight real tropes, namely
0, 12, 34,
At
/t
fj,
ifi,
takes
all real
As
values.
and remains
returns to
at a node, alternately
fi
(fi, fi,
have
all
cc2
after
fi
has
changed their
its original
(fi, /i)
signs.
The
/jl
pencil of tangents
value twice and then the point of contact has described four arcs
and passed through four nodes. Hence the elliptic segments have
four sides and four corners each.
The collineations which interchange the nodes at the ends of a
Hence the nodes 0, 34,
side are either 34 and 56 or 45 and 63.
12, 56 taken in this order are the corners of two elliptic segments,
being joined consecutively by hyperbolic segments so also are the
;
nodes of the one other real tetrad of the same group-set, namely
Similarly the tetrads of nodes 0, 45, 12, 63 and
35, 54, 46, 36.
35, 43, 46, 65 are connected
by
elliptic
segments.
Each tetrad
two nodes
generators through
trices
with respect to the six pairs of direcnamed after them, only (12) of
one system and (45) (56) (64) of the other system are real, since
the correlations associated with a^ = 0, and x2 = 0, separately, transform a real line into an imaginary one. Hence the nodes in this
124
[CH. XI
degenerate case
and
this gives
case.
IIP
Hi
mm
Fig. 16.
As
is
a that each
elliptic
segment
contains a point where the two inflexional tangents have fourpoint contact, but are imaginary.
These points
lie
by
fours
on
b.
One hyperbolic
sheet.
(kg
/j,,)
(ks
fi
respectively.
2)
(ji^,
/j, 2 ,
are +, +,
The
it is
whence
""4
and
fa
fi 2
> A;s
real
only
if
the signs of
cyclical order of
impossible for
imaginaries,
jt6
are
A.)
Ml 3
and
>
>A;4
/^
The tangents
= 3,
4, 5,
""6
to
""
fat
be equal, or to be conjugate
The
surface
is real,
and everywhere hyperbolic there are four real principal asymptotic curves, and the discussion is the same as in I b except that
there is only one real sheet
for of thirty- two associated points
only sixteen are real, corresponding to the different signs of ws xit
#, #6 and these can be reached by continuous variation of the
elliptic coordinates fi lt n 2 and accordingly lie on the same sheet.
;
Equations of II a and II
b.
(*ft . s
>
h)>
(*7> S,
a *),
>
72-73]
125
If
a, /3, 7,
are real.
The
substitution of
(ioc,
II a has the
effect of
evident that
all
By
It is
b.
reasoning similar to
two real
points.
73.
III.
is
In
this case
There
not
arise.
The
is
that (k x
y^) (k fa)
x
and (k2 /Aj) (k2 /i2 ) must be of the same sign hence ^ and fa
may be imaginary, but if real do not occur alternately with kx and k2
and so may be equal. Of thirty-two associated points only eight
;
nodes and four real tropes. Two singular conies cut in two nodes
whose parameters are kx and k2 terminating two hyperbolic seg,
ments, and the other two conies cut in the other two nodes, so that
there are four hyperbolic segments altogether.
elliptic
nodes.
The complete
The
surface
is
represented in
Fig. 17.
to
126
The equation
(ix,
$y, ih,
a,
j8,
of III
for (x, y, z,
t)
7, 8),
are
Four
obtained from
and
la by
substituting
S) for (, $, 7,
/S, 7,
(a,
8),
XI
8).
a, /3,
The
7, 8),
74.
IVa.
is
t)
[CH.
are imaginary.
the case
when the
elliptic coordinates
If one
imaginary.
()
of
thirty-two
are,
/j,
IV b.
Four
r^
and if one of thirty- two associated lines is real, then eight are.
There are no real pencils of tangents to the surface but the line
-1
(X, X, n, /a) is real provided X and /a
are conjugate imaginaries,
and this line either passes through a node or lies in a trope.
Hence there are four real nodes and four real tropes, but the
surface
As
is
imaginary.
CHAPTER
XII.
75.
By
is to be understood a method
numbers and equations, in which the
'
principles
figurations
in the
A
is
It
may
linear equations.
Two
(\x
4-
fiy) called
the
oo
equations
tv
we obtain a
= \x + fxy + vz + pt
s
(s
= 1,
a space consists of
linear equation.
all
The
2, 3, 4,
ws
5)
so that
linear manifolds
The
number
128
[CH. XII
75-76]
LINEAR MANIFOLDS
129
76.
We
now proceed
to
six
dimensions.
fifteen lines
3,
...,
They determine
5, 6.
and twenty planes 123 ...,
4,
and fifteen spaces 1234.... The line 12 cuts the opposite space
These fifteen diagonal points lie by
3456 in a diagonal point P12
.
12
Pu P
56
lie
3456, 1256, 1234, and any three spaces have a line in common.
transversal lines and may be denoted by
Thus the line (12.34.56) contains the points
Again the three lines (12.34.56), (12.35.46),
-P 12 Pa, Pm(12 45 36) meet in the point P12 and are contained in the same
These
are
(12.34.56),
called
etc.
space 3456.
of three,
Pys
* 36
-*26
"l5
Pss
Pie
14
34
24
H.
130
the
cardinal
[CH. XII
space.
transversal lines
fifteen
in them.
This
the configuration
is
we
shall
surface
this
the section by an
is
when
Further,
in
the section
is
by
a tangent space, one more node appears and we have the general
Kummer
surface.
ANALYTICAL METHODS.
77.
represented by five
is
six,
xlt
as2 ,
#!
The equation
which
is
xs + UiXt + u s x5
+u
xe =
added to the
is
fix their
"i
The
+ xt + x5 + xe = 0.
is
+ v^x + u
unaltered
To
ocs
of any space
UiXx
coefficients.
+# +
coordinates
+ + w + + + u = 0.
2
may be
and x2 = xs = xt = x6 = x6
and similarly for the other points.
Hence the space 3456 has coordinates (1,-1, 0, 0, 0, 0) and
equation x-, = x2
from this it may easily be proved that the
point
are 1^
in space coordinates
in point coordinates,
diagonal point
]2
is
x1 = xi
m1 + m2 = 0;
,-1-^ =
m3 + m4 = m5 -|-m6 = 0;
= Ui = m3 = Ui = - = w,
= xi
xs = x4 xB = x6
,
is
is
x3 = xi = xi = xe
#!
+ # + x = Xi + x + x = 0.
2
ANALYTICAL METHODS
76-78]
An
131
arbitrary space
+a
a^xx
x2 + as xs
+a
xt + as xs + a6 tee =
(a x
a 2 ) xl
+ (a + a ) x + (a + a ) x =
3
+ M _ M + 4 _ + U
+a a +a a +a
Ml
(h
'
THE 16 6 CONFIGURATION.
78.
lines
Thus, corresponding
(p. 10).
of transversal lines
12.36.45
13.42.56
14.53.62
15.64.23
16.25.34.
We
The
first
= the
2as #s =
three points
five
points of
are coplanar.
determined by the
is
U1 + U2
a1 + a
Now
1l
_U +U
~ +a
S
U 2 + Uj
'
a2 + a4
_U +m
s
ae
+ as
+a )
2
(a 2
+a)
4
(at
ui + u1
at + a
x
_ u +u
a + az
'
= 0, 2as = 0, we have
+ a ) + (a + a ) (a + a ) (a + a ) =
3
in consequence of 2a
(!
'
+ U 2)
(w 2
leading to 2,u s3 =
Q,
( 4
+ Mi) + (s + Me)
(e
+ u)
+U)=
3
this result
two.
(b
shows that
three points passes through the other
is satisfied,
92
the
132
[CH. XII
coplanar points
on
a conic.
One
"ZagXs
= 0,
= 0, the
"2usXs
is
replaced by us + Xas
six
X being arbitrary.
new planes meet in
us are
SMs ga =
>
is
It follows that
Kummer
the
fifteen,
configuration of
sixteen
points
and
sixteen planes *.
79.
The name
when the
variety
is
relation is linear
is
f(x,
y, z, t)
= 0,
y,
z', t')
of
it,
first
- x) Bf/dx + (y- y') dfldy' +(z- z') dfjdz + (t- t') df/dt' = 0,
1
which, by analogy,
{as, y', z',
Change
the
and
the
coordinates
i
in a series of
variety
called
is
t').
considered,
(x',
is
is
= t+f {x,y,z,t)+f
3
+...
space
is
xdfjdx'
+ ydfjdy' + zdfjdz +t = 0,
small quantities of the second order being omitted. This cuts the
space t =
in the polar plane of the point (x, y', z') with respect
* Similar analytical treatment can be applied to the theories of lines on a cubic
and Pascal's figure. See Richmond, Camb*
78-79]
to the cone
= 0.
Thus
133
(#, y, z, 0)
all
the oo
in a tangent space
be called
a double
A finite
point.
number
in
the case of
two of the
a tangent plane,
and
tangent spaces
coincide.
The
The degree
= z/z' = t/t'
cuts the
by the equation in k
of this equation
is
let')
= 0.
One
root
is
zero,
is
and
= 0,
0)
z,
= 0,
/,(*, y,
z,
these
All
/,(*, y,
equation of a space
0)
= 0.
If
the
general
is
Ix
+ my + nz +p + qt =
a single equation
< {I,
m,
n,
p,q)=0
may be
coordinates
space
(I',
m,
n,
d^/dl'
is
p', q')
is
Thus
if
= 0,
the origin
df/dy
is
= 0,
d//dz
= 0,
df/dt=0.
form
=/,(*, y,z,t)
+/,+
....
The
section
oo
134
[CH. XII
is
y, z, t)
+ ....
The
of a plane to be
Ix
+ my + nz + t+p =
=$
The
m,
2 (I,
p)
(f><)
+ ....
first
is
fa
now
(I,
that
therefore infer
quadratic,
m,
n,
p)
all
= 0,
We may
surface.
We
n,
repeated
in the
80.
+ u2 + u3 + ui +u + u =
6
ls
2g =
is
= 0.
form
+ w ) {u + u ){u + u ) + (m + u ) (m + M ) (Ui + u ) =
3
This equation
is satisfied
by
+ = 0, u + u = 0,
2
is
to
say,
79-80]
shown
by
to lie
sixes
135
spaces.
It follows
are
proportional
ur2 us2
to
etc.
U1 = U 2 = U3 =
!/,
=M =U
5
Hence each
or
is
cardinal space
all
we get a plane touching the fifteennodal surface along a conic containing six nodes. We recognise
in the arbitrary space section
that the section must be a quartic surface with fifteen nodes and
ten tropes.
By
new node
the
at
point
The
of contact.
section
is
we get
now a
2w3 = 0, 1u v = 0,
The
a tangent space.
is
= 0,
conditions are
2.uv 2
2,1?
a great
{(#!
many
X.2 )
ways.
One
- (x + X4 - 5 3
is
}*
in which no use
xs
There are
is
made
It
may be
written
...
x+y
and
is
+ z + x + y + z' =
(1).
136
[CH. XII
is
+ (#/<)*
where
so,'
(2),
By
starting with
reciprocal processes
we
shall
On
quartic variety,
of the
'
cone
'
is
is
CHAPTER
XIII.
GEOMETRY ON A SURFACE.
field of
geometry,
it.
we
Here a
investi-
An
surface
important step
when the
made
is
in the
theory of an algebraic
in the
therefore necessary.
But there
is
common
may
to several surfaces.
is
defined by
138
The
being algebraic.
suffice, for
a given curve
[CH. XIII
is
that in
may
not be
may
is
means of a
x=y=z=0
point
arbitrary values of x:y:z determine one value of the remaining coordinate, which must therefore be given by an equation
of the form
*X ( x y> z ) = ylr 0. V> z )
Then
>
The complete
cones
f= 0,
^=
0,
intersection consists
may be common
to the
$ = 0.
The theory
is
simplified
When
surface alone.
it
may happen
the
Kummer
surface
it
for
is
is
not true.
82.
<J>
be written as
in
10
in the form
<^2 +
so that the point
2(f) 3
< 4
0,
x=y=z=0
cone
is
$2 = 0.
Any
f{x, y,z)
and a monoid
tx (*,
y, %)
= $ (*,
y, z)
7.
81-82]
139
On
find
V fa fa fa
2
so that, in the
which
+ fa =
we
is
a curve
upon
fafa
/=
plane curves
and
F G
2
polynomials in
x, y,
factors
all
necessary
a, factor of an expression
and G are homogeneous
where
F
(fa
fafa),
z; but this does not define the form of/ with
Now
sufficient precision.
factors
that/ must be
sufficient condition is
of the form
is
fa
fa<j>
is
x1 x2 x3 xi x6 xe
let
be the product of some of these
and T the product of the rest, then /must be a factor of
;
F X H Y.
fafa = is a
2
fa = y xz=0, we
adopt a
Then
and
fa
Let
U be the
and substitute
of coordinates
1,
\ (u + v),
uv,
(p. 18),
= (u- k ) (v - k )
fafa = IIa;s =U (u ks) (v k ) =
xs
hexagram circumscribing
new system
Il say.
U becomes Fwhen
P*U-Q?V =
uv
is
integral
and symmetric
in u,
v,
and
this value,
It is a
being symmetric in
(P2 U+ Q>V)/2PQ,
u, v, is rational in x, y, z.
wx
directly,
is rational is ex1 xi x3 xi xB x6
pressible exactly in one of these two forms, that is, without
p. 56.
140
[CH. XIII
extraneous
factors,
Expressing
this differently,
F X-G Y=0,
3
or
is
uv
of these
At present we assume
two forms.
this
theorem in order
It
P>JU+QJV=0\
+ 4> = ^UV)
t
<f>.2
P</U+Q</V=0)
and
fct
^-VzM"
xs
#123
This
is
= (u k ) (v - ks ) = UgVn
(s
= 1,
2, 3, 4, 5, 6),
F*JX + G*/Y=Q.
identities
an example.
3 i>
e,
Let
then
(u
(u
v)
(fa
^3) fa V#2aVi23
234
in
82-83]
141
83.
every algebraic
the form
x Vpi +
x
\ \/p
...
= o,
= 2\ p
s
is
From the nature of the coordinates used, must be homogeneous in the linear forms. The number of factors in any product p s is called the order of , or of the equation.
After substitution for the radicals in terms of u and v every
term of the equation takes the form P*JU+ Q\/V, [/"and V being
the same for
all
the terms.
formerly written
u and
F*JX + GV Y =
',
that
is,
it
is
geometrical distinction.
It
is,
order
is
odd.
Thirdly,
we have to
u and
tution in terms of
of radicals
*JU and
tJV.
v every
Now
It
start
142
[CH. XIII
V#m
xr xs x
all
it
product
11
22
33
= 44 =
55
= 66 = 123456 = identity.
is
The symbol
is
one which
= 0.
Of the
sixteen characteristics
there
is
is
()
=0.
a/3
When
equation
order,
the
order,
is said to
parity,
and
characteristic,
linearly independent
with undetermined
84.
coefficients.
= \ Vpi + X, \fpi +
= 0.
83-84]
Let
and '
family.
if
u and
<1>.
<1>
their
2 is
the
rational
is
Vpip
vx x2 x3 xi x5 xg which
as
surface
but
expressions in terms of
143
can be rationalised by
0,
<E>
in the
= 0,
curves
'
only.
Any
two curves of the same family form the complete intera surface of order n.
section with
By making
Every curve
is the
curve of contact
Every curve
is
is
n.
and
surface
is
= 0,
2
of order 4m
after rational-
n and
of order
is
therefore the
hence
curve
where
The 2mn
it
meets the
intersections of
curve and surface all lie on <E> and are contacts except at nodes.
Hence any two curves of orders 2m and 2n intersect at an even
number, 2p, of nodes and at mn p other points.
Let S1 = 0, 82 = be the surfaces of order n, tangent along two
of the same family, and let 8 =
be the
curves i = 0, 2 =
surface of the same order containing both curves then, in virtue
;
of <&
= 0,
/ = #!, 2 =
a
whence
S& = S
in virtue of
2,
<>
0,
,,
= ,
identity
SAs/Sf + G*.
O=
4,
The
first
two theorems of
S2
along
this identity
many
touching Sj and
From
the following
is
144
two
the same as
V x x6
teristic
on
when
is
number
is
by wx^-^ or by
which
has zero characV^s
[CH. XIII
of V#i
Va^a^a^
either rational or is
which
is
rational
<I>.
From
this
theorem
is
Every curve and not more than four singular conies together
form a complete intersection.
For, if =
is the equation of any curve, it can be rationalised
by finding a product p of the same parity and characteristic whose
order has the same parity as that of
then *Jp =
is an
equation of even order, even, and of zero characteristic and therefore, by the preceding theorem, is rational on <3> and represents an
and the singular
algebraic surface cutting <J> in the curve =
We shall show in the next section,
conies in the tropes p = 0.
by examining all the different cases that arise, that p need not
contain more than four factors.
Thus when a family of curves is given by the equation
;
= X
the
first
^=
g
= 0,
is
to find a product
=P
<E>
>Jpps
of least
a rational integral
is
cut out
X
having
P + X P + ...=0
1
The curves
less
the tropes
s,
after
p = 0.
is
one
making
* = 0.
use of
The
surface
p=
2\ P =
8
passes through
all
planes
0,
84-85]
145
85.
shall
exceed
four.
When
three distinct kinds and those of odd order are of only two distinct
kinds.
""
for
is
example
x-^x^x^x^,
Since an odd
conies.
number
\n +
of the planes
p=
pass through
each node, the curves of the family pass through all the nodes.
If the characteristic is not zero it determines two associated
each consists of four pairs, giving four even and four odd
products of order two and the same characteristic. /j may be
rationalised by means of any one of the four products of the same
octads
two
conies.
of the system.
Secondly
let
cording as
In this case
may be
characteristics
all
sixteen
afi
t/x a p
is
not.
rational on
If they
and
Then
since
pxap
is
10
146
node
is
common
[CH. XIII
by pairs
in
The curves
We
and the
which the curves of each family pass.
a typical Rosenhain tetrad and suppose
number
We
of nodes through
that x
is
even and xu
is
odd.
85-86]
numbers
147
conies
148
properly choosing
equation
Sin-4
the
may be reduced
number
[CH. XIII
to
+ ty + 1 - tn - |(i 2 + s) =
+ i (n - s).
2
We notice
all
intersections
number
of variable
is
The
surfaces
again in
many ways
may be
defined in various
It is possible in
ways.
to
extent arbitrary, yet they cut the curve in a definite linear series
of groups of points.
If
arbitrarily chosen.
is
This series
we put
//.
= 4,
points of which
p 1 may
is
= \ (n + 1)
be
In
group
2p-2 = n
or
-s,
p-l + t(n-8),
is
CHAPTER
XIV.
QUARTIC CURVES.
Taking
first
if
even,
We
have then the family of plane sections, of which only four are
In attempting to form an equation of zero
characteristic we see from the multiplication table of the group
that the two factors of each term must be equal, and the product
is simply one of the sixteen linear forms.
This shows that the
equation cannot be odd, and therefore no quartic passes through
linearly independent.
all
sixteen nodes.
Two
common
Hence
the other in four points, and the curves have four variable inter-
150
sections.
The two
[CH.
XIV
which
is
S + 2\S + X'Si = 0.
1
This
p. 1493.]
Two
curves of
families
with
different
characteristics
pass
yx13 x23
Va;14 a;24,
^^u^,
v&ioze
from the associated octad. But since the equation of the surface
expressible as a linear relation between any three terms from
each set of four, only two out of each set are linearly inde-
is
pendent.
87-88]
151
88.
tetrad, as in the
diagram
and
79)
z, t
(cf.
p.
\/xt
+ 'X"Jyz = 0,
'Jyi
+ pt/zx^O,
t,
We
<t>
know
may be
that Vxyzt
is
rational on
xyzt
and
<j)
<I>,
is
2
<$>
conies.
The equations of
whence
+ v\/xy = 0,
+ v*xy 0.
vB - fiG = (yy - fiz) (t - /mvx)
\C-vA = (\z - vx) (t - v\y)
fi,A-\B = (/MX- \y) (t - \fiz),
C=zt+2v<t>
Then
5 = 0, 0=0
yyt
In
this
is
way we get
contact of A, B, O,
^ =0)
fivx = 0]
'
aoyzt
= 0.
152
[CH.
XIV
The product
(y/xt
is
rational on
<t>,
+ X Vyz) ( Vyt +
whence we
/j,
*Jzx) (*Jzt
+ v V#y)
S=
can
where
a quadric.
is
lying on 8;
ABC = S*+G,
A and G
Evidently
in fact
/a,
v,
an even tetrahedron of
tropes, cut
of \,
A, B,
in the
same
way, so that
then
88-89]
153
| 89.
z'
is
we may take
the equations
to be
+ X >Jy'z' = 0,
\/zx + fi "Jz'x = 0,
*Jxy + v vx'y' = 0.
*Jyz
Further
<> = 0,
\/%x', "Jyy',
and we may
'
face to be
Vaac'
*Jyy' 4- 'Jzz'
0.
- \y') + \xx' = 0,
B = (z - fix') (x - /iz') + fiyy' = 0,
G = (x vy) (y vx') + vzz = 0.
1
= {y- Xz )
(z
Hence
and
G contain
the line
vy' fiz' = 0)
fiy + vz fivx' = 0j
x
generators
is
lie
a cubic
curve.
common
in the plane
\x + fiy +
meets
vz
common
= fivx' +
to
vKy'
+ Xfiz
where
it
4>.
154
As
XIV
is
[CH.
rational on
+ \ >Jy'z') (\/zx +
<E>,
/j,
*Jz'x')
{*Jxy
+ v \/x'y')
in
ABC=S*+G<S>.
Since
and
<J>
S,
<E>), it
touches
S and
<t>
of
at
therefore lies
it.
on
on
(B,
is
entirely on
lie
<1>),
lie
plane repeated.
six bitangents
same two
points.
from different
make up
90.
SEXTICS
THROUGH
SIX NODES.
two
arbitrary coefficients.
five
linearly
independent curves.
The sextic meets one trope at six nodes and every one of
the fifteen others at two nodes and two points of contact.
The
inscribed cubic surface cuts the one trope in a plane cubic passing
through the six nodes, and every other trope in a conic and the
line joining the
Two
conic
cuts
them
in the
Among
when one
conic is
89-90]
THROUGH
SEXTICS
NODES
SIX
155
four
all
lie
up
the inscribed cubic surface breaks up into the three tropes con-
By
any
sextic of the family in six coplanar points, other than the six
common
nodes, and the three lines in which this plane cuts the
three tropes
lie
been shown
to
also
= 0, S2 = 0,
S=
we have
is
the identity
repeated.
We
is
yH, zH,
teristic
same
and
The equation
parity.
order, charac-
is
so that
<j>
is
= xyzt -
2
<f>
=
Then
is
repre-
cubic to be
$,
= xyz +
2 (ax
+ by + cz + dt)(j> + (ax +
ax
Another sextic
by
(a'x
+ cz + dt)
is
+ cz + dt =
of the family
*J~xyz
by
b'y
0.
is
= 0.
156
and
consisting of a conic
XIV
on the curve
a'x + b - b'y + c -
(a
Write
lie
[CH.
c'z
+ d d't) \/t =
a plane section of
3>.
for abbreviation
= ax + by + cz + dt
2 = a'x + b'y + c'z + d't,
2u
+ u*t
$! = xyz
=
S xyz + 2w $ + ut
S = xyz + (u + u )(j> + uiu 2
S S (u uj
+ w t),
Wj
then
-t-
1 (j>
whence
1,
and
(</>
(<f>
showing that the three cubics have a common curve in the plane
i = 2
Two sextics from different families cut in two nodes and eight
other points. We may take their equations to be
.
s/x
= 0,
\Jy
= 0.'
+u
\fxzt + u
*Jyzt
They both
lie
(\/yzt
<j>
lie
intersection
<f>
+ ux = 0,
+ uy = 0,
y=
y) =
0.
The
conic in
cuts
<3>
in the
<f>
four
common
and u
We
with $.
are thus led to consider the four sextics
t
+ u\lx = 0,
/3 "Jztss + n\jy = 0,
a vyzt
+ u*Jz = 0,
8 ^xyz + uJt = 0,
<y"Jtxy
u=
= z,
90-91]
SEXTICS
157
cut out by four quadrics, one through each conic of a Gbpel tetrad,
and the same plane section of the quadric containing the four
conies.
They pass by threes through the pairs of points u = 0,
= 0, etc., and the remaining twenty-four intersections lie
x = 0,
by fours on the six concurrent planes
<j)
xja
SEXTICS
91.
The
pendent
is
sixteen, leaving
by cubic
The number
of
surfaces.
The
on the trope
which completes the Rosenhain tetrad, and therefore has three
contacts with this plane and one contact with every other trope.
The same is true of the inscribed cubic which therefore contains
three lines lying in one trope.
Two
other points.
we have the
all
touch
<!>
at the
same four
points,
and
identity
SA -S> = (?*,
where
is
a quadric.
common
Now
points of the three cubics are singular points on G<& = 0.
the four contacts count for sixteen intersections and ten more are
at ten nodes of
be a node on G.
<f>
Hence G is a
<J>
which must
S.2 along
with
lie
Since the inscribed cubic has four nodes the lines joining them
it and therefore touch <E> where they cut the sextic
entirely on
of contact.
Hence
inscribed in
<&,
lines
the
<.
the corners
The
six
six
of a tetrahedron
edges are torsal
and lying
in a tritangent plane.
The cubic
possesses only
158
[CH.
XIV
one tritangent plane beside those containing the nodes, and in the
present case this has been shown to be one of the tropes of <t>*.
In order to construct the nodes of an inscribed cubic surface
we have only to take any three points on one conic of 3> and draw
the six tangent planes through pairs of them. These planes meet
92.
terms.
or
u^fx~t
b'y
+ v*Jyz=0.
on
u=
=v
meets
4> lie
The
<I>
for
tritangent planes.
The planes
xyzt
the equation of
<I>
may be
=$
xyzt
is
2uvc{>
for
+ v*yz =
a double line
there
is
a pencil of
quartics
xyzt <f> 2
*
+ \ (u'xt +
2uvcj)
+ v*yz) =
These theorems are easily proved by taking the tetrahedron of nodes for
+ Humbert, Liouville, ser. 4, ix, 103.
reference.
91-93]
all
touching
<3>
By
159
the form
(yz
+ Xm ) (xz + Xu ) 2
we
for
X=
oo
nodal line w
The
- Xw) =
!
(</>
= =
v.
S& =S*+ G,
for each of the twelve common points of contact of Si, S2 <& is
counted four times among the points common to S, G, <J>, leaving
,
S2
and
of
S + GQ> =
2
to be divided between
93.
is
order four and zero characteristic, and the curves are cut out by
The surfaces
must be made to pass through the four corners of the tetrad
and six more points on each conic, making twenty-eight conquartic surfaces through an odd tetrad of conies.
relation
(I>
= 0.
linearly
six
inde-
We
contact of inscribed
Two
Kummer
The
identity
and <S2 have also sixteen nodes each. Hence the quartics touching
a given Kummer surface along an octavic curve passing through
Of these inscribed
the nodes, are also Kummer surfaces.
quartic surfaces ten have double lines and are Pliicker surfaces
all
(p. 68).
Hence the
CHAPTER
XV.
WEDDLE'S SURFACE.
BIEATIONAL TRANSFORMATION OF SURFACES.
94.
Let P, Q, R,
x, y, z,
t,
of the
in
other,
and
T) of the
n variable points common
is then
unique and therefore rational both ways, and the equations constitute a rational transformation between two spaces *.
If (x, y,
z, t)
describes a surface
fix,
y, z, t)
= 0,
F(X, Y,Z,T) =
into
which/
is
transformed.
The plane
sections
aX + bY+cZ+dT=0
correspond to the linear system of curves cut out on
by the
family of surfaces
aP + bQ + cR + dS =
0.
94]
/=
is
rationally in terms of
161
X, Y,
If a simple point of
is
For
Z, T.
this reason
called birational.
is
we have
approximately
TjSm
More
is
mi-ic poly-
transformed into a
An
2m-ic curve.
Pm = 0,
Qm =
0, i?TO
'
curve
on/
If only oo
X Y
:
X=Y=Z=0.
is
satisfying
aX + bY+cZ=0,
corresponding to points (x, y, z) in the neighbourhood of the
curve, and this is equal to the number of points in which
with
the curve cuts the residual intersection of /=
aP + bQ + cR = 0.
h.
11
weddle's surface
162
XV
[ch.
X =T
P = P + P,+
1
...,
Q = Q + Qs + ...,
R = Ri + i2 + ...,
S = $<> + $! + $ +
1
since (0,
is
0,
0) is not
.,
From
this it
linearly transformed
is
by the equations
X/P = Y/Q^Z/R^l/Sc
x
Next,
let (0, 0, 0)
S = 0.
We
and by means
i/s1
/ the four
denominators can be expressed as linear functions of x and y.
Hence to the pencil of tangent lines to / correspond the points
of a straight line on F, that is, a rational curve of order 1, as
was shown
before.
95.
Among
most useful are those in which the order of the new surface is
as low as possible, and also the order of the surfaces employed
in the transformation.
We require a linear system of curves
having as many intersections as possible at base points so we
shall consider only those systems which are contained in the
:
94-95]
163
number
of variable
intersections
points *
The base
may be
2\2
2X
1,
2fi,
The number
system
is
of orders
number
the
fi,
of coincident
/jl,
that
2,
it
out having a
/x-ple point.
intersections of
2fi?,
and
N = tf -
- 2 (2X2 +
2N = 2n* - 2
A
is
(2X
l)2
birational transformation of
by means
s (p.
148)
2\)
- 2
2fi?
- %v\
16 -2s
2s
so that
16-2S
2s
is
- 2
(2/*)
- 22^
Kummer's
2
.
surface
is
effected
if
the dimension
is
and
sufficiently great.
* Cf.
Humbert,
cendental methods.
who obtains the same results by transSee also Hutchinson, Amer. Bull, vn, 211.
112
weddle's surface
164
fc-ple
and
Take
passes.
point the
number
XV
[ch.
of conditions for a (k
this
node
If
z.
has
l)-ple point
is
K* + l)(*+2)-l,
since the term zh
is
absent.
Hence when
is
arbitrary, subject
is
.F=0 may be
cients in
F+G^> = 0,
replaced by
<X
i(x-i)x(x +
We
of these conditions.
the coefficients in
In the second
<
fi
1) conditions
among
terms of order
3>,
and
< p1
in
F + G<&
terms of
of
be used, leaving
J/t(/.+ l)0*
conditions
i)
X(X +
case, the
may
alone.
is
l).
among the
+ 2)-i0*-2)0t-l)/* = A*
coefficients in
alone.
is
therefore
Now
(p.
D = l + Hn -s)-2\(\ + l)2
Hence
Now
for
JV
t /J?-t^v(v +
- 2D = - 2 + %v.
and therefore
is
iV> 4 +
Xv.
we must have
l).
95-96]
165
96.
The
last result
is,
surface) into another quartic surface, all the base points must be
at nodes*.
Further, the multiplicities at the base points must
be
(p.
2s
2 (2\ +
and then the dimension
l) 2
-2s
+ S
(2/^)
of the system
2w2
8,
exactly
3, so that in this
case the linear system of curves by which the transformation is
effected is complete, for it is determined entirely by its base points.
We
of
is
n and
express 2n2
as the
= 0,
/a
0,
sum
of 2s
Taking n
3,
=3
s
1
is
way
+ 1 + 4=
by
effected
is
10,
sextic
curves passing
version f.
Next, taking n =
1
3, s
= 5,
(ra
4) = 10.
+ 1+1 +
The
orjly
way
is
+ 1 + 1 = 10,
2s
or
*
= 2 <y - 4),
= s + 4,
This theorem follows at once from the fact that the rational curve into which
is transformed, lies on an "adjoint" surface of order
N-i.
t Neither, Math. Ann. in, 557; Cayley, Proc. London Math. Soc. in, 170;
Papers, vn, 230.
Coll.
weddle's surface
166
and since
admissible value
Any two
is s
giving n
5,
[ch.
0,
1,
3, 4, 5, 8,
xv
the only
= 3.
This
is
Ann.
in, 517).
weddle's surface*
97.
We
surface
now
<3>
is
'
'
etc. as
6 which are
'
and
1, 2, 3, 4, 5,
odd,'
By
lines.
these planes.
of
Hence
Rummer's
surface.
weddle's surface
96-97]
We
167
weddle's surface
168
birational transformation
between
<!>'
and
[ch.
xv
W by means of quadric
W*.
points.
sections.
triple
= 18
Of these ten are at nodes, leaving eight variable interHence the class of the developable is eight. Of the
xa touches the
tropes,
and
is
therefore a
..
+ 1+9 = 24.
all its
properties
Ku miner's
surface,
of
<3>,
which
The remaining
For
Bendiconti Lincei,
ser. 4, vi, 3.
De
Paolis,
Memoire Lined,
ser.
4,
i,
576
97-98]
169
for bitangents
example
Complete quadrilaterals can be inscribed in
over, for
having two
98.
P = ^P
Let
rs
xr xs =
Q = 2Qn xr xs =
R = ~ZRrs xr xs
= 1, 2, 3, 4)
{P = Pgr etc.)
(r, s
aP + bQ + cR + dS = 0.
Any
surface
YV
is
transformed, by taking P, Q, R,
as
U
new
a surface
V(P,Q,R,S) = 0.
the six base points a group of eight associated points. The locus
of (*') is another surface W' = 0, and there must be an identity
of the form
(P,
Q,
R,
8)=
WW
j_
d(P,
\30\i
Q,R,S) _
Q
#2> ^3> ^4)
i,
247.
weddle's surface
170
Now J=0
vertices
is
[ch.
is
XV
family
aP + bQ + cR + dS = 0,
for the conditions for
and on eliminating
at once that
a, b, c,
J contains
{x^
ddS/dxs =
x2 xs xt) are
,
(s
= 1,
is
2, 3,
4)
this it follows
the points.
all
J and
and then
and W, also
W,
coincides with J.
identity!*
&(P,Q,R,S) = J>
which admits of direct verification.
Let four of the base points be taken for tetrahedron of reference
and let the remaining two be (e^ e2 es> e4 ) and {f\,fz,f<i,fi). Then
the equation of Weddle's surface can be expressed in the convenient form |
,
Vl
^"1
98]
171
is
the last paragraph the elements of the leading diagonal are zeros,
and the equation has the form
= 0,
z
z
which
x'
y'
the same as
is
*Jxx'
where the
In
a, b, c, d.
+ >Jyy + 4z3 = 0,
coordinates
fact
etc.
points
points
(e)
relations
e2 es x
+ e&y + e
e^z
having
by means
a
common
of these quadrics
quartic
surface
having
points
into
a-
fourteen,
is
birationally transformed
surface
fifteen,
whose reciprocal
or
respectively.
On
solving the
first
(aP + bQ
+ cR + dS) =
doc.
for &i
a?2
x3
xt we find
VX
z "
%yz'
sixteen
is
nodes
weddle's surface
172
[ch.
xv
where
<
expressible in the
is
equivalent forms
kyy'zz = % ^zz'xx' =
1
(fy*
we observe
-ty*
4sxx'yy' =
0,
2/
= 0,
= 0;
method
of trans-
formation.
on Weddle's
lying
surface.
= 0, x^x^ = 0, and
Thus plane
b c d we
xx x2 xs xt
:
on a cubic curve
sections of
Kummer's
Caspary* gives the equation in terms of the tetrahedra whose corners are
the nodes A, B, C, D, E,
and any point
of the surface, namely
m-
+ 3 (2xPa 3
F= Sxyzt - ixz - iyH + 3yh2 - x2
where
and the
six
dF/dz
tp 3 ) dF/dt=0,
fi,
/(d^-p^+p^-p^+p^-psd+p^O.
A parametric expression of the surface is
x :y z t= U+ V: vU+uV v U+v?V: v U+u
where
W=f, V*=f{v).
:
Darboux
Coll.
V,
(Richmond.)
CHAPTER
XYI.
THETA FUNCTIONS.
99.
shown
is
of any point on
and
\/{x
2/
that
If,
<E>
6 ).
by
pairs
is,
then,
a ),
which are
With
finite
v1
=l dx/y+\
v2
dx'/y',
1
rx
rx
I
of the curve.
xdxjy +
x' dx'/y',
and then invert, that is, solve these equations for x and x'. It can
be proved that x + x' and xx' are uniform periodic functions of
The latter property is obvious since the integrals are
v 1 and v 2
.
THETA FUNCTIONS
174
[CH.
XVI
The
theta functions, in
are
shown
let
x=
If
<$>
x'
(! , v),
(/>'
(v x ,
at present
Vs).
Vi
(l
"a)
dxjy,
va
= <>
we have
xdx/y,
^)
= *o')
'.
Vi
and
v2
<j>
(v lt v s ),
being connected by
<' ("1,
v 2 )=
'.
Now
are projectively related to the coefficients in the canonical equation of the complex,
Kummer's
it
same
and therefore into one another, and can be uniformly
represented in terms of the same pair of integrals.
sextic
The
sum
of the
(ax + b) dx/y
Namely
of integrals
when
the summation
is
extended to
all
2/
(x
- k ) (x - kt) (x - k) (x - k )
3
the
99-100]
175
and
let
(x)
= yf(x),
its coefficients.
The
by
F(x) =
^-ff' = 0,
in each intersection
F'
which
is
0,
the same as
- yjrS<f>)/F'(x).
F'
Now X (ax + b) (<j}Sylr tyhfy) j (x) vanishes when the summation is extended over all the roots of F(x) =
because the degree
Sx/y = 2 (c/>0>
in
x of the numerator
is
two
at least
less
Hence
%(ax + b)8x/y = 0,
which proves the theorem.
The
as
100.
We
theta functions*.
shall
define
expressions,
One
elliptic
exp (aw2
+ 2nu)
periodic properties.
It is convenient to
modify
this slightly
and
write
6 (u)
),
all
positive
is
present purpose, and for references to the original authorities, see Baker,
Functions (1896) Krazer, Lehrbuch der Thetafunktionen (1903).
;
A belian
THETA FUNCTIONS
176
n by n
replacing
either
or 1
+ 1a
[CH.
XVI
$ may
be
'
functions.
We
formal expression,
its
by interpreting the
letters differently,
exponent, written in
2iri (n^ij
The condition
full,
is
+n u)+
is
for
iri
(T n ??! 2
2^2?^ n?
+ T^nf),
u2 + /32
|/S1;
respectively,
r^,
n2
w,,
u2 by
where a lt a2 &,
,
/32
Wi
+ ^otj,
a p(u),
n2
+ ^a2
are integers.
It
is
We proceed
to
many
relations.
101.
By
definition
Bap
which t
in
is
Til
\ T21
all
+ a)
},
and
ttm-(?i
^12
T22/
suffixes 1 and 2.
The summation is for
and n 2 from oo to + oo
i and a 2 are
integers which may be taken to be either
or 1, since the integer
parts of \ a may be absorbed in n
and /82 are also integers, and
since the addition of even integers to /3 can at most change the
sign of the function, it will generally be supposed that /3 X and /S2
letters distinguished
all
integer values of
by
are either
is
or
1.
This being
so,
&
the matrix
a,
a2 \
/V
afi.
100-101]
177
terms without altering the value of the function, since the series
absolutely convergent the general exponent is now
is
2tm
from which
it
('
+ ft, that is
Til 5i + tm 2 + ft u
TaSi +T S + ^
pair of quantities rd
22
called a period
is
Mu).
**(-) = (_-)"
a) 2
follows that
+ ft) = (-yee^(u),
(u + to) = (-) 5 P exp (- 27rt'aw -
8e(u
#0
The
of these
first
taking n
is
easily
new
a for a
7rtVa 2 )
a/3
().
is
(n
is
ftlt ft 2
by a
or
The one
1.
Any
period.
may be
(rot
in which all
Corresponding
+ ft), and
no two
+ t5 +
easily verified
I ft)
= exp
{- wia (u
We may
for the
9 a+Ei p+ ^ (u)
same values
Of the sixteen
Hence six thetas
from
h.
12
THETA FUNCTIONS
178
Whence
since the
[CH.
sum
XVI
of two half
is a
There is a close connection between the squares of the sixteen
Accordthetas and the sixteen linear forms considered in 15.
ingly we adopt a notation for the characteristics which brings out
clearly the analogy. This is sufficiently indicated by the schemes
periods
I)
"2
101-102]
+ /3)
u = \ (ja +
179
when
and a'/3' lie on the same row or on the same column. Hence
what was before an incidence diagram becomes now a table of half
if a/3
102.
in the product
tt/3
(m)
6^ (v)
is
2m (m +
Now
c,
Write
or d respectively.
m+n=
in
where
ix
sum
the
a,b,
c,
2fj, + a,
n=2v + a,
or d.
tt/3
(v)
may be arranged
as
is
(//,
is
now with
integers
a (u) = 2
(v) for
(_)3+a0
Hence giving
o+
which a
-
(u
a the values a,
+
is
2iriT (n
},
+ v) - ( u _ v
b, c,
+ \ a)
summed
).
results,
we
find
o*
()
ea+5 ( + ) ; ( - v)
= Z a+-a (u-v)(-ye (u + v)
e+ () = 2 (-ye+v
di
* Cf. Clifford,
Abelian Functions,
"On
p.
526.
122
THETA FUNCTIONS
180
XVI
[CH.
the values a, b, c, d
we obtain a matrix of sixteen elements which is evidently the
product of the two matrices whose elements are
By
a+;
and
(-)*"
O - )
(u
Written in
respectively.
giving a and
(a.
+ v)
full
(a,
ft
= a,b, c, d),
ft = a, b, c, d),
a
they are
b (U - V), a {U - v)'
a (U - V), 6 {U - v)
e d (-), e(-t;)
(u-v), d (u-v)_
a(u + v),- a (u + v),-a (u + v), a (u + v)~
-b (u + v), (u + v),- b (u + v), b (u + v)
- (u + v),-@ (u + v), (u + v), (u + v)
d (u + v), d (u + v), d (u + v), d (u + v)_
d ( - V),
(u - v),
(u - V), d (U - V),
(u-v), e. (-),
a (u-v), (u-v),
c
'
-'<
and we recognise that they have the same form as the matrices
which were multiplied together to give the sixteen linear forms
(cf.
We
a/s
() 6 afi (v)
(a/3).
eM (u) e (v),
ecb () ecb (v)~
odd (v) d dd (v),
ba (u) e ba (v),
-6ab {u)8ab (v), 0<fo(w) 0<to(), - (w). (), 6 bd (u) 6bd (v)
- ebe (u) ebc (v\ &cd (u) ecd o), e^ () edb (), - eM (u) eaa (v)
- Qca, (") Oca (v), ~ Obb (") &bb (), ^ad () &ad (v), 6dc (u) do (v)_
ais
flsc
C(!
From
relations
this,
among
103.
ap
0,
nearly
all
the
Firstly put v
u and
x =e(2u), y
(2m), g
(2m),
= d (2u),
= d (0),
y
then 6\ s (u) actually becomes the linear form denoted by (rs) in
which a, ft, y, 8 have been replaced by x y z, t respectively.
Xo
= a(0),
= b (0),
= o(0),
,
all
the
102-103]
181
"
"0(O)M)>
0,
0,
8 da (0)
0,
6* (v), -
CC
(0)
6 M (0) 6 M ()
(u),
o,
ed
(o)e cd
(u),
o,
-ehb
(0)0 m
(u),
e ad (0)d ad
and the
relations
equations of
(u),
0^
(0)
all
the irrational
dc
().
Rummer's
= a (2u),
y=
(2u),
= e (2u), t= d (2u)
The parameters
so that
we may
sixteen conies.
period,
say that
8^
(n)
is
the nodes.
From
x, y, z,
The
all
even.
dd
respectively,
and are
characteristics
Hence
for all
four functions
(-2k) = 6 (2w),
and from the periodic property
and
to.
2w - 7ri2ra
!1
(2m),
+ /3 to the arguments
Thus the
x, y, z,
ratios of the
surface,
but
u+ period
give the same point.
THETA FUNCTIONS
182
[CH.
XVI
The
the parameters.
104.
^ (u + to"
4-
^ (w).
Obviously when the order and characteristic are given, the sum
number of theta functions is another of the same kind.
Again, from this equation it follows that the product of any two
of any
theta functions
teristic are
is
By
a'/3'
Wa +
we
a', |3+/3'
is
n of the
n,
the
general
Kummer
where
either
(a/3)
and
103-104]
The converse
even or odd
of this theorem
theta
number
to
To prove
also true,
is
when equated
function,
183
zero, represents
this it is sufficient to
of linearly independent jl
is
equal to the
number
the two
2(+/8)=(-)*2(u).
so that
and in u2 independently,
say.
cients of e evi+v^
we
find
flffm,
-"n^a e
7cin 2
fi
J3A-n n
A
lL
pTria,
z
>
p nia 2
both even.
a x and m 2
a 2 are
^ (m + to) = (-)=3
and equating
exp {-
2irira (u
irinu
find
coefficients of e
eirinra
we
iriafi rrirra^
+ \to)} ^ (u),
A ni+Wj7h
i!
maJ
Since however
must be exactly r2
number
1),
or (r2
%(u + v) = Q, e( + tO = o
have Irs common
solutions.
Consider
definite
first
parity.
= v' =
t Beference, footnote
p.
186 below.
THETA FUNCTIONS
184
[CH.
XVI
arguments and each of the nodes only one, additive periods being
always neglected.
Hence the
2 (rs
Again
it
total
number
-p) + 2p =
of solutions
is
2rs.
e2(+.)eJ(t.-.) S e(),
and of
definite
Under
2r,
zero characteristic
a similar hypothesis
pairs of equations
''
v') = 0>
^(u+O-o)'
eg,
@2
e2(-*)=i
ls(-")=
*2<-)=i e
=
or
(+*')
e(
5.(t.+o-or
It
i
; -) =
s
<>.
105.
The whole
may
coordinates
may be equated
It is
ser. 4,
ix, 29.
104-105]
185
= #'/$' = e"
S/4>
F (u),
(M)
where
a uniform entire
is
= F(u M
+ l) = e " iau F(u),
Ffa + l,
F(u lt u2
F (% + t
F (U + T
2 ),
F (u),
u2 + t31 ) = e (6i+.+*
1T)
12 ,
ui)
U.2
+ T&) = e
2,ri (ft'v.+e'u.-Hi')
i? (
M).
By comparing
b,
aTu>
b',
If
we
are integers.
c'
write
ar 12
6t 12
= aru +
e,
+ ct &'ru - c't
c' = ar w + e, the last
22
we
21
result
gives
where /
periods
is
is
another integer.
22
22
2
12
+/= 0,
in general it is
among
the
assumed that
must
lation]:.
This
to
fill
is
the
gap
on the surface.
in
a continuous
theory of
algebraic
(p.
140)
curves
relations
J Krazer,
p. 116.
THETA FUNCTIONS
186
[CH.
XVI
One
other theorem*
zeros,
additive
being
periods
importance, namely
and n have 2mn common
of fundamental
is
Then the
disregarded.
surface
The tetrahedron
is
of
By
This
of tropes
is
relation f
and corresponds
obtained.
'Jasas'
after z
is
form of
to a rationalised
\/yy'
*Jzz'
on',
y,
0,
linear functions of
y'.
Again, if we take any one of the sixteen thetas of the first order,
and equate the non-homogeneous coordinates to the negatives of
we
Kummer's
If the
fundamental sextic in k
surface^.
is
5
6
/<;
*-6&
6
,
-X
|X!
2z
-2;/
-2y =
X
-4<z-\2
2i/ + \ 3
-4>cc-\4
2x
iX
2z
2y
+ ^\
0.
2x
\
-X
B
* Krazer, p. 42.
xi,
129.
t Gopel, Crelle (1847), xxxv, 291. This is the historical origin of the transcendental theory. Baker, Abelian Functions, pp. 338, 466.
J Baker, Proc. Camb. Phil. Soc, ix, 513 and xn, 219.
105]
187
xs
= 6\t
{u),
(a
= l,
2,
...&),
many
Kummer's
surface*.
suggested by the
hyperelliptic theta functions, for which the elements of the symmetrical matrix t are connected by | (p 1) (p 2) relations,
generalisation in a different direction (
is
and the absolute constants upon which the function depends are
independent cross ratios of the roots of the fundaSelecting any one of the 2 2p thetas
equation.
there are \p (p + 1) functions 3 2 log 0/du r dus (r, s, =1 ...p) connected by \p (p 1) relations. This set of relations represents a
the 2p
mental (2p
+ 2)-ic
p-fold in space of
* Wirtinger, (for
\p
p = 3),
(p
1) dimensions.
und Physik
ix, 521.
(1890),
i,
113.
CHAPTER
XVII.
106.
characteristic.
independent
is
22
= 4,
The number
and
(2)
From
it
follows that
e ? (u - v)
(
ttp
? (u
v
III j
+ v) = o
"-
'
().
*
first
order,
O - v) 0p (u + v)
arguments
Wj,
u2
e* ( + v) = o
At every point
v)
{U
+ V) = 0,
may
The advantage
is
that by means of
the equation
eafi
(u-v)=0
106]
TANGENT SECTIONS
homogeneous coordinates,
189
da x\doa and
and
z*
further,
expressible rationally in
terms of
oc,
y,
and w2 are finite, so that jdu^ and fdu2 are " integrals of the first
kind" for the curve. The number of such integrals is equal to
the deficiency of the curve, and so the plane section is of deficiency 2
Uy
at least.
The equations
0(u-v) = O,
0(u + v) = O
therefore
satisfies
have
(p.
^-{0(u-v)0(u + v)} = 0,
-{O(u-v)0(u + v)}=O,
this point
is
contact.
we can
In the identity
(p.
179)
= e( - v) a (u + v) + b (u- v)(u + v)
+ (u-v)c (u + v) + d (u- v) d (u +
replace u by u + v and v by u v then
( + V) 0o (U - V) = a (2) a ( + 6 (2) l (2)
+ (2d) c (2m) + d (2d) e d (2m)
= a (2d) * + 6 (2d) y + c (2d) + d (2d)
0,(u) 6>(d)
v),
u,
and (+
v)
may be
same functions
regarded as the
2
surface with respect to the fundamental quadric x
+ y + z + t = 0.
2
mental complex.
(0
Incidentally
we
and
notice
(d) is a
that
if
(u v)
= 0,
ray of a fundamental
given by
0(2w)
O,
(0
= even).
* Krazer, p. 117.
190
Let
(it)
0(u-v) = O,
if (P),
plane
= (PXi P )
2
and
(v),
is
is
an odd
0(u
(v),
[CH. XVII
then
+ v) = O;
+ P) lies on the
+ \P), which
joining (u) and (v + $P)
Hence the
line
Hence the
six
(u-) = O
u=
+ odd
half period.
Univocal curves.
The tangent
is
represented by
<">( -t;) =
where (u) is
and (v) is any pair of constants.
0,
points satisfy
u-v) = 0,
( u v) is an
<>(-
and
since
(n)
(u v)
in)
e(u-ti) = o,
is
n-ic surface.
<)(_ u -v)
of order 4w
There are m2
= o.
(u - v) ' (- u - v)
(- u - v) ' (m - v)
numerous geometrical theorems
of considerable interest.
107.
We
v^
COLLINEAK POINTS.
and u2 are integrals of the
first
the curve
0(u-v) = O,
*
4, ix, 154.
kind for
COLLINEAR POINTS
106-107]
191
so,
collinear points
+ u' =
const.
By
+u =
considering a bitangent
we
find that
+u =
2v (mod. P),
known
double point.
putting
u'
= u,
0(iP)
The
= O.
0(u-v)0(u + v) = O,
0(u-v')0(u + v') = O,
and are the solutions of the four
0(u-v) = O\ n
0(u-v') = O) K)
0(u-v) = O)
0(u + v') = 0]
'
'
+ v) = O\
+ v') = O\ Kh
0(u + v) = O)
(u - v') = 0)
0(u
0(u
Let (a) and (b) be the two solutions of (1), (a) and (6) of (2),
and (d) of (3), (- c) and ( d) of (4). Then the four collinear
have parameters (a), (b), (c), (d), and the
(u v) =
points on
same points on (u v') = have parameters (a), (b), ( c), ( d).
Hence any two tangent sections cut in four points, whose parameters on the two curves are the same except for two changes
(c)
of sign.
If then (a), (/3), (7), (8) are the (pairs of) parameters of four
tangent planes through a line which cuts the surface in (+ a),
192
may
(+ b), (+ c), (+ d) we
parameters
(
(a)
(-a)
(b)
(-c)-(d)
(c)
(d)
(c)
(d)
0(u-8)=O.
(-a) (-6)
(a)
(b)
sum
Since the
8 (u
v)
is
a) = 0,
0(u-ff) = O.
0(u- 7 = O,
( c)
b)
(d)
on the section 6 (u
we have
2v
a-b -c + d= 2a +P
-a + &-c + d = 2/3 + P
-a-& + c + d = 2-y+P
a + b + c + d = 2S + P
Substitute for
6>(a-a) =
a, ft, y,
0{(a
lt
2
6>(&-/3)
0,
then
[CH. XVII
= 0,
0(c-7) = O,
+ & + c-d)-iPs =
}
0(d-8) = O,
(s.=
O,
l, 2,
3,4)
implying that four theta functions vanish for the same point
^ (a + b + c d). But singular conies do not intersect except at
nodes, and so
we
infer that
iP = P
Hence
(
c),
= PS = P4
(mod. P).
6),
( d) are that
\ (- a +
must
all
Of
is,
must
repre-
course, in these
b, c, d may be changed.
2Q = P + P2 + P3 + P4 wefind
= |( -/3- 7 + 8) + iQ,
& = -H-a + /3- 7 + S) + !Q,
c = (-a-/3 + -y + S) + Q,
d = ( a + /3 + y + 8) + lQ,
a
and
^(-a + b + c + d) = ^(-a + l3 + y +
(mod. P)
8)
+ %Q,
By way
of illustration
we
same
form.
functions.
COLLINEAR POINTS
107]
(1)
and (+
193
b).
$(a-b) + u, \ (a + b)
= u, d = u the arguments become
%(a-b), $(a + b)u.
;
if
however we put
(4H-*
<(-; )-.
for Oi,
0(a-v) =
showing that
and (6).
= b = u,
showing that (u
function.
(3)
(w),
then
0(u+v) = O,
and contains
(a)
two
= d = v.
0(u)
will
= 2v,
First put
The
a+b
(v) is
undeter-
*(^-)-o
0(u-v) = O,
O,
(2)
cases.
0(b-v) =
O,
If
is
This
is
6(v)
Q,
= 0,
d = v;
= O,
v)
Secondly
then the conditions of collinearity are
0(u-v) = O,
0(u
= b = c = u,
d=
+ v) = O,
is
Then the
v.
fU
V\
+a=+6= + c = w.
It
we take
conditions are
fSu
n
-o,
V\
a
= o,
*(2^)
at
is
0(u)
h.
= O,
13
[CH. XVII
194
and the
and 0(2w) = O;
is
0(2u) = O,
where
an odd
(a/3) is
characteristic.
108.
Let
ASYMPTOTIC CURVES.
0(u
v)
0(u-v) = O,
O,
(u
0(i>
= 0,
dv ) = 0,
dv^)
a
The
(u
One
of these
Making use
is
0W
(u
of this
we
+ v) du, + 0
0W{u +
v),
0(u + v)
0(u-v)
di^
du 2
(u
= dv,)
dv
0,
'
2)
we deduce
alternative
0v(u-v),
and
first
+ v) du = 0.
(u
or
the
= 0.
given by
From
v)
-v)du + 0
l
(u - v) du 2 =
0,
showing that the two inflexional tangents coincide and the locus
of u is either a cusp locus or an envelope of asymptotic curves
the cusp locus consists of isolated points at the nodes, and the
envelope, which is the parabolic curve, consists of the sixteen
If (u) is a general point of the surface we must
singular conies.
take the second alternative and find the integrated equation*
u
* Reiehardt,
v, 465.
= v + k.
(1886), u.
107-108]
Hence the
ASYMPTOTIC CURVES
195
0(2u-k) =
where the constant k
is
O,
satisfies
0(k) = Q.
The sixteen points (k + \P) are the points of contact of the curve
with the tropes.
The equation should be written
0(2u + k)0(2u-k)
O,
and
curve
is
all
"
the nodes.
inter-
Thus
+ k) = 0,
0(2u-k) = O;
(2m
is
collineations.
The equations
0(2u-k) = O,
0(k)
= O,
= k + k',
\ (k + k') + \P.
2u
that
is,
may
(2u
+ k') = 0,
it
132
196
the same, and
order
We
8.
we get the
six
[CH. XVII
have
#aj3
(m
As an example
may
be given.
it
in
4n
number
points,
of plane (w+l)-ics
cuts the tangent section again in two points collinear with the point of
contact.
If the point of contact is on the 4w-ic the remaining in 2 intersections lie on a plane n-ic through the point of contact.
109.
INSCRIBED CONFIGURATIONS.
bc+d
+d
2y= ab + c + d
28 =
a + b + c + d;
x = d a = c +/3 =6 +y = -a + S
y = c + = d /3 = a + <y = -b + S
z = b+a. = a + $ = d-y=c+8
t=a-a=b /3 = cy= d + 8,
2a
2@ = -a+b-c
thus
tx
line are
108-109]
showing
INSCRIBED CONFIGURATIONS
the incidences
all
for
197
may
is
Klein's tetrahedra.
If
is
now we suppose
that
does not
(t)
lie
on the
conic,
but
by the
corners of a tetrahedron.
x1 ,y ,z
( 34).
hedra
is oo
t2
5
.
Instead of taking
0(u-8) = Q;
+ a = 6 + 8 = 3/
b+a= c + 8 = z
when
, 8, b
are given c
From
are
(a), (8)
is
"
conjugate
" to
the points
rationally determined.
(6), (c),
is
we
see that
d and
two pairs correspond, giving
is
a,
b,
c,
six conjugate sets, incident with the six edges of the principal
tetrahedron.
Expressed in terms of
x, y, z,
t,
=x+y+zt
2/3= x- y + z -t
27= x + y z t
2S= x + y + z + t
=x+y+z+t
26= x y + z + t
2a
=
2d=
2c
2<x
x+yz + t
x+y + z-t
set is given
by
a= h + y + z\
b = S y\
S = S
c = 8-z)
arbitrary and 8{y) = = 6 (z).
j
where 8
is
Bohn's theorem.
The
0(uv) = O
we can
write
of sixteen points of
Rummer's
198
configuration.
Let
(/) be any
[CH. XVII
six points
on
the conic
6 (u)
then the plane
^(a + b + c + d + e +/),
=v
^(a + b + c + d + e-f)
%(-a + b + c + d + e+f)
and these lie in the planes obtained from the first by changing the
signs of two of the parameters, and so on. The group of operations
is precisely the same as that which deduces thirty-two lines from
a given one in Klein coordinates. The sum of the parameters
of the above six coplanar points is 4u, showing that they lie on
a conic through the point of contact of their plane *.
Humbert's tetrahedra.
158)
is
Humbert's
of
tetrahedra "f"
on the conic
()
0.
The plane (y + z) contains the point (y) since the difference of the
parameters represents a point on the conic, and for a similar reason
contains (z). The second tangent plane through these two points
(y
meet by threes
in the points
= (-x + y + z)
(b)= (x-y + z)
(c) =
( + y - z)
(a)
(<*)
(*
+ y + )
the
line
joining
them
(d),
and since
passes
through
the point of
contact.
t Humbert,
'
109]
INSCRIBED CONFIGURATIONS
199
The
lines.
may
The
points of contact of
between
its
ABC]
A'BC]
A'B'G']
ABC]
AB'G)
A'BC)
ABC'}
A'B'G]
CHAPTER
XVIII.
110.
We have seen in the case of the Wave surface that in consequence of a special situation of the nodes in each trope and
a corresponding relation among the coefficients ks of the quadratic
complex, the point coordinates may be expressed in terms of
In the present section we seek the correspondelliptic functions.
ing relation among the periods of the theta functions. Starting
with the more general problem of linearly transforming the
arguments of the theta functions into arguments of
functions,
we
find
of this form
we
of which the
Wave
Kummer
different relations
surfaces of
elliptic
among
surface
which
is
the
first,
and
Kummer
u2
surfaces
Tn ai +
T 21 a! +
Ti 2 0,
TjgOs
first
order
+ /3
+ /32
1
Assume
U = <)\U\ + gi^,
110-111]
ELLIPTIC SURFACES
periods 12 and
integers
m m
8
'
S T n + st t = m l + m/fi'
g^n +92^12= nitl + m 'D,'
f
2i
=m n + m 'n'
s
whence, on eliminating g 1
Tn
g2
g2
=m
,
i Cl
+ m i'Q,';
X2, Xi',
= 0,
201
Hence eight
'
202
[CH. XVIII
= (T7 + S) -1 u
(!!)
Then the
first
the matrix
= (TY + 8)-
(Ta
+ /3).
The
condition
for
the
transformation*
so
it is
ordinary
form exists,
this
7a =
=
8a
yS-y = r = a.8 y&,
017
/3S-8y8
The integer r
is
S-J8N
(S
/3\
\y
a)
\-y
8J
/r
0\
\0
r)'
is
often
each matrix having four rows since the right side is a numerical
multiple of the unit matrix, the order of the factors on the left
;
may
0\/r
5 -ySx (8
8/ \7 a)
V-7
/
which
is
0\
\0 r)
= r=8~0L yfi,
aP-J3a = = y8-8y.
a8 J3y
of this
358.
reproduction
THE INVARIANT
111-112]
112.
When
203
THE INVARIANT.
exists,
Kummer
It is evident
case.
may
function
when
Mi
ft
and
so
on
/a
y\
\/3
SJ
this
M,
21
. . .
M M M \_ /m, m m
/M,.
From
21
satisfied
by
'
a, /3, 7,
we deduce
204
If r
= 1,
[CH. XVIII
that
also linear, so
is
we
have in this case o-2 A = A', where o-Sl; whence p=l = <r and
A' = A, proving the invariance.
In the general singular case in which A is not a square, m,,m,',
not necessarily integers, can be found in many ways to express the
given singular relation in determinant form. The lines of the
preceding proof may then be followed and lead to the same two
properties of formal and numerical invariance.
It can be proved* that by a linear transformation the singular
relation can be reduced to one of the canonical forms
-iAT +T
sa
11
=0,
- i (A - 1) t + T + T* = 0,
Both of these are included
in the
form
from which
it
for
V 1
that
and
(p. 176).
it
follows
-1=0,
We shall suppose
&T]2
for this has invariant
done
then different
A=
It?.
113.
The
is
and
k.
PARAMETRIC CURVES.
distinguished by different
them
of
Kummer
= const.,
surfacesf
u^
= const,
112-113]
On
(A,
205
elliptic
a2 )
(ai,
PARAMETRIC CURVES
is
Pa)
is
= !,
Wg =
0t 2
i= a 1 + P
m2 = (h + Q 2
1 ,
where (Q1;
2 ) is
independent.
=
w2 = a2
Mi
+ Pi - Q u
are given by
W2
<m^
where
&
1,
C&2,
are
2/c
common
points.
We
the relation
(i + Tn a +
+ ftu 2 + T-nCli + T
+ /3 )
22 CC2
TiaOta
we
1 tt 1
12
a 1 a2
+T
2
22 or 2 )},
find
(! + TjS + /8
showing that is a
1(
m2 )
= (!, m )
2
exp ( ikirioiih
2&7n.Ya
2
),
The
number of zeros* not differing by multiples of 1 and Tx is 2k, and
a parameter curve w2 = const, cuts one of the coordinate planes in
2k points, and
is
If a surface
of order k be
is
1.
arbitrary point of
If,
further,
w2 = a2 ,it
* Krazer,
made
Oj, it will
will
to pass
contain
p. 41.
206
Now
it.
2&2
1 is
[CH. XVIII
number
exactly the
Kummer's
surface
k.
it,
1,
cut out by
is
surfaces through
A--ic
any curve
may be
where
tion
obtained by using
denned by
all
is
2
}
summa-
22 exp - faik-
+ l-a,,) + I
{2u, + (2*, + a ) k~\ 2t u X {2 + (2/ + a ) k~\ 2rw
where (a a ) = (10), (11), (01), (00) in succession.
The two parametric curves u = a and w = a are given by
1
(v,.
(j/j
lt
<*!>}
i0
],
2,
the equation
Sii (ku!
kdj
%i (ku 2
The
0.
k.
114.
The curves w
UNICURSAL CURVES.
= o +^P
1
and
= a + \P are
When Oj is small
w,
2)
(!
u1
krn )
left side is
characteristic,
(Pj,
+ ka^
ka
near.
the points
system and
the same,
is
therefore of
k.
coordinates
are therefore
functions of a
new
proportional to
variable u 2
\P2
113-114]
UNICUBSAL CURVES
207
When
is
odd, there
lie
on the curve
Mi=i( Tni +
T 12 a 2
),
a2 \
l& OJ
(<*i
<h\
Vft
1/
Vft
<*i
+l\
+1
<*i
U+i
a 2 +l
1
the first parameters of these nodes are either equal to the preceding value of m x or differ from it by | (t 1s + 1), which is here equal
to the period (k + 1) t12 on account of the singular relation. The
for
sum
product
is
Hence there
first
family of
M2
= i (Taii + TjjOa + /3 )
2
aA
/a,
a 2\
\o
&)
\i
&)
/Oj
+1
/ x
+l
A+U Vi A +
i/
node itself counting as two, so that the two unicursal curves cut in
one node and \ (k 2 - 1) other points.
A surface S of order \ (k + 1) can be drawn through these
|(A; 2 +1) common points and ^(k + 1) arbitrary points on each
2
S then passes
curve, for this makes ^(k+l) + l conditions.
through both curves. Now each curve meets the common face of
the two Rosenhain tetrahedra in three nodes and \ (k 3) points
of contact, so that 8 meets the conic in that face in k + 3 points,
and therefore contains it entirely. Hence two unicursal curves of
different systems, together with one conic, form the complete intersection with a surface of order %(k + 1).
208
When k
is
[CH. XVIII
= i ( Tu a + Tu a2 + &)
i
a2\
/,
aA
lo
oy
vo
1/
/a, a.A
/Oj a s \
U u u
or
= h (T
si
ai
+ Ts2s + &)
same nodes.
Two unicursal
2 points. A
of these common
115.
When k
is
the diagram, in which the four tetrads and one set of coplanar
nodes are indicated.
Hence three groups of the involution consist of two nodes and
\ (k 2) points counted twice, and one group consists of \k points
counted twice.
When
is
114-115]
209
two
sets of three
to
2,
that
;
is conies,
intersecting
Three of these pairs of conies cut any singular conic in three pairs
of nodes the remaining two conies touch the singular conic at the
double points of the involution to which the nodes belong.
There exist on the surface two pencils of elliptic quartic curves
(intersections of pairs of quadric surfaces), obtained by making one
parameter or the other constant. These results agree with what
has been proved before by other methods.
Next suppose k = 3.
Each pencil of parameter curves consists of sextics* cutting
each singular conic in groups of three points of an involution
;
2, 4, 6,
are coincident,
them.
passes through
surface of invariant 9
among the
coefficients
k s or modular equation,
,
|(t-i-y 4 = 0,
is
(r,s=l,
2,
3)
xvi, 249.
14
[CH. XVIII
210
scribing
1,
and inscribed in
3, 5,
2,
Similarly there
6.
4,
is
INTERMEDIARY FUNCTIONS.
116.
Consider
now Kummer's
the usual way, except that the periods are connected by a singular
relation
whose invariant
is
They
not a square.
are characterised
general exist.
F (u) = 0,
is
\og{F(u + Ta
hyperelliptic
where
+ /3)/F(u)}
By multiplying
1.
by the exponential of a
2 1, and making the periodic conditions consistent we can arrange that
is
linear in
ult u2
quadratic in
v^,
F( Ul + l,u2) = F(u),
u2 +l) = F (u) exp
F{ih.,
2-Kimu^
12
u2 ) +
const.},
1-22)
12
u2)
+ const.
J,
where m, n11 n n n^,n^ are integers and the periods are connected
by the relation
,
WiiTia
+ (rh2 + it)
in this relation
tion of order
t-22
all
=n
21
Tn
+ (n m + mT
must be
or n^
zero,
-t
2
12 ,
and then
but when
may be made
;
+ integer.
tj2 ra Tn T22
,
t 12
12 )
and
1,
functions different from theta functions can exist and have the
preceding properties. By a linear transformation it is possible
= and reduce the singular relation to the form
to make
WsiT n
+ (^22 - Mil) tm - n a Ta =
0,
115-116]
INTERMEDIARY FUNCTIONS
211
for
in-
variants.
U =n
\n\
where
in
+
= n^n^ n
\n\
21 u 1
12
n 2i
is
U = nu.
is
is
de-
w T = TIT,
and
n 22 u 2
Then
* = *(&),
U + l) = $(U),
<I>(t/i + l, U = (U
* (Cfi + Tu, U + t = $ U) exp \n\ ^ + const.},
=
* #1 + t 17 + t^)
U) exp - 2m \n JJ + const.},
where
2)
12 ,
21 )
<I>
2-jri
characteristic.
The
We
equalities,
as
Since
follows.
the
relation
"si Til
is
positive,
T 22
ra 12 ra 21 )
>
- %l) T - n
(22
12
12
we have
(n n
+n
22 )
4 (nn n 22
0.
then
the con-
> 0,
t"
for,
but
|ra|
and
= |ra||T"|,
> 0,
n = n-aU^, n n n > 0.
rn " = WnTn" + n t " > 0,
2
|T"|
=
|
so that
Lastly,
|rcT"|
t"
21
12
and
it
21
22
?? 22
>
Vn u n22 - n 12 n
2l
142
212
SINGULAR CURVES.
117.
Taking
<& (
characteristic,
equation
is
we
[CH. XVIII
and
whose
F (u) =
where
and
F is
This
differs
F(u) = e"m 2
the preceding relation gives
of
A SiiSi
whence
If
2 "*
and
-4 Sl+%1
,2+ n M
in terms
follows
it
pendent functions
matrix n is the area of the parallelogram whose corners, referred
to rectangular axes, are (0, 0), (n^, Wj. 2 )> (au ^22), (%+%, n 12 + n22),
that is, n
The nodes through which the singular curve F (u) = passes
are given by the half-periods for which F vanishes.
These are
seen from the periodic relation to be \ (tS + #) where
I
|.
fi/3
+ ayS + an/3
Hence
all
n"
n*j
> 'vVi22 -
WiaWai,
n n _1 =
I
12
1
11'
cases.
By adding
its factors.
117-118]
SINGULAR CURVES
213
K< +
- n^tia - n n
On putting n,,' = n = 2, n = n = 0, the second
plane section, and we get the order n n + n
i
22u'
12 ).
21
'
22
12
21
curve becomes a
22 .
The
118.
5.
invariant 5 depends
satisfying
nn
The
+ n^ = V4 (nu n^ - n
12
n 21 )
5.
ingly
we take n = n 12 = w =
equation F(u)
21
the characteristic
nw =
1,
The
2.
is
corresponding
+n
= 3,
and if
through the nodes \ (ra + /3)
rin
22 ,
where
<*i
A + i/3
+ 2/3i + 2a /3
Now it is easily
2
congruence
(!
+ O & + !& =
(mod. 2)
+a
2)
&+
oti/3.,
(mod. 2)
to
dd
ac
ba
cb
ab
-da
cc
bd
be
cd
db
aa
ca
bb
ad
dc
form a Weber-hexad.
pass
and
through the ten nodes since five of them
the
projected
from
node
is
It
so must pass through the hexad.
(aa) by a quadric cone passing through the intersections of the
tropes (cd), (dc), (bd), (db), (cb), (cd) taken consecutively and
therefore touching the remaining trope (be) which passes through
the node (aa). The fact that such a conic can be drawn is a
consequence of the singular relation, and it is easy to express the
lie
214
The values
i11
= 2, n^=l,
? ]2
that
(i
is,
+ as)
= n = 1,
21
+ /3)
where
A + a A = 1 (mod. 2)
A + A = 1 (mod.
2
<*
dd
[CH. XVIII
2),
118-119]
215
2t
+ rm = 0.
The
(a/3)
given by
that
is
a'/3'
aft
(!
')
First take
2),
or
1.
'
the congruence
since a 2
and ft are
arbitrary.
lj
or
lj
(l
(ba)
(act)
The
tetrad.
l)
(ab)
(l
lj'
(bb)
alternative congruence
*,&
= (>
Hence there
exists
is
will
By
The
By taking
different values
for.
a2
'
and
ft'
Again taking
the group-set.
we
/u
i.W
\n n
nj
-2\
2
a' ft
passing through
so that it
Thus there
is
is
+a
ft'
+ a/ft + 'ft + a
2
ft
or
1,
[CH. XVIII
216
BIRATIONAL TRANSFORMATIONS OF
INTO THEMSELVES.
120.
KUMMER SURFACES
identity
be transformed birationally
for
the general
among
Kummer
can
surface
different orders
these, for
The equation
tetrad of nodes.
is
so that
by using
= y'y = z!z =
different tetrads
we
t't,
obtain in this
Again
way an
the
(s
infinite
equation of
= 1,
2, 3,
is
substitution.
There are
obtain in this
Kummer's
two
way another
4)
fifteen equations
we
group of transformations of
a one-one relation between the
infinite
is
surfaces.
There
exist special
Kummer
We
120]
217
the same.
is
in
same
for different
polygons.
Under these
I.
218
The
third
[CH. XVIII
how
The general
by equating
Thus we
are led
known
new
must be of the
first order.
former case
it
cases, according
as the transformation
is
ul = Mj + const,
w/ = M
and by
Kummer's
surface
(i,
2)
surface
+ const.,
= w + const.,
uj = m + const.
w2
'
is
the transformation
Wi'
= Hj + ai,
w 2'=w2 +a a
is
Humbert, Liouville,
120]
219
and
into
this again
itself,
among
the
Kummer
surface.
INDEX.
Theorem 188
Algebraic curves on
Abel's
Elliptic coordinates 55
Summer's
surface
Kummer
surfaces 200
Fresnel 100
Geiser 12
Baker
Ball 38, 40
Gopel 19
even tetrads
Battaglini's complex 94
79, 85
Gopel's biquadratic relation 186
Group 4
Harmonic complex 94
Hermite 202
Hess 45
Hexads 80
Hierholzer 170
Hudson 139
Humbert 3, 76,
Caporali 45
90,
194, 216
Identical relations
179
Incidence diagram 7
Intermediary functions 210
Irrational equations of Kummer's surface 35
5, 44,
131; the
60 16 42
Congruences of bitangents
Cremona 132
53, 63, 66
Kantor
94,
Kummer
Darboux
De
14, 77, 87
170
Lie-Scheffers 102
Paolis 168
Desmic tetrahedra
216
Klein 36, 38, 40, 42, 49, 53, 56, 57, 58,
1,
43
Line-geometry 37
INDEX
222
Martinetti 45
Schroeter
Matrices 24
Schur 94
Minchin 37
Segre 93
3,
12
Serret 31
Niven 107
Node 14
Singular conic 14
gtaude 36
Stephanos 3, 45
Study 3, 83
Sturm
Painleve 184
Summer's
complex surface 68
functions 173
Transformation of Rummer's surface
163
of theta functions 201
Trope 14
Pliicker's
Summer's
surface 115
Keichardt 194
Eeye 12, 38, 63
Eichmond
170
Bodrigues 32
Bonn
Bosenhain 31, 68
odd tetrads 78, 83
Bouth 37
Salmon-Fiedler 63
Schilling 122
Schottky 170
CAMBEIDGE
Wave
surface 100
Weber
Zeuthen 72