Professional Documents
Culture Documents
Advisory Board
Nicholas Ayache
James S. Duncan
Alex Frangi
Hayit Greenspan
Pierre Jannin
Anne Martel
Xavier Pennec
Terry Peters
Daniel Rueckert
Milan Sonka
Jay Tian
Kevin Zhou
Titles:
Balocco, A., et al., Computing and Visualization for Intravascular Imaging and Computer
Assisted Stenting, 9780128110188
Dalca, A.V., et al., Imaging Genetics, 9780128139684
Depeursinge, A., et al., Biomedical Texture Analysis, 9780128121337
Pennec, X., et al., Riemannian Geometric Statistics in Medical Image Analysis, 9780128147252
Wu, G., and Sabuncu, M., Machine Learning and Medical Imaging, 9780128040768
Zhou K., Medical Image Recognition, Segmentation and Parsing, 9780128025819
Zhou, K., et al., Deep Learning for Medical Image Analysis, 9780128104088
Zhou, K., et al., Handbook of Medical Image Computing and Computer Assisted Intervention,
9780128161760
RIEMANNIAN
GEOMETRIC
S TAT I S T I C S I N
MEDICAL IMAGE
A N A LY S I S
Edited by
XAVIER PENNEC
STEFAN SOMMER
TOM FLETCHER
Academic Press is an imprint of Elsevier
125 London Wall, London EC2Y 5AS, United Kingdom
525 B Street, Suite 1650, San Diego, CA 92101, United States
50 Hampshire Street, 5th Floor, Cambridge, MA 02139, United States
The Boulevard, Langford Lane, Kidlington, Oxford OX5 1GB, United Kingdom
Copyright © 2020 Elsevier Ltd. All rights reserved.
No part of this publication may be reproduced or transmitted in any form or by any means, electronic or mechanical, including
photocopying, recording, or any information storage and retrieval system, without permission in writing from the publisher.
Details on how to seek permission, further information about the Publisher’s permissions policies and our arrangements with
organizations such as the Copyright Clearance Center and the Copyright Licensing Agency, can be found at our website:
www.elsevier.com/permissions.
This book and the individual contributions contained in it are protected under copyright by the Publisher (other than as may be
noted herein).
Notices
Knowledge and best practice in this field are constantly changing. As new research and experience broaden our understanding,
changes in research methods, professional practices, or medical treatment may become necessary.
Practitioners and researchers must always rely on their own experience and knowledge in evaluating and using any information,
methods, compounds, or experiments described herein. In using such information or methods they should be mindful of their
own safety and the safety of others, including parties for whom they have a professional responsibility.
To the fullest extent of the law, neither the Publisher nor the authors, contributors, or editors, assume any liability for any injury
and/or damage to persons or property as a matter of products liability, negligence or otherwise, or from any use or operation of
any methods, products, instructions, or ideas contained in the material herein.
ISBN: 978-0-12-814725-2
Contents
Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xvii
Part 1 Foundations of geometric statistics
Chapter 1 Introduction to differential and Riemannian geometry . . . . . . . . 3
Stefan Sommer, Tom Fletcher, Xavier Pennec
1.1 Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
1.2 Manifolds . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
1.3 Riemannian manifolds . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
1.4 Elements of analysis in Riemannian manifolds. . . . . . . . . . . . . . . . . . 20
1.5 Lie groups and homogeneous manifolds . . . . . . . . . . . . . . . . . . . . . . . 24
1.6 Elements of computing on Riemannian manifolds . . . . . . . . . . . . . . 29
1.7 Examples . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
1.8 Additional references . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36
Chapter 2 Statistics on manifolds . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39
Tom Fletcher
2.1 Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39
2.2 The Fréchet mean . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40
2.3 Covariance and principal geodesic analysis. . . . . . . . . . . . . . . . . . . . . 44
2.4 Regression models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50
2.5 Probabilistic models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71
Chapter 3 Manifold-valued image processing with SPD matrices . . . . . . 75
Xavier Pennec
3.1 Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75
3.2 Exponential, logarithm, and square root of SPD matrices . . . . . . . . 79
3.3 Affine-invariant metrics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81
3.4 Basic statistical operations on SPD matrices . . . . . . . . . . . . . . . . . . . . 96
3.5 Manifold-valued image processing . . . . . . . . . . . . . . . . . . . . . . . . . . . 103
vi Contents
Stephen M. Pizer, Junpyo Hong, Jared Vicory, Zhiyuan Liu, J.S. Marron,
Hyo-young Choi, James Damon, Sungkyu Jung, Beatriz Paniagua,
Jörn Schulz, Ankur Sharma, Liyun Tu, Jiyao Wang
6.1 Introduction to skeletal models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 233
6.2 Computing an s-rep from an image or object boundary . . . . . . . . 236
6.3 Skeletal interpolation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 239
6.4 Skeletal fitting. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 242
6.5 Correspondence. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 245
6.6 Skeletal statistics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 245
6.7 How to compare representations and statistical methods . . . . . . . 253
6.8 Results of classification, hypothesis testing, and probability
distribution estimation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 256
6.9 The code and its performance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 265
6.10 Weaknesses of the skeletal approach. . . . . . . . . . . . . . . . . . . . . . . . . . 266
Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 267
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 267
Chapter 7 Efficient recursive estimation of the Riemannian barycenter
on the hypersphere and the special orthogonal group with
applications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 273
Rudrasis Chakraborty, Baba C. Vemuri
7.1 Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 273
7.2 Riemannian geometry of the hypersphere . . . . . . . . . . . . . . . . . . . . . 274
7.3 Weak consistency of iFME on the sphere . . . . . . . . . . . . . . . . . . . . . . 276
7.4 Experimental results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 283
7.5 Application to the classification of movement disorders . . . . . . . . 284
7.6 Riemannian geometry of the special orthogonal group . . . . . . . . . 287
7.7 Weak consistency of iFME on so(n). . . . . . . . . . . . . . . . . . . . . . . . . . . . 288
7.8 Experimental results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 291
7.9 Conclusions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 293
Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 294
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 294
Chapter 8 Statistics on stratified spaces . . . . . . . . . . . . . . . . . . . . . . . . . . . 299
Aasa Feragen, Tom Nye
viii Contents
Contributors
Martin Bauer
Florida State University, Department of Mathematics, Tallahassee, FL, United States
Rudrasis Chakraborty
University of Florida, CISE Department, Gainesville, FL, United States
Benjamin Charlier
IMAG, Univ. Montpellier, CNRS, Montpellier, France
Institut du Cerveau et de la Moëlle Épinière, ARAMIS, Paris, France
Nicolas Charon
Johns Hopkins University, Center of Imaging Sciences, Baltimore, MD, United States
Hyo-young Choi
UNC, Chapel Hill, NC, United States
James Damon
UNC, Chapel Hill, NC, United States
Loic Devilliers
Université Côte d’Azur and Inria, Epione team, Sophia Antipolis, France
Aasa Feragen
University of Copenhagen, Department of Computer Science, Copenhagen, Denmark
Tom Fletcher
University of Virginia, Departments of Electrical & Computer Engineering and Computer
Science, Charlottesville, VA, United States
Joan Glaunès
MAP5, Université Paris Descartes, Paris, France
Polina Golland
Massachusetts Institute of Technology, Computer Science and Artificial Intelligence Lab,
Cambridge, MA, United States
Pietro Gori
Télécom ParisTech, LTCI, équipe IMAGES, Paris, France
Junpyo Hong
UNC, Chapel Hill, NC, United States
Sarang Joshi
University of Utah, Department of Bioengineering, Scientific Computing and Imaging Institute,
Salt Lake City, UT, United States
xiv Contributors
Sungkyu Jung
Seoul National University, Seoul, Republic of Korea
Zhiyuan Liu
UNC, Chapel Hill, NC, United States
Marco Lorenzi
Université Côte d’Azur and Inria, Epione team, Sophia Antipolis, France
J.S. Marron
UNC, Chapel Hill, NC, United States
Stephen Marsland
Victoria University of Wellington, School of Mathematics and Statistics, Wellington,
New Zealand
Nina Miolane
Université Côte d’Azur and Inria, Epione team, Sophia Antipolis, France
Stanford University, Department of Statistics, Stanford, CA, United States
Jan Modersitzki
Institute of Mathematics and Image Computing, University of Lübeck, Lübeck, Germany
Fraunhofer MEVIS, Lübeck, Germany
Klas Modin
Chalmers University of Technology and the University of Gothenburg, Department of
Mathematical Sciences, Göteborg, Sweden
Marc Niethammer
Department of Computer Science, University of North Carolina at Chapel Hill, Chapel Hill, NC,
United States
Biomedical Research Imaging Center (BRIC), Chapel Hill, NC, United States
Tom Nye
Newcastle University, School of Mathematics, Statistics and Physics, Newcastle upon Tyne,
United Kingdom
Beatriz Paniagua
UNC, Chapel Hill, NC, United States
Xavier Pennec
Université Côte d’Azur and Inria, Epione team, Sophia Antipolis, France
Stephen M. Pizer
UNC, Chapel Hill, NC, United States
Thomas Polzin
Institute of Mathematics and Image Computing, University of Lübeck, Lübeck, Germany
Contributors xv
Laurent Risser
Institut de Mathématiques de Toulouse, CNRS, Université de Toulouse, UMR CNRS 5219,
Toulouse, France
Pierre Roussillon
ENS Cachan, CNRS, Université Paris-Saclay, CMLA, Cachan, France
Jörn Schulz
Arctic University of Norway, Tromsø, Norway
Ankur Sharma
UNC, Chapel Hill, NC, United States
Stefan Sommer
University of Copenhagen, Department of Computer Science, Copenhagen, Denmark
Anuj Srivastava
Florida State University, Tallahassee, FL, United States
Liyun Tu
UNC, Chapel Hill, NC, United States
Baba C. Vemuri
University of Florida, CISE Department, Gainesville, FL, United States
François-Xavier Vialard
Laboratoire d’informatique Gaspard Monge, Université Paris-Est Marne-la-Vallée, UMR CNRS
8049, Champs sur Marne, France
Jared Vicory
UNC, Chapel Hill, NC, United States
Jiyao Wang
UNC, Chapel Hill, NC, United States
William M. Wells III
Harvard Medical School, Department of Radiology, Boston, MA, United States
Miaomiao Zhang
Washington University in St. Louis, Computer Science and Engineering, St. Louis, MO,
United States
Ruiyi Zhang
Florida State University, Tallahassee, FL, United States
Introduction xvii
Introduction
Introduction
Over the last two decades, there has been a growing need in
the medical image computing community for principled meth-
ods to process nonlinear geometric data. Typical examples of data
in this domain include organ shapes and deformations resulting
from segmentation and registration in computational anatomy,
and symmetric positive definite matrices in diffusion imaging.
In this context, Riemannian geometry has gradually been estab-
lished as one the most powerful mathematical and computational
paradigms.
This book aims at being an introduction to and a reference
on Riemannian geometric statistics and its use in medical image
analysis for researchers and graduate students. The book provides
both descriptions of the core methodology and presentations of
state-of-the-art methods used in the field. We wish to present this
combination of foundational material and current research to-
gether with examples, applications, and algorithms in a volume
that is edited and authored by the leading researchers in the field.
In addition, we wish to provide an overview of current research
challenges and future applications.
Beyond medical image computing, the methods described in
this book may also apply to other domains such as signal process-
ing, computer vision, geometric deep learning, and other domains
where statistics on geometric features appear. As such, the pre-
sented core methodology takes its place in the field of geometric
statistics, the statistical analysis of data being elements of nonlin-
ear geometric spaces. We hope that both the foundational mate-
rial and the advanced techniques presented in the later parts of
the book can be useful in domains outside medical imaging and
present important applications of geometric statistics methodol-
ogy.
Contents
Part 1 of this edited volume describes the foundations of Rie-
mannian geometric computing methods for statistics on mani-
folds. The book here emphasizes concepts rather than proofs with
the goal of providing graduate students in computer science the
xviii Introduction
The editors:
Xavier Pennec
University Côte d’Azur and Inria, Sophia Antipolis, France
Stefan Sommer
DIKU, University of Copenhagen, Copenhagen, Denmark
Tom Fletcher
University of Virginia, Charlottesville, VA, United States
February, 2019
Introduction to differential and
1
Riemannian geometry
Stefan Sommera , Tom Fletcherb , Xavier Pennecc
a University of Copenhagen, Department of Computer Science, Copenhagen,
1.1 Introduction
When data exhibit nonlinearity, the mathematical description
of the data space must often depart from the convenient linear
structure of Euclidean vector spaces. Nonlinearity prevents global
vector space structure, but we can nevertheless ask which mathe-
matical properties from the Euclidean case can be kept while still
preserving the accurate modeling of the data. It turns out that in
many cases, local resemblance to a Euclidean vector space is one
such property. In other words, up to some approximation, the data
space can be linearized in limited regions while forcing a linear
model on the entire space would introduce too much distortion.
The concept of local similarity to Euclidean spaces brings us
exactly to the setting of manifolds. Topological, differential, and
Riemannian manifolds are characterized by the existence of local
maps, charts, between the manifold and a Euclidean space. These
charts are structure preserving: They are homeomorphisms in the
case of topological manifolds, diffeomorphisms in the case of dif-
ferential manifolds, and, in the case of Riemannian manifolds,
they carry local inner products that encode the non-Euclidean ge-
ometry.
The following sections describe these foundational concepts
and how they lead to notions commonly associated with geome-
try: curves, length, distances, geodesics, curvature, parallel trans-
port, and volume form. In addition to the differential and Rieman-
nian structure, we describe one extra layer of structure, Lie groups
that are manifolds equipped with smooth group structure. Lie
groups and their quotients are examples of homogeneous spaces.
The group structure provides relations between distant points on
the group and thereby additional ways of constructing Rieman-
nian metrics and deriving geodesic equations.
Riemannian Geometric Statistics in Medical Image Analysis 3
https://doi.org/10.1016/B978-0-12-814725-2.00008-X
Copyright © 2020 Elsevier Ltd. All rights reserved.
4 Chapter 1 Introduction to differential and Riemannian geometry
1.2 Manifolds
A manifold is a collection of points that locally, but not glob-
ally, resembles Euclidean space. When the Euclidean space is of
finite dimension, we can without loss of generality relate it to Rd
for some d > 0. The abstract mathematical definition of a man-
ifold specifies the topological, differential, and geometric struc-
ture by using charts, maps between parts of the manifold and Rd ,
and collections of charts denoted atlases. We will discuss this con-
struction shortly, however, we first focus on the case where the
manifold is a subset of a larger Euclidean space. This viewpoint is
often less abstract and closer to our natural intuition of a surface
embedded in our surrounding 3D Euclidean space.
Let us exemplify this by the surface of the earth embedded in
R3 . We are constrained by gravity to live on the surface of the earth.
This surface seems locally flat with two dimensions only, and we
use two-dimensional maps to navigate the surface. When travel-
ing far, we sometimes need to change from one map to another.
We then find charts that overlap in small parts, and we assume
that the charts provide roughly the same view of the surface in
those overlapping parts. For a long time, the earth was even con-
sidered to be flat because its curvature was not noticeable at the
scale at which it was observed. When considering the earth sur-
face as a two-dimensional restriction of the 3D ambient space, the
surface is an embedded submanifold of R3 . On the other hand,
when using maps and piecing the global surface together using
the compatibility of the overlapping parts, we take the abstract
view using charts and atlases.
the earth. The sphere with radius 1 can be described as the set of
unit vectors in R3 , that is, the set
S2 = {(x 1 , x 2 , x 3 ) ∈ R3 | (x 1 )2 + (x 2 )2 + (x 3 )2 = 1} . (1.1)
Notice from the definition of the set that all points of S2 satisfy the
equation (x 1 )2 + (x 2 )2 + (x 3 )2 − 1 = 0. We can generalize this way
of constructing a manifold to the following definition.
Courage, sire,
Is constant industry for happiness.
When I become a monk——
CHILPERIC.
Nay, no confession,
No putting reasons to your Overlord.
[to his nobles]
You need not shake your spears so stormily,
We leave you a stout leader for your wars,
[to Carloman] And you, your liberty. What use of it
You make is of no moment to the world,
And does not raise my curiosity,
Who for myself have found in meat and drink,
In sleep and long, long abstinence from care
The pleasure proper to me. Pepin, come!
[Enter Geneviva.]
My wife!
Is Geneviva come to me?
GENEVIVA.
Now the dull monk has left you. Rouse your head!
I have been taking thought how best to trim
My beauty for you. Boniface was slow
In giving counsel; slowly I took up,
Handled and dropt my jewels. Of a sudden,
When Pepin’s voice was heard upon the stair,
I laid these blossoms in a ruddy knot
Thus hasty on my bosom. Come to me.
My lord, you owe me many hours of love,
So many hours I have been beautiful
In vain. You do not see me when I sing,
You miss the marks of music in my face,
You do not love the hunt, and you have never
Ridden beside me in the morning light.
You see me but as now when I am vexed
And haughty for caresses.
CARLOMAN.
GENEVIVA.
CARLOMAN.
GENEVIVA.
Assuredly, but the cold font has left
No chill upon my heart. Think not of that,
Think of our marriage-day. You leave me lonely
While Boniface enthralls you.
CARLOMAN.
GENEVIVA.
[Carloman folds her in his arms quietly. Then with great effort bends over
her and speaks]
CARLOMAN.
Marcomir
Is restless for a pilgrimage to Rome.
I think we shall be starting presently:
And afterward ... If I am long away ...
GENEVIVA.
[breaking from him]
Oh, think a little! Can you leave this hair
So crisp and burnished? When the sun is bright
Across your shield, it has no livelier flash—
Confess, it has not? But you come to me
Stale, weary from your dreams and abstinence,
And tingle my suspicion.
CARLOMAN.
If these dreams
Were growing all the world to me!—You start,
You turn away, you will not understand.
The fear of hurting you has made me keep
So distant from you lately, and my eyes
You thought were worn with vigil and with books
Have burnt with tears at night for many a month
To think you have not known the tyrant-joy
That moves a soul to change and severance,
Except upon the day when for my sake
You parted from your home: but by the rapture
That made such tumult in the daughter’s grief
When she became a bride, your husband now
Implores your comprehension.
All thou hast,
So the Church teaches, family and spouse,
The child thou hast begotten, thine own life
Thou must abhor, if thou would’st have new days
Of blessing on the Earth. I feel this law
Is written in my very heart of hearts,
There is such haunting freshness deep below
The sorrow of farewell.
GENEVIVA.
[defiantly] My God is Love—
The God who made a bower in Paradise,
Who wedded Eve and Adam, who abode
In the sweet incense of His Church to bless
My marriage.
[Carloman stretches out his hand to support her.]
Have no fear that I shall fall,
I cannot swoon while I remember it—
How in the songful hush a restless hand
Grew tight about my fingers, and a vow
Thrilled all the girl in me to womanhood,
And stung the future lying at my heart
To joy and frankness. That was years ago ...
[She breaks into a bitter laugh]
O Carloman, you know not what you do,
You know not what I am, nor what a blank
Of mercy there is in you!
CARLOMAN.
Were I dead,
You would not be so violent: in a trance
Of resignation you would think of me,
With tears, not gasping laughter.
GENEVIVA.
[pacing the room excitedly] Pilgrimage!
Did you say, pilgrimage? To think of you
Growing each day more cramped about the mouth,
More full of resolution in the eyes.
What shall I do? Pray for you—but the dead,
You have just told me, should be left unmourned,
Forgotten as last summer’s autumn-leaves.
[facing him coldly] My lord, I am no reliquary-urn;
There is no widow in me.
CARLOMAN.
GENEVIVA.
CARLOMAN.
No, Geneviva. I have little speech;
But when the secret crept into my soul
I loved you, it was not to Marcomir
I spoke: and if another secret now
Is breaking through my nature, do not think
That he will be the spokesman.
[noticing her agitation] Hermann died
I think by his own hand; he courted death.
What can a man prize in captivity?
[as Geneviva grows more agitated]
There! I will speak no more of him. Your maids—
[turning to summon her attendants].
GENEVIVA.
[She looks after Carloman, who walks out, stroking his chin].
To think he dared
To lean above me with those burning eyes
Unconscious what they glassed. I did not learn
From him the magic that was born in me,
I learnt it when great Hermann passed in chains,
And he is dead. I promised I would go
To-day and visit him. How could he die?
[Marcomir enters.]
Why, you are deadly pale!
[She recoils, and says in a faint voice]
It is the hour
Fixed for our visit.
MARCOMIR.
GENEVIVA.
MARCOMIR.
GENEVIVA.
He lies in chains?
Are the brows restful?
MARCOMIR.
GENEVIVA.
MARCOMIR.
[facing her again] Oh, a lifetime, if
It please you! I am going to a place
Where love is held of little consequence.
GENEVIVA.
MARCOMIR.
GENEVIVA.
MARCOMIR.
To any coast you have not trod, wherever
The flowers are different from the flowers you wear,
To some Italian convent. Geneviva,
I am not framed to see you minister
To other men; but when long years are passed,
It may be in a fresco, I shall find
Some figure of a lady breaking bread
To mendicants, and kneel and pray to her
That she may bless me also: but till then ...
[covering his eyes]
O God, you shall not tempt me, though I feel
Just how your hair burns in a fiery wreath
Above your brow, and how your eyes are soft
With blue, and deeper blue, as through the hills
The valley stretches azure to the close.
You shall not tempt me, though I almost hear
Your bosom taking record of your breath,
And I could sit and watch that tide of life
Rising and falling through the lovely curves,
Till I was lost in ecstasy.
GENEVIVA.
Oh, hush!
But then you love me. It was in a fit ...?
MARCOMIR.
Of devilish malice.
GENEVIVA.
In a jealous fit?
You shall remain.
[She goes up to him: he takes her hands in his, kisses them coldly, and puts
them away.]
MARCOMIR.
GENEVIVA.
MARCOMIR.
True, I swerved;
I have confessed my sin, and now must bear
The settling of my spirit on the Cross.
GENEVIVA.
So many favours!
MARCOMIR.
GENEVIVA.
MARCOMIR.
Again, again.
GENEVIVA.
MARCOMIR.
He forgets;
But, Geneviva, if a thousand years
Broke over me, when Time had cleared his storms
I should look up and know your face by heart.
GENEVIVA.
MARCOMIR.
[bitterly, as he turns away] A soul so wide
In innocence, so regal, on the day
He wedded, he appointed me your squire!
GENEVIVA.
[following him]
He keeps you with him, you can read his heart,
You know what way he travels, when his soul
Flies homeward. Tell me—’tis the only knowledge
I crave for in the world—does Carloman
Still hold me in affection? I beseech,
Tell me the truth. He loves you——
MARCOMIR.
Yes, he loves,
He does not use me for his purposes.
[perceiving PEPIN]
Not Carloman—his brother on the stair
Laughs at your light behaviour. So you lose
One last poor opportunity.
[Re-enter Pepin.]
PEPIN.
Good even.
Well, my fair sister, you have heard the news,
Wept [glancing at Marcomir]
and found consolation.
But to think
The son of Charles Martel should be a monk!
GENEVIVA.
A monk!—a pilgrim?
PEPIN.
MARCOMIR.
PEPIN.
Oh, no impiety;
A crazy fit: he must get near to God,
So puts away all intercourse with man:
And while I rule he thinks to thrill the world
With some convulsive movement from his prayers.
Ha, ha! But you shall queen it as before.
GENEVIVA.
[Exit Pepin.]
[turning to Marcomir]
You are a murderer: this act of yours
Will leave me very lonely.
MARCOMIR.
I repent.
GENEVIVA.
[Carloman and Boniface cross from right to left at the back of the hall.
Geneviva intercepts them.]
Farewell!
CARLOMAN.
[arrested] O Geneviva!
GENEVIVA.
Not my name,
Never my name again. Say, holy father—
They take new titles who renounce the world?
CARLOMAN.
[with flushing eagerness]
Then you too will renounce it? oh, the joy!
There is a strange new passion in your eyes.
Speak to me ... but you cannot! I could take
No leave of you in your fierce, worldly mood;
Now all is changed.
GENEVIVA.
CARLOMAN.
BONIFACE.
Do you choose,
Lady, a mere retreat among the nuns,
Or, like your husband, do you break all ties
That bind you to the earth?
GENEVIVA.
CARLOMAN.
[nervously]
Pepin will guard him.
GENEVIVA.
MARCOMIR.
PEPIN.
Come, an end to this!
Brother, if you are wise you will not leave
This woman in the world. Convents are made
To tame the pride of such and keep them cool.
CARLOMAN.
Marcomir, farewell!
You will be monks together. When my husband
Forgets me, you must bring me to his thoughts
Recall that day we hunted and you fell;
I stayed to tend you; but the whole live day
My voice rang through the woods for Carloman
Until I wearied you; he was not found;
But you remember how I cried for him.
MARCOMIR.
GENEVIVA.
O insolence!—
The virginal chill heart!—No intercession!
[to Carloman]
Our marriage is dissolved. How great a stranger
You have become to me! I should grow mad
To breathe by you another single hour.
[to Boniface]
And you, old man, who stand with such meek eyes,
Though you have robbed me of my name of wife,
And made my boy an orphan—go your way!
I cannot curse you, but I prophesy:
Dishonour motherhood, plant virgin homes,
Give to religion the sole charge of love,
And you will rear up lust of such an ice
As Death himself will shiver at.
[to Pepin] Lead on!
Now there is hope you may become a King,
There should be some high festival to keep
To-night in everlasting memory.
Lead me away.
PEPIN.
The thing to do
Is simply just the sole thing to be done.
There should have been no tears, no taking leave;
A freeman can do anything he will.