Multidimensional scaling 2nd ed Edition Trevor F. Cox

Multidimensional scaling 2nd ed Edition Trevor
F. Cox pdf download
https://ebookname.com/product/multidimensional-scaling-2nd-ed-
edition-trevor-f-cox/
Get Instant Ebook Downloads – Browse at https://ebookname.com

Instant digital products (PDF, ePub, MOBI) available
Download now and explore formats that suit you...
Scaling Methods 2nd ed Edition Peter Dunn-Rankin
https://ebookname.com/product/scaling-methods-2nd-ed-edition-
peter-dunn-rankin/
Essentials of sports nutrition 2nd ed Edition F Brouns
https://ebookname.com/product/essentials-of-sports-nutrition-2nd-
ed-edition-f-brouns/
Community Based Adaptation to Climate Change Scaling it
up 1st Edition E. Lisa F. Schipper
https://ebookname.com/product/community-based-adaptation-to-
climate-change-scaling-it-up-1st-edition-e-lisa-f-schipper/
Prion Diseases and Copper Metabolism Bse Scrapie and
Cjd Research 1st Edition D. Brown (Auth.)
https://ebookname.com/product/prion-diseases-and-copper-
metabolism-bse-scrapie-and-cjd-research-1st-edition-d-brown-auth/

Matthew James and Didache Hubertus Waltherus Maria Van
De Sandt
https://ebookname.com/product/matthew-james-and-didache-hubertus-
waltherus-maria-van-de-sandt/
Mr Timothy First Edition. 1St Edition Louis Bayard
https://ebookname.com/product/mr-timothy-first-edition-1st-
edition-louis-bayard/
Yoshio Nishina Father of Modern Physics in Japan 1st
Edition Dong-Won Kim
https://ebookname.com/product/yoshio-nishina-father-of-modern-
physics-in-japan-1st-edition-dong-won-kim/
The Neurocognition of Language 1st Edition Colin M.
Brown
https://ebookname.com/product/the-neurocognition-of-language-1st-
edition-colin-m-brown/
Economic and Social Transformation in China Challenges
and Opportunities Routledgecurzon Studies on the
Chinese Economy 1st Edition Angang Hu
https://ebookname.com/product/economic-and-social-transformation-
in-china-challenges-and-opportunities-routledgecurzon-studies-on-
the-chinese-economy-1st-edition-angang-hu/

Firewall Policies And VPN Configurations 1st Edition
Laura E. Hunter
https://ebookname.com/product/firewall-policies-and-vpn-
configurations-1st-edition-laura-e-hunter/

MONOGRAPHS ON STATISTICS AND APPLIED PROBABILITY
General Editors
D.R. Cox, V. Isham, N. Keiding, T. Louis, N. Reid, R. Tibshirani, and H. Tong
1 Stochastic Population Models in Ecology and Epidemiology M.S. Barlett (1960)
2 Queues D.R. Cox and W.L. Smith (1961)
3 Monte Carlo Methods J.M. Hammersley and D.C. Handscomb (1964)
4 The Statistical Analysis of Series of Events D.R. Cox and P.A.W. Lewis (1966)
5 Population Genetics W.J. Ewens (1969)
6 Probability, Statistics and Time M.S. Barlett (1975)
7 Statistical Inference S.D. Silvey (1975)
8 The Analysis of Contingency Tables B.S. Everitt (1977)
9 Multivariate Analysis in Behavioural Research A.E. Maxwell (1977)
10 Stochastic Abundance Models S. Engen (1978)
11 Some Basic Theory for Statistical Inference E.J.G. Pitman (1979)
12 Point Processes D.R. Cox and V. Isham (1980)
13 Identification of Outliers D.M. Hawkins (1980)
14 Optimal Design S.D. Silvey (1980)
15 Finite Mixture Distributions B.S. Everitt and D.J. Hand (1981)
16 Classification A.D. Gordon (1981)
17 Distribution-Free Statistical Methods, 2nd edition J.S. Maritz (1995)
18 Residuals and Influence in Regression R.D. Cook and S. Weisberg (1982)
19 Applications of Queueing Theory, 2nd edition G.F. Newell (1982)
20 Risk Theory, 3rd edition R.E. Beard, T. Pentikäinen and E. Pesonen (1984)
21 Analysis of Survival Data D.R. Cox and D. Oakes (1984)
22 An Introduction to Latent Variable Models B.S. Everitt (1984)
23 Bandit Problems D.A. Berry and B. Fristedt (1985)
24 Stochastic Modelling and Control M.H.A. Davis and R. Vinter (1985)
25 The Statistical Analysis of Composition Data J. Aitchison (1986)
26 Density Estimation for Statistics and Data Analysis B.W. Silverman (1986)
27 Regression Analysis with Applications G.B. Wetherill (1986)
28 Sequential Methods in Statistics, 3rd edition
G.B. Wetherill and K.D. Glazebrook (1986)
29 Tensor Methods in Statistics P. McCullagh (1987)
30 Transformation and Weighting in Regression
R.J. Carroll and D. Ruppert (1988)
31 Asymptotic Techniques for Use in Statistics
O.E. Bandorff-Nielsen and D.R. Cox (1989)
32 Analysis of Binary Data, 2nd edition D.R. Cox and E.J. Snell (1989)
© 2001 by Chapman & Hall/CRC

33 Analysis of Infectious Disease Data N.G. Becker (1989)
34 Design and Analysis of Cross-Over Trials B. Jones and M.G. Kenward (1989)
35 Empirical Bayes Methods, 2nd edition J.S. Maritz and T. Lwin (1989)
36 Symmetric Multivariate and Related Distributions
K.T. Fang, S. Kotz and K.W. Ng (1990)
37 Generalized Linear Models, 2nd edition P. McCullagh and J.A. Nelder (1989)
38 Cyclic and Computer Generated Designs, 2nd edition
J.A. John and E.R. Williams (1995)
39 Analog Estimation Methods in Econometrics C.F. Manski (1988)
40 Subset Selection in Regression A.J. Miller (1990)
41 Analysis of Repeated Measures M.J. Crowder and D.J. Hand (1990)
42 Statistical Reasoning with Imprecise Probabilities P. Walley (1991)
43 Generalized Additive Models T.J. Hastie and R.J. Tibshirani (1990)
44 Inspection Errors for Attributes in Quality Control
N.L. Johnson, S. Kotz and X, Wu (1991)
45 The Analysis of Contingency Tables, 2nd edition B.S. Everitt (1992)
46 The Analysis of Quantal Response Data B.J.T. Morgan (1992)
47 Longitudinal Data with Serial Correlation—A state-space approach
R.H. Jones (1993)
48 Differential Geometry and Statistics M.K. Murray and J.W. Rice (1993)
49 Markov Models and Optimization M.H.A. Davis (1993)
50 Networks and Chaos—Statistical and probabilistic aspects
O.E. Barndorff-Nielsen, J.L. Jensen and W.S. Kendall (1993)
51 Number-Theoretic Methods in Statistics K.-T. Fang and Y. Wang (1994)
52 Inference and Asymptotics O.E. Barndorff-Nielsen and D.R. Cox (1994)
53 Practical Risk Theory for Actuaries
C.D. Daykin, T. Pentikäinen and M. Pesonen (1994)
54 Biplots J.C. Gower and D.J. Hand (1996)
55 Predictive Inference—An introduction S. Geisser (1993)
56 Model-Free Curve Estimation M.E. Tarter and M.D. Lock (1993)
57 An Introduction to the Bootstrap B. Efron and R.J. Tibshirani (1993)
58 Nonparametric Regression and Generalized Linear Models
P.J. Green and B.W. Silverman (1994)
59 Multidimensional Scaling T.F. Cox and M.A.A. Cox (1994)
60 Kernel Smoothing M.P. Wand and M.C. Jones (1995)
61 Statistics for Long Memory Processes J. Beran (1995)
62 Nonlinear Models for Repeated Measurement Data
M. Davidian and D.M. Giltinan (1995)
63 Measurement Error in Nonlinear Models
R.J. Carroll, D. Rupert and L.A. Stefanski (1995)
64 Analyzing and Modeling Rank Data J.J. Marden (1995)
65 Time Series Models—In econometrics, finance and other fields
D.R. Cox, D.V. Hinkley and O.E. Barndorff-Nielsen (1996)

66 Local Polynomial Modeling and its Applications J. Fan and I. Gijbels (1996)
67 Multivariate Dependencies—Models, analysis and interpretation
D.R. Cox and N. Wermuth (1996)
68 Statistical Inference—Based on the likelihood A. Azzalini (1996)
69 Bayes and Empirical Bayes Methods for Data Analysis
B.P. Carlin and T.A Louis (1996)
70 Hidden Markov and Other Models for Discrete-Valued Time Series
I.L. Macdonald and W. Zucchini (1997)
71 Statistical Evidence—A likelihood paradigm R. Royall (1997)
72 Analysis of Incomplete Multivariate Data J.L. Schafer (1997)
73 Multivariate Models and Dependence Concepts H. Joe (1997)
74 Theory of Sample Surveys M.E. Thompson (1997)
75 Retrial Queues G. Falin and J.G.C. Templeton (1997)
76 Theory of Dispersion Models B. Jørgensen (1997)
77 Mixed Poisson Processes J. Grandell (1997)
78Variance Components Estimation—Mixed models, methodologies and applications
P.S.R.S. Rao (1997)
79 Bayesian Methods for Finite Population Sampling
G. Meeden and M. Ghosh (1997)
80 Stochastic Geometry—Likelihood and computation
O.E. Barndorff-Nielsen, W.S. Kendall and M.N.M. van Lieshout (1998)
81 Computer-Assisted Analysis of Mixtures and Applications—
Meta-analysis, disease mapping and others D. Böhning (1999)
82 Classification, 2nd edition A.D. Gordon (1999)
83 Semimartingales and their Statistical Inference B.L.S. Prakasa Rao (1999)
84 Statistical Aspects of BSE and vCJD—Models for Epidemics
C.A. Donnelly and N.M. Ferguson (1999)
85 Set-Indexed Martingales G. Ivanoff and E. Merzbach (2000)
86 The Theory of the Design of Experiments D.R. Cox and N. Reid (2000)
87 Complex Stochastic Systems
O.E. Barndorff-Nielsen, D.R. Cox and C. Klüppelberg (2001)
88 Multidimensional Scaling, 2nd edition T.F. Cox and M.A.A. Cox (2001)

CHAPMAN & HALL/CRC
Multidimensional
Scaling
TREVOR F. COX
Senior Lecturer in Statistics
University of Newcastle Upon Tyne, UK
AND
MICHAEL A. A. COX
Lecturer in Business Management
University of Newcastle Upon Tyne, UK
SECOND EDITION
Boca Raton London New York Washington, D.C.

This book contains information obtained from authentic and highly regarded sources. Reprinted material
is quoted with permission, and sources are indicated. A wide variety of references are listed. Reasonable
efforts have been made to publish reliable data and information, but the author and the publisher cannot
assume responsibility for the validity of all materials or for the consequences of their use.
Neither this book nor any part may be reproduced or transmitted in any form or by any means, electronic
or mechanical, including photocopying, microfilming, and recording, or by any information storage or
retrieval system, without prior permission in writing from the publisher.
The consent of CRC Press LLC does not extend to copying for general distribution, for promotion, for
creating new works, or for resale. Specific permission must be obtained in writing from CRC Press LLC
for such copying.
Direct all inquiries to CRC Press LLC, 2000 N.W. Corporate Blvd., Boca Raton, Florida 33431.
Trademark Notice: Product or corporate names may be trademarks or registered trademarks, and are
used only for identification and explanation, without intent to infringe.
No claim to original U.S. Government works
International Standard Book Number 1-58488-094-5
Library of Congress Card Number 00-060180
Printed in the United States of America 2 3 4 5 6 7 8 9 0
Printed on acid-free paper
Library of Congress Cataloging-in-Publication Data
Cox, Trevor F.
Multidimensional scaling / Trevor F. Cox, Michael A.A. Cox.--2nd ed.
p. cm. -- (Monographs on statistics and applied probability ; 88)
Includes bibliographical references and indexes.
ISBN 1-58488-094-5 (alk. paper)
1. Multivariate analysis. 2. Multideminsional scaling. I. Cox, Michael A.A. II. Title.
III. Series.
QA278 .C7 2000
519.5′35--dc21 00-060180
CIP
Visit the CRC Press Web site at www.crcpress.com

Contents
Preface
1 Introduction
1.1 Introduction
1.2 A look at data and models
1.2.1 Types of data
1.2.2 Multidimensional scaling models
1.3.1 Similarity/dissimilarity coefficients
1.3.2 Distribution of proximity coefficients
1.3.3 Similarity of species populations
1.3.4 Transforming from similarities
to dissimilarities
1.3.5 The metric nature of dissimilarities
1.3.6 Dissimilarity of variables
1.3.7 Similarity measures on fuzzy sets
1.4.1 The spectral decomposition
1.4.2 The singular value decomposition
1.4.3 The Moore-Penrose inverse
1.3 Proximities
for mixed data
1.4 Matrix results
2 Metric multidimensional scaling
2.1 Introduction
2.2 Classical scaling
2.2.1 Recovery of coordinates
2.2.2 Dissimilarities as Euclidean distances
2.2.3 Classical scaling in practice
2.2.4 How many dimensions?
2.2.5 A practical algorithm for classical scaling
2.2.7 Classical scaling and principal components
2.2.6 A grave example

2.3
2.4
2.5
2.6
2.7
2.8
2.2.8 The additive constant problem
Robustness
Metric least squares scaling
Critchley’s intermediate method
Unidimensional scaling
2.6.1 A classic example
Grouped dissimilarities
Inverse scaling
3 Nonmetric multidimensional scaling
3.1 Introduction
3.1.1 Rp
space and the Minkowski metric
3.2 Kruskal’s approach
3.2.1 Minimising S with respect to the disparities
3.2.2
3.2.3 Kruskal’s iterative technique
3.2.4
3.2.5 STRESS1/2, monotonicity, ties and
A configuration with minimum stress
Nonmetric scaling of breakfast cereals
missing data
3.3 The Guttman approach
3.4 A further look at stress
3.4.1 Interpretation of stress
3.5 How many dimensions?
3.6 Starting configurations
3.7 Interesting axes in the configuration
4 Further aspects of multidimensional scaling
4.1 Other formulations of MDS
4.2 MDS Diagnostics
4.3 Robust MDS
4.4 Interactive MDS
4.5 Dynamic MDS
4.6 Constrained MDS
4.6.1 Spherical MDS
4.7 Statistical inference for MDS
4.8 Asymmetric dissimilarities
5 Procrustes analysis
5.1 Introduction
5.2 Procrustes analysis
5.2.1 Procrustes analysis in practice

5.2.2 The projection case
5.3 Historic maps
5.4 Some generalizations
5.4.1 Weighted Procrustes rotation
5.4.2 Generalized Procrustes analysis
5.4.3 The coefficient of congruence
5.4.4 Oblique Procrustes problem
5.4.5 Perturbation analysis
6 Monkeys, whisky and other applications
6.1 Introduction
6.2 Monkeys
6.3 Whisky
6.4 Aeroplanes
6.5 Yoghurts
6.6 Bees
7 Biplots
7.1 Introduction
7.2 The classic biplot
7.2.1 An example
7.2.2 Principal component biplots
7.3 Another approach
7.4 Categorical variables
8 Unfolding
8.1 Introduction
8.2 Nonmetric unidimensional unfolding
8.3 Nonmetric multidimensional unfolding
8.4 Metric multidimensional unfolding
8.4.1 The rating of nations
9 Correspondence analysis
9.1 Introduction
9.2 Analysis of two-way contingency tables
9.2.1 Distance between rows (columns)
in a contingency table
The theory of correspondence analysis
9.3.1 The cancer example
9.3.2 Inertia
9.3

9.4 Reciprocal averaging
9.4.1 Algorithm for solution
9.4.2 An example: the Munsingen data
9.4.3 The whisky data
9.4.4 The correspondence analysis connection
9.4.5 Two-way weighted dissimilarity coefficients
9.5 Multiple correspondence analysis
9.5.1 A three-way example
10 Individual differences models
10.1 Introduction
10.2 The Tucker-Messick model
10.3 INDSCAL
10.3.1 The algorithm for solution
10.3.2 Identifying groundwater populations
10.3.3 Extended INDSCAL models
10.4 IDIOSCAL
10.5 PINDIS
11 ALSCAL, SMACOF and Gifi
11.1 ALSCAL
11.1.1 The theory
11.1.2 Minimising SSTRESS
11.2.1 The majorization algorithm
11.2.2 The majorizing method for
11.2.3
11.3.1 Homogeneity
11.2 SMACOF
nonmetric MDS
Tunnelling for a global minimum
11.3 Gifi
12 Further m-mode, n-way models
12.1 CANDECOMP, PARAFAC and CANDELINC
12.2 DEDICOM and GIPSCAL
12.3 The Tucker models
12.4 One-mode, n-way models
12.5 Two-mode, three-way asymmetric scaling
12.6 Three-way unfolding
12.3.1 Relationship to other models

Appendix: Computer programs for
multidimensional scaling
A.l Computer programs
A.2 The accompanying CD-ROM
A.2.1
A.2.2 Data and output
A.2.3 To run the menu
A.2.4 Program descriptions
To manipulate and analyse data
A.5.1 Data format
Installation instructions
A.3 The data provided
A.4
A.5 Inputting user data
A.6 Error messages
References

Preface
It has been a pleasure for us to write the second edition of this
book on multidimensional scaling. The second edition extends the
first with recent references, a new chapter on biplots, a section on
the Gifi system of nonlinear multivariate analysis and an extended
version of the suite of computer programs.
Multidimensional scaling covers a variety of techniques, with its
main development having rested in the hands of mathematical psy-
chologists and the journal Psychometrika having championed the
publication of articles in the subject. Multidimensional scaling has
now become popular and has extended into areas other than its
traditional place in the behavioural sciences. Many statistical com-
puter packages now include multidimensional scaling.
The book has a review style to it which has been necessitated
in attempting to cover several areas, but wanting to keep the size
of the book of manageable proportions. The techniques covered
have been applied to interesting data sets, hopefully giving insight
into the data and the application of the theories. We hope readers
will try out some of the techniques themselves, using the suite of
computer programs provided. These run under DOS or Windows;
a full Windows version will be available by the end of 2000.
Again, in this edition, we thank the many authors who have
contributed to the theory of multidimensional scaling - not just
the giants of the subject, Arabie, Benzécri, Carroll, Coombs, de
Leeuw, Gower, Greenacre, Groenen, Guttman, Harshman, Heiser,
Hubert, Kiers, Kroonenberg, Kruskal, Meulman, Ramsay, Schöne-
mann, Shepard, Sibson, Takane, ten Berge, Torgerson, van der
Heijden and Young, but every one of them. For without them,
this book would not exist. Also, we would like to thank those who
pointed out errors in the first edition, especially Jos ten Berge.
Newcastle upon Tyne
June 2000
Trevor F. Cox
Michael A. A. Cox

CHAPTER 1
Introduction
1.1 Introduction
Suppose a set of n objects is under consideration and between each
pair of objects (r,s) there is a measurement S
,
, of the “dissimilar-
ity” between the two objects. For example the set of objects might
be ten bottles of whisky, each one from a different distillery. The
dissimilarity S
,
, might be an integer score between zero and ten
given to the comparison of the rth and sth whiskies by an expert
judge of malt whisky. The judge would be given a tot from the
rth bottle and one from the sth and then score the comparison:
0-the whiskies are so alike she cannot tell the difference, to 10-the
whiskies are totally different. The judge is presented with all forty-
five possible pairs of whiskies, and after a pleasant day’s work,
provides the data analyst with a total set of dissimilarities {
S
,
,
}
.
Indeed Lapointe and Legendre (1994) understand the importance
of a proper statistical comparison of whiskies, using data from a
connoisseur’s guide to malt whiskies written by Jackson (1989). In
the same spirit, two much smaller analyses of whiskies are given in
Chapters 6 and 9.
A narrow definition of multidimensional scaling (often abbre-
viated to MDS) is the search for a low dimensional space, usu-
ally Euclidean, in which points in the space represent the objects
(whiskies), one point representing one object, and such that the
distances between the points in the space, {dTS},
match, as well as
possible, the original dissimilarities {
S
,
,
}
. The techniques used for
the search for the space and the associated configuration of points
form metric and nonmetric multidimensional scaling.
An example
A classic way to illustrate multidimensional scaling is to use jour-
ney times between a set of cities in order to reconstruct a map of
the cities. Greenacre and Underhill (1982) use flying times between

Southern African airports, Mardia et al. (1979) use road distances
between some British cities.
For illustration here, the journey times by road between twelve
British cities were subjected to multidimensional scaling, using
classical scaling, which is described fully in Chapter 2.
shows the configuration of points produced by the technique. There
is a striking similarity between the positions of the points repre-
senting the cities and the positions of the same cities seen in a
geographical map of Great Britain, except of course the cities in
appear to be reflected about a line and rotated from the
geographical map usually presented in an atlas.
BRIG
LOND
NOTT
OXFO
EXET
ABER
NEWC
9
&
P
LIVE
INVE
Figure 1.1 A map of British cities reconstitutedfromjourney time by road.
ABER - Aberystwyth, BRIG - Brighton, EDIN - Edinburgh, EXET - Ex-
eter, GLAS - Glasgow, INVE - Inverness, LIVE - Liverpool, LOND - Lon-
don, NEWC - Newcastle, NOTT - Nottingham, OXFO - Oxford, STRA -
Strathe1yde.
Multidimensional scaling is not only about reconstructing maps,
but can be used on a wide range of dissimilarities arising from
various situations, as for example, the whisky tasting experiment
or other situations as described later in the chapter.
A wider definition of multidimensional scaling can subsume sev-
eral techniques of multivariate data analysis. At the extreme, it
Figure 1.1
Figure 1.1

covers any technique which produces a graphical representation
of objects from multivariate data. For example the dissimilarities
obtained from the whisky comparisons could be used in a clus-
ter analysis to find groups of similar whiskies. This text does not
attempt to cover all these possibilities, as there are many books
covering multivariate data analysis in general, for example Mardia
et al. (1979), Chatfield and Collins (1980), Krzanowski (1988)and
Krzanowski and Marriott (1994, 1995). The aim here is to give an
account of the main topics that could be said to constitute the
theory of multidimensional scaling.
Much of the theory of multidimensional scaling was developed
in the behavioural sciences, with Psychometrika publishing many
papers on the subject. It is a tribute to the journal that multidi-
mensional scaling techniques are becoming a popular method of
data analysis, with major statistical software packages now incor-
porating them into their repertoire.
1.2 A look at data and models
Several types of data lend themselves to analysis by multidimen-
sional scaling. Behavioural scientists have adopted several terms
relating to data which often are not familiar to others.
1.2.1 Types of data
Variables can be classified according to their “measurement scale”.
The four scales are the nominal scale, the ordinal scale, the interval
scale and the ratio scale.
Nominal scale
Data measured on the nominal scale are classificatory, and only
different classes are distinguishable, for example, hair colour, eye
colour.
Ordinal scale
Data on the ordinal scale can be ordered, but are not quantitative
data. For instance, whisky from bottle number 3 might be judged
to be of better quality than that from bottle number 7.

Interval scale
Quantitative data where the difference between two values is mean-
ingful are measured on the interval scale. For example, tempera-
ture in degrees Celsius, the difference in pulse rate before and after
exercise.
Ratio scale
Data measured on the ratio scale are similar to those on the in-
terval scale, except that the scale has a meaningful zero point, for
example, weight, height ,temperature recorded in degrees Kelvin.
Multidimensional scaling is carried out on data relating objects,
individuals, subjects or stimuli to one another. These four terms
will often be used interchangeably, although objects usually refers
to inanimate things, such as bottles of whisky, individuals and
subjects refering to people or animals, while stimuli usually refers
to non-tangible entities, such as the taste of a tot of whisky.
The most common measure of the relationship of one object
(stimulus, etc.) to another is a proximity measure. This measures
the “closeness” of one object to another, and can either be a “sim-
ilarity” measure where the similarity of one object to another, sTs,
is measured, or a “dissimilarity” measure where the dissimilarity,
6,,, between the two objects is measured.
Suppose for the whisky tasting exercise, several more judges are
brought in and each one of them compares all the pairs of whiskies.
Then the available data are Srs,i where T , S refer to the bottles of
whisky, and i refers to the ith judge. The situation now comprises
a set of whiskies (stimuli) and a set of judges (subjects).
Number of modes
Each set of objects that underlie the data for multidimensional scal-
ing is called a mode. Thus the dissimilarities STs,i from the whisky
tasting above are two-mode data, one-mode being the whiskies and
the other the judges.
Number of ways
Each index in the measurement between objects etc. is called a
way. So the &,,i above are three-way data.
Thus data for multidimensional scaling are described by their
number of modes and number of ways. With only one whisky judge,
the data are one-mode, two-way, which is the commonest form.

The entries in a two-way contingency table form two-mode, two-
way data. An appropriate method of analysis is correspondence
analysis described in Chapter 9. Another form of two-mode, two-
way data is where n judges each rank m stimuli. These data can be
subjected to unfolding analysis described in Chapter 8. The two-
mode, three-way data obtained from the judges of whisky can be
analysed by individual differences models of Chapter 10. Three-
mode, three-way, or even higher-mode and -way data can be ana-
lysed by using some of the methods described in Chapter 12. Data
with large number of ways and modes are not very common in
practice.
Coombs (1964) gives a classification of types of data. This was
updated by Carroll and Arabie (1980) who classify data and also
classify types of multidimensional scaling analyses. In so doing,
they have constructed a useful review of the area. Other useful
reviews have been given by Greenacre and Underhill (1982), de
Leeuw and Heiser (1982), Wish and Carroll (1982), Gower (1984)
and Mead (1992). An introductory book on multidimensional scal-
ing is Kruskal and Wish (1978). Fuller accounts of the subject are
given by Schiffman et al. (1981), Davidson (1983), Young (1987)
and Borg and Groenen (1997) among others.
This book attempts to cover the main constituents of multidi-
mensional scaling, giving much, but not all, of the mathematical
theory. Also included in the book is a CD-ROM enabling the reader
to try out some of the techniques. Instructions for loading the CD-
ROM and running the programs are given in the appendix.
1.2.2 Multidimensional scaling models
Some models used for multidimensional scaling are outlined before
fuller definition and development in later chapters. The starting
point is one-mode, two-way proximity data, and in particular, dis-
similarity measurements.
Suppose a set of n objects have dissimilarities {dTS} measured
between all pairs of objects. A configuration of n points repre-
senting the objects is sought in a p dimensional space. Each point
represents one object, with the rth point representing object r. Let
the distances, not necessarily Euclidean, between pairs of points be
{dTs}.Then as stated before, the aim of multidimensional scaling
is to find a configuration such that the distances {dTs} “match”,

as well as possible, the dissimilarities {d,,}. It is the different no-
tions of “matching” that give rise to the different techniques of
multidimensional scaling.
Classical scaling
If the distances in the configuration space are to be Euclidean and
d,, = 6
,
, r,s = 1,...,n
so that the dissimilarities are precisely Euclidean distances, then it
is possible to find a configuration of points ensuring the equality in
(1.1).Classical scaling treats dissimilarities {S,,} directly as Euc-
lidean distances and then makes use of the spectral decomposition
of a doubly centred matrix of dissimilarities. The technique is dis-
cussed fully in Chapter 2.
Metric least squares scaling
Least squares scaling matches distances {d,,} to transformed dis-
similarities {f(dTs)}, where f is a continuous monotonic function.
The function f attempts to transform the dissimilarities into dis-
tances whereupon a configuration is found by fitting its associated
distances by least squares to { f ( S r s ) } . For example, a configuration
may be sought such that the loss function
c,
C,(d,.S - (a+P d T d 2
c,
c,
d?s
is minimized where a and ,8 are positive constants which are to be
found.
Classical scaling and least squares scaling are examples of “met-
ric scaling”,where metric refers to the type of transformation of the
dissimilarities and not the space in which a configuration of points
is sought. Critchley’s intermediate method (Critchley, 1978) is
* an-
other example of metric scaling and is also described in the second
chapter.
Unidimensional scaling
A special case of multidimensional scaling occurs when the con-
figuration of points representing the objects is sought in only one
dimension. This is unidimensional scaling. The single dimension
produces an ordering of the objects which can be useful in an anal-
ysis. An example of this is given in Chapter 2 where the technique

is used on classic data relating to the works of Plato. Unidimen-
sional scaling can be plagued with a plethora of local minima when
attempting to minimise the loss function.
Nonrnetric scaling
If the metric nature of the transformation of the dissimilarities
is abandoned, nonmetric multidimensional scaling is arrived at.
The transformation f can now be arbitrary, but must obey the
monotonicity constraint
for all 1IT , s, T ‘ , s‘ In.
Thus only the rank order of the dissimilarities has to be preserved
by the transformation and hence the term nonmetric. Nonmetric
multidimensional scaling is discussed in Chapter 3.
Procrustes analysis
Suppose multidimensional scaling has been carried out on some
dissimilarity data using two different methods giving rise to two
configurations of points representing the same set of objects. A
Procrustes analysis dilates, translates, reflects and rotates one of
the configurations of points to match, as well as possible, the other,
enabling a comparison of the two configurations to be made. Pro-
crustes analysis is covered in Chapter 5.
Bip1ots
Biplots attempt to plot not only a configuration of points repre-
senting objects, but also axes within the plots that represent the
variables upon which the dissimilarities were calculated. In the sim-
plest case, the axes are linear, but with generalization the axes can
be curvilinear. Chapter 7 explains the theory.
Unfo1ding
Suppose n judges of m types of whisky each rank the whiskies in
order of their personal preference. Unfolding attempts to produce
a configuration of points in a space with each point representing
one of the judges, together with another configuration of points in
the same space, these points representing the whiskies. The con-
figurations are sought so that the rank order of the distances from
the ith “judge” point to the “whisky” points, matches, as well as
possible, the original whisky rankings of the ith judge. This is to

be the case for all of the judges. Unfolding analysis is the subject
of Chapter 8.
Correspondence analysis
Data in the form of a two-way contingency table can be analysed
by correspondence analysis. A space is found in which row pro-
files can be represented by points, and similarly, another space is
also found for representing the column profiles. Distances in these
spaces reproduce chi-square distances between row/column pro-
files. Full discussion is given in Chapter 9.
Indvidual differences
Again, if m judges each compare all pairs of whiskies, then either
m separate multidimensional scaling analyses can be carried out
or an attempt can be made at a combined approach. Individual
differences models produce an overall configuration of points rep-
resenting the whiskies in what is called the group stimulus space,
together with a configuration of points representing the judges in a
different space called the subject space. The position of a particular
judge in the subject space depends on the weights needed on the
axes of the stimulus space to transform the configuration of points
in the group stimulus space into the configuration that would have
been peculiar to that judge. Individual differences models are the
subject of Chapter 10.
Gifi
The Gifi system of nonlinear multivariate analysis extends various
techniques, such as principal components analysis. It has links to
multidimensional scaling which are explored in Chapter 11. Mul-
tidimensional scaling based on alternating least squares scaling
(ALSCAL) and by “majorization a complicated function” (SMA-
COF) are also discussed in Chapter 11.
Chapter 12 gives a brief summary of further multidimensional
scaling models that involve data of more than 2 ways and 1mode.
1.3 Proximities
Proximity literally means nearness in space, time or in some other
way. The “nearness” of objects, individuals, stimuli needs definition
and measurement prior to statistical analysis. In some situations,
this is straightforward, but in others, difficult and controversial.

Measures of proximity are of two types: similarity and dissimilar-
ity with the obvious interpretation of measuring how similar or
dissimilar objects are to each other.
Let the objects under consideration comprise a set 0. The sim-
ilarity/dissimilarity measure between two objects is then a real
function defined on 0 x 0,giving rise to similarity s,,, or dissimi-
larity S,, between the rth and sth objects. Usually S,, 20, s,, 20,
and the dissimilarity of an object with itself is taken to be zero,
i.e. S,, = 0. Similarities are usually scaled so that the maximum
similarity is unity, with s,, = 1.
Hartigan (1967) gives twelve possible proximity structures, S,
that might need to be considered before a particular proximity
measure is chosen. These are listed in Cormack (1971) and also
below.
SI
s2
s3
s4
s5
S6
s 7
S8
s 9
SIO
s11
s12
S defined on 0 x 0 is Euclidean distance,
S defined on 0 x 0 is a metric,
S defined on 0 x 0 is symmetric real-valued,
S defined on 0 x 0 is real-valued,
S is a complete ordering 5 on 0 x 0,
S is a partial ordering 5 on 0 x 0,
S is a tree r on 0 (a partial similarity order (r,s) 5 (T’, s’)
whenever sup, (r,s) 2sup, (r’,s‘), see Hartigan or
Cormack for further details),
S is a complete relative similarity ordering 5, on 0;for
each T in 0, s 5, t means s is no more similar to r
than t is,
S is a partial relative similarity order dTon 0,
S is a similarity dichotomy on 0 x 0 in which 0 x 0 is
divided into a set of similar pairs and a set of dissimilar
pairs,
S is a similarity trichotomy on 0 x 0 consisting of similar
pairs, dissimilar pairs, and the remaining pairs,
S is a partition of 0 into sets of similar objects.

Structure SI is a very strict structure with dissimilarity defined
as Euclidean distance. Relaxing this to the requirement of a metric
gives S2, where it is recalled that S,, is a metric if
S
,
, = 0
S
,
, = S,,
S
,
, 5 S,t +St,
if and only if r = s,
for all 15 r , s 5 n,
for all 15 r,s,t 5 n.
Relaxing the metric requirement to S,, being symmetric real-valued
or real-valued gives structures S3 and S4. Losing ratio/interval
scales of measurement of S
,
, leads to the nonmetric structures S5
to S12. Of these the highest structure, S5, has a complete ordering
of the {S,,}. The lowest structure, S12, simply partitions 0 into
sets of similar objects.
Choice of proximity measure depends upon the problem at hand,
and is often not an easy task. Sometimes similarity between two
objects is not based on any underlying data recorded on the ob-
jects. For example, in the whisky tasting exercise, the judge simply
uses taste and smell sensations to produce a score between zero
and ten. The similarity/dissimilarity measurement is totally sub-
jective. It is extremely unlikely that the dissimilarities arrived at
by the judge would obey proximity structure S1, since they are all
integer-valued. The only possibility would be if the whiskies could
be represented by integer points on a one dimensional Euclidean
space and differences between points generated all forty-five dis-
similarities correctly. It is even unlikely that S2 would be satisfied.
The most likely structure is S3, or possibly S5 if actual scores were
ignored and only the rank order of the dissimilarities taken into
account.
In other situations, similarities (dissimilarities) are constructed
from a data matrix for the objects. These are then called similarity
(dissimilarity) coefficients. Several authors, for example Cormack
(1971), Jardine and Sibson (1971), Anderberg (1973), Sneath and
Sokal (1973), Diday and Simon (1976), Mardia et al. (1979), Gor-
don (1999), Hubdek (1982), Gower (198513), Gower and Legendre
(1986), Digby and Kempton (1987),Jackson et al. (1989),Baulieu
(1989), Snijders et al. (1990) discuss various similarity and dis-
similarity measures together with their associated problems. The
following synthesis of the work of these authors attempts to outline
the main ideas behind forming dissimilarities from a data matrix.

Table 1.1 Dissimilarity measures for quantitative data
Euclidean distance
Weighted Euclidean
Malialariobis distarice
City block metric
Minkowski metric
Canberra metric
Divergence
Bray-Curtis
i
Bhattacliaryya distance
Wave-Hedges
Angular separation
Soergel
Correlation

Let X = [IC,~] denote the data matrix obtained for n objects on p
variables. The vector of observations for the rth object is denoted
by Xr7 and SO X = [x:].
Quantitative data
gives a list of possible dissimilarity measures for quan-
titative data that are in particular, continuous, possibly discrete,
but not binary.
Binary data
Object s
a + b
0 c + d
Object r
a+c b+d p = a + b
I + c + d
When all the variables are binary, it is usual to construct a sim-
ilarity coefficient and then to transform this into a dissimilarity
coefficient. The measure of similarity between objects r and s is
based on the above table. The table shows the number of variables,
a, out of the total p variables where both objects score “I”,the
number of variables, b, where r scores “1”and s scores “O”, etc.
gives a list of similarity coefficients based on the four
counts a,b, c,d. Various situations call for particular choices of co-
efficients. In practice, more than one can be tried hoping for some
robustness against choice. HubAlek (1982) gives the most compre-
hensive list of similarity coefficients for binary data and groups
them into five clusters based on an empirical evaluation using data
on the occurrence of fungal species of the genus Chaetorniurn.
Nominal and ordinal data
If, for the ith nominal variable, objects r and s share the same
categorization, let S r S i = 1,and 0 otherwise. A similarity measure
is then p-’ xis,,i. Of course, if other information is available re-
garding the relationship of various categories for the variables, then
s,,i can be given an appropriate value. For example, if the vari-
able “bottle shape” has categories: standard (st); short cylindrical
Table 1.1
Table 1.2

Table 1.2 Similarity coeficients for binary data
Braun, Blanque
a
Inax{ (a+b),(a +c ) }
S r s =
2a
2 a + b + c
Czekanowski, Smensen, Dice S r s =
Hamnian
Jaccard coefficient
Kulczynski
Kulczynski
S
,
, = ~
b + c
4(ad - bc)
{ ( a+d ) 2 +( b +c)2}
Michael S r s =
Mountford
Mozley, Margalef
Ochiai
Phi
Rogers, Tariinioto
Russell, Rao
Simple matching coefficient
Sirripson
Sokal, Sneath, Anderberg
Yule
2a
a(b +c) +2bc
a(a +b +c +d )
(a +b)(a+c)
a
Srs =
S r s =
Srs =
[(a+b)(a +c)]
ad - be
[(a+b)(a+c)(b+d)(c+d ) ] i
a + d
a +2b+ 2c +d
a + b + c + d
a + d
a + b + c + d
min{ (a +b),(a +c)}
a +2(b +c)
S r s =
S r s =
a
S r s =
Srs =
a
S r s =
a
S r s =
ad - be
ad +be
S r s = ~

(sh); tall cylindrical (ta); and square section (sq), the following
“agreement scores” may be appropriate for bottles r and s.
bottle r
st sh ta
0.5 1.0 0.3 0.0
ta 0.5 0.3 1.0 0.0
sq 0.0 0.0 0.0 1.0
bottle s :L 1
1.0 0.5 0.5 0.0
So if bottle r is “tall cylindrical” and bottle s “standard” then
s,,i = 0.5, for example.
If a variable is ordinal with k categories, then k - 1 indicator
variables can be used to represent these categories. The indicator
variables can then be subjected to similarity coefficients in order
to give a value to s,,i. For instance, if a bottle variable is “height
of the bottle” with categories: small; standard; tall; long and thin,
then the variable might be categorized as follows.
Indicator variable
category I1 I2 13
small 0 0 0
standard 1 0 0
tall 1 1 0
long and thin 1 1 1
If bottle r is “standard” and bottle s is “long and thin”, then
using the simple matching coefficient to measure similarity for this
variable, s,,i=0.33. For further details see Sneath and Sokal (1973)
or Gordon (1999).
1.3.1 Similarity/dissimilarity coeficients for mixed data
When data are of mixed type where binary, categorical, ordinal
and quantitative variables might be measured, the similarity and

Random documents with unrelated
content Scribd suggests to you:

him. Like Joan of Arc, he "hears voices" in the trees and from the
rocks, the winds, the waters, the animals, and the birds. When he
returns to his tribe and convinces the braves that he has received
the Spirit, from that day he is entirely trusted. The greatest chief
must consult him concerning every movement; whether it be the
distant chase, change of location, or of war. He is Sir Oracle.
STRANGE VISITORS IN OLD ST. LOUIS.
The writer does not speak at random or by hearsay of Indian life. He
saw and studied something of it, more than half a century ago,
before civilization had wrought the changes now seen. Indians are
profound believers in the immortality of the soul. Some suspend
their dead in the leafy treetops, that they may the more easily
ascend to "the happy hunting-grounds." The custom of many is to
kill the favorite horse and bury it with all accoutrements and
implements of war, as well as their finest garments, believing the

spirit will need them and receive greater honor. The leading thought
of the Indian seems to be that all material things have a spirit that is
immortal. The Indian burying-grounds are sacred spots and seldom
if ever are desecrated in savage life, even by their worst enemies.
Some of the beautiful little islands in the rivers of the Far West have
thus been used, as the many ruins testify. It has long been noted
that Indians in war will risk their own lives to carry off and bury their
dead and prevent mutilation of bodies.
Is the Story of the Flathead Chiefs of 1831-32 Authentic?
So strange and so without precedent in savage life was the mission
of the Indians to St. Louis, that many have doubted the truthfulness
of the report, and have called it "visionary." Fortunately the reader
need not be in doubt in regard to the entire truthfulness of the event
as reported. The Christian people of that time believed and acted
upon it in a way to convince every honest mind of their earnestness.
It may be said the incident made a profound impression in the
religious world, and the history we are to recite of the after-results
mark it as one of the providential events guiding the nation by
unseen hands to its destiny.
Had such a notable event occurred in modern days, it would have
entered at once into current literature. That it did not at the time is
no disparagement of its truthfulness. There is one strong chain of
evidence regarding the mission of the Nez Perces chiefs, not easily
broken; that is, the written evidence of George Catlin. Aboard the
steamer Yellowstone, upon which General Clark sent his savage
friends, there happened to be a celebrated artist, George Catlin,
then on one of his visits to the West to paint Indian pictures and
study Indian life. These Nez Perces chiefs at once attracted him, and
they became intimate friends—during the long journey he made
pictures of them. Indians are not great talkers, and he did not learn
much from them as to the object of their long journey. From others
afterward he heard of their strange mission to St. Louis, and
believing he had secured two historic pictures, he first wrote General
Clark, and afterward met him, and was assured by him that such

was the mission of the four Flathead chiefs. Catlin, in his
Smithsonian report for eight years, in 1885, says:
"These two men, when I painted them, were in beautiful Sioux
dresses, which had been presented them in a talk with the
Sioux, who treated them very kindly, while passing through the
Sioux country. These two were part of a delegation that came
across the Rocky Mountains to St. Louis a few years ago to
inquire for the truth of a representation which they said some
white man had made among them, that the white man's religion
was better than theirs, and that they would all be lost if they did
not embrace it. Two of the old and venerable men of the party
died in St. Louis, and I traveled two thousand miles,
companions with those two fellows, toward their own country,
and became much pleased with their manners and dispositions.
When I first heard the report of the object of their mission, I
could scarcely believe it, but upon conversing with General
Clark, on a future occasion, I was fully convinced of the fact."
The two pictures are now numbered 207 and 208 in the Smithsonian
Institution, and highly prized. H. H. Hcotes Min (no horns on his
head), who made the notable banquet speech, died near the
Yellowstone River on the journey home, and but one, the youngest
of the four, Hee-Ah-K. S. Te Kin (the rabbit skin leggins), lived to
reach his tribe beyond the Rockies. As was customary with the
Indians, a large band was sent along the trail far away to the Rocky
Mountains to meet the expected delegation of chiefs with "the book
of heaven." Their legends say, "Rabbit Skin Leggins shouted when
far off, 'A man will be sent with the book.'" The world of to-day may
well give thanks, that both Christian men and women were "sent
with the Book" at that earnest and honest appeal. Christianity is
broad, and its command is to "preach the Gospel to every creature."
The Nez Perces Indians, who, in blind faith, sent for teachers, were
blessed in the act above all Indian tribes in the land, and the
blessing has followed them from that day to this. In another
connection in a later chapter will be read facts in proof of their

condition, and showing the effect of the Gospel verses upon Indians.
Indian men, like the whites, are made up of good and bad. The
missionaries were bright, shrewd men and women, and they easily
saw that so fair a land could not much longer be held by savages in
its unfruitful condition. They bent themselves to the heavy tasks laid
upon them, to do the best they could for their savage wards. The
true story for our pages, however, does not take us into any large
study of missionary work, but mainly along the lines of Christian
patriotism.
The author in answer to any critics of the missionaries to the Indians
will relate a simple incident in his own experience, which dates
fourteen years after their advent in Oregon. It shows how the seeds
of Christianity they planted made of savages unselfish and humane
men. It was on a Saturday, after days of weary traveling, we came
to a little valley where we at once resolved to rest for a couple of
days. It was such a little paradise that we named it "The Valley of
Blessing." On Sunday morning, with a single companion, the writer
wandered for miles up the narrow valley, enjoying its luxurious
surroundings. To the right was a mountain whose rugged sides were
covered with dwarf firs and cedars; while rocks were piled on rocks
looking like ancient castles in ruins. Flowering vines climbed to the
tops of the trees, and their fragrance filled the air. A clear stream
divided the valley where flocked myriads of birds from the mountain,
as they drank and bathed, whistled cheerily to their fellows in the
mountain home. As we were admiring all this wilderness of beauty,
on rising a little eminence, we came suddenly in view of four
Indians, digging at a short distance away. We immediately dropped
behind the hill, but not before we had been observed by the Indians.
We were watchful and well armed, but the old Indian gave us a
peace signal, and we approached the spot. The company was made
up of an aged Indian, eighty or more, his grandson, and two half-
breeds. They were digging a grave and were silent as we stood until
its completion. The old Indian then invited us to look at the corpse
under the shade of a near-by tree. We were astonished to find it the
emaciated body of a white man. It was wrapped in a well-tanned

buffalo skin, white and clean. The four Indians took the body and
placed it in the grave, and the old man, removing his cap, to our
astonishment, said, "Now, maybe some white man who knows
religion will make a little prayer over the poor fellow!" The half-
breeds, perhaps not understanding the English the old chief spoke,
began pushing in the sand with their moccasined feet. Thus the
Argonaut of 1850 was laid to his final rest, with only the wild birds to
sing his requiem. The old Indian had brought along a smooth board
to place at the head of the grave, and at his request, I wrote:
John Wilson, St. Louis, Mo., 1850.
Left by his company and nursed by Hoo Goo Chee.
He told us Wilson had traveled as long as he was able, and begged
his companions to leave him there alone to die. He told the chief he
had no complaint to make of his fellows. We mention the incident to
show that the beautiful trait of unselfishness has a place even
among Indians. The old chief could easily have buried the body near
his mountain home without bearing it the long distance to be near
the road, where the grave could be seen by his friends. He might
have used an old blanket instead of the costly dressed robe the
Indian prized so highly. Here we found a savage who, like the
Flatheads, had heard of "the white man's book of heaven," had
practically caught its unselfishness and humanity, and its spirit of
love.
It is well to remember that the Indian has no literature, and has ever
been dependent upon his enemies to write his history and his
achievements. They have chosen to write only of his savagery. This
is not fair treatment by the United States government, incited by
justice, and the wholesome Christian sentiments of the land has
during the past thirty years done much to correct all abuses of its
savage wards.

Multidimensional scaling 2nd ed Edition Trevor F. Cox

CHAPTER III
The Effect of the Banquet Speech. How it Stirred Christian
People. The American Board Acts. Drs. Parker and Whitman
Go on a Voyage of Discovery. His Indian Boys. His Marriage
and Journey through Savage Lands to Oregon.
The Indian oration at the St. Louis banquet was translated by a
young man present, William Walker, who was an Indian chief, but a
white man, and it was first published some months later in "The
Christian Advocate" in New York, with a ringing editorial from its
editor, Rev. Dr. Fiske, headed, "Who will Carry the Book of Life to the
Indians of Oregon?"
The effect was electrical among religious people in the East. The
Methodist Foreign Missionary Society were prompt to act, and the
very next year sent two able-bodied, earnest Christian ministers,
Jason and Daniel Lee, with one layman to aid them. They reached
their field by the long, round-about waterway, via London and the
Hawaiian Islands. For many years they did effective work, far up on
the Willamette River. The American Board, then under the control of
Congregational and Presbyterian churches, was more cautious. It
was an unheard-of proposition to come from savage life so far away
from civilization, and they wanted time to investigate. The Rev. Dr.
Samuel Parker of Utica, New York, became restive under the delay,
believing fully in the call of the Indians, and resolved to join some
trading company to the Far West and go to Oregon. In 1834 he
reached the border upon the Missouri, but the fur-traders had
departed. He returned home and renewed his efforts to arouse the
American Board to action. He found Marcus Whitman, M. D., as
much of an enthusiast in the work as he, and the Board resolved to
send the two men upon a voyage of discovery in 1835, and to have
them return and report upon the possibility of establishing missions

in that well-nigh unknown land. So in 1835, the minister and the
young physician were on the western border in time to join a
company of American fur-traders, bound for Green River, in what is
now northern Utah. Upon reaching this point they met some two
thousand Indians, representing various tribes living within five to
seven hundred miles. There were large delegations of Oregon
Indians to trade their furs for articles needed. When the object of
the missionaries was explained to the Indians, they received the
news with such enthusiasm as to dispel every doubt from the minds
of the missionaries of the wisdom of their course and the Indians'
sincerity in asking for Christian teachers. Upon consultation they
agreed that it was wise to make no delay in reporting to the
American Board. While Dr. Parker was to continue his journey to
Oregon with the Indians, Dr. Whitman was to return with the convoy,
make the report, and return the next year with reinforcements to
begin mission work. The Indians showed such confidence in Dr.
Whitman's promise to come to them after one more snow, that they
selected two of their brightest, most intelligent, and muscular boys
about eighteen years to accompany him, and help him on his way
the coming year. Dr. Parker, with his Indian guides, reached Oregon,
over which country he traveled extensively. He organized no mission,
but studied the situation fully, so as to be able to make a wise report
for the future guidance of the American Board.
Finding a ship sailing next year for the Hawaiian Islands he did not
wait for Whitman and his company. Dr. Parker was a scholarly man
and a keen observer, and upon his return wrote a book of great
value. It was a true description of Indian life and conditions, the
wealth of forest and the prospective finds of coal and minerals in the
hills and mountains, the magnificence of rivers, the healthfulness
and mildness of the climate. The book passed through six editions,
and was interesting reading, but it was of a far-away land, and
induced little or no immigration at that time.

AN INDIAN WELCOME.
Dr. Whitman and his two Indian boys joined the fur company for
escort on its return trip. While on the plains a scourge of cholera
broke out, and the Doctor's skill and his untiring work to save the
lives of the men, won all their hearts, and they united in giving him
a cordial invitation to join them in the spring, upon their annual visit
to Green River. This was gladly accepted, as such an escort was a
necessity in that day. The Doctor and his two Indian aids reached
Rushville, New York, late on a Saturday night in November, 1835. His
return was unexpected, and his first appearance to his friends was
when he marched up the aisle of the church with his Indian boys, as
they sang the opening hymn. His good old mother was so
astonished that she spoke right out in meeting, "If there ain't Mark
Whitman!" It is easy to conceive that such an incident called out a
wide interest and inquiry, which was just what the Doctor desired,
enthused as he was himself in the importance of the work before
him. The Doctor had taken great pains all summer to instruct his
Indian boys in English, and they proved apt pupils. He put them at
once in school, where they made rapid progress, and were general

favorites. Never was the enthusiastic young Doctor more active than
in the fall and winter months in making his preparations. The
American Board had resolved to establish a mission in Oregon, and
they notified him that they preferred to send married men into the
missionary field. This was unexpected but welcome news to
Whitman, and was in accordance with the last advice from Dr.
Parker: "Bring with you a good wife." He had already in his own
mind made his selection in the person of Miss Narcissa Prentice, a
daughter of Judge Prentice, of Angelica, New York, but owing to the
privations and perils of the journey, and the isolated life among
savages, he had hesitated to ask her to make such sacrifice. One
can easily imagine his happiness, when upon fully explaining all, he
found her with a courage equal to his own, and an abounding
enthusiasm for the prospective work. After a time the clear-headed
men of the Board, doubtless guided by their clearer-headed wives,
raised a point, and said, upon such an expedition, so full of care and
responsibility and danger, it would not do to send a woman unless
accompanied by another of her sex. Here was a new dilemma. Time
was passing, and candidates for such perils were not plentiful. The
day of the wedding was postponed, and Whitman endeavored to
meet the requirements. He finally heard of Dr. Spalding and his
newly wedded wife, who were en route to the Osage Indian Mission.
He learned their proposed route and set out to find them. Whether
through chance or Providence, he succeeded. It was a cold day and
a driving snow, when in his sleigh he sighted them ahead, after a
long chase. When in hallooing distance he shouted, "Ship ahoy, you
are wanted for Oregon!" Hearing the cheery, pleasant voice, they
halted, Whitman driving his sleigh by the side of theirs, and he at
once bounded into the subject of which he was full. Dr. Spalding
proposed that they go to the hotel in the town just ahead, where
they could talk the matter over without freezing. By a glowing fire
Dr. Whitman retold the story of the Flatheads, about whom they had
read; of his journey to the Far West to verify the facts, and the
result, and of the two Indian boys ready to escort them to Oregon,
where they would meet with an enthusiastic reception such as he
and Dr. Parker had received on Green River. Whitman was often

called "The Silent Man," but when aroused and enthused, he was an
eloquent pleader. And with all at stake, as in this instance, he was
doubtless at his best. They listened with profound attention. Mrs.
Spalding was an educated woman, of great decision of character, an
earnest "Christian," and a firm believer in a power higher than
herself ready to guide her in life's duties. They were silent for a
moment, when she arose and said, "I desire a few moments to
myself for prayer," and retired to her room. The two men sat by the
fire canvassing all the dangers of the expedition and the hopefulness
of the outlook. Dr. Spalding afterward wrote, in speaking of the
meeting, "I do not think she was gone from us more than ten
minutes before she returned, her face all aglow with happiness and
enthusiasm, and said, 'Yes, we will go to Oregon!'" He continues, "I
gently expostulated, 'My dear, we must consider your health in such
a hazardous undertaking.' She replied, in the words, 'Go ye into all
the world and preach the Gospel to every creature, there is no
exception made for ill health.' And no words of mine could alter her
determination." Mrs. Spalding had been a semi-invalid for months,
but her faith and Christian courage were strong. It was her prompt
decision which decided the fate of the Oregon mission, of the four
notable characters, and we may add, the fate of questions so great
and grave to the nation, as to be unfathomable by man's wisdom.
The wedding day was again fixed. In this case there was more than
usual interest in the bride, for her friends all knew of her destination.
The late Mrs. H. P. Jackson, of Oberlin, Ohio, sister of the bride, has
told me in letters, of the events of the pleasant occasion. The two
Indian boys, dressed in their best, were guests of honor. Dr.
Whitman introduced them to his wife, and says Mrs. Jackson, "When
he told them she would go with them to their far-away home in
Oregon, and teach them, they did not try to conceal their delight."
Narcissa Prentice was the eldest daughter of Judge Prentice, an
influential, earnest Christian man, then residing in Angelica, New
York. The daughter was well educated, loved for her womanly
qualities and famed in all the country around for the sweetness of

her voice. She was the leader of the church choir of the village, and
the people crowded the building the evening before their departure
to bid the little party a good by and give them a blessing. After a
good social time, the minister, the Rev. Dr. Hull, called the meeting
to order, and gave out the old familiar hymn:
"Yes, my native land, I love thee,
All thy scenes I love them well;
Friends, connections, happy country,
Can I bid you all farewell?" Etc.
The late Martha J. Lamb, editor of the "Magazine of American
History," who wrote the report of the farewell gathering, says:
"The great audience joined in singing the opening stanzas, but
soon they began to drop out by ones and scores, and sobs were
heard all over the audience. The last stanza was sung by one
voice alone in a clear, sweet soprano, and not a faulty note; it
was the voice of Narcissa Whitman."
It was the last time her old friends heard the sweet voice, for
daylight found them braving the winter storm on their way to
Oregon.
The late Eli G. Coe, of Illinois, then a young man, drove them in his
sleigh to the mountains, en route to Pittsburg, where they were to
take boat for St. Louis. He has given me a delightful sketch of the
journey, upon which he marks Whitman "The Silent Man, ever
thoughtful of all his guests," and Mrs. Whitman, "The lovely little
woman who was the life of the company, who often dispelled gloom,
and made all forget the winter cold, by a song of cheer."
Their route was from Pittsburg down the Ohio River to the
Mississippi up to St. Louis, thence up the Missouri River to near
where Fort Leavenworth now stands. The journey had no mishaps
until they reached "The Big Muddy," as the Missouri has long been
called. Those who navigated it half and three-quarters of a century

ago, will never forget the journey. It was sand bars on sand bars,
forever shifting with each freshet, and snags galore! The engineer
stood constantly at his lever, to answer the bell, a leadsman stood in
the bow casting a lead and calling in loud, singsong the depth of
water, until suddenly, like an electric shock, came the sharp, "five
feet scant," and the bell rang, and the wheels reversed with a
suddenness that aroused every one, until he got used to it. They
were hung on snags, "hard aground" on sand bars, and as a
consequence were four or five days behind the time at Leavenworth.
The reader will recollect that the fur-traders had given Dr. Whitman
a cordial invitation to join them in the spring, and he was impatient
but helpless in the delay. To the great discomfiture of the
missionaries upon reaching the landing, they learned that the fur
company had left four days before. What added to Whitman's
trouble was, that at St. Louis he had been told he could get all the
provisions he lacked at the fort, and upon inquiry, found nearly
everything sold, and that he would have to start in poorly equipped
with provisions, without a hope of being able to add to his stock,
except by chance and courtesy of the traders.
This was the first great test of the courage of Dr. Whitman. Dr.
Spalding was outspoken, "We must turn back and never think of
such madness as to brave a journey among savages without an
escort." Whitman said little, but rapidly made his preparation, simply
declaring, "We will go on." Mrs. Spalding nobly seconded Whitman,
and said, "I have started for Oregon, and to Oregon I will go or
leave my body upon the plains." Mrs. Whitman was alike cheerful.
So soon as harness could be adjusted, the loads packed, and the
cattle rounded up, the man of courage gave the order, and the little
train began to move through the deep mud of the Missouri River
bottoms. We learned after that the fur company waited one day over
the stipulated time. But they had in some way learned at St. Louis
that the Doctor was going to bring with him some American women
for the journey, something never heard of before, and as they were
expecting to have to fight their way at times, they did not care for

such encumbrances, anxious as they were to have the services of
the good Doctor. Thus it was a gloomy start for the brave little
company. Dr. Whitman had made ample preparation for the comfort
of the women in a spring-wagon, "the brides' wagon," fitted up with
various little comforts and a protection in every storm. But it is
doubtful whether two cultivated American brides before, or since,
ever made so memorable a wedding journey. The party consisted of
the two brides and their husbands, Dr. W. H. Gray, two teamsters,
and the two Indian boys. We may add that somewhere in the Sioux
country the boys picked up three other Nez Perces friends; one of
them, Samuel, was added permanently to the company. Mrs.
Whitman writes, "When the boys get together they make a great
chattering."
They were in an Indian country from the first day's start, and met
great numbers of savages, out on their hunts, many moving to new
camps, and some on the war-path. At no time were the missionaries
molested, but on the contrary, were treated with great courtesy, and
as Mrs. Whitman wrote, "They seemed greatly surprised to see white
women in the party." The Indian boys were soon in their element,
and of inestimable value; they could swim the rivers like ducks, and
took all the care of the loose stock, and were wise in the ways of
plains' life. They could explain to any suspicious Indians the coming
of "the great medicine men" they were taking to their people, and in
a hundred ways were helpers to the little company. Mrs. Whitman,
from the outset, rode on horseback with her husband, only
occasionally resting in the wagon, and for company to Mrs. Spalding,
who was yet an invalid.
We make no pretense of writing a continued narrative of the
journey, but just enough to catch its spirit. We have seen in it a
dreary and discouraged start, and none but a hero with heroines to
encourage him would have entered upon it. They had now been a
whole month on the way making forced marches, the trail of the fur-
traders getting fresher every day, until finally hearing they were in
camp on Loupe Fork, the wagons pushed on and joined them. The

Doctor and Mrs. Whitman were behind helping to hurry forward the
loose stock. Finally, late at night, the Indian boys begged the Doctor
and his wife to ride on to camp and leave them to drive the stock in
at daylight. But they refused to leave them. Picketing their horses
out to graze, then with their saddles for pillows, they lay upon the
warm ground looking up at the stars and slept. At daylight they rode
into camp and were courteously received and praised as "a plucky
set."
The two American women, who had so alarmed the old plainsmen
as a burden and an encumbrance, by their tact and kindness soon
won them as friends, and nothing was left undone that the rough
old fellows could do for their comfort. They had succeeded so
admirably in passing safely for a month alone through the Indian
country, that they began to have confidence in themselves. But they
learned that they had not yet reached the point of real danger, and
were glad to be protected by such a stalwart troop. The Indians had
a great respect for these pioneer traders, who were veterans of the
plains and splendidly armed. The greatest anxiety was for the safety
of their stock at night, when picketed out to graze. The Indians
especially coveted the oxen and cows, which required careful
guarding to prevent stampeding. Cattle when frightened at night
lose all sense, breaking away and running as long as they can stand,
becoming easy prey for the savages, while horses and mules almost
invariably break for the tents and wagons, and the company of men.
Camp at night is always made by driving the wagons in a circle, with
tents pitched inside. The wagons make a protection from an enemy,
and all their contents are in easy reach.
The year 1836 was a peaceable year among the Indians, and the
buffalo and other game was so plentiful as to make small temptation
for Indian depredation upon the white man's stock during this
portion of the journey, but we may add they cast longing eyes at all
times upon every good horse the white man rode.
In the Buffalo Country

The company had now reached the buffalo country, and soon began
to see great herds containing thousands, and even tens of
thousands. Every spring the buffalo journeyed northward to the
valleys and plains to feed on the rich grasses. It is a feast occasion,
one of the greatest the Indian enjoys. Tribes travel four and five
hundred miles from their homes to meet the buffalo, and lay in a
supply of dried meat, calf skins, and robes, and never forgetting to
feast for a month while laying up winter stores. It is a novel and
exhilarating sight to view the annual Indian migration to meet these
noble wild cattle of the plains—the whole tribe, old and young, dogs
and loose horses, with all their movable worldly goods brought with
them packed on poles drawn by ponies. They settle down in the little
valleys near springs, or along running waters, and arrange for work
in advance with as much system as the farmer in the spring plows
and sows. The buffalo country has generally, by mutual consent,
been regarded as "peace grounds," but the desire for revenge has
many times made it the scene of bloody contests and massacres.
Hunting buffalo in those days, either by the Indians or white men,
was not sport, but butchery. They were in such immense herds that,
when running from their enemies, those in the rear could not get out
of the way, and were an easy prey to any kind of weapon of death.
The buffalo bull is the most gallant and noble among animals. On
the march he leads, brings up the rear, and marches on the flanks,
while all the cows and calves are kept in the center of the herd and
protected from the bands of wolves, mountain-lions, and bears
which linger around ready to devour the straying members of the
herd. By a wonderful provision of nature, the buffalo calves are
practically all of the same age, so that a herd in the long summer
outing is not much detained upon its way, for the little one trots
gayly beside its mother in a few hours. But while the little fellows are
thus comparatively helpless, those who have witnessed the scene,
bear testimony to the courage of the great, strong-necked, sharp-
horned bulls who will attack a grizzly or a whole pack of wolves, or a
mountain-lion regardless of his own danger. At such times he is even
at night a sleepless, faithful picket ever on duty. He walks backward
and forward along his picketed line like a trained soldier, and when

the ground is wet, he treads a deep path in the sod, and the picket
line of a sleeping herd can easily be traced long afterward, and often
is referred to as "Indian trails." One would suppose that such nobility
would command respect. But it never did. Even such explorers and
writers as Parkman and his men never seem to have enjoyed the
day unless, in addition to the calves they killed for food, they were
able to tell of the slaughter of many "savage old bulls." At the time
of which I write buffalo were seen by the million. Fourteen years
later, when the writer visited the same region, they could be seen in
single herds covering a thousand acres. When frightened and
running, they were turned from their course with the greatest
difficulty.
A train on the trail they were crossing was only safe in halting and
allowing it to pass. The pressure from the rear was so great that the
front could not halt. Some of the old plainsmen told of "a
tenderfoot's" experience, who was going to have some "rare sport,
and his pick of an entire bunch." He observed a large herd quietly
grazing and saw by making a detour, up a dry ravine, where he
would be hidden from view, he could get immediately in their front.
He succeeded, and tying his mule behind him, concealed himself in
the edge of some bushes upon the bank of the creek. He did not
have long to wait, something in the rear frightened the herd and it
began to come directly toward him. As soon as in reach, he began to
fire and kill. It would break the ranks for an instant only, and he at
once saw death impending, as there was not a tree large enough to
climb. He had shot until his gun was hot, but all in vain. Just then his
old mule tied in the bushes opened up his musical "honk, honk,"
such as only a thoroughly frightened mule can utter, and the whole
herd opened right and left, and the man was saved.
Some have expressed a wonder that these noble animals, in such
myriads, should so soon have disappeared. It is easily seen, in the
fact of the improved firearms used by the Indians, and that they
killed, for food, skins for clothing, and robes for the market, only the
cows and calves. They selected only the choice cuts of the meat,

and left the great bodies for the wolves and other varments. They
could tan only the skins of cows and calves for clothing and for
tepee covers. It was a sickening sight to pass over the place of
slaughter, and thus see hundreds of bodies, with only tongues and
choice cuts and skins taken. American hunters were equally
sacrificial. Half a century later the writer rode over the same land
and saw Indians, all across the region, with carts and pack ponies
gathering up bones of the buffalo. Passing stations along the Great
Northern and Northern Pacific railroads, one passes ricks of bones
half a mile long on each side, and as high as the tops of the cars,
waiting for shipment East as fertilizers, and horn handles for knives
and other uses in the arts. Only two living wild herds of buffalo are
now reported, one small one in Texas, and one carefully protected
by the government in Yellowstone Park. It would have been wise
and humane had they been protected sooner by the strong arm of
the law.
But it was the great good fortune to our missionaries to meet the
buffalo herds. They started out poorly provided, and would soon
have been in distress, for they had added three Nez Perces Indian
boys to their company, and the pure air and exercise upon the plains
provokes great appetites. It was equally good for the fur-traders,
who had calculated upon the event. So the whole train stopped and
began to kill and "jerk" meat. The Indian boys were in their element
and veterans in the business, and laid in bountiful supplies. While it
is fresh and juicy few animals furnish more nutritious food. A buffalo
porterhouse steak, cooked over coals at the end of a forked stick,
when the thermometer of appetite is up to "one hundred degrees in
the shade," is a royal feast to be remembered. If however kept up
long enough, the good old-fashioned pig with lean and fat strips on
his ribs, is quite a relief. But the dried meat was the staple food of
the little company from that time on. Mrs. Whitman cheerfully and
jokingly writes in her diary, "We have dried buffalo meat and tea for
breakfast, and tea and jerked buffalo for supper, but the Doctor has
a different way of cooking each piece to give variety to the
entertainment."

Mrs. Whitman kept carefully a daily diary of events of travel, which
was luckily preserved, and passed into the hands of her sister, Mrs.
Jackson, of Oberlin, Ohio, which I have been permitted to read and
from which have copious selections in my larger work, "How Marcus
Whitman Saved Oregon," after which it was passed on to the
Whitman College Library, where it is preserved as a precious
treasure. The notable feature of this diary is its self-sacrificing spirit
and good cheer. The scorching sun, the clouds of alkaline dust that
stung the eyes and throat, the impure water they were compelled to
use, the myriads of mosquitoes and buffalo gnats, all of which the
author so well remembers as the dreariest things encountered in a
long life, did not daunt the spirit of this delicate little woman. Not a
word of complaint can be found in that daily diary, which was never
written for the public eye, or for effect. The nearest to it was once,
after being without flour or bread for weeks, she writes, "O for a few
crusts of mother's bread; girls, don't waste the bread in the old
home!" Men and women are all human, and I have no desire to
picture my characters as perfect beings. They doubtless had their
faults, but none who have not experienced some of the difficulties of
that pioneer band, who, tired and worn with travel, sought sleep
while hungry (after shaking out their blankets to be sure no snakes
were within them), can censure. I repeat, it takes such experience to
fully appreciate the heroism and unselfishness of such consecrated
lives.
The old pioneers were wise geographers and surveyors. There were
two things necessary for life upon the plains, viz., water and grass.
They studied their maps and saw the Platte, North and South Forks,
reaching northward and westward. So they made their trails along
the banks, cutting off bends, avoiding impossible sloughs and hills,
but keeping an eye upon the river in the distance, and ever working
nearer to it when a detour had been made. The two Plattes thus
furnish supplies for from five to six hundred miles. Travellers struck
across the divide for the Sweetwater and its tributaries, until the foot
of the Rockies is reached.

As the eyes of our travelers had rested for a month upon the snow-
covered peaks of the great stony mountains, one can imagine it was
a day of rejoicing when they began the ascent. The trail up "the
South Pass" was so easy a grade that the horses and cattle scarcely
felt the strain. One looking at it would surmise that this break in the
great mountain was not an accident, but it was left for a great
highway between the oceans, to make one family, and a United
Nation. Striking mountains, after the long dreary summer upon the
alkaline plains, hard as mountain-climbing is, was yet a change to be
appreciated. I recollect distinctly, it turned our little company of
sturdy men (a few years later) into rollicking boys who whooped and
sang to get the echoes, and rolled great stones, until their arms
ached, crushing down the mountain-side.
PACIFIC SPRING—JULY 4, 1835
A Notable Celebration

Here on the top of the Rockies, or just beyond the summit, is a
spring appropriately named "The Pacific Spring," for its pure, ice-cold
water bubbles up and in a silvery stream winds its way westward. It
is a beauty spot as the author well remembers. A little valley upon
the mountains, covered with grass and wild flowers, with grand
views of valleys and mountains reaching farther away than the eye
can follow. Here the missionaries halted and allowed the fur-traders
to pass on. It was the Fourth Day of July, a day ever memorable in
the mind of every patriotic American. True they were but
missionaries, and far from home and friends, but they were home-
lovers and patriots. So spreading their blankets upon the bunch
grass, they brought out the American flag, unfurled it, and with
prayer and song dedicated the fair land thence to the Pacific, to God
and the Union. It was a prayer and song which after history proved
a prophecy; and one in which the actors in this little celebration took
so brave a part as to deserve their names enrolled among the
nation's royal benefactors. God rules the world, and all history shows
that he oftenest leaves the great and strong, and takes the weak
and humble to accomplish his grand purposes. Eternity will reveal
whether that dedication was one of the agencies which brought the
after grand results. Certain it was, that it was the agency of Dr.
Whitman and his heroism in carrying out that vow years after, and
stirred up a spirit never before experienced, and aroused the nation
to action.
No stage could have been grander for such a celebration. Behind
were the long stretches of the great plains, and still beyond the
civilization of the continent, the hope of the Christian world; while
before was the wilderness in all its wildness, reaching to the Pacific.
The Rockies towered about them, glittering in the sunshine! The
craggy peaks of the Wind River mountains loomed up in the north,
with the Coast Range visible, like floating clouds in the far west. The
luxurious grass, the towering pines, and flowers that perfumed the
air, made the spot beautiful, while the history of the event is a fit
theme for a grand national epic or painting. There have been many

historic celebrations of the nation's birth, some upon battle-fields
where victory perched upon the "the banner of beauty and glory,"
but none more impressive than when upon that mountain top, in
1836, Mrs. Whitman's musical voice echoed from the rocks and
trees,
"The star-spangled banner in triumph shall wave,
O'er the land of the free, and the home of the brave."
They had now entered upon the scenic stage of their journey, and it
was a delightful change from the dead levels of the plains. They
luxuriated in the pure ice-cold water, and magnificent scenery, but it
was well for them that they knew none of the weary climbs ahead.
We will not pause to note events from thence to Green River.
There they met with exciting and interesting savage life in all its
realities. They found at "the rendezvous" two thousand Indians in
camp, waiting for the coming of the traders. A thousand or more
were from the Oregon country, and among them friends and
relatives of the Indian boys, who had come the long distance to
meet and welcome them, as well as to trade. They gave the boys a
royal greeting, as they regarded them as heroes and great travelers.
They were proud of their accomplishments in speaking like the
"Bostons," and when the missionaries vouched for their earnest,
faithful services, the Indians were proud of their boys. Here they
stayed for nearly two weeks waiting for the completion of the
trading. The Indians regarded the missionaries as their guests, and
taxed themselves to the utmost to amuse them by wild games and
feats of horsemanship and mimic battles. They scoured the hills and
woods for game, brought fish from the river, and seemed to think
even that not doing enough. They at all times treated "the white
squaws" with the greatest courtesy. Mrs. Whitman marks this in her
diary. She says:
"One of the chiefs brought his wife to our tent, and taking off
his cap and bowing gracefully, introduced her as politely as any

civilized man. Such encourages me to believe that much can be
done for these poor people, and I long to be at work."

CHAPTER IV
"Old Click-Click-Clackety-Clackety," the Historic Wagon. Breaking
Camps and its Incidents, and the End of the Journey.
Breaking camp at Green River was a noisy and gleeful occasion.
Half-starved Indian ponies, when they have rested a few weeks,
generally rebel when packs are cinched with a "diamond hitch"
around their well-marked ribs. Upon this occasion amusement was
diversified and enjoyable, even to the actors. But both Indians and
traders were no novices in such business, and soon the companies
bade good by to each other and started along the trails to their
widely scattered homes. It was the great exciting social event of
Indian life, this distant visit to trade. The Indians there met friends
and relatives, exchanged gossip, gathered the few luxuries and
necessaries of life for the year to come. They brought with them
squaws and some of their children, and enjoyed their outing in their
savage way as much as the élite do the seashore or Saratoga, and
judging of both, one would say they had more fun. The Oregon
Indians were all anxious to be escorts to "the Boston teachers."
There were two intelligent traders from Oregon, Messrs. McKay and
McLeod, who offered escort to the little company, which was gladly
accepted, and they were of invaluable service in that most difficult
portion of the journey. The faithful Indian boys, however, held their
places of honor and trust to the last. Mrs. Spalding had for some
time been on horseback, and enjoyed it more than the wagon,
traversing the rocky roads. There was no longer need of two
wagons, and one was left at the rendezvous; but "the brides'
wagon" pulled out with the pack-train. My young readers may think
it an uninteresting object to write about, but they must remember it
is "the brides' wagon," fitted up with all the little accommodations
for the first two white women who braved the dangerous journey

across the great stony mountains to the Pacific. True, it was battered
and worn, dust and mud and storms had robbed it of style. It is well
for those who ride in palace cars and whizzing 'autos to remember
the days of their great grandfathers and grandmothers, who, amid
privations and perils, with the parting blessings of Puritan homes,
pulled across the Alleghanies in rough wagons and hewed out
homes, and built this great empire of the Middle West. The more
often we remember the heroines of the past the more we will enjoy
this grandest inheritance of the present ever left to any people. But
there was more than sentiment to this wagon as we shall see later
on. It figuratively blazed the way, and "marked a wagon-road to the
Columbia," and years after silenced the eloquence of America's
greatest orator!
The battered old wagon was a source of amusement to the Indians,
who rode in troops by its side to see the wheels go round, and hear
its clatter. Especially was it a novelty to the younger Indians, who at
once named it "Old Click-Click-Clackety-Clackety." There was a plain
wagon-road from the Missouri to Green River, and from thence to
Fort Hall—there it stopped. The royal owners of Oregon had long
before prophesied and decreed, "there would never be a wagon-
road to the Columbia!" They did not want one.

THE RUGGED TRAIL TO OREGON.
The company reached Fort Hall safely, which was an outpost of the
English Company, and only a pack trail led westward to the
Columbia. Captain Grant, in command of the post, knew his
business, and that was never to allow a wagon to go beyond Fort
Hall. He at once told the company of the dangers and perils of the
journey, of the impracticability of hauling a wagon. If tried it would
so detain them that they would be caught in the snows upon the
mountains and perish. His earnestness and arguments were such
that he convinced most of them, who favored abandoning the

wagon. Even Mrs. Whitman joined others in the entreaty to Dr.
Whitman to leave the wagon and move on. "The Silent Man" said
little, but went on with his preparations, and when the pack-train
moved out, "Old Click-Click-Clackety-Clackety" clacked in the rear as
usual. The real facts are, that Captain Grant had scarcely overstated
the dangers and difficulties of the undertaking. From the day they
left Fort Hall until the memorable baptism of the wagon in Snake
River, the old wagon is one of the constant themes of Mrs.
Whitman's diary. We read, "Husband had a tedious time with the
wagon to-day. It got stuck in a creek and he had to wade to get it
out. After that in going up the mountain the wagon upset twice."
She describes the steep up and down mountain trails where at times
the mules had to be unhitched and the wagon lowered with ropes
(as the writer a few years later was compelled to do). She adds, "I
wondered that the wagon was not turning somersaults all the time.
It is not grateful to my feelings to see him wearing himself out with
such fatigue. All the mountain part of the way he has walked in
laborious attempts to take the wagon." About one week later Mrs.
Whitman writes, gleefully, "The axletree of the wagon broke to-day. I
was a little rejoiced, for we are in hopes it will now be left." She
adds, in her next note, "Our rejoicing was in vain; they have made
the wagon into a cart with the back wheels, and lashed the front
wheels to the sides, determined to take it through in some shape or
other." "Worse yet" (she writes a week later), "The hills are so steep
and rocky, husband thinks it best to lighten the load as much as
possible, and haul nothing but the wheels, leaving the box and the
trunk!" What do you think of that, my girl readers? The brides' trunk,
that came from the far-away home, with all its mementoes and
tender memories to be sacrificed, and "only the wheels" taken! But
the gallant McLeod solved the problem and ordered the trunk packed
on one of his mules, and it made the journey safely, and the old
wagon made into a cart, but its wheels and every iron sacredly
preserved, was still a wagon; and under a power impressed upon
one brave soul it moved on its great way, marking a wagon-road and
a highway between the oceans. Those may smile who will, but they
do not think deep, nor do they estimate how small and seemingly

insignificant events shape the greatest events in a personal, and
even national, life.
The last note of Mrs. Whitman's diary referring to the wagon says:
"August 13. We have just crossed the Snake River, the packs
were removed from the ponies and placed on the tallest horses,
while two of the highest were selected for Mrs. Spalding and
me. Mr. McLeod gave me his and rode mine. The river is divided
into three channels by islands, the last, a half a mile wide, and
our direction was against the current, which made it hard for
the horses, as the water was up to their sides. Husband had a
difficult time with the cart, as both mules and cart upset in
midstream, and the animals got tangled in the harness, and
would have drowned but for the desperate struggle for their
release. Two of the strongest horses were taken into the river
and hitched to the cart, while two men swam behind and guided
it safely to the shore."
There they were at Fort Boise, beyond the Snake, and in Oregon!
The wagon-road was made! It was within easy reach of their future
home. There it was decided to leave the cart until spring, together
with half a dozen footsore cattle, which could be sent for, or
exchanged for others at Fort Walla Walla. Packs were now divided
and the patient mules, which had long drawn the cart, became
packers.
An old wagon is the common rubbish in every farm-yard, and if my
reader enters a protest to the large place I have given it, or to
protest against Marcus Whitman for his persistent refusal to take the
advice of his companions, I will state in simple defense, I believe
Whitman was an inspired man! He never once made such claim,
even to the wife he almost adored. Later on, as we shall see, he
obeyed the same voice under far more trying circumstances, when
called to make his midwinter ride to save Oregon.

When his friends insisted in saying, "It is like going down into the
valley and shadow of death; wait until spring," his only answer was,
"I must go now!" Who can fathom such mysteries in any other way
than that I have mentioned. The chances are, he never dreamed of
making a trail for a great transcontinental traffic. It is not at all likely
that ever the thought came to him that he should guide a great
immigrant train over the same route a few years later and the brides'
wagon proved a notable factor in his success.
The Last March
The incumbrances left behind, the company moved on as rapidly as
the loose stock could be driven. It was still a wild, rugged road, but
much of the country traversed was beautiful. They were all now on
horseback, and all their worldly possessions on pack-saddles. The
weather was delightful, game abundant, and there was now no
danger of starving, although they had long been without all the
luxuries common to civilization. But best of all, they were buoyed up
by the near completion of a nearly seven months' journey of
hardships and danger. The day before they were to reach Fort Walla
Walla, the Doctor and Mrs. Whitman rode ahead of the company,
and camped under the trees on the bank of the river, eight miles
from the Fort. At daylight they were upon the road. Who can
imagine the delight of the tired travelers, as they came in sight, at a
distance, of human habitations and civilization! They spurred their
horses into a gallop and rode to the gates of the Fort just as the
occupants were sitting down to breakfast. The men and women of
the Fort came at once and admitted them through the gates, and
gave them a cordial welcome, and did their best to make them feel
at ease.
Mrs. Whitman writes in her diary:
"September 1, 1836. We reached here this morning just as they
were sitting down to breakfast. We were soon seated at table
and treated to fresh salmon, potatoes, tea, bread, and butter;

what a variety thought I. You cannot imagine what an appetite
those rides in the mountain air give a person."
She playfully adds that,
"While at breakfast a rooster perched himself upon the
doorstep, and crowed lustily. Whether it was in honor of the
arrival of the first two white women, or as a general compliment
to the company, I know not, but he pleased me."
The rest of the company reached the fort during the afternoon. Here
they all were, and none missing, right upon the scene of their
probable future labor.
The Cayuse Indians who had earnestly interceded for teachers were
the owners of a great tract of fertile land on both sides of the Walla
Walla River. Adjoining them, one hundred miles distant, was the Nez
Perces, to whom all the missionaries felt indebted and attracted,
because of the boy friends who had so faithfully served them during
the long journey, and as well for their amiable dispositions. The
Cayuse were smart Indians, whose wealth was in horses, which
roamed over their rich pastures, and without care, kept fat the year
through. But the Cayuse were not like the Nez Perces, always to be
relied upon. They were sharp traders, and notably tricky. But our
missionaries found they could do nothing by way of settlement until
they presented their credentials and consulted with the ruling
authorities—the English Hudson Bay Company at Vancouver, two
hundred and fifty miles down the Columbia. They were urged to stop
and rest before making the long journey, but so eager were they to
get to their work, and to make preparations for the winter, that they
declined the kind invitation. Large boats were secured, and strong-
armed, experienced Indian rowers soon bore the party to their
destination, through a land, and along rivers romantically interesting.
They found great bands of Indians on their route, especially at the
rapids, and The Dalles, where many found employment, as boats

and goods had to be carried for miles to smooth water. Dr. Whitman
at once marked The Dalles as an ideal place for a mission.[1]
Dr. McLoughlin, the chief factor of the Hudson Bay Company,
received the party most cordially, and bade them welcome. He was
known among the Indians as "The great white head chief." He was a
giant in stature, a gentleman of culture and education, and a man
with a soul as large as his body. From the outset there seemed to be
a freemasonry attachment between Whitman and McLoughlin. They
were much alike, physically and mentally. They were both physicians
and men with high moral character, stamped in every act of their
lives. McLoughlin carried out fixed principles in all his dealings with
the Indians; he never allowed them cheated in any trade; he lived
up to every promise made; and the savage tribes, in every quarter,
obeyed his commands like good soldiers do their general. Whitman
laid bare the whole case, how and why they were there, and
concealed nothing. His ideas freely given were, that he believed
savages must first be taught to build homes, plant and sow, and
raise cattle, sheep, and stop their roaming life. This was directly
what the Hudson Bay people did not want. They wanted furs and
skins, and to get them whole tribes must each year migrate to the
distant hunting and trapping regions. Dr. McLoughlin, while anxious
to serve the missionaries, was yet true to his company. He had
placed the Methodist missionaries Jason and Daniel Lee the year
before far up the Willamette, and he explained to Dr. Whitman that
The Dalles was not the place for a mission, and that it would be far
better for the company and for the missionaries, to settle in a more
distant quarter. It all resulted in Dr. Whitman going to the Cayuse on
the Walla Walla, and Dr. Spalding to the Nez Perces, one hundred
and twenty-five miles further on.
McLoughlin was so impressed with the honesty and earnestness of
his guests, that he gave them liberty to draw upon him for anything
he could furnish for their use and comfort. Such an unlooked-for
kindness was greatly appreciated. And we may add, as far as Dr.
McLoughlin could execute the promise, it was sacredly fulfilled. It is

Welcome to our website – the ideal destination for book lovers and
knowledge seekers. With a mission to inspire endlessly, we offer a
vast collection of books, ranging from classic literary works to
specialized publications, self-development books, and children's
literature. Each book is a new journey of discovery, expanding
knowledge and enriching the soul of the reade
Our website is not just a platform for buying books, but a bridge
connecting readers to the timeless values of culture and wisdom. With
an elegant, user-friendly interface and an intelligent search system,
we are committed to providing a quick and convenient shopping
experience. Additionally, our special promotions and home delivery
services ensure that you save time and fully enjoy the joy of reading.
Let us accompany you on the journey of exploring knowledge and
personal growth!
ebookname.com

Multidimensional scaling 2nd ed Edition Trevor F. Cox

More Related Content

Similar to Multidimensional scaling 2nd ed Edition Trevor F. Cox (20)

Recently uploaded (20)

Multidimensional scaling 2nd ed Edition Trevor F. Cox