Download the Full Ebook and Access More Features - ebooknice.

com

(Ebook) Graph Algorithms for Data Science (MEAP

v7) by Toma Bratani

https://ebooknice.com/product/graph-algorithms-for-data-
science-meap-v7-50699820

OR CLICK HERE

DOWLOAD EBOOK

Download more ebook instantly today at https://ebooknice.com

 Instant digital products (PDF, ePub, MOBI) ready for you
Download now and discover formats that fit your needs...

Start reading on any device today!

(Ebook) Graph Algorithms for Data Science: With examples

in Neo4j by Bratanic, Tomaž ISBN 9781617299469, 1617299464

https://ebooknice.com/product/graph-algorithms-for-data-science-with-
examples-in-neo4j-55836718

ebooknice.com

(Ebook) Vagabond, Vol. 29 (29) by Inoue, Takehiko ISBN

9781421531489, 1421531488

https://ebooknice.com/product/vagabond-vol-29-29-37511002

ebooknice.com

(Ebook) Grandmaster Repertoire 5: The English Opening 1.c4

c5 - Volume Three by Mihail Marin ISBN 9781906552596,
1906552592
https://ebooknice.com/product/grandmaster-repertoire-5-the-english-
opening-1-c4-c5-volume-three-2112606

ebooknice.com

(Ebook) The Modern English 1.c4 c5, 1...Nf6, 1...e6,

Volume 2 by Kiril Georgiev, Semko Semkov ISBN
9786197188233, 6197188236
https://ebooknice.com/product/the-modern-
english-1-c4-c5-1-nf6-1-e6-volume-2-53518506

ebooknice.com
(Ebook) Graph-Theoretic Concepts in Computer Science: 44th
International Workshop, WG 2018, Cottbus, Germany, June
27–29, 2018, Proceedings by Andreas Brandstädt, Ekkehard
Köhler, Klaus Meer ISBN 9783030002558, 9783030002565,
https://ebooknice.com/product/graph-theoretic-concepts-in-computer-
3030002551, 303000256X
science-44th-international-workshop-wg-2018-cottbus-germany-
june-2729-2018-proceedings-7319936
ebooknice.com

(Ebook) Graph Data Science with Neo4j: Learn how to use

Neo4j 5 with Graph Data Science library 2.0 and its Python
driver for your project by Scifo, Estelle ISBN
9781804612743, 180461274X, 9781804612763, 180461234X
https://ebooknice.com/product/graph-data-science-with-neo4j-learn-how-
to-use-neo4j-5-with-graph-data-science-library-2-0-and-its-python-
driver-for-your-project-56106314
ebooknice.com

(Ebook) Graph Data Science (GDS) For Dummies by Alicia

Frame, Zach Blumenfeld ISBN 9781119909330, 1119909333

https://ebooknice.com/product/graph-data-science-gds-for-
dummies-48372456

ebooknice.com

(Ebook) Experimental Algorithms: 14th International

Symposium, SEA 2015, Paris, France, June 29 – July 1,
2015, Proceedings by Evripidis Bampis (eds.) ISBN
9783319200859, 3319200852
https://ebooknice.com/product/experimental-algorithms-14th-
international-symposium-sea-2015-paris-france-
june-29-july-1-2015-proceedings-5141820
ebooknice.com

(Ebook) Boeing B-29 Superfortress ISBN 9780764302725,

0764302728

https://ebooknice.com/product/boeing-b-29-superfortress-1573658

ebooknice.com
Graph Algorithms for Data Science
1. MEAP_VERSION_7
2. Welcome
3. 1_Graphs_and_network_science:_An_introduction
4. 2_Representing_network_structure_-
_design_your_first_graph_model
5. 3_Your_first_steps_with_the_Cypher_query_language
6. 4_Exploratory_graph_analysis
7. 5_Introduction_to_social_network_analysis
8. 6_Projecting_monopartite_networks_with_Cypher_Projection
9. 7_Inferring_co-
occurrence_networks_based_off_bipartite_networks
10. 8_Constructing_a_nearest_neighbor_similarity_network
11. 9_Node_embeddings_and_classification
12. 10_Link_prediction
13. 11_Knowledge_graph_completion
MEAP VERSION 7
Welcome
Thanks for purchasing the MEAP for Graph Algorithms for Data
Science.

This book has been written for anyone with no experience with
graphs to more experienced graph users to augment their
understanding of graph algorithms and their role in the various
analysis. With traditional analytics, you are trying to make sense of
data points, whereas with graph analytics, you are more focused on
analyzing connections between data points. Graph algorithms are
designed to investigate those connections between data points and
help you explore who is well connected, who has the most influence,
how communities form, and more.

When I first started experimenting with graph analytics five years

ago, there weren't many tools available, and you would often need
to combine a few tools helping you construct a graph and then
analyze it. Not only that, but there weren't many tutorials or courses
available that would explain the graph analysis workflow in simple
terms. For this reason, I've decided to write this book that would
minimize the number of tools needed to get started and explain
graph algorithms and analysis workflow in basic terms.

I will present the advantages of using a graph database and teach

you how to construct a graph from structured and unstructured data
using NLP techniques. Along the way, you will learn how to use
Cypher query language to manipulate graph structure and extract
valuable insights. Next, I will walk you through the typical graph
algorithms like PageRank and community detection/clustering
algorithms and demonstrate how to use them in practice. The last
part of the book will focus on graph machine learning, specifically
how to translate graph topology and structure into machine learning
model input by using node embedding models and graph neural
networks.

Please let me know your thoughts and ideas in the liveBook

Discussion forum on what's been written so far. Your feedback is
invaluable in improving and increasing the understanding of this
book.

Thanks again for your interest and for purchasing the MEAP!

—Tomaž Bratanič

In this book

MEAP VERSION 7 About this MEAP Welcome Brief Table of Contents

1 Graphs and network science: An introduction 2 Representing
network structure - design your first graph model 3 Your first steps
with the Cypher query language 4 Exploratory graph analysis 5
Introduction to social network analysis 6 Projecting monopartite
networks with Cypher Projection 7 Inferring co-occurrence networks
based off bipartite networks 8 Constructing a nearest neighbor
similarity network 9 Node embeddings and classification 10 Link
prediction 11 Knowledge graph completion
1 Graphs and network
science: An introduction
This chapter covers:
Introducing networks and graphs
Introducing node degree characterization of a network
Spotting graph-shaped tasks
Introducing machine learning on graphs

Networks are everywhere, and they do matter. First of all, where are
these networks? Communication networks are one example. For
example, the internet consists of routers. Routers analyze the
incoming data, determine the optimal path to the destination, and
forward the data to the next device along the route. Another example
are the social media platforms. You use those platforms to connect
with other users. Most of your connections are local, ranging from
your family and friends to coworkers. And then you have some
connections from distant friends that can span oceans and
continents. When you map all those connections, what you end up
with is referred to as a social network.

Figure 1.1. World-wide social network.

Also very interestingly, your biological existence depends on
networks. Proteins are called the building blocks of the body. They
form the machinery that helps sustain life. Proteins rarely act alone
as their functions tend to be regulated. The identification of protein
interactions can lead to a better understanding of diseases and the
development of drugs and treatments. The process of mapping those
interactions results in protein-protein interaction networks, also
known as PPI.

Figure 1.2. Protein-protein interaction network.

One can also map the connections between neurons in a brain as a
network, also known as the connectome. Not only is it possible, but
scientists from Google AI and a research team from Janelia Research
Campus in Virginia have released a detailed network of the neuronal
connectivity of an animal as well as the human brain.[Shan Xu et al,
2020]

I hope by now, you have realized that networks are everywhere. You
just have to open your eyes to see them. When looking at the world,
you can spot networks occurring in politics, markets, art, and even
the dependencies between code modules form as a network. If
you’ve ever developed any code beyond Hello World, you probably
imported other libraries or code from your other modules.

Figure 1.3. Python code module dependency network.

In Figure 1.3, you can observe that the Pandas library has an
external dependency on the NumPy library. All the other
dependencies are between internal code modules in the Pandas
library and are several levels deep.
Now we can move on to the second question. Why do networks
matter? The usefulness of networks and the underlying graph theory
lies in its ability to model many real-world situations and applications.
In Figure 1.1, we visualized a network where the nodes represent
people, and the relationships represent friendship links. Instead of
treating entities as independent from one another in your analysis,
you must realize that we live in a connected world. Using the graph
approach to analysis, you consider that the world is connected and,
therefore, the entities are often not independent. Your job as a
network scientist is to map and understand those connections in real-
world scenarios. In the social network example, given the friendship
network topology, you can find influencers or leaders. Information
about a network’s leaders might help you spread your information
throughout the whole graph more efficiently. You could also try to
find the nodes that would break the network if they were removed.
Finding the most critical nodes can help you disrupt or upgrade the
resilience of a network. Another example would be to find segments
or communities of like-minded people. The information about the
segments could be used as input to your recommendation engine or
to improve your target advertising persona. Understanding the
segments of customers you are dealing with is also very helpful in a
business domain. Instead of looking at people’s friendship links, you
might be interested in their purchasing and product usage behavior
patterns. There are so many real-world network applications that it is
impossible to squeeze them into a single chapter or even a book.

1.1 Introduction to graph theory

Before we delve more into network science and its applications, you
will first have to learn a bit of graph theory. First of all, what is the
difference between graphs and networks? The term graph is used
more in mathematical terminology and refers to a general language
for describing and analyzing entities and their relations or
interactions. On the other hand, the term network is used to describe
a real-world dataset containing entities and their relationships, such
as a social network, a communication network, or a road network. In
practice, the distinction between the terms graph and network is
blurred, and the terms are often used interchangeably. In this book, I
will try to stick to using the term graph when introducing
mathematical theory and concepts and use the term network when
describing real-world entities and their relationships.

1.1.1 What is a graph?

Interestingly, if you google images of a graph, you will find similar
images like the following:

Figure 1.4. A sample bar chart visualization.

Even the first definition of a graph in the Merriam Webster dictionary
describes a diagram representing a variable compared to that of one
or more other variables. In this book, we will refer to any diagram or
a graph that visualizes a series of one or more points, lines, or areas
as a chart. Furthermore, in this book, the term graph is reserved to
describe a set of nodes (also known as vertices) and relationships
(also known as connections, edges, or links).

The history of graph theory can be traced to the 18th century when
the Swiss mathematician Leonhard Euler solved the Königsberg
bridge problem. [Leonhard Euler, 1736] In that time, seven bridges
spanned over the forked river in Königsberg. According to folklore,
the puzzle arose of whether a person could take a walk through the
town in a way in which they would cross each bridge exactly once.
Euler argued that no such path exists. Euler’s solution is simple, once
you look at the problem from a graph perspective.

Figure 1.5. A graph visualization of the Königsberg bridge problem.

The problem of finding a way that crosses every bridge exactly once
turns into finding a path through the graph that traverses every
relationship exactly once. For a walk that crosses every relationship in
a graph exactly once to be possible, either none or precisely two
nodes can have an odd number of edges attached to them. However,
in the Königsberg bridge problem, all nodes have an odd number of
relationships, making a walk that crosses every bridge once
impossible. In solving this puzzle, Euler started a field that is today
known as graph theory.

1.2 How to spot a graph-shaped

problem
Before beginning with network analysis, you must ask yourself if a
graph-based approach is suitable for your problem. While almost any
use-case can be modeled as a graph, a specific set of scenarios is
uniquely suited for graph-based analysis.

The first scenario deals with self-referencing relationships between

entities of the same type. In relational databases, a self-referencing
relationship occurs between data points within the same table. One
such example would be a Facebook friendship network, where you
have a single entity type and multiple relationships between them.

Figure 1.6. Facebook friendship network.

Figure 1.6 represents a small friendship network, where there is a

single type of nodes called User that can have one or more
relationship to other users. Graphs are commonly encouraged to use
in analysis when you are dealing with many data joins. As far as I
have seen, the most frequent and straightforward demonstration of
graphs is the friend of a friend query. The idea behind the friend of a
friend query is that you are interested in persons who are two or
more hops or joins distant from the original person. Having many
joins when dealing with larger datasets might be computationally
expensive in traditional relational databases but is quick and easy
when storing the data in a graph database due to treating
relationships as a first-class data structure. The trick is that instead of
computing joins between data entities by scanning foreign key
indexes at query runtime, native graph databases store a physical
representation of relationships at import. This trick delivers a better
query performance when traversing relationships between your data
points. There are many scenarios for a graph with self-referencing
relationships:

Detecting social networks influencers

Analyzing consequences of vulnerabities in a dependency
network
Inspecting organizational hierarchies

Another fairly common graph scenario is discovering paths or routes

between entities or locations. Most of you have probably used
navigational systems to find the most optimal route for your travels.

Figure 1.7. Transportation network.

Figure 1.7 visualizes an example transportation network between
cities in Belgium and Netherlands. The cities are represented as
nodes, while the transportation modes are represented as
relationships between cities. For example, you could bike from
Antwerp to Rotterdam in about 330 minutes or take the train from
Hague to Amsterdam for 37 minutes.

Like mentioned, you could use a transportation network to calculate

the most optimal route based on your specifications. The route could
be optimized by time, distance, or cost. You could also analyze the
network as a whole and try to predict traffic congestions based on
the network structure or find critical connections that would disrupt
the whole network if, for example, an accident occurred. In a
relational database, you would have to hypothesize the order of
relationships you must join to find an available path between two
entities. In the example in Figure 1.7, there are three relationship
options you could choose to traverse. You could hop from one city to
another using the road, railroad, or bike network. Another problem
you might face with traditional databases is that you don’t know
beforehand how many relationships you must traverse to get from
node A to node B. Not knowing beforehand precisely which and how
many relationships you must traverse could lead to potentially
complex and computationally expensive queries. Treating your data
as a graph helps you mitigate those two problems.

Finding optimal routes can be applied on following scenarios:

Logistics and routing

Infrastructure management
Finding optimal paths to make new contacts

Another very powerful use-case for graphs is examining indirect or

hidden relationships. Consider the following graph:

Figure 1.8. User-item network.

Figure 1.8 represents a network of customers and their purchases.
The purchased products can also be categorized into groups like
apparel or technical goods. You can observe that both Clair and
Aditya bought a phone that falls into the technical goods category.

While there are no direct relationships between customers, you can

compare their purchasing patterns and find similar customers.
Essentially, you could define segments of customers using this
approach. These types of graphs are also frequently used in
collaborative filtering recommender systems, where you search for
similar customers or products commonly purchased together. You
could also examine how many purchases have items that span across
many product categories. There are many more scenarios where
user-item networks come in handy, like movie recommendations on
Netflix or song recommendations on Spotify. Essentially, any scenario
where a person is rating, purchasing, or voting for an item can be
modeled as a user-item network. As another example, think of app
store reviews or parliament members voting on laws and resolutions.
You could then investigate how similarly members of parliament vote
and compare that to their political party association.

1.3 Machine learning on graphs

In the last couple of years, the field of machine learning on graphs
has taken of. The main idea behind graph machine learning is to
manually define or automatically learn the node representations and
encode them in the embedding space.

Figure 1.9. Encoding node position into the embedding space. Copyright
(c) 2017 Manan Shah, SNAP Group
Figure 1.9 demonstrates the idea of encoding nodes in a network into
embedding or euclidian space. In a traditional machine learning
workflow, each data point is represented as a vector of integers or
floating points. The vectors are then fed into a machine learning
model during training and inference. The primary challenge of
machine learning on graphs is finding a way to represent or encode
network structure as a vector to be easily fed into a machine learning
model.

For example, let’s say you have been given the task of predicting a
person’s net worth based on their characteristics and attributes. The
dataset contains features that describe each data point and the
target variable that needs to be predicted. With supervised
classification, the training data contains both the features as well as
the target variable value, which you can use to train your machine
learning or deep neural network model. Once you have trained the
model, you can use it to predict the net worth of previously unseen
data points and examine how well it works.

Figure 1.10. Traditional machine learning approach, where you treat each
data point as independant.
Figure 1.10 shows an example dataset, where each data point is
described by features such as age, hobby, and education. The data
points are considered independent, which means they are not related
or connected. You would then train a machine learning model based
on the available features to predict a person’s net worth.

You might know that people are very interconnected, and many
people will tell you that networking is a vital part of getting more and
better job or other opportunities in life. Since you haven’t encoded
any of the networking attributes as data point features, you will skip
all that information about connections that might help you more
accurately predict a person’s net worth better. Essentially, you treat
each data point as independent and ignore its context. Here is where
graph machine learning and node representation learning come into
play.

First, you need to map those connections or relationships between

people. The relationships could be of various types like friends,
coworkers, family, or others. Similarly, you could also define the
strength of a connection. For example, if you hang out with a friend
almost every day, the relationship is stronger than if you only meet
once every couple of months. In graph terminology, the strength of
the relationship is characterized by its weight. For simplicity’s sake,
the following example will not differentiate between relationship
types or weights.

The most basic network characteristic of a node is its degree. A node

degree is a local characteristic of a node and is simply the count of its
relationships. A local characteristic of a node does not take into
account the whole network. Specifically, node degree considers only
direct neighbors, but other local characteristics of a node could, for
example, consider friends of a friend. If you consider only direct
neighbors, you can say that you are examining a node’s
neighborhood one hop away. Likewise, if you would also consider
friends of friends, you are considering all nodes that are at most two
hops away.
In the example of predicting a person’s net worth, one could argue
that a person with more connections will have more opportunities in
life, which could correlate with their net worth. Another benefit of
having many connections is having access to more experts in their
respective fields. Talking to experts and understanding their points of
view could help you make better decisions, which could also correlate
to your net worth.

Figure 1.11. Add the number of connections each person has as a feature.

Left hand-side of Figure 1.11 visualizes mapped connections between

data points. In this example, you can say that connections represent
friendships between persons. We hypothesized that the number of
friendship links might correlate with one’s net worth. To incorporate
the count of relationships each node has (node degree) as a model
feature, you simply count the connections and add them as an
element in the data point representation table.
With the node degree example, you have just added a single
"graphy" feature to describe a data point, while the rest of the
machine learning workflow remains the same. While simply counting
the number of connections does not require specialized storage or
tools, it is a nice example to get you thinking about relationships
between the data points and how you could use them in your
machine learning workflow.

Graph algorithms usually refer to more global operations that

consider the whole network as an input. Imagine that the information
in the network can only flow through existing connections. For
example, in Figure 1.11, if node A wants to communicate with node
I, the information must flow through nodes E and G. A graph
algorithm called Betweenness centrality can be used to identify the
node’s influence over the information flow in the network. The
assumption behind the Betweenness centrality is that the information
always flows along the shortest paths between pairs of nodes. Then,
the node’s importance over the information flow is simply the count
of those shortest paths that pass through the node. The
Betweenness centrality can be used to identify bridges between
different communities or, in our example, find nodes with a strong
influence over the information flow of the network.

Figure 1.12. Add the betweenness centrality as a feature of a person.

In the example in Figure 1.12, node E is the bridge between the
upper and bottom communities. First of all, it makes node E be the
vital connection or a bridge between the two communities. If node E
was removed from the network, the two communities couldn’t
communicate anymore. Secondly, as a lot of information flows
through it, it is probably well informed and has access to the type of
information other nodes do not. Lastly, it can withhold passing the
information forward and effectively cutting the communication lines
between half of the network.

In the context of social networks, a person with better access to

information can make better decisions. Not only that, but they can
exert their influence over the information flow. As a practical
example, say that you are a manager who reports directly to the CEO
of the company. All the information from the CEO to your managees
flows through you. You get to hear both sides of the story but have
the option to decide which information you will pass along to the CEO
or your subordinates. Since data or information is sometimes
regarded as being a more valuable resource than oil, one’s influence
over the information flow could correlate with their net worth.
Similarly, as with the node degree, you could use the Betweenness
centrality score as one of the model’s features that predicts a
person’s net worth.

So far, both node degree and Betweenness centrality values are

examples of manual feature engineering. Manual feature engineering
is a process of identifying relevant data points representations that
could correlate with the target variable by hand. Recent
advancements in machine learning on graphs focus heavily on
automatically encoding a node’s network position. The field of
research that studies automatically encoding nodes' position in a
network as a vector is called node representation learning. The key
here is the learning part, where you train an embedding model or a
neural network to describe a node in the network with a feature
vector.

You might be familiar with image or text embedding models, which

perform similarly to node embedding models. Both images and text
can be represented as a graph.

Figure 1.13. Both image and text can be represented as a graph.

Text and images have a pre-determined graph shape used to derive
word or image representations by the embedding models. Without
going into specifics of text or image embedding models, they
essentially derive a pixel or word vector representation by observing
its neighborhood. Most, if not all, real-world networks don’t have a
pre-determined shape and can vary significantly from domain to
domain. While the node embedding models take great inspiration
from both word and image embedding models, they are designed to
work on graphs with any shape and form.

Probably you have come across an anecdote that a person is an

average of their best friends. I’ve also read some articles that not
only your direct friends, but also friends of friends, influence your life
options and choices. Instead of manually describing relevant person
features as we did before, we can utilize a node embedding model
that will automatically encode a node’s neighborhood as a vector that
can be used in a downstream machine learning flow.

Figure 1.14. Learn node representation automatically by aggregating its

neighborhood.
Some node embedding models encode only the nodes' position in the
network, while others also consider nodes' properties. For example,
graph neural networks consider both node properties and their
position in a network to derive the final vector representation. In the
example scenario of predicting a person’s net worth, you could utilize
a graph neural network to derive a node’s vector representation that
could be used to predict a person’s net worth more accurately.

Through the practical examples in this book, you will learn how to
spot graph-shaped problems and construct a graph. Next, you will
learn how to calculate and interpret both the local and global node
characteristics like the node degree and the Betweenness centrality.
The last part of the book is dedicated to machine learning on graphs,
where you will learn how to improve your model’s accuracy by
encoding the network’s structure as your data point feature sets.

1.4 Summary
Networks are everywhere and they do matter
A bar or line chart is not regarded as a graph in this book
Problems that require a graph-based approach have
interconnected data points such as self-referencing relationships
in a social network or paths in a transportation network
 Sometimes the relationships between data points are not
explicitly defined but can be inferred based on indirect patterns,
as in the example of users purchasing products
Node degree attribute represents the count of relationships a
node has
Node degree is a local characteristic that examines the node’s
direct neighborhood
Graph algorithms like the Betweeness centrality move beyond
the direct neighborhood of a node and inspect the whole
network
Node embedding models are used to automatically encode a
node’s network position as a vector
Encoding network information for a downstream machine
learning task can greatly improve your accuracy

1.5 References
[Shan Xu et al, 2020] Xu CS, Januszewski M, Lu Z, Takemura S-Y,
Hayworth KJ, Huang G, Shinomiya K, Maitin-Shepard J, Ackerman D,
Berg S, et al. A connectome of the adult Drosophila central brain.
bioRxiv. 2020 [accessed 2021 Jan 21]:2020.01.21.911859.
www.biorxiv.org/content/10.1101/2020.01.21.911859v1.
doi:10.1101/2020.01.21.911859

[Leonhard Euler, 1736], "Solutio problematis ad geometriam situs

pertinentis". Comment. Acad. Sci. U. Petrop 8, 128–40, 1736

[Erdős Rényi, 1959] P. ERDŐS-A. RÉNYI, On random graphs.

I,Publicationes Mathematicae (Debrecen),6 (1959), pp. 290–297.

[S.Brin and L. Page, 1998] S. Brin and L. Page. The anatomy of a

large-scale hypertextual web search engine. Computer networks and
ISDN systems, 30(1-7):107–117, 1998.

[Albert, R., Jeong, H. & Barabási, 1999] Albert, R., Jeong, H. &
Barabási, AL. Diameter of the World-Wide Web. Nature 401, 130–131
(1999). doi.org/10.1038/43601

[Barabási and Albert, 1999] Barabási, A.L., & Albert, R. (1999).

Emergence of Scaling in Random Networks. Science, 286(5439),
509–512.
2 Representing network
structure - design your first
graph model
This chapter covers:
Math and text representation of graphs
Introducing graph databases
Labeled-property graph model schema design
Extracting information from text

In the previous chapter, you learned the basics of graph theory.

Before you dig more into practical network analysis, you first have to
learn how to practically represent network structures. The most basic
graph representation is the mathematical data structure adjacency
matrix.

Figure 2.1. Adjacency matrix representing a network structure

An adjacency matrix is a square matrix, where the matrix elements

indicate whether pairs of nodes are connected or not in the graph.
The adjancency matrix dimensions are equal to the number of nodes
in the graph. It can also be expanded to represent weighted graphs.
Instead of having zeroes indicating the presence of the relationships,
you store the relationship weight as the matrix element. You will not
be using it in the examples of this book, but you can check out the
adjacency matrix appendix if you want to learn more. Another
mathematical structure to represent networks is called the edge list
data structure.

Figure 2.2. A directed unweighted graph represented with a edge list

data structure

An edge or relationship list is a simple data structure where each row

represents a relationship of a given network. The first row in the
edge list of Figure 2.2 represents a directed relationship from node A
to node B. There are five rows in the edge list, which represent five
relationships of the network. It also supports representing multi-
graphs, where you can have multiple relationships between a given
pair of nodes. An edge list can be expanded to hold the information
about relationship weights.

Figure 2.3. A directed weighted graph represented with a edge list data
structure
The value of the relationship weight is stored in a separate column.
You could also store additional information about the relationship in
the edge list, such as the time component. One limitation of the edge
list is that it does not allow isolated nodes to be present. Isolated
nodes are nodes without any relationships. This limitation can be
solved by introducing a node list next to the edge list.

Figure 2.4. With addition of the node list, you can represent isolated
nodes in a network

By introducing the node list next to the relationship list, you can
describe networks with isolated nodes present. The network in Figure
2.4 has an isolated node E. Node E is isolated because it is described
in the node list but has no entries in the relationship list. The node
list can also be expanded to store various properties of the nodes.
For example, the node list in Figure 2.4 contains information about
nodes' age. Node and relationship lists are very frequently used as
input to network visualization tools. When trying to visualize a
network, you could store the size and the color of the visualized node
as additional properties in the node list. Node and edge lists are
useful when you have a defined graph structure that doesn’t require
additional data manipulations or transformations. However, in
practice, I have noticed that you often need to transform and
manipulate the network data to fit your problem best. A typical
example would be translating indirect relationships into direct ones.
While you could use a scripting language for data transformations
directly with node and edge lists, I recommend using a graph
database and dedicated graph-pattern query languages.

Before you learn more about graph databases, we will also look at
the text representation of simple networks. The text representation of
networks comes in handy when you want to communicate the
network structure via text quickly. We will borrow the syntax from the
Cypher query language. Cypher’s syntax provides a visual way to
match patterns of nodes and relationships in the graph using ASCII-
Art syntax. Its syntax describing nodes and relationships is also the
basis for the future Graph Query Language (GQL), which aims to
unify the graph-pattern query language the same as SQL did for
relational databases. An example node representation in Cypher looks
like the following:

(:Person {name:"Thomas"})

To depict nodes in Cypher, you surround a node with parentheses, for

example (node). The colon is used to describe the type of a node. In
the above example, the node type is defined as Person. Nodes can
also have properties, which are depicted as key-value pairs inside the
brackets of a node. There is a single key value pair inside the curly
brackets, {name:"Thomas"}, that represents the name property of the
node. Relationships in Cypher are surrounded with a square bracket.
-[:FRIEND{since:2016}]->

Similarly to nodes, you describe the type of relationship with a colon.

Relationships can also have properties defined as key-value pairs
inside the curly brackets. A relationship can never exist in solitude
without existing source and target nodes. Cypher syntax is frequently
used to describe patterns of a network. For example, you can specify
a simple friendship network with the following syntax:

(:Person {name:"Thomas"})-[:FRIEND {since:2016}]->(:Person

{name:"Elaine"})

This Cypher syntax describes a friendship relationship between

Thomas and Elaine and can be visualized as the following network.

Figure 2.5. Example friendship network between Thomas and Elaine

Both Thomas and Elaine are persons, and they are friends since
2016. If you look carefully, you can observe a direction indicator of
the relationship at the end of the text representation. With it, you can
differentiate between directed and undirected relationships. If you
want to describe the friendship relationship as undirected, all you
have to do is omit the relationship direction indicator.

-[:FRIEND{since:"2016"}]-

I need to add a small disclaimer here. Many of the graph databases

don’t directly support storing undirected relationships. I will show you
how to deal with undirected relationships in a graph database in
Chapter 3.
Try to represent the relationship you have with the organization you
are employed at with the Cypher syntax. There are two types of
nodes present in this graph pattern, a person and an organization or
a business. You can also add additional node or relationship
properties as you see fit.

I could describe my relationship with the Manning publication using

the following Cypher syntax:

(:Person {name:"Tomaz"})-[:WRITES_FOR {since:2020}]->

(:Organization {name:"Manning Publication"})

2.1 Graph databases

A native graph database is typically a type of NoSQL database
designed to store graph representations. There are also some
implementations of graph layers on top of SQL databases, but we
won’t cover them here. The key difference between a native graph
database and other databases is that native graph databases support
index-free adjacency. Index-free adjacency ensures that you can
traverse relationships without using an index. This allows the
performance of graph traversal to be independent of the overall
graph size. The query cost is only associated with the part of the
graph touched or walked by the query. For example, if you were to
traverse all the relationships of a node, the query performance is only
dependant on the number of connections a node has. On the other
hand, traditional relational databases traverse relationships by
performing join operations. Join operation is typically an intersection
operation between two sets, where the relational database uses an
index to detect where those two sets overlap. As a consequence, the
size of the data or the overall graph affects the query performance.
Also, the cost of queries in a relational database grows exponentially
with the number of joins. From a performance perspective, the index-
free adjacency versus a traditional join operation is the most
important thing to consider when thinking about using a native graph
database. In general, the graph databases can be split into two
categories based on the underlying graph model they use.

2.1.1 RDF Graph Database

First, you will learn about the Resource Description Framework (RDF)
graph model. While most RDF graph databases do not support index-
free adjacency, it is still relevant to learn about their underlying graph
model. RDF databases, also known as triplestores, are designed to
store and allow the retrieval of triples using graph-pattern queries. A
triple is a data entity that consists of subject, predicate, and object.
The triple data structure can be visualized as following:

Figure 2.6. A graph visualization of a triplet with two nodes(Subject and

Object) and a relationship connecting them (Predicate).

Although not explicitly specified in a triple, by RDF standard, there

can be three kinds of nodes in an RDF graph:

IRI: An IRI or an internationalized resource identifier denotes

any real-world entity or concept. An IRI is used to identify nodes
in the graph in an unambiguous way. You can describe a single
node in multiple triples by using the same resource identifier
(IRI).
Literal: Literals are values used to represent datatypes like
strings, numbers, or dates.
Blank node: Blank nodes are nodes without an identifier used in
special RDF modeling scenarios.
Predicates can be interpreted as relationships between two nodes or
as defining an attribute value. As the underlying data structure is a
triple, an attribute value is stored as a literal node, and the predicate
forms a relationship between the subject node and the literal object
node. Each predicate has a resource identifier that defines the type of
relationship. The relationships are always directed as they point from
the subject to the object node. For example, let’s say that you want
to represent the following information as an RDF graph:

Thomas is 40 years old

Thomas is friend with Elaine
Thomas is a person

Figure 2.7. Table and network visualization of a RDF graph, where

Thomas is 40 years old and is friends with Elaine

You can observe that you are doing a complete data normalization by
representing a network with an RDF graph model. RDF’s abstraction
level is a triple, where the subject and object of the triple are
represented as nodes, and the predicate is represented as a
relationship. Consequently, there is no internal structure available on
the nodes and relationships, meaning that you are not dealing with
internal node or relationship properties. but store the properties as a
separate node with a literal value. Although, in practice, you might
consider literal node values as node attributes, the underlying triple
data structure treats them as separate nodes. In the example in
Figure 2.7, the node Thomas is disambiguated by using a resource
identifier "http://schema.org/Thomas". Resource identifiers follow the
structure of a URI. For this example, I have made up the resource
identifiers values. I have also omitted the schema.org prefix in the
network visualization for readability purposes. The node Thomas has
three relationships. Two of them point to another resource node,
while the age relationship is pointing to a literal value and can be
interpreted as a node attribute.

In my opinion, WikiData is the most famous graph database based on

the RDF graph model. WikiData acts as central storage for the
structured data of its sister projects Wikipedia and others. It features
a SPARQL endpoint that you can use to extract relevant data. This
book will teach you the most basic SPARQL queries that will allow you
to fetch relevant data from WikiData and enrich your graph with
additional information. You can also explore the WikiData graph with
your favorite internet browser. For example, you can look up and find
information about Tom Hanks on the WikiData webpage.

Figure 2.8. WikiData web page with information about Tom Hanks
available at www.wikidata.org/wiki/Q2263
Tom Hanks is a resource node in the WikiData graph. Its id is Q2263
and in the Figure 2.8 is always regarded as the subject of the triple
data structure. That being said, this does not limit Tom Hanks to
appear as a object of the triple data structure in any other examples.
You can observe that Tom Hanks is a citizen of both the United States
of America and Greece.

As mentioned, most RDF graph databases do not support index-free

adjacency. For this reason, I will not use them as a source of truth for
network analysis in this book. However, a lot of world knowledge is
stored in RDF databases such as WikiData and DBpedia. Both
WikiData and DBpedia provide publicly available APIs that you can
use to retrieve information and enrich your existing graph with it. In
later chapters, you will use the WikiData SPARQL API to enrich your
graph. Also, the RDF has a couple of standard serialization formats
that are helpful when exchanging network data, so if you ever come
across them, you will know the underlying structure they represent.

2.1.2 Labeled-property graph database

The other category of graph databases is based on the labeled
property graph model (LPG). A key difference between the labeled
property graph model and the RDF model is that with LPG, both
nodes and relationships can have internal structures. The internal
structures are node or relationship properties stored as key-value
pairs. Nodes also have a special type of property called a label, which
is used to represent node roles in your domain. For example, you
could use the label to categorize whether a node represents a person
or an organization. All relationships are directed and have exactly a
single type assigned to them. As mentioned, relationships can also
store properties. To demonstrate a simple LPG graph model, I will use
the same information as in the RDF example:

Thomas is 40 years old

Thomas is friend with Elaine
Thomas is a person

Figure 2.9. Labeled-property graph model representing the example data

As you can observe, the key difference from the RDF approach to
graph modeling is that labeled-property graph(LPG) supports both
node and relationship properties stored as key-value pairs. In the
example in Figure 2.9, the age information is now stored as an
internal node property. Another key difference is that you can group
or categorize nodes into distinct sets using a node label. In Figure
2.9, both Thomas and Elaine nodes have a label that indicates they
are categorized as persons. You might have observe that the Cypher
query syntax you learned before is used to describe LPG model
domain.

In some domain use cases, you might still want to represent literal
values as separate nodes. A typical example is the fraud detection
scenario, where you are interested in examining customers who
share the same address, social security number, or phone number.

Figure 2.10. Labeled-property graph model representing a fraud

investigation domain

Nothing is stopping you from modeling your graph where some literal
values are represented as separate nodes. The graph model depends
on your task, and with the LPG model, you can represent a literal
Random documents with unrelated
content Scribd suggests to you:
 “The cavalry barracks, as a whole, are the least overcrowded, and
have the freest external movement of air. Next come the infantry; and
the most crowded and the least ventilated externally are the Guards’
barracks; so that the mortality from consumption, which follows the
same order of increase in the different arms, augments with increase
of crowding and difficulty of ventilation.”[18]

Her own well-trained mind was in extreme contrast with the type of mind
which she describes in the following story:—

“I remember, when a child, hearing the story of an accident, related

by some one who sent two girls to fetch a ‘bottle of sal volatile from
her room.’ ‘Mary could not stir,’ she said; ‘Fanny ran and fetched a
bottle that was not sal volatile, and that was not in my room.’”

All her teaching, so far as I know it, is clearly at first-hand and carefully
sifted. It is as far as possible from that useless kind of doctrine which is a
mere echo of unthinking hearsay. For instance, how many sufferers she must
have saved from unnecessary irritation by the following reminder to nurses:—

“Of all parts of the body, the face is perhaps the one which tells the
least to the common observer or the casual visitor.
“I have known patients dying of sheer pain, exhaustion, and want
of sleep, from one of the most lingering and painful diseases known,
preserve, till within a few days of death, not only the healthy colour of
the cheek, but the mottled appearance of a robust child. And scores of
times have I heard these unfortunate creatures assailed with, ‘I am
glad to see you looking so well.’ ‘I see no reason why you should not
live till ninety years of age.’ ‘Why don’t you take a little more exercise
and amusement?’—with all the other commonplaces with which we
are so familiar.”

And then, again, how like her it is to remind those who are nursing that “a
patient is not merely a piece of furniture, to be kept clean and arranged
against the wall, and saved from injury or breakage.”
She was one of the rare people who realized that truth of word is partly a
question of education, and that many people are quite unconscious of their
lack of that difficult virtue. “I know I fibbs dreadful,” said a poor little servant
girl to her once. “But believe me, miss, I never finds out I have fibbed until
they tell me so!” And her comment suggests that in this matter that poor little
servant girl by no means stood alone.
She worked very hard. Her books and pamphlets[19] were important, and
her correspondence, ever dealing with the reforms she had at heart all over
the world, was of itself an immense output.
Those who have had to write much from bed or sofa know only too well the
abnormal fatigue it involves, and her labours of this kind seem to have been
unlimited.
How strongly she sympathized with all municipal efforts, we see in many
such letters as the one to General Evatt, given him for electioneering
purposes, but not hitherto included in any biography, which we are allowed to
reproduce here:—

“Strenuously desiring, as we all of us must, that Administration as

well as Politics should be well represented in Parliament, and that vital
matters of social, sanitary, and general interest should find their voice,
we could desire no better representative and advocate of these
essential matters—matters of life and death—than a man who, like
yourself, unites with almost exhaustless energy and public spirit,
sympathy with the wronged and enthusiasm with the right, a
persevering acuteness in unravelling the causes of the evil and the
good, large and varied experience and practical power, limited only by
the nature of the object for which it is exerted.
“It is important beyond measure that such a man’s thoughtful and
well-considered opinions and energetic voice should be heard in the
House of Commons.
“You have my warmest sympathy in your candidature for Woolwich,
my best wishes that you should succeed, even less for your own sake
than for that of Administration and of England.—Pray believe me, ever
your faithful servant,
“Florence Nightingale.”

And also the following letter written to the Buckinghamshire County Council
in 1892, begging them to appoint a sanitary committee:—
 “We must create a public opinion which will drive the Government,
instead of the Government having to drive us—an enlightened public
opinion, wise in principles, wise in details. We hail the County Council
as being or becoming one of the strongest engines in our favour, at
once fathering and obeying the great impulse for national health
against national and local disease. For we have learned that we have
national health in our own hands—local sanitation, national health.
But we have to contend against centuries of superstition and
generations of indifference. Let the County Council take the lead.”

And how justly, how clearly, she was able to weigh the work of those who
had borne the brunt of sanitary inquiry in the Crimea, with but little except
kicks for their pains, may be judged by the following sentences from a letter
to Lady Tulloch in 1878:—

“My Dear Lady Tulloch,—I give you joy, I give you both joy, for this
crowning recognition of one of the noblest labours ever done on
earth. You yourself cannot cling to it more than I do; hardly so much,
in one sense, for I saw how Sir John MacNeill’s and Sir A. Tulloch’s
reporting was the salvation of the army in the Crimea. Without them
everything that happened would have been considered ‘all right.’
“Mr. Martin’s note is perfect, for it does not look like an
afterthought, nor as prompted by others, but as the flow of a
generous and able man’s own reflection, and careful search into
authentic documents. Thank you again and again for sending it to me.
It is the greatest consolation I could have had. Will you remember me
gratefully to Mr. Paget, also to Dr. Balfour? I look back upon these
twenty years as if they were yesterday, but also as if they were a
thousand years. Success be with us and the noble dead—and it has
been success.—Yours ever,
“Florence Nightingale.”

We see from this letter how warmly the old memories dwelt with her, even
while her hands were full of good work for the future.
The death of Lord Herbert in 1868 had been a blow that struck very deeply
at her health and spirits.
In all her work of army reform she had looked up to him as her “Chief,”
hardly realizing, perhaps, how much of the initiating had been her own. Their
friendship, too, had been almost lifelong, and in every way ideal. The whole
nation mourned his loss, but only the little intimate group which centred in his
wife and children and those dearest friends, of whom Miss Nightingale was
one, knew fully all that the country had lost in him.
It may be worth while for a double reason to quote here from Mr.
Gladstone’s tribute at a meeting held to decide on a memorial.

“To him,” said Gladstone, “we owe the commission for inquiry into
barracks and hospitals; to him we are indebted for the reorganization
of the medical department of the army. To him we owe the
commission of inquiry into, and remodelling the medical education of,
the army. And, lastly, we owe him the commission for presenting to
the public the vital statistics of the army in such a form, from time to
time, that the great and living facts of the subject are brought to
view.”

Lord Herbert had toiled with ever-deepening zeal to reform the unhealthy
conditions to which, even in times of peace, our soldiers had been exposed—
so unhealthy that, while the mortality lists showed a death of eight in every
thousand for civilians, for soldiers the number of deaths was seventeen per
thousand. And of every two deaths in the army it was asserted that one was
preventable. Lord Herbert was the heart and soul of the Royal Commission to
inquire into these preventable causes, and through his working ardour the
work branched forth into four supplementary commissions concerning
hospitals and barracks. When he died, Miss Nightingale not only felt the pang
of parting from one of her oldest and most valued friends, but she also felt
that in this cause, so specially dear to her heart, she had lost a helper who
could never be replaced, though she dauntlessly stood to her task and helped
to carry on his work.
 CHAPTER XXI.
Multifarious work and many honours—Jubilee Nurses—Nursing Association
—Death of father and mother—Lady Verney and her husband—No
respecter of persons—From within four walls—South Africa and
America.

Her activities were so multitudinous that it is difficult even to

name them all in such a brief sketch as this. Besides those at which
we have already glanced, prison reform, help to Bosnian fugitives,
Manchester Police Court Mission for Lads, Indian Famine Fund—
merely glancing down two pages of her biography, I find all these
mentioned. She was herself, of course, decorated with the Red
Cross, but M. Henri Dunant’s magnificent Red Cross scheme for
helping the wounded on the battlefield may be said to have been
really the outcome of her own work and example. For it was the
extension of her own activities, by means of the Red Cross Societies,
which throughout the European continent act in concert with their
respective armies and governments.
She was the first woman to be decorated with the Order of Merit,
which was bestowed on her in 1907, and in the following year she
received, as the Baroness Burdett Coutts had done, the “Freedom of
the City of London,” having already been awarded, among many like
honours, the French Gold Medal of Secours aux blessés Militaires,
and the German Order of the Cross of Merit. On May 10, 1910, she
received the badge of honour of the Norwegian Red Cross Society.
But there was another distinction, even more unique, which was
already hers. For when £70,000 came into Queen Victoria’s hands as
a gift from the women of her empire at the time of her Jubilee, so
much had the Queen been impressed by the work of the Nursing
Association and all that had been done for the sick poor, that the
interest of this Women’s Jubilee Fund, £2,000 a year, was devoted to
an Institution for Training and Maintaining Nurses for the Sick Poor;
and the National Association for Providing Trained Nurses, which
owed so much to Miss Nightingale, was affiliated with it, though it
still keeps its old headquarters at 23 Bloomsbury Square, where for
so many years would arrive at Christmas from her old home a
consignment of beautiful holly and other evergreens for Christmas
festivities. H.R.H. the Princess Christian is President of the Nursing
Association, and Miss Nightingale’s old friend and fellow-worker, Mr.
Henry Bonham Carter, is the Secretary. The influence of Miss
Florence Lees, described by Kinglake as “the gifted and radiant pupil
of Florence Nightingale,” who afterwards became Mrs. Dacre Craven,
and was the first Superintendent-General, has been a very vitalizing
influence there, and the home owes much also to her husband, the
Rev. Dacre Craven, of St. Andrew’s, Holborn. Miss Nightingale’s
warm friendship for Miss Florence Lees brought her into peculiarly
intimate relations with the home, and both the Association and the
Queen’s Jubilee Institute are the fruit of Miss Nightingale’s teaching,
and a noble double memorial of the national—nay, imperial—
recognition of its value.
The Royal Pension Fund for Nurses also, in which Queen
Alexandra was so specially interested, helped to crown the fulfilment
of Miss Nightingale’s early dream and long, steadfast life-work.
But equally important, though less striking, has been the growing
harvest of her quiet, courteous efforts to help village mothers to
understand the laws of health, her pioneer-work in regard to all the
dangers of careless milk-farms, her insistence on the importance of
pure air as well as pure water, though she had always been careful
to treat the poor man’s rooftree as his castle and never to cross his
doorstep except by permission or invitation.
After the death of her father at Embley in 1874—a very peaceful
death, commemorated in the inscription on his tomb, “In Thy light
we shall see light,” which suggests in him a nature at once devout
and sincere—she was much with her mother, in the old homes at
Embley and Lea Hurst, though Lea Hurst was the one she loved
best, and the beech-wood walk in Lea Woods, with its radiant
shower of golden leaves in the autumn, for which she would
sometimes delay her leaving, is still specially associated with her
memory: and her thoughtfulness for the poor still expressed itself in
many different ways—in careful gifts, for instance, through one
whom she trusted for knowledge and tact; in her arrangement that
pure milk should be sent daily from the home dairy at Lea Hurst to
those in need of it.
With faithful love she tended her mother to the time of her death
in 1880, and there seems to be a joyous thanksgiving for that
mother’s beauty of character in the words the two sisters inscribed
to her memory: “God is love—Bless the Lord, O my soul, and forget
not all His benefits.”
After her mother’s death, when the property had passed into the
hands of Mr. William Shore Nightingale, she still visited her kinsman
there and kept up her interest in the people of the district.
Among the outward events of her life, after her return from the
Crimea, one of the earliest had been the marriage of her sister
Parthenope, who in 1858 became the second wife of Sir Harry
Verney,[20] and her home at Claydon in Buckinghamshire was
thenceforth a second home to Miss Nightingale. It need hardly be
said that in Sir Harry Verney’s various generous schemes for the
good of the neighbourhood, schemes in which his wife cordially co-
operated, Miss Nightingale took a warm and sympathetic pleasure.
His keen interest in army reform was, of course, a special ground of
comradeship. Miss Nightingale divided her time chiefly between her
own home in South Street, Park Lane, and visits to the rooms that
were reserved for her at Claydon. One of her great interests while at
Claydon, soon after her sister’s marriage, had been the building of
the new Buckinghamshire Infirmary in 1861, of which her sister laid
the foundation; and her bust still adorns the entrance hall.
Mrs. Tooley reminds us that not only was Lady Verney well known
in literary and political circles, but also her books on social questions
had the distinction of being quoted in the House of Commons. She
gives many interesting details with regard to the philanthropic and
political work of Sir Harry Verney and his family, but it is hardly
necessary to duplicate them here, since her book is still available.
Lady Verney’s death in 1890, after a long and painful illness,
following on that of her father and mother, bereaved Miss
Nightingale of a lifelong companionship, and might have left her very
lonely but for her absorbing work and her troops of friends.
How fruitful that work was we may dimly see when we remember
that—to instance one branch of it only—in ten years the death-rate
in the army in India, which her efforts so determinately strove to
lessen, fell from sixty-nine per thousand to eighteen per thousand.
[21] She strove—and not in vain—to improve the sanitary conditions
of immense areas of undrained country, but she also endeavoured to
bring home to the rank and file of the army individual teaching.
She gives in one of her pamphlets a delightful story of men who
came to a district in India supposed to be fatal to any new-comer,
but, strong in their new hygienic knowledge, determined not to have
cholera. They lived carefully, they grew their own garden produce,
they did not give way to fear, and all, without exception, escaped.
To return for a moment to Britain, since a separate chapter is
reserved for India. She was before her day in contending that foul
air was one of the great causes of consumption and other diseases.
And her teaching was ever given with courtesy and consideration.
How strongly she felt on this and kindred subjects, and how practical
her help was, we see clearly in her letters and pamphlets. She
delighted in making festivities for companies of nurses and of her
other hard-working friends. And in St. Paul’s fine sense of the
phrase, she was no “respecter of persons”: she reverenced
personality, not accidental rank. She had no patience with those
visiting ladies who think they may intrude at all hours of the day into
the homes of the poor, and her quick sense of humour delighted in
many of the odd speeches which would have shocked the prim and
conventional. She thought the highest compliment ever paid to her
staff of nurses who visited in the homes of the poor was the speech
of the grubby ragamuffin, who seemed to think they could wash off
even the blackness of the Arch-fiend and, when being scrubbed,
cried out, “You may bathe the divil.”
But with all her fun and relish of life, how sane, how practical, she
was!
Do you remember how she laughed at the silly idea that nothing
was needed to make a good nurse except what the “Early Victorian”
used to call “a disappointment in love”?
Here are other of her shrewd sayings from her Nursing Notes:—

“Another extraordinary fallacy is the dread of night air.

What air can we breathe at night but night air? The choice is
between pure night air from without and foul air from within.
Most people prefer the latter.... Without cleanliness within and
without your house, ventilation is comparatively useless....
And now, you think these things trifles, or at least
exaggerated. But what you ‘think’ or what I ‘think’ matters
little. Let us see what God thinks of them. God always
justifies His ways. While we are thinking, He has been
teaching. I have known cases of hospital pyæmia quite as
severe in handsome private houses as in any of the worst
hospitals, and from the same cause—viz., foul air. Yet nobody
learnt the lesson. Nobody learnt anything at all from it. They
went on thinking—thinking that the sufferer had scratched his
thumb, or that it was singular that ‘all the servants’ had
‘whitlows,’ or that something was ‘much about this year.’”

If there had been any hope at first that Miss Nightingale might
grow strong enough to stand visibly among those who were being
trained as nurses by the fund raised in her honour, that hope was
now past, and when the great new wing of St. Thomas’s was built—
the finest building for its purpose in Europe—the outward reins of
government had to be delivered over into the hands of another,
though hers was throughout the directing hand. And the results of
her work are written in big type upon the page of history.
In India and America she is acclaimed as an adored benefactress,
but what has she not done for our own country alone? To sum up
even a few of the points on which I have touched: she initiated sick
nursing among the poor, through her special appeal was built the
Central Home for Nurses, she was the pioneer in the hygienic work
of county councils, and, besides the great nursing school at St.
Thomas’s, to her was largely due the reform of nursing in
workhouses and infirmaries. And in 1890, with the £70,000 of the
Women’s Jubilee Fund, the establishment of the Queen’s Nurses
received its charter.
In affairs of military nursing it is no exaggeration to say that she
was consulted throughout the world. America came to her in the
Civil War; South Africa owed much to her; India infinitely more; and
so vital have been the reforms introduced by Lord Herbert and
herself that even as early as 1880, when General Gordon was
waging war in China during the Taiping Rebellion, the death-rate as
compared with the Crimea was reduced from sixty per cent. to little
more than three in every hundred yearly.[22]
We have seen that, though she was so much more seriously
broken in health than any one at first realized, that did not prevent
her incessant work, though it did in the end make her life more or
less a hidden life, spent within four walls, and chiefly on her bed.
Yet from those four walls what electric messages of help and
common sense were continuously flashing across the length and
breadth of the world! She was regarded as an expert in her own
subjects, and long before her Jubilee Fund enabled her to send forth
the Queen’s Nurses, she was, as we have already seen, busy writing
and working to improve not only nursing in general, but especially
the nursing of the sick poor; and unceasingly she still laboured for
the army.
 Repeated mention has been made of General Evatt, to whose
memory of Miss Nightingale I am much indebted.
General Evatt served in the last Afghan campaign, and what he
there experienced determined him to seek an interview, as soon as
he returned to England, with her whom he regarded as the great
reformer of military hygiene—Florence Nightingale. In this way and
on this subject there arose between them a delightful and enduring
friendship. Many and many a time in that quiet room in South Street
where she lay upon her bed—its dainty coverlet all strewn with the
letters and papers that might have befitted the desk or office of a
busy statesman, and surrounded by books and by the flowers that
she loved so well—he had talked with her for four hours on end,
admiring with a sort of wonder her great staying power and her big,
untiring brain.
He did not, like another acquaintance of mine, say that he came
away feeling like a sucked orange, with all hoarded knowledge on
matters great and small gently, resistlessly drawn from him by his
charming companion; but so voracious was the eager, sympathetic
interest of Miss Nightingale in the men and women of that active
world whose streets, at the time he learned to know her, she no
longer walked, that no conversation on human affairs ever seemed,
he said, to tire her.
And her mind was ever working towards new measures for the
health and uplifting of her fellow-creatures.
We have seen how eager she was to use for good every municipal
opportunity, but she did not stop at the municipality, for she knew
that there are many womanly duties also at the imperial hearth; and
without entering on any controversy, it is necessary to state clearly
that she very early declared herself in favour of household suffrage
for women, and that “the North of England Society for Women’s
Suffrage is the proud possessor of her signature to an address to Mr.
Disraeli, thanking him for his favourable vote in the House of
Commons, and begging him to do his utmost to remove the injustice
under which women householders suffered by being deprived of the
parliamentary vote.”[23]

Florence Nightingale’s London House, 10 South Street, Park Lane (house

with balcony), where she died, August 14, 1910.
 Whatever could aid womanly service—as a voice in choosing our
great domestic executive nowadays undoubtedly can—had her
sympathy and interest; but what she emphasized most, I take it, at
all times, was that when any door opened for service, woman should
be not only willing, but also nobly efficient. She herself opened many
such doors, and her lamp was always trimmed and filled and ready
to give light and comfort in the darkest room.
It has been well said that in describing a friend in the following
words, she unconsciously drew a picture of herself:—

“She had the gracefulness, the wit, the unfailing

cheerfulness—qualities so remarkable, but so much
overlooked, in our Saviour’s life. She had the absence of all
‘mortification’ for mortification’s sake, which characterized His
work, and any real work in the present day as in His day. And
how did she do all this?... She was always filled with the
thought that she must be about her Father’s business.”
 CHAPTER XXII.
India—Correspondence with Sir Bartle Frere—Interest in village girls—The
Lamp.

We come now to Miss Nightingale’s most monumental

achievement of all, the reform of sanitary conditions in India—a
reform ever widening and developing, branching forth and striking
its roots deeper. Her interest in that vast population, that world-old
treasury of subtle religious thought and ever-present mystical faith,
may perhaps have been in part an inheritance from the Anglo-Indian
Governor who was counted in her near ancestry. But there can be
little doubt that her ardent and practical desire to improve the
conditions of camp life in India began in her intimate care for the
soldiers, and her close knowledge of many things unknown to the
ordinary English subject. The world-wide freemasonry of the rank
and file in our army enabled her to hear while at Scutari much of the
life of the army in the vast and distant dominions of Burma and
Bengal, and she had that gift for seeing through things to their
farthest roots which enabled her to perceive clearly that no mere
mending of camp conditions could stay the continual ravages of
disease among our men. The evil was deeper and wider, and only as
conditions were improved in sanitary matters could the mortality of
the army be lessened. She saw, and saw clearly, that the reason
children died like flies in India, so that those who loved them best
chose the agony of years of parting rather than take the risks, lay
not so much in the climate as in the human poisons and
putrefactions so carelessly treated and so quickly raised to murder-
power by the extreme heat.
Much of this comes out clearly in her letter to Sir Bartle Frere, with
whom her first ground of friendship had arisen out of their common
interest in sanitary matters.
 What manner of man Sir Bartle was may be divined from a letter
to him written by Colonel W. F. Marriott, one of the secretaries of the
Bombay Government, at the time of his leaving Bombay:—

“The scene of your departure stirred me much. That bright

evening, the crowd on the pier and shore as the boat put off,
the music from the Octavia, as the band played ‘Auld Lang
Syne’ as we passed, were all typical and impressive by
association of ideas. But it was not a shallow sympathy with
which I took in all the circumstances. I could divine some of
your thoughts. If I felt like Sir Bedivere, left behind ‘among
new men, strange faces, other minds,’ you must have felt in
some degree like King Arthur in the barge, ‘I have lived my
life, and that which I have done may He Himself make pure.’ I
do not doubt that you felt that all this ‘mouth honour’ is only
worth so far as it is the seal of one’s own approving
conscience, and though you could accept it freely as deserved
from their lips, yet at that hour you judged your own work
hardly. You measured the palpable results with your
conceptions and hopes, and were inclined to say, ‘I am no
better than my fathers.’ But I, judging now calmly and
critically, feel—I may say, see—that though the things that
seem to have failed be amongst those for which you have
taken most pains, yet they are small things compared with
the work which has not failed. You have made an impression
of earnest human sympathy with the people of this country,
which will deepen and expand, so that it will be felt as a
perpetual witness against any narrower and less noble
conception of our relation to them, permanently raising the
moral standard of highest policy towards them; and your
name will become a traditional embodiment of a good
governor.”[24]

Frere had seen that the filthy condition of many of the roads, after
the passing of animals and the failure to cleanse from manure, was
of itself a source of poison, though the relation between garbage
and disease-bearing flies was then less commonly understood, and
he was never tired of urging the making of decent roads; but this,
he knew, was only a very small part of the improvements needed.
His correspondence with Miss Nightingale began in 1867, and in
that and the five following years they exchanged about one hundred
letters, chiefly on sanitary questions.
It was part of her genius always to see and seize her opportunity,
and she rightly thought that, as she says in one of her letters, “We
might never have such a favourable conjunction of the larger planets
again:
“You, who are willing and most able to organize the machinery
here; Sir John Lawrence, who is able and willing, provided only he
knew what to do; and a Secretary of State who is willing and in
earnest. And I believe nothing would bring them to their senses in
India more than an annual report of what they have done, with your
comments upon it, laid before Parliament.”
In order to set in motion the machinery of a sanitary department
for all India, a despatch had to be written, pointing out clearly and
concisely what was to be done.
Frere consulted Miss Nightingale at every point about this
despatch, but spoke of the necessity for some sort of peg to hang it
on—“not,” he said, “that the Secretary of State is at all lukewarm,
nor, I think, that he has any doubt as to what should be said, or how
—that, I think, your memoranda have fixed; the only difficulty is as
to the when....
“No governor-general, I believe, since the time of Clive has had
such powers and such opportunities, but he fancies the want of
progress is owing to some opposing power which does not exist
anywhere but in his own imagination.
“He cannot see that perpetual inspection by the admiral of the drill
and kit of every sailor is not the way to make the fleet efficient, and
he gets disheartened and depressed because he finds that months
and years of this squirrel-like activity lead to no real progress.”
The despatch with its accompanying documents went to Miss
Nightingale for her remarks before it was sent out. Her commentary
was as follows:—

“I find nothing to add or to take away in the memorandum

(sanitary). It appears to me quite perfect in itself—that is, it is
quite as much as the enemy will bear, meaning by the enemy
—not at all the Government of India in India, still less the
Government of India at home, but—that careless and
ignorant person called the Devil, who is always walking about
taking knowledge out of people’s heads, who said that he was
coming to give us the knowledge of good and evil, and who
has done just the contrary.
“It is a noble paper, an admirable paper—and what a
present to make to a government! You have included in it all
the great principles—sanitary and administrative—which the
country requires. And now you must work, work these points
until they are embodied in local works in India. This will not
be in our time, for it takes more than a few years to fill a
continent with civilization. But I never despair that in God’s
good time every man of us will reap the common benefit of
obeying all the laws which He has given us for our well-being.
“I shall give myself the pleasure of writing to you again
about these papers. But I write this note merely to say that I
don’t think this memorandum requires any addition.
“God bless you for it! I think it is a great work.”[25]

It was a great work, and it might have been delayed for scores of
years, with a yearly unnecessary waste of thousands of lives, if she
had not initiated it.
 Florence Nightingale in her Last Days.

(From a drawing from memory. Copyright A. Rischgitz.)

Her words to Sir Bartle Frere at the outset had been: “It does
seem that there is no element in the scheme of government (of
India) by which the public health can be taken care of. And the thing
is now to create such an element.”
As early as 1863, in her “Observations on the Sanitary State of the
Army in India,” she had written:—

“Native ‘caste’ prejudices appear to have been made the

excuse for European laziness, as far as regards our sanitary
and hospital neglects of the natives. Recent railroad
experience is a striking proof that ‘caste,’ in their minds, is no
bar to intercommunication in arrangements tending to their
benefit.”

Sir C. Trevelyan justly says that “a good sanitary state of the

military force cannot be secured without making similar
arrangements for the populations settled in and around the military
cantonments; that sanitary reform must be generally introduced into
India for the civil as well as the military portion of the community.”
And now that the opportunity arrived, all was done with wise and
swift diplomacy. The way was smoothed by a call from Frere on his
old friend Sir Richard Temple, at that time Finance Minister at
Calcutta, asking him to help.
Those who know India best, and know Miss Nightingale best, are
those who are most aware of the mighty tree of ever-widening
health improvement that grew from this little seed, and of the care
with which Miss Nightingale helped to guard and foster it.
“She was a great Indian,” her friend General Evatt repeated to me
more than once, “and what a head she had! She was the only
human being I have ever met, for instance, man or woman, who
had thoroughly mastered the intricate details of the Bengal land-
purchase system. She loved India, and she knew it through and
through. It was no wonder that every distinguished Indian who
came to England went to see Miss Nightingale.”
 She bore her ninety years very lightly, and made a vision serene
and noble, as will be seen from our picture, though that does not
give the lovely youthful colouring in contrast with the silvery hair,
and we read of the great expressiveness of her hands, which, a little
more, perhaps, than is usual with Englishwomen, she used in
conversation.
It was a very secluded life that she lived at No. 10 South Street;
but she was by no means without devotees, and the bouquet that
the German Emperor sent her was but one of many offerings from
many high-hearted warriors at her shrine.
And when she visited her old haunts at Lea Hurst and Embley she
delighted in sending invitations to the girls growing up in those
village families that she had long counted among her friends, so that
to her tea-table were lovingly welcomed guests very lowly, as well as
those better known to the world.
Her intense and sympathetic interest in all the preparations for
nursing in the South African campaign has already been touched
upon, as well as her joy that some of her own nurses from among
the first probationers at St. Thomas’s were accepted in that
enterprise with praise and gratitude.
It would be a serious omission not to refer my readers to a very
moving letter which she wrote to Cavaliere Sebastiano Fenzi, during
the Italian War of Independence in 1866, of which a part is given in
Mrs. Tooley’s book, and from which I am permitted to quote the
following:—

“I have given dry advice as dryly as I could. But you must

permit me to say that if there is anything I could do for you at
any time, and you would command me, I should esteem it
the greatest honour and pleasure. I am a hopeless invalid,
entirely a prisoner to my room, and overwhelmed with
business. Otherwise how gladly would I answer to your call
and come and do my little best for you in the dear city where
 I was born. If the giving my miserable life could hasten your
success but by half an hour, how gladly would I give it!”

How far she was ahead of her time becomes every day more
obvious; for every day the results of her teaching are gradually
making themselves felt. For example, it can no longer, without
qualification, be said, as she so truly said in her own day, that while
“the coxcombries of education are taught to every schoolgirl” there
is gross ignorance, not only among schoolgirls, but also even among
mothers and nurses, with regard to “those laws which God has
assigned to the relations of our bodies with the world in which He
has put them. In other words, the laws which make these bodies,
into which He has put our minds, healthy or unhealthy organs of
those minds, are all but unlearnt. Not but that these laws—the laws
of life—are in a certain measure understood, but not even mothers
think it worth their while to study them—to study how to give their
children healthy existences. They call it medical or physiological
knowledge, fit only for doctors.”
In her old age, loved and honoured far and wide, she toiled on
with all the warm enthusiasm of a girl, and the ripe wisdom of
fourscore years and ten spent in the service of her one Master, for
she was not of those who ever tried to serve two. And when she
died at No. 10 South Street, on August 10, 1910—so peacefully that
the tranquil glow of sunset descended upon her day of harvest—the
following beautiful incident was recorded in Nursing Notes, to whose
editor I am specially indebted for bringing to my notice the verses in
which the story is told[26]:—

“At Chelsea, under the lime tree’s stir,

I read the news to a pensioner
That a noble lord and a judge were dead—
‘They were younger men than me,’ he said.

“I read again of another death;

The old man turned, and caught his breath—
‘She’s gone?’ he said; ‘she too? In camp
We called her the Lady of the Lamp.’

“He would not listen to what I read,

But wanted it certain—‘The Lady’s dead?’
I showed it him to remove his doubt,
And added, unthinking, ‘The Lamp is out.’

“He rose—and I had to help him stand—

Then, as he saluted with trembling hand,
I was abashed to hear him say,
‘The Lamp she lit is alight to-day.’”

F. S.
 CHAPTER XXIII.
A brief summing up.

Those who write of Florence Nightingale sentimentally, as though

she spent herself in a blind, caressing tenderness, would have
earned her secret scorn, not unflavoured by a jest; for she stood
always at the opposite pole from the sentimentalists, and perhaps
had a little of her father in her—that father who, when he was giving
right and left, would say to some plausible beggar of society who
came to him for wholesale subscriptions, “You see, I was not born
generous,” well knowing that his ideas of generosity and theirs
differed by a whole heaven, and that his were the wider and the
more generous of the two.
She had a will of iron. That is what one of her greatest admirers
has more than once said to me—and he knew her well. No doubt it
was true. Only a will of iron could have enabled a delicate woman to
serve, for twenty hours at a time, with unwearying tenderness and
courage among the wounded and the dying. Even her iron resolution
and absolute fearlessness could not prevent her from taking Crimean
fever when she insisted on visiting a second time the lonely typhus
patient outside Balaclava, at a moment when she was worn out with
six months of nursing and administration combined. But it did enable
her to go back to her post when barely recovered, and, later in life,
even when a prisoner within four walls, who seldom left her bed,
that will of iron did enable her to go on labouring till the age of
ninety, and to fulfil for the good of mankind the dearest purpose of
her heart. Nothing is harder than iron, and that which is made of it
after it has been through the furnace has long been the very symbol
of loyalty and uprightness when we say of a man that he is “true as
steel.”
 Yes, iron is hard and makes a pillar of strength in time of need.
But he who forges out of it weapons and tools that are at once
delicate and resistless, knows that it will humbly shoe the feet of
horses, and cut the household bread, and will make for others
besides Lombardy a kingly crown. And when iron is truly on fire,
nothing commoner or softer nor anything more yielding—not even
gold itself—can glow with a more steadfast and fervent heat to
warm the hands and hearts of men.
The picture of Miss Nightingale that dwells in the popular mind no
doubt owes its outline to the memories of the men she nursed with
such tenderness and skill. And it is a true picture. Like all good
workmen, she loved her work, and nursing was her chosen work so
long as her strength remained. None can read her writing, and
especially her Nursing Notes and her pamphlet on nursing among
the sick poor, without feeling how much she cared for every
minutest detail, and how sensitively she felt with, and for, her
patients.
But such a picture, as will have been made clear by this time,
shows only one aspect of her life-work. One of her nearest intimates
writes to me of her difficulties in reforming military hospitals, and
her determination therefore to give herself later in life to the reform
of civilian nursing; but in reality she did both, for through the one
she indirectly influenced the other, and began what has been
widening and unfolding in every direction ever since.
Those who knew her best speak almost with awe of her
constructive and organizing power. She was indeed a pioneer and a
leader, and girt about with the modesty of all true greatness.
Like Joan of Arc, she heeded not the outward voices, but, through
all faults and sorrows, sought to follow always and only the voice of
the Divine One. This gave her life unity and power. And when she
passed on into the life beyond, the door opened and closed again
very quietly, leaving the whole world the better for her ninety years
in our midst. “When I have done with this old suit,” says George
Meredith, “so much in need of mending;” but hers, like his, was a
very charming suit to the last, and even to the end of her ninety
years the colouring was clear and fresh as a girl’s.
Like all strong, true, disinterested people, she made enemies—
where is there any sanitary reformer who does not?—yet seldom
indeed has any one, man or woman, won deeper and more world-
wide love. But that was not her aim; her aim was to do the will of
her Commander and leave the world better than she found it.
Seldom has there been a moment when women have more
needed the counsel given in one of the letters here published for the
first time, when she begs of a dear friend that her name may be that
“of one who obeys authority, however unreasonable, in the name of
Him who is above all, and who is Reason itself.”
And as we think of the debt the world owes to Florence
Nightingale and of all she did for England, for India, and not only for
the British Empire, but for the world, we may well pause for a
moment on the words that closed our opening chapter, in which she
begs her fellow-workers to give up considering their actions in any
light of rivalry as between men and women, and ends with an
entreaty:—

“It does not make a thing good, that it is remarkable that a

woman should have been able to do it. Neither does it make
a thing bad, which would have been good had a man done it,
that it has been done by a woman.
“Oh, leave these jargons, and go your way straight to God’s
work, in simplicity and singleness of heart.”

The well-remembered words of Ruskin’s appeal to girls in “Sesame

and Lilies,” published but a few years earlier, were evidently in Miss
Nightingale’s mind when she wrote the closing sentences of her
tribute to Agnes Jones—sentences which set their seal upon this
volume, and will echo long after it is forgotten.
 “Let us,” she writes, “add living flowers to her grave, ‘lilies
with full hands,’ not fleeting primroses, nor dying flowers. Let
us bring the work of our hands and our heads and our hearts
to finish her work which God has so blessed. Let us not
merely rest in peace, but let hers be the life which stirs up to
fight the good fight against vice and sin and misery and
wretchedness, as she did—the call to arms which she was
ever obeying:—

‘The Son of God goes forth to war—

Who follows in His train?’

“O daughters of God, are there so few to answer?”

 APPENDIX.
LIST OF PUBLICATIONS BY FLORENCE
NIGHTINGALE.
Letter (on the Madras Famine): The Great Lesson
of the Indian Famine, etc. 1877.
Life or Death in India. A Paper read at the Meeting
of the National Association for the Promotion
of Social Science, Norwich, 1873, with an
Appendix on Life or Death by Irrigation. 1874.
Notes on Hospitals: being two Papers read before
the National Association for the Promotion of
Science ... 1858, with the evidence given to
the Royal Commissioners on the state of the
Army in 1857 (Appendix, Sites and
Construction of Hospitals, etc.).
Do., 3rd Edition, enlarged, and for the most part
rewritten. 1863.
Notes on Matters affecting the Health, Efficiency,
and Hospital Administration of the British
Army, founded chiefly on the experience of
the late war. 1858.
Notes on Nursing: What it is, and what it is not.
1860.
New Edition, revised and enlarged, 1860; another
Edition, 1876.
 Miss Florence Nightingale ovy knitra o oŝctr̂ ování
nemocnŷch. z anglického pr̂ eloẑila. Králova,
1872.
Des Soins à donner aux malades ce qu’il faut faire,
ce qu’il faut eviter. Ouvrage traduit de
l’Anglais. 1862.
Notes on Nursing for the Labouring Classes, with a
Chapter on Children. 1861.
Do., New Edition, 1868 and 1876.
Observations on the ... Sanitary State of the Army
in India. Reprinted from the Report of the
Royal Commission. 1863.
On Trained Nursing for the Sick Poor.... A Letter ...
to The Times ... April 14, 1876.
Sanitary Statistics of Native Nursing Schools and
Hospitals. 1863.
Reproduction of a printed Report originally
submitted to the Bucks County Council in the
year 1892, containing Letters from Miss
Florence Nightingale on Health Visiting in
Rural Districts. 1911.
Statements exhibiting the Voluntary Contributions
received by Miss Nightingale for the Use of
the British War Hospitals in the East, with the
mode of their Distribution in 1854, 1855,
1856. Published, London, 1857.

A LIST OF SOME OF THE BOOKS CONSULTED

In case any of my readers wish to read further for themselves:—
Kinglake’s Invasion of the Crimea. (William
Blackwood.)
Memoir of Sidney Herbert, by Lord Stanmore.
(John Murray.)
Life of Sir Bartle Frere, by John Martineau. (John
Murray.)
Letters of John Stuart Mill, edited by John Elliot.
(Longmans.)
William Rathbone, a Memoir by Eleanor F.
Rathbone. (Macmillan.)
The Life of Florence Nightingale, by Sarah Tooley.
(Cassell.)
Felicia Skene of Oxford, by E. C. Rickards. (John
Murray.)
Memoir of Sir John MacNeill, G.C.B., by his
Granddaughter. (John Murray.)
Agnes Elizabeth Jones, by her Sister. (Alexander
Strahan.)
A History of Nursing, by M. Adelaide Nutting, R.N.,
and Lavinia L. Dock, R.N. (G. P. Putnam and
Sons.)
A Sister of Mercy’s Memories of the Crimea, by
Sister Aloysius. (Burns and Oates.)
The Story of Florence Nightingale, by W. I. W.
(Pilgrim Press.)
Soyer’s Culinary Campaign, by Alexis Soyer.
(Routledge.)
Kaiserswerth, by Florence Nightingale.
Florence Nightingale, a Cameo Life-Sketch by
Marion Holmes. (Women’s Freedom League.)
 Paterson’s Roads, edited by Edward Mogg.
(Longmans, Green, Orme.)
The London Library, No. 3, vol. of The Times for
1910.
Nursing Notes, by Florence Nightingale, and other
writings of Miss Nightingale included in the
foregoing list.

A BRIEF SKETCH OF GENERAL EVATT’S

CAREER.
[As given in Who’s Who.]

Evatt, Surgeon-General George Joseph Hamilton, C.B., 1903; M.D.,

R.A.M.C.; retired; Member, Council British Medical Association, 1904;
born, 11th Nov. 1843; son of Captain George Evatt, 70th Foot;
married, 1877, Sophie Mary Frances, daughter of William Walter
Raleigh Kerr, Treasurer of Mauritius, and granddaughter of Lord
Robert Kerr; one son, one daughter. Educated, Royal College of
Surgeons, and Trinity College, Dublin. Entered Army Medical Service,
1865; joined 25th (K.O.S.B.) Regiment, 1866; Surgeon-Major, 1877;
Lieutenant-Colonel, R.A.M.C., 1885; Colonel, 1896; Surgeon-General,
1899; served Perak Expedition with Sir H. Ross’s Bengal Column,
1876 (medal and clasp); Afghan War, 1878-80; capture of Ali Musjid
(despatches); action in Bazaar Valley, with General Tytler’s Column
(despatches); advance on Gundamak, and return in “Death March,”
1879 (specially thanked in General Orders by Viceroy of India in
Council and Commander-in-Chief in India for services); commanded
Field Hospital in second campaign, including advance to relief of
Cabul under General Sir Charles Gough, 1879; action on the Ghuzni
Road; return to India, 1880 (medal and two clasps); Suakin
Expedition, 1885, including actions at Handoub, Tamai, and removal
of wounded from MacNeill’s zareba (despatches, medal and clasp,
Khedive’s Star); Zhob Valley Expedition, 1890; commanded a Field
Hospital (despatches); Medical Officer, Royal Military Academy,
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!

ebooknice.com