Full text of "Rural water supplies and their purification"

See other formats

ter Supplies
Purification

HOUSTON^

MAIN LIBRARY-AGRICULTURE DE.I

RURAL WATER SUPPLIES
AND THEIR PURIFICATION

ALEXANDER CRUIKSHANK HOUSTON,
M.B., D.Sc., F.R.S.ED.

Director of Water Examination^ Metropolitan Water Board

ILLUSTRATED

LONDON

JOHN BALE, SONS & DANIELSSON, LTD.

OXFORD HOUSE

i, GREAT TITCHFIELD STREET, OXFORD STREET, W.

1918

BY THE SAME AUTHOR :
u STUDIES IN WATER SUPPLY."

(Messrs. MACMILLAN & Co.) 1913.

u RIVERS AS SOURCES OF WATER SUPPLY."
(Messrs. JOHN BALE, SONS & DANIELSSON, LTD.) 1917.

MAIN LFBRARY-AGRICtlLTURE DEPT

THIS LITTLE BOOK IS, WITH PERMISSION,

RESPECTFULLY DEDICATED BY THE WRITER TO

SIR WILLIAM OSLER, BART., M.D., F.R.S.,' F.R.C.P.(LOND.);,

REGIUS PROFESSOR OF MEDICINE, OXFORD UNIVERSITY,

IN APPRECIATION OF

HIS INVALUABLE SERVICES TO MEDICINE,
HIS HIGH LITERARY GIFTS AND HIS UNAFFECTED
MANNERS AND SUNNY GENEROUS DISPOSITION ;
QUALITIES WHICH ATTRACT HIS MANY FRIENDS
AS FORCIBLY AS OPEN SPACES, GREEN FIELDS
AND PLEASANT WATERS APPEAL TO LOVERS OF
THE COUNTRY.

419832

PREFACE.

THE kindly reception given to the writer's
previous books, " Studies in Water Supply "
(Messrs. Macmillan and Co., 1913), and " Rivers
as Sources of Water Supply " (Messrs. John Bale,
Sons and Danielsson, Ltd., 1917), has led him to
try and fulfil a want which has long been felt by
rural dwellers.

Despite the enormous improvements in our
public water supplies, not only as regards quality
and quantity of water, but in respect of its
availability to isolated communities, there remains
a large, but scattered population, whose only source
of a primary necessity of life is the utilization of
rain-water, or the water from shallow wells, neigh-
bouring rivers, streams, or lakes. These private
isolated supplies, lying outside the direct control of
Public Authorities, are frequently polluted and a
source of potential danger to health. It is to be
feared that in many of these cases there is a regret-
table disregard of possible consequences, and
although a minority of individuals remains anxious
to remedy existing evils, adequate knowledge of

PREFACE

how best to proceed is often lacking. Even those
who are usually careless as regards water supply
may soon learn wisdom, or prove amenable to
friendly advice, but unfortunately their would-be
advisers are themselves often at a loss what course
to suggest.

The writer has always been glad to give his
advice freely to professional men — for example,
clergymen and doctors in the country concerned
about the welfare, as regards water supply, of their
parishioners and patients. In answering these
appeals for help, which are often of a pressing and
sometimes of a pathetic character, the writer has
often felt how great the gulf is between theory and
practice, how hopeless it is to attempt, within the
limits of ordinary correspondence, to deal satis-
factorily with so complex a subject, and how useless
it is to give references to books which do not
contain the exact kind of information really required.

A life-long experience of water questions has left
the writer in the position of a learner, and in his
spare time he has endeavoured to acquire a more
practical knowledge of rural requirements. In
trying to teach others, the would-be teacher speedily
learns how much has to be self-learnt, but the
writer's purpose will have been served, if, in this
little book, he can usefully hand on to others the
outcome of some of his own experiences (see p. 1 28).

vi.

PREFACE

The future of this country holds possibilities
which can only be gauged by guess work. One
effect of the War may be to increase enormously the
number of people who like to do things well, and to
decrease correspondingly the number of those who
prefer to let things " slide." There has been un-
questionably a great awakening, and amidst the
excitement and sadness of a colossal war one sees
everywhere signs of quickened interest. Beyond
all doubt, in the ordinary affairs of life, there is a
growing tendency to learn and work and succeed,
not merely from recognition that ignorance and
idleness and failure may spell disaster, but because,
let psychologists explain it as they may, some
influence is at work impelling people to adopt a
saner outlook on life, which was not felt, or in much
less degree, in pre-war days. One direction in
which a change is noticeable is in the growing
appreciation of the value of sanitary science, and in
this connection the question of water supply will
receive its full share of attention. The writer is of
opinion, as already suggested, that in the future it
will become, to a progressive extent, a pleasure and
not a burden to do things efficiently (e.g., to take
a homely illustration, cooking). So also is he
optimistic enough to believe that " scamping" and
ignorance in connection with water supply will
largely disappear. It is true that pessimists declare
that when the War is over there will be a rapid
reversion to the old somewhat careless ways of

vii.

PREFACE

living. Others, however, hope and believe that the
tendency will be quite the other way. Much
depends on how we are taught to regard the
ordinary affairs of life. Education, discipline,
example, if not mere suggestion, can render our
daily tasks almost pleasurably exciting. Speaking
irrespective of any party the writer ventures to
think that a little more imagination (the greatest
gift on earth) on the part of our rulers is needed.
Many hours are spent daily, and spent well, in
cooking operations. Surely it is worth while to
devote a few minutes, say in the morning and
evening, to purifying water for domestic use.

Again, no one knows to what extent our mode
of life may change. At heart we are all country-
folk, and after the War there may be an unprece-
dented rush for the peaceful country, stimulated,
perhaps, by the standardization of houses, more
equitable terms for the purchase of houses and
land, war weariness, shattered nerves, reduced
incomes, difficulty of soldiers reverting to town life
after their experiences at the Front, successful
results of allotment gardening, tariff on imported
foodstuffs, improved railway facilities for the con-
veyance of market produce and recognition of the
value of co-operation, revived love of Nature, alarm
at the falling birth-rate (for which town life is
partly responsible), increasing distrust of over-
centralization, simpler standards of living and dislike

VI 11.

PREFACE

of ostentation, growth of democratic (in the best
sense of the word) ideas, reversion to more primi-
tive (but perhaps saner) ideals, and other factors.
Should these events come to pass, rural water
supplies will become of enormous importance, and
means of purifying contaminated water will be
eagerly sought after. For it is hardly conceivable
that a large, new and widely-scattered rural popula-
tion can be supplied economically now (possibly
not for many years to come) from the existing or
even contemplated public sources of water supply.
Moreover, this book will have failed in its object
if it does not convince the reader that almost any
water supply can be purified to any standard of
safety required.

It would be a grave mistake to suppose that the
writer is wavering in his allegiance to public water
supplies — quite the contrary is the case — but com-
plete belief in responsible control does not imply dis-
belief in making the best of existing circumstances.

The writer feels that some words of explanation
are needed to his potential readers. The actual
expert will perhaps find little in these pages by way
of instruction, unless, indeed, the writer's long
experience as a bacteriologist and epidemiologist
may here and there prove to be of some slight use.
Next come those who by their capabilities and
training (not necessarily technical) are admirably
qualified, prospectively, if not immediately, to help

ix.

PREFACE

themselves, and others, as regards the safety of water
supplies. These may possibly find this little book
a useful stepping-stone to further progress. Lastly,
but most importantly, come those who have grown
a little rusty, those who are inexperienced but
zealous to learn, and those who, whatever their
limitations may appear to be, are determined to
overcome all obstacles. These are the readers to
whom the writer specially makes his appeal.

A moment's reflection will suggest the wide circle
of country dwellers liable to be affected by non-
public water supplies. The dreamy philosopher,
the ascetic, the man of science, the millionaire, the
sportsman, the invalid, the clergyman, the doctor,
the farmer, the rich and the poor.

There is another aspect of the case which the
writer feels deserves recognition.

The principles which underlie the choice and
treatment of water supplies, be they public or
private, are very much the same, and the writer in
studying rural requirements has widened consider-
ably the scope of his own knowledge of the subject
of water supply generally.

Is it too much to hope then that the more
fortunate and more numerous consumers of publicly
controlled water supplies may find in this treatise
some point of importance to stimulate their interest
in a subject which may be as old as the hills, but

PREFACE

which still possesses unexhausted fields of possibili-
ties ? Even the partial realization of such a hope
would far more than repay the writer for the time
and labour he has devoted to the study of rural
water supplies.

Perhaps the best years of the writer's life have
been spent in the service of the Metropolitan Water
Board ; his work has been rendered additionally
pleasant for the reason that although primarily con-
cerned with London's Water Supply, the policy of
his Board has always been in the direction of
improving knowledge of water supplies gener-
ally. Not less grateful is the writer to the members
of his staff, whose active co-operation and valuable
help have been of so pleasant and intimate a
character that he welcomes success more for their
sakes than for his own.

Since writing these notes, the author has read in
the Morning Post of November 9, 1917, a letter
from Lord Sydenham, entitled " Home Settlement
for Ex-Service Men."

The following extracts are of interest : —

" It has been said that few of our sailors and
soldiers will wish to settle on the land, and that
during the period of reconstruction after the War
there will be an immense demand for labour. We
cannot be certain as to what the extent and the

xi.

PREFACE

nature of the labour demanded will be. Nor can
we assume that the men themselves will necessarily
be willing to fall in completely with preconceived
ideas. They have fought for our freedom, and
saved the Empire from disaster. We must consider
the effect that the War has had upon them. Their
nerve-shaking experiences, and their altered outlook
on life, are bound to make a certain number of them
welcome the conditions of life on the land, instead
of in urban surroundings. We must ensure that
the necessary opportunities are provided."

"The Royal Colonial Institute has during the
past two years issued and circulated some hundreds
of thousands of leaflets in the endeavour to let our
sailors and soldiers know what opportunities there
are for them at Home and Overseas. The inquiries
arising from these leaflets are extremely interesting,
because they show that so far the number of men
desiring to settle at home is equal to, if not greater
than, that of those who wish to emigrate to the
Dominions. This fact alone proves the urgent
need for immediate action on the part of the
Government."

In conclusion, the writer ventures to hope that
this book may be of some little use in post-war
settlement, so far as rural water supplies are con-
cerned. (See offer of help on p. 128.)

January, 1918. A. C. HOUSTON.

xn.

CONTENTS.

The book is arranged as follows : —

The first three Chapters are devoted to the study
of rain-water. All the chief methods for sterilizing
and purifying waters are here dealt with in detail,
and it is assumed that the reader will master their
contents as the subsequent Chapters IV. and V.,
(well-waters, springs, rivers, brooks and lake-water)
deal chiefly with generalizations.

In Chapters VI. and VII. the results of actual
experiments are considered, and in Chapter VIII. a
description of the apparatus required is given.

Next comes some miscellaneous information
(nature of chemicals, symbols, atomic weights,
weights and measures, conversions) notes on lime
and chlorine, cost of chemicals, &c., which may

be useful to the non-expert reader.

•
In advance of each Chapter there is a summary

of contents, and at the end a few concluding
remarks.

The index is at the end of the book.

The non-expert reader should not be turned aside
by the amount of technical matter necessarily included
in this book. References to the concluding notes at
the end of each Chapter (pp. 12, 34, 46, 56, 67, 88,
101, 113) and to Miscellaneous Information (p. 114)
will simplify things considerably. See also conclud-
ing remarks on p. 128.

xiii.

ILLUSTRATIONS.

PAGES

FIG. i. — Suspended matter in 0-5 c.c. of rain-water

collected from a roof in London ( x 50 diam.) 8, 9

FIG. 2. — Small conical flask ... ... ... 15

FIG. 3. — One hundred c.c. measuring cylinder ... 15

FIG. 4. — Burette ... ... ... ... 16

FIG. 5. — Burette stand ... ... ... ... 16

FIG. 6.— Small bottle ... ... ... ... 18

FIG. 7. — Cheap form of dispenser's hand balance ... 20

FIG. 8. — A more delicate balance ... ... ... 20

FIG. 9. — Suspended matter in 0-5 c.c. of a river water

during a flood (x 50 diam.) ... ... 60,6 1

FIG. 10. — (a) Anabsena x 400 ; (b) Eudorina x 400 ; (c)
Sponge growth after treatment with acid to
show spicules X 50 ... ... ...66,67

FIG. ii. — (a) Stephanodiscus x 400; (b) Glenodinium
X 150 ; (c) Pandorina x 400 ; (d) Synura x
400 ... ... ... ... ...66,67

FIG. 12. — Mixing pan ... ... ... ... 103

FIG. 13. — Floating-arm method of drawing off clear

water ... ... ... ... ... 103

FIG. 14. — Screw clip ... ... ... ... 103

FIG. 15. — Rubber bung with distal connections ... 104

FIG. 16. — Cork floater at proximal end of rubber tube... 105

FIG. 17. — Rigid outlet method of drawing off clear water 106

FIG. 18. — Small sand and gravel filter ... ... 108

FIG. 19. — Plan of rain-water tank (Kershaw)... ... 125

The writer is greatly indebted to Mr. Kershaw for per-
mission to quote his views on the utilisation of rain-water
(with plan of rain-water tank, p. 125), and to Messrs. Baird
and Tatlock (scientific instrument makers, laboratory fitters,
chemical dealers, &c.), 14, Cross Street, Hatton Garden,
E.G. i, for the loan of blocks in connection with Figures 2, 4,
5, 6, 7, 8, 12, and 14.

xv.

RURAL WATER SUPPLIES AND
THEIR PURIFICATION.

CHAPTER I.
RAIN -WATER.1

First Report (1916) of the Committee formed (1912) for the
Investigation of Atmospheric Pollution — Composition of
normal air — Gaseous and solid impurities of the atmo-
sphere— Results of analyses of rain-water — Questions of
taste — Physical appearances and chemical composition
of rain-water — Health considerations — Sterilization and
purification of rain-water — Concluding remarks (p. 12).

AS a result of a conference of delegates of
municipal authorities and others held in con-
nection with the Smoke Abatement Exhibition in
1912, a Committee was formed for the Investigation
of Atmospheric Pollution. The Committee's first

1 The writer feels that he is justified in placing this source
of water supply first, because it may be the sole refuge of
dwellers in certain rural areas where there is neither a public
water supply nor reasonable hope of providing at a non-
prohibitive cost a private supply (superficial or deep). To
this must be added the fact that a rain-water supply, although
often impure and most difficult to render palatable, is prac-
tically free from the possibility of human fa3cal pollution.
(Set, however, qualifying remarks on p. 10.) Most of our

WATER SUX'PI.IES AND THEIR PURIFICATION

report1 was published in 1916 and contains a great
deal of useful information, bearing directly or in-
directly, on the quality of rain-water.

The following table, taken from the report,
shows the composition of normal air : —

TABLE I.— COMPOSITION OF NORMAL AIR.

Oxygen 20-94

Nitrogen 78-09

Argon 0-94

Carbon dioxide 0-03

Helium, krypton, neon, &c traces

lOO'OO

Unfortunately, falling rain is not only affected by
the presence of these gases, but also by the gaseous
and solid pollutions of the atmosphere resulting
from animal and vegetable life and the manifold
operations of mankind.

Coastguard stations have rain-water supplies for domestic
use. From a British point of view, one of the most interest-
ing water supplies in the world is that of Gibraltar.
For non-potable purposes (e.g., flushing) a brackish water is
used. There is also an emergency distilling plant for furnish-
ing, if required, a potable supply from sea-water, but the
principal drinking supply is from the rain, which apparently
is partly collected from a cemented rock surface, and also
from corrugated iron sheets laid on sandy slopes. The rain
falling on these surfaces is led into underground storage
reservoirs, the locality of which is, of course, necessarily kept
secret. (See " Water Supply," by Dr. W. P. Mason. John
Wiley and Sons.) Information as regards the Bermuda
rain-water supply will be found on p. 124.

1 Lancet, February 26, 1916. Offices of Committee, 47,
Victoria Street, Westminster, London, S.W.

RAIN-WATER

The Committee give the results of the analysis
of the solid matters intercepted by the ventilator
filters of a large institution in London :—

TABLE II. — ANALYSIS OF SOLID MATTERS FROM
VENTILATOR FILTERS.

Sodium chloride trace

Calcium carbonate ... ... ... 2*17 per cent.

Ferric oxide ... ... ... ... 2-44 „

Calcium sulphate ... ... ... ... 5-09 „

Alumina ... ... ... ... ... 834 ,,

Magnesium carbonate 0-33 „

Sand 37-99 „

Carbon ... ... 35-48

Ammonium sulphate ... ... ... 5-77 tt

Tar (extracted by CS2) 1-49 „

Fibrous matter ... ... ... ... 0-95 ,,

100-05 »

The large proportion of carbon, together with
sticky tarry matter, explains why the surface of
all solids, both inside and outside buildings, are
coated with filthy sooty material in large towns.
Not less important in relation to destructive effect
on stone buildings is the presence of sulphur
compounds.

One of the chief sources of these suspended
impurities in the air is the imperfect combustion
of bituminous coal, and the analyses of samples
of the sooty matters produced are given in the
report. (See Table III.)

These are the matters which cling to the walls of
the chimneys and also pass in the form of smoke
into the atmosphere accompanied by many noxious

RURAL WATER SUPPLIES AND THEIR PURIFICATION

gases, e.g., formaldehyde, sulphuretted hydrogen,
carbonic oxide, carbonic dioxide, sulphur oxides,
ammonia, &c.

TABLE III. — ANALYSES OF SOOTY MATTERS.

Sample No. I Sample No. 2

SO8 6-2 percent. ... 5-8 per cent.

H2O ... ... 4-4 „ ... 4-0

CaO ... ... 1-4 „ ... 2-5 „

Fe2O3 2-2 „ ... 2-5 „

Cl ... ... 77 „ ... 7-9 „

Ether extract (tar) ... 28-6 ,, ... 17-7 „

NH3 (combined) ... 4-7 „ ... 3-8 ,,

Other constituents ... 44-8 ,, ... 55*8 „

The oxides of sulphur and ammonia, whether
existing separately as gases or combined to form
the soluble solid ammonium sulphate, appear to
exercise a most destructive action on buildings.

The report sets forth very clearly the advantages
of gas and electric heating and cooking stoves over
coal fires, so far as atmospheric pollution is con-
cerned.

" The products of combustion from a gas fire
consist mainly of CO2 and water vapour, together
with a minute quantity of sulphur oxides. There is
a total absence of those tarry and carbonaceous
substances that are produced by the use of raw
coal, and as the sulphur is so exceedingly small in
amount the gaseous method of heating is compara-
tively innocuous. As regards heating capacity, a
coal fire consuming 2 1 Ib. of coal per day would be

RAIN-WATER

about equal to a gas fire consuming 200 cubic feet
of gas per day, but whereas the coal would evolve
about 1,500 grains of sulphur in various forms, as
well as tar, carbon, &c., already alluded to, the gas
would contribute only 50 or 60 grains of sulphur in
the form of sulphur oxides. Electric stoves do not
evolve products of combustion, and therefore their
use also tends to a cleaner air.

" Apart from the polluting effect of the 20 to 30 Ib.
of sulphur for each ton of coal consumed, it is
regrettable that this valuable material should be
wasted. Its total amount must be huge. The
Committee recognizes that much of this waste of
material is avoided when coal is utilized in the
gasworks.''

In view of what has been said, it is small wonder
that rain-water is apt to be seriously contaminated,
especially near large towns and manufacturing
centres. Rain literally washes the air, and so
inevitably contains, either in solution or suspension,
many impure substances which are really foreign
to a non-vitiated atmosphere.

On the other hand, the reader must preserve a
sense of proportion and remember that the impurities
which contaminate rain-water (collected over a clean
area) are drawn into the respiratory passages with
every breath we take, and yet town dwellers seem
to thrive on the respiratory " capture" (potentially)
of substances which mixed with or dissolved in rain-
water appear to be of almost poisonous significance.

RURAL WATER SUPPLIES AND THEIR PURIFICATION

No doubt the respiratory tract is a purifying and
filtering agency, but so, to some extent, is the
alimentary tract. At all events, the amount of rain-
water we require for drinking purposes is small and
the contained impurities relatively slight when
compared with the huge volume of air which we
breathe and the aggregate amount of apparently
deleterious substances liable to be absorbed or
retained in the process of respiration. Of course,
when rain is collected from areas which concentrate,
so to speak, past atmospheric pollutions and add
to them other contaminations, which are not truly
of aerial origin, the case may be different.

The Committee in seeking to investigate the
pollution of the atmosphere adopted the expedient
of collecting, at a number of centres, the total
rainfall for each month of the year in a standard
circular gauge vessel of four square feet superficial
area, and submitting the liquid to special analysis.

The results are expressed as metric tons (2,205 1°-)
per square kilometre (about 250 acres).

There may be many advantages in expressing
the results in this way, but it would be a great
convenience if, as well, they were expressed as
parts per 100,000 or grains per gallon of rain-water.
As the volume of water collected is always
measured, this could readily be done.

A useful table of the deposit per acre per month
is given on p. 31 of the report.

It should be noted that the nine columns are not

RAIN-WATER

independent of each other. The tar, carbonaceous
matter and ash come under the heading of insoluble
matter. The loss on ignition and ash refer to
the soluble matter, and included in this are the
sulphates, chlorine, and ammonia.

TABLE IV. — DEPOSIT PER ACRE PER MONTH (RAIN-WATER).

Carbon-

aceous

Loss

Tar

other
than

Ash

on
ignition

Ash

Total
solids

S04

NH,

tar

Ib.

CLASS A

0-23

4'5

3'4

6-8

4*5

I '4

0-23

(The Malvern

type)

CLASS B

0-9

I3-5

5'4

0-9

(The Ravens-

court Park or

Cheadle type)

CLASS C

1-8

180

10-8

1-8

(The Liverpool

or Embank-

ment Gar-

dens type)

CLASS D

Over

(The Old ham

2-25

225

2-3

type)

One inch of rain per acre is equal to 22,622 gallons
(about 101 tons, say 100 tons), and 2 to 3 inches
of rainfall per month is a good average figure to
take. Adopting 2*5 inches, 250 tons, or 2,240
x 250 = 560,000 Ib. of rain-water contain the
number of Ibs. of the constituents given in the
table. On this basis the relatively pure Malvern
type of rain-water yields, in parts per 100,000 and
grains per gallon, the following results : —

RURAL WATER SUPPLIES AND THEIR PURIFICATION

TABLE V.— ANALYSIS RELATIVELY PURE RAIN WATER.

Parts per Grains per

100,000 gallon

Insoluble matter: —

Tar 0*041 ... 0*0287

Carbonaceous (other than tar) 0*8 ... 0*56

Ash 1*6 ... ri2

Soluble matter : —

Loss on ignition o'6 ... 0*42

Ash 1*21 ... o'847

Total solids 4-1 ... 2*87

Included in soluble matter: —

Sulphur as SO* 0*8 ... 0-56

Chlorine as Cl 0-25 ... 0*175

Ammonia as NH 0*041 ... 0*0287

For present purposes the Malvern (comparatively
pure) type is best chosen as an example, because
in the vicinity of tqwns Public Water Supplies are
nearly always available, and in such cases it is quite
unnecessary, if not indefensible, to use rain-water
for drinking purposes.

The gauge vessel used by the Committee is
enamelled, and there is a device to safeguard it,
as far as possible, against the " droppings" of birds.

When rain is collected from roofs, the roof
surfaces, and more especially the gutters, harbour
all sorts of accumulated filth (e.g., manure, dust,
grit, decaying vegetation, " droppings " of birds
and small animals, &c.).

It follows that the first " washings," after the
cessation of a period of fine weather, are apt to
be very foul, and 4< separators " (see p, 126) are often
provided which reject this portion and allow of

FIG. i. — Shows the suspended matter in 0-5 c.c. of a sample of rain-water
collected in London (x 50 diam.)-

RAIN-WATER

the collection only of rain-water after this pre-
liminary scouring has taken place.

In considering the possibility of using rain-
water for potable purposes, regard must be had to
questions of taste, chemical composition, physical
qualities, and, most important of all, considerations
of health.1

Speaking generally, there is nothing in the
chemical composition of rain-water to preclude its
use for drinking purposes apart from its physical
qualities and taste.

As regards taste, even pure rain-water is apt to
have a flat, insipid taste, and impure samples are
so objectionable as almost to create feelings of
nausea (see p. 63).

The physical appearances of rain-water are often
most uninviting. It is frequently highly coloured
and contains much suspended matter.

The analysis of a town collected sample of rain-
water gave results (in grains per gallon) as
follows : —

1 The writer feels that he labours under the disadvantage
of trying to appeal both to the expert and non-expert reader.
For the former, advice is hardly needed unless indeed the
writer's experience as an epidemiologist and bacteriologist
may prove to be of some little value. As regards the latter,
an endeavour will be made to show that the ordinary man of
affairs is quite competent to safeguard the purity of any
ordinary rural water supply, although he may have at the
outset to seek expert help.

RURAL WATER SUPPLIES AND THEIR PURIFICATION

TABLE VI.— ANALYSIS OF TOWN RAIN-WATER.

Ammoniacal nitrogen 0*14

Albuminoid nitrogen ... ... ... ... 0*015

Chlorine 0*45

Oxygen absorbed from permanganate ... 0*22

Total hardness 37

Colour (m.m. brown, 2-ft. tube) 160

The high figures for ammoniacal nitrogen,
oxidizable matter and colour are especially note-
worthy.

It is of interest to give for comparative purposes
the corresponding figures (grains per gallon) for the
raw Thames and Thames-derived filtered waters
for the ten-year period ended March 31, 1916.

TABLE VII. — ANALYSIS OF Raw THAMES AND THAMES
Filtered WATER.

Raw Thame
water

Ammoniacal nitrogen 0*005
Albuminoid nitrogen o-oii

s Thames-derived
filtered water

0'OO02
... 0-0043

PIS

Oxygen absorbed from permanganate 0-147
Total hardness 16-3
Colour (m.m. brown, 2-ft. tube) ... 71

... 0-059
... 15-6
... 19

From a health point of view, rain-water has the
advantage that it is practically free from the
possibility of human excremental pollution. That
is, of course, excluding those cases where the
rain-water is stored in pervious underground tanks,
unfavourably situated as regards drains, cesspools,
and other sources of dangerous contamination. On
the other hand, roof collected rain-water may be

RAIN-WATER

contaminated by the "droppings" of birds, the
excreta of rats, mice, and other lower animals, and
a multitude of flies, insects, &c. Some diseases of
the lower animals are communicable to man, and
it would seem not unwise to regard unpurified
rain-water as potentially unsafe, if not actually
dangerous.

The first thing to consider then is how rain-water
can be best sterilized, and, secondly, how it can be
so purified as to be physically free from objection
as regards colour, suspended matter and taste. It
is assumed in what follows that the rain-water
has been collected in bulk in storage cisterns, tanks
or butts and all that is needed is to periodically
purify such proportion of the bulk quantity as will
meet the daily requirements of the household. It
would obviously be out of place to encroach on the
domain of the architect or the engineer and describe
methods of separating the first "washings" and
the best means of collecting and storing rain-water
in bulk (see, however, notes under Miscellaneous
Information, p. 126). It is, however, taken for
granted that the expert in these matters has pro-
vided against the possibility of the stored rain-
water becoming contaminated from outside sources.
For example, a pervious underground tank, badly
situated as regards external sources of pollution,
should, as already suggested, at once be condemned.

In conclusion, the chief points to be noted are
as follows : —

RURAL WATER SUPPLIES AND THEIR PURIFICATION

Putting on one side the dirty discoloured appear-
ances of rain-water, its unpleasant taste, and the
presence of sulphur compounds and tarry matters,
there is nothing chemically objectionable in its use
for drinking purposes.

On the other hand, if the mode of collection of
rain-water is open to criticism (as it usually is),
many undesirable impurities may be added to its
normal composition.

Bacteriologically it is potentially unsafe, or at
least open to suspicion, as, under ordinary con-
ditions, it is liable to be polluted with excrementitious
matter. On the other hand, it is practically free
from the possibility of the worst kind of contamina-
tion, namely, the discharges of human beings (see^
however, qualifying remarks on p. 10).

Assuming it to be possible to remove, or reduce
satisfactorily, the taste,1 suspended matter and colour,
and to kill all harmful microbes without leaving in
the water any noxious substance, rain-water is
perfectly safe for drinking purposes.

The methods of rendering rain-water safe and
palatable will next be described.

1 The question of taste is fully considered on p. 63. The
non-technical reader may find it convenient to concentrate
attention on this page (12) and pass over lightly the preceding
technical paragraphs.

RAIN-WATER

CHAPTER II.

RAIN-WATER— (Continued).

Excess lime method of sterilization — Determination of dis-
solved carbonic acid gas, &c., in rain-water — Estimation
of sulphates or permanent hardness — 10 gallons taken as
the unit — Straining necessary — Amount of lime to be
used — Eight to twelve hours' contact — Neutralization of
the excess lime — Sodium phosphate and sodium bi-
carbonate method — Aluminium sulphate and sodium
carbonate method — " Carbonic acid water " method —
Summary — Simpler "blind " methods — Questions of
duration of contact — Acids as neutralizing agents —
Concluding remarks (p. 34).

EXCESS LIME METHOD.1

Rain-water is very soft and normally contains
so few substances capable of combining with lime
that the use of this substance as a sterilizing agent
is specially attractive in view of its cheap and
innocuous character. In the doses about to be
recommended, a lime-treated water has no taste
(due to lime) and the use of lime is " hallowed by
precedent." From a purely scientific point of view,
these matters may seem to be relatively un-

1 For fuller information the reader is referred to p. 78 of
the author's little book " Rivers as Sources of Water Supply "
(Messrs. John Bale, Sons and Danielsson, Oxford House,
83-91, Great Titchfield Street, Oxford Street, London, W.i).

RURAL WATER SUPPLIES AND THEIR PURIFICATION

important, yet the average person may consider
that they dominate the whole position.

The first thing to determine is how much of
the lime is used up and therefore rendered
non- bactericidal by the dissolved carbonic acid,
bicarbonates, magnesium salts, &c., in the rain-
water. It is well worth acquiring this knowledge
(despite its technical character) because once it has
been mastered all lime sterilization processes become
matters of extreme simplicity and certainty.

Add 20 c.c. of standard lime water (i c.c. =
O'ooi gramme CaO)1 to 700 c.c. rain-water (= 2
grains per gallon) in a stoppered bottle, shake and
allow any precipitate formed to settle.

Rinse out a small flask and 100 c.c. (cubic
centimetres) measuring cylinder (see figs. 2 and 3)
respectively with the settled liquid.

In all chemical operations " rinsing " is of vital
importance, and if the point is not pressed hereafter,
it is assumed that this warning will not be neglected.

Measure out 70 c.c. (cubic centimetres) of the
settled liquid, pour it into the flask, and add a few
drops of phenol phthalein solution (see p. 115); the
liquid ought to turn bright pink. If it does not do

1 Any first class firm of chemical dealers employing well-
trained chemists can supply the standard or other solutions
or chemicals mentioned in this treatise. On p. 114 will be
found a few elementary notes on the various chemicals,
materials, &c., mentioned in this little work. No apology
is needed to the expert if the non -expert reader finds them
of some little use.

RAIN-WATER

so, repeat the experiment using more lime water
(in measured amount).

It should perhaps be explained here that standard
solutions are made up on the metric gramme and

FIG. 2.

c.c.

(about '/s***) (about fufl size)

FIG. 3. — 100 c.c. measuring cylinder.

cubic centimetre basis. If one uses 100 c.c.
(100,000 milligrammes) of water for analysis the
results read as parts per 100,000 parts ; if 70 c.c.

RURAL WATER SUPPLIES AND THEIR PURIFICATION

{70,000 milligrammes) the results read as grains
per gallon, because a gallon weighs 70,000 grains.

Rinse out a burette (see fig. 4) fixed in a burette
stand (see fig. 5) with a little standard sulphuric
acid (i c.c. = O'ooi gramme CaO) and. fill up to
the 0*0 c.c. mark with the acid.

FIG. 4,

FIG. 5.

Add the acid, a few drops at a time, to the
contents of the flask until the pink colour just dis-
appears on shaking, and then take the reading
(say, 4 c.c. of acid used).

Next add a very few drops of methyl orange solu
tion (see p. 115) and proceed with the titration until
the yellow colour begins to change to pink. This

RAIN-WATER

requires a little practice, and it is desirable for com-
parative purposes to add the same amount of methyl
orange to 70 c.c. of rain-water contained in a separate
flask and compare the tints during titration. Now
take the final reading, that is, the total number
of cubic centimetres of acid used up (say, 7 c.c.
required altogether). Then multiply the first phenol
phthalein reading by 2, and subtract the total methyl
orange reading from it 14x2 ==8 — 7==!. Now
as 2 were added (see p. 14) and i remains, the
CaO used up (i.e., rendered inoperative) is equal to
i grain per gallon and this must be allowed for in
calculating the dose of lime required for sterilization
purposes. For example, if the dose for sterilization
purposes is 3 grains, then 4 grains would be required
for each gallon. Should the phenol phthalein
reading doubled be less than the total reading
(methyl orange), this shows that there is no excess
of lime (CaO), and the operation must be repeated
and more lime-water used.

If it is proposed to remove the permanent hard-
ness (due to sulphates, not affected by boiling) as
well as to sterilize the rain-water, the sulphates
present should be determined. This, however,
involves precipitation with baric chloride (using and
therefore wasting at least 1,000 c.c. of rain-water)
and the very accurate weighing of the precipitate
of barium sulphate produced. Another and simpler

* For further information see table on p. 55.
2 17

RURAL WATER SUPPLIES AND THEIR PURIFICATION

plan is to estimate the permanent hardness by
means of the soap test and allow i *o6 (anhydrous)
or 2*86 (crystals) grains of sodium carbonate for
every degree of hardness. The procedure is as
follows : —

Measure out (fig. 3) 70 c.c. of the rain-water,
pour the liquid into a small flask (fig. 2), and boil
until the volume of the water has been reduced
about one-half. Add 25 c.c. of previously boiled
and cooled distilled water and transfer
contents to a measuring cylinder (fig. 3)
and, when cool, add more distilled
water so as to bring the volume of
liquid back to the original 70 c.c.
Transfer this to a small bottle (fig. 6)
and from a burette (figs. 4 and 5) run
in a few drops of standard soap solution
(i c.c. = i grain per gallon CaCO3) at
intervals, shaking vigorously after each
addition until a lather which remains "unbroken"
for a few minutes is obtained. The number of
cubic centimetres of soap used represents the de-
grees of permanent hardness (grains per gallon)
of the water. Suppose 4 c.c. are needed, then
4 x 2*86 = n*44 grains of soda crystals per gallon
are required. Although a useful chemical exercise
the determination of the permanent hardness with
a view to its removal is usually not a matter of
great importance.

RAIN-\VATER

Let it be supposed that 10 gallons is the amount
of water to be dealt with.

Ten gallons of rain-water are transferred from
the rain-water tanks or butts into a suitable vessel.

In Chapter VIII. will be found a description of
the vessel recommended, but the writer feels that
his purpose will have failed if it does not stimu-
late others to improve greatly on the tentative
suggestions here offered.

The water should be strained through wire gauze,
having a sufficiently fine mesh to hold back all but
microscopic objects, and if it is passed through clean
linen or fine muslin as well so much the better.

The lime should not be added as calcium oxide
or quicklime (CaO), but as slaked lime or hydrate
of lime (CaO, H2O). The latter contains theoreti-
cally over 75 per cent, of oxide of calcium (CaO),
but as it is practically impossible to get it absolutely
pure, 60 per cent, (factor 1*67) would be a much
safer figure to adopt.

It will be remembered that 4 grains of calcium
oxide (CaO) were needed, so 4 x 1*67 x 10
= 66'8 grains (say, i drachm, 7 grains) are required.1
It is best to weigh out this amount, mix it into a

1With a small Apothecaries' balance (having a pan of
2 1 in. diam.) the maximum amount of substance (say, of
calcium hydrate) that can be conveniently " massed " on the
pan is about 100 grains. For larger amounts, a letter weight
balance, or for still larger amounts a kitchen balance may be
used. Figs. 7 and 8 represent a cheap dispenser's and a
more delicate balance, respectively.

RURAL WATER SUPPLIES AND THEIR PURIFICATION

paste with a little rain-water in a small cup and
transfer the mixture to the 10 gallons of rain-water,
rinsing the cup out several times afterwards to ensure
that all the lime has been washed out. Then stir

FIG. 7.

FIG. 8.

thoroughly for five to ten minutes and leave for
eight to twelve hours to allow of sterilization taking
place.

Of course, if preferred, the lime may be added as

RAIN-WATER

" milk of lime." Weigh out ij oz. (Imperial) of
slaked lime on an ordinary letter weight balance,
and, if accuracy is aimed at, add to it 1 2 grains of
lime weighed on Apothecaries' scales. Transfer
the lime to a 12 oz. bottle, having a mark at
10 oz. capacity and containing, say, 5 oz. of water.
Shake vigorously for several minutes, then make
up with water to the 10 oz. mark. One oz. of
the shaken mixture, measured quickly, will be the
correct dose, approximately, for TO gallons.

Another method is to use lime-water. The
solubility of CaO is nearer i in 900 than i in 1,000
but having regard to loss of strength on keeping,
the latter is the safer figure to adopt. Further,
for the sake of simplicity, the dose may be taken
at 70 instead of 66*8 grains per 10 gallons. Then
the proportions are 0*9 gallon (=7 pints, 4 oz.)
of lime-water to 9 gallons of rain-water. The
lime-water may be prepared in bulk as follows :
Fit a barrel, capable of holding rather more
than 30 gallons, with a stopcock, about 3 inches
from the bottom. Raise a foot or two from the
ground for " drawing-off" purposes. Pour 2 or
3 gallons of rain-water into the barrel and add
an excess (say, i Ib.) of slaked lime (CaO, H2O),
mixing the lime and water very thoroughly. Then
bring up to 30 gallons, stirring all the time and
place a tightly fitting lid on the mouth of the barrel.
The clear liquid drawn off from the tap will be
approximately of the strength of i in 1,000.

RURAL WATER SUPPLIES AND THEIR PURIFICATION

The next stage is neutralization of the excess
lime. Rinse out a 100 c.c. measuring cylinder and
small flask with the liquid. Then measure out
70 c.c. and transfer to flask and estimate the calcium
oxide (CaO), using phenol phthalein and methyl
orange in the way indicated previously (p. 16).
Suppose the phenol phthalein and methyl . orange
" readings " are 3*5 and 4 respectively ; then
3-5 x 2 = 7-4 = 3 x I0 (f°r 10 gallons) = 30
is the number of grains of calcium oxide (CaO/
which require neutralization. There is a con-
siderable number of ways of effecting this object.
(See, however, first paragraph, p. 31.)

(i) Sodium hydrogen phosphate (Na2 HPO4, 12
H2O)1 may be used, the practically insoluble calcium
phosphate being precipitated as a flocculent white
deposit and sodium hydrate being left in solution.
(See Table XVII., p. 92.)

1 68 716 (factor 4*26)

3 CaO + H20 + 2 Na2HP04, 12 H2O = Ca3 (PO4),
+ 4NaHO+ i2H20.

The amount required is obviously 30 x 4*26 =
127*8 grams (say, 2 drachms, 8 grains)2 of sodium
phosphate.

1 The writer has purposely chosen, as far as possible, pre-
parations which are used medicinally and are to be found in
the British Pharmacopoeia.

2 Weights may be purchased from 10,000 grains down
to T^j grain which avoid all the complications of scruples,

RAIN-WATER

As it is perhaps undesirable to leave sodium
hydrate (NaHO) in solution, sodium bicarbonate
(NaHCO3) may be added to change it into carbonate
of soda (Na2CO3), which in its turn will act on the
calcium sulphate (CaSO4 permanent hardness).

1 60 336 (factor 2-0).

4 NaHO + 4 NaHCO3 == 4 Na.2 CO3 + 4 H,O.

The amount required is 30 x 2*0 = 60 grains
(i drachm) of sodium bicarbonate (NaHCO3), which
theoretically would produce about 76 grains of
sodium carbonate (Na2CO3) or enough to remove
about 7 degrees of permanent hardness.

This is in excess of the permanent hardness of
country rain-water, but sodium carbonate in small
doses is quite innocuous.

Na.2CO3 + CaSO4 = Na2SO4 + CaCO8.

The sodium sulphate (Na2SO4) produced is soluble,
innocuous and does not cause curdling of soap.
The calcium carbonate (CaCO3) is soluble only to
the extent of about 1*4 grains per gallon.

127*8 grains (say, 2 drachms, 8 grains) of sodium
phosphate are therefore added to the excess-limed
rain-water and the mixture stirred for a few minutes,

drachms and ounces (Apothecaries' weights) and ounces and
pounds (Imperial weights). Both as regards weights and
measures it seems a great pity that we do not finally adopt the
simple and scientific systems practised by Continental
nations.

RURAL WATER SUPPLIES AND THEIR PURIFICATION

and then 60 grains (i drachm) of sodium bicarbonate
added as well and the mixture again stirred.

After settlement, preferably for eight to twelve
hours, the clear liquid is ready for domestic use.

(2) Alternatively, Aluminium sulphate crystals
(A12 (SO4)3, 1 8 H2O) may be used to neutralize the
excess lime. (See p. 78.)

1 68 666 (factor 3*96)

3 CaO + A12 (S04)3, 1 8 H2O = A12O3 + 18 H2O
+ 3 CaSO4.

The amount required is thus 30 x 3*96 = 118*8
grains (nearly 2 drachms) of crystallized sulphate of
alumina. As this would not remove the permanent
hardness formed (CaSO4 being soluble), it may be
desirable to add subsequently carbonate of soda
(crystals).

408 858 (factor 5'ii).

3 CaSO4 + 3 Na2CO3, 10 H2O = 3 Na2SO4 +
3 CaCO3 + 30 H2O.

Hence 30 X 5*11 = 153*3 grains (2j drachms,
3 grains) of sodium carbonate are required.

If it is required as well to remove the original
permanent hardness of the rain-water, an additional
28*6 grains of sodium carbonate (crystals) for each
degree of hardness is required per 10 gallons.

1 1 8* 8 grains (say, 2 drachms) of aluminium sul-
phate are therefore added to the excess-limed rain-

RAIN-WATER

water, the mixture stirred for a few minutes and
then 153*3 grains (2^ drachms, 3 grains, or more if
the original permanent hardness has to be removed)
of sodium carbonate added as well and the mixture
again stirred.

Unfortunately, sodium carbonate has a solvent
action on aluminium hydrate, and instead of acting
on the sulphates may have a tendency to interfere
with the coagulating effect of the aluminium sul-
phate. In the case of lime-treated peaty waters, at
all events, the writer has obtained better clarification
by its omission.

After settlement, preferably for eight to twelve
hours, the clear liquid is ready for domestic use.

(3) Instead of the phosphate or aluminium sul-
phate methods, the excess CaO could be neutralized
directly and simply with sodium bicarbonate (NaH
CO3), although the clarifying action is apt to be
much less satisfactory (see p. 96).

1 68 504 (factor 3). 318 (factor '63).

3 CaO + 6 NaHCO3 = 3 Na2 CO3 + 3 CaCO3 +
3H20.

30 x 3 = 90 grains (\\ drachms) of sodium
bicarbonate (NaHCO3) required which by producing
56*7 grains of Na2CO3 (90 x "63 = 56*7) would
remove about 5 to 6 degrees of permanent hardness.

Na2CO3 + CaSO4 = Na2SO4 + CaCO3.

This is about the hardness of a town rain-water,

RURAL WATER SUPPLIES AND THEIR PURIFICATION

but country rain-water would be softer. Any slight
excess of Na2CO3, however, is, as already suggested,
harmless, possibly even beneficial.

Another plan would be to place the 90 grains of
sodium bicarbonate (NaHCO3) in a flask with a little
water, fit the neck with a rubber bung and bent glass
tube, one end of which is placed in the lime-treated
water. With a spirit lamp heat the contents of the
flask and boil briskly for a few minutes ; the carbonic
acid gas (CO2) will be driven off and combine with
the CaO to form the relatively insoluble carbonate
of lime.

6 NaHCO3 = 3 Na2CO3 + 3 H2O -f 3 CO2
3 C02 + 3 CaO = 3 CaCO3.

Speaking of the first named plan, 90 grains
(ij drachms) of NaHCO3 are added to the excess-
limed rain-water and the mixture stirred for several
minutes. After settlement, preferably for eight
to twelve hours, the clear liquid is ready for
domestic use.

(4) Another very convenient way of neutralizing
the excess CaO is to prepare carbonic acid water
(see p. 99). For this purpose Gasogene or Sparklet
Syphons may be used. Full instructions are issued
with the apparatus and all that need be said here
is that the water used should be above suspicion.
Perhaps the simplest way is to bring 30 oz. of
rain-water to the boil and add to it 10 oz. of non-
26

RAIN-WATER

boiled rain-water. When cool, the mixture will be
free from the germs of water-borne epidemic disease,
and suffice in volume for the 2 -pint size of apparatus.
After the water has been charged with carbonic
acid gas, a portion of the excess-limed rain-water
(say, 2 oz.) is placed in a small bottle or flask and
phenol phthalein added, when the liquid will turn
bright pink. A little of the carbonic acid water is
run into any convenient vessel and 10 c.c. of it
drawn up into a graduated 10 c.c. pipette, and
this is added very gradually to the liquid in the
flask until the pink colour disappears. Half the
amount used multiplied by Soo1 gives the volume,
in cubic centimetres, required to neutralize 10 gallons
of the excess limed rain-water.

CaO + CO2 = CaCO3.

After settlement, preferably for eight to twelve
hours, the clear liquid is ready for domestic use.

An excess of carbonic acid water should prefer-
ably not be added as then the relatively insoluble
carbonate of lime (CaCO3) is re-dissolved, forming
bicarbonate of lime, thus hardening the water.

CaCO3 + H2O, CO2 = CaCO3, H2CO3.

There would, of course, be no objection to
adding an excess of carbonic acid water subse-
quently, that is, after the clear liquid has been
separated from the deposited carbonate of lime.

1 Because 2 oz. = ^J^ part of 10 gallons.
27

RURAL WATER SUPPLIES AND THEIR PURIFICATION

Indeed, such a procedure would give "bite" and
flavour to the final product.

SUMMARY AT THIS STAGE.
Ten gallons of rain-water require to be treated.

(a) Determine, in the way already described, the
amount of CaO used up by the CO2, &c., in the
rain-water (p. 14).

(b) Determine the permanent hardness (see p. 18).

(d) Estimate, in the manner previously set forth,
the amount of CaO left over after sterilization
per 10 gallons (p. 22).

(e) For sodium phosphate and sodium bicarbonate
treatment, multiply figure obtained under (d) by
4*26 and 2*0 respectively (p. 22).

For aluminium sulphate and sodium carbonate
(see, however, remarks on p. 25) treatment, multiply
figure obtained under (d) by 3*96 and 5*11
(4- 28*6 for every degree of permanent hardness)
respectively (p. 24).

(g) For direct sodium bicarbonate method,
multiply figure obtained under (d) by 3 (p. 25).

RAIN-WATER

(h) For carbonic acid water method use a
Gasogene or Sparklet apparatus and add a
neutralizing quantity of the CO2 charged water to
the excess-limed rain-water (p. 26).

The purification processes here described offer
to the student of chemistry no difficulties either in
principle or practice. Nevertheless, there are many
persons who will be discouraged by the technical
terms used, and by the apparent complexity of the
procedure. For these some simpler (although
admittedly " blind ") methods may be described.

A fairly safe figure for the amount of oxide of
lime (CaO) used up by the dissolved carbonic acid
gas, &c., in freshly collected samples of rain-water
would be one grain per gallon. Sixteen samples of
freshly collected samples of town rain-water gave
an average of 0*8. Of course, in the case of rain-
water stored for a long time in tanks containing
much decomposing organic matter the water might
be almost saturated with CO2, and, if the tanks
were made of concrete, bicarbonates would be
formed which would also use up a certain pro-
portion of the lime (CaO). In these cases actual
tests ought certainly to be made.

A reasonably safe figure for bactericidal purposes
would be about 3 grains per gallon of lime (CaO)
with about twelve hours' contact. Very impure
samples may possibly require a larger dose, or more
prolonged contact. On the other hand, the dose
suggested is in excess of ordinary requirements.

RURAL WATER SUPPLIES AND THEIR PURIFICATION

The reason for giving about twelve hours for
sterilization and about twelve hours for subsequent
neutralization, clarification, &c., is that the first part
of the process may conveniently be started in the
morning and the final part in the evening, the com-
pletely treated water being ready for drinking on
the following morning. This involves, of course, the
tanks or vessels for treatment being in duplicate.
The important point is that rapid chemical steriliza-
tion processes labour, in the writer's experience,
under the serious disadvantage of ^/^r-dosing, that
is, of using far more of the chemical than is actually
required owing to the duration of contact being
unduly curtailed. This is apt to be specially true
when the water to be sterilized is at a low tempera-
ture. Under abnormal conditions short contact
may be absolutely unavoidable, but, generally
speaking, the practice is to be condemned.

Slaked lime or hydrated lime (CaO, H2O) con-
tains nominally over 75 per cent, of oxide of lime
(CaO), but 60 per cent, would be a safer figure
to adopt (factor 1*67).

Hence (i + 3) x 1*67 = 6*68 grains (i drachm,
7 grains) per gallon is the amount of slaked or
hydrated lime (CaO, H2O) required to sterilize
rain-water within twelve hours.

Dealing in round numbers, 67 grains (i drachm,
7 grains) of slaked lime are therefore added to
10 gallons of rain-water, the mixture stirred for five
to fifteen minutes and then left for twelve hours to

effect sterilization.

RAIN-WATER

Such a water could be drunk with perfect safety
without any further treatment, the amount of excess
lime being too small to give rise to any taste of lime,
or to produce any injurious effects when ingested.

On the other hand, .the process is incomplete
and might be considered to call for neutralization
of the excess lime, with, perhaps as well, clarification
by coagulants.

Hence, the following additional processes are
put forward tentatively.

For the sake of simplicity it is assumed that,
after sterilization, the excess lime (CaO) is equal
to 30 grains per 10 gallons.

Add to the sterilized 10 gallons of rain-water one
or other of the following substances : —

Sodium hydrogen Aluminium sul- Sodium bi- Neutralizing

phosphate, 128

phate, 119 grains

carbonate,

quantity of

grains (2 drachms,

(about 2 drachms).

90 grains

carbonic

8 grains). Then

Then (see, how-

(l£ drachms).

acid water.

sodium bicarbo-

ever, remarks on

nate, 60 grains

p. 25) sodium car-

(i drachm).

bonate, 153 grains

(2\ drachms, 3

grains).

The first two processes aim not only at neutraliza-
tion, but clarification and softening. The last two
exercise a softening influence, but their clarifying
effect is negligible.

In provisionally fixing the bactericidal dose of
excess CaO at 3 grains per gallon, regard has been
had to the duration of contact (eight to twelve
hours), providing a wide margin of safety, and the

RURAL WATER SUPPLIES AND THEIR PURIFICATION

variable quality of the rain-water. With a contact of
one to seven days the dose may safely be reduced
from three (assuming this to be sufficient with eight
to twelve hours' contact) to from about two to
mere traces. The matter requires furthur study,
but the doses about to be suggested probably err on
the side of safety, as the beneficial effect of storage
would be (so to speak) superimposed on the
germicidal action of the lime. Say, 2, i, 0*5, 0-25,
0*125, 0*0625 and 0*03125 grains of CaO per gallon
for one, two, three, four, five, six, and seven days'
contact (plus i grain per gallon in each case for
CO2, &c., see p. 29).

Putting on one side questions of clarification, &c.,
the "rain-water purifier" might perhaps not un-
reasonably prefer to rely on time rather than on
dose, and have tanks in duplicate for alternate use
to hold either seven, six, five, four, three, two or
one days' supply. (See Table XVI., p. 86.)

Working on a ten-gallon a day basis the dose
and tankage accommodation would, if the fore-
going assumptions are sound, be approximately
as shown in Table VIII.

Of course, especially with the smaller doses here
given, it is desirable first to determine accurately
the amount of CaO used up by the dissolved CO2,
&c., in the rain-water, and then calculate the
amount of hydrate of lime to be added so as to give
an excess of CaO of 2, i, 0*5, &c., &c.

Indeed, having regard to the innocuous, if not

RAIN-WATER

beneficial, effect of minute doses of lime, a minimum
dose of 0*25 excess CaO, with not less than four
days' contact, would seem to be the wisest course
to pursue.

TABLE VIII.— LIME, TIME AND TANK CAPACITY IN RELATION
TO STERILIZATION.

Ten Gallons a Day Basis, Tanks being in Duplicate and used Alter-
nately. (Doses based partly on experimental data, and partly on
empirical grounds?)

Hydrate of lime (CaO, H2O) <* parity

<Srains> (gallons)

1 day's contact 50'!, say 50 (approximately 2 CaO in excess) 10

2 » 33'4 » 33 j» i »> i« 20

25-05 „ 25 „ 0-5

20-875 », 21 „ 0-25 „

18788 „ 19 „ 0-125 »»

17744 „ 1 8 „ 0-0625 „

17-222 „ 17 „ 0-03125 „

30
40

50
60

For health or taste purposes, and putting on one
side questions of clarification or hardness, water
treated in this manner could be safely drunk without
any further treatment.

Some reference should perhaps also be made to
the use of acids for neutralization purposes. The
writer sees no objection to the use of either mineral
or organic acids in this connection, provided the
treatment rests in safe hands. Of the mineral acids,
phosphoric acid (the dilute phosphoric acid of the
Pharmacopoeia, acidum phosphoricum dilutum may
be used), has the advantage of producing the
practically insoluble calcium phosphate, and so re-
ducing the hardness of the " lime-treated " water.
3 33

RURAL WATER SUPPLIES AND THEIR PURIFICATION

On the other hand, the organic citric or tartaric acids
are crystalline, and so can be added by weight.
They are medicinal preparations and practically
non-poisonous. In combination with lime, the salts
produced have an anti-scorbutic tendency, and, in
the amounts involved, are perfectly innocuous.
Each grain per gallon of excess CaO requires
1*24 and 2*66 grains of citric and tartaric acids
respectively.

In conclusion, the chief points to be noted are as
follows : —

Rain-water can be sterilized successfully with an
innocuous dose of lime, say, 67 grains of slaked lime
per gallon of water. (See also Tables XL, XIV.,
XVI. and XVII.)

If questions of neutralization of the excess lime
and of clarification and softening are considered
necessary or desirable, these results can be achieved
by comparatively simple means.

Granting the approximate accuracy of certain
assumptions, the processes involved require no
chemical knowledge, and become a matter of mere
simple weighing of the necessary ingredients.

The non-expert reader may find it desirable to
pay most attention to pp. 29 to 34.

Consideration may next be given to the steriliza-
tion of rain-water by means of heat and by the use
of chlorine compounds.

RAIN-WATER

CHAPTER III.

RAIN-WATER (Continued).

Sterilization by means of heat — three parts boiled, one part
not boiled method — Clarifying methods — Sterilization by
means of chlorine — Bleaching powder — " Chloros " —
Electrolytic compounds — Prejudice against "doping"
waters — Doses of chlorine — How administered — Use of
sodium sulphite — Combination with clarification methods
— Combined lime and chlorine methods — How best
applied — Neutralization of the proportion treated with
lime — Less chance of taste — Liquid chlorine — Electro-
lytic compounds — Ozone — Ultra-violet rays — Household
devices—" Halazone " — Concluding remarks (p. 46).

STERILIZATION BY MEANS OF HEAT.

This method does not aim directly at clarifying a
water, or eliminating taste, or purifying it chemically.

Nevertheless, from the point of epidemic water-
borne disease, it has no equal, and it differs from all
other kinds of treatment, inasmuch as it is applicable
to every kind of water (rain-water, river-water,
surface-water, spring- water, well-water, &c.).

The old rule was to boil a water violently for at
least five minutes ; this procedure was based partly
on empiricism, and partly on a knowledge of the
limitations of human nature.

It has the disadvantage of robbing a water of

RURAL WATER SUPPLIES AND THEIR PURIFICATION

most of its dissolved gases, and therefore rendering
it less palatable.

A perfectly safe rule to adopt is to bring three
parts of water barely to boiling point (212° F. ;
1 00° C.), and add one part of " unboiled" water.
The mixture within five minutes will be innocuous,
whatever the temperature of the water was ante-
cedently, and the dissolved gases will only be partly
dissipated. For example, even with water initially
at the freezing point, the temperature of the mixture
would be 167° F. (75° C.), a temperature more than
sufficient to kill the typhoid bacillus within five
minutes (see Miscellaneous Experiments, p. 98).

Of course, if it is desired as well to reduce the
colour and precipitate the bulk of the suspended
matter, heat may be used for sterilization purposes,
and aluminium sulphate as a coagulant. About
3 to 6 grains of aluminium sulphate and 3-9 to
7*8 grains of sodium carbonate crystals may be used
for this purpose per gallon of rain-water.

Perhaps a few words should be added as regards
distillation processes. Practically any water may be
used for distillation purposes, and distilled water is
absolutely safe from a bacteriological point of view.
Most people dislike the taste of distilled water, but
the writer disagrees with those who consider it
actually injurious to health. Dr. W. P. Mason1
quotes Surgeon-General W. C. Braisted, of the
U.S. Navy, as saying: —

1 " Water Supply," by Dr. W. P. Mason (John Wiley and Sons).

RAIN-WATER

" The use of distilled water on ships of the Navy
has always tended to the very best of health condi-
tions. In my opinion the use of distilled water
offers the ideal drinking fluid for human con-
sumption."

Of course, unless the water is brackish, or con-
tains an injurious proportion of salts, or for some
other reason cannot be drunk directly, there is no
occasion to distil it, heating in the way already
described being all that is required. As distillation
is only required in quite exceptional circumstances,
and as the provision of a still and condensing and
heating apparatus is a mere matter of purchase,
involving no special knowledge, the matter need
not be considered further. Any responsible firm
dealing in laboratory apparatus can supply full
information on the subject.1

STERILIZATION BY MEANS OF CHLORINE.*
There are at least four ways of utilizing this
method : —

(1) Bleach solution. Bleaching powder or chloride
of lime should contain about 33 per cent, of avail-
able chlorine.

(2) Chloros (solution of sodium hypochlorite)
contains 10 to 15 per cent, of available chlorine.

1 For large plants to deal with sea-water, p. 1820 of
" Kelly's London Directory" (1917), may be consulted.

2 Bromine and Iodine have also been employed for sterili-
zation purposes, but their use cannot be considered here.

RURAL WATER SUPPLIES AND THEIR PURIFICATION

(3) Liquefied chlorine gas (100 per cent, value).

(4) Electrolytic compounds. These are prepared
electrolytically from various chlorides (e.g., sea-
water), and the percentage of available chlorine
depends on the concentration of salt and other
factors.

It seems a great pity that so much prejudice
surrounds the use of chlorine for sterilization pur-
poses.1 As a pioneer (Lincoln Water Supply,
1905) on this subject, the writer has had to contend
with a great deal of well-meant criticism. After all,
it is only natural to look with suspicion on what
many persons regard as a "doped" water. It is
foolish to try and ride roughshod over the opinions
of others merely because they do not coincide with
our own. Yet if this War has taught us anything,
it is the triumph of expediency over sentimentality.
When all has been said, the safety of a water supply
comes first, and if this end can only be achieved by
chemical treatment, all other considerations should
occupy a secondary position.

Rain-water varies so much in chemical composi-
tion that it is difficult to suggest a dose of chlorine
suitable for all, or, perhaps, even a majority of,
cases. For example, unless suitably protected from
outside sources of contamination, rain-water may
contain much decomposing vegetable matter, many
insects, and even the bodies of mice, rats, and birds.

JSee Chapter V., "Studies in Water Supply," and
Chapter IV., " Rivers as Sources of Water Supply."

RAIN-WATER

Strictly speaking, the strength of the bleaching
powder and chloros in terms of available chlorine
should always be determined. But as this involves
skilled chemical knowledge, experience and tech-
nique, it has been assumed in what follows that
bleaching powder contains 33 per cent, (factor 3),
and chloros 12 per cent, (factor 8*3) of available
chlorine.

On the whole, it would seem wisest to employ a
somewhat larger dose than is actually required,
and to counteract the effect of this by prolonged
contact, and, if necessary, the final use of an
" anti-chlor."

A dose of 0*5 part of available chlorine per
million parts of water (i in 2 millions) with twenty-
four hours' contact, and then the addition (if found
desirable) of a de-chlorinating dose of sodium
sulphite is recommended. This corresponds with
1*05 grains of bleaching powder or 0*00664 oz. of
chloros per 10 gallons of water.

As these are inconveniently small amounts to
deal with (by liquid or measure) it is desirable to
work with standard solutions, as follows : —

Weigh out 1 0*5 grains (breaking up any lumps)
of bleaching powder and transfer to a small bottle
containing 2 oz. of rain-water and some lead shot.
Shake thoroughly for several minutes, allow to
settle and pour the more or less clear liquid into a
10 oz. measuring glass. Add another 2 oz. of
rain-water to the small bottle, shake vigorously,

RURAL WATER SUPPLIES AND THEIR PURIFICATION

allow to settle and pour the clear liquid into the
measuring glass. Repeat this procedure twice
more so as to ensure that practically all of the
active part of the bleaching powder has been
extracted, and then finally make up the measuring
glass to 10 oz. with rain-water ; stir, allow to settle,
and transfer the clearer portion to a stoppered bottle.
The dose is now obviously i oz. for every 10
gallons of rain-water.

Alternatively, the 10-5 grains may be put in a
mortar and mixed repeatedly with rain-water, the
clearer portion being transferred each time to a
10 oz. measuring cylinder, Finally, rain-water is
added up to the 10 oz. mark, the mixture stirred,
then allowed to settle, and the clear portion poured
into a stoppered bottle.

As regards the chloros, if this is diluted with
19 volumes of rain-water ( = i in 20), then 1^0624
(say i) drachm of the mixture per 10 gallons gives
a dose in available chlorine of 0*5 per million.

Ten gallons of rain-water are placed in a suitable
vessel and either i oz. of bleach solution or i drachm
of diluted chloros added, and the mixture stirred for
a few minutes.

After twenty-four hours, measure out 8 oz. of
the now-sterilized rain-water into a convenient
bottle or flask. Take the temperature of the liquid,
if it is below 10° C. add a little freshly prepared
potassium iodide and starch solution (see p. 116).
If above 10° C., the liquid should, before testing, be

RAIN-WATER

cooled, if practicable, with a mixture of ice and salt
to below 10° C., as otherwise the delicacy of the
test is much impaired.

If there is no blue reaction, or only the very
faintest tint, no sodium sulphite need be added and
the liquid may be drunk with impunity. On the
other hand, if there is a decided blue reaction,
sodium sulphite solution (o'i percent.) should be run
in from a burette and the number of cubic centimetres
needed to remove the blue colour " read off."
Obviously, this amount X 200 will give the dose of
sodium sulphite required. Those who are averse
to making actual tests might proceed on the basis
that each part of available chlorine requires about
3-5 parts of sodium sulphite. With the dose here
suggested (0*5 per million) 1*225 grains of sodium
sulphite would be required. As, however, all, or
nearly all, the available chlorine is likely to have
been used up at the end of twenty-four hours, it is
probable that about J grain of sodium sulphite
would be sufficient. It is perhaps unnecessary to
warn the reader that on no account must the sodium
sulphite be added until the chlorine has exercised
its sterilizing action. This, of course, would have
the effect of destroying the bactericidal effect of the
chlorine.

The chlorine treatment may be combined with
clarification, but if aluminium sulphate is used, care
must be exercised not to add more than will be
neutralized by the alkalinity of the rain-water.

RURAL WATER SUPPLIES AND THEIR PURIFICATION

For example, after the addition of the bleach or
chloros solution add from 30 to 60 grains (per 10
gallons) of aluminium sulphate according to the
quality of the rain-water, and then, to avoid any
risk of acidity, from 39 to 78 grains of sodium
carbonate crystals.

666 858 (factor about i -3)

A12 (S04)3, 1 8 H20 + 3 Na2 CO3, 10 H2O -*
A12 O3 + 3 Na2 SO4 + 3 CO2 + 28 H2O.

If this method is adopted, it is best to do without
the sodium sulphite, as its addition at the end of
the twenty-four hours necessarily means disturbing
the precipitate produced by the aluminium sulphate.

Of course, if it is added as a solution, it might be
possible to stir so gently and superficially as to
secure fair mixing without appreciable disturbance
of the precipitate.

Alternatively and preferably, the clear liquid
might be " run " into another vessel and the sodium
sulphite added to it. Speaking generally, the non-
expert use of sodium sulphite is almost contra-
indicated, unless in infinitesimal doses, or in those
cases where the excess of chlorine is so marked as
to impart a chlorinous taste to the treated water.
In passing, the writer is prone to admit that it is
not always easy for the non-expert worker to steer
between the Scylla of an impure water, and the
Charybdis of an imperfectly treated supply. Never-
theless, "microbial" safety stands first, and even

RAIN-WATER

gross blunders with the chemicals here recom-
mended could hardly render a water actually
injurious to health.

Yet another alternative plan may be suggested.
The dose of chlorine may be doubled, the duration
of contact halved, and then wood charcoal (say 20 to
200 grains per gallon, according to the impurity of
the water) added together with aluminium sulphate
and a neutralizing quantity of sodium carbonate.
The charcoal interferes, but not seriously, with
precipitation and it has the effect of removing the
chlorinous and the sooty taste.

COMBINED LIME AND CHLORINE METHODS.

It is obvious that these two processes may be
combined in a variety of ways.

For example, 5 gallons of rain-water may be
placed in each of two vessels (A) and (B).

(A) receives 33 grains (| drachm, 3 grains) of
hydrate of lime.

(B) receives f oz. of bleach solution (10*5 grains
of bleaching powder in 10 oz. of water). This
equals a dose of 075 in i million (instead of
0*5 per million) as the duration of contact is only
twelve hours.

After twelve hours (B) receives an anti-
chlorinating dose of sodium sulphite, say 0*4594

RURAL WATER SUPPLIES AND THEIR PURIFICATION

grain, allowing for about half the chlorine having
been used up.

The contents of (A) and (B) are then poured
into a third vessel (C) capable of holding 10 gallons,
and 45 grains (2 scruples, 5 grains) of sodium
bicarbonate added to neutralize the lime and
remove (as carbonate) the permanent hardness, and
the mixture left to settle for twelve hours.

Alternatively (C) may be treated as follows: —
(a) With 64 grains (i drachm, 4 grains) of

sodium phosphate and 15 grains (4 scruple, 5 grains)

of sodium bicarbonate, or,

(6) With 60 grains (i drachm) of aluminium
sulphate and 77 grains (i drachm, £ scruple, 7
grains) of sodium carbonate, or,

There is less chance of a chlorinous taste with
this combined method, although it loses a little in
point of simplicity.

As regards the use of liquid chlorine, special
apparatus has to be provided, and although the
writer has had, on the whole, a favourable expe-
rience of its use on a large scale, he is inclined to
think that for very small volumes of water some
of the other processes here described are for the
present to be recommended. It is quite possible,

RAIN-WATER

however, that some ingenious inventor will design
a form of apparatus which will sterilize even small
volumes of water simply, cheaply and effectively.

Little need be said about electrolytic compounds.
Where electric current is available, it may be used
either in connection with an ozonizing apparatus to
produce ozone (a most powerful bactericidal agent),
or with an elect rolyzer to form hypochlorites from
salt solution, or else with a quartz-mercury lamp to
yield ultra-violet rays. The forms of apparatus
on the market, however, for fulfilling these objects
are usually designed for dealing with larger volumes
of water than are being considered in this treatise.
Waters containing much suspended matter should
be filtered before being exposed to sterilization by
means of ultra-violet rays. The writer, however,
is rather trying to avoid purification processes which
require filtration.

The fact that there has been no wide-spread
adoption of any household device for sterilization
purposes seems to point to inventors having failed to
convince the public, or their advisers of the cheap-
ness and "fool-proof" efficiency of the apparatus
on the market.

As regards dose, enough has already been said
to indicate the probable amount of active substance
required for sterilization purposes.

Before closing this section, reference should

RURAL WATER SUPPLIES AND THEIR PURIFICATION

perhaps be made to a new chlorine compound,
p-sulphondichloraminobenzoic acid (called kalasone]
recommended by Dakin and Dunham (British
Medical Journal, May 26, 1917) for sterilizing
small quantities of water. Halazone can be made
up in tablets, and to sterilize a quart of water one
or two tablets, according to its quality, are said to be
required. Apparently sterilization takes place within
thirty minutes (see also Halazone for Water Sterili-
zation, British Medical Journal, August n, 1917).

In conclusion, the chief points to be noted are as
follows : —

Rain-water can be sterilized by means of heat
(3 pints boiling, i pint unboiled) and clarified with
coagulants (e.g., aluminium sulphate). (See also

P. 98.)

Various chlorine preparations (e.g., bleaching
powder) may be used successfully for sterilization
purposes (say, 0*105 grain per gallon), and any
excess of active chlorine removed by means of an
"anti-chlor. " (e.g., sodium sulphite). Aluminium
sulphate may be employed as well for clarification
purposes. (See also Chapters VI. and VII.)

The lime and chlorine methods may be combined
with some advantages.

Household devices for sterilizing waters by means
of ultra-violet rays, ozone, &c., have not achieved
wide popular success.

RAIN-WATER

A substance termed " Halazone " has recently
been introduced in tablet form for household
sterilization purposes.

The diligent and intelligent reader will, no
doubt, by this time have come to the conclusion
that, although the purification of rain-water may
be a complex process, putting on one side com-
plicating questions (e.g., turbidity, taste, &c.), safety
may be secured always by adding 3 parts of
boiling to i part of unboiled water, and usually by
adding either 67 grains of £&£**/ lime, or i "05 grains
of bleaching powder to 10 gallons of rain-water.

In the remaining chapters, other sources of water
supply will be briefly dealt with, but, to save
repetition, it will be assumed that the sterilization
and purification processes (see also Actual Experi-
ments, Chapters VI. and VII.) described under rain-
water have been carefully studied. The necessity
for sterilization has, so far, been assumed, but, of
course, there are waters which are pure, bacterio-
logically, but unsatisfactory as regards colour,
suspended matter, &c. In these cases, the addition
of from 3 to 6 grains per gallon of aluminium
sulphate is usually found to be most satisfactory,
but in the case of soft water an alkali should be
added as well — say 3*9 to 7*8 grains of sodium
carbonate per gallon. The precipitate must be given
time to settle and the clear liquid decanted for
domestic use.

RURAL WATER SUPPLIES AND THEIR PURIFICATION

CHAPTER IV.
WELL-WATER AND SPRING- WATER.

Well-water — Shallow wells usually impure and often hard —
Excess lime method — Determination of hardness — Trial
experiments — Estimation of alkalinity — Practical difficul-
ties— Neutralization of the excess lime — Citric and
tartaric acid — Duration of contact — Heat, chlorine,
electrolytic compounds, ozone and ultra-violet rays —
Combined lime and chlorine method — Relative volumes
— Trial experiments — Springs — Often very pure —
Caution, however, necessary — Methods of treatment
(when required) broadly comparable with those already
described — Usually tasteless and palatable — Concluding
remarks (p. 56).

WELL-WATER.

Shallow1 wells are so often contaminated in the
neighbourhood of dwellings, that their probable
impurity should be taken almost for granted.

Such waters, despite their unsafe character, are
frequently bright, clear and sparkling, and pleasant
to taste.

Their topographical relation to drains and cess-

1 Deep wells, especially if sunk through impervious strata,
are usually perfectly safe sources of water supply. Owing to
the excessive and speculative cost of sinking they are seldom
found in connection with private individual supplies. In any
case the purity of the water is usually beyond question.

WELL-WATER AND SPRING-WATER

pools is most important. Hardly less vital is the
kind of soil in which they are sunk (a good filtering-
medium or otherwise), the nature of their lining
(pervious or impervious), and the possibility or
probability of surface impurities reaching the supply.
For example, a leaking or overflowing cesspool,
situated at a higher level than, and not far distant
from, a well, the intervening soil being highly porous
or fissured, and the well itself shallow and unpro-
tected either laterally from the surface downwards
or vertically from below upwards from the rapid
ingress ofunpurified water, is an extreme illustration
of obvious dangers not unfortunately always avoided.
Medical officers of health in particular, but medical
men generally, are specially trained in these matters,
and if any doubt exists, their advice should be
sought and taken, especially as it usually errs on the
side of " safety."

The principles of purification remain much the
same whatever class of water is dealt with, but the
details require modification.

As a rule, well-waters are hard and not infre-
quently much of the hardness is what is known as
*' temporary," that is, it can be removed by boiling.

It is assumed in what follows that the preceding
chapters have been carefully studied.

EXCESS LIME.

Here, consideration must be given not only to
dissolved carbonic acid gas (CO2) but to bicarbon-
4 49

RURAL WATER SUPPLIES AND THEIR PURIFICATION

ates, which may be present in considerable amount ;
otherwise all the lime added may be neutralized and
rendered non-bactericidal.

The first step is the determination of the total
hardness of the water.

Pour 70 c.c. of the water into a small bottle and
from a burette " run " in a few drops of standard
soap solution (i c.c. = i grain per gallon CaCO3) at
intervals, shaking vigorously after each addition
until a lather which remains unbroken for a few
minutes is obtained. The number of c.c. used gives
the total hardness of the sample. The permanent
hardness is estimated in the way explained on
p. 1 8. The difference between the two is the
"temporary" hardness and is commonly due to
calcium bicarbonate. The action of lime (CaO)
on calcium bicarbonate is as follows : —

100 56 (factor '56)

CaCO3, H2CO3 + CaO « 2 CaCO3 + H2O.

Suppose the temporary hardness was found to be
12, then 12 x 0*56 = 672 grains per gallon of CaO
would be required. Add, say, 1*0 for dissolved
carbonic acid and, say, 3*0 for bactericidal purposes
(=1072).

It will be remembered that this must be multiplied
by 1*67 if hydrate of lime is used (1072 x 1*67

= I7'9).

In order to obtain a more reliable figure, it is
desirable to add this amount to a gallon (or, say,

WELL-WATER AND SPRING-WATER

one-tenth of it to one -tenth of a gallon) and then
estimate the excess CaO with phenol phthalein and
methyl orange in the way described on pp. 14 to 16.

Suppose the phenol phthalein reading is 5 and
the methyl orange reading 7, then 5 x 2 = 10
— 7 = 3 grains per gallon of excess CaO, or the
correct dose.

If found to be less or greater than 3, an obvious
calculation will determine the correct amount to add.

Another way of estimating the dose of CaO
required for sterilization purposes is to estimate
the alkalinity. Seventy c.c. of the water are
placed in a small flask, methyl orange added
and then standard acid (i c.c. =cvooi gramme CaO)
run in slowly from a burette until the yellow tint is
beginning to change to pink (see p. 116). The
number of cubic centimetres of standard acid solu-
tion used gives the number of grains per gallon of
CaO likely to be needed to combine with the bicar-
bonates in the sample of water. To this must be
added say i grain for dissolved CO2 and 3 grains
for bactericidal purposes.

Of course, the matter may be much more complex
than is here indicated. For example, lime acts on
magnesium salts as well as on bicarbonate of lime,
and magnesium salts are not uncommon constituents
of water.

Only a complete analysis by a highly skilled
analyst could be expected to yield the fullest infor-
mation, but determination of the alkalinity or the

RURAL WATER SUPPLIES AND THEIR PURIFICATION

temporary hardness, is an approximate guide to the
amount required. Then one or more trial experi-
ments with calculated amounts of lime, with subse-
quent estimation of the excess of lime actually left
in the water, leaves no doubt as to the actual
amount of lime required.

Unfortunately, most well-waters vary in com-
position from time to time, and this circumstance
can only be met by checking the results periodically,
or adding such an excess of lime as will cover all
variations, so far as sterilization is concerned.

With waters having a very high temporary
hardness, the excess lime should be considerable,
as a relatively slight increase in the hardness might
use up all the caustic alkalinity and destroy the
bactericidal action.

Those who do not feel wholly competent to carry
out these tests are advised to send a sample of the
water in question to a skilled analyst and ask him
to determine what amount of lime it is necessary
to add to it, so as to leave 3 grains of lime (as
CaO) per gallon in it in excess.

The fact is that with rain-water and very soft
moorland waters the excess lime method may be
used almost " blindly," because nearly all the lime
is available for bactericidal purposes. With hard
waters the case is widely different, most of the lime
being exhausted in precipitating carbonates from
bicarbonates.

For the sake of description, let it be assumed

WELL-WATER AND SPRING-WATER

that 1 8 grains (J scruple, 8 grains) of hydrate of
lime per gallon have to be added so as to leave
an excess of 3 grains of CaO.

Working as before with a 10 gallon unit, 180
grains (3 drachms) of hydrate of lime are added to
the water, the mixture well stirred and left to sterilize
for eight to twelve hours.

As regards subsequent treatment, it should be
remembered that well-waters do not usually require
clarification, and so the sodium phosphate and
aluminium sulphate methods are hardly needed.

Neutralization of the excess lime should be
effected by means of sodium bicarbonate, the
amount required being 30 x 3 = 90 (ij drachms).
This would produce 567 grains of sodium carbonate
or enough to remove five to six degrees of per-
manent hardness. This is not at all an unlikely
amount for a well-water, but even if it were less,
the presence of a small excess of sodium carbonate
would be of no importance.

Alternatively, "carbonic acid water" might be
used to neutralize the excess of CaO in the way
already explained under rain-water.

In either case, after addition of the sodium
bicarbonate or the "carbonic acid water," the water
should be left, preferably for eight to twelve hours,
so as to allow the precipitate produced to settle to
the bottom of the vessel.

As explained under rain-water, citric or tartaric
acid might be employed for neutralization purposes,

RURAL WATER SUPPLIES AND THEIR PURIFICATION

the amounts required being 37*2 (J drachm, 7
grains), and 79*8 grains (i drachm, i scruple)
respectively.

The sodium bicarbonate method, however, seems
most suitable as producing a beautifully soft water.

As described under rain-water, the bactericidal
dose of lime (CaO) is largely governed by the time
it is allowed to act. Hence, by using tanks capable
of holding one to seven days' supply (in duplicate)
the dose could be so reduced that no neutralization
of the excess of lime (CaO) would be really re-
quired, or, if needed, the amount of sodium bicar-
bonate necessary would be reduced proportionately.

On the other hand, as rain-water is usually very
soft and well-water often very hard, the utmost care
would have to be taken in the latter case to avoid
any risk of all the lime being used up by the bicar-
bonates, &c., in the water. This, of course, would
have the effect of robbing the lime of its bactericidal
power.

As regards heat, chlorine and electrolytic com-
pounds, ozone and ultra-violet rays, the remarks
made under rain-water apply broadly to other
sources of supply.

The combined lime and chlorine method is well
suited for well-waters, having a high temporary
hardness as perhaps as much as one-half or more
may be sterilized by means of lime, the subsequent
mixture of the limed and chlorinated waters pro-

WELL-WATER AND SPRING-WATER

ducing a sterile soft and tasteless water. Theo-
retically, whatever the number of grains per gallon
of temporary hardness was found to be, roughly
one-half (56 to 100), that amount of CaO would be
required for neutralization purposes. For example,
if the excess lime were equal to 3 grains per gallon,
6 grains of temporary hardness would be removed
by it. Hence, if the temporary hardness were 12,
one gallon of the limed water would be neutralized
by 0*5 gallon of the chlorinated water (i.e., say 66
to 33 per cent.).

Obviously, the temporary hardness in grains per
gallon, divided by twice the number of grains per
gallon, of excess CaO gives the divisor for ascer-
taining the proportion of i gallon which i gallon of
limed water will neutralize.

In practice, it is best to mix the water in the
amounts as calculated, and then determine the alka-
linity with phenol phthalein and methyl orange as
previously described (p. 14). Where pp = the phenol
phthalein reading and mo = = the methyl orange
reading, the following results may be obtained : —

TABLE IX. — ESTIMATION OF BICARBONATES, CARBONATES AND
HYDRATES (BASED ON A.P.H.A. TABLE).

(< = less than ; > = greater than).

(a) pp = 0 -«,

bicar bonates

Carbonate
0

tlydrates
0

(£) 2 pp < mo

mo - 2 pp

2PP

2Pp

(d) 2 pp > mo

2 (mo - pp)

2 pp — mo

(e) pp = mo .

RURAL WATER SUPPLIES AND THEIR PURIFICATION

A (d) or (e) result implies that more chlorinated
water is needed. A (c) result points to perfect
neutralization. An (a) or (b) result indicates that
the chlorinated water is in excess. Practically,
under the conditions of experiment, results (a) and
(e) could not possibly be obtained, as the former
would mean that no lime had been added and the
latter that carbonate of lime is absolutely (not
merely relatively) insoluble.

SPRINGS.

Spring-water is often very pure, as the source of
supply may be far distant and the passage of the
water through the soil tends to rob it of any im-
purities. This, however, is not always the case,
and if any doubt exists on so important a matter,
a sample, or samples, should be sent to a competent
expert, accompanied with any topographical or
other notes which may assist in the interpretation
of the results obtained.

If impure, its treatment may be carried out on
lines broadly parallel with those suggested under
well-waters. Putting on one side the so-called
" ochre " and medicinal springs and other excep-
tional cases, the water from springs is usually
bright, clear, sparkling, and most pleasant to drink.

In conclusion, the chief points to be noted are as
follows : —

WELL-WATER AND SPRING-WATER

The principles of purification remain the same
for all kinds of water, but in the treatment of well-
water it should be remembered that the question of
sterilization is frequently the most important factor
to decide upon, clarification and removal of taste
being seldom required.

Well-waters are often so hard that a sterilizing
process, which involves softening as well, has many
advantages (see Table XIX., p. 98). On the
other hand, many persons may consider chlorine
sterilization simpler (for dose, see p. 122).

Spring-water is often as pure as shallow well-
water is impure, but when polluted, the treatment
in the two cases may, as a rule, be carried out on
broadly parallel lines.

The topographical surroundings of wells and
springs are always of great importance.

The next chapter deals with the sterilization and
purification of river, brook and lake-water for
domestic use, and with questions of taste.

RURAL WATER SUPPLIES AND THEIR PURIFICATION

CHAPTER V.
RIVERS, BROOKS, AND LAKE-WATER.

(THE TASTE OF WATERS.)

Rivers, brooks, and lakes — Usually agreeable to taste — Very
variable as regards colour, physical appearances, chemical
composition, and bacteriological qualities — Pollutions
known and unknown (accidental) — Comparison between
waters impure but purified, and initially pure but liable
to chance contaminations — Soft waters and metallic
poisoning — Topographical survey essential — Suggestions
for treating different kinds of impure water — A few words
of advice — The taste of water — Concluding remarks
(P. 67).

RIVERS, BROOKS, AND LAKE-WATER (natural or
artificial).

These supplies present the widest variations
in quality, both chemically and bacteriologically.
They usually have an agreeable taste, in the
absence of excessive algal growths, although the
taste of very soft peaty waters, in the case of per-
sons used to a very hard water, is at first considered
mawkish. When the conditions are reversed, the
hard water, to begin with, seems to grip the mucous
membrane of the mouth and to have almost a
metallic taste. These sources of water supply may
be so pure as to be absolutely safe for drinking pur-
poses (excepting accidents) without any purification

RIVERS, BROOKS AND LAKE-WATER

whatsoever. On the other hand, they may be
grossly polluted with excremental matters and so be
extremely dangerous sources of supply. The com-
paratively recent (1905) typhoid epidemic at Lin-
coln should be a warning of the potential danger
attached to surface water supplies. Here, with a
population of about 50,000 there were over 1,000
cases of typhoid fever and over 100 deaths. Surface
waters vary greatly in chemical composition and
physical qualities, and some are as hard as others
are soft.

The mere fact that they are surface sources (in
greater or less measure) of supply suggests caution,
as, even if normally pure, they are always exposed
to chance contaminations.

The purest burn in the lonely Highlands of
Scotland may be fouled by a chance and reckless
wayfarer. Should such a vagrant be a typhoid
" carrier," the burn water will, for the time being,
be infinitely more dangerous than an initially impure
supply which has been subjected to adequate purifi-
cation processes. Indeed, it might be argued, with
some show of reason, that uniformly well purified
waters of doubtful or even dangerous antecedents,
may be safer than non-purified supplies of happier
origin, if the latter are ever exposed to accidental
pollutions of specific sort. As an example of the
former, an adequately stored and efficiently filtered
initially impure river-water might be taken, and, for
the latter, a pure Highland burn in the track of

RURAL WATER SUPPLIES AND THEIR PURIFICATION

shooters and beaters, some of which might be
irresponsible and conceivably at the same time
diseased.

When peaty water is piped to a house, it should
be remembered that these waters are often acid and
act energetically on lead, and that lead is a cumu-
lative and dangerous poison. In such cases, the
water should be rendered alkaline by means of lime
or soda before passing through lead pipes, or the
pipes used (owing to their composition, or their
internal coating) should preclude the possibility of
metallic poisoning.

It is most important that these surface sources of
supply should be carefully surveyed to see whether
there is any possibility or probability of excremental
pollution between their areas of origin and points
of abstraction for domestic use.

The mere fact, let us say, of your having found
it convenient to fix your residence near a stream
should suggest the probability of others having
been similarly influenced. If the habitations are
situated higher up on the water-shed the danger is
yours, if lower down, it is your bounden duty to see
that the danger is not theirs.

Impure soft, peaty and discoloured waters may
well be treated with lime, aluminium sulphate and
sodium carbonate, in the way described under rain-
water, and the treatment, if properly carried out,
sterilizes and clarifies the water and leaves it very
soft.

FIG. 9. — Suspended matter in 0-5 c.c. of a river water during a flood

(X 50 diam.)

RIVERS, BROOKS AND LAKE-WATER

For contaminated hard waters which are not
highly coloured, lime and bicarbonate of soda treat-
ment yields satisfactory results.

Instead of lime, chlorine may be used as the
sterilizing agent, and here it should be remembered
that peaty discoloured waters contain much oxidiz-
able matters, which may use up a considerable
proportion of the chlorine.

River-waters are usually very turbid in flood-time
and the use of lime, aluminium sulphate and
sodium carbonate is attended with excellent results.
Fig. 9 (for description see p. xv.) shows the sus-
pended matter in o'5 c.c. of a sample of river-water
during a flood.

The lime and chlorine methods may also be
combined in a variety of ways according to the
particular requirements of the case.

Careful consideration of the suggestions made
under rain-water and well-water will enable the
reader to decide which is the best course to adopt.

Important points to be considered are : —

(a) What is the source of the supply ; is it
uniformly contaminated, or are chance accidental
pollutions the only danger to health ? In this
connection, remember that although the excreta of
the lower animals is a most undesirable form of
contamination, the greatest danger lies in the dis-
charges of human beings, particularly those who

RURAL WATER SUPPLIES AND THEIR PURIFICATION

have the misfortune to be typhoid " carriers"
(3 per one thousand according to some authorities).

(b) What is the character of the supply? Hard
or soft, clear or discoloured, free from suspended
matters or otherwise, rich in organic matter or the
reverse, &c.

(c) If the verdict is unfavourable, questions of
economy, sentiment and trouble must be ruthlessly
set aside and the safety of the supply secured by
sterilization at all costs. Absence of taste, freedom
from colour and suspended matter, softness and
reduction in the amount of oxidizable matter are
less important than the destruction of all the germs
of disease.

(d) If in doubt, sterilize the water or secure an
alternative supply, or consult a competent expert.1
If the suggested methods of sterilization and puri-
fication appear too complex, seek the advice of some
person of experience in these matters. Never
adopt a laissez faire attitude in the matter of water
supply.

Before closing this chapter some reference may
be made to questions of taste, as this is a subject

1 Readers who are ambitiously inclined, and desire to test
for themselves the bacteriological qualities of water supplies
are referred to pp. 138 to 188 of the author's little book,
" Studies in Water Supply " (Messrs. Macmillan and Co., St.
Martin's Street, London).

RIVERS, BROOKS AND LAKE-WATER

which is too often neglected ; yet to many persons
it is so important that they prefer to drink a pleasant
flavoured water although it is known to be exposed
to pollution, rather than a bacteriologically " safe "
water which has an unpleasant taste or smell.

The Taste of Water. — This is a fascinating but
difficult study, and the number of persons with a
discriminating and reliable taste is comparatively
few.

Rain-water has a peculiar, mawkish, unpleasant,
sooty taste, and in the neighbourhood of towns
may be almost undrinkable. One can recall the
taste vividly to the remembrance of most persons,
however old, by reminding them of the days of
their childhood when, if they really were children,
they sucked icicles broken off from roofs, gutters,
waterspouts, &c. It is a faint but unforgettable
taste and one most difficult to eliminate. Rain-
water shaken vigorously with air in a partially
filled bottle for hours retains it almost, if not quite,
unimpaired. None of the chemical processes of
purification previously described appear at all
satisfactory. Permanganate of potassium is most
disappointing in this connection. If enough is
added to give a faint pink tinge persisting for
several minutes, the taste is seldom abolished and
may hardly be even modified. If more is added,
a permanganate taste is acquired which is, if any-
thing, worse. Even when a large excess is added

RURAL WATER SUPPLIES AND THEIR PURIFICATION

and then after twenty-four hours' contact the excess
removed by means of sodium sulphite, the results
are disappointing. Slow sand filtration may
improve matters very slightly, but slow filtration
through charcoal is much more effective. Al-
ternatively, wood charcoal powder may be added to
the water, the mixture well stirred and then
aluminium sulphate added as a coagulant. Both
the charcoal and alumina are precipitated. As
regards amounts, 20 grains per gallon of wood
charcoal powder may suffice, but in obstinate cases
it may be necessary to increase the dose greatly —
it may be even tenfold. The aluminium sulphate,
if lime has been used as the bactericidal agent,
may be added in amount sufficient to neutralize the
excess CaO (3*96, say 4, grains of aluminium
sulphate for each grain of CaO). If this does not
produce a good precipitate, more may be used up
to a total of say 6 grains per gallon, care being
taken to add sodium carbonate as well in the
proportion of 1*29 (say 1*3) grains of sodium car-
bonate crystals for each grain of aluminium sulphate
used in excess of that portion required for neu-
tralization of the excess lime (CaO). If the settled
liquid is unsightly, due to imperfect settlement of
the charcoal, it should be filtered. In the writer's
experience, the worst cases of taste can be cured
by the foregoing method. If a slight super-dose
of chlorine (say i in i million) is added to rain-
water, the liquid usually acquires a chlorinous taste

RIVERS, BROOKS AND LAKE-WATER

which may mask the sooty flavour, but is perhaps
equally if not more unpleasant. If enough sodium
sulphite is added to remove the excess of chlorine,
the chlorinous taste may disappear, but the sooty
flavour is apt to become again noticeable. The
1 ' after-taste " of stale chlorinated waters does not
seem to lend itself readily to remedial treatment.

If a very strong super-dose of chlorine is tried
and then the excess chlorine removed by means of
sulphites, the chlorinous taste vanishes but the sooty
taste still persists.

Wells and Springs. — These waters are usually
highly palatable, although in the former case the
water may be derived from most doubtful sources
of supply. Slow passage through the soil, oxi-
dation processes, and solution of various soluble
salts and gases give to well waters a most
acceptable "bite" and flavour.

Brooks and Burns. — These are usually well
flavoured, although when very soft and peaty, the
taste is an acquired one.

Rivers. — Much the same may be said of rivers,
although in stagnant " reaches " a great develop-
ment of plant and animal life may occur. These,
by their decomposition and the setting free of oily
matters and offensive gases, may give rise to a
variety of most unpleasant tastes and odours.

5 65

RURAL WATER SUPPLIES AND THEIR PURIFICATION

Lakes, Lochs and Reservoirs are generally
pleasant to drink, but are liable, seasonably or
occasionally, to active algal growths which may
render the water temporarily almost undrinkable.
It is curious that potassium permanganate, so
disappointing in the case of rain-water, is almost a
specific1 for this trouble in doses of 0*5 part per
million parts. The writer has not found chlorine
preparations satisfactory in this connection. Some
aquatic plants (e.g., chara) give rise to most offen-
sive tastes and odours as do some animals (e.g.,
the sponges). Figs. 10 and u (for description, see
p. xv.) illustrate some of the growths associated with
taste troubles.

Of course, the writer is here dealing with
ordinary, not exceptional supplies.

Deep wells may sometimes have a most un-
pleasant taste due to the presence of iron and
sulphuretted hydrogen. Aeration and filtration
usually remove these troubles. Some springs
contain so much iron as to be called "ochre
springs." Again, some surface and deep supplies
may contain so high a proportion of chlorides,
sulphates, &c., as to be practically undrinkable.
Lastly, there are medicinal springs which, owing to
their taste and purgative action, could only be
tolerated for curative purposes. These, however,

1 See p. 100, " Studies in Water Supply" (Messrs. Macmillan
and Co.).

(A) Anabaena, x 400.

(B) Eudorina, x 400.

(A) Stephanodiscus x 400.

(B) Glenodinium, x 150.

FIG. n.

RIVERS, BROOKS AND LAKE-WATER

are exceptional waters, which lie outside the
purview of this treatise.

In conclusion, the chief points to be noted are
as follows : —

All surface sources of water supply, even if
normally free from the taint of sewage pollution,
are open to accidental contamination.

In selecting a method of treatment, consider in
the first place safety from disease and then questions
of clarification, softening, taste, &c. (see Tables XII.,
XV., and XVI. and Miscellaneous Experiments,

P. 98).

Never forget the importance of a topographical
survey of the source of any water supply.

It is sometimes easy, at other times very difficult,
to remove the taste from waters having an un-
pleasant flavour. Aeration, filtration and the use of
permanganate of potassium and charcoal (see p. 64)
are all of value in particular cases. A chlorinous
taste, due to active chlorine, may be removed by
means of sulphites, &c., but the after- taste of a
stale chlorinated water is less amenable to treat-
ment. If strong doses of either permanganate or
chlorine are added to rain-water, the excess being
subsequently neutralized, the permanganate or
chlorinous taste may disappear, but the sooty
flavour is apt to become prominent again.

RURAL WATER SUPPLIES AND THEIR PURIFICATION

Consideration will now be given to the results
of actual experiments on the sterilization and
purification of rain-water, &c. The reader will
thus be able to judge how far practice agrees with
theory. It is hoped that the examples given will
stimulate the novice to become, if not a water
expert, at all events a convert to the belief that
nearly all waters may be purified to any standard
of safety required.

The writer obviously cannot hope to render
matters which are intrinsically difficult to the
trained expert transparently simple to the average
reader. His hope lies in stimulating interest, and
in affording information to those who, with or
without further advice, are determined to advance
their knowledge of rural water supplies. (See con-
cluding remarks on p. 128.)

THE RESULTS OF ACTUAL EXPERIMENTS

CHAPTER VI.
THE RESULTS OF ACTUAL EXPERIMENTS.

Series I. Sterilization of rain-water with bleach solution. —
Series II. Sterilization of rain-water by means of lime
— Series III. Purification, clarification and softening of
impure river- water by means of lime, aluminium
sulphate and sodium carbonate. — Series IV. (I., II.,
and III.). Experiments in Series I., II., and III.
repeated on a confirmatory B. coli basis. — Series V.
Lime, time and sterilization. — Concluding remarks
(p. 88).

There is nothing quite so convincing as the
results of actual experiments, and the following
notes are given as examples of the sterilization
and purification of various kinds of water by one
or other of the methods previously described.

SERIES I.1

The Sterilization of Town Rain-water by means
of Bleaching Powder Solution.

The dose, in terms of available chlorine, was
(a) i in i million, (b) i in 2 millions, and (c) i in

1 In Series I., II., III., a war brand of bile-salt medium was
used which proved unsatisfactory from the point of view of
presumptive B. coli results. In all the other experiments
confirmatory tests were applied.

RURAL WATER SUPPLIES AND THEIR PURIFICATION

4 millions. This is equal in grains per gallon to
(a) 0*07, (&) 0*035, and 00 0*0175. With bleaching
powder of 33 per cent, strength the amount of
actual substance would be (a) o'2i, (b) 0*105, and
(c) 0*0525. The rain-water was first purposely
inoculated with a little fresh human faeces and then
examined bacteriologically for B. coli ; 500 c.c.
(17*65 ozs.) of the sample were next poured
into each of three stoppered bottles and chlorine
added in the doses already stated. It may be
desirable to explain to the non-expert reader the
reason why the B. coli test is applied in these
cases. Excremental matters contains B. coli in
enormous numbers, and although the ordinary
faecal type of B. coli is believed to be relatively
or absolutely harmless, its presence in any number
in a water is presumptive evidence of potential
danger to health, because, if the pollution is of
human origin, B. coli is liable at any time to be
accompanied by bacteria which are unquestionably
pathogenic (e.g., the typhoid bacillus). Add to
this the fact that B. coli is a more hardy germ
than the typhoid bacillus and it becomes at once
apparent that a sterilization process which rids a
water of B. coli is unquestionably safe as regards
the microbes of epidemic water-borne disease
(e.g., typhoid fever).

Experiment i. — After inoculation, the rain-water
contained B. coli in o'oi, but not in o'ooi c.c.

THE RESULTS OF ACTUAL EXPERIMENTS

After eight and twenty-four hours, B. coli tests
were again made with the following results : —

After eight hours.

(a) + 100 - IOG.C. (£) + 100 - IOC.G. (c) + 100
- 10 c.c.1

After twenty-four hours.
(a) - 100 c.c. (b) - 100 c.c. (c) + TOO- loc.c.

It is apparent that after eight hours the improve-
ment, bacteriologically, was 10,000 times even
with the smallest dose (i in 4 millions).

In twenty-four hours, the improvement was at
least 100,000 times in (a) and (b) and 10,000 times
in (c).

Experiment 2. — After inoculation, the rain-water
contained B. coli in 0*01, but not 0*001 c.c. The
results, after eight and twenty-four hours, were
exactly the same as in Experiment i except that in
twenty-four hours (b) as well as (c) contained B. coli
in 100 c.c.

Experiment 3. — After inoculation, the rain-water
contained B. coli in 0*000 1, but not 0*0000 1 c.c.

1 Note that + = positive result (or present in), and —
= negative result (or absent from), the amount of water
stated.

RURAL WATER SUPPLIES AND THEIR PURIFICATION

After eight hours.

(a) + 10 c.c. (6) + 10 c.c. (c) + i c.c,
— i c.c. — i c.c. — 0*1 c.c.

(Improved 100,000 (Improved 100,000 (Improved 10,000

times) times) times)

After twenty-four hours.

(a) - 100 c.c. (b) + 100 c.c. (c) + 10 c.c.
— 10 c.c. — i c.c.

(Improved at least (Improved I million (Improved 100,000

10 million times) times) times

Experiment 4. — After inoculation, the rain-water
contained B. coli in o'oi, but not o'ooi c.c. The
results, after eight and twenty-four hours, were
exactly the same as in Experiment 2.

Experiment 5. — After inoculation, the rain-water
contained B. coli in o'oi, but not o'ooi c.c. The
results, after eight and twenty-four hours, were
exactly the same as in Experiment i except that in
twenty-four hours (c) as well as (a) and (b) contained
no B. coli in 100 c.c.

Experiment 6. — After inoculation, the rain-water
contained B. coli'm o* i, but not o'oi c.c. The results,
both after eight and twenty-four hours, were the
same in each case, namely, 10 c.c. — i c.c. The
improvement was thus 100 times.

THE RESULTS OF ACTUAL EXPERIMENTS

Experiment 7. — After inoculation, the rain-water
contained B. coli in o'oi, but not o'ooi c.c.

After both eight and twenty-four hours, the
results were : —

(a) + 100 — 10 c.c. ; (ft) + i — o'i c.c.;
(c) + 0*1 - 0*01 c.c. The improvement was (a)
10,000 times; (b) 100 times, and (c) 10 times.

Experiment 8. — After inoculation, the rain-water
contained 13. coli in o'oi, but not o'ooi c.c. The
results, both after eight and twenty-four hours, were
as follows : —

(a) - 100 c.c.; (b) + 10 - I c.c.; (c) -f I - o'l c.c.

The improvement was thus (a) at least 100,000
times ; (b) 1,000 times ; and (c) 100 times.

Experiment 9. — After inoculation, the rain-water
contained B. coli in o'ooi, but not o'oooi c.c. The
results were as under : —

After eight hours.

(a) + 100 - 10 c.c. ; (b) + 10 - i c.c.; (c] + 10
— i c.c. The improvement was thus (a) 100,000
times; (b) and (c) 10,000 times.

After twenty-four hours.

(a) - 100 c.c. ; (b) + 100 - 10 c.c.; (c) + 100
- 10 c.c. The improvement was thus (a) at least
i million times, (b) and (c) 100,000 times.

RURAL WATER SUPPLIES AND THEIR PURIFICATION

It is worth noting that the samples of rain-water
in Experiments 6 and 7 were, quite apart from the
artificial faecal contamination, extremely dirty.

A parallel series of experiments were carried
out, but without the addition of faeces, for judging
questions of taste.

All the samples had a sooty taste before the
addition of the bleach solution.

After twenty-four hours they either had a
chlorinous or a combined chlorinous and sooty,
or a sooty taste only. The addition of sodium
sulphite to those giving a chlorinous taste was of
use inasmuch as it removed the chlorinous taste,
but the sooty taste remained quite appreciable.

The chief bacteriological results are set forth in
the table on next page.

The results show that with a very dirty sample
of rain-water a dose of i in i million with twenty-
four hours' contact is hardly sufficient (on a pre-
sumptive basis), but with a very good sample
i in 4 millions may actually suffice. Practically, a
dose of i in 2 millions would probably be ample
in the great majority of cases.

It should also be noted that the positive B. coh
results were based on a presumptive basis.1 Ex-
perience has shown that presumptive results may
be sometimes misleading, because the positive

1 Presumptive = acid and gas formation in the lactose bile-
primary medium. Confirmatory = isolation of a " lactose +
indol + " microbe.

THE RESULTS OF ACTUAL EXPERIMENTS

TABLE X. — THE STERILIZATION OF RAIN-WATER WITH BLEACH
SOLUTION. DOSE : (a) i IN i MILLION, (b) i IN 2 MILLIONS,

(<r) I IN 4 MILLIONS, IN TERMS OF AVAILABLE CHLORINE.

Smallest

amount (in

Same as Ccl. 2,

Same as Col. 2,

Experi-
ment

c.c.) of water
yielding
a positive
B. colt result
after inocula-
tion, but

but after
eight hours of
chlorine treat-
ment.
(See, however,
remarks at

Improve-
ment after
eight hours
(times)

but after
twenty-four hours'
chlorine
treatment
(See, however,
remarks at end

Improve-
ment after
twenty-four
hours
(times)

before addition

end of Series I.)

of Series I.)

of chlorine

Cols, i

+ O'OI

+ IOO

10,000

None in 100

100,000

(at least)

> J

100,000

+ IOO

10,000

+ O'OI

+ IOO

10,000

None in 100

100,000

(at least)

2t>

+ IOO

10,000

»'

+ O'OOOI

+ 10

100,000

None in 100

10,000,000

(at least)

+ IOO

1,000,000

•J

+ I

10,000

+ 10

100,000

+ O'OI

+ IOO

10,000

None in 100

100,000

(at least)

+ IOO

10,000

-f o-oi

+ IOO

10,000

None in loo

100,000

(at least)

5t>

»>

100,000

+ O'l

+ 10

IOO

+ 10

IOO

'»

7<*

+ O'OI

+ IOO

10,000

+ IOO

10,000

+ 1

IOO

+ I

IOO

+ O'l

+ O'OI

None in 100

100,000

None in 100

100,000

(at least)

S/>

+ 10

1,000

+ 10

1,000

8<r

H- i •

IOO

-f- i

IOO

+ O'OOI

+ IOO

100,000

None in 100

1,000,000

(at least)

+ 10

10,000

+ IOO

100,000

+ 10

RURAL WATER SUPPLIES AND THEIR PURIFICATION

result obtained may be more apparent than real.
That is, that acid and gas may show in the primary
bile-salt medium1 and yet on subculture no growths
occur, or a growth of microbes other than B. coli.
It is possible or even probable that the results
would have been completely satisfactory2 if they
had been based on confirmatory instead of pre-
sumptive evidence. These remarks also apply
to Series II. and III.

SERIES II.

The Sterilization of Town Rain-water by
means of Lime.

The rain-water was first purposely inoculated
with a little fresh human faeces and then examined
bacteriologically. Then 2 grains of hydrate of lime
(CaO, H2O) were added to 48 ounces (= about
6*7 grains per gallon) of the faecally contaminated
water and cultures made after eight and twenty-
four hours on the shaken sample. Before shaking,
however, a little of the clear liquid was withdrawn
to estimate the amount of lime (CaO) left over in
excess.

1 This remark, however, does not apply, or only to a very
modified extent, to the bile-salt media of pre-war days.
Then a presumptive result (if at all decided) was nearly
always confirmed on subculture. Now with the current
brands of bile-salt, the results, in the writer's experience,
are much less to be trusted.

3 Reference to Series IV. (I., II., and III.) will show the
correctness of this assumption.

THE RESULTS OF ACTUAL EXPERIMENTS

The method of determining the excess lime has
already been described (see pp. 16, 55).

The chief results are set forth in the accompanying
table : —

TABLE XL— THE STERILIZATION OF RAIN-WATER WITH LIME. Two
GRAINS OF HYDRATE OF LIME (CaO, HaO IN 48 oz. (= ABOUT
67 GRAINS PER GALLON).

Experi-
ment

Smallest amount (in c.c.)
of water yielding a posi-
tive B. coli result after
inoculation, but before
addition of lime

Same as Col. 2, but after
eight hours of lime
treatment. (See, how-
ever notes at end of
Series I.)

Improvement after
eight hours.
(Times)

Same as Col. 2, but after
twenty-four hours of
lime treatment. (See,
however, notes at end
of Series I.)

Improvement after
twenty-four hours.
(Times)

Actual excess of caustic
lime (CaO) in the water
in grains per gallon

Cols, i

+ O'OOI

+ 10

10,000

+ 100

100,000

2-4

+ O'OI

+ 10

1,000

+ 10

1,000

3'5

+ 0*01

+ 100

10,000

+ 100

10,000

4'3

+ o-ooo i

+ 10

100,000

+ 10

IOO,OOO

3'4

-f O'OI

+ 10

1,000

+ 10

1,000

4'5

+ O'OI

-f 100

10,000

+ 100

10,000

4'5

+ O'l

+ 10

100

-f- 100

I,OOO

3*5

+ O'OI

+ 10

1,000

+ 100

10,000

3-2

Average

about 3 7

It is apparent from the table that a dose of 6 to
7 grains per gallon of hydrate of lime (CaO, H2O),
leaving in the water an excess of caustic lime (CaO)
of about 3 to 4 grains, can improve to a remarkable
extent bacteriologically, in from eight to twenty-
four hours, an artificially fsecally contaminated
rain-water. A somewhat larger dose than was
here employed is needed to sterilize the water
absolutely (as judged by presumptive results), but

RURAL WATER SUPPLIES AND THEIR PURIFICATION

it is obvious from the table that the results obtained
were so good as only to miss perfection by a
narrow margin.

The bacteriological cultures were made on
44 shaken" samples (i.e., so as to include the
sediment).

In practice, however, the deposit would not be
drunk, and human faecal contamination of rain-
water is most unlikely to take place, and never to
the extent here artificially produced. See, however
concluding remarks under Series I., which indicate
that much better results would probably have been
obtained if confirmatory tests had been applied, or
a pre-war brand of bile-salt used in preparing the
medium.

SERIES III.

The Purification, Clarification, and Softening of
an Impure River Water by means of Lime>
Aluminium Sulphate and Sodium Carbonate,

The water was first examined for B. coli and for
hardness and colour. Then 9 grains of calcium
hydrate (CaO, H2O) were added to 80 ounces
contained in a stoppered bottle, this being judged
to be about the amount necessary to neutralize the
dissolved CO2 and bicarbonates in the water and
yet leave enough caustic lime (CaO) in excess to
effect sterilization. After vigorous shaking the
mixture was left for twelve hours. Next, a little

THE RESULTS OF ACTUAL EXPERIMENTS

was withdrawn and used to estimate the excess
lime (CaO). The bottle was then shaken and
further B. coli cultures made so as to estimate the
improvement bacteriologically. Then 40 ounces
were poured into a bottle and first 3 grains of
aluminium sulphate and then 7*5 grains of sodium
carbonate were dissolved in the water, it being
judged that these amounts would serve to neutralize
the excess of lime and clarify and soften the water.
After settlement the clear liquid was examined for
colour and hardness.

The chief results are shown in the accompanying
table (Table XII.).

It will be seen that the chemical results were
very satisfactory, the colour and hardness being
reduced on the average 65 per cent, and 79 per
cent, respectively.

Bacteriologically, a great improvement was
effected, although evidently a slightly larger dose
of lime, or a longer contact, was required to produce
entirely satisfactory results (on a presumptive basis).
See, however, concluding remarks under Series I.,
which indicate that much better results would
probably have been obtained if confirmatory tests
had been applied, or a pre-war brand of bile-salt
used in preparing the medium.

SERIES IV.

In order to clear up the point as to whether the
results in Series I., II., and III. would not have

*oO c §"

g ^ Jr "ri • £ &.

BJifM

_:s ^ o-jj —

1 2

1 ^-

1 3

I?,

1 5

"1 c'^S'S

< g"* w

*G*

tn'c>

2
»^> M

mooi

s
O O *"

co o **

«oo *-

T3
COOO S

-a a

vo co^

*** '^ c

r^ -<a--^

O\ ^""Q

°* co^

ON fOg

3.1

M 4>

""* w

1-1 <u

ex,

V
U

'G"
.2

•g

^^1

c
o

"§

•"

63 w

S-

*«

J£,

£.

*0*

a &

l§

^|o

"trt"

l^>

_„__

^ <

(-5 M

C JJ *T3

??!

^
s

•J3

ATE SUBSEQUENl

'ED WATER ( =

allest amount of wat
elding a positive B. t
See, however, notes a
Series I.)

!!j

Pol

2
1

1—4

§2!

5 >

S*>»>~'

%™-*

rv?

C ^J

^, :

« :

^ :

j-> ;

G *^

U w

ill i

a» C
|| ^

I||

B I '«

4) C

-S J>

D >— 3

4-> S ta

t» "rt *j

S ttf

S ts

* "cS 4J

v "a

III 1

w w

,£* 4> C
II 1

M< g"
M

<S w "*•
1

is ^ 0

il S

v2 ** *c

o 5

THE RESULTS OF ACTUAL EXPERIMENTS

been still better if confirmatory B. coli tests had
been applied, a further set of experiments were
undertaken.

RAIN-WATER. — SERIES IV. (I.).

This corresponds with Series /., except that the
B. coli results were based on confirmatory
evidence.

TABLE XIII.— COMPARE WITH TABLE X.

DOSE (a) I IN I MILLION, (b) I IN 2 MILLIONS, (f) I IN 4 MILLIONS,
IN TERMS OF AVAILABLE CHLORINE.

Experi-
ment

Smallest amount (in c.c.)
of water yielding a
positive B. coli result,
after inoculation, but
before addition of
chlorine

Same as Col. 2, but after
eight hours of
chlorine treatment

Same as Col. 2, but
after twenty-four hours
of chlorine treatment

Cols, i

-f O'OI C.C.
»
»

+ 10 c.c.

+ I „

+ o-i ,,

-f- IOO C.C.

+ I „

+ o-i „

za
2b

-f O-OI C.C.
»
II

+ O*I C.C.
M

+ I C.C.

+ o-i „
+ o-i „

3<*
3*

+ O-OI C.C.

J>
»

+ 10 C.C.

+ I „

+ O'l „

+ IO C.C.

+ o-i „
+ o-i „

4<*
4b

4<r

+ O'OOI C.C.
»»
»

— IOO C.C.

+ IOO C.C.

— IOO C.C.

— loo „
+ loo „

5*
5*
5'

+ O'l C.C.

II
II

— IOO C.C.
II

— IOO C.C.
»»

The first three samples of rain-water, as it turned
out, happened to be very dirty, and, although an
improvement was effected even with the minimum
6 81

RURAL WATER SUPPLIES AND THEIR PURIFICATION

dose, the maximum dose failed to produce complete
sterilization.

With the last two samples completely satisfac-
tory results were obtained with the middle dose of
i in 2 millions in the first experiment and with the
minimum dose of i in 4 millions in the last experi-
ment. For all ordinary samples of rain-water a
dose of i in 2 millions would probably be quite
safe.

SERIES IV. (II.).

This corresponds with Series //., except that the
B. coli results were based on confirmatory
evidence.

TABLE XIV.— COMPARE WITH TABLE XI.
RAIN-WATER. Two GRAINS OF HYDRATE OF LIME (CaO, H2O) IN

48 OZ. (= ABOUT 6'7 GRAINS PER GALLON).

Experi-
ment

Smallest amount
(in c.c.) of water
yielding a positive
B. coli result,
after inoculation,
but before addition
of lime

Same as Col. 2,
but after
eight hours of
lime treatment

Same as Col. 2,
but after
twenty-four hours
of lime treatment

Actual excess of
caustic lime
(CaO) in the
water in grains
per gallon

Cols, i

+ O'OI C.C.

+ 100 C.C.

— IOO C.C.

3-0

-fo-ooi „

- loo ,,

2-3

4- o-i „

- 100 „

2'5

The results seem to show conclusively that
a dose of about 6*7 grains of calcium hydrate
(CaO, H2O) per gallon of rain-water would be
perfectly safe as regards sterilization.

THE RESULTS OF ACTUAL EXPERIMENTS

SERIES IV. (III.).

This corresponds 'with Series III., except that the
B. coli results were based on confirmatory
evidence, and questions of hardness and colour
were not again considered,

TABLE XV.— COMPARE WITH TABLE XII.
RIVER-WATER. HYDRATE OF LIME (CaO, H2O) 9 GRAINS PER

80 OZ. (= l8 GRAINS PER GALLON.)

Experi-
ment

Smallest amount of water
(in c.c.) yielding a positive
B. coli result, before
addition of lime

Same as Col. 2, but after
twelve hours' lime
treatment

Actual excess of caustic
lime (CaO) in the water
in grains per gallon

Cols, i

... -f- O'l C.C.

— 100 C.C.

I '4

... +0-1 „ ...

2'8

... + O'OI „

The results show clearly that a dose sufficient
to leave less than 3 grains of excess lime (CaO)
in the treated water is more than enough for
sterilization purposes.

Speaking generally, the assumption that the good
results obtained in Series I., II., and III. would have
been better still if confirmatory B. coli tests had
been applied is completely borne out by the results
set forth in Series IV. (I., II., and III.).

RURAL WATER SUPPLIES AND THEIR PURIFICATION

SERIES V.
Lime, Time and Sterilization.

It will be remembered that in Chapter II. it was
pointed out that the dose of lime for sterilization
purposes could be very greatly reduced by pro-
longing the contact for several days.

The following experiments serve to emphasize
this point : —

Experiment i. — Rain-water : Lime added in
such proportion as to leave an excess (CaO) of 2*1
grains per gallon. Inoculated with fresh human
faeces. Initial B. coll determination (+ o'ooi c.c.)
After twenty-four hours, no B. coli even in 100 c.c.

Experiment 2. — Same as Experiment i. Initial
B. coli determination (+ 0*0 1 c.c.). Excess lime
(CaO) 1*9 grains per gallon. After twenty-four
hours, no B. coli even in 100 c.c.

Experiment 3. — Same as Experiment i. Initial
B. coli determination (+ o'oi c.c.). Excess lime
(CaO) 2 '6 grains per gallon. After twenty-four
hours, no B. coli even in 100 c.c.

Experiment 4. — Same as Experiment i. Initial
B. coli determination (+ o'oooi c.c). Excess lime
(CaO) 0*7 grain per gallon. After twenty-four
hours, no B. coli even in 100 c.c.

Experiment 5. — Same as Experiment i. Initial
B. coli determination (+ o'ooi c.c.). Excess lime

THE RESULTS OF ACTUAL EXPERIMENTS

(CaO) 0*4 grain per gallon. After twenty-four hours,
no B. coli even in 100 c.c.

Experiment 6. — Same as Experiment i. Initial
B. coli determination (+ o'ooi c.c.). Excess lime
(CaO) 0*3 grain per gallon. After twenty-four
hours, no B. coli even in 100 c.c.

Experiment 7. — Same as Experiment i. Initial
B. coli determination (-f o'ooi c.c.). Excess lime
(CaO) o'i grain per gallon. After twenty-four
hours, B. coli present in 10 c.c. After forty-eight
hours, no B. coli even in 100 c.c.

Experiment 8. — Same as Experiment i. Initial
B. coli determination (+ 0*001 c.c.). Excess lime
(CaO) 0*2 grain per gallon. After twenty-four
hours, B. coli present in 100 c.c. After forty-eight
hours, no B. coli even in 100 c.c.

Experiment 9. — In this experiment an impure
river-water (no artificial contamination was made)
was substituted for rain-water. Initial B. ^/^deter-
mination (+ 0*0 1 c.c). Lime added in such proportion
as to leave an excess of 2*4 grains (CaO) per gallon.
After twenty-four hours, no B. coli even in 100 c.c.

Experiment 10. — Same as Experiment 9. Initial
B. coli determination (+ 0*1 c.c.). Excess lime
(CaO) 1*2 grains per gallon. After twenty-four
hours, no B. coli even in 100 c.c.

Experiment n. — Same as Experiment 9. Initial
B. coli determination (+ o'oi c.c.). Excess lime

RURAL WATER SUPPLIES AND THEIR PURIFICATION

(CaO) o'6 grain per gallon. After twenty-four
hours, B. £0/2* present in 100 c.c. After forty-eight
hours, no B. coli even in 100 c.c.

Experiment 12. — Conditions same as in Experi-
ment i (rain-water). Initial (i.e., after inoculation)
B. coli determination (+ 0*1 c.c.). Excess lime
(CaO) 0*1 grain per gallon. After twenty-four
hours, B. coli present in i c.c. After forty-eight
hours, B. coli present in 100 c.c. After seventy-two
hours, no B. coli even in 100 c.c.

Experiment 13. — Conditions same as in Experi-
ment i. Initial B. coli determination (+• 0*1 c.c.)
Excess lime (CaO) cri grain per gallon. After
twenty - four hours, B. coli present in TOO c.c.
After forty-eight hours, B. coli present in 100 c.c.
After seventy-two hours, no B. coli even in 100 c.c.

TABLE XVI. — LIME, TIME AND STERILIZATION.

Experi-
ment

Excess lime (CaO) in
grains per gallon

Initial B, coli result

When B. coli was killed

2'I

+ 0-001 C.C.

Within twenty-four hours

I'9

-|- O-OI

» >!

2'6

-j- O-OI

» »

-j- O'OOOI

»> >»

o*4

+ O'OOI

» J»

0-3

-j- O'OOI

>» »*

O'l

-j- O'OOI

Within forty-eight hours

0*2

+ 0-COI

>» 11

2 -4

+ o-oi ,

Within twenty-four hours

I '2

+ 0-1

>» »

0-6

+ o-oi ,

Within forty-eight hours

O'l

+ 0-1

Within seventy-two hours

o-i

+ o-i ,

»> >»

O'2

+ 0-1

5> J>

THE RESULTS OF ACTUAL EXPERIMENTS

Experiment 14. — Conditions same as in Experi-
ment i. Initial B. coli determination (4- o'l c.c.).
Excess lime (CaO) 0*2 grain per gallon. After
twenty-four hours, B. coli present in 10 c.c. After
forty-eight hours, B. £0/z present in 100 c.c. After
seventy-two hours, no B. coli even in 100 c.c.

The chief results are tabulated in Table XVI.

It is apparent that some waters at all events may
be sterilized with very minute traces of lime (in
excess) provided the contact is not less than from
one to three days. It does not follow, however,
that all waters will behave in a precisely similar
manner (especially as some absorption of CO2 from
the air would take place), and in practice it would
be unsafe to rely on these fractional doses, unless the
treatment was kept under scientific observation.
Nevertheless, the results do indicate that if rain-
water is so reasonably free from colour, taste and
suspended matter as not to call for any special
treatment, and if it can be stored for several days,
lime treatment presents remarkable advantages.
For example, the dose required for sterilization pur-
poses is so small as not to require any neutralization
subsequently, and, of course, a water treated in this
way could not possibly have any taste of lime and
would be innocuous in all respects. In the case
of really hard waters, however, this " small dose and
long contact" method of treatment is less attrac-
tive, for the reason that any slight increase in the

RURAL WATER SUPPLIES AND THEIR PURIFICATION

temporary hardness might more than rob the
treated water of any excess of lime (CaO), and
in so doing destroy its bactericidal action.

In conclusion, the chief points to be noted are
as follows : —

With reasonably pure samples of rain-water a
dose of i in 2 millions (in terms of available chlorine),
acting for eight to twenty-four hours, is amply
sufficient for sterilization purposes (Series I., also
Series IV. (I.)).

Rain-water may be effectively sterilized with a
dose of slaked lime (calcium hydrate) of, say, 67
(to provide a margin for safety) grains per gallon
acting from eight to twenty-four hours (Series II.,
also Series IV. (II.)).

Hard impure river-water may be readily sterilized
by means of lime if enough is added to leave over
an excess of, say, 3 (to provide a margin for
safety) grains of CaO per gallon. The excess lime
(CaO) may be neutralized, and the water softened
(reduction 79 per cent, in the experiments) and
rendered less brown (reduction 65 per cent, in
the experiments) by the addition subsequently
of aluminium sulphate and sodium carbonate
(Series III., also Series IV. (III.)).

Extremely small doses of lime (CaO) in excess
(considerably less than i grain per gallon) are

THE RESULTS OF ACTUAL EXPERIMENTS

capable of sterilizing rain-water and other waters,
if the duration of contact is prolonged over several
days (Series V.).

In the next chapter (so as to avoid over-
burdening the present one) the results of further
experiments on the purification of water will be
given.

RURAL WATER SUPPLIES AND THEIR PURIFICATION

CHAPTER VII.

THE RESULTS OF ACTUAL EXPERIMENTS—

(Continued).

Series VI. Lime and phosphate method of sterilization. —
Series VII. Sterilization and clarification of soft peaty
waters. — Series VIII. Lime and sodium bicarbonate
method of treatment. — Series IX. Miscellaneous experi-
ments.—Concluding remarks (p. 101).

SERIES VI.

Excess Lime and Sodium Hydrogen Phosphate
Method of Purification.

Experiment i. — Forty-eight ounces of rain-water
were inoculated with fresh human faeces. The
initial B. coli determination was + 0*01 c.c. The
colour was 80 and the hardness 4*4 grains per
gallon. Three grains (= 10 grains per gallon) of
calcium hydrate (CaO, H2O) were then added, the
mixture well shaken and the precipitate allowed to
settle. After twelve hours a little of the clear
liquid was withdrawn and the excess lime (CaO)
determined. It was equal to 7*2 grains per gallon.
After shaking, B. coli tests were made, but no
B. coli could be found even in 100 c.c. of the water.

THE RESULTS OF ACTUAL EXPERIMENTS

Next, 24 ounces of the mixture were treated with
2-5 grains (= about 17 grains per gallon) of sodium
hydrogen phosphate and then with i grain (= about
7 grains per gallon) of sodium bicarbonate. When
the copious precipitate produced had been allowed
to settle, the colour and hardness were found to
be 50 (reduction about 37 per cent.) and 5*2
respectively.

Experiment 2. — Same as Experiment i. Initial
B. coli determination (+ o'ooi c.c.), colour 150 and
hardness 3*9 (grains per gallon). Three grains of
calcium hydrate added to 48 ounces as before.
After twelve hours, excess lime (CaO), 6*8 grains per
gallon ; no B. coli even in 100 c.c. To 24 ounces
of mixture 4*5 (= 30 grains per gallon) and 2*2
(= about 15 grains per gallon) of sodium hydrogen
phosphate and sodium bicarbonate added, respec-
tively. After settlement, colour and hardness 60
(reduction 60 per cent.) and 4*8 respectively.

Experiment 3. — Same as Experiment i. Initial
B. coli determination (+ i c.c.). Colour 84 and
hardness 5*8 (grains per gallon). Here only 2 grains
of calcium hydrate were added to 48 ounces of
water (= about 6*7 grains per gallon). After twelve
hours, excess lime (CaO) 1 7 grains per gallon ; no
B. coli even in 100 c.c. To 24 ounces of mixture
added ri grain of sodium hydrogen phosphate and
then 0*5 of sodium bicarbonate. After settlement,

RURAL WATER SUPPLIES AND THEIR PURIFICATION

colour and hardness determined and found to be
45 and 5*2 (grains per gallon) respectively.

The chief results may be tabulated as follows : —

TABLE XVII. — RAIN-WATER. LIME AND SODIUM PHOSPHATE
TREATMENT.

Experi-
ment

Initial B. coli
result

Initial
colour

Initial
hard-
ness

Excess
lime
(CaO)

Final B, coli
result

Final colour

Final
hard-
ness

+ O'OI C.C.

4 '4

7-2

— 100 C.C.

5'2

(about 37 per

cent, reduc-

tion)

+ o-ooi „

ISO

3'9

6-8

4'8

(60 per cent.

reduction)

+ 1

5-8

5'2

(about 46 per

cent, reduc-

tion)

It will be seen that an excess of lime (CaO) of less
than 2 grains per gallon (Experiment 3) sufficed
for sterilization purposes. The colour was also
considerably reduced. The hardness was higher
in the first two experiments, where the dose of lime
was somewhat large, but in the last experiment
there was a slight reduction in the hardness as a
result of the treatment.

Some authorities object to the use of phosphates
as precipitants, as phosphates tend to encourage
bacterial growths. The writer thinks that the
importance of this matter has been exaggerated, as
the only microbes likely to be affected would be
harmless water bacteria.

THE RESULTS OF ACTUAL EXPERIMENTS

SERIES VII.

The Sterilization and Clarification of a Soft Peaty
Moorland Water.

As a matter of fact, the water in question con-
tained very few B. coli, so it was first inoculated
with a trace of fresh human faeces, so as to see what
would take place in the case of peaty waters ex-
posed to excremental pollution. The B. coli, colour,
and hardness results were then determined.

(A) Eighty ounces then received 3*35 grains
(= 67 grains per gallon) of calcium hydrate
(CaO, H2O). The mixture was well shaken and
then allowed to settle. After twelve hours, a little
of the clear liquid was withdrawn for estimation
of the excess lime (CaO). After shaking, B. coli
cultures were made. Then 40 ounces of the
mixture were treated with aluminium sulphate and
sodium carbonate, and the colour and hardness
results re-determined.

Five hundred cubic centimetres (about
17*65 ounces) of the water were placed in each ot
three bottles, and these received bleach solution in
doses (in term^ of available chlorine) of (a) i in
i million, (b) i in 2 millions, and (c) i in 4 millions.
After seventeen hours, B. coli cultures were made.

Experiment i. — Initial B. coli result ( + cri c.c.).
Colour 1 20. Hardness 3*4 (grains per gallon).

RURAL WATER SUPPLIES AND THEIR PURIFICATION

(A) Excess lime (CaO) 3*1. B. coli result, after
twelve hours (— 100 c.c.). Forty ounces received
3*1 grains of aluminium sulphate and 4 grains of
sodium carbonate. After settlement, colour 40, and
hardness 8*0 (grains per gallon).

(B) After seventeen hours of chlorination, the
results were as follows : (a) — 100 c.c. ; (b) — 100 c.c. ;
(c) + 10 c.c.

Experiment 2. — Same as Experiment i, except
that less lime (2*5 grains) was used. Initial B. coli
result (+ 0*001 c.c.). Colour 114 ; hardness 273.

(A) Excess lime (CaO) 2*0. B. coli results,
after twelve hours (— 100 c.c.). Forty ounces
received 1*96 grains of aluminium sulphate and
2 '55 grains of sodium carbonate. After settlement,
colour 60 and hardness 7*0.

(B) After seventeen hours of chlorination, the
results were as follows : (a) + 100, not 10 c.c. ;
(6) 4- 100 and 10, not i c.c. ; (c) + 100, 10 and i,
not 0*1 c.c.

Experiment 3. — Same as Experiment 2, 2*5 grains
of lime being used. Initial B. coli result (+ o'oi c.c.).
Colour no; hardness 2-5.

(A) Excess lime (CaO) 2*2. B. coli results after
twelve hours ( — 100 c.c.). Forty ounces received
2*18 grains of aluminium sulphate and 2*8 grains of

THE RESULTS OF ACTUAL EXPERIMENTS

sodium carbonate. After settlement, colour 90 and
hardness 8*26.

(B) After seventeen hours of chlorination, the
results were as follows: (a) — 100 c.c. ; (6)
+ 100 - 10 c.c. ; (c) +' i c.c. — cri c.c. The
chief results may be tabulated as follows : —

TABLE XVIII. — STERILIZATION AND CLARIFICATION OF MOORLAND WATERS,

Experi-
ment

Initial B. coli
result

Initial

colour

Initial
hardness

+ 0*1 C.C.
+ 0-001 „

1 2O
114

3'4

273

4- o-oi „

110

2*5

(A) Part of experiment

(B) Part of experiment. Final B. coli result

Excess
CaO

Final B. coli

result

Final
colour

Final
hardness

<«>.„.

i in i million

i in 2 millions

i in 4 millions

3'i

— IOO C.C.

8-0

— IOO C.C.

+ IO C.C.

2'0

7-0

+ ,,

+ 10 „

-f- I C.C.

2'2

8-26

~~ >»

+ loo „

As regards the A part of the experiment, it will
be seen that an excess of 2 grains per gallon of
CaO was quite effective bacteriologically. The
water, however, was considerably harder after than
before the treatment, and in the last experiment
the reduction of colour was disappointing, looking
as if the lime (CaO) had acted on the colouring
matter in the water in some way so as to render

RURAL WATER SUPPLIES AND THEIR PURIFICATION

it less " coagulable." (See also remarks on p. 25.)
In respect of the B part of the experiment, a dose
of chlorine of i in i million was sufficient in two
out of the three cases.

SERIES VIII.
Lime and Sodium Bicarbonate Treatment.

In illustration of this method of treaiment, a
hard well-water was chosen (total hardness 23 grains
and permanent hardness 6*4 grains per gallon).
The well-water in question is sunk in the upper
porous chalk and is known to be liable to pollution.
In order to increase the B. coli normally present
in the water it was inoculated with a little fresh
human faeces. Its alkalinity (reduced to CaO)
was found to be 10 grains per gallon : that is,
70 c.c. of the water required for its neutralization
(see p. 51), 10 c.c. of standard acid (i c.c. =
0*001 gramme CaO). Allowing 3 c.c. for CO2 sterili-
zation, &c., the lime (CaO, H2O) required was
theoretically 13 x 1*67 = 2171 grains per gallon.

Experiment i. — Forty ounces of well-water re-
ceived 5*42 grains of calcium hydrate (CaO, H2O).
The initial B. coli determination was + 0*1 c.c.
After twelve hours, a little of the clear liquid was
drawn off and the excess lime (CaO) was found
to be 0*9. After shaking, B. coli cultures were
made and no B. coli were found, even in 100 c.c.

THE RESULTS OF ACTUAL EXPERIMENTS

Next 0*34 grain of sodium bicarbonate were added
to 20 ounces of the mixture, and, after settlement,
the hardness was determined and found to be
6*3 grains per gallon.

Experiment 2. — Same as Experiment i, but
7*42 grains of calcium hydrate (CaO, H2O) added.
The initial B. coli determination was + croi c.c.
After twelve hours, a little of the clear liquid was
drawn off and the excess lime was found to be 4*0.
After shaking, B. coli cultures were made and no
B. coli were found, even in 100 c.c. 1*5 grains of
sodium bicarbonate were next added to 20 ounces
of the mixture, together with a trace of sodium
carbonate to ensure the softening effect. After
settlement, the hardness was again determined and
found to be 2*5 grains per gallon.

Experiment 3. — Same as Experiment i, but
6*42 grains of calcium hydrate (CaO, H2O) added.
The initial B. coli determination was + croi c.c.
After twelve hours, a little of the clear liquid was
drawn off and the excess lime was found to be
2 '8. After shaking, B. coli cultures were made and
no B. coli were found, even in 100 c.c. 1*05 grains
of sodium bicarbonate were next added to 20 ounces
of the mixture, and after settlement the hardness
was determined and found to be 27. The chief
results may be tabulated as follows : —

RURAL WATER SUPPLIES AND THEIR PURIFICATION

TABLE XIX.— HARD WELL-WATER. LIME AND SODIUM BICARBONATE
METHOD. TOTAL HARDNESS 23 (PERMANENT HARDNESS 6-4) GRAINS
PER GALLON.

Experi-
ment

Initial B. coli
result

Excess lime (CaO)
in grains per
gallon

Final B. eWz re-
sult, after twelve
hours' contact

Final total hardness, in
grains per gallon

+ O'l C.C.

0'9

— 100 C.C.

6'3

(nearly 73 per cent.

reduction)

+ O'OI C.C.

4'0

) J

2'5

(about 89 per cent.

reduction)

2-8

(about 88 per cent.

reduction)

It will be noted that even so small an excess as
0*9 of lime (CaO) proved effective, although, in
practice, it would hardly be safe to rely on so small
a dose in the absence of skilled supervision and the
application of frequent bacteriological tests. More-
over, in cases where there is a fair amount of
permanent hardness there is no good reason for
running things too fine, because more lime simply
means more bicarbonate of soda, greater security
and a beautifully soft water as the final product.

SERIES IX.
Miscellaneous Experiments.

Experiment i. — An impure river- water in flood
(initial colour 208, initial hardness 207, initial
B. coli result + 0*1 c.c.) was treated as follows:
Three parts were heated and brought barely to the
boiling point and then added to one part, which

THE RESULTS OF ACTUAL EXPERIMENTS

was not heated. After five minutes, cultures were
made and no B. coli were present, even in 100 c.c.
of water. Some of the remaining water was
treated with aluminium sulphate (in the proportion
of 6 grains per gallon) ; after settlement, the colour
was 44 (nearly 79 per cent, reduction) and the
hardness 21-9.

Experiment 2. — Same as Experiment i, except
that clarification by aluminium sulphate was not
tried and the sample was collected on a different
date. Initial B. ^//result + croi c.c. Final B. coli
result negative 100 c.c. These two experiments
show that by adding three parts of " boiled " to one
part of " unboiled" water, the mixture is rendered
perfectly safe to drink. In Experiment i, it is also
shown that aluminium sulphate has a powerful clari-
fying action.

Experiment 3. — A river-derived water having an
average hardness of 19*4 grains per gallon was
treated with calcium hydrate. The next day the
clear liquid was decanted and the caustic alkalinity
estimated. It was found to be 2*2. The hardness
was also determined and found to be 77. No
bacteriological tests were made, it being clear from
the results of the experiments already recorded
that this excess would suffice for sterilization pur-
poses. Two ounces of the liquid were then placed
in a small flask, phenol phthalein solution added,
and " carbonic acid water," freshly drawn from a

RURAL WATER SUPPLIES AND THEIR PURIFICATION

sparklet syphon, very gradually run in by means
of a graduated pipette, until the pink colour dis-
appeared. Half the number of cubic centimetres
of "carbonic acid water " required were next added
for each 2 ounces of the lime-treated water. After
the precipitate of carbonate of lime produced
(CO2 + CaO == CaCO3) had settled, the hardness
was re-determined and found to be 6*5.

Experiment 4. — This experiment was on all
fours with Experiment 3. The excess lime (CaO)
was 3*8, and the hardness 12*4. After neutraliza-
tion with "carbonic acid water" the hardness was
reduced to 6*0.

Experiment 5. — Same as Experiments 3 and 4.
Excess lime (CaO) 3*6, hardness 117. After
neutralization with the "carbonic acid water" the
hardness was reduced to 6*2.

Experiment 6. — Same as Experiments 3, 4 and 5.
Excess lime (CaO) 4*9, hardness 14*6. After
neutralization the hardness was reduced to 7*2.

Experiments 3 to 6 show that if lime is used as
a sterilizing agent, the excess CaO can be neutralized
by means of "carbonic acid water" derived from
a sparklet syphon or gasogene, the carbonate of
lime produced being precipitated and the water
"softened" in consequence. The volume of "car-
bonic acid water" required depends, of course, on
the concentration of the carbonic acid gas in it and

100

THE RESULTS OF ACTUA '

the amount of excess lime (CaO). In the above
experiments about 2*8 to 5*6 ounces were required
per gallon.

It should perhaps be stated that in practically all
these experiments comparatively small quantities
were dealt with, and a cheap dispensing pair of
scales were used for weighing purposes.

In conclusion, the chief points to be noted are
as follows : —

Rain-water can be sterilized with very small
doses of lime and the excess lime (CaO) neutralized
by means of sodium phosphate.

Soft peaty moorland waters may be sterilized
with lime (about 2 grains per gallon in excess) and
the excess neutralized with aluminium sulphate ; or
else sterilized with chlorine (about i in i million).

Hard well-water is conveniently sterilized and
softened by means of lime and sodium bicarbonate.

Sterilization can be effected by adding three
parts of boiling water to one part of unboiled
water.

"Carbonic acid water" can be used for neutraliz-
ing the excess lime (CaO) in a lime-sterilized water.

The concluding chapter deals with a description
of the apparatus required for the sterilization of
water on a domestic scale.

101

' KURAt WATER SUPPLIES AND THEIR PURIFICATION

CHAPTER VIII.
DESCRIPTION OF APPARATUS (see Figs. 12 to 18).

Ten gallon a day basis — Description of vessel recommended
— Floating arm method of drawing-off the liquid —
Rigid outlet method — Notes on filtration — Apparatus
described — Cleaning of filter — Disposal of precipi-
tate in large vessel — Storage of purified water — " Long
time lime" method of treatment— Tanks required —
Questions of chlorine— Concluding remarks (p. 113).

The writer has assumed throughout a daily con-
sumption for drinking purposes (per household) of
ten gallons; where more is needed the tanks or
vessels required may be correspondingly increased
in size, and where less, provision has been made for
more than one day's supply.

The tanks or vessels must be in duplicate so as
to allow of their alternate use for purposes of
treatment.

The 10 gallon glazed earthenware vessels known
technically as " mixing pans" (see Fig. 12) serve
admirably for this purpose.

The tubulure is fitted with a perforated rubber
bung, but the attachments depend very much on
circumstances.

(a) Floating Arm Method (see Fig. 13). — Where it

102

DESCRIPTION OF APPARATUS

is desired to draw off a very clear liquid, a glass tube
is inserted in the rubber bung. At the distal (outlet)
end is a piece of rubber tubing, a strong pinch-cock
or screw clip (see, Fig. 14) and another piece of

FIG. 12.

O-^?

FIG. 13. — Floating arm method.

FIG. 14.

glass tubing, which, if thought necessary, may be
covered at its free end with one or more folds of
fine linen, bound on with a rubber band. Alter-
natively, instead of a piece of plain glass tubing, a

103

RURAL WATER SUPPLIES AND THEIR PURIFICATION

" thistle head " tube may be used and the interior
packed with absorbent cotton wool or asbestos
fibre, kept in place with a piece of linen stretched
over the mouth and held in position with a rubber
band (see Fig. 15).

-Rubber bung
with distal*
attachments

X"""^-i

- / \

"*>vH

Screw clip here*
(not drawn)

G/ass tube

Linen and
rubber band

(reduced)

FIG. 15.— Rubber bung with its distal connections.

At the proximal end (inside) is a piece of rubber
tubing nearly long enough to extend right across
the vessel (see Fig. 13). Its free end is fixed firmly
by means of a very long pin (say 2 inches long) to a
large cork bung (say 2 to 3 inches in diameter) and
it is desirable to stick into the bung two other pins.
so as to provide three more or less equidistant legs
(see Fig. 16). This allows the cork floater to rest

104

DESCRIPTION OF APPARATUS

securely on the bottom, when the liquid has been
drawn down to its lowest level, without disturbing
the sediment. Next, the rubber tube is pinched and
two or three holes cut out with sharp scissors at
different angles close to the cork bung. Obviously,

Rubber tube
pinned to
cork

Hole in tube

_Tube leading'
" to rubber
bung

FK;. 16.— Cork floater.

when the vessel is filled with water the cork floats,
and when the exit stopcock is opened the liquid
escapes from near the surface through these holes,
As the liquid falls in level so also does the cork,
until finally the pins strike and rest on the bottom
of the vessel and all the liquid excepting about the

105

RURAL WATER SUPPLIES AND THEIR PURIFICATION

last gallon (more or less, according to the length of
the pins and the position of the holes) is withdrawn
without disturbing the sediment.

(b) Rigid Outlet Method. — For all practical pur-
poses it is sufficient to have a fixed "draw-off"
slightly above the level of the precipitate (say
J to i inch). The proximal end of the tube should
not project inside the vessel as the precipitate
settles on it, and at the very end of the " drawing-
off " process, it is liable to be detached and sucked
down the tube (see Fig. 17). The distal end of the

V/ece oftinen
FIG. 17.— Rigid outlet method.

tube may, as in the " floating arm " method, be
covered with linen or a thistle-head tube may be
used, the interior being packed, if thought desirable,
with absorbent cotton wool or asbestos fibre. Where
the precipitate, per se, is absolutely harmless (e.g.,
a precipitate of carbonate of lime) these straining
processes seem unnecessary, unless possibly on
aesthetic grounds, and even in the case of alumina
their use can hardly be considered at all essential, as
practically all the precipitate is left at the bottom of

106

DESCRIPTION OF APPARATUS

the vessel. In all purification processes it is a good
thing to ask oneself the question : Is the mixture in
the freshly prepared and stirred up condition
injurious from the chemical point of view? If not,
it is out of the question to regard the final product
as a source of danger to health.

(c) Filtration Methods. — This book is concerned
chiefly, if not entirely, with non-filtration processes
of water purification. This does not necessarily
mean that the author has no faith in filters, but the
subject of filtration has already been dealt with by
numerous writers (e.g., Dr. Sims Woodhead). If,
however, it is thought desirable to filter finally
the sterilized liquid, the consumer must decide
whether it is better to use one of the many patent
filters on the market or to construct one himself.

The writer does not propose to enter into a dis-
cussion of the respective merits of the different types
of patent domestic filters, but only to indicate very
briefly how a small filter may conveniently be
constructed.

The mixing pans already described may be had in
small sizes (one gallon). The tubulure is fitted
with a rubber bung, and a bit of glass tubing to
which is attached a piece of rubber tubing controlled
by means of a screw clip (see Fig. 14). The bottom
is filled with gravel varying in size from a bean to
a very small pea, the coarse material being under-
most. The gravel should extend well above the

107

RURAL WATER SUPPLIES AND THEIR PURIFICATION

level of the outlet tube, as otherwise the sand will
be sucked downwards and appear in the filtrate.
Next is placed a layer (say \ inch) of coarse sand and
then fine sand on the top of this (say 2 to 3 inches).
The whole is then well covered with a layer of pea-
sized gravel to prevent the onflowing water from
disturbing the surface of the sand. It is essential
that the gravel and sand should be scrupulously
clean, and if any doubt exists as regards its purity,

Screw clip

FIG. 18.— Small filter.

the filter should be filled with boiling water and the
outlet tube opened a little and more hot water poured
on to take its place, until the temperature of the
outgoing water is 70° C. and still rising. When
it is certain that the whole of the contents of the
filter has been exposed to this temperature, the
outlet tube may be closed and in five minutes the
filter may be safely assumed to be free from any of
the microbes causing epidemic disease, and the

108

DESCRIPTION OF APPARATUS

water finally run off. It is a good plan to place a
small piece of flat slate on the surface of the filter as
the onflowing liquid impinging on it is spread over
a larger area. It is desirable as well to alter from
time to time the position of the slate and the delivery
tube. As regards the latter, a piece of copper wire
may be tied firmly on to the glass tube and the
rubber tube fastened loosely to it as well. Then by
bending the wire slightly the direction of the flow of
water may be altered at will. When filtering, the
outlet tube should be completely open and the rate
of filtration controlled by the screw clip on the
delivery tube. The speed of filtration may be made
as slow as circumstances permit, but must never be
so fast as to cause "ponding" on the surface of the
filter with a consequent risk of an overflow. The
first few ounces of filtrate should be rejected or
poured back on to the surface of the filter. Once a
week (or more frequently, if thought desirable), the
bits of gravel should be removed from the surface of
the gravel and also about half an inch of the top
layers of sand. This is readily done by means of a
little scoop — a bit of bent tin does well for this pur-
pose. The sand and gravel should be washed
separately, first in cold, and then in very hot water
and replaced and any loss made good. About once
a month it may be desirable to remove the whole of
the material from the filter and replace it with fresh
gravel and sand, or the same material after it has
been thoroughly cleaned. It will be noted that this

109

RURAL WATER SUPPLIES AND THEIR PURIFICATION

is the " percolation " as opposed to the " submerged "
method of filtration and it has the advantage of
encouraging aeration. At the same time, such a
filter may actually not achieve as good results as a
dirty " mature" filter, the reason being that in the
former case practically all the non-sporing bacteria
have been banished, whereas in the latter case
microbes are present in enormous numbers and some
species exercise a purifying action on the water. In
the present case, however, the water is supposed to
be sterile to start with, and the filter is only required
in order to remove the suspended matters from the
water. Of course, any ingenious person can make
a filter out of almost anything, e.g., a flower pot can
be readily adapted for this purpose.

A few words must be said about the disposal of
the precipitate left in the 10 gallon vessel. One
way is to run the whole of it to waste. If this
method is decided upon, a couple of wedge-shaped
pieces of wood should be inserted laterally towards
the side of the vessel farthest from the outlet.
Next, a pail should be placed underneath the out-
let, the precipitate well stirred up and the bung
removed, so as to allow the contents to escape from
the tilted vessel with a rush. Some more water
may be used to flush out the residue, but this is
hardly necessary in actual practice.

Another way is to leave the precipitate and
supernatant water in situ, and make an allowance
for its presence by adding fresh chemicals in reduced

110

DESCRIPTION OF APPARATUS

amount to the next lot of water. This entails
knowledge of its volume ; and this information is
readily obtained either by measuring the amount
drawn off and deducting it from the total, or ascer-
taining beforehand what volume of water the vessel
holds up to the ''draw-off'' level. Within limits,
the old precipitate rather helps in the production of
a further deposit, but, of course, the time comes
(say in a week) when the whole of the sludge must
be swept out to waste. Generally speaking, the
first method should be adopted in cases where an
accumulation of the deposit is for some reason
undesirable. On the other hand, where the pre-
cipitate is per se innocuous (e.g., carbonate of lime)
the final flushing operation may be delayed for
some considerable time.

The writer has not dealt so far with the vessel or
vessels in which the purified liquid is to be collected,
This obviously depends upon circumstances ; for
example, if the whole of the water is required daily,
it may be collected in one large vessel and the
whole of the plant duplicated for use each alternate
day. On the other hand, if only a small proportion
is needed daily it may be convenient to fill a series
of Winchester quart bottles (carefully dated) and
delay the preparation of a fresh lot of purified water
until their contents have been nearly exhausted.
No one of intelligence can fail to find out by experi-
ence the method best adapted for his (or her)
particular requirements. Needless to say the

111

RURAL WATER SUPPLIES AND THEIR PURIFICATION

vessels should be scrupulously clean, and if more
than one day elapses between collection and con-
sumption they should be glass stoppered.

When what may be described as the " long time
lime" method of treatment (see Chapter II.) is
adopted, duplicate tanks capable of holding, say
seven days' supply, are provided. Here the dose
of lime required is so small as not to need neutrali-
zation and the tanks may be drawn on, on alternate
weeks, just when the water is actually needed.
Galvanized iron tanks may be used for this purpose,
the outlet pipe being placed laterally a few inches
from the bottom, Another outlet pipe may be pro-
vided for periodical flushing out of the accumulated
deposit, and this should preferably be placed on the
actual bottom (i.e., facing upwards). In adding the
lime, enough should be employed for the volume
of water down to the level of the lateral outlet pipe,
not for the total contents, which include, of course,
the bottom water which is not drawn off, except
occasionally. It is quite unnecessary to enter into
any details as regards supports, height from the
ground, &c., these being matters which the con-
sumer is in the best position to deal with. When
the foregoing method is adopted, it is assumed that
the water is so reasonably clear to start with as not
to require any filtration process. At the same time,
there is no reason why the liquid drawn off from
the large tank should not be filtered, if this is
thought desirable.

112

DESCRIPTION OF APPARATUS

There is not much, if any real, objection to the
use of chlorine as a sterilizing agent in connection
with galvanized iron tanks. The zinc may be
gradually dissolved, but, in the writer's opinion,
the danger of zinc poisoning has been greatly
exaggerated. The tanks could, of course, first be
painted with " bituminous " or other suitable paint,
but the water would be apt at first to have an
appreciable taste. On the whole, however, the
writer prefers lime to chlorine in those cases where
it is proposed to reduce the dose of chemical to the
lowest possible limit by greatly extending the period
of contact, for the reason that the lime never ceases
to act bactericidally until it is all carbonated,
whereas the chlorine loses its effectiveness much
more speedily.

In conclusion, the chief points to be noted are as
follows : —

The vessels or tanks and their connections
required for the domestic purification of water may
be of the simplest kind.

The method of working them presents no real
difficulties.

If filtration is considered desirable, and a sand
filter is preferred to a patent domestic filter, a home-
made one is easily prepared, and readily kept sweet
and clean.

8 113

RURAL WATER SUPPLIES AND THEIR PURIFICATION

For the "long time lime" method of sterilization,
galvanized iron tanks, holding, say seven days'
supply, may be used.

MISCELLANEOUS INFORMATION.
The following notes are meant for the non-
expert reader.

The word lime is used by some persons to describe
quicklime or calcium oxide (CaO = 56, sol. about
i in Qoo)1 ; others apply it only to slaked lime,
caustic lime, calcium hydroxide, or calcium hydrate
(CaO, H2O = 74) ; yet others use it in con-
nection with calcium carbonate or carbonate of
lime (CaCO3 = 100). Chalk, whiting, marble,
limestone, &c., are composed chiefly of carbonate
of lime (sol. about i in 50,000).

Calcium bicarbonate or bicarbonate of lime
(CaCO3, H2CO3 = 162). The chief cause of
the temporary hardness of waters. Precipitated
on boiling, or on addition of calcium hydrate.

Carbonic acid gas (CO2 = 44). Occurs in water
free as dissolved carbonic acid gas, or fully-bound
(as, e.g., in carbonate of lime, CaCO3), or half-
bound (as, e.g., in bicarbonate of lime, CaCO3,
H2CO3). Can be used to precipitate calcium
hydrate as calcium carbonate. If in excess it
re-dissolves the carbonate to form the soluble
bicarbonate of lime.

1 See Table XX, p. 117.
114

MISCELLANEOUS INFORMATION

Aluminium Sulphate or sulphate of alumina
(A12 (SOJ8, 1 8 H2O = 666). Very soluble, forms a
flocculent precipitate with alkalies, much used in
the purification and clarification of waters.

Sodium Carbonate or carbonate of soda (Na2CO
anhydrous = 106 ; Na2C03, 10 H2O crystalline
= 286). Very soluble, sometimes called "washing
soda," frequently used for removing the permanent
hardness of water due to sulphates, sodium sulphate
being formed and the carbonates precipitated.

Sodium Bicarbonate or bicarbonate of soda
(NaHCO3 = 84). Very soluble, sometimes called
" baking soda," can be used to precipitate calcium
hydrate as calcium carbonate, sodium carbonate
remaining in solution.

Sodium Hydrogen Phosphate (Na2HPO4, 12 H2O
= 358)- Very soluble, can be used to precipitate
calcium hydrate as calcium phosphate.

Phenol Phthalein Solution. — A solution prepared
by dissolving 0*2 gramme of phenol phthalein in
60 c.c. of alcohol (90 per cent.) and making up to
100 c.c. with distilled water. Used as an indicator ;
colourless in acid and bright pink in solutions of
hydrates and carbonates.

Methyl Orange Solution. — A solution prepared
by dissolving 0*2 gramme of methyl orange in dis-
tilled water, adding 25 c.c. of alcohol (90 per cent.),
and making up to 100 c.c. with distilled water.

115

RURAL WATER SUPPLIES AND THEIR PURIFICATION

Used as an indicator ; yellow colour with alkalies
and pinkish with acids.

Potassium Iodide and Starch Solution. — Rub
2 grammes of potato starch with enough distilled
water to form a paste. Add 100 c.c. of distilled
water and bring mixture gently to boiling point.
Allow to settle, decant relatively clear liquid and
add two or three crystals of potassium iodide.
Used to detect the presence of active chlorine, &c.
Citric Acid (H3C6H5O7, H2O - 208-5), Tar-
taric acid (H2C4H4O6 — 149) and diluted Phos-
phoric acid (B.P.) can be used to neutralize
calcium hydrate.

These, together with the standard lime (i c.c.
== O'ooi gramme CaO), sulphuric acid (i c.c. =
o'ooi gramme CaO), and soap (i c.c. = i degree of
hardness), solutions may' be obtained from any first
class firm of chemical dealers employing well-trained
chemists.

Calcium Hypochlorite, in the form of bleaching
powder or chloride of lime. Should contain about
33 per cent, of available chlorine. A most powerful
germicide.

" Chloros" a solution of sodium hypochlorite,
containing aboift 12 to 15 per cent, of available
chlorine. Used, like bleaching powder, for steriliz-
ing impure water.

Sodium sulphite (Na2SO3, 7H2O = 250), can
be used as an " anti-chlor." (i.e., to remove any
excess of active chlorine).

116

MISCELLANEOUS INFORMATION

TABLE XX.— SYMBOLS AND

ATOMIC WEIGHTS.

Symbols and Atomic Weights of the

Chief Elements mentioned

in this Treatise.

Symbol Atomic weight

Aluminium

Al ... 27

Calcium...

Ca ... 40

Carbon ...

C ... 12

Chlorine

Cl ... 35

Hydrogen

H ... i

Iodine ...

I ... 126

Magnesium

Mg ... 24

Manganese

Mn ... 55

Nitrogen

N ... 14

Oxygen ...

O ... 16

Phosphorus

P ... 31

Potassium

K ... 39

Sodium ...

Na ... 23

Sulphur ...

S ... 32

TABLE XXL— WEIGHTS AND MEASURES.

17 minims — i cubic centimetre
About 15 grains = i gramme

gallon = 10 Ib. = 70,000 grains = 160 ounces = 4546 cubic
centimetres

FLUID MEASURES.

i fluid drachm = 60 minims = 54*68 grains (3*552 c.c.)
8 fluid drachms = i fluid ounce
20 fluid ounces = i pint
2 pints = i quart

4 quarts = i gallon (160 ounces)

SOLID WEIGHTS.
i cwt. = 112 Ib. i ton = 2240 Ib.

About 109 grains = \ ounce
About 219 grains = \ ounce
About 328 grains = £ ounce
About 437 grains = i ounce
i 750 grains --£lb. = 4 ounces
3500 grains = % Ib. = 8 ounces
5250 grains = f Ib. = 12 ounces
7000 grains = i Ib. = 16 ounces

117

RURAL WATER SUPPLIES AND THEIR PURIFICATION

The above notes relate to the avoirdupois or
imperial standard weights and measures and they
must not be confused with apothecaries' weight.

In the apothecaries' weight : —

TABLE XXII.— APOTHECARIES' WEIGHTS.

20 grains = i scruple
3 scruples (60 grains) — I drachm
8 drachms (480 grains) = I apoth. ounce
(Note that the Imperial ounce = 437*5 grains.)

When working with liquids in small amounts, use
measures divided into fluid drachms (8 to the
fluid ounce) and minims (60 to the fluid drachm).
For larger amounts work with the pint (20
ounces), quart (2 pints = 40 ounces), or gallon
(4 quarts = 160 ounces.)

It is to be hoped that the day is not far distant
when we shall finally adopt the simple and scientific
way of measuring liquids in terms of litres and
cubic centimetres. In all scientific laboratories the
International system (both for weights and measures)
has been in vogue for many years.

When dealing with weights in small amount, the
weights (apothecaries') will be in grains, scruples
(20 grains), and drachms (60 grains), and the
ordinary apothecaries' scales may be used.

For heavier weights, use the imperial ounce
(437*5 grains) or its fraction (letter-weight balance),
and pound (16 ounces, 7,000 grains), or its fractions
(kitchen balance). Of course, the different balances
and their weights encroach, so to speak, on each

118

MISCELLANEOUS INFORMATION

other, but, generally speaking, the kitchen balance
should not be used below J Ib. and the letter-weight
balance below ounce.

CONVERSIONS.

Grammes (or cubic centimetres) into grains,
ounces or pounds, multiply by 15*432, 0*03528 and
0*0022046 respectively.

Grains, ounces or pounds into grammes (or cubic
centimetres), multiply by 0*0648, 28-35 an<^ 453'6
respectively.

Degrees Centigrade into degrees Fahrenheit,
multiply by 9, divide by 5, and add 32.

Degrees Fahrenheit into degrees Centigrade, sub-
tract 32, multiply by 5 and divide by 9.

Parts per 100,000 into grains per gallon, multiply
by 7 and divide by 10.

Grains per gallon into parts per 100,000, multiply
by 10 and divide by 7.

"Hardness" parts per 100,000 (CaCO3) into
degrees of hardness = Clark's scale = grains per
gallon, multiply by 7 and divide by 10.

Grains into scruples (apothecaries'), drachms
(apothecaries'), ounces (imperial), and pounds
(imperial), divide by 20, 60, 437*5, and 7,000
respectively.

Scruples (apothecaries') into drachms (apothe-
caries'), ounces (imperial), and pounds (imperial),
divide by 3, 21-875 (say, 21-9), and 350 respectively.

119

RURAL WATER SUPPLIES AND THEIR PURIFICATION

Drachms (apothecaries') into ounces (imperial)
and pounds (imperial), divide by 7*29167 (say, 7*3),
and 1 16*667 (sav» 1167) respectively.

Ounces (imperial) into pounds (imperial), divide
by 1 6.

All these little difficulties may be overcome by
the purchase of a set of weights from 10,000 grains
down to i/io grain. Their use renders it un-
necessary to consider the " grain value " of scruples,
drachms and ounces (apothecaries' weights), and
ounces and pounds (imperial weights). The weights
are arranged in the simple and scientific manner
pertaining to the gramme weight (and its multiples
and fractions), which apparently only our insular
prejudice prevents us from finally adopting.

Capacity in gallons of rectangular vessels :
Multiply length by breadth, by depth in inches,
and divide by 277*5.

Capacity in gallons of cylindrical vessels : Square
of half the diameter in inches multipled by 3*1416 ;
multiply product by depth in inches ; divide answer
by 277*5.

EXCESS LIME (CaO) METHOD.

Each grain per gallon of excess lime (CaO)
requires for its neutralization —

Sodium hydrogen phosphate 4*26 grains (2 grains
of sodium bicarbonate subsequently added to 'form
sodium carbonate), or—

120

MISCELLANEOUS INFORMATION

Aluminium sulphate crystals 3*96 grains
(5*11 grains of sodium carbonate crystals subse-
quently added (see, however, p. 25) to precipitate
the sulphates + an extra 2*86 grains for every
degree of permanent hardness in the water before
treatment), or —

Sodium bicarbonate, 3 grains, or —

4 'Carbonic acid water." Half the amount required
to remove the pink colour from a known amount
of the lime-treated water to which phenol phthalein
has been added, or —

Citric acid, 1*24 grains, or—
Tartaric acid, 2*66 grains.

All these substances are very soluble (sodium
bicarbonate much less soluble) and may be added,
if preferred, as solutions. One fluid ounce would
be the correct dose for 10 gallons, if the foregoing
figures are in each case multiplied by 800 and that
amount dissolved in 80 ounces (i Winchester quart
or J gallon) of water. Hot water may be used to
accelerate solution, except in the case of sodium
bicarbonate, which unfortunately is the least soluble
of those specified.

Dose of excess lime (CaO) for sterilization pur-
poses. About 3 grains (excess CaO) per gallon
with about twelve hours' contact, but a good deal

121

RURAL WATER SUPPLIES AND THEIR PURIFICATION

depends upon the quality of water being dealt
with.

Generally speaking, this is considerably in excess
of the dose actually required, but in these matters
it is well to err on the side of safety. Given
several days' contact the dose may be reduced to
a fraction of i grain, and even with only twelve
hours' contact i to 2 grains may suffice, but in these
cases skilled estimations of the actual excess of
lime (CaO) and frequent bacteriological tests are
desirable. Only an expert can be trusted to sterilize
a water with the absolute minimum dose.1

Dose of Chlorine for sterilization purposes, in
terms of available chlorine. About 0*07 and
0*0234 grain per gallon according to the quality of
the water to be treated, with about twelve hours' con-
tact. In terms of actual materials (bleaching powder
and chloros), multiply by 3 and 8 respectively.

COST OF CHEMICALS, APPARATUS, &c.

The pre-war retail prices of the chief chemicals,
apparatus, &c., mentioned are taken from well-
known dealers' price lists. (See Table XXI 1 1. )

Of course, the price of these reagents varies
greatly according to their degree of purity, and all
of them, if bought in larger quantities, would cost far
less.

1 For amount of lime, as slaked lime, see pp. 19, 30, 33, 34.

122

MISCELLANEOUS INFORMATION

The price of the standard sulphuric acid, lime
water, and soap solution would be about is., is.
and 2s. 3d. a Ib. respectively.

The phenol phthalein and methyl orange solu-
tions would each cost about 5d. an ounce.

The potassium iodide and starch solution would
cost about 2S. 6d. a Ib.

TABLE XXIII.— TABLE OF COSTS.

Per Ib. (7,000 grains

Sodium phosphate (crystals) .. ... 8d.

„ carbonate ,,

„ bicarbonate (powder)

Citric acid (crystals)

Tartaric acid „

Bleaching powder (in tins) ...

Calcium hydrate ...

Aluminium sulphate (crystals)

Wood charcoal1 (powder) ...

6d.

is. gd.
is. 6d.

5d.

44.
2s. 3d.

5d.

Dispensing hand scales (with weights, 2 drachms
to i grain) cost about 55. to 73. 6d.

A better and more delicate dispensing balance
(nominally sensitive to ^ grain) costs about i6s. 6d.
to 255.

Letter-weight and kitchen balances are (at all
events the latter) household necessities. Their
price varies greatly, according to quality, but is
comparatively small.

_r —

1 Wood charcoal can be prepared quite easily at home.
Fill an empty tobacco tin (or any other convenient tin) with
bits of ordinary firewood. If the lid fits very tightly, bore
one or two small holes through the tin. Place on a red fire,
The escaping gases will catch fire and burn, but after a time
this action ceases, and the tin becomes red hot. Leave for a
little longer and then remove tin and set it aside to cool.

123

RURAL WATER SUPPLIES AND THEIR PURIFICATION

A cheap burette stand (to hold two burettes)
costs about 2s. gd.

50 c.c. burettes cost 35. 6d. to 55. each.

The necessary small flasks, bottles, &c., cost only
a few pence each.

The lo-gallon and i -gallon mixing pans cost
about 22s. and 35. respectively.

Galvanized iron tanks holding 10, 25, 50, 75 and
100 gallons cost about 125., 125. 6d., 175., 2os., and
255. respectively.

The following useful practical notes (with plan)
on the Utilization of Rain-water for Domestic
Purposes, taken from pp. 22 to 25 of Kershaw's
Book on " Sewage Purification and Disposal," are
of interest from many points of view : —

"It may not be out of place at this point to refer
briefly to rain-water collected from the roofs of
cottages and out-buildings as a drinking water
supply for places where water is scarce. It seems
strange that this source of water supply is not more
frequently utilized in England. In Bermuda (pop.
in 1907, 21,000; rainfall about 48 in. p.a.) the
only water supply is obtained from the rainfall
falling on roofs and specially constructed "catches,"
from which it flows into large tanks cut out of the
limestone and rendered with Portland cement. All
roofs of dwelling-houses have an inclined fillet of
cement running round them to collect the rain
falling on them at one point. Unfortunately, the
water is seldom filtered. Fig. 19 shows a plan and

124

MISCELLANEOUS INFORMATION

125

RURAL WATER SUPPLIES AND THEIR PURIFICATION

section of a rain-water tank constructed by excava-
tion in rock, the walls and floors being rendered."
* * * * *

"The following table (taken from Quantity
Surveyors Diary and Tables (Metchim and Sons,
Princes Street, London, S.W.) gives the run-off of
rain from a roof having an area of 100 square feet
with varying degrees of rainfall : —

TABLE XXIV.— SHOWING DAILY YIELD OF WATER FROM
100 SQUARE FEET OF ROOF WITH VARYING RAINFALL.

Mean
rainfall

Loss from
evaporation

Requisite
capacity of
tank

Mean daily
yield of
water

Mean daily
yield wettest
year

Mean daily
yield driest
year

Inches

Per cent.

Cubic feet

Gallons

2*1

3'3

1-6

2-8

1-9

3*4

2'2

3'9

5-5

2 '5

6-1

3*6

5'5

7-1

4'3

"Thus, with a roof area of 1,000 square feet, the
mean daily yield in a district having an annual
rainfall of 25 inches would be equivalent to 2*8 x 10
= 28 gallons.

" Rain-water Separators. — As the first washings
from roofs generally contain soot, dust, bird drop-
pings, decayed leaves, &c., it is advisable to turn
to waste the first washings off the roof, before
running the rain-water to the storage tank. To
effect this, automatic contrivances termed Rain-
water Separators have been devised : that form

126

MISCELLANEOUS INFORMATION

known as the ' Roberts ' separator, made by
Rogers, of Haslemere, being perhaps as well known
as any.

" These separators are guaranteed to take a rain-
fall at the rate of 2 inches per hour, and to work
with a rainfall of J inch per twenty-four hours on
the areas given in the table below. In tropical
countries where the rainfall sometimes exceeds
2 inches per hour, and in smoky cities, large
separators are needed.

"It should be remembered that the separator
must correspond to the size of the roof if it is to
work properly.

TABLE XXV. — GIVING PARTICULARS OF RAIN-WATER SEPARATORS.

Area of roof in square feet

Distance between the
levels of inlet and
outlet pipes

Free to any
railway
station in
Great Britain

£ s. d.
300
4 10 0
600

7 10 o

900
10 10 0

No.
I

3
5
7
9
ii

In the country

600 to 1,000
1,000 „ 3,000
3,000 ,, 5,000
5,000 „ 7,000
7,000 ,, 9,000
9,000 ,,11,000

In the city or tropics

700 to 2,000
... 2,000 ,, 4,000
... 4,OOO ,, 6,OOO

... 6,000 ,, 8,000
... 8,000 ,, 10,000

Pure outlet
Inches

::: f ::.

'.'.'. 8 '.'.'.
... 9 ...

... 10

Foul outlet
Inches
II
13 -
13* -

:55i ::

16 ...

" Each further increase of 2,000 square feet in
area of roof adds 305. to the cost. The makers
observe that the separators need no attention
except washing out at intervals of about three
months in the country and one month in town.

" With regard to the construction, size, and posi-
tion of the storage tank, this should be constructed
of cement concrete, or brickwork in cement with

127

RURAL WATER SUPPLIES AND THEIR PURIFICATION

excess manholes, and placed underground for
reasons of temperature. According to the makers
of the separators, it should be capable- of holding
a rainfall of about 4 inches, or, say, i cubic foot for
each three superficial feet of roof, so that for a roof
having an area of about 2,000 square feet the
storage tank should hold 666 cubic feet or about
4, 1 50 gallons ; a larger capacity is desirable, how-
ever, if it can be obtained, but much depends upon
what purposes the water is used for. In the case
of most houses a tank holding 100 days' supply
would be * drought-proof.' The water before being
used for drinking purposes should be passed through
an efficient filter.

" There is no question that many rural districts
might make far more use of rain-water for domestic
purposes than is now the case."

In this brief account of Rural Water Supplies and
their Purification the Author has doubtless failed
in many particulars, but if the reader cares to write
and explain his (or her) difficulties, or to offer any
suggestions or criticisms, the Author will endeavour
to answer any such communications to the best of
his ability.

A. C. HOUSTON,

19, Fairhazel Gardens,
London, N W.

128

INDEX

INDEX.

Acids :

For estimating alkalinity, etc., 51, 55,

Il6, 122

For neutralizing lime (CaO) :
Citric, 34, 116, 121, 123
Phosphoric, 33, 116
Tartaric, 34, 116, 121, 123
Actual experiments:

Results of (Chaps. VI.andVII.),69-iol
Algal (and other) growths :

See Growths (in water)
Alkalinity :

Determination of, 51, 55
Aluminium sulphate:

For clarification, 36, 41, 46, 47, 78,

93,99

For neutralizing CaO, 24, 28, 31
Analysts :

Should be consulted, 52, 62
Animals :

Contrast between pollutions of human

and " lower animal" origin, 61
" Anti-chlor. " :

See Sodium sulphite
Apparatus :

Description of, 102-114

Filtration methods, 107

Floating-arm method of draw-off, 102

Mixing pan, 102

Rigid outlet method, 106
Atmospheric pollution:

Committee's investigations, I

Composition of, 2

Gas and electric heating v. coal fires, 4
Atomic weights, 117
Awakening :

Great awakening in progress, vii.

Bacillus coli :

Why its destruction is evidence of

safety, 70
Presumptive and confirmatory tests

(see these headings)
Reasons for applying the test, 70
Bacteria (microbes, bacilli) :
See Bacillus coli

See Typhoid fever (typhoid bacillus)
Baird and Tatlock :

Reference to, xv.

Bale, Messrs. John, Sons and Daniels-
son, Ltd.:

Publishers of this book and " Rivers as
Sources of Water Supply," i., ii., v.
Bermuda :
, Population of, 2, 124

Rain-water, supply of, 2, 124
Birds (also rats, mice, insects, etc.) :
In relation to pollution of rain-water,

8, n, 38, 126

Bleaching powder (chloride of lime) :
See Chlorine

Strength in available chlorine, 39
Braisted (Surgeon-General) :

U.S. Navy and distilled water, 36
Brooks (burns):

See Rivers
Buildings :

Destructive action (on) of sulphur and
ammonia, 4

Call

Of the country, viii.
Cambridge University Press:

Publishers of "Sewage Purification
and Disposal," 124

129

RURAL WATER SUPPLIES AND THEIR PURIFICATION

Capacity :

In gallons of rectangular and cylin-
drical vessels, 120
Carbon :

In air, 3
Carbon-dioxide :

In air, 2

In water, 14, 49
Carbonic acid:

Amount of lime needed to combine
with CO2 in water, 29

" Carbonic acid water," 26

In air, 2

In water, 14, 49
Changes :

Impending, vii.
Chapter I. :

Rain-water, 1-12

Summary, i

Concluding remarks, 12
Chapter II. :

Rain-water (continued), 13-34

Summary, 13

Concluding remarks, 34
Chapter III. :

Rain-water (continued), 35-47

Summary, 35

Concluding remarks, 46
Chapter IY.:

Well-water and spring-water, 48-57

Summary, 48

Concluding remarks, 57
Chapter Y. :

Rivers and lakes (taste of water),58-68

Summary, 58

Concluding remarks, 67
Chapter YI.:

Actual experiments, 69-89

Summary, 69

Concluding remarks, 88
Chapter YIL:

Actual experiments (contd.)t 90-101

Summary, 90

Concluding remarks, 101
Chapter YIII. :

Description of apparatus, 102-114

Summary, 102

Concluding remarks, 113

Charcoal :

As an " anti-chlor.," 43
For removing taste, 64
How to prepare it, 123

Chemicals :

Atomic weights, &c., of, 117
Description of, 114

Chloride of lime (bleaching powder) :

See Chlorine
Chlorine (for sterilization purposes) :

As bleach solution, 37
As chloros, 37

As electrolytic compounds, 38
As liquid chlorine, 44
Chlorine sterilization and clarification
by means of aluminium sulphate, 41
Ditto plus charcoal, 43
Combined with lime sterilization, 43
Dose for sterilization purposes, 39, 122
Potassium iodide and starch test for

active chlorine, 40, 116
Prejudice against its use, 38
Preparation of bleach solution, 39
Results of actual experiments with
bleach solution and rain-water, 69
Results of actual experiments with

lime and rain-water, 76
Sodium sulphite as an "anti-chlor.,"

4i> H6

Strength of bleaching powder (or
chloride of lime) and of chloros in
terms of available chlorine, 37
"Chloros":

For sterilization purposes, 37
Strength in available chlorine, 37
See also Chlorine
Coagulants :

See Aluminium sulphate
Coal:

Fires wasteful and polluting, 5
Products of combustion, 4
Coastguard stations:
Water supply of, 2
Colon bacillus:

See Bacillus colt

Combined lime and chlorine methods
of sterilization, 43

130

INDEX

Committee on atmospheric pollution :

Expression of results, 6

Standard gauge, 6
Comparison

Between drinking and breathing im-
purities, 5
Concluding remarks:

(See end of each chapter)
Conclusions

At end of each chapter

See concluding remarks
Confirmatory B. coli test :

See Chap. VI., Series L, II., III. and

IV. (I., II., III.), 69-83
Contamination :

Of atmosphere, 3

Of water, i, 5, 8, 10, 12, 38, 48, 59,

61, 70-101
Contents :

Arrangement of, xiii.
Conversions :

English weights and measures into
Continental weights and measures,
&c., 119
Cost of

Chemicals, apparatus, &c., 122
Country :

Rural yearnings foreshadowed, viii.

Dakin

On sterilization, 46
Dedication

(To Sir William Osier), iii.
Deposit

From rain, nature of, 7
Dose

Of chlorine for sterilization, 39, 43,

69, 81, 122

Of lime for sterilization, 17, 19, 29,
3i, 33. 43. 76, 82, 83, 86, 88, 90,
93, 96, 101, 121
Dunham

On sterilization, 46

E
Elements :

Symbols and atomic weights of, 117

Estimations of

Amount CO2 required to neutralize

CaO, 26
Amount of lime (CaO) rendered inert

by certain substances in water, 14, 49
Bicarbonates, carbonates and

hydrates, 55

Excess lime (CaO), 14, 55
Permanent hardness, 18
Sulphates, 17
Total hardness, 50
Excess lime method

As applied to rain-water, 13, 76, 82
As applied to river-water, 61, 78, 83,

88,99

As applied to well-water, 49, 96

Dose required (see Dose)

How added to water, 19

How estimated, 14, 55

Lime, time and sterilization, 32, 84

Neutralization of excess, 22-31, 120

Simple " blind " methods, 29
Excremental pollution :

Human, I, 10, 12, 61

Lower animals, 8, n, 38, 126
Experiments :

The result of actual experiments, 69,
101

Factors for

Amount to neutralize active chlorine of

Sodium sulphite, 41
Amount to neutralize aluminium sul-
phate of

Sodium carbonate, 24
Amount to neutralize CaO of
Aluminium sulphate, 24, 12 1
Sodium bicarbonate, 25, 121
Sodium phosphate, 22, 120
Available chlorine in bleaching pow-
der, 39

Available chlorine in chloros, 39
Figures :

Description of, xv.
Filtration methods :
Home-made filter, 107
Patent filters, 107
Future

Possibilities r; rural water supplies.vii.

131

RURAL WATER SUPPLIES AND THEIR PURIFICATION

Oases

In air from coal combustion, 4
Gibraltar:

Water supply of, 2
Growths (in water) :

Algal and other growths, 66

Anabsena, 66-67

Chara, 66-67

Eudorina, 66-67

Glenodinium, 66-67

Pandcrina, 66-67

Sponges, 66-67

Stephanodiscus, 66-67

Synura, 66-67

Halazone

For sterilization purposes, 46, 47
Hardness :

Estimation of, 18, 50

Hard waters, 49

Permanent, 18

Removal of, 96

Soft waters, 13, 93

Temporary, 50

Total, 50
Heat

As a sterilizing agent, 35, 98
Help

Offer of, 128
Hypoehlorites from salt solution :

Hypochlorites (calcium), see Bleaching
powder

Hypochlorites (sodium), see Chloros

Illustrations :

Description of, xv.
Indebtedness to, xv.

Indicators :

Methyl orange, 16, 55, 115
Phenol phthalein, 14, 55, 115
Potassium iodide and starch solution,
40, 116

Information (lack of)

In relation to purification of rural
water supplies, vi.

Iron (in waters), 66

Kershaw, G. B. :

Author of 'Sewage Purification and

Disposal," 124
Views on rain-water supplies, xv., 124

Lakes (reservoirs) :

See Rivers
"Lancet":

Reference to, 2
Lime:

Internet Archive Audio

Featured

Top

Images

Featured

Top

Software

Featured

Top

Books

Featured

Top

Video

Featured

Top

Mobile Apps

Browser Extensions

Archive-It Subscription

Save Page Now

Full text of "Rural water supplies and their purification"

See other formats