Skip to main content

Full text of "Observations on the Construction of Healthy Dwellings"

See other formats


This is a digital copy of a book that was preserved for generations on library shelves before it was carefully scanned by Google as part of a project 

to make the world's books discoverable online. 

It has survived long enough for the copyright to expire and the book to enter the public domain. A public domain book is one that was never subject 

to copyright or whose legal copyright term has expired. Whether a book is in the public domain may vary country to country. Public domain books 

are our gateways to the past, representing a wealth of history, culture and knowledge that's often difficult to discover. 

Marks, notations and other maiginalia present in the original volume will appear in this file - a reminder of this book's long journey from the 

publisher to a library and finally to you. 

Usage guidelines 

Google is proud to partner with libraries to digitize public domain materials and make them widely accessible. Public domain books belong to the 
public and we are merely their custodians. Nevertheless, this work is expensive, so in order to keep providing tliis resource, we liave taken steps to 
prevent abuse by commercial parties, including placing technical restrictions on automated querying. 
We also ask that you: 

+ Make non-commercial use of the files We designed Google Book Search for use by individuals, and we request that you use these files for 
personal, non-commercial purposes. 

+ Refrain fivm automated querying Do not send automated queries of any sort to Google's system: If you are conducting research on machine 
translation, optical character recognition or other areas where access to a large amount of text is helpful, please contact us. We encourage the 
use of public domain materials for these purposes and may be able to help. 

+ Maintain attributionTht GoogXt "watermark" you see on each file is essential for in forming people about this project and helping them find 
additional materials through Google Book Search. Please do not remove it. 

+ Keep it legal Whatever your use, remember that you are responsible for ensuring that what you are doing is legal. Do not assume that just 
because we believe a book is in the public domain for users in the United States, that the work is also in the public domain for users in other 
countries. Whether a book is still in copyright varies from country to country, and we can't offer guidance on whether any specific use of 
any specific book is allowed. Please do not assume that a book's appearance in Google Book Search means it can be used in any manner 
anywhere in the world. Copyright infringement liabili^ can be quite severe. 

About Google Book Search 

Google's mission is to organize the world's information and to make it universally accessible and useful. Google Book Search helps readers 
discover the world's books while helping authors and publishers reach new audiences. You can search through the full text of this book on the web 

at |http: //books .google .com/I 


















Late Royal Engineers, C.B., Hon.DX.L., F.R.S., Assoc. Inst. C.E,, Hon. M.I.B.A., M.I.M.E. 

F.G.S., F.C.S.f F.R.G.S.y ^c. Formerly Secretary Railway Department Board of Trade , 

Assistant Inspector-General of Fortifications^ Assistant Under Secretary of State 

for Wary Director of Ptiblic Works and Buildings, 6fc. &*c. 


[i4// rights reserved"] 




The researches of the physiologist and of the 
medical man into the laws which govern the preva- 
lence of diseases have enabled them, by the gradual 
accumulation of information, to lay down the principles 
upon which the healthy construction of houses should 
rest. It is the duty of the architect, the builder, the 
engineer, and the surveyor to apply these principles, 
and their correct application is as essential to the efifi- 
cient construction of a dwelling as is the quality or 
strength of the materials which are used to build the 

The object which the author has in view in this trea- 
tise is to present in a condensed form a short resume 
of the very scattered information which exists bearing 
on the construction of healthy dwellings, whether 
houses, hospitals, barracks, asylums, or prisons; this 
information was embodied originally in a series of 
lectures delivered to the officers of Royal Engineers 
in their educational establishment at Brompton Bar- 
racks, Chatham. The author has expanded the in- 
formation thus collected into this volume, which 
contains partly an enunciation of principles, and partly 
brief sketches to elucidate the application of these 

principles. The field embraced is so vast that it was 


V fX an d 

VI Preface. 

beyond the scope of this work to enter into the details 
which a complete text book would entail; and those 
whose business calls upon them to apply practically 
the principles of sanitary construction, must seek the 
further information they may need from one or other 
of the numerous separate treatises already in existence 
on each subject. 

There is necessarily little that is new in this work. 
It contains much that the author has derived from 
personal experience, as well as from the opportu- 
nities he has had of becoming acquainted with the 
progress of sanitary construction in many European 
countries, in the United States of America, and in our 
Indian dominions, much also has been collected from 
published books ; and he trusts that the several 
authors of thes^ books will accept the acknowledge- 
ment of their works in this preface, and pardon 
him for not having invariably furnished references to 
the extracts in the body of the work. The following, 
among these works, may be usefully referred to by 
those who would pursue the subject further; as well 
as by architects, builders, engineers, and surveyors, 
who are concerned in the application of details, viz : — 
The Reports of the Barrack and Hpspital Improve- 
ment Commission, the Reports of the Commission on 
Cubic Space in Workhouses, and Pollution of Rivers, 
Army Sanitary Commission, Indian^ Sanitary Reports, 
Reports of Board of Health, Reports of Privy Council, 
Reports of Local Government Board, Dr. Parkes on 
Hygiene ; the works of General Morin, M. Tresca^ M. 
Joly, Dr. De Chaumpnt, F. Sander (Leipzig), R, Raw- 

Preface. vii 

linson, Richardson, Baldwin Latham, H umber on Water 
Supply, Bailey Denton on Sanitary Engineering, 
Dr. Edward Smith's Manuals, Hellyer on Healthy 
Houses, Teale's Sanitary Defects, Buchan's Plumbing, 
Hood on Warming and Ventilation, Leeds on Venti- 
lation (New York), Edwards on Fireplaces, Rankine s 
Tables, Balfour Stewart on Heat, Box on Heat, Peclet 
sur la chaleur, Symons' Rainfall, Buchan s Meteorology, 
Ansted's Geology, Tidy's Chemistry, Hurst's Archi- 
tectural Surveyors Handbook, Dr. Angus Smith's 
works, also numerous papers on Sanitary science in the 
Proceedings of the Royal Society, Chemical Society, 
Institutions of Civil and Mechanical Engineers and 
Surveyors, Society of Arts, etc., and many others. 

b 2 




Preliminary Observations 1 

Conditions which Regulate the Healthiness of a Site . 4 

Effect of Soils and Local Conditions on Healthiness of Site 20 


Conditions which Regulate the Healthy Arrangements of 
Dwellings on a given Area 30 

Purity of Air in an Occupied Building .... 37 

Cubic Space and Floor Space .48 

X Contents. 



Movement of Air 62 


Preliminary Considerations upon the Inflow and Removal 

OF Air 77 

Simple Ventilation without Warmed Air .... 85 

Observations on Warming . 95 

Action of Open Fireplaces on Ventilation . . .120 


Ventilation in combination with Warmed Air in Rooms 

WHERE Open Fireplaces are not in use . . .136 

Conditions affecting the Internal Arrangements of Buildings 156 

Conditions affecting Materials and Details of Construction 180 

Purity of Water 198 

Removal of Refuse 225 

Contents. xi 



Drains in a Dwelling 239 

Conditions to be observed in Main Sewerage Works . .256 

Disposal of Water-carried Sewage 273 

Conclusion 286 

^ND£X •..*....•.• £\j L 

■» — 

» mm 



he science of the 

science of counter- 

which arise in the 

-lese influences may 

at, light, electricity, 

il, air, water, food. 

conditions of sex, 
ofession, family, or 

tudy of all these, 
iisideration of those 
conditions un-iw* — isses, in which the 

architect and the engineer are ino. . ally concerned. 

The statistician and the physiologist supply the data which 
enable us to determine what is a healthy condition of life ; 
and their researches show that the majority of diseases, and 
the greater part of the low health, which prevail in any 
country arise from causes which are within man's control. 

For instance, epidemic diseases are observed to occur in 
very different degrees of intensity at different periods, amongst 
groups of population exposed to certain unhealthy conditions. 





Hygiene is generally defined as the science of the 
preservation of health — that is to say, the science of counter- 
acting the influences injurious to health which arise in the 
surroundings in which man is placed. These influences may 
be classed as: — 

1st. Physical ; such as conditions of heat, light, electricity, 

sound, &c. 
2nd. Chemical ; such as conditions of soil, air, water, food. 
3rd. Biological or individual ; such as conditions of sex, 

age, inheritance, constitution, temperament. 
4th. Social ; such as conditions of profession, family, or 

Hygiene, in its wider sense, includes a study of all these. 
The present treatise is limited to a consideration of those 
conditions included in the first two classes, in which the 
architect and the engineer are more especially concerned. 

The statistician and the physiologist supply the data which 
enable us to determine what is a healthy condition of life ; 
and their researches show that the majority of diseases, and 
the greater part of the low health, which prevail in any 
country arise from causes which are within man's control. 

For instance, epidemic diseases are observed to occur in 
very different degrees of intensity at different periods, amongst 
groups of population exposed to certain unhealthy conditions. 



2 Preliminary Observations. 

Sometimes they take the form of pestilences, and immediately 
afterwards, the conditions remaining the same, they subside 
and all but disappear, again to renew their ravages at some 
future periods. 

Whilst we have no knowledge of why these epidemics 
break out at one time and not at another, there are certain 
well-defined conditions which influence materially, not only 
their actual intensity, but also their frequency. 

Thus intermittent fever was observed to disappear from 
places which formally suffered from it, after drainage of the 
soil and improved cultivation. 

By cleanliness, fresh air, and diminished crowding, the very 
worst forms of pestilential fever, which used to commit ravages 
similar to those of the plague, disappeared entirely from 
English gaols. 

The breathing of foul air contaminated by the breath of 
other persons appears to be the special agent which de- 
velops consumption, and-diseases of that class. 

Consumption and tubercular disease used to be rife in 
the British army, because barrack rooms were crowded and 
unventilated, and the atmosphere close or foul during the 
hours of sleep, when the system is more peculiarly predisposed 
to its effects. Out of such an atmosphere, which the men had 
been breathing night after night, they were taken and ex- 
posed on guard to wet and cold, and the disease soon 
developed itself. The cessation of these preventible causes 
has led to a great diminution of the disease. 

Zymotic diseases, namely — fevers, diarrhoea, cholera, 
dysentery, &c., are most intensely active where there is 
over-crowding, and the repeated breathing of air already 
breathed, such air being further contaminated by moisture 
and exhalations from the skin ; equally poisonous are emana- 
tions proceeding from animal excretions, or from decaying 
vegetable nniatter, together with moisture, the want of drainage, 
and a foul state of latrines, urinals, cess-pits, and me^nure 

Preliminary Observations, 3 

heaps. Moreover, cholera and dysentery are intimately 
connected with the condition of the water supply : while 
an epidemic prevails, the question whether a given population 
shall suffer or escape may almost be predicted from a chemi- 
cal analysis of the drinking water. 

Hence, the causes of the deteriorated health which He more 
especially within the scope of the present work, arise from 
poisons in the soil we live on, the air we breathe, or the 
water we drink — emanating from decomposition, which is 
the result of the previous occupation of the locality by some 
form of animal or vegetable life. 

When, therefore, the degree of health in a community, 
a household, or an individual, falls below the ascertained 
standard of good health, it is the duty of every individual 
in that community to seek out the removeable causes in 
operation which are injurious to health, and to remove 

Practical sanitary science is thus embodied in the words, 
pure air, pure water — these conditions include a pure subsoil. 
Where these exist, the highest d^ree of health of which 
the human constitution is capable may be anticipated. 
These conditions imply a healthy site, a healthy dwelling, 
and the rapid removal from its vicinity of everything that 
is liable to putrefy. 

Whilst it is for the physiologist and chemist to point 
out the prevalence of the poison, it is the function of the 
engineer and architect to devise means for obviating its 
baneful effects. 

B % 




The healthiness of a site depends, not only on the position 
itself^ but also on what lies around it. 

The selection both of the position for, and the form of, 
a dwelling must largely depend on the climate ; that is to 
say, on temperature, on rainfall, on moisture of soil, on the 
nature and prevalence of winds. 

The engineer may modify the conditions and temperature 
of the soil, and diminish atmospheric damp by drainage ; he 
may alter the moisture and temperature of the air by plant- 
ing or removing forests; he may produce changes in the 
immediate surroundings of a locality; but those general 
climatic conditions of a country which are due to position on 
the globe and to the vicinity of seas or continents are beyond 
the control of an engineer. 

I. Temperature of Atmosphere, 

In classifying the temperature of a country, it may be 
said that when the mean temperature of the locality lies 
between the thermal equator and the isothermal line of 
80° Fahrenheit, the climate is torrid ; between the isotherm 
of 80° and that of 60° the climate is hot ; between the 
isotherm of 60° and that of 40® the climate is temperate ; 
between the isotherm of 40° and 24° Fahrenheit, the climate 
is cold. A greater degree of cold represents a polar climate. 

But temperature decreases with the altitude of a district 

Healthiness of a Site. 5 

above the sea. This decrease is, however, subject to many 
variations dependent upon latitude, situation, dampness and 
dryness of atmosphere, upon the surroundings of the locality, 
whether water^ or forest, or desert, or mountains, and even 
upon the season of the year and hour of the day. A general 
rule, subject to these variations, has been laid down as follows ; 
viz. each 300 feet of height above the sea lowers the tem- 
perature about 1° Fahrenheit. 

Similarly, in passing from the equator towards the pole, 
the mean temperature may be said to be lowered by about 
9*^ Fahrenheit for every 10° of latitude. A greater variation 
of climate, however, than that due to latitude arises from 
other causes ; for instance, the relative proximity of a place 
to the ocean, the temperature of which prevents an extreme 
degree of cold in countries bordering thereon ; also, the 
effect of mountain ranges, which deprive the winds which 
pass over them of their moisture, and allow of a more com- 
plete radiation of heat from the ground, during the long 
winter nights, and thus produce an intense degree of cold. 

The position of a site as compared with the level of the 
adjacent country affects its temperature. For instance, when 
the air in contact with declivities of hills and rising ground 
becomes cooled by the ground, the cold air will flow down 
the sides of hills into the valleys, displacing the warmer air 
and forming as it were pools of cold air. Thus rising ground 
is, never exposed to the full intensity of the cold. 

It is beyond the scope of this treatise to enter fully into 
the subject of climatic conditions and temperature. It may 
however be observed, that the death-rate in any locality has 
been stated to increase with the increase in the difference of 
the mean temperature of January and July of that locality. 

One of the principal dangers to health in torrid and hot 
climates lies in the fact that heat and moisture especially 
favour the decomposition of animal and vegetable matter ; 
and these climates are, from this cause, liable to be eminently 

6 Conditions which regulate the 

unhealthy. A high range of mean temperature is therefore 
an important element in tropical and subtropical climates. 
In the temperate and cold latitudes the conditions of decom- 
position are not so intense ; and where they exist they are 
more easily controlled. 

2. Temperature of Soil. 

Daily changes of temperature do not affect the soil to 
a greater depth than three feet, varying with the daily 
range of temperature. 

The annual variation is dependent on the conductivity 
and specific heat of the soil, but it does not penetrate below 
40 feet, and below %^ feet it is very small. The mean 
temperature of the soil follows slowly the mean temperature 
of the air. The highest annual temperature of the trap 
rocks at Calton Hill, Edinburgh, at a depth of 24 feet, 
takes place about the 4th of January, and the greatest 
cold about the 13th of July. At Greenwich, which is on 
the tertiary gravels, the highest temperature at a depth 
of %^,6 feet occurs on the 30th of November, and the lowest 
on the first of June. At a depth of 12.8 feet the highest 
temperature occurs on the 25th of September, and the 
lowest on the 27th of March. For all practical purposes 
the temperature in the soil at a depth of from six to eight 
feet, may be said to be practically fixed all the year round, 
because it follows so slowly the summer and winter changes 
that it never attains summer heat or winter cold. 

The temperature of the earth increases with the depth. 
This rate of increase of temperature varies in different 
geological formations. 

In the Paris basin it has been estimated at 1° Fahrenheit 
for every 55 feet of depth. In England it has been stated 
at 1° for every 54i feet. It is also stated that it varies with 
the latitude, being lower in the higher latitudes, and higher 
towards the equator. 

Healthiness of a Site. 7 

The mean rate of increase over the globe may be approxi- 
mately assumed at i® Fahrenheit for every 50 feet in depth. 

The internal heat exercises, under certain conditions, an 
influence over the mean temperature of the surface soil of 
a locality. 

Where the rainfall is tolerably evenly divided over the 
year, the average annual temperature of the soil will be that 
of the climate of the locality. But in countries with distinct 
wet and dry seasons, the mean temperature of the soil will 
not necessarily be the same as that of the mean temperature 
of the air. Snow, being a bad conductor, prevents the 
passage of the heat from the earth into the air, and thus in 
countries where snow lies for some time on the ground, the 
mean temperature of the earth exceeds that of the air. 

The temperature of water in permanent springs is neces- 
sarily derived from the subsoil line of fixed temperature, and, 
except in the cases just alluded to, it will not be found to 
vary more than i® or a° from the mean temperature of the 
locality; therefore the temperature of a permanent spring 
may be assumed to afford a certain guide to the mean 
temperature of a district. 

Thus, in England, the permanent springs range in tem- 
perature from 49° to 5 1*', the mean annual temperature being 
50°. In India the springs will be found to varj^* in different 
parts, according to the temperature of the locality ; in some 
instances they attain a temperature of from 70° to 80°. 

But whilst the mean temperature of the ground depends on 
the climate, soils have a very varying conducting capacity for 
heat ; loam, clay and rocks are better conductors than sand, 
and by allowing the sun^s heat to pass more rapidly down- 
wards, do not become heated to so high a degree. 

The conducting capacity of the soil has a very important 
bearing upon the comfort, if not upon the health, of those 
wiio live upon it. 

The following table shows the relative power of soils to 


Conditions which regulate the 

retain heat : sand being the worst conductor, lOO is allotted 
to it as the standard : — 

Sand, with some lime . . . 100*0 

Pure sand 95-6 

Light clay 76*9 

Gypsum 73-2 

Heavy clay 71-11 

Clayey earth 68.4 

Pure clay 66'7 

Fine chalk 6i-8 

Humus 490 

The great retentive power of the sands is thus evident, 
and the comparative coldness of the clays and humus. 

Under exposure to the sun's rays, herbage lessens the 
absorbing power of the soil, and radiation is more rapid 
from it, because a portion of the heat is lost by the evapora- 
tion which goes oh from the pores of plants, and the leaves 
are rapidly robbed of their heat by the adjacent air. 

Changes of temperature take place slowly in trees as 
compared with the temperature of the air. Trees acquire 
the maximum temperature after sunset, whilst the maximum 
temperature of the air occurs between 2 and 3 p.m. Hence 
the influence of trees is to make the night warmer and 
the days colder ; and the heat is more evenly distributed 
over the twenty-four hours in countries covered with vegeta- 
tion than in those free from it. 

Evaporation goes on slowly under trees ; but the vapour, 
not being so liable to be removed by the wind, accumulates 
among the trees. Hence, whilst forests diminish evaporation, 
they increase humidity, and they keep the summer tempera- 
ture lower, and the winter temperature higher than it would 
be without them. For these reasons, forests may act in the 
same way as a range of hills, to increase rainfall in the 
summer by causing condensation in the case of warm moist 

3. Moisture of Soil, 

The condition which, more than any other, governs the 
healthiness of the soil, is the relation which the ground air, 

Healthiness of a Site. 9 

or air in the soil, bears to the ground water ; that is to say, 
the presence or absence of moisture in the soil. 

The moisture in the soil, or ground water, depends upon 
the amount and mode of incidence of the rainfall; for, as 
a rule, rainfall is the parent of the ground water. 

Rain varies greatly, both as regards frequency and rate 
of fall. In some places it never rains. The heaviest re- 
corded average annual rainfall is said to be 600 inches. 
This occurs on the Khasia Hills, which rise abruptly opposite 
the Bay of Bengal, and are separated from it by 200 miles 
of swamps. As much as 700 inches have fallen in a year. 
Five-sixths of the quantity of rain falls in about half the 
year; 264 inches have been recorded to have fallen in the 
month of August alone, and 30 inches to have fallen in one 
period of 24 hours. 

The basin of the Indus contains districts where the rain- 
fall is sometimes as low as six inches in the year. 

In England the average rainfall is 32 inches, but the 
average gives a wrong impression of the condition of different 
parts of the country. 

In the west of Great Britain and Ireland, in the immediate 
neighbourhood of hills, the average rainfall is above 75 
inches ; and in some localities 150 inches have been observed : 
in some years it is even higher. In the east of Great Britain, 
from 20 to 28 inches of rain falls. The amount of rainfall 
is affected by proximity to the sea, as well as by mountain 
ranges and hills, from these latter, and other causes, it varies 
materially at places a short distance apart, and therefore 
each locality must be considered separately. 

The average of a series of years is not what an engineer 
must look to. He has to deal with the maximum or mini- 
mum. In questions of water supply, the minimum of the 
yearly fall is what must influence his calculations. But in 
questions of the removal of water, he must look to the 
maximum, not of the yearly fall only, but to the greatest 

lo Conditions which regulate the 

amount which may fall in a limited period. The driest 
years test water supply ; the wettest years test works. 

In the ^4 hours ended 9 a.m., on the 15th of July, 1875, 
there was recorded as having fallen at Newport, 5.33 inches ; 
Tintern, 5.31 inches ; Cardiff, 4.7 inches. 

There are other instances on record of rain having fallen in 
England at the rate of four inches in an hour on a limited area. 
In fact the rain in this country occasionally falls as heavily as 
in India, but a heavy fall does not last for so long a time. 

It has been estimated from observations in this country 
that the maximum annual rainfall exceeds by one-third, 
and the minimum annual rainfall is less by one-third, than 
the mean rainfall of a series of years. It has also been 
observed that the average annual rainfall of three consecutive 
dry years amounts to 80 per cent, of ,the mean annual rain- 
fall of a series of years. 

The time of the year at which the rain falls, and the 
resulting effects on the air and the soil, is a point of great 
sanitary importance. 

In dry seasons the re-evaporation is rapid. In India, 
rain may fall at such intervals in a dry season as to allow 
of entire re-evaporation. In wet seasons water falls on water 
and flows off in floods. 

Experiments would appear to shew that in this country, 
on an average, nearly two-thirds of the mean annual rainfall 
goes back at once into the atmosphere as evaporation ; but 
these experiments were chiefly made on parts of the country 
where the rainfall does not materially vary from the mean 
annual rainfall. A further portion evaporates slowly from the 
soil, or is absorbed by vegetation. The water which is not 
evaporated, or absorbed by vegetation, percolates into the 
soil, to flow out in springs, streams, and rivers, to the sea,^ 

* The Upper Thames basin contains 3676 square miles; the mean daily flow 
of water over Teddington Lock during ten years, added to the water pumped 
into London by the Water Companies, was nearly 140,000,000 cubic feet; the 

Healthiness of a Site. 1 1 

whence it again passes into the atmosphere, and returns as 
rainfall or as dew. For it is evident that if the humidity of 
the atmosphere be assumed to be constant for an average of 
years, the evaporation over the whole globe must equal the 

The proportion of evaporation and percolation in different 
soils varies — 

1. With the time of year in which the rain falls. 

2. With the quality of the soil, with its capacity for heat, 

and with the character and extent of the vegetation 
with which it is covered. 

In England, on an average of years, the spring is the 
driest and the autumn the wettest part of the year. The 
driest months are March and April, the two wettest months 
are October and November. 

The greatest percolation takes place after a wet period, 
when the soil becomes saturated. In the summer there is 
scarcely any percolation. A series of experiments on per- 
colation in England, extending over 14 years, showed that in 
five of the years there was no percolation during a continuous 
period of seven months; and that in one year only, viz. 
i860, did percolation take place every month. 

In a series of experiments by Mr. Dickenson, it appeared 
that from April to September, 93 per cent, of rainfall was 
evaporated, and 7 per cent, absorbed ; whilst from October to 
March, 25 per cent, was evaporated and 75 per cent, absorbed. 

Great percolation follows the thawing of snow, and the 
greatest percolation is due to frequent small falls of snow. 

The nature of the soil affects the percolation. Through 
sand the percolation is great. The evaporation from sand 
in this country has been shown by experiment to be 16 

average rainfall in the Upper Thames basin during those years was 29.5 inched 
= 690,000,000 cubic feet per diem. The proportion which the water brought down 
by the river bore to the rainfall during the period was as i : 4'9 ; but probably 
some flowed away underground in the porous strata in which the river is situated. 


Conditions which regulate the 

per cent., and the percolation 84 per cent., whereas with 
clay and loam^ the percolation was found to be ^7, and 
the evaporation 73 per cent.^ 

In hot climates the relative power of various soils to retain 
heat would alter these proportions ; for instance, the evapora- 
tion in such cases would be large from sand if unprotected by 

The presence or absence of vegetation exercises an im- 
portant influence on percolation. When vegetation is rapid, 
as in the case of growing crops in the spring, it arrests 
percolation ; when the ground is covered with forests, the 
moisture is retained nearer the surface. 

* The following Table gives the result of various experiments on percolation 
through different soils : — 


Dalton . . 


Maurice . 
Gasparin . 

Rister . . 

Greaves .... 

Laws & Gilbert . . 

Dickenson & Evans 
Evans .... 


j Earth 

) Surface grass 

! Gravelly loam 
Surface grass 

Earth . . . 

Earth . . . 

I Impervious subsoil ) 
Sand, cropped i ' ' ' 

Mean . . 

( Loam, gravel, and sand, turfed 
I Sand 

f Loam, clay subsoil, built 
in 20" deep 

. Loam, clay subsoil, built 
I in 40" deep 

Loam, clay subsoil, built 
w in 60" deep . . . 

Surface soil at Nash Mills 

(SoU . 




3 years 







o • 










i . 

(4 V 

75 -o 











Healthiness of a Site. 13 

Dr. Ebermeyer found at Salzburg that the percolation was 
in May, %^,%" 
June, 53.1 

July, ^3.4 Vper cent, less through turf than through bare 
Aug., 29.1Z earth, 

Sept., 7^.7^ 
and that the difference was least in January. 

Ebermeyer's experiments, moreover, showed that in the 
summer half year, forest soil is moistest, bare open ground 
less moist, turf driest. 

Forests retain the moisture, protect the soil, and in moun- 
tainous countries retard the flow of torrents. Turf produces 
to some extent the same effect. 

The power of retention of water in the soil exercised by 
the planting of trees was exemplified in the Island of Ascen- 
sion. That island formed a convenient point for ships to call 
at for obtaining water on their way home from the East 
Indies. It was a barren rock, to which formerly the water 
had to be conveyed in ships. Some years ago, trees were 
planted on the Island. These have thriven, and now the rain 
which falls, instead of passing away at once into the atmo- 
sphere by evaporation, is retained in a sufficient quantity to 
fill tanks with water for the supply of the ships which call 
there. The problem of the water supply of some others of 
our insular possessions might be solved by similar pro- 

On the other hand, all growing vegetation evaporates a 
large quantity of water. In order to form i lb. of woody 
fibre, a plant evaporates i^oo lbs. of water ; consequently a 
country covered with forest evaporates an enormous quantity 
of water out of the soil, in order to produce a growth of 

When the rainfall has penetrated from \% inches to % feet 
into the ground, the loss from evaporation is comparatively 
small ; but it varies with the nature of the soil. In the chalk 

14 Conditions which regulate the 

formation, water will rise by capillary attraction from the 
level at which the chalk is saturated, to a considerable height 
above that point; while a bed of sand will be dry at the 
height of about a foot above the water standing in it. 

So long as there is water sufficiently near the surface of the 
soil to keep it moist by attraction, evaporation will continue. 
Clay and similarly retentive soils do not give off vapour as 
copiously as free open soils; therefore a given quantity of 
moisture will occupy a longer period in passing off these soils 
than is the case with free soils under similar conditions. 

The capacity of soils to retain water varies greatly. Im- 
permeable granite or marble will hold about a pint of water 
per cubic yard. Pure sand will hold 40 or 50 gallons, or 
about 17 per cent, of its weight when dry, and the ordinary 
red sandstone rock a 7 gallons per cubic yard, or from 7 to 
8 per cent, of its weight when dry. London clay will hold 50 
per cent, of its own weight when dry, Oxford clay about 
30 per cent. Experiments made by Mr. B. Latham in 1879, 
for the Royal Agricultural Society, gave the following per- 
centages of water by weight which surface soils selected from 
different localities will absorb, viz. open gravel from 9 to 13 
per cent. ; gravelly surface-soil 48 per cent. ; light sandy soils 
from 23 to ^d per cent. ; loamy soil 43 per cent. ; yellow marl 
subsoil %^S) per cent. ; stiff land and clay soils from 43.3 to 
57.6 per cent.; sandy and peaty soils from 61.5 to 80 per 
cent. ; peat 103 per cent. The conditions of two localities may 
thus vary greatly, although there may be an apparent general 
similarity in the soils. 

Under-draining facilitates the passage of the water from 
the surface into the ground ; and a smaller evaporation and 
greater percolation takes place in drained lands. 

A drained field will consequently have a temperature as 
much as 6® or 7° Fahrenheit higher than an adjacent un- 
drained field. The cause is obvious. To convert water into 
vapour absorbs 960** Fahrenheit of heat from its vicinity; 

Healthiness of a Site. 15 

thus each cubic foot of water evaporated will lower the tem- 
perature of something like 3,000,000 cubic feet of air i^ The 
lower the water in the soil, the less the evaporation, and the 
warmer the adjacent air. 

The discharge of underdrains in a free soil of chalk, gravel, 
and sand was found, by Mr. Bailey Denton, to be %\ times as 
rapid as that from underdrains in a clay soil ; the drains in 
the clay being 25 feet apart ; and those in the free soil placed 
at irregular intervals widely apart. 

The rate at which a soil allows of the percolation of water 
regulates the distance apart at which underdrains should be 
placed, for the purpose of lowering the subsoil water in land. 
In free soils a single drain will lower the water for a large 
area ; in clay soils numerous drains are necessary. The dis- 
charge from open soils is more regular than from clay; 
although clay soils often give out a large proportion of the 
rainfall immediately after it occurs. Barometric pressure 
may affect the discharge from drainage outlets ; an increased 
discharge has been observed to follow a fall in the barometer 
without any fall of rain on the surface. 

In a country where the proportions of mean annual rainfall 
vary so greatly as they do in Great Britain at comparatively 
short distances apart, the areas of heavy rainfall have an 
important bearing on water supply. On these areas the rain- 
fall is more continuous, and the actual amount evaporated will 
therefore be less than in the drier parts of the country. 
Hence the proportion to the total rainfall, of rainfall which 
can be collected from these areas, will be many times greater 
than what could be obtained under the most favourable 
circumstances in the drier parts of the country. 

4. Aeration of Soil. 

The level of water in the subsoil regulates the amount of 
ground air. 

Air permeates the ground, and occupies every space not 

1 6 Conditions which regulate the 

filled by solid matter, or by water. Thus, it is the same thing 
to build on a dry gravelly soil, where the interstices between 
the stones are naturally somewhat large, as to build over a 
stratum of air. The air moves in and out of the soil in pro-' 
portion to barometric pressure, and with reference to the wind. 
If there is much water in the soil, the air carries with it watery 
vapours, and is cold, and such a site is called damp. 

The fact of this continual free passage of air in and out of 
the ground makes it important that, not only should the 
ground lived on be free from water, but that it should also 
be free from impurities. It would be just as healthy (indeed 
probably far healthier) to live over a pigstye than over a site 
in which refuse has been buried, or in which sewer water has 
penetrated, or over a soil filled with decaying organic matter ; 
thus, before building on any ground, its nature should be 
carefully examined. 

It must, however, be remembered that in proportion as 
there is a free movement of air in the soil, so is the process of 
decay, and consequently the removal of decaying matter, 
more rapid. Louis Cr^teur, in his work * Hygiene in the 
Battle Field,' gives his experience in disinfecting the pits 
where dead were buried near S^dan. The bodies were buried 
in chalk, quarry rubble, sand, argillite, slate, marl, or clay 
soils, and the work of disinfection lasted from the beginning 
of March till the end of June. In rubble the decay had taken 
place fully, but in clay the bodies were surprisingly well kept, 
and even after a very long time the features could be identified. 

Experiments have demonstrated that there is a considerable 
quantity of carbonic acid in the ground : a result frequently 
of the decay of organic matter. Water which comes up from 
deep springs always contains carbonic acid, but it has been 
shown that the ground air frequently contains 50 per cent, 
more carbonic acid than the ground water ; it seems, there- 
fore, that the latter is supplied with its carbonic acid from the 
ground air. Recent experiments in India have demonstrated 

Healthiness of a Site. " 1 7 

that the carbonic acid in air drawn in a particular locality 
from a depth of three feet was one-half that obtained from a 
depth of six feet. 

These experiments have also shown that there are con- 
siderable variations in the amount of carbonic acid present in 
the soil of localities in close proximity ; the amount was found 
to be doubled in a distance of 50 yards, with apparently 
similar soil. The processes going on in the soil at these two 
spots must have differed materially; and if such processes 
affect health, persons inhabiting a building over one of these 
sites would be exposed to different hygienic conditions from 
persons living over the other. 

These facts show the immense importance which the soil 
on which dwellings are placed exercises upon health, espe- 
cially in cold or damp weather, when the air in the dwelling 
is warmer than the air outside ; for then the upward move- 
ment of this warmer air will draw in air to supply its place, 
from the ground on which the dwelling stands. This move- 
ment would be prevented if the whole surface under the 
dwelling were covered with an impervious material. But it is 
difficult to find a building material impervious to air. 

The following table shows the volume in cubic feet per 
hour of air which passed through a square yard of wall sur- 
face of equal thicknesses built of the following materials, the 
pressure being obtained by a difference of temperature of 72° 
Fahrenheit inside, and 40° Fahrenheit outside : 

Wall built of Sandstone 4.7 cubic feet. 

do. Quarried Limestone 6.5 „ „ 

do. Brick 7-9 ,, »» 

do. Limestone lo.i „ „ 

do. Mud 144 „ „ 

Concrete is the most convenient covering to be placed 
under a building ; it is not impermeable; the lime in it will 
absorb the carbonic acid from the ground air for a certain 
time, until it is all converted into carbonate of lime, and 


1 8 Conditions which regulate the 

then its beneficial action would be diminished. Asphalte is 
more impermeable than concrete. 

The amount of the air drawn from the soil may be checked 
by raising the floors above the surface of the ground, and 
affording free circulation of air from outside between the 
raised floor and the ground. In Burmah, the dwellings are 
raised on poles. In Italy and France it is usual to build a 
basement on open arches, which are used for wood or other 
stores, over which the dwelling is constructed. 

So little, however, has the influence of ground air been 
appreciated in this country that it is only comparatively 
recently that the occupation of cellars with walls abutting on 
the soil, as habitations, has been prohibited ; and even within 
%o years rooms in the basements of barracks have been used 
as barrack-rooms for soldiers. The plan of allowing the earth 
to rest against the walls of rooms in basements is still unfor- 
tunately common. 

These considerations show the importance of forming a 
deep open area round a house, and carrying it below the level 
of the basement floor ; between the floor and the ground there 
should always be ventilation to the outer air. Similarly, a 
trench should always be dug round a tent. Apart from the 
advantage which this affords for draining off the water, the 
sides of the trench enable the atmospheric air to permeate 
freely to the ground imjmediately under the tent, especially 
when the tent stands on a gravelly soil. 

Whilst a permanently low water level (say 15 feet) in the 
soil may be termed healthy, and a permanently high water 
level (say under 5 feet) may be termed unhealthy, a fluc- 
tuating water level is very unhealthy, especially when the 
fluctuations are rapid. The unhealthiness mainly shows 
itself when the level of the ground water falls. 

Fever chiefly occurs in flooded districts when the floods 
have receded, and in some of the Indian districts where the 
earth surface is covered by water and secluded from the air 

Healthiness of a Site. 19 

there is no cholera ; but when the water falls and the surface 
of the country is drying up cholera reappears. The unhealthi- 
ness may be due to the decay of organic matter left by 
the receding water; but it must be remembered that under 
such a condition of flooding, impure water and filth disappear 
for the time ; and when the water falls, the surface and the 
water sources are again subject to pollution, and soon become 
rife with impurities. 

It is for these several reasons that it is desirable to keep 
the permanent level of water in the soil, where habitations are 
placed, as low as possible ; that is to say, to drain it, so as to 
allow the air to have free play in the soil. But where the 
ground water cannot be maintained permanently at a low 
level, then keep it at an even level. 

Thus the presence or absence of moisture determines very 
much the degree of healthiness of soils. In any country, that 
area over which fogs appear soonest after nigfitfall should be 
avoided for camping, and should be drained before building. 
The water level of every camping ground should be examined 
by digging holes ; but a correct idea can be obtained as to 
where water is nearest the surface, from observing where the 
vegetation is greenest, where midges prevail in the day time, 
and where fogs appear soonest at nightfall. 

Apart from its effect on the moisture of the soil, vegetation 
has an influence of its own on the healthiness of a site. 
Plants in respiration absorb oxygen and throw off carbonic 
acid, but the action of the cells in which the green matter 
termed chlorophyll is formed, which gives colour to vegeta- 
tion, is to absorb carbonic acid and to eliminate oxygen. 

This action is due to the sun's rays. Vegetation is thus 
beneficial, but when in the vicinity of dwellings it should 
always be vigorous, healthy, and green; fallen leaves and 
decayed vegetation should be rapidly removed like other 
refuse from the vicinity of dwellings. 

C % 




Dr. Parke's gives the following table to show the relative 
healthiness of various geological formations, based upon their 
relative permeability — 

Permeability of 

Emanations into air. 


Primitive rocks, clay state, 
millstone grit. 




Gravel and loose sands, with 
permeable subsoils. 











Sands, with impermeable 

by subsoils. 



Clays, marls, alluvial soils. 




Marshes, when not peaty. 



Dr. Parkes states that cholera is unfrequent on granite, 
metamorphic, and trap rocks. Granite districts are usually 
sparsely peopled and the element of overcrowding as a cause 
of disease is absent. And a careful collation of facts shows 
that so far as cholera is concerned, the site and the geological 
formation have nothing to do with it : it occurs where there 
is a population in a filthy condition. And as a rule, local 

Effect of Soils and Local Conditions. 21 

conditions modify the effect of soil and geological formation. 
For instance, granitic and other impermeable formations are 
termed healthy because the impurities, instead of passing into 
the soil, are carried off rapidly by rainfall ; but if filth is 
allowed to accumulate they will be unhealthy. 

Thus during the first visitation of cholera, one of the places 
which suffered most severely, owing to' its filthy local condi- 
tion, was Megavissey, on the granitic formation in Cornwall. 

There was much sickness and mortality in 1859-60 at Hong 
Kong. The peninsula of Kowloon was selected as a sanato- 
rium. It was of granite formation, freely exposed to the 
winds; it was reputed to possess every qucllity for health. 
Huts were built, and the troops were moved into them. 
They suffered severely from fever. 

This arose from the disturbance in a tropical climate of 
the surface soil impregnated with decaying organic matter. 
Until soil of that nature has been opened and oxygenised, it 
is in the highest degree deleterious. 

Brushwood is a source of danger near camps in a hot 
climate, but the immediate result of the removal of brushwood 
has been found to cause fever, owing to the disturbance of 
decaying organic matter occasioned thereby. 

In cold countries, the clayey soils are cold ; and as they 
prevent the percolation of rainfall, they are also damp, and 
favour the production of rheumatism and catarrhs ; for these 
reasons sands are generally the healthier soils in this climate. 
Sand or gravel soils are, however, only healthy if kept entirely 
free from sewage and decaying organic matter, and from 
excess of water. If this is not seen to, then expect typhoid 
fever from foul subsoil air and polluted well water. More- 
over, sand is easily polluted, because the polluted matter on 
the surface percolates freely into sand with the aid of rain 

On the other hand, in hot countries, sands are objectionable 
from their heat, unless they are covered with grass. They do 

22 Efi-'ct of Soils and Local Conditions on 

not allow the heat to pass through, but radiate it slowly, and 
the air is hot over them day and night. 

A clay soil is a cold soil, and the air over it is always ' 
moister than over dry sand, but a clay soil cannot be easily 
polluted by sewage water like sand. In some cases fever has 
been observed to stop on passing from gravel to clay. 

On the other hand, the Indian experience, in some cases, 
has shown that fever death-rates are highest in alluvial clay 
soils and water-logged ground ; whilst the general death-rate 
was highest on porous wet soils ; and that porous wet soils 
possess no advantage in the way of escape from fever if they 
be deluged with' water. 

Pervious beds, such as sand and gravel interlaced with 
impervious beds, such as clay or shale, have the great dis- 
advantage of sweating out water at the outcrop ; this is a 
frequent cause of fever. 

A wet hill slope should always on this account be avoided, 
if at all practicable, but if it must be used, then it must be 
efficiently drained. 

The following instance will explain this. Figure 2 
shows the slope of the ground falling towards the plain of 

The formations are rock below and above, traversed by a 
belt of clay and shale. 

The 79th Highlanders were placed on the clay, and as the 

Healthiness of Site. 23 

material was soft, their huts were placed on terraces cut out of 
the hill side, and were thus embedded in the ground, and the 

Fig J 

Huts on hill-side at Balaclava. 

floors censequently were always damp. There was no roof 
ventilation. This regiment had half the men down with fever. 
The 42nd Highlanders were placed on the rock, and as it 
was hard they did not cut into the rock, but preferred building 
their huts on projecting terraces, so that they were quite dry, 
and air circulated freely round. This regiment did not suffer 
from fever, 

Fig. 3. 79th hut 

The huts on the clay were subsequently altered (Fig. 3), 
so as to allow of a clear circulation of air. Drainage and roof 
ventilation were provided, as shown in the sketch, and fever 
no longer prevailed. 

24 Effect of Soils and Local Conditions on 

In connection with this, it should be mentioned that where, 
from circumstances, tents and buildings must be placed on the 
side of a hill, a plateau should be formed to receive them, and 
a broad space be left between the hill and the tents or build- 
ings. A trench should also be cut to carry off the moisture 
between the tent and plateau, and no accumulation of refuse 
should be allowed between the hill and the tent or building. 

Although the general features of a site may be unhealthy, 
when it is absolutely necessary for any reasons, military, 
political, or otherwise, that it should be occupied, much 
may be done to remedy its unhealthiness. 

For instance, if temporary occupation only is contemplated, 
probably cutting off the water which may flow from higher 
levels, or the adoption of measures already mentioned, such as 
digging trenches round tents or huts, would be all that could 
be done : but if the ground is to be permanently occupied, 
not only the area to be built on, but an area extending to 
probably lOO yards round it on every side should be 
thoroughly under-drained, and the mouths of the drains 
so arranged as to allow the aeration of the soil^ as well as 
the removal of the subsoil water. Dwellings should be raised 
above the level of the ground, and provided with ventilated 
air spaces underneath. 

5. Effect of the Conditions of the Adjacent Districts 

on Healthiness of a Site. 

Elevated positions are generally healthy, but when these 
positions are exposed to wind blowing over marshes or mala- 
rial ground, their very elevation is a source of danger. 

In order to provide a healthy station at Jamaica, an ele- 
vated site from 3,500 to 4,000 feet above the sea level, at a 
place called New Castle, was selected for barracks. It was 
situated on the crest of a spur of land falling rapidly from the 

Healthiness of Site. 25 

Blue Mountains southwards towards the deep damp valleys 
and ravines, filled with tropical vegetation, which connect the 
range with the lower country. The sides of the ridge sloped 
down at angles of 40° and 50°. The surface was clay mixed 
with vegetable matter. The ridge was so narrow that the huts 
were placed on terraces cut out of the slopes of the hill, 
with but a few feet of space between the back of the hut and 
the soil supporting the terrace above. Even in temperate 
climates such a position contributes to fever. The result was 
that in the yellow fever epidemics in 1856 and 1867, those 
huts which were so placed that the malaria blowing up the 
valleys must necessarily strike them, yielded a large per- 
centage of yellow fever even at this high elevation. 

Algeria perhaps offers some of the best illustrations of the 
manner in which engineering operations have remedied the 
evils of the proximity of marshes. 

Bond stands on a hill overlooking the sea; a plain of a 
deep rich vegetable soil extends southward from it, but little 
raised above the sea level. The plain receives not only the 
rainfall which falls on its surface, but the water from adjacent 
mountains, and is consequently saturated with wet. The 
population living on it and near it suffered intensely from 
fever; entire regiments were destroyed by death and in- 
efficiency. It was at last determined to drain the plain. 
The result of this work was an immediate reduction of the 
sick and death-rate. 

Such instances might be multiplied. 

Irrigation, if applied in excess, is unhealthy, and should 
not be allowed near habitations. In Northern Italy irrigated 
rice grounds are not allowed within 1000 yards of small 
towns, and are required to be placed further from large cities ; 
but in these the water is allowed to stagnate in the subsoil. 
Where the water is not allowed to stagnate irrigation may be 
carried on with less danger. 

Fondouc, in Algeria, is situated on sloping ground, imme- 

26 Effect of Soils and Local Conditions on 

diately above the marshy plain of the Mitidja, and has 
mountain ranges behind it. It was first occupied in 1844, 
and in the succeeding year half the population was swept 
away by fevers and dysentery. During the first %o years the 
mortality was 10 per cent. The surrounding marsh has been 
cultivated, and there are now upwards of 10 square miles 
round the town under cultivation, producing cereals, cotton, 
tobacco, and wine. 

The cultivation consists of ploughing and trenching com- 
bined with irrigation, so that water in excess is not applied. 
The mortality now is only 20 per 1000. 

In India, wherever water is applied in excess for irrigation, 
so as to become stagnated in the subsoil, ague, spleen dis- 
ease and fever prevail. 

But it is possible to improve such localities by draining 
away the superfluous stagnant subsoil water. 

In the northern Doab districts in the North-West Provinces 
a great diminution of the excessive fever mortality for which 
these districts were noted has followed the extension of drain- 
age works, by which the water which formerly stagnated on 
and in the land is now led away by continuously flowing 

But it is not sufficient to make the drains. All drainage 
cuts are liable to become injured, if open, by vegetation, and 
in all cases by decay, by atmospheric causes and other means. 
If not properly maintained and cleared out, the evils they are 
created to remove will recur. 

For instance, at Bond, from which, as already mentioned, in 
consequence of drainage works the fever disappeared. 

The drains were left to atmospheric influences ; they be- 
came partially obstructed and irregular, and did not allow the 
water to reach the outfall ; the result was a violent outbreak 
of fever at Bond attended with great loss of life both civil and 
military, an enquiry took place, the drainage was rectified, 
and since then Bond has been healthy. 

Healthiness of Site. 27 

. The engineer thus has it in his hands to mitigate the evils 
of a marshy district by providing for the removal of stagnant 
water, and to prevent the evils which arise from irrigation by 
combining drainage with works intended for irrigation. 

6. Summary of Conclusions as to a Healthy Site. 

The following is a brief summary of the conclusions to 
which the considerations above adduced point — 

1. Clay soils should, if possible, be avoided. 

2. Ground at the foot of a slope, or in deep valleys, which 
receives drainage from higher levels, should be avoided. It 
predisposes its occupants, even in temperate climates, to 
epidemic diseases. 

3. High positions exposed to winds blowing over low 
marshy ground, although miles away, are in certain climates 
unsafe, on account of fevers. Indeed, it sometimes happens 
that a site in the immediate vicinity of a marsh, or other local 
cause of disease, especially if protected by a screen of wood, 
is safer than an elevated and distant position to leeward. 

4. Elevated sites situated on the margin or at the heads of 
steep ravines, up which malaria may be carried by air currents 
flowing upwards from the low country, are apt to become 
unhealthy at particular seasons. Such ravines, moreover, 
from want of care, are often made receptacles for decaying 
matter and filth, and become dangerous nuisances. In 
tropical climates these ravines convey malaria, and occasion 
aggravated remittent, or even yellow fevers, at an elevation 
which would be otherwise exempt from the action of tropical 

5. Ground covered with rank vegetation, especially in 
tropical climates, is unhealthy, partly on account of the 
afnount of decaying rnatter in the soil, partly because the 
presence of such vegetation is in itself a mark of the presence 
of subsoil water. Or of a humid atmosphere. 

28 Effect of Soils and Local Conditions on 

6. In warm climates, muddy sea be,aches, or river banks, or 
muddy ground generally, if it be subject to periodical flood- 
ing, and marsh land, especially if it be partly covered with 
mixed salt and fresh water, are peculiarly hazardous to 

7. A porous subsoil, not encumbered with vegetaflon, with 
a good fall for drainage, not receiving or retaining the water 
from any higher ground, and the prevailing winds blowing 
over no marshy or unwholesome ground, will, as a general 
rule, afford the greatest amount of protection from disease 
which the climate admits of. 

8. To test the healthiness of a site an enquiry into the rate 
of sickness and mortality in the district will afford valuable 
information. But care should be taken not to be guided by 
the mortality alone. The nature of the diseases, and the facility, 
or otherwise, with which convalescences and recoveries take 
place, must also be taken into account. 

To sum up these conclusions for this country for a site 
tQ be selected for occupation. The local climate should be 
healthy ; the soil should be dry and porous ; it should be 
protected from the north and east by shelter at a sufficient 
distance to prevent stagnation of air or damp, — otherwise 
the shelter from cold and unhealthy winds, which is an evil 
recurring only at intervals, will be purchased by loss of 
healthiness at other times. The ground should fall in all 
directions, to facilitate drainage ; it should not be on a steep 
slope, for high ground rising near a building stagnates the air 
just as a wall stagnates it : the natural drainage outlets 
should be sufficient and available. There should be nothing 
to prevent a perfectly free circulation of air over the district ; 
there should be no nuisances, damp ravines, muddy creeks or 
ditches, undrained or marshy ground close to the site, or in 
such a position that the prevailing winds would blow the 
effluvia over it. 

The site should, moreover, be thoroughly under-drained, 

Healthiness of Site. 29 

except possibly in a case where the ground is so elevated and 
porous as to ensure that water never remains in it ; and if 
there is higher ground adjacent, the water from the higher 
ground should be carefully cut off by underground catch- 
water drains, and led away from the vicinity of the site. 

The object to be attained in laying out the ground is the 
rapid and effectual removal of all water from the buildings 
themselves, and from the ground in their vicinity, so that 
there shall be no stagnation in or near the site. 

It is no doubt impossible always to procure a perfect site 
for building ; but it will be necessary in the construction of 
buildings upon a given site to discount any departure from 
these qualifications by additional sanitary precautions in the 
building — i. e. by increased expenditure. 





However healthy a site may be, evil may accrue from the 
undue crowding of buildings upon it. 

The deteriorating effect of residence in towns has been 
frequently noticed. The Registrar-General has shown that a 
population of 12,89^,98^^ persons living on 3,183,965 acres in 
the districts comprising the chief towns of England, showed 
an average death-rate for ten years of 34.4 per 1000 ; whilst 
a population of 9,819,^84 living on 34,135,^56 acres in 
districts comprising small towns and country parishes, showed 
an average death-rate for a similar period of 19.4 per 1000. 

Dr. Morgan's paper on the deterioration of races in great 
cities shows that of the adult population of London 53 per 
cent., of that of Birmingham 49 per cent., of that of Man- 
chester 50 per cent., and of that of Liverpool 6% per cent., 
were immigrants from the country settled in the town, and 
that the majority of the incomers were men and women in 
the prime of life. 

The mortality in these four towns averaged 26 per 1000 
against 19 per 1000 in the adjacent country districts; the 
mortality of persons under the age of 15 being 40.7 per 1000 
in these towns against 22, per 1000 in the country districts. 

The marriages in the city population were four times as 
numerous as in the agricultural counties, but the births in the 

Arrangements of Divellings on a Given Area, 31 

town population only exceeded those in the agricultural 
population by one-sixth. 

A statistical analysis by Mr. Francis Galton of the details 
of 1000 town families and 1000 country families, selected 
from the town of Coventry and the adjacent agricultural 
population, showed that the town population supplied to 
the next generation only three-quarters of the number of 
adults supplied by the equally numerous country population ; 
and that in two generations the adult grandchildren of artisan 
townsfolk were little more than half as numerous as those of 
labouring people who lived in healthy country districts. In 
large, closely-built centres of population the ratio would 
probably be considerably increased against the town 

The greater unhealthiness of towns is largely due to the 
too close proximity of the dwellings, the consequent absence 
of fresh air, and the saturation of the subsoil with impurities 
passing into it from the closely occupied surface. 

The subsoil of all large Indian cities has been saturated 
with the filth of generations, just as the subsoil of large cities 
in ancient times had been saturated. 

From this saturation these ancient cities became foci of 
disease, and were abandoned. 

Many large cities in India contain saltpetre factories, the 
nitrogen being derived from the previous contamination of 
the soil with decaying animal matter. 

The careful paving of a town area, coupled with adequate 
drainage to carry off the water falling on the gravel surface, 
is a great protection to health. 

It is essential for health that buildings should have free 
circulation of air all round them, and as much sunlight as 

Rows of back-to-back dwellings which do not admit of 
thorough ventilation should not be permitted ; there should 
be a clear space sufficient for a free circulation of air at the 

32 Conditions which regulate the Healthy 

back of every dwelling house, from the level of the lowest 
floor or basement, and along its whole width between it and 
the building behind it. 

In temperate and cold climates buildings should, if possible, 
be arranged so as to allow the sun to shine on both of the 
principal sides of the building during the course of the day. 

Indeed in the general arrangement of buildings or tents or 
huts, it may be laid down as a safe rule that the larger the 
space that is allowed between them for circulation of air the 
more healthy will the occupants be. Hence, if for any reason 
it be necessary to lodge a large number of persons on a given 
limited area, it is preferable for health to place them in 
buildings of several storeys high, designed to afford free 
thorough ventilation, and so distributed on the site as to 
admit of free circulation of air all round each building, rather 
than in dwellings of lesser height which must be in close 
proximity, not admitting of free circulation of air. 

The following are examples of the death-rates in some of 
the most densely peopled districts in the metropolis, as com- 
pared with less dense localities. 

St. Anne's, Soho .... 

Strand . 





Model Lodging Houses, viz. , 
Industrial Lodgings Company* 
Metropolitan Association , . 

* Registrar's District, Strand. 

' „ „ Kensington. 

* Lewisham. 


Number of 


per acre. 






1 140 

Number of 


per acre. 














* The population per acre in one of the buildings of this Company is 2,200. 

Arrangements of Dwellings on a Given Area. 33 

The examples drawn from towns are from places where 
paving and draining have been more or less carried out, and 
where, nevertheless, the influence of surface overcrowding on 
health is obvious on a comparison being made with less 
crowded districts. 

The superior healthiness of the Model Lodging Houses is 
due partly to the careful provision of sanitary arrangements, 
but mainly to the fact that the numerous storeys in these 
buildings, whilst affording accommodation for a dense popula- 
tion on a limited area, are provided with free through venti- 
lation ; in addition to this, ample space is provided all round 
the structure for the circulation of air, and impurities are not 
allowed to be retained on the open area round the buildings. 

With armies, a camp is formed of a number of tents or 
huts with circulation of air all round each. But if the force 
be large, a too close proximity of tents may be, and certainly 
has been, a common cause of camp diseases. A camp is a 
temporary town, without paving or proper drainage. It is 
only by paving and drainage that the deleterious influence of 
surface overcrowding in towns can be reduced to a minimum. 
But paving and drainage cannot be carried out to a sufficient 
extent in camps to enable the surface to be crowded with 
safety to health ; and therefore in laying out a camp the 
extent of space allotted to it should be as large as the 
nature of the ground, or of the service, will admit ; great care 
should be exercised in the selection of places for the deposit 
of refuse, and after a temporary camp has been occupied for 
some time, the site should be abandoned for a new one. 

The Quartermaster-General's instructions for camping, 
issued at the commencement of the Crimean War, authorised 
densities of population on the camp surface equal to 54^ and 
1037 inhabitants per acre. The lowest of these densities is 
nearly double that of St. Anne's, Soho, one of the most 
densely populated districts in England, where the population 
occupies houses of ordinary construction. It includes jiot 


34 Conditions which regulate the Healthy 

only the ground actually covered by tents, but all the open 
spaces in the camp. The ground actually covered by tents in 
these plans of encampment gave a density of population equal 
to 1632^ per acre, or a space of little over 5x5 feet for each 

A comparison of these authorised densities for camps, 
which had neither drainage nor paving, with the dense 
populations in towns already mentioned, affords an index of 
what would be likely to be the influence on health of surface 
overcrowding in the camps. 

The surface area per tent for different densities of popula- 
tion per square mile is as follows : — 

Square yards per tent. 

Tents per acre. 


Troops per acre, assuming 
1 2 men per tent. 








The number of troops to be placed on a given area must 
be determined by local circumstances, but the above table 
will be useful in enabling a correct judgment to be formed 
upon one very important element in the sanitary state of 
camps — namely, density of population. 

The manner of arranging tents is of importance to health, 
as well as to cleanliness. 

Battalion tents should never be arranged in double line ; 
short single lines are best. The tents in line should be 
separated from each other by a space at the very least equal 
to a diameter and a half of a tent, and the farther the lines 
can be conveniently placed from each other the better. 
These are all matters which are necessarily more or less 
subject to military considerations, and therefore the object 
in pointing them out is to furnish a sanitary standard at 
which to aim, rather than to suggest an absolute rule. 

Arrangements of Dwellings on a Given Area. 35 

The construction of permanent buildings is a matter 
over which the architect and the engineer have a more 
complete control than over the location and arrangement 
of a camp. 

The general locality of houses, hospitals, asylums, prisons, 
or barracks, and even towns, is no doubt settled by the special 
circumstances of each case ; or, in the case of government 
buildings, by military or political considerations ; but the 
actual site within such general locality, and the arrangement 
of buildings on the site, is a matter which falls more imme- 
diately within the province of the architect or engineer to 

The health of any building is dependent upon free-moving 
pure air, outside and inside its walls ; anything which interferes 
with this first condition of health is injurious. 

If the building is placed in a town, the health of the 
inmates is governed by the same conditions as those of 
the rest of the population of the town. 

Thus certain barracks in manufacturing towns recently 
showed a death-rate of io«88 per 1000, as compared with 
a death-rate of 6«98 per 1000 at Aldershot. But there is 
some ground for fearing that the continued occupation of 
Aldershot as a camp, on a porous soil, without paving or 
adequate drainage round the huts, is leading to a gradual 
deterioration of the health of the camp. 

Where commercial, political, military, or other necessities 
require that a building containing a large number of inmates 
should be placed in a town, additional precautions must 
be taken to render it as little unhealthy as possible. 
The enclosure should be sufficient to allow of ample space 
being reserved for a supply of fresh air between the enclosure 
wall and the inhabited buildings, and between the buildings 

The sources of impure air within the enclosure, such as 
ash-pits, manure-pits, &c., should be reduced to a mini- 

D % 

36 Healthy Arrangements of Dwellings, &c, 

mum, and so placed that the air in their vicinity shall not 

In the design of any building intended for habitation the 
first consideration is, how can the ground at the disposal 
of the architect or engineer be best utilised, so as to 
secure pure flowing air and sunlight over every part of the 



Having explained the conditions which govern the purity 
of the outer air, it is in the next place necessary to consider 
what is the accepted meaning of purity of air in an inhabited 

In order to appreciate the enormous difference between the 
purity of air out of doors and the purity of air in a confined 
space, it is necessary to consider what are the constituents of 
the outer air. 

Standard of Purity in Air. 

Air taken under the most favourable circumstances, in free 

open spaces or on elevated ground, consists of the following 

constituents (Angus Smith) : — 

I Oxygen, 1^09 \,o %\\ parts ; izo9*6 mean. 
1000 parts \ „ 

I Nitrogen, 789 to 791 parts. 

Moreover, every analysis of air shows the presence in 
varying proportions of carbonic acid, vapour of water, organic 
matter, ammonia, suspended matter. 

The purity or impurity of the air, and its effect on health, 
depends upon the greater or less degree in which these 
various subsidiary matters are present in the air. 

The most important of the gaseous impurities in air which 
have an influence on health is carbonic acid COg. It is, 
however, less important on account of its own special action, 
than because of its use as a measure of the purity of the air. 

38 Purity of Air in 

. The average amount of CO2 has been taken at 0-400 
volumes per 1000 in normal air, although it is not unfre- 
quently as low as -a, and sometimes as high as -5, or more. 

M. Reiset obtained from a year's observation, at a station 
in the country far from dwellings, and situated at about 
four miles from Dieppe, an average of •2942 per 1000. The air 
above a crop of red trefoil in the month of June gave 'izSgS ; 
and at a height of one foot from the soil, in a barley-field 
in July, '2829, per 1000 : the corresponding amounts at the 
country station being •2915 and •3933 per 1000 respectively. 
The presence of 300 sheep near the apparatus raised the 
proportion to •3178 per 1000, and at Paris in May 1873-75-79 
the mean amount was '3027 per 1000. 

The presence of from i«5 to 2 per cent, of this gas pro- 
duces in many persons severe headache ; but as much as 
3 per cent., or even more, has been found just endurable 
under certain circumstances. A candle will be extinguished 
with 2'5 per cent. ; and it may be assumed that the presence 
of 5 per cent, or over will cause death. 

In this connection, it may be mentioned that carbonic 
oxyde COj is eminently poisonous. Less than 0*5 per cent, 
has produced poisonous symptoms, and i per cent, rapidly 
produces fatal results. This gas is formed by the imperfect 
combustion of carbon. 

The deleterious action of gases disengaged from marshes 
has been attributed to the presence of sulphuretted hydrogen, 
but the action of these gases is not very accurately 

There are, moreover, various suspended matters in air 
which produce disease from mechanical causes, such as the 
dust which in Egypt produces a sort of ophthalmia. Bron- 
chitis and lung disease prevail in many factories, arising from 
the inhalations by the workmen of the dust of coal, sand, and 
steel, or of particles of cotton or hemp. Stone masons suffer 
from inhalation of stone-dust. 

an Occupied Building. 39 

The Guards suffered largely about 18 years ago from lung 
disease ; one of the contributing causes was assumed to be 
the quantity of pipeclay they inhaled in the process of 
cleaning their white cloth fatigue jackets. 

House-painters suffer from the dust of white lead ; though 
in this, as in many cases, the persons suffer as much from 
swallowing particles, in consequence of not washing off the dirt 
. from their hands before eating, as from breathing the dust. 

Of all the impurities of air, that which stands highest in 
the scale of injury to health is organic matter. An undue 
proportion of carbonic acid may indeed kill outright, but to 
the presence of organic matter diseases of impure air are 
mainly traceable. 

Malaria appears to arise from the poison of decaying moist 
vegetable matter in marshes and forests. Typhoid fever is 
traceable to the poisonous air which arises from the putrefy- 
ing substances in sewers. In camps men have had typhoid 
fever in consequence of their tents being placed near to 
manure heaps, or on damp soil. Phthisis and other diseases 
may result from breathing air rendered impure by the putre- 
fying organic matter thrown off from the human body in 
the process of breathing and transpiring. 

So long as air is in movement out of doors, the products 
of vegetable and animal waste are being continually removed 
from the air by oxidation. They are also washed out by 
rain, or removed by snow and hail. Much of the oxidation 
is probably due to the action of ozone, and would not be 
^ effected by ordinary or inactive oxygen. 

Ozone is oxygen in an altered or allotropic condition, and 
appears to be formed by the passage of the electric spark 
through dry oxygen or by slow oxidation of phosphorus and 
other essential oils in presence of moisture. Ozone is 
insoluble in water. 

Ozone is rarely, if ever, absent in fine weather from the air 
of the country ; but it is more abundant, on the whole, in the 

40 Purity of Air in 

air of the mountain than of the plain. It is also said to occur 
in larger quantity near to the sea than in inland districts. It 
has been found to an unusual amount after thunderstorms. 

There is great variety of opinion as to the conditions which 
produce ozone. According to some observers, the amount of 
ozone in the air is greater in winter than in summer, and 
greater in spring than in autumn ; but according to other 
observers, it is greater in spring and summer than in autumn 
and winter. Ozone has usually been found more abundantly 
in the air at night than by day; but, again, some careful 
observers have found the reverse of this statement to be true. 

No connection has yet been proved to exist between the 
amount of ozone in the atmosphere and the occurrence of 
epidemic and other forms of disease. 

Ozone is rarely found in the air of large towns, unless in a 
suburb when the wind is blowing from the country ; and it is 
only under the rarest and most exceptional conditions that it 
is found in the air of the largest and best ventilated apart- 
ments. It is, in fact, rapidly destroyed by smoke and other 
impurities which are present in the air of localities where 
large bodies of men have fixed their habitations. 

The permanent absence of ozone from the air of a locality 
may, however, be regarded as a proof that the air is adul- 
terated air. Its absence from the air of towns and of large 
rooms, even in the country, is probably the chief cause of the 
difference which every one feels when he breathes the air of a 
town or of an apartment, however spacious, and afterwards 
inhales the fresh or ozone-containing air of the open country. 

The amount of ozone in the atmosphere is extremely small, 
and an excess of ozone is destructive to life; thus the 
respiration for a very short time of oxygen containing about 
i-240th part of ozone is certainly fatal to all animals ; whilst 
similar animals will live in good health for months after 
respiring oxygen alone for 37 hours, the carbonic acid being 
removed during the experiment. 

an Occupied Building. 41 

The action of breathing and transpiring upon the air is 
as follows : — 

I. The oxygen is diminished. 
%. The. carbonic acid is increased. 

3. A large amount of watery vapour, is added. 

4. There is an evolution of ammonia and organic matter. 

5. A considerable amount of suspended matter is set free, 

consisting of epithelium, and molecular and cellular 
matter, in a more or less active condition of putrefac- 
tion. At the same time, portions of epithelium are 
constantly being given off from the skin, and even 
pus cells from suppurating surfaces ; as, for instance, 
with surgical cases in hospitals. 
The oxygen is, of course, diminished in the direct ratio 
of the consumption of carbon and hydrogen in the system. 
As regards the amount of carbonic acid, a subsistence diet, 
sufficient for the internal work of the body only, is a little 
under 3000 grains of carbon daily, yielding about 13-6 cubic 
feet, or about 0-57 cubic feet per hour of CO2. 

Angus Smith, in his experiments, was unable to find more 
than 0-4 per hour of- COg given off; but the experiments of 
Pettenkofer showed that in a state of repose an adult gave off 
about o«7, and in a state of active work 0*9 to i-o or more. 
The constitution and usual diet of the person experimented 
on no doubt influences the result. 

These numbers correspond pretty closely with theoretical 
calculation, but if the number 0'6 be taken to allow for differ- 
ence of age, weight, and sex, it will be well within the mark 
in the calculation. 

The amount of vapour varies, but taking the amount from 
skin and lungs together, it may be assumed at about 30 oz. 
per diem, or about 550 grains per hour, enough to saturate 
about 90 cubic feet of air at a temperature of 63° Fahrenheit. 
The amount of organic matter has been variously estimated, 
but there are hardly any trustworthy experiments on record. 


Purity of Air in 

As already mentioned, this organic matter is highly poison- 
ous ; and it is as much from the presence of this as from 
carbonic acid in re-breathed air that injury arises. 

The air of towns is rendered impure chiefly from the 
presence of suspended matters. 

The experiments of Dr. Angus Smith show that in towns 
the oxygen is not less than in the country districts, and that 
carbonic acid is not materially in excess. 

See the following table : — 

In Manchester, 


In fog and frost . . . 
Outer circle, not raining 

Suburb, in wet weather 

Per 1000 



\ 209*800 

( 209*600 


Per 1000 vols. 

Streets 0.403 

Where fields begin . . . 0*369 
Streets in fog 0*679 

In London, 


Open places, summer . 
Streets, November . . 




On Thames . 
Parks, open . 
Streets . . 



It is therefore to other impurities that the oppression from 
town air is attributable. For instance — The presence of sul- 
phuric acid in the air is very noteworthy. 

Numerous analyses of various sorts of coal showed that 
whilst there was a mean of i«7 per cent of sulphur in the 
several coals, no more than 0-% per cent remained in the ash. 
Therefore the burning of 1000 tons of coals of this description 
would send 15 tons of sulphur into the air as sulphurous acid ; 
and this soon becomes converted into sulphuric acid ; this is 
sufficient in quantity to render the rain water which is col- 
lected in towns very frequently acid. 

It has been estimated that the coal consumed in Glasgow 
and its vicinity gives off sufficient sulphur acids to amount 
to 300,000 tons of oil of vitriol annually. The quantity of 

an Occupied Building. 43 

coal estimated to be consumed annually in London is about 
5,000,000 tons, which, from this calculation, would send into 
the air 75,000 tons of sulphurous acid. London air contains 
about 19 grains of sulphurous acid in a cubic yard of air. It 
contains moreover, an enormous quantity of soot, fine carbon 
and tarry particles of coal ; of the two latter almost I per 
cent, is given off in combustion. This rarely rises above 600 
feet from the ground. London air also contains much sus- 
pended organic matter, independently of the sewer and other 
emanations, and independently of the ammonia given out 
by the manure of the enormous number of horses kept in 
London. It is noteworthy that the mud from a paved street 
in London was found on analysis to contain nearly 90 per 
cent, of horses' dung ; the mud on the new wood pavements 
consists almost entirely of horse dung^. In addition to such 
matters town air contains vestiges of food, clothing, and 
building materials ; as well as dust from manufactories. 

An important cause of the impurity of air in town houses 
especially arises from the use of coal-gas in rooms. 

The products of the combustion of gas are carbonic acid, 
carbonic oxide, compounds of ammonias, and various com- 
pounds of sulphur, which are injurious to health. 

The products of combustion vary much with the quality of 
the gas and the completeness of the process, but 100 cubic 
feet will unite with from 90 to 164 cubic feet of oxygen, and 
produce iioo cubic feet of carbonic acid, and from %o to 50 
grains of sulphuric acid, so that 100 cubic feet of coal-gas 
consume the oxygen or destroy the vital qualities of 800 cubic 
feet of air, and raise the temperature of 3i'a90 cubic feet of 
air 100° Fahrenheit. 

With imperfect combustion, 67 per cent, of nitrogen, 16 per 
cent, of water, 7 per cent, of carbonic acid, and 5 to 6 per cent. 

^ The wood pavement laid down in Regent Street about 30 years ago was re- 
moved because it had become so saturated with ammonia that the emanations 
tarnished the plate in silversmith^s shops. 

44 Purity of Air in 

of carbonic oxide, with sulphurous acid and ammonia, are 
thrown into the atmosphere, but the quantity of carbonic oxide 
will be materially reduced with more perfect combustion. 

It follows that each cubic foot of gas burnt per hour may 
be assumed upon an average to vitiate as much air as would 
be rendered impure by the respiration of an individual. 

An oil lamp burning 154 grains of oil per hour consumes 
the oxygen of y% cubic feet of air, and produces a little 
more than -5 cubic feet of carbonic acid. And a candle of 
6 to the lb. burns about 170 grains per hour. The com- 
bustion of these does not produce the compounds of sulphur 
which result from the use of coal gas. 

In the open country, the atmospheric currents continually 
disperse the various substances thrown off in breathing. 

The movement of the air is stated in the Registrar 
General's reports to be about 1% miles an hour, on an average, 
or rather more than 1 7 feet per second. It will rarely be 
much below 6 feet per second. 

Imagine a frame about the height and width of a human 
body, measuring about 6 feet by li, or 9 square feet; multi- 
plying this by the velocity of movement of the air at 6 feet 
a second, it will appear that in one second 54 cubic feet, in 
one minute 3^40 cubic feet, in one hour 196,400 cubic feet, of 
air would flow over one person in the open. 

In a room the conditions are very different. In barracks, 
in a temperate climate, 600 cubic feet is the space allotted by 
regulation to each soldier; and when in hospital from 1000 
to iJ5oo cubic feet to each patient. 

If it were desired to supply in a room a volume of fresh air 
comparable with that supplied out of doors, it would be 
necessary to change the air of the room from twice to six 
times in every minute, but this would be a practical impossi- 
bility ; and even if it were possible, it would entail conditions 
very disagreeable to the occupants. 

It is thus evident that when considering the condition of 

an Occupied Building. 


air indoors, it is necessary to seek a standard of admissible 
impurity in the air, rather than a standard of purity of air, 
comparable with that which exists out of doors. 

In judging of the amount of impurity which may be allowed 
in an inhabited air-space, the sense of smell, when carefully 
educated, affords the best indication of the relative purity and 
impurity of different kinds of aih 

The accompanying table obtained from results of experi- 
ments communicated by Dr. de Chaumont to the Royal 
Society shows the conclusions at which he arrived from a very 
large number of observations on the air of barracks and 
hospitals. The method employed in judging of the quality 
of the air was to enter directly from the open air into the 
room in which the air was to be judged, after having been at 
least 15 minutes in the open air. It will be seen how closely 
the state of the room, as detected by the sense of smell, 
agrees with that which would be expected from the carbonic 
acid as shown by analysis. 

Sense of Smell. 



Carbonic Acid per 
1000 volumes. 

In air 





In air 





In room. 


over outer 



A little smell 

Close or disagreeable smell . 

Very close, or offensive and ) 
oppressive smell . , . j 

Extremely close, when the J 
sense of smell can no longer > 
differentiate ) 

















In these experiments. Dr. de Chaumont takes •oooiz of 
carbonic acid per cubic foot as the standard of impurity, in 
addition to -0004 carbonic acid per cubic foot as the normal 
amount of CO2 in the outer air. 

The experiments were made in barracks and in hospitals. 

46 Purity of Air in 

and a result came out from the experiments confirmatory of 
the opinion that, in the case of sick men, more air is required 
to keep the air space pure to the senses than is necessary in 
the case of men in health. It appeared that, in barracks, the 
mean amount of respiratory carbonic acid, when the air was 
pure to the senses, was •196 per 1000 volumes, but in hospitals 
it was only '157 ; or, in other words, whilst in the hospitals the 
air would have smelt somewhat impure when the COg was 
•157, in the barracks with that amount, it was fresh. 

On these grounds it would therefore appear that whilst the 
standard for impurity for healthy persons may be regulated 
by allowing an excess of •ooo!Z per cubic foot of COg over 
that in the outer air, it would be desirable to limit the excess 
in the case of sick to •00015 per cubic foot. 

In addition to the proportion of carbonic acid, and of the 
impurities of which its presence affords a rough test, there are 
conditions of temperature and humidity necessary for good 

Temperature. The dry bulb thermometer in this climate 
ought to read 6'^ F. to 6^ F., and ought not, if possible, to 
fall much below 60° F, 

The wet bulb ought to read 58° F. to 6i° F. That is 
to say in this country the difference between the two ther- 
mometers ought not to be less than 4° F. or more than 
8° F. A greater degree of dryness in the air, provided the 
supply of air be ample, is not however found objectionable. 

In the open air, in healthy weather, it is often 8° or 9° or 
more. The difference is of course increased in hot and dry 

Vapour ought not to exceed 4*7 grains per cubic foot at a 
temperature of S'^^ or 5-0 grains at a temperature of 6^ F. 

The limit of humidity is 75 per cent, or under. 

When the outer air is saturated, as in wet weather, the 
reduction of the humidity in a room will depend on the 
increase of temperature of the air admitted. 

an Occupied Btnlding. 47 

The capacity of the air for moisture increases enormously 
with the temperature, and that which would saturate air at 
50° F. would give only 71 per cent, at 60° F. Thus, at 50° F. 
a cubic foot of air is saturated by 4«i grains; but at 60° F. 
it requires 5-8 grains, so that 4-1 grains would give only 
71 per cent. 

If therefore the outer air is at a temperature of 50°, and if 
the temperature inside the room be maintained at a comfort- 
able standard, say 6^ to 6^"^^ the incoming moisture would 
never cause an excess of humidity. 

In the case of an external atmosphere, saturated at or above 
the temperature within, such as occurs occasionally in hot 
climates, it would be necessary to let in an unlimited quantity 
of air through every possible aperture. 

vl' .^ 



The purity of the air within an inhabited space, enclosed 
on all sides, is necessarily vitiated by the emanations pro- 
ceeding from the bodies of those who inhabit it, and especially 
by the effect on it of their respirations. With persons 
suffering from disease, especially infectious fevers, or from 
wounds, or sores, these emanations are greater in quantity 
and more poisonous in quality, than from persons in health. 
Stagnation in the movement of the air would lead to rapid 
putrefaction of these emanations. 

Vitiated air does not necessarily mix with the whole air of 
the room with rapidity. Any one may satisfy himself of this 
by comparing the upper part of a heated room with the 
lower, or by examining outlets for the escape of air. The 
top of a room will sometimes be found to contain much more 
carbonic acid than the lower part. Therefore under certain 
conditions the law of diffusion does not act so rapidly as to 
prevent an occasional difference between the amount of COg 
in the upper and the lower air of a room ; the organic emana- 
tions diffuse themselves more slowly, and without absolute 
uniformity, and are deposited on the cool walls, ceilings, floor, 
and furniture, where they may be easily collected if desired. 

It would be desirable, if it were practicable, to remove the 
exhalations with such rapidity as to prevent deposition, and 
to permit as little admixture as possible with the air of the 
room ; but this is not practicable. 

and Floor Space. 49 

In considering theoretically the condition of a room in 
which sources of impurity exist, and which is furnished with 
any kind of ventilating arrangements, tha two extreme suppo- 
sitions (both inadmissible) are 

(i) That all exhalations are immediately removed com- 
pletely out of contact of the persons in it, so that the occupants 
of the room are in the same condition as to purity of air as if 
they were out of doors in a brisk wind. 

(2) That the ventilation is so unequal that the spaces 
immediately surrounding the persons do not get ventilated at 
all ; and that the occupants of the room practically live in an 
air which may become saturated with noxious matter. 

The actual state of things must be something between these 
two. And it is probable that the best condition actually 
attainable would approximate, not very closely, but still in 
some tolerable degree, to the ideal condition in which all 
diffusible emanations should be instantaneously and uniformly 
diffused through the whole space. 

Now, supposing this ideal condition to subsist, it is perfectly 
easy to show that the degree of purity of the air would ulti- 
mately depend in no way on the size of the room, but solely 
on these two things, viz. (a) the rate at which emanations 
are produced : (/3) the rate at which fresh air is admitted. 

Demonstration ^. 

Suppose P units of diffusible poison are produced per 

Also suppose A cubic feet of air introduced per hour. 

The same number must necessarily escape per hour. 

The condition of the room having become permanent, the 
quantity of poison escaping is the same per hour as the 
quantity produced (otherwise the condition of the room 
would be changing). 

* By the late Professor Donkin of Oxford. 


50 Cubic Space 

Hence P units of poison escape per hour ; and since this 
quantity is carried away in A cubic feet of air, the escaping 

P . . 

air necessarily contains -j units of poison per cubic foot, what- 

ever be the size of the room. The escaping air may or may 
not be a sample of the average of the room. On the sup- 
position of uniform diffusion^ it is a sample ; hence : — 

On the supposition of uniform diffusion, the air in the room 


ultimately contains -^ units of poison per cubic foot, whatever 

be its size. Thus, on this supposition, the final condition of 
the air depends only on the rate of production of poison, and 
on the rate of admission of fresh air^ and in no way on the 

But if the mode of ventilation be bad, the diffusion will not 
be uniform; and not only so, but there is theoretically no 
limit (except that of saturation) to the quantity of poison 
which may remain as a constant quantity in the room, 
however abundant may be the supply of fresh air. 

It seems hardly conceivable, though it is mathematically 
possible, that the whole quantity of poison remaining per- 
manently in the room could be reduced by any contrivance 
below that of uniform diffusion. 

The condition referred to above, as permanent^ is a state 
which, theoretically, would never be actually attained, but to 
which the actual condition would continually approximate as 
a limit. 

Suppose, as before, that P units of poison are produced in 
the room per hour when it is occupied. Suppose also that 
the fresh air itself contains / units of poison per cubic foot. 
Let c be the number of cubic feet in the room ; and suppose 
that at a given time the room begins to be occupied, and that 
A cubic feet per hour of fresh air are introduced, so that the 
same volume of air per hour also escapes. 

Then, if x be the number of units of poison per cubic foot 

and Floor Space. 51 

in the air of the room at the end of t hours, it can be shown ^ 
(see demonstration below) that on the hypothesis of uniform 

where e is (as usual) the number a«7i8. The numerical 
value of the last term in this expression diminishes rapidly 
as / increases, and will become insensible after a number of 
hours depending on the ratio oi A to c. Thus the quantity 
of poison per cubic foot increases continually from the initial 

^ Demonstration of the formula used above. 

C « content of room in cubic feet. 

A •- number of cubic feet of fresh air introduced per hour. 
P = number of units of poison produced in room per hour. 
p — number of units of poison in a cubic foot of fresh air. 
/ B time (in hours) since beginning of occupation. 
X B number of units of poison per cubic foot in the room at time /. 
During the next instant dt, Adt cubic feet of air are introduced, and the same 
quantity escapes. 

The escaping air contains x units of poison per foot, so that Axdt is the quantity 
of poison which escapes. 

During the same instant, Apdt units are introduced with the fresh air, and Pdt 
units are produced in the room. Hence the whole increase of poison in the room is 


but the increase per cubic foot is dx^ so that the whole increase is cdx ; hence 

cdx = (P + Ap—Ax) dt. 

Integrating this equation, and determining the constant of integration by the 
condition that x = p when / = o, we obtain the expression for x given above. 

The following may be added. 

Suppose a room of c cubic feet, containing initially n' units of poison per cubic 
foot, to be shut up for t hours with a man in it who produces P units per hour. 
At the end of that time, how much fresh air (containing p units per cubic foot) 
must be added to the air of the room in order to reduce the quantity of poison 
per cubic foot to v units? 

It is easily found that the number of cubic feet required is 

c(v'—v) + Pt ^ 
w^p ' 

and this formula shows clearly that if v « v, that is, if the room is to be brought 
back to its initial condition, the quantity required is independent of c, that is, of 
the size of the room. 
This demonstration was furnished by the late Professor Donkin of Oxford. 

£ 2 

52 Cubic Space 

value pi and tends to the final or permanent value / + -2" * 

which it will attain sensibly after a finite number of hours, 
though never rigorously. 

The size of the room then does not affect the permanent 
condition of the air ; but everything else being the same, the 
larger the room is, the longer it will be (after beginning to be 
occupied) before it attains sensibly its final or permanent 
condition of impurity. 

If, in the final condition, the number of units of poison per 

cubic foot be tt, then tt =/ -f — i 

whence A = > 

TT — / 

which gives the number of cubic feet of fresh air per hour 
required to maintain this condition. 

For example ; suppose a man produces 6 units of carbonic 
acid per hour, and fresh air contains •004 such units per cubic 
foot, if it is required to maintain a room (of whatever size), 
constantly occupied by one man, in such a condition that the 
units of carbonic acid in a cubic foot shall never exceed '006, 

then A == — ^ = 3000, 

.006— .004 ^ 

that is, 3000 cubic feet of fresh air must be supplied per hour. 

In this case, at the end of / hours after the room begins to 

be occupied, the number of units of carbonic acid per cubic 

foot is , 3>ooo< 

.000 — 002 X € T"' 

where c is the number of cubic feet of space in the room. 

Thus, suppose the room contains 1000 cubic feet of space, 
then the units of carbonic acid per cubic foot are 

at first *oo4, 

after i hour .... -005900, 

„ a hours .... '005995, 

•I 3 w .... -0059997 ; 

and Floor Space. 53 

so that after two hours the room would have sensibly reached 
the final condition of '006 units per cubic foot. If the room 
contained only 100 cubic feet of space, the approximation to 
the final state would be much more rapid. 

In considering the question of purity of air in an enclosed 
space, it is necessary to take into consideration the sources 
of vapour inside the room. Every man gives off from lungs 
and skin each hour enough to raise the humidity from 70 per 
cent, to complete saturation in 500 cubic feet at 60° F., and 
to raise it to 82 per cent, in 1500 cubic feet. Now to reduce 
this amount to 73 per cent, would take 3000 cubic feet of air 
saturated at 50° F. But the vapour given off by the body is 
not the only source of humidity. Humidity may arise from 
the combustion of lights, or the vapour of liquids used in the 

According to this theoretical assumption of temperature 
and moisture, a room containing an air space of 1000 cubic 
feet, occupied by one individual, would require to be supplied 
with 3000 cubic feet per hour, in order to maintain it in a 
proper condition of purity and humidity. 

Thus, upon the assumption made, the theoretical calcula- 
tions, based first on carbonic acid, and secondly upon humidity, 
lead to similar conclusions in each case. 

In a warm climate the natural changes of temperature, and 
consequent alteration of the conditions of the movement of 
air, differ widely from those in temperate and cold climates. 
In warm climates these figures may be applicable. 

But in our temperate climate, a careful practical examina- 
tion of the condition of barrack-rooms and hospitals, judged 
of by the test of smell, shows that arrangements which appear 
to provide for a volume of air much less in amount than 
that obtained by calculation will keep the room in a fair 

These results have pointed to about 1 200 cubic feet of air 
admitted per hour in barrack'-rooms occupied by persons in 

54 Cubic Space 

health. This need not be set down to errors in calculation or 
in theory. 

There are many data which cannot be brought into the 
theoretical calculation. 

For instance, the carbonic acid disappears in a newly- 
plastered or lime-washed room, and could be recovered from 
the lime, therefore a newly cleaned lime-whited room will 
present different conditions from a long occupied dirty room. 
Quicklime washing destroys fungi in dirty walls, as also does 
sulphurous acid fumigation. Now air has the jsame property, 
especially dry air ; and hence opening windows, turning down 
beds, and all such measures, act directly on the subsequent 
state of the air. Therefore an enormous effect is produced on 
all the elements of the above calculation if the windows of a 
room are kept open for several hours a day, instead of being 

Besides this, the conditions under which the air flows in and 
out of a room are so varied. The walls and ceiling themselves 
allow of a considerable passage of air. Examples of the 
porosity of materials have been already given. The ceiling 
affords a ready instance of porosity ; an old ceiling is black- 
ened where the plaster has nothing over it to check the passage 
of air, whilst under the joists where the air has not passed so 
freely> it is less black. On breaking the plaster, it will be found 
that its blackness has arisen from its having acted like a filter, 
and retained the smoky particles while the air passed through. 

Moreover, the porosity of the walls materially influences the 
moisture, for a porous wall may absorb much moisture ; and 
on this account rooms with walls of polished impervious 
material require much more air to pass through them. In the 
absence of sufficient ventilation, when the walls are colder 
than the air, moisture condenses on the walls. 

Ill-fitting doors and windows allow of the passage of a 
considerable quantity of air. 

In a temperate climate, where the changes of temperature 

and Floor Space. 55 

of the outer air are rapid and considerable, these means of 
producing the outflow of air from and the inflow of air into a 
confined space are in constant operation. A sleeping-room is 
very warm when occupied at night ; a rapid fall of tempera- 
ture occurs outside, and at once a considerable movement of 
air takes place. 

The majority of occupiers of sleeping-rooms in England 
close their windows at night; they also often block up the 
chimney by a register or otherwise, to prevent the blacks 
falling. These rooms have no special inlet or outlet for 
changing the air. In the morning they would no doubt come 
under Dr. de Chaumont's definition of ' very close ' ; and if it 
were not for the continual insensible change of air which 
passes through the walls, and the door and window chinks, 
&c., the occupants would be asphyxiated. A well-built house, 
unprovided with special means for the inflow of fresh air, is 
from the very completeness of construction a real source of 

For these reasons, the form of a building is important, 
especially where rooms have to be occupied by large numbers 
of persons. 

The air, thus insensibly coming in, should be taken from 
pure sources. Thus, barrack-rooms with outside walls are 
better than rooms opening out of a corridor, or on each side 
of a corridor. The air in a corridor becomes, after a time, 
saturated with impurities, and the interchange of air from it 
to the barrack-room becomes in time only an interchange of 
impure air. This is especially noteworthy in hospitals, where 
fresh air is of even more importance. The following Fig. 4 
shows an arrangement of barrack-rooms built little more than 
twenty-five years ago, which illustrates this point ; the outer 
wall is only a quarter of the whole wall-space. 

Fig. 5 shows the form of rooms which have been adopted in 
all recent barracks, in accordance with the principles here laid 
down. In this case the windows afford means for sweeping 


Cubic Space 

the bad air out of the room, so that the occupants shall have 
the opportunity of starting every day with fresh air ; while 
the length of wall exposed to outer air, as compared with the 
inside walls, is as four to one. 

Fig. A- 

These considerations bear essentially upon the construction 

of buildings occupied by lai^e numbers of persons, such as 

barracks, workhouses, schools, or asylums ; but they are 

especially applicable to hospitals, where, as has been already 

Fig. 5- 

shown, the emanations from a given number of sick are more 
perceptible than those from individuals in health j and for 
this reason, in addition to numerous other reasons, the pa- 
vilion form of hospital construction presents advantages over 
other forms. In private dwellings the same conditions of 
occupation do not prevail ; and in these, therefore, considera- 

and Floor Space. §7 

tions of comfort, and convenience of internal arrangement, may- 
be allowed to have more weight in the design than the outer 
form of the building. 

The foregoing observations will have shown that whatever be 
the cubic space, the air may be assumed to attain a permanent 
degree of purity^ or rather impurity, theoretically dependent 
upon the rate at which emanations are produced, and the rate 
at which fresh air is admitted ; and that therefore the same 
supply of air will equally well ventilate any space, but the 
larger the cubic space, the longer it will be before the air in it 
attains its permanent condition of impurity. Moreover, the 
larger the cubic space, the more easily will the supply of fresh 
air be brought in without altering the temperature, and with- 
out causing injurious draughts. 

One of the chief difficulties of ventilation arises from the 
draughts occasioned thereby. Every one approves of ventila- 
tion in theory ; practically no one likes to perceive any move- 
ment of air. 

Large rooms, in addition to the advantage afforded of 
enabling the air to be changed with more comfort to the 
occupants than small rooms, also present the advantage of a 
larger wall-surface, and of more numerous windows, which 
allow of a larger insensible ventilation ; thus larger rooms will 
have, even in proportion to their occupants, an apparently less 
degree of impurity than small rooms. Although the uniform 
diffusion of carbonic acid is comparatively rapid in the air of 
a room, the organic emanations given out do not in practice 
diffuse themselves either rapidly or uniformly. They hang 
about in corners where there are obstructions to the flow of 
air, or near the ceiling, in which case they cool and fall down, 
and mix with the air of the room, thus increasing the im- 
purities in the lower part of the room. Consequently there is 
no advantage in mere height in a room unless combined with 
means for removing heated air from the upper part. Indeed 
a lofty room with a space above the top of the windows or 

58 Cubic Space 

ventilating openings to which air loaded with emanation can 
ascend, remain stagnant, cool, and then fall down, is a positive 

In a room with more than one occupant, it is necessary 
that a certain floor-space should be allotted to each occupant, 
for the purpose of allowing the currents of air to remove the 
emanations from one occupant without interfering with his 
neighbour, and to prevent the inconvenience of too close 
juxtaposition. The cubic space for soldiers in barrack-rooms 
occupied by day and night was fixed by the Royal Com- 
mission of 1 858 at 600 cubic feet per man ; if we assume a 
room to be i:? feet high, that allows 50 superficial feet per 
occupant, and if the room be 20 feet wide, with beds on each 
side, the width across each bed, that is the linear bed-space, 
would be 5 feet. 

In a room 10 feet high and 20 feet wide, a width of 6 feet 
of linear bed-space would be afforded ; but with rooms higher 
than 12 feet, it would be unadvisable to diminish the floor-space ; 
thus the floor-space necessarily to some extent governs cubic 
space. In warm climates a larger cubic space is given, mainly 
with the object of obtaining a larger floor-space. In some 
cases as much as 80 feet of floor-space per occupant has 
been given in barracks, dependent on local conditions of the 
healthiness of the site, and of the plan of the buildings. 

The Royal Commission on Cubic Space in Workhouses 
considered where so many persons have to be lodged at the 
expense of the ratepayers, it was necessary to exercise the 
most rigid economy of space, and to supplement the deficiency 
of space by the strictest attention to ventilation and warming ; 
and they reported, that for dormitories in workhouses occupied 
only at night by occupants in health, 300 cubic feet would be 
sufficient, provided the wards did not contain more than two 
rows of beds, and that the height, if above 12 feet, was not 
reckoned in the calculation. This would allow a minimum 
floor-space of 25 feet per occupant, or with dormitories 17 feet 

and Floor Space. 59 

wide, a bed-space of about 3 feet. This allowance is based 
upon the assumption that the most watchful care is bestowed 
on the efficiency of the ventilation. 

In ordinary hospitals the cubic space is practically de- 
pendent on the floor-space, for on this depends the distance of 
the sick from each other, the facility of moving about the sick, 
shifting beds, cleanliness, and other points of nursing. If 
there be a medical school attached to the hospital, the 
question of area has to be considered with reference to 
affording the largest amount of accommodation practicable 
for the teacher and his pupils. 

A ward with windows improperly placed, so as not to give 
sufficient light, or where the beds are so placed that the nurse 
must necessarily obstruct the light in attending to her patients, 
will require a large floor-space, because the bed-space must be 
so arranged, and of such dimensions, as to allow of sufficient 
light falling on the beds. In well-constructed wards with 
opposite windows, the greatest economy of surface area can 
be effected, because the area can be best allotted with re- 
ference both to light and to room for work. 

In a ward %4 feet in width, with a window for every two 
beds, a 7 feet 6 inch bed-space along the walls would probably 
be sufficient for nursing purposes. This would give 90 square 
feet per bed, and there should be as little reduction as possible 
below this amount for average cases of sickness; but this 
space is too small for fever or lying-in wards. 

The practice in regard to area differs considerably in dif- 
ferent hospitals. In the naval hospitals it is about 78 square 
feet per bed. In the Herbert Hospital, where there is no 
medical school, it is 99 square feet per bed. The cubic space 
which results from this, with wards 14 feet high, is 1160 cubic 

In the Royal Victoria Hospital at Netley, where there is a 
medical school, it is 103 square feet. In St. George's Hospital 
it is about 70 square feet. 

6o Cubic Space 

From this minimum, it varies to 138 square feet in Guy's 
Hospital. In the new H6tel Dieu, at Paris, the space per bed 
is from 104 to no square feet, and in the new St. Thomas's 
Hospital it is 112 square feet. This latter area is considered 
sufficient both for nursing and teaching purposes. 

In fever hospitals, and in wards for bad surgical cases, where 
the emanations from the patients are considerable, it is found 
desirable to afford a larger floor-space, varying from 150 to 
200 superficial feet, or occasionally more, according to the 
position and shape of the structure; and this entails an 
enlarged cubic space. 

In cases of fever, if a separate ward is not available, a bed 
should be removed on each side of the fever patient. 

The most recent form for lying-in wards adopted in Paris is 
for each patient to be placed in a small separate room, about 
T a X 14 and 1 1 feet high, opening through a lobby to the open 
air; each room being provided with a small scullery, also 
opening into the lobby. The waste pipe from the sink dis- 
charges with an open end over a trapped gully out of doors, 
and all foul refuse is received in moveable receptacles and 
removed at once. 

These rooms are left vacant, and with open windows and 
doors for a certain time after each occasion of being used. 

On the other hand, in workhouse hospitals, where the 
strictest economy is sought, and where the cases are generally 
of a more chronic character than in ordinary hospitals, the 
Royal Commission on Cubic Space in Workhouses required 
850 cubic feet per inmate, with a minimum of 70 square feet 
of floor-space, and a clear space of six feet across each 
bed, and that no bed should be placed in the middle of the 

In special workhouse hospitals for fever and small-pox 
patients, 3000 cubic feet, and a minimum floor-space of 166 
square feet, is provided. 

For lying-in wards in workhouses a minimum of 1200 

and Floor Space. 6i 

cubic feet and ico square feet is the standard ; but these wards 
in workhouses are seldom continuously occupied. 

In wards partially occupied by day and by night for aged, 
chronic, and infirm cases, with the use of a day-room, 500 
cubic feet and 42 superficial feet were specified as necessary. 

These sizes were adopted upon the condition that the 
ventilation would be adequate and carefully watched. 

It will thus be seen that the question of floor space in a 
hospital ward must be settled with reference to the existence 
or non-existence of a clinical school in the building, and the 
number of pupils likely to follow the medical officer. Excluding 
a medical school, and assuming that the locality of a hospital is 
healthy, the floor-space may be fixed at about 90 square feet 
per bed in this climate, with the understanding that the area 
shall be increased if the building is designed for a medical 
school, or where from unavoidable circumstances an unfavour- 
able site must be selected. 

Dormitories in schools should certainly not afford less floor- 
space than from 50 to 60 square feet. In prison cells, where 
the prisoner is confined continuously, the superficial area per 
occupant should not be less than from 80 to \%o feet, accord- 
ing to the character of the prison and other circumstances. 

In the case of rooms occupied in the day-time, or for a 
portion of the day only, such as day-rooms in hospitals, 
workshops, or schools, a smaller cubic space is sufficient. 
The reason is obvious. In the case of schools, for instance, 
the occupants leave the room empty occasionally, so that the 
air can be periodically renewed. 

Thus the proportion of floor-space and cubic space in any 
room must be regulated to a certain extent with reference to 
its shape and to the conditions of its occupation, as well as to 
its capacity for ventilation. 



Air of the composition before mentioned, viz. aio of 
oxygen to 790 of nitrogen, is a heavy body. At a tem- 
perature of 3^°, and with the barometer at 30 inches, which 
is about the mean sea level, dry air weighs 5^6*85 grains per 
cubic foot. The pressure of the atmosphere on any surface 
is nearly 14*7 lbs. to the square inch ; and a column of air of 
about 87«6 feet in height, under these conditions, will balance 
a column of mercury •! (or one tenth) of an inch in height. 

The molecules of air are but feebly attracted to each other, 
and small increases of temperature, or slight diminutions of 
pressure separate the particles from one another, and thus 
one cubic foot of expanded air weighs less. Similarly, small 
decreases of temperature bring the particles nearer together, 
and make the 'cubic foot of cold air heavier than the standard 
above mentioned. This expansion and contraction are equal 
for equal increments or decrements of temperature. 

This increase of volume amounts to 0-365, or about three- 
eighths of the original bulk, in the process of being heated 
from the freezing to the boiling point of water ; or nearly 
•00203 for every degree of Fahrenheit. 

Thus, if the air inside a room were 20^ Fahrenheit warmer 
than the air outside, the air in the room would be expanded 
to a iZ5t^ part more in bulk, and would to that extent be 
specifically lighter than the outside air. 

This dilatation of air by heat and its contraction by cold 
are expressed by the formula Mi = (i-^at) M 

Movement of Air, 


when M = volume at 3^® and the barometer at 30 inches, 
M-i = volume at the temperature of t degrees above 32^ 
a = co-efficient derived from experiments on the pro- 
portion of the increase of volume of air for each 
degree of elevation of temperature ( = •00203 for 
each degree of Fahrenheit). 
When temperature is decreasing the formula is 

When the temperature of air and the space it occupies 
increases, its density, that is its weight per cubic foot, 
decreases in the ratio expressed in the following formula, 
assuming barometric pressure constant, 

The following table shows the density of air at different 

Weight of air per cubic foot under 30 inches 
pressure of Mercury. 


Dry Air. 

Air saturated with 



























It follows that since warmed air expands and becomes 
lighter, and since cooled air contracts and becomes heavier, 

^ The weight of a cubic foot of air at different temperatures and pressures may be 
found approximately by the formula weight in lbs. » 

1.3253 X Height of Barometer in inches of mercury 
459 + Temperature Fahrenheit 

64 Movement of Air. 

the colder air has a tendency to press upwards the warm and 
lighter air, and to occupy the place of the warmer air. In an 
enclosed space, the rate at which the warmer air will thus 
be pressed upwards by the inrush of colder air, and be forced 
out, will depend upon the form, size, and materials of the 
openings which permit its escape. 

Everywhere the heating and cooling of the air is going on ; 
the sun's rays, the proximity of a warm body, the vicinity 
of a cool shaded surface, all cause changes of temperature, 
and thus create movements or currents in the air. 

In a room, as air is warmed by the bodies of the occupants, 
it ascends ; it comes in contact with the walls of the room or 
with the glass of the windows ; it cools, and falls down. A 
draught experienced near the window, does not necessarily 
show that air is coming in through the window, it may simply 
result from the cooled air which is falling. 

It is on this law of the dilatation of air that all the move- 
ment of air depends, from the winds and hurricanes to the 
ventilation of houses, except where air is propelled by fans or 
by other mechanical appliances. 

It is also noteworthy that the air saturated with vapour is 
lighter than dry air, and air will therefore flow upwards more 
easily in proportion as the ascending column is saturated with 
vapour ; and therefore breathed air which contains moisture 
from the occupants of a room ascends more easily than the 
drier unbreathed air. 

The law which regulates the movements of the air in a con- 
fined space, when the temperature is higher than that of the 
outside air, depends upon the following considerations : — 

I. Upon the difference of temperature of the air inside the 
confined space, as compared with that outside. 

a. Upon the area and other conditions of the apertures 
through which the warmed air can flow out and the cooler 
outside air can flow in to take its place. 

3. Upon the height of the column of ascending warmed air. 

Movement of Air. 65 

If AB represent the height of a column of air of the outside 
temperature Z^, and A C the height of a column of the same 
quantity of air expanded by the warmer temperature /, then 
the velocity at which the warmer air ascends will be that 
which would be acquired by a body falling from C to B. 



That is, V = ^/^g x BC, 
if F = velocity of ascending air in feet per second ; 
H — height of shaft in feet ; 
/ = temperature in shaft ; 
ty = temperature out of doors ; 

a = the coefficient of dilatation of air, which for 1° Fahren- 
heit = •002036 ; and for 1° Centigrade = •003665 ; 


The theoretical equation becomes F= ^^igHa^t^t^, 
Therefore the velocity, and consequently the volume of air 
varies with the square root of the difference between the 
temperature inside and outside the shaft. 

The actual movement of air in a chimney is very different, 
owing to the resistartce from friction to which the moving air 
is subjected. The friction varies directly with the square of 
the velocity of the air-currents, and with the length of the 
channel or flue, and inversely with the diameter or area of 
the flue ; and is, moreover, much influenced by the material 
of which the sides of the flue are constructed. With a sooty 
flue, or a flue with rough sides, the velocity, with equal tem- 
peratures, has been found to be one-half that of a smooth 
clean flue. 

The velocity is, moreover, diminished by the friction caused 
by impediments to the ingress of the fresh air required to 
supply the place of that which flows out; and therefore an 
efficient system of ventilation requires that the extraction of 


66 Movement of Air, 

air should be accompanied by convenient arrangements for 
the supply of fresh air to take its place, and vice versA. 

If these resistances be represented by the constant K^ to be 
determined for each case depending on the (orm and material 
of the shafts and air-channels, and on other considerations, 
such as their freedom from dirt, &c., the formula may be 
generally expressed as follows — 

P^clet, in his treatise on the application of heat, has given 
the following formula to include some of the resistances — 

" D+2gHK ' 
when Z> = diameter of shaft, circular flue, or square root of 
area of rectangular flue ; K = the coefficient of resistance. 

And he determined the coefficient of resistance, correspond- 
ing to this formula, due to pottery chimneys to be •0127 ; sheet- 
iron chimneys to be -005 ; and cast-iron chimneys to be •0025. 
This formula gives rather too high results. Phipson suggested 

"" Z-hi6Z> 
where L = length of evacuation channels. Hurst gives 

^ " n-\^KL ' 

where the dimensions are in feet, and K = -02 for clean glazed 
earthenware flues, -03 wood flues, -06 sooty flues. The diffi- 
culty of obtaining a uniform coefficient for the resistances will 
be made apparent from the fact that in flues of the size of 
ordinary chimneys, soot or accumulations of dust on the sides 
seriously affect the velocity ; General Morin found that the 
presence of a cobweb in a flue almost entirely checked the 
passage of air. The main conditions to be attended to in 
the design of air channels are that they should be as straight 
as possible, with smooth sides, and with such an area as will 
prevent the necessity of maintaining a high velocity in the 

Movement of Air, 67 

The circumstances which affect the flow of air are thus so 
varied that it is preferable, in estimating the amount of air 
removed for purposes of ventilation from buildings already 
constructed, to measure the actual volume of the air in the 
flues or air-passages ; that is to say, to cause it to pass along 
a channel — the size and area of which are known — and then 
to measure the velocity with which the air passes through 
this channel. The multiple of the area into the velocity in 
a given time gives the volume which passes through in that 

There are various ways of measuring the velocity. It may 
be measured by puffs of vapour of turpentine ; by balloons 
filled with hydrogen, and weighted to be of the exact specific 
gravity of air. In ordinary cases, the most convenient 
method is by means of an anemometer. An ordinary form of 
anemometer is that of vanes fixed to a spindle, the revolutions 
of which are recorded by a counter. The vanes are turned 
by the direct action of the current of air, and the number of 
revolutions which are recorded by the counter gives the 
velocity. Of course the value of these revolutions has to be 
ascertained in the first place by direct experiment; that is, 
by forcing a known bulk of air, at a uniform rate, through 
a channel of a given size ; and ascertaining the number of 
revolutions made by the vanes. The most convenient ap- 
paratus for this purpose is a graduated vessel constructed on 
the principle of the ordinary gas-holder, arranged to move 
with a uniform speed, from which a known quantity of air can 
be expelled at will through a channel of convenient dimen- 
sions in connection with it. 

For anemometers of this pattern the formula which must be 
applied to ascertain the velocity takes the following form — 

where V = velocity of air ; 

« = a constant number showing the minimum speed 
of current which will move the vanes, and which, 

F % 

68 Movement of Air. 

sHould be contrived to be as small as possible, 
* by means of light vanes and delicate m.achinery ; 
3 = a constant coefficient ; 
N = number of turns of the spindle in i second. 
Fletcher's Anemometer is another very convenient form for 
measuring the speed of air in heated flues. 

The instrument consists of two parts ; firstly, of two metal 
tubes of about ^rs of an inch internal diameter, open throughout, 
and of any length ; secondly, of a manometer, or pressure- 
gauge. Of these tubes, the end of one is straight and plain, 
while that of the other is bent to a right angle. When in 
use these tubes are placed parallel to each other, and so that 
their ends are exposed to the current of air to be measured. 
They lie at right angles to the current, which thus crosses the 
open end of the one, and blows into the bent end of the 

By this means a partial vacuum is established in the 
straight tube, whilst the pressure of the current forces the air 
into the bent tube ; a differential manometer, attached to the 
outer ends of the tubes, shows the excess of pressure in the 
bent one over that in the straight one. The manometer 
used is a simple U-tube set vertically, containing ether, fitted 
with microscopes and vernier scales, by which the difference 
of level of the surfaces of the ether in the two limbs can 
be measured to j^Vtt of ^"^ inch. This difference of level 
between the columns of ether becomes a measfure of the 
speed of the current passing the ends of the anemometer 
tubes ; the law which governs the speed is expressed generally 
by the formula «/ = v''/x28».55 The corrections to be made 
for small variations of barometric pressure and temperature 

Movement of Air. 69 

are unimportant. The corrections when required are em- 
bodied in the following formula — 

39-92 459+^ 

where p is the height of the column of liquid driven up the 
tube measured in inches, and v is the velocity measured in feet 
per second of air at a temperature of / degrees Fahr., under 
a pressure of h inches of mercury. 

Tables of the velocities corresponding with the readings are 
supplied with the anemometer, and also a table of correction 
for temperature. 

The variations of temperature to which the manometer 
Itself is exposed are not great, being those of the external 
atmosphere only. 

This is a very convenient form of anemometer, because the 
pressure-tubes may be of any length or diameter, and may be 
connected with the manometer by india-rubber tubing of any 
length. Therefore the readings may be observed in any 
convenient place at a distance from the flues. 

The pressure-tubes should project into the air-current to 
the extent of one-sixth part of its diameter, to measure the 
average velocity. It is desirable, after observing the height of 
the ether in the tube, then to reverse the connection with the 
manometer, and to take a second reading; if the smaller 
reading be deducted from the greater, twice the height of 
the column supported by the difference of pressure is ob- 
tained, and the error of observation is halved. But in this, 
as in the anemometer with vanes, there is a difficulty in 
accurately observing low velocities, i. e. under i foot per 

It is worth noting that a sheet of light tracing-paper moved 
through the air at 2 feet per second takes up an angle of 45°, 
and affords a ready means of measuring that velocity; and 
for smaller velocities the angle assumed by the flame of a 


Movement of Air. 

candle affords a fairly accurate index according to the follow- 
ing table. 

Velocity of flow 

of air. 
Feet per second. 

Angle of inclination 

of flame of candle 

with horizon. 











In a shaft open at the top, containing air at a temperature 
above that of the outside air, and in which means exist for 
keeping the incoming air at a similar uniform temperature 
above that of the air outside, the velocity of the upward 
current will vary with the height of the shaft. In order, 
therefore, to effect the extraction of the air in an enclosed 
space with the least expenditure of heat, the shaft should be 
made as high as possible. Any diminution in height or in 
area piust be compensated for by additional heat in the shaft ; 
this means an additional consumption of fuel to keep up the 

Therefore the economical application of the law of dilatation 
of air depends on the height available for the shaft, upon its 
area, and upon its being as free from friction and other resist- 
ances as possible, and upon the difference of temperature 
which can be obtained indoors and out. It will thus to some 
extent depend upon climate : where the outer air is cold, and 
a comparatively high indoor temperature must be maintained 
for comfort, it is most economical ; but where the difference 
between the indoor and outside temperature is small it is 
less economical. Thus General Morin found that whilst in 
winter the ventilation in the Conservatoire des arts et metiers 
by extraction-shafts was continuously maintained with an 
expenditure of i lb of fuel for 8700 cubic feet of air removed, 
in summer i lb of fuel removed only 3000 cubic feet. 

Movement of Air. ji 

Arrangements for the change of air in a confined space 
which depend on difference of temperature are thus the 
simplest form of ventilation. Such arrangements can, of 
course, be applied only where there is a difference between 
the inside temperature and that outside — whether the air 
inside is warmer or cooler than that outside. But where 
fuel must be used solely for obtaining movement of air, it 
can be applied more economically in the propulsion of air 
by mechanical means. 

Propulsion of the air by means of fans or of pumps may 
be used either to force the air into a room, or to extract it 
from a room. 

Theoretically, the propulsion of air into a room would expel 
all the foul air through the cracks of windows and doors, even 
if no special apertures were made for its removal, and the 
existence of pressure in the room would tend to prevent 
draughts of cold air from doors and windows ; but in practice, 
in the ventilation of hospital wards, the system of propulsion, 
i.e. forcing the fresh air into the room, and allowing the 
vitiated air to find its way out, has not been generally found 
successful as a means of purifying the air. The air forced in 
seems to seek the first place of escape, and unless the system 
is combined with an efficient system of extraction, much of 
the vitiated air will remain in corners and dead angles. It is 
therefore advisable always to combine with a system of pro- 
pulsion for the inflowing air some method of extraction of 
vitiated air. Where the circumstances allow of it, it will be 
found simpler to dispense with propulsion, and to rely upon 
the action of extraction-shafts to draw in the air required 
through adequate channels provided for the ingress of fresh 
air. In cases, however, where a large volume of air is required 
to be passed continuously into a confined space, and the 
channels are limited in size, it may be found advisable to 
assist the movement of the inflowing air by propulsion. 

The compression of the air caused by propulsion raises the 

72 Movement of Air. 

temperature of the inflowing air. Some experiments of 
General Morin showed that in a system of ventilation where 
the pumping in of the air gave a pressure equivalent to % 
inches of water, the temperature was raised from 20° to ^5° 
Fahrenheit, between the temperature at the place from which 
the air was drawn into the fan and the temperature at the 
inlet into the room. 

The system of extraction by fans is of the highest value in 
cases where it is desired to remove particular impurities with 
great rapidity ; such, for instance, as in workshops where the 
dust of cotton, steel filings, or injurious emanations produced 
locally are sought to be removed at once. 

The friction of air varies as the square of the velocity multi- 
plied by the pressure against the sides of the passage. This 
pressure being uniform, its total amount depends upon the 
total surface ; that is, the length multiplied by the perimeter 
of the cross section. The force required to propel air through 
any passage is therefore equal to the square of the velocity 
into the total surface multiplied by the coefficient of friction. 
It is more convenient to state the force in lbs. per square inch, 
or per square foot, or as so many inches of water pressure. 

The best form of the formula for practical purposes of 
ventilation seems to be — 

^ KV^PL 
H^—^ ' 


i7= head of pressure in feet of air of same density as the 
flowing air ; 

L = length of the pipe or passage in feet ; 

P = perimeter of cross section in feet ; 

A = area of pipe or passage in square feet ; 

V = velocity taken in thousands of feet per minute ; 

K = coefficient of friction = 0-03 ; but this varies accord- 
ing to the nature of the sides of the passage. 

This formula is perfectly general, and may be used for any 

Movement of Air. 73 

fluid ; H always being the head, stated in feet, of the flowing 

The weight of i foot of air, under the pressure of one 
atmosphere, at 32^° Fahrenheit, is equivalent to 0*0807 lbs. 
per square foot. The weight of i inch of water at 39°* i 
Fahrenheit equals 5'2 lbs. per square foot ; therefore i foot 
of air, at '>/f Fahrenheit, under the pressure of one atmo- 
sphere, exerts a pressure equivalent to 0*0154 inches of water. 

For circular passages, taking D for the diameter, the for- 
mula becomes — 



These formulae are only applicable to passages whose 
diameter is small in proportion to their length. For short 
passages the actual length should be increased, for purposes 
of calculation, by about 50 diameters of the passage : thus the 
formula for short circular passages becomes 

for short irregular-shaped passages, 


It is beyond the limits of this treatise to enter into the 
question of the form of propeller which should be used. It 
may however be observed, that the best fans have produced 
from 70 to 75 per cent, of useful effect in proportion to the 
power employed. 

There is a third system, which can only be applied in special 
cases — ^viz. that of ventilation by means of compressed air, 
adopted in the Mont Cenis and St. Gothard tunnels during 
construction. The compressed air, after having been utilised 
to work the boring machines, escaped into the tunnel, and 
provided fresh air for ventilation. 

Compressed air has also been applied to produce a current, 

74 Movement of Air. 

and thus to extract vitiated air, somewhat on the principle of 
the steam jet which causes the draught in a locomotive 
chimney, but acting by its momentum only. 

H M — the volume ; 

V = the velocity of the air injected into the channel ; 

M^ = the volume of the additional air drawn into the 
channel by the movement of the injected air ; 
M + M^ = J/i=the total volume of air passing along 
the channel ; 

Fi = mean velocity of the air, as it flows along the 
channel ; 

MV^{M ^M^) Fi = M^V^. 
If d = density of injected air ; 
D = density of air in channel ; 
s = section of orifice of injector ; 
6* = section of orifice of channel ; 

M = and M^ = • 

g S 

The system was used for the ventilation of some of the 
galleries of the Exhibition at Paris in 1867, and was subjected 
to experiment by General Morin ; and he shows that the use- 
ful effect varies inversely with the cube of the velocity of the 
air in the injector. 

In a channel having a diameter of about 8 feet, with a jet 
of injection of about 5 inches diameter, the velocity in the 
channel being from 6 to 9 feet per second, and the velocity of 
the jet of injection about %%o feet per second, the useful effect 
of the jet only equalled ^V of the power of the jet ; and as 
the jet only utilised half the useful effect of the engine, the 
general result was to utilise only ^V of the motive power 
expended. Therefore this method of propelling air is not 
advantageous; but it may be resorted to in special cases 
where difficulties exist in applying other methods of pro- 


Movement of Air. 75 

There remains the method of extraction by the movement 
of the atmosphere alone. If an open tube or shaft be carried 
up from a room or enclosed space to a point above, where the 
top is exposed to the free movement of the atmosphere, an 
upward current will prevail in the shaft so long as there is a 
movement in the atmosphere. The movement is of course 
unequal in its action. It is powerful when the wind is high. 
In calm weather it is very small; but in this country, as 
already mentioned, the average velocity of the atmosphere is 
above 17 feet per second, and it is rarely quite at rest. 

It is very difficult to measure the relation which the current 
in a tube or shaft caused by this method of extraction bears 
to the velocity of the wind, because there are so many conflict- 
ing elements to be considered. The formulae for calculating 
the velocity of wind in some of the standard anemometers 
are not entirely satisfactory for the very low velocities. The 
action of wind, whilst it tends to exhaust the air through the 
tube, is, at the same time, acting on all other openings in the 
building, either to exhaust or to force in air. Hence gusts of 
wind will sometimes cause a reverse action in the tube, in 
consequence of some other opening acting temporarily as a 
means of extraction. 

The temperature inside and outside must also be considered. 

If the atmosphere should be without perceptible movement 
in cold weather, when the temperature indoors is maintained 
for comfort above that out of doors, the difference of tempera- 
ture will cause an upward movement in the shaft. In hot 
weather, if the shaft is colder than the outer air, a down 
current may ensue ; but if, in hot weather, there should be 
little or no movement in the shaft, this occurs at a time when 
the windows can be kept open, and the air be renewed by 
this means. 

The friction in the shaft varies inversely with the area ; and 
with small tubes it forms a very perceptible element of re- 
tardation. Experiments made with tubes three inches in 

76 Movement of Air. 

diameter tend to show that the velocity obtained in the tube 
was about \ of that of the wind ; larger diameters, on the 
other hand, produce better results. 

These results were obtained in a place supplied, as far as 
possible, with fresh air to replace that removed, in a manner 
independent of the movement of the atmosphere, and with 
the top of the tube freely exposed on all sides. 

In consequence of the numerous causes of disturbance 
enumerated above, this method of extraction, when applied 
to a house, could not be relied on to act on all occasions with 
certainty as an extraction-shaft. But it can be relied on to 
ensure in one way or other, and to a certain extent, a continual 
change of air. 

A tube or shaft with an open top acts best. It is, however, 
necessary to protect the top, to prevent rain from entering the 
tube ; but a cover tends more or less, according to its shape, 
to delay the current in the tube or shaft. A 1cowl with a 
curved head, arranged to move round with the wind, so as 
always to present its back to the wind, would appear to be 
the form of covered top best adapted to facilitating extraction, 
especially if gradually enlarged at its mouth into an oval 
shape, with an aperture somewhat larger than the tube. 




The efficient ventilatioa of a building depends upon the 
efficient ventilation of each room in that building. The first 
matter for consideration therefore is the manner in which the 
ventilation of a room can be best secured. 

The most efficient way of thoroughly renewing the air of a 
room is by means of open windows and doors, when these are 
so placed as to ensure a thorough draught ; and open windows 
should always be resorted to when the weather permits. 

But the windows cannot always be kept open. In cold 
climates the windows must be closed to keep the rooms warm. 
In hot climates the windows must be kept closed to keep the 
rooms cool. In the one case warmed air, in the other cooled 
air, should be admitted independently of the windows. More- 
over, windows are so placed in a room as to meet the require- 
ments of light, and do not therefore necessarily occupy the 
most advantageous position for the continuous admission of 
air. Therefore every room should have special arrangements 
for the admission and extraction of air. This entails inlets 
and outlets, and in some cases ducts and flues leading thereto. 

The rate and temperature at which air is removed and 
admitted materially affects the comfort of the occupants of a 

78 Preliminary Considerations upon 

The velocity of the air as it flows in and out of a room, as 
measured at the openings for admission or exit, should not 
exceed i foot, or at most % feet, per second ; firstly, in order to 
prevent a sensible draught being felt ; and secondly, because 
a low velocity is favourable to the uniform diffusion of the 
incoming air through the room. 

To avoid friction, it is convenient that the velocity in the 
channels leading to the main extracting shafts should not 
exceed from 3 feet to 4i feet per second, and the velocity 
in the main extracting shafts themselves should not exceed 
from 6 to 7 feet per second. 

This latter velocity will, under general circumstances, and 
where the extraction is effected by means of a heated shaft, 
be provided by a difference of temperature between the inside 
and outside of from 30° to '>^^ Fahr. In special cases, on 
grounds of construction or otherwise, it may be found neces- 
sary to exceed this. 

These velocities would be regulated by the size given to 
the inlets, outlets, supply channels, and extracting shafts, as 
compared with each other respectively, and with the quantity 
of air to be supplied and removed; which quantity would 
depend upon the number of occupants of the rooms to be 
provided for, and the amount of air to be allotted to each ; 
on the conditions required for the ventilation of corridors, 
lobbies, staircases, water-closets, and other subsidiary accom- 
modation ; and on any other conditions necessary to be 
adopted in fixing the supply. The areas thus obtained should 
be the free areas, exclusive of gratings or other impediments. 

Air should be introduced and removed at such parts of the 
room as not to cause a sensible draught. Air flowing against 
the body, at or even somewhat above the temperature of the 
air of a room, will cause an inconvenient draught ; from the 
fact that as it removes the moisture of the body, it causes 
evaporation or a sensation of cold. 

Air should not as a rule be introduced near the floor level. 

the Inflow and Removal of Air. 79 

The openings would be liable to be fouled with sweepings and 
dirt. The air, unless very much above the temperature of the 
air of the room, would produce a sensation of cold to the 

The orifices at which air is admitted should be above the 
level of the heads of persons occupying the room ; the current 
of inflowing air should be directed towards the ceiling, and 
should be as much subdivided as possible by means of 
numerous orifices. 

Air admitted near the ceiling very soon ceases to exist as 
a distinct current ; and will be found at a very short distance 
from the inlet to have mingled with the general mass of the 
air, and to have attained the temperature of the room, partly 
owing to the mass of warmer air in the room with which the 
inflowing current mingles, partly to the action of gravity, 
where the inflowing air is colder than the air in the room ; 
and where there are open fireplaces, the result is acce-» 
lerated by the action of the fire. 

The ventilation of rooms is affected by their situation. In 
a house with a central staircase carried from the ground floor 
to a lantern light in the roof, when the air in it is warmer 
than the outside air, it becomes a powerful shaft ; and being 
of such a large size, tends to draw in the air from adjacent 
rooms, and even down the chimneys, especially when the 
chimney is cold, or when its size is greater than required for 
the removal of the air provided for its supply. In the latter 
case, a double current, one up, one down, and both sluggish, is 
sometimes established in the chimney. 

A smoky chimney will therefore often be cured by being 
reduced in area, so as to reduce the volume of air required to 
fill it, or by an adequate provision of fresh air to the room in 
which it is placed, or by a supply of air in an adjacent hall or 

The effect of a high wind is to diminish the air-pressure in 
a house when all windows are closed on the side of the outer 

8o Preliminary Considerations upon 

air-pressure : hence although some of the chimneys may have 
an accelerated draught due to the velocity of the wind, the 
action of the air-pressure may be to cause other of the 
chimneys to smoke, the remedy for which would be to open 
slightly the windows on the side of the air-pressure, so as to allow 
the pressure to act equally through the house. It will fre- 
quently happen that the open fires in the rooms which require 
to be fed with fresh air, and the warm air shaft of the central 
staircase, will combine to draw in the air from every available 
opening. They will draw up the air from the basement ; they 
will draw it, if they can, from the water-closet and sink-pipes, 
unless the water-closets and sinks are in projecting buildings 
cut off by lobbies ventilated from the open air. 

For these reasons it is of importance not only to shut off 
the staircase from the basement, but to provide fresh air to 
the staircase ; and when open fires are used, to supply each 
room in which there is an open fire-place with its own supply 
of fresh air. If the temperature of the room is to be kept 
up to a pleasant heat, this must to some extent be warmed 

In a building with halls and passages, the inflow of cold air 
through the doors when opened must be guarded against by 
the independent supply of fresh air to, and the extraction of 
vitiated air from, all parts of the building, and by the main- 
tenance of a proper temperature in all parts. By this means 
draughts occasioned from opening doors would be avoided. 
An inflow of cold air occurs only where the ventilation and 
warming are defectively arranged. 

The fresh air should be obtained from places where there 
are no adjacent sources of impurity ; especially not from the 
vicinity of ash-pits, manure-heaps, gully-gratings, or other 
sources of foul air. The inlets from the outer air should be at 
least two feet from the ground, and the surface near should be 
paved and sloped away from the inlet, so as to carry off wet 
rapidly. It has been estimated that the impurities of town 

the Inflow and Removul of Air. $>\ 

air are very much diminished at ioo yards height, And ife not 
found above 600 yards in height — a London fog will rarely be 
perceptible at a height of above 100 yards— coniSequently to 
bring pure air into a large town the simplest way would be to 
draw it down from a height by means of a high shaft or tower. 
The Houses of Parliament could be supplied with very much 
purer air than is now provided for them by bringing the air 
from the top of the Victoria Tower. 

The temperature at which air should be delivered in a room 
depends somewhat on the temperature of the outer ain With 
a temperature out of doors of 86\ air delivered at 60^ would 
be felt as too cold. Similarly, in the arctic expedition, with 
an out-of-doors temperature —40° below zero, a temperature 
of -f 40° was felt to be hot. 

In this climate the temperature indoors should be main- 
tained at from 58^ to 66°, and the warmed inflowing air 
should be supplied at a temperature a few degrees above this, 
but as nearly as possible not to exceed from 68° to 75°, or at 
most 80°. The hygrometric condition of the air must also be 
considered. If the air is too dry, it may, after it has been 
warmed, be passed over vessels containing water, to allow it 
to take up additional moisture — but the natural dampness of 
an English climate renders this less necessary here than it 
may be elsewhere. It has not unfrequently happened that 
complaints of the oppressiveness of dry air artificially warmed 
have been caused rather by the insufficiency of the supply of 
warmed air than from its dry condition. 

The channels for the admission of air should, as a rule, be 
short, direct, and accessible. 

Long channels collect dirt, and form a refuge for insects. 
But if it is necessary to make them long, they should be easily 
accessible for cleaning. Deep underground channels and 
receptacles will modify materially the temperature of the 
inflowing air, because in winter the temperature of the earth 
within a short distance of the surface is warmer, and in 



Preliminary Considerations upon 

summer it is cooler than that of the air ; channels for the 
purpose hi utilising this effect of temperature should be of a 
rectangular shape, so as to afford a lai^e surface in propor- 
tion to their area ; and they should be of a good conducting 

Underground channels should always be impervious to 
ground air ; and, as has been already shown, there are very 
few materials impervious to air. 

The liability of air in underground channels to mix with 
ground air would be diminished if the air were supplied to the 
channels or receptacles by propulsion, and retained in them 
under some pressure. 

Undei^round channels should be perfectly dry. Damp 
channels overcharge the air with moisture ; and thus they 
interfere with the warming of the air ; they induce the pre- 
sence of animal and v^etable life, which dies and decays, 
and renders the air impure. 

In towns where the atmosphere is fuU of particles of soot 
and other impurities, the inflowing air deposits these particles, 

and rapidly blackens any surfaces it impinges on. The im- 
purities may be removed by passing the air through a filter 
made of cotton wool laid lightly, to a thickness of about half 
an inch, on a copper wire frame. (See Figs. 7 and 8.) 

The cotton wool must be renewed at intervals, dependent on 

ike Inflow and Removal of Air. 83 

the state of the atmosphere. In London fogs it should not be 
left on for more than a few days. In clear weather in London 
it will last two or three weeks. 

Sliced sponge acts equally as an air filter, and may be easily 
washed, dried and replaced. 

A system has been introduced of endeavouring to remove 
impurities from the inflowing air by causing the current to 
come in contact with a surface of water. In damp weather 
this is liable to the inconvenience that the water keeps up the 
saturation of the air with moisture. In dry hot weather the 
system which has recently been adopted for washing smoke 
by passing it through a chamber filled with spray, might be 
adapted for washing the impure air of towns. 

Extraction-shafts, when for ventilation only, should be 
placed so as not to occasion unpleasant draughts ; but their 
position must depend also upon other conditions, wkich will be 
further alluded to ; meanwhile it may be observed that it is 
advisable that as far as possible extraction-shafts from dif- 
ferent rooms, the operation of which depends on temperature, 
should be independent of each other. Where the rate of 
extraction is sluggish, and the temperature is low in the shaft, 
there may arise conditions in one room whicli may determine 
a reverse current, and then, when there is not a complete 
separation, the bad air from one room may be introduced into 
another room. And even when the flow of air into an 
extraction-shaft has been rapid througli the outlets from all 
the rooms communicating with it, smells from one room 
have been diffused into other rooms along the line of the 

Extraction-flues in which the temperature exceeds that of 
the outer air should be arranged whenever possible to con- 
duct the air upwards, so as to assist its flow. Every descent 
in the flue has to be compensated for in some way, either by 
extra height of shaft, temperature, or expenditure of force. 

Thus in a chimney-flue, where the extraction depends upon 

G % 

84 . Inflow and Removal of Air^ 

the heat in the flue if a portion of the flue near the fire is 
horizontal, it is found in practice that the flue cannot be 
depended on to act efficiently unless the vertical height is 
double the horizontal length. 

The best form for an extraction-flue is the circular form, 
because it affbrds a maximum of area with a minimum of 
perimeter, or surface, and therefore causes a minimum of 

The various considerations enumerated above apply equally 
whatever be the manner in which the air is admitted or 

It should, however, be accepted as an axiom, that by the 
best ventilating arrangements we can only obtain in our rooms 
air of a certain standard of impurity ; and that ventilating 
arrangements are only makeshifts to assist in remedying the 
evils to which we are exposed from the necessity of obtaining 
an atmosphere in our houses different in temperature from 
that of the outer air. To obtain that temperature we sacrifice 
purity of air ; therefore, whenever the outer temperature 
permits it, windows should be widely opened, so as to replace 
the air of the room by fresh air, as often as possible. 



The simplest way of obtaining a change of air in a room 
is to take advantage of the movement in the air produced by 
a change of temperature, or by the action of the winds. 

Wherever there is an outlet and an inlet for air in a room, 
this system will operate. The open fireplace is one example 
of it ; the sun burner is another example, but the system is 
also applied in every room in which there is an opening at 
the upper part, out of which the warmed air can pass, and an 
opening either level with it or below it, through which fresh 
air can flow in. 

Thus an ordinary sash window is the simplest example. 
If the top sash is lowered and the bottom sash raised, the 
warmed air passes out of the room at the top, and the cooler 
outer air flows in below. Hence, for an inlet for air to an 
ordinary room, provided with a fireplace, but unprovided 
with special inlets, a very simple plan is to cut a slit above 
the lower bar of the upper sash of a window, so as to leave a 
clear space of about a quarter of an inch along its whole length, 
through which the fresh air will be drawn in in an upward 

Ventilators which act similarly to direct the current of air 
towards the ceiling are sometimes introduced into the upper 
panes of window-frames, such as hopper ventilators, or 


Simple Ventilation 

Moore's louvred panes, but these are all makeshifts. Every 
room should have special inlets and outlets for air, arranged 
so as to be independent of the windows ; although the win- 
dows, when they can be kept open, are the best means for 
the renewal of the air in a room. 

The relative position and arrangement of the inlets and 
outlets regulate both the comfort and efficiency of the venti- 

It is therefore necessary to consider what currents are 
caused in a room by ventilating openings. 

In the first place, the question should be considered apart 
from an open fireplace, and apart from the question of 
warming the inflowing air. 

If a room has two outer walls on opposite sides, and if an 
opening be made in each wall, and if the wind blows against 
one of the walls, there will be an increase of pressure against 
that wall, and a diminution of pressure against the wall 
opposite ; consequently air will be forced in through the inlet 
on the side against which the wind blows, and be extracted 
on the other side. 

Hollow ventilating beam. 

I " n II n H II II 11 H I 


\ ^ 

Fig. 9. 

Fig. 10. 

Barrack-rooms have been occasionally ventilated by hollow 
beams, carried across the rooms from one outside wall to the 
other, communicating with the open air at both ends, and 

without Warmed Air. 87 

also provided with openings into the room, but having a 
wooden partition placed across in the centre of the beam, so 
as to compel it to act both as an inlet and an outlet when the 
wind is blowing against either outer wall, as in Fig, 9. 

The action becomes more efficient if the beam be dispensed 
with, and the openings in the opposite walls retained. The 
most convenient form for such openings is the Sherringham 
ventilator; Figs. 10, 11. 

Fig. I 

Tills consists of an iron air-brick or box inserted close to 
the ceiling of the room, and affording a direct communication 
with the external air. In order to prevent the air from 
coming in by stray currents, there is placed at the mouth of 
the opening within the room a hopper-shaped valve, hinged 
at its lower side and opening towards the ceiling ; the result 
of which arrangement is, that the inflowing current is thrown 
up towards the ceiling, and difTused to a greater or less extent 
in the general mass of air within the room. 

This ventilator may, under certain conditions, act as an 
outlet ; but when the room is shut up, it would, especially 
with a fire in the grate, act as an inlet for fresh air. 

Considered as an inlet, its principle and position are both 
good, but acting by itself It is not a sufficient ventilator for 
rooms with a large number of occupants. 

Inlets have been formed by vertical tubes, the opening to. 


Simple Ventilation 





the outer- air being made near the floor, and the tube being 
employed as a means of carrying the point at which the air is 
allowed to enter the room to a height of 5 or 6 feet or more 
above the floor. This is convenient in cases where necessities 
of construction make it desirable to place the opening to the 
outer air low down as compared with the point of entrance of 
the air into the room. The tube has the advantage of directing 
the inflowing current upwards towards the ceiling — but in con- 
sequence of the friction of the sides of the tube the velocity 
with which the air enters the room is less than it would be if 

it came in through a shorter channel 
and a more direct inlet inclined up- 
wards like a Sherringham ventilator. 
The main objection to these tubes 
is that they form very convenient 
receptacles for dirt, insects, cobwebs, 
and dust, which after a time may 
injuriously afi*ect the air passing 
through them. Moreover, inlets of 
this shape do not readily lend them- 
selves to act the part of outlets when 
occasion requires, which is so con- 
venient a feature of the Sherringham ventilator. Upon 
the whole Sherringham's is the most convenient form, and it 
IS easily cleaned. 

Where a room has two outside walls, and is provided with 
openings on both sides, this inflow and outflow of air is 
almost certain to go on continuously, in consequence of the 
movement of the outer air. 

Where rooms have only one outer wall, other conditions 
prevail. The Sherringham ventilator, in such cases, would 
be found to act on occasions both as an outlet and an inlet ; 
but its action would not be sufficiently energetic, consequently 
an additional outlet becomes necessary. This is best pro- 
vided by means of a vertical shaft or tube carried from near 


Fig. i». 


without Warmed Air. 89^ 

the ceiling to above the roof. In a room possessing a 
chimney-shaft, and with few occupants, an Arnott's ventilator 
is convenient; its main use being to assist ventilation on 
those occasions where the fire is low or not lighted. The 
merits which Dr. Arnott's chimney-valve possesses have led 
to its extensive introduction into barracks, hospitals, private 
houses, &c. It consists of an oblong metal frame inserted 
into the room chimney near the ceiling. Its object is to take 
advantage of the upward draught of the chimney in drawing 
the upper strata of the air of the room through the frame into 
the flue, while to prevent down-draughts of smoke into the 
room, a light silk or talc flap-valve, supported behind a 
perforated metal plate, is placed in the opening of the box 
into the room. This valve, like every other, requires certain 
conditions for its action. If the throat of the chimney be 
very wide, the quantity of air and smoke which pass up the 
shaft from below will be more than the chimney can accom- 
modate at its narrowest part, where the ventilator is placed, 
and smoke will consequently pass through the valve into the 
room. Wherever, therefore, Arnott's valve is to be used, the 
throat of the chimney between the fireplace and the valve 
must be contracted to such an extent as to leave a balance in 
the draught to be supplied by air passing through the valve. 
As, however, the amount of this balance — in other words, 
the number of cubic feet of air which can pass through the 
valve into the chimney per hour — is very limited, this form of 
ventilator is not adapted to be the only outlet for a barrack- 
room, dr for any room with several people in it. 

Under such circumstances, as well as in rooms where this 
is not available, an outlet should be provided by • a shaft 
carried from near the ceiling to above the roof. (Fig. 13.) 

The velocity in these shafts, when acting freely of them- 
selves, is dependent of course on the diff'erence of temperature 
between the air in the room and the air without, on the 
amount of movement in the outer atmosphere, on the adequate 


Simple Ventilatio7i 

Fig. 13. 

supply of fresh air in place of that removed, and on other cir- 
cumstances. When the temperature is nearly equal, as, for 

instance, when the windows are 
open, there is very little upward 
draught, except as the result of 
movements in the atmosphere with- 
out, but when the windows are 
open the room is being ventilated 
without the shaft. At other times a 
current varying from 2 J to 5 feet 
per second may be relied on, and 
often a stronger current will be found 
to prevail. 

It sometimes happens, especially 

in rooms with a very short shaft, 

as, for instance, rooms near the roof, 

that from the action of the wind 

the current becomes irregular, and is 

occasionally reversed, and produces down-draughts — so that 

the shaft becomes temporarily an inlet, and the Sherringham 

ventilator an outlet 

A reverse current in the shaft may be occasioned also if 
there is a large open fire in the room inadequately supplied 
with fresh air; or if a large and lofty hall or staircase is in 
proximity to the room — these may act like large shafts to 
draw in the air from the room and down the ventilating 

Such an effect equally provides the room with fresh air; 
but to prevent inconvenience to the occupants, it is advisable 
to terminate the shaft a little below the ceiling with a solid 
bottom and with side openings covered by inverted louvres, so 
that any down current would be directed towards the ceiling. 

The sizes which were adopted for outlets and inlets of 
this description in the case of barrack-rooms are, for shafts 
or tubes in rooms next the roof, a sectional area of one inch 

without Warmed Air, 91 

to every 50 cubic feet of room-space; for the floors next 
below the upper floor a sectional area of one inch to 55 cubic 
feet of room-space ; and, where the barrack consists of three 
floors, for the lower floors a sectional area of one inch to 
60 cubic feet of room-space. For inlets one square inch for 
every 60 cubic feet of content of the room as the clear inlet 
area exclusive of gratings. This is the opening to the outer 
air — the opening into the room should afford an area larger 
than this, so as to reduce the velocity of the inflowing current. 
As the velocity of outflow and consequent inflow depends 
on temperature, provision must be made for closing either 
wholly or partially the inlets in cold weather; and it is 
assumed that in very mild weather the sluggishness of the 
ventilation would be assisted by opening the windows. In 
barrack-rooms, whatever be the weather, the windows are 
opened during a certain time every day. These sizes were 
adopted on the basis; of the cubic space allotted being 600 
cubic feet per man, and that a volume of fresh air not less 
than 1200 cubic feet per occupant per hour should be 
admitted. The contamination of the air varies with the 
number of occupants. In hospital wards, in which the volume 
of air to be removed per occupant is greater than in barracks, 
the same sizes were still adopted ; because in hospitals, the 
cubic space per occupant being much greater, the change of 
air per occupant would be correspondingly greater. But in 
places where the cubic space is much less than this, such as 
in some schools, workhouse dormitories or in common lodging- 
houses, a larger proportional area to cubic space may be 
found advisable. 

There are forms of shafts which have been devised to act 
both as inlets and outlets, but in order to do so they require 
fixed conditions. Alter these fixed conditions and any of 
them may become wholly outlet or wholly inlet. The con- 
dition essential to their operation is, that the room to which 
they are applied be closed, and in a closed room their action 


Simple Ventilation 

is singular^ If a number of people be crowded into a room 
with the fireplace closed and the doors and windows shut, and 
if a tube of an apparently sufficient area to afford ventilation 
for the inmates be carried from the ceiling of the room above 
the roof of the building, there will be an irregular effort at 
effecting an interchange between the air of the room and the 
outside air. The outer air will descend, and the inner air will 
ascend, in fitful, variable, irregular currents, and the room will 
be badly ventilated, if ventilated at all. 




Fig. 14. 

l^ig- 15- 

But singularly enough, no sooner is the tube divided longi- 
tudinally from top to bottom by means of any division, 
however thin, than its action becomes immediately changed — 
a current of air descends into the room continuously on one 
side of the partition, and a current of foul air ascends from the 
room continuously on the other side of the partition. One 
half of the tube supplies fresh air to the inmates of the room, 
and the other half removes foul air, so that if the size be 
properly adjusted the air in the room is kept sweet. (Fig. 14.) 

Watson's Ventilator (Fig. 15) supplies this principle in its 
elenuentary form. It consists of a square tube with a division 

without Warmed Air. 


down the centre, one side affording a tube slightly higher than 
the other : it has no means of diffusing the descending current. 
Mackinnel's ventilator (Fig. i6) professes to be an improved 
application of the same principle. It consists of two tubes, 
one within the otherj leaving a space between them. The 
inner tube is the longer, and projects above the outer tube at 
its upper end ; .the inner tube also projects a little below the 






Fig. 1 6. Fig. 17. 

opening of the outer tube in the ceiling, to give support to a 
circular flange projecting parallel with the ceiling, and con- 
cealing the opening of the outer tube. The action of this 
contrivance is as follows: — The greater length of the inner 
tube determines the upward current to take place in it ; it 
therefore becomes the foul air shaft. The outer tube becomes 
the fresh air inlet, and the descending current striking against 
the flange, is thrown out in the plane of the ceiling, and so 

Muir's ventilator (Fig. 17) consists of a square tube, like 

94 Simple Ventilation without Warmed Air. 

Watson's, divided into four parts, A^ A, B, By by partitions, in- 
serted diagonally. These partitions are carried above the top 
of the tube, and the box is completed outside and above the 
roof by louvres instead of solid sides. The object of this 
arrangement of divisions and louvres is to secure not only up- 
ward and downward currents at ordinary times, but to take 
advantage of any movement of the external air, light, winds, 
&c., which, by striking through the louvres on any angle, would 
cause a stream of air to be projected down into the room, and 
would assist the extraction of the air on the side away from 
the wind. 

All these ventilators act as desired in a closed room, but as 
soon as a door or window is opened they become simply 
upcast shafts ; they cease to supply air, and the air supply 
comes in from the other openings. Again, if there be a fire- 
place in the room, with a strong fire in it, and the doors and 
windows shut, the fire will supply itself from these ventilators, 
and they will become inlets. 

It is obvious that these plans possess certain advantages in 
cases where they are applicable. In single rooms standing 
apart, such as churches, chapels, schools, libraries, &c., warmed 
by stoves or hot water pipes, and where the doors are kept shut 
for hours at a time, any of them will answer as ventilators. 

In stables, also, of a certain construction, they will be more 
or less applicable. 

In the case of living-rooms in houses they would not be 
applicable, both on account of the difficulty and cost of intro- 
ducing the apparatus into a number of detached rooms on dif- 
ferent floors, and on account of the existence of open fireplaces. 

Inlets and outlets may be varied in form, but those men- 
tioned are amongst the simplest forms. It must be borne in 
mind that so long as efficient action without draughts can be 
obtained, the best form to adopt is that which is simplest and 
most easily cleaned. 



It will have been apparent in the preceding chapters that 
the question of ventilation cannot be separated from that of 
the temperature of air. In a cold climate the air required to 
replace vitiated air must be warmed. In a hot climate cool 
air must be supplied. The question must be considered in 
the aspects of health, comfort, and economy. 

Comfort and health are practically synonymous ; and for 
these purposes in this climate the day-temperature of a room 
should be maintained at something between 58° and 66°. The 
night-temperature should not be so high, but it is not desirable 
that the night-temperature should fall below 40°. It should, 
however, be noted, that with a high temperature the quantity 
of oxygen present in the air is diminished. Thus, a cubic foot 
of dry air at 3^° weighs 566*85 grains, and if the proportion of 
nitrogen and oxygen be assumed to be by weight 77 and 23 
per cent., and the slight amount of carbonic acid be neglected, 
there will be in a cubic foot — 

436*475 grains of nitrogen. 
130-375 » oxygen. 

As a man draws, on an average, when tranquil, 16.6 cubic 
feet per hour into his lungs, he will thus receive 130-375 x i6-6 
= 2164-2 grains of oxygen per hour. 


Observations ofi Warming. 

At a temperature of 80° the foot of air weighs 5 1^*3^ grains, 
and is made up by weight of — 

397.61 grains of nitrogen, 
118.77 „ oxygen, 


Therefore, in an hour, if a man withdraws 16-6 cubic feet, 
he will receive 118.77x16.6=1971.6 grains of oxygen per 
hour. Or, in other words, in an hour he would receive 
192.6 grains of oxygen less with the higher temperature; 
that is to say, he would inhale an amount of oxygen equal 
to about 90 per cent, of the amount he would breathe at the 
lower temperature. 

If saturated with moisture, a cubic foot of air will contain 
130 grains of oxygen at 32°, and iia grains at 100°. 

The application of heat in an economical manner requires a 
consideration of the conditions under which heat is generated 
and diffused. 

The standard unit of heat is the amount required to raise 
one pound avoirdupois of water at 32° one degree Fahrenheit. 
The specific heat of a body is the number of units of heat 
necessary to raise that body 1° Fahrenheit. The number of 
units of heat necessary to raise the temperature of one pound 
avoirdupois of the following bodies 1° Fahrenheit is — 


Unit of heat, 

constant volume . . .169 

constant pressure . . .238 

Water at 32° loo 

Pine- wood -65 

Oak .57 

Ice -50 

Olive oil .31 

Unit of heat. 

Marble and Chalk .... •21 

Glass , •19 

Cast iron .13 

Wrought iron -ii 

Copper .09 

Lead .03 

It takes 966*1 British units of heat to evaporate ilb. of 
water under one atmosphere. 

In order to find the quantity of heat required to produce a 
given rise of temperature in a given weight of a given sub- 

Observations on Warming. 97 

stance, it is necessary to multiply together the rise of temper- 
ature, the weight, and the specific heat of the substance. 

But the specific heat of bodies is somewhat greater at high 
temperatures than at low temperatures. Thus if the specific 
heat of water at 32° be taken as i, its specific heat at 446° 
Fahrenheit will be i«o5. i lb. of air at 31^° contains i(Z-38 cubic 
feet, and i cubic foot of air at 32° weighs 'oS lbs. At 6:z° its 
volume would be increased to 1-061, and its weight would be 
•076 lbs., whilst at 104° the volume would be i«i46 and the 
weight -07 lbs., and at 212° the volume would be 1*365 and the 
weight -059 lbs. All gases, dry air, and vapours out of contact 
with their generating fluids, expand very nearly alike, and the 
following formula is a fair expression of the rule of expansion — 

458.4+ r 

in which M=z volume of gas or air, at temperature J", and m = 

volume of air at new temperature t. 

The rate of expansion increases slightly with the pressure. 

When the pressure is not constant, the volume of any gas 

varies in the inverse ratio of the pressure — the temperature 

remaining nearly constant. The pressure is the total pressure 

above a vacuum. Thus one cubic foot of air has the pressure 

of the atmosphere of 14.7, or say about 15 lbs., per square inch 

upon it to begin with, and thus its volume at an additional 

I X 15 
pressure P above atmospheric pressure would be — p • 

Where there is a change both in temperature and pressure, 

and P = pressure corresponding to temperature T and to 

volume Jf, and p = pressure corresponding to temperature t 

and volume m^ the rule becomes — 

-_ P 458-4+/ 
m=Mx — X *^ 

/ 458-4+^ 
When water is present the rule becomes 



Observations on Warming. 

in which J/, P^ 7", and F are the volume, pressure, tempera- 
ture, and elastic force of vapour corresponding to each other 
in one case, and w, /, /, / in the other case. 

The following table shows approximately the cold produced 
by dilation and the heat produced by compression of air. The 
volume at atmospheric pressure at the sea level and at 60® 
Fahrenheit being i«o. 


No. of Inches 
of Mercury. 

Volume of 
the Air. 

Actual Temperature 

of the Air 

during the process. 


Difference due 
to compression 

or expansion. 





- 36° 





+ 33° 

- »7° 




+ 60° 





+ 94° 

+ 34° 




+ 124° 

+ 64' 





+ 115° 

Thus if air at a temperature of 60** be compressed to half 
its volume, or to a pressure of two atmospheres, its temperature 
is raised to 175° Fahrenheit. If, on the other hand, it be 
expanded so as to occupy double the space, the temperature 
would be decreased to nearly 36° Fahrenheit below zero ; and 
hence if air be compressed, and if the compressed air be 
allowed to cool down to the temperature of normal air, and 
the air be then allowed to expand, a degree of cold will be pro- 
duced equal to the difference between that caused by the com- 
pression and the normal temperature. This affords an efficient 
means of supplying cooled air for ventilating purposes. 

The heating power of different kinds of fuel varies consider- 
ably, depending chiefly upon the proportion of carbon and 
hydrogen which they contain. One atom or i3«5 lbs. of 
hydrogen combines with one atom or 100 lbs. of oxygen to 
form ii2'5 lbs. of water. The heat evolved by this com- 
bination is (^%^^'>i^ units per pound of hydrogen. One atom or 
75 lbs. of carbon unites with % atoms or !200 lbs. of oxygen to 
form :&75 lbs. of carbonic acid. The heat evolved in this 

Observations on Warming. 99 

process, according to Dulong's experiments, is 12,906 units 
per lb. of carbon. One atom or 75 lbs. of carbon combining 
with one atom or 100 lbs. of oxygen forms 175 lbs. of carbonic 
oxide. The heat evolved in the process, according to Dulong's 
experiments, is ^1495 units per lb. of carbon, and, according to 
the same experimenter, i lb. of carbon in the form of carbonic 
oxide burning to carbonic acid evolves 10,400 units of heat. 

If the constituents of fuel be classed under C = carbon, 
H = hydrogen, and O = oxygen, and refuse, the theoretical 
evaporative power or total heat of combustion in units of 
evaporation per unit of weight of fuel from chemical analysis 
may be briefly expressed by the formula — 

Theoretical evaporative power = 156"+ 64 (//" — ^) 

as shown in the following table of examples of theoretical 
evaporative powers of fuel.^ 

Carbon 15 

Hydrogen 64 

Various hydrocarbons from 20 to 2'i\ 

Charcoal and coke „ 12 to 14 

Coal, best qualities : — anthracite 15 

bituminous .... from 14 to 16 
oxygenous about 13} 

n If 

V ,» brown „ la 

Peat, absolutely dry , 10 

Wood „ „ 7j 

The oxygen required for combustion is supplied hy air. 

^ The following are the results which have been obtained by direct experiment 
by various experimenters of the value of several of the above substances as gene- 
rators of heat. 

Units of hfeat 
per lb. of fuel. 

Hydrogen burning to water will give 62,535 

Petroleum will give 20,240 

Carbon burnt to carbonic acid 12,906 to 14,040 

)* » oxide 2,495 

Coal (mean of 97 varieties) i3>oo7 

Charcoal from wood , 1 2,000 

Coke 10,970 

Wood, perfectly dry 6,480 

Wood, in ordinary state of dryness (i.e. with 20 °/q of water) 5,040 

Peat (dried naturally) ; 7,150 

Peat (dried artificially) . 8,736 

H % 

lOO Observations on Warming. 

One atom or loo lbs. of oxygen combines with % atoms or 350 
lbs. of nitrogen to form 450 lbs. of air, equivalent to about 
5800 cubic feet of air at 6jj° Fahrenheit. One pound of hydro- 
gen requires for its combustion 8 lbs. of oxygen, equivalent 
to 36 lbs. or about 470 cubic feet of air ; and one pound of 
carbon requires for its combustion 2*69 lbs. of oxygen, equiva- 
lent to 13 lbs. or 157 cubic feet of air at 62° Fahrenheit. The 
net weight of air which is chemically necessary for the complete 
combustion of a unit of weight of fuel may be expressed by 
the formula — 


where C stands for carbon, H for hydrogen, O for oxygen. 

Nitrogen passes through a fire without material alteration, 
and for purposes of combustion the oxygen alone is available. 
Experience shows that the air which has passed through a fire 
retains a considerable proportion of its normal quantity of 
oxygen, and therefore for practical purposes of combustion 
the supply of air should be increased beyond the quantity 
which theoretical considerations show to be necessary, in the 
ratio of li to i or a to i. 

According to this view, i lb. of coal or charcoal requires for 
its combustion about 300 cubic feet of air at 6a° Fahrenheit ; 
I lb. of dry wood requires about 160 cubic feet of air. The 
efficiency of a furnace may be diminished by from *% to '5 of 
its value through unskilful firing. 

In a close stove perfect combustion depends on the area of 
the grate, and other apertures for admitting air, with reference 
to the fqel used, to the height of chimney or other means for 
drawing or propelling the air through the fuel, and to the power 
of regulating the inflow of air by dampers or doors. In open 
fireplaces, whilst a blazing fire will be best obtained by a 
grating under the fire, yet if the air be properly guided to the 
back as well as to the front and sides of the upper part of the 
fuel in an open fire, a bright fire will be obtained; as for 

Observations on Warming. loi 

instance, with Dr. Arnott's pain of a closed bottom to the 
fireplace. In all cases when the supply of air is insufficient, 
carbonic oxide is formed. 

When carbonic oxide is in course of formation in an 
ordinary grate burning coal, the fact is made known by the 
blue-coloured flame sometimes seen. When this occurs it is 
an evidence that there is an insufficient supply of oxygen at 
that part of the fire for supporting combustion. The small 
number of units of heat evolved in the formation of carbonic 
oxide shows the wasteful effect of burning coal with an 
insufficient supply of air. 

The effect of water in a combustible is to diminish the 
actual weight available for producing heat, as well as to 
absorb such a proportion of the heat generated as may be 
necessary for evaporating the water ; therefore fuel should be 
as dry as possible. For instance, a fire supplied with damp 
wood will give out scarcely any heat. 

When oxygen and carbon combine, the volume of carbonic 
acid formed is nearly the same as that of the oxygen con- 
sumed ; therefore when a combustible contains carbon only, 
the volume of gas in the chimney is the same as that of the 
air entering the fire expanded to the volume due to the 
increased temperature. 

The air will be heated to a temperature varying with the 
volume of air admitted, and with the heating power of the 
fuel. If half the air admitted be consumed, the temperature 
of the air as it leaves a coal fire will be 'Xi^^'^ Fahrenheit ; but 
if only one quarter of the air were consumed the temperature 
would be 1159°. These temperatures are rapidly lost by 
radiation, and the temperature in the chimney of a well con- 
structed steam boiler is not above S^d^. Nor is it desirable 
that a greater temperature should be obtained, because be- 
yond that temperature the increase of the volume of air by 
expansion limits the discharging power of the chimney. 

Thus, assuming a chimney of 32^ feet height, with an outside 

102 Observations on Warming. 

temperature of 62^ and a temperature of air in the chimney of 
192°, the velocity of the expanded hot air at exit is 24-69 feet ; 
whilst that of the cold air entering the chimney is 1973 feet 
per second. 

At a temperature of 582® in the chimney, the velocity of 
the expanded hot air at exit would be ^%'(i feet ; whilst that 
of the cold air entering the chimney would be 26«3 feet ; and 
at 712° in the chimney the velocity of the expanded hot air at 
exit would be 58*44 feet per second, but in consequence of the 
expansion of the air, the velocity of the cold air entering the 
chimney would be only 25-99 ^^^^ P^^ second, and higher 
temperatures would show a greater diminution. 

Radiant heat passes through moderate thicknesses of air 
without sensibly heating the air, so that it may be assumed 
that air cannot be heated by the radiant heat from a flame or 
an open fire ; but the radiant heat warms the bodies which 
intercept the rays, and these bodies warm the air. For mode- 
rate temperatures the emission of heat from bodies by radiation 
is proportional to the difference of temperature, but for high 
temperatures and great differences of temperature, the propor- 
tionate emission is much greater. The radiant power of a body 
is equal to its absorbing power, and varies with the nature of 
the surface. The units of heat emitted or absorbed from a 
square foot of surface per hour, for a difference of 1° Fahrenheit 
are — 

Silver, polished 0265 

Lead, sheet -1328 

Iron, ordinary sheet •5661 

Glass -5948 

Cast iron, new -6480 

Building stone, Plaster. Wood, Brick . . -7358 

Sand, fine '740o 

Water 1*0853 

Heat is emitted and absorbed in an accelerating ratio in 
proportion as the difference of temperature increases between 
the body from which the heat is radiated and the body which 
receives the heat ; and with the same difference of tempera- 

Observations on Warming. 


ture between the recipient and the radiant, the effect of the 
radiant will be greater according to the increased temperature 
of the recipient. In. other words, the ratio of the emission of 
heat increases with the temperature. 

Excess of tem- 
perature of the 
radiant over that 
of the recipient 
in degrees 

Temperature of recipient of radiant heat 
in degrees Fahrenheit. 







Ratio of heat emitted in units of heat. 


2 07 






Mr. Anderson made some experiments in 1875 on hot 
water pipes at low temperatures ^, and found that with a con- 
stant difference of temperature of 50° Fahrenheit, between 
the surface of the pipes and the air — 

With the temperature of air = 






The total heat units given ) 
out by pipes was == \ 






For the several reasons above mentioned, it is more econom- 
ical to effect the warming of a given space by means of a highly 
heated surface than by a surface emitting a lower temperature. 

* The total amount of heat radiated per square foot per hour by heated iron 

pipes may be found by the expression my.c? (a*— i), and the heat carried off by 
convection or contact vnth air in atmospheric air by the formula — 

in which a « 1-0077, ^ — temperature of air surrounding the pipes, / = difference 
of temperature between the air and the pipes, both in centigrade degrees,^ = height 
of barometer in millimetres, and m a coefficient of radiation. 
Reduced to British units and 30 inches of the barometer, the formula becomes 
u = total units emitted per square foot per hour by radiation and convection 

— mx 1.00427^ (i.oo427<— i) + 0.2853 X / 1.233. 
P^clet found the value of »i = 124.76 for iron. 

Mr. Anderson's experiments on hot water pipes gave m ^ 122 for cast-iron pipes, 
and m » 250 for wrought-iron pipes. Proc. Inst. C. £. vol. xlviii. 

104 Observations on Warming. 

An open fire warms the air in a room by first warming the 
walls, floor, ceiling, and articles in the room, and these in 
their turn warm the air. Therefore in a room with an open 
fire, the air of the room is, as a rule, less heated than the 
walls. In this case the warming of the air depends on the 
capacity of the surfaces to absorb or emit heat ; except that 
the heat received by the walls may be divided into two parts, 
one part heating the air in contact with the wall, and the 
other passing through the wall to the outer surface, where it is 
finally dissipated and wasted. 

In a close stove heated to a moderate temperature, the 
heat, as it passes from the fire, warms the surface of the 
materials which enclose and are in contact with the fire and 
with the heated gases, and transfers the heat through the 
materials to the outer surface in contact with the air ; and the 
air is warmed by the agency of this outer surface. If heated 
to high temperatures it gives out radiant heat, which passes 
through the air and warms the objects on which the rays 

With hot-water pipes, the heat from the water heats the 
inner surface of the pipe, and this surface transfers its heat to 
the outer surface through the material of the pipes. The rate 
at which the heat can pass from the. inner to the outer surface, 
and be thus utilised instead of passing away straight into the 
chimney, depends on the heat evolved by the fire, on the 
extent of surfaces exposed to the heat, on th^ir capacity to 
absorb and emit heat, and on the quality of the material 
between the inner and outer surface as a good or bad con- 
ductor of heat. 

The passage of heat through a body by conduction varies 
directly with the quality of material and with the difference 
between the temperature of the inner surface exposed to the 
heat and the outer surface exposed to a cooling influence, and 
inversely as the thickness between the surfaces. 

Observations on Warming. 105 

\l E =. loss by conduction, 

C = conducting power of the material, 
f = temperature of heated surface, 
/ = temperature of cooler surface, 
G = thickness of material, 

then E = ^-S^. 

The following is the quantity of heat in units transmitted 
per square foot per hour by a plate one inch in thickness, 
with a difference of 1° Fahrenheit between the temperature 
of the two sides of the plate : — 

Copper 515 

Iron 233 

Lead 113 

Stone (calcareous) 13*68 

Glass 6-6 

Brickwork 4-83 

Plaster 3-86 

Wood, fir parallel to the fibres ... T'37 

„ fir perpendicular to the fibre's . . '75 

Wool -33 

Thus, Other things being equal, copper is a better material 
than iron for conveying the heat from the fire to water or air ; 
and coverings of brickwork, wood, or woollen fabrics are 
better adapted than iron for retaining the heat. The pro- 
perty which appears more than any other to make materials 
good non-conductors of heat is their porosity to air, and the 
fact that they retain air in their pores ; air being a very bad 
conductor of heat. 

The direct warming of the air may be effected by stoves 
with brick or iron flues, or by hot-water or steam pipes. The 
sizes of the heating surfaces for this object must be propor- 
tioned to the volume of air required and to the degree of heat 
to be maintained ; and it should also bear a proportion to 
the thickness and the capacity of the material for absorbing 
and radiating heat and for transferring heat from" one surface 
to the other. When a large volume of air is supplied and 

io6 Observations on Warming. 

removed for ventilation, rapidity in transferring the heat from 
the fuel to the air is an important consideration. 

Brick stoves and flues are worse conductors of heat than iron 
stoves or flues ; therefore the heat generated in a brick stove 
passes more slowly to the outer surface ; but the surface of a 
brick stove parts with the heat which reaches it somewhat more 
rapidly than do the surfaces of an iron stove. The slow con- 
ducting power of the material, and the greater thickness, of a 
brick stove, prevent alternations which may take place in the 
fire from being felt so much as with iron stoves or flues ; and 
therefore the brick stove warms the air more equably, without 
sudden variations ; the air so warmed is free from objectionable 
effects; and where they can be conveniently applied, it is advis- 
able to use brick stoves for warming air for ventilating purposes. 

With an iron flue-pipe from a stove, almost the whole heat 
which any fuel is capable of developing may be utilised by 
using a long flue-pipe, horizontal for the greater part of its 
length, to convey the products of combustion to the outer air. 
The heat given out by- a stove-pipe varies with the tempera- 
ture from end to end, being of course greatest at the end next 
the stove, where the loss of heat is very rapid ; and the 
amount of heat given out per square foot will vary at each 
point as the distance from the stove increases ; and thus- in 
dividing the pipe into lengths, each giving out an equal 
amount of heat, the length of the portion of pipe at the end 
further from the stove would be considerable, whilst the length 
required to give out an equal amount of heat near the stove 
would be very short. But not only does the amount of heat 
given out vary greatly from end to end of the flue-pipe, but 
the proportions into which the heat divides itself between the 
walls and the air vary greatly with the temperature. 

Thus, with a stove-pipe heated at the end nearest the stove 
to a dull red heat of 1 230*^ Fahr., and of sufficient length to 
allow the heat to be diminished to 150° at the further end, it 
would be found that at the stove end of the flue-pipe 92 per 

Observations on Warming. 107 

cent, of the total heat emitted by the pipe is given out by 
radiation to the walls, and only 8 per cent, to the air ; but at 
the exit end the heat is nearly equally divided, the walls 
receiving ^^ and the air 45 per cent. Taking the whole 
length of such a pipe, the walls would receive 74 per cent, and 
the air %6 per cent, of the heat emitted. But with a flue-pipe 
heated to lower temperatures the air might receive half the 
heat or even more. 

When therefore the object is to heat the walls rather than 
the air, which is sometimes the case, the temperature of the 
pipes should be high, and for this purpose stove pipes are 
more effective than hot-water pipes, or low-pressure steam 
pipes. Hot-water pipes heated under pressure on Perkins' 
system, and high-pressure steam pipes would radiate a pro- 
portion of heat to warm the walls. 

At high temperatures there will be practically little dif- 
ference of effect between horizontal and vertical flue-pipes, 
because the heat given out is principally that due to radiation, 
which is independent of the form and position of the radiant. 

An adequate proportion of flue-pipes to the form and size 
of the stove involves a large surface for the flue-pipe ; and 
with a careful observance of proportion, as much as 94^ per 
cent, of the heat in the fuel has been utilised, only 5J per 
cent, being carried away. 

In deciding on the proper height of the flue-pipe as a 
chimney for a stove, it is necessary to consider the three 
cases of (i) a vertical flue-pipe and of a pipe with a uniform 
upward slope, {%) of a pipe with the part next the stove ver- 
tical while the rest is horizontal, and (3) of a pipe with the 
part next the stove horizontal and the vertical part at the 
exit end. 

The mean temperature of the internal air in the flue 
regulates the discharge which, after allowing for friction, is 
due to the difference in vertical height between the stove end 
of the flue and the chimney end. Therefore, in the first case 

io8 Observations on Warming. 

— viz. with a flue either vertical or with a uniform slope — the 
mean temperature of the flue may be taken. In the second 
case, the high temperature at the vertical part of the flue may 
render a shorter vertical flue equally efficient with the inclined 
flue ; but in the case where the horizontal part of the flue is 
nearest to the fire, and the vertical part the furthest removed 
from it, the temperature in the vertical part of the flue will be 
lower, being further from the stove, and the height of the vertical 
portion must be greater than that of the uniformly inclined flue. 

There are however several serious objections to iron stoves, 
especially for small rooms. A long flue-pipe is unsightly, and 
on that account often inadmissible. 

Iron stoves heat rapidly, and easily become red-hot, there- 
fore the effect produced is unequal both on the air and on 
persons in the vicinity of the stove. An iron stove cools down 
with rapidity when the fire is low. The flue-pipe gives out 
unequal degrees of heat in the different parts of its length. 
With an iron stove the temperature at i8 inches from the stove 
has been found to exceed by 37° Fahrenheit that observed at 
6 feet from it ; and with a red-hot stove the difference in that 
short distance was in some cases as much as 45^ 

Carbonic oxide has been found in air heated by iron stoves. 
This can only occur provided the stove be very highly heated ; 
but a high temperature is a liability to which iron stoves are 
subject. Iron very highly heated may take up the oxygen 
from the carbonic acid prevalent in the air of an occupied 
room, and thus reduce the carbonic acid to the condition of 
carbonic oxide. Moreover, the quantity of dust in a room, 
which almost always contains organic matter, may under 
these conditions of temperature somewhat influence the 
presence of carbonic oxide. 

General Morin alleges that he found these eflfects to be 
nearly three times as great with cast-iron as with wrought- 
iron stoves. It has also been alleged that the carbonic oxide 
generated in the fire may permeate through the cast iron of 

Observations on Warming. 109 

the stove, if very highly heated or of inferior porous metal, into 
the surrounding air. Carbonic oxide may also be produced 
by the oxygen of the air acting on the carbon in the cast iron 
if heated to a red heat. The effect would be diminished by 
the presence of moisture in the air. Consequently the use of 
vessels containing water on metal stoves has been recom- 
mended. Whatever may be the value of these allegations, 
the use of surfaces of iron heated to a red heat for warming 
air for ventilating purposes is seriously objectionable. Hence, 
whenever iron stoves or cockles are used for heating air, care 
should be taken to prevent the iron from attaining a high 
temperature, and with this object all iron stoves should have 
a lining of fire-brick, so as to prevent the fire from coming in 
direct contact with the iron ; such an arrangement prevents 
these inconveniences, and preserves greater regularity in the 
heating of the air. An advantageous arrangement would be 
to provide the iron with an outer surface of glazed enamel, 
because the iron would convey the heat rapidly from the fire 
to the surface, while the enamel surface would emit the heat 
more rapidly than the iron surface. 

Hot-water pipes for warming air are free from many of the 
objections arising from the direct application of heat to iron, 
because the heat can be regulated with exactness. 

Water boils at ii%^ Fahrenheit under the atmospheric 
pressure of I4'7 lbs. per square inch, or 30 inches of mercury, 
i.e. at about the sea level. Under one-half that pressure, 
viz. 7«3 lbs. per square inch, or 15 inches of mercury, it boils 
at 180° ; and under a pressure of four atmospheres above the 
ordinary atmospheric pressure, or 44 lbs. per square inch, it 
boils at 291°; and under a pressure of 10 atmospheres, or 
132 lbs. per square inch, it boils at 357° Fahrenheit. Thus 
a high temperature may be obtained from water without 
generating steam by heating it under pressure. Salt water, 
saturated with \i*% lbs. of salt per 100 lbs. of water, boils at 
227"*, and freezes at about the zero of Fahrenheit. 

no Observations on Warming. 

Steam is generated under the mean atmospheric pressure 
of 30 inches of mercury when the boiling-point is attained, 
i. e. Tti'f ; but before the water becomes vapour, such further 
amount of heat equal to 966° is absorbed and becomes latent 
as would sujffice to raise the water to 1178° if it did not turn 
into steam. The temperature of 11 78**, which is thus required 
to produce steam, is necessarily constant, and consequently a 
greater or less amount of heat becomes latent according as 
the pressure is below or above the atmospheric pressure. 
This latent heat is given out on the reconversion of steam 
into water. 

Pipes may be heated either by hot water or by steam. 
The higher the temperature, the greater is their comparative 
effect on the warming of air. Thus pipes heated by hot water 
under pressure convey heat to the air with greater rapidity 
than pipes heated by hot water at low pressures ; and steam 
pipes are more effective than hot-water pipes ; and steam at a 
high pressure is more effective than low-pressure steam. One 
advantage of heating by steam or by water under considerable 
pressure is the high temperature obtained in the pipes, and 
the consequent radiation of a larger proportion of heat to the 
walls of a room, than takes place with pipes heated under 
ordinary pressures. When steam is employed to warm air 
for ventilating purposes it is often necessary to adopt special 
means to moderate the temperature in warm weather. 

There is moreover some practical inconvenience attending 
the use of high-pressure steam in localities where an ex- 
perienced workman is not near at hand. For this reason, 
hot-water pipes have been generally preferred for warming 
ordinary houses. 

The efficient action of hot-water pipes depends upon the 
upward flow of the heated and expanded water, as it passes 
from the boiler, being made as direct as possible, and so pro- 
tected as to lose little heat between the boiler and the place 
where the heat- is to be utilised. The return pipe, which 

Observations on Warming. iit 

brings back the water to the boiler after it has been cooled 
down by the abstraction of heat in warming the air, should be 
passed in to the bottom of the boiler as directly, and in as 
uniform a line from the place where the heat has been used, as 
possible. The velocity of flow in the pipes will depend upon 
the temperature at which the water leaves the boiler, the height 
to which the heated water has to rise, and the temperature at 
which it passes down the return pipe back into the boiler. 
The efficiency of a hot-water apparatus will be regulated by 
these conditions, by the sizes of the pipes, and by such other 
conditions as affect the flow of water in pipes. 

It will be evident that to obtain an equal velocity of flow 
when the height of the vertical column is small, the tempera- 
ture at which the water returns to the boiler must be lower 
than when the vertical column is long. Therefore when the 
boiler or source of heat is very near the level of the pipes 
for heating the air, the average temperature which can be 
obtained in the pipes will be lower than when the vertical 
column is long. Hence, the heating surface of the boiler, 
the area of the grate which regulates the flow of air to the 
fuel, and the surface of pipe which enables the heat from the 
boiler to be utilised, must be regulated with reference to this 
difference of level. 

It may further be assumed that with small pipes, the 
temperature being constant, the velocity of flow in the pipe 
necessary to furnish a given amount of heat will vary in the 
ratio of the length of the pipe. 

When the water circulates through the pipes by virtue of 
the difference of temperature of the flow and return currents 
only, it is impossible to count upon a greater mean tempera- 
ture in the pipes than from i6o° to i8o°, because above that 
temperature the water in the boiler begins to boil, and causes 
an overflow of the supply cistern and escape of steam at the 
air pipes. In order to obtain a sufficient velocity of circu- 
lation for long distances, or with small differences of level, a 

112 Observations on Warming. 


forced circulation may be resorted to, as has been done by- 
Messrs. Easton and Anderson at the County Lunatic Asylum 
at Banstead. 

There two pipes are laid side by side, one of which com- 
municates with the boilers and is termed the flow-pipe; the 
other, termed the return-pipe, is connected with the feed- 
cistern for the boilers, which cistern is situated above the 
level of the boilers. 

Both pipes are connected with the various coils to which 
the heated water is desired to be conveyed, by valves which 
can be opened or closed at wilL An Archimedean screw 
pump is fixed on the return-pipe near the point where the 
pipe ascends into the cistern. This pump is always kept at 
work. When the communications between the flow and 
return pipes are closed, the screw simply slips through the 
water; as soon as any communication is opened, the screw 
draws the water along the pipe, and forces it into the cistern, 
thus ensuring a constant circulation. 

By Perkins' system the water is heated under considerable 
pressure, and a higher temperature is thus obtainable than 
with ordinary pressures. 

In its simplest form the apparatus consists of a continuous 
or endless iron tube of about one inch diameter, closed in all 
parts and filled with water. The joints are screw-joints 
connected within a socket forming a right and left hand 
screw. About one-sixth part of the tube is coiled in any 
suitable form and placed in the furnace, forming the heating 
surface, and the other five-sixths are heated by the circu- 
lation of the water which flows from the top of this coil ; and 
after having been cooled in its progress through the building, 
returns to the bottom of the coil to be re-heated. 

Water when heated from 39^1 to 212^ Fahrenheit expands 
about 5 per cent, of its original bulk. Therefore, in order 
to provide for the increased volume of the water when 
heated, a tube called an expansion tube is placed above the 

Observations on Warming-, 113 

highest level of the smaller tubes which convey the heat to 
the distant parts of the building. 

This tube is of larger diameter than those used as heating 
surfaces, and its length and capacity are proportioned to the 
quantity of tube to which it is attached. 

The filling tube of the apparatus is placed on a level with 
the bottom of this expansion tube so as perfectly to fill all 
the small tubes, and yet prevent the possibility of filling the 
expansion tube itself. The expansion tube being then left 
empty, allows the water as it becomes heated to expand with- 
out endangering the bursting of the smaller tubes. 

The apparatus is filled by an opening connected with the 
lowest line of tubing, so that the water, as it rises, drives the 
air before it, and out through an opening in the expansion 
tube. Great care must be taken to expel all air from the 
pipes by repeatedly forcing water through them. When the 
pipes are filled, both the opening in the filling tubes and the 
opening in the expansion tube are closed by screw-plugs. 

The form and size of the furnace varies according to the 
locality and the work the pipes have to do. 

A temperature of as much as 300° Fahrenheit can be 
obtained in the tubes. 

Steam in lieu of hot water is especially applicable, either 
where steam for other purposes is in use, as for instance where 
the exhaust steam from an engine is available, or where heating 
is required on a large scale. 

Where exhaust steam is used it may be assumed that the 
capacity of heating is almost in a ratio with the fuel ex- 
pended under the boiler. There is some cooling in passing 
through the engine, and along the pipes to the rooms requir- 
ing to be heated, but this is slight if the pipes and cylinder 
are covered with a good non-conducting material, so that the 
condensed water may be taken back hot into the boiler. 

So far as economy in heating is concerned, there is not 
much difference between the use of exhaust steam and of that 


H4 , Observations on Warming'^ 

taken direct from the boiler. For instance, if the steam is 
taken from the boiler direct to the pipes at five atmospheres, 
the temperature would be a little over 300°. If, on th^ other 
hand, a comparative capacity of steam were allowed to pass 
' through the engine to create power, and discharged into the 
pipes at one atmosphere, it would decrease in temperature to 
something over 7,12^ ^ but it would increase in bulk according 
to the expansion ; and by an increased heating surface a larger 
proportion of beat can be practically utilised than shown by 
the difference of the temperature. 

As regards t^e extent to which exhaust steam can be 
carried, Mr. Boultqn, who has perfected this system, mentions 
a case in which fropi an engine of ^5 horse power nominal, 
with a 17-inch cylinder, the exhaust steam travels about iioo 
yards in a direct line, and passes along various branches,^ 
amounting in the aggregate to about ^^386 yards, or i J miles 
of li-inch pipes, and then into the tank to warm the feed 
water for the boiler. 

The use of steam for heating on a large scale is exemplified 
in many of our principal buildings, and amongst others in the 
Houses of Parliament. An interesting application of steam 
for heating, is that adopted at Lockport, a town in the United 
States, of which a description has been given by Mr. George 
Maw; the object i^ this c^se being to avoid the necessity of 
having separate fires itx each house. 

Two hundred houses were heated from the central supply 
through about three miles of pipiag, radiating from the boiler- 
house ; this building contained two boilers each 16 ft. by 5 ft., 
and one boiler of about half the size. The steam during the 
winter was maintained at a pressure of 35 lbs. to the inch, with 
a consumption of four tons of anthracite coal in 24 hours. 

This boiler pressure of 35 lbs. was maintained through the 
entire length of the three miles of main piping up to the 
points of consumption, at each of which there is a cut-off 
under the control of the consumers. 

Observations on Warming. 115 

The first 600 feet of mains from the boilers are 4 inches 
in diameter. There are 1400 feet of 3-inch pipes ; 1500 ft. of 
2j-inch pipes ; and 2000 feet of a-inch pipes. The supply- 
pipes from these mains to the houses are li-inch in diameter, 
and within each house f-inch pipes are used. In addition to 
the cut-off tap from the main under the control of the con- 
sumer there is a pressure-valve regulated to a 5 lb. pressure 
under the control of the company, and beyond this is an 
ingeniously constructed meter, which not only indicates the 
total consumption in cubic feet of steam, but also the quantity 
of steam used in each apartment. At each 100 feet of main 
an expansion-valve like an ordinary piston and socket is 
inserted, allowing an expansion in each section of 100 ft. of 
i|-inch for the heat at 35 lbs. pressure. No condensation 
occurs in the mains. They are covered with a thin layer of 
asbestos paper next the iron, then a wrapping of Russian felt, 
and are finally wrapped round with Manilla paper like smooth 
light brown paper, and the whole encased in timber bored out 
three-quarters of an inch larger than the covered pipes, and 
laid along the streets like gas pipes. 

The distribution of heat in the apartments is by means of 
radiators, consisting of inch pipes 30 inches long placed ver- 
tically either in a circle or in a double row, and connected 
together top and bottom ; they are furnished with an outlet 
pipe for the condensed water which escapes at a temperature 
a little below boiling, sufficient in quantity for the domestic 
purposes of the house, or it may be used as accessory heating 
power for horticultural and other purposes. 

The steam was also applied at a distance of over half a mile 
from the boilers for motive power, and it was stated that two 
steam engines of lo-horse and 14-horse power were occasionally 
worked from the boilers at a distance of half a mile with but 
a slightly increased consumption of fuel. 

The laid-on steam was used for cooking purposes, for boiling, 
and even baking ; in a house three-quarters of a mile from the 

I % 


Observations on Warming. 

boilers, a bucket of cold water could be raised to boiling heat 
in three minutes, by the passage of the steam through a 
perforated nozzle plunged into the bucket. 

As in the case of gas supply, the Steam Supply Company 
lay their pipes up to the houses, the consumer paying for all 
internal pipes, fittings, and radiators. 

. In a moderately-sized eight-roomed house the expenses of 
these amounted to 150 dollars, or a trifle over £'>p^ and in 
larger houses with more expensive fittings to 500 dollars, or 
about ;^ioo. 

The working expenses consist of but little more than the 
coal and the wages of two firemen. 
















• * 






































^^^^^ ___^^ 

Difference of \^ 20° 30" 40" so** eo** 70** ao" so** 100° no" 120° 130° 140° 150*^160 Mo'\9^\9Q2Qfi^TtTivpcralun. 

Fig. 18. 

The annexed diagram (Fig. i8), resulting from Mr. Ander- 
son's experiments, is published in the Journal of the Institution 
of Civil Engineers for 1877, and shows the total units of heat 
given out by cast-iron and wrought-iron pipes per square foot 
of surface per hour for various differences of temperature, 
applicable either to hot-water or steam pipes. 

Suppose, for example, it is required to know how much 

Observations on Warming. 117 

heat will be given out by 4-inch pipes at 190® in a room, the 
temperature of which is 60°, the difference of temperature 
being 130°: look along the line of abscissae for 130, and the 
ordinate then gives 2327 units for 4-inch pipes, and '>^^6 units 
for a-inch wrought-iron pipes per square foot per hour. 

When the form of apparatus for heating has been decided 
on, the amount of heating surface to be afforded for purposes 
of ventilation and warming depends mainly upon the volume 
of air to be admitted and removed, and the temperature 
desired to be maintained ; but in any given building there 
are other circumstances to be taken into account — viz. the 
position, aspect, subsoil, temperature of locality, shape and 
size of building, extent, position, and thickness of walls, size 
and form of windows, skylights, and suchlike matters, which 
affect either the temperature of the incoming air or the con- 
ditions which determine the loss of heat in a building. 

A rough empirical rule is suggested by Mr. Anderson for 
hot-water pipes in this country — viz. that i square foot of 
heating surface is required for every 6^ cubic feet of content 
of the space to be warmed, and in a greenhouse i square foot 
to every 24 cubic feet. This empirical rule does not however 
appear to be based on the important sanitary considerations 
of the renewal of air. A similar rough approximate rule for 
pipes heated with exhaust steam has been given by Mr. 
Boulton as follows : i superficial foot of steam pipe for each 
6 superficial feet of glass in the windows ; i superficial foot 
of steam pipe for every 6 cubic feet of air removed for 
ventilating purposes per minute, i superficial foot of steam 
pipe for every 120 superficial feet of wall, roof, or ceiling, 
allowing about 15 per cent, on the amount thus obtained for 
contingencies. He states that, roughly speaking, the exhaust 
steam due to one-horse power can be made to warm 30,000 
cubic feet of space. 

If the fresh air for ventilation is warmed in underground 
chambers, such places should be so formed as to be dry and 

ii8 Observations on Warming. 

well-drained, and the channels leading therefrom constructed 
of such form and materials that the air shall lose as little 
heat as possible in transmission. 

When air is warmed, its capacity for moisture is increased; 
thus if air at 3a** Fahrenheit be in a condition of complete 
saturation from moisture, which is represented by 100 degrees, 
and if the air be then raised to a temperature of 5^^°, it would 
have only 46 degrees of saPturation, and appear to be dry; 
whereas if the temperature be raised to 72° Fahrenheit, the 
degree of saturation would be diminished to 23 degrees,, and 
the air would appear to be extremely dry. 

In a damp climate, up to a certain point, the drying of the 
air caused by raising the temperature is as much a source of 
comfort as the warmth. But when the outer air is dry, the 
additional dryness caused by warming the air may in some 
cases become inconvenient. 

For reasons already mentioned, the open fire does not 
directly warm the air to so great an extent as stoves or 
heated pipes, therefore the dryness is generally not found to 
be objectionable with an open fire. But with stoves or hot- 
water pipes, it is frequently necessary to provide vessels of 
water on or near the stoves or heated pipes, to supply ad- 
ditional vapour to the air when required. 

The conclusions which follow from these considerations are, 
that where a room is heated by warmed air passed through 
flues into the room, without any source of heat in the room itself, 
the air imparts its heat to the walls. The air is thus warmer 
than the walls. When a room is warmed by an open fire, on 
the other hand, the warming is effected by the radiant heat 
from the fire ; the rays from the fire pass through the air 
without sensibly warming it, the radiant heat warms the 
walls and furniture, and these impart their heat to the air. 
Therefore the walls in this case are warmer than the air. 
Consequently in two rooms, one warmed by an open fire and 
one by hot air, and with air at the same temperature in both 

Observations on Warming. 119 

rooms, the walls in the room heated by hot air would be some 
degrees colder than the air, and therefore colder than the 
walls of the room heated by ail open fire, and these colder 
walls would therefore abstract heat from the occupants by 
radiation more rapidly than would be the case in the room 
heated by an open fire. And to bring the walls up to the 
same temperature in the former case — ^viz. in the room heated 
with hot air — the air of the room would require to be heated 
to an amount beyond that necessary for comfort, and therefore 
to a greater amount than is desintble oii sanitary grounds, 
seeing that warmed air contains less oxygeii than cooler air. 
As sick persons are more sensitive to such influences than 
persons in health, this would appear to be the reason why 
in hospital wards the Warming by means of an open fire has 
been always recognised as preferable to warming by hot air. 

To a certain extent, dependent on the temperature of the 
heating surface, a similar effect takes place in rooms warmed 
by close stoves, hot-water pipes, or low-pressufe steam pipes ; 
but in a room warmed by steani pipes heated to a high tem- 
perature, the effect on the walls would tend to approximate 
in some degree to that produced by the open fire in pro- 
portion as the temperature of the pipes is raised. Where 
the object is to warm the walls as well as the air, high-pressure 
steam is a more advantageous mode of heating than hot-water 



One pound of coal is far more than sufficient, if all the 
heat of combustion is utilised, to raise the temperature of a 
room, 2,0 feet square and 12, feet high, to 10 degrees above the 
temperature of the outer air. If the room were not ventilated 
at all, and the walls were composed of non-conducting mate- 
rials, the consumption of fuel to maintain this temperature 
would be very small ; but, in proportion to the change of the 
air of the room and to the escape of heat through the walls, 
windows, ceiling, &c., so would the consumption of fuel neces- 
sary to maintain that temperature increase. If the volume 
of air contained in the room above mentioned were changed 
every hour, one pound of coal additional would be required 
per hour to heat the inflowing air, so that to maintain the 
temperature at 10 degrees above that of the outer air during 
I z hours would require 1 2 lbs. of coal. 

The principle of the ordinary open fireplace is that the 
coal shall be placed in a grate, to which air is admitted from 
the bottom and sides to aid in the combustion of the coal ; 
and an ordinary fireplace, for a room of 2,0 feet square and 1 2 
feet high, will contain from about 15 to 30 lbs. at a time, and, 
if the fire be kept up for 12 hours, probably the consumption 
will be about 100 lbs., or the consumption may be assumed at 
about 8 lbs. of coal an hour. 

As already mentioned, the radiant heat from the fire does 
not warm the air of the room ; the rays from the fire war^m 
the sides and back and parts adjacent to the grate, they warm 

Action of Open Fireplaces on Ventilation. 121 

the walls, floors, ceiling, and furniture of the room, and these 
impart heat to the air. The form and material of the fire- 
place can thus assist materially the warming of the air. The 
rays should impinge more freely on the walls and floor than 
on the ceiling. A projecting chimney-piece with a surface 
favourable to the absorption and emission of heat would be 
more favourable to the warming and circulation of the air 
than one which would allow the rays to pass to the ceiling. 
In an ordinary fireplace the sides should be splayed, as in 
the Rumford form of grate ; the sides and back should be of 
non-conducting material, with a surface favourable to the rapid 
absorption and emission of heat. Thus brick or tiles are better 
than iron for this purpose. Similarly, the degree to which the 
materials of the walls or floor of the room are unfavourable to 
conduction but favourable to the absorption and emission of heat 
will have a bearing on the capacity of the room for warmth. 

One pound of coal may be assumed to require, for its per- 
fect combustion, 160 cubic feet of atmospheric air; 8 lbs. 
would require i,:z8o cubic feet ; but at a very low computa- 
tion of the velocity of the gases in an ordinary chimney-flue 
the air would pass up the chimney at a rate of from 4 to 6 
feet per second, or from 14,000 to iio,ooo cubic feet per hour ; 
with the chimneys in ordinary use, a velocity of from 10 to 
15 feet per second often prevails, giving an outflow of air of 
from 35,000 to 40,000 cubic feet per hour. This air comes 
into the room cold, and when it is beginning to be warmed it 
is drawn away up the chimney, and its place filled by fresh 
cold air. A room 20 feet square and \% feet high contains 
4,800 cubic feet of space. In such a room, with a good fire, 
the air would be removed four or five times an hour with 
a moderate draught in the chimney, and six or eight times with 
a blazing fire. The atmosphere of the room is thus being 
cooled down rapidly by the continued influx of cold air to 
supply the place of the warmer air drawn up the chimney ; 
and General Morin estimated that of the heat generated by 

122 Action of Open Fireplaces 

fuel in an ordinary open fireplace about one eighth only is 
utilised in the room. The very means adopted to heat the 
room tends to produce draughts, because the stronger the 
direct radiation, or rather the brighter the flame in open 
fireplaces, the stronger must be the draught of the fire and 
the abstraction of heat- 

A fireplace is thus powerful enough to draw into the room all 
the air it wants ; and for this purpose will use indiscriminately 
all other openings, whether inlets or outlets if necessary. 

The only way to prevent draughts is to adopt means for pro- 
viding fresh warmed air to supply the place of that removed. 

Fig. 19. Sketch of Experiment made by Mr. Campbell in 1S57 showing the 
movement of currents of air in a room with an open fireplace. 

The way in which an ordinary open fireplace acts to create 
circulation of air in a room with closed doors and windows, is 
as follows : — The air is drawn along the floor towards the 
grate ; it is then warmed by the heat which pervades all 
objects near the fire, and part is carried up the chimney with 
the smoke, whilst the remainder, partly in consequence of the 
warmth it has acquired from the fire, and partly owing to the 
impetus created in its movement towards the fire, flows up- 
wards towards the ceiling near the chimney-breast It passes 
along the ceiling, and as it cools in its prepress towards the 

on Ventilation. 123 

opposite wall, descends to the floor, to be again drawn towards 
the fireplace. 

It follows from this, that with an open fireplace in a room, 
the best position in which to deliver the fresh air required to 
take the place of that which has passed up the chimney, is 
above the projecting chimney-piece, and at any convenient 
point in the chimney breast, between the chimney-piece and 
the top of the room, for the air thus falls into the warmer 
upward current, and mixes with the air of the room, without 
perceptible disturbance ; and it also follows that if two fire- 
places be placed in a room, they should, unless the room be 
very large, both be on the same side of the room, and if shafts 
are required for ventilating a room in addition to the open 
fireplace, they also should be placed on the same side of the 
room as the fireplace, but as far from it as possible. 

The open fireplace thus presents special advantages for 
securing efficient ventilation by means of the circulation of air 
which it creates. It makes the room in which it is in use 
independent of other means of extraction of air, unless the 
room is very crowded, or beyond the size for which the fire- 
place is calculated. 

The air thus extracted must be drawn into the room from 
somewhere, and unless arrangements be made for supplying 
the room with warmed fresh air, cold air finds its way into 
the room through the special inlets, if any are provided ; if not, 
through the chinks of the windows and doors, or wherever it 
can get in most easily. The large volume of fresh air required 
to supply that drawn up the chimney cannot always be 
warmed with sufficient rapidity by contact with the walls and 
furniture only ; the temperature in different parts of the room 
is therefore frequently very unequal, and the occupants are 
subjected to draughts ; and if there are two fireplaces in the 
same room unprovided with special means for the admission 
of fresh air, one of which is not lighted, the air is often drawn 
down the vacant chimney. 


Actioft' of Open Fireplaces 

It IS therefore desirable in cold weather to replace some 
part at least of the air drawn up the chimney by air partially 
warmed. This can be effected in various ways, either by a 
service of hot-water pipes carried into all the rooms, to coils 
placed in front of the fresh-air inlets, so as to warm the fresh 
air as it enters the room ; or by bringing warmed air into the 
room by flues from a central heating apparatus. But the simplest 
plan is to utilise some of the waste heat which in the case of 
an ordinary open fireplace passes away unused up the chimney. 

Fig. 20. Section of a room with a ventilating grate and warm-air flue, 
showing action of Are in producing circulation of air. 

The ventilating fireplace was designed with the object of 
obviating the above-named objections to the common fire- 
place, of utilising spare heat, and of providing such adequate 
means of ventilating the soldiers' rooms in cold weather, when 
the windows are shut, as would not be liable to be deranged. 

Fresh air is admitted to a chamber formed at the back of 
the grate, where it is moderately warmed by a large heating 
surface, and then carried by a flue, adjacent to the chimney- 
flue, to the upper part of the room, where it flows into the 
currents which already exist in the room. These grates have 
never been patented, and therefore manufacturers do not care 
to suggest their use. 

on Ventilation. 125 

The body of the stove is of the best cast iron, and consists 
of three pieces, properly connected by screws. The first piece 
forms the moulded projecting frame ; the second the body of 
the grate; and the third the nozzle or connexion with the 
smoke-flue, the bottom flange of which is bolted to the back 
of the grate. The stoves are of three sizes. The largest has 
an opening for fire of i ft. 9 in. wide, and was intended for 
rooms containing from 7,500 to 10,000 cubic feet ; it weighs 
about 3 cwt. I qr. 10 lbs. The second, or medium size, has 
an opening for fire i ft. 5 in. wide, and was intended for rooms 
containing from 3,600 to 7,500 cubic feet; it weighs about 
2 cwt. 3 qrs. 5 lbs. The third, or smallest size, has an opening 
for fire 1 ft. 3 in. wide, and was intended for rooms contain- 
ing 3,600 cubic feet and under ; it weighs about % cwt. 2 qrs. 

The figures appended (Figs. 21-24) show an elevation, 
section, and plan of the second or medium-size stove, the 
extreme dimensions of which are 40 inches wide by 42 inches 
high. The projecting moulded frame enables the stove to 
be applied to any existing chimney-opening. 

The fireplace has a lining of fire-lumps in five pieces — two 
sides, one back-piece, and two bottom-pieces, moulded to the 
form shown in the woodcut. The object of this fire-clay 
lining or cradle is to prevent the contact of the incandescent 
fuel with the iron, and to preserve a high uniform temperature 
in the vicinity of the fuel to assist the combustion. 

The bottom is partly solid, being made of two fire-lumps 
placed one on each side, and supporting an intermediate cast- 
iron fire-grating, which occupies about one-third of the bottom 
of the grate ; by this means, whilst the draught is checked 
by the solid part of the bottom of the grate, and the con- 
sumption of fuel reduced, a sufficient supply of air is obtained 
for combustion through the grating to secure a cheerful fire. 
A clear space, half an inch deep, is formed between the back 
piece of fire-lump and the iron back of the grate, through 
which a supply of air passes from the ash-pit under the grate, 


Action of Open Fireplaces 

and through a slit in the fire-lump, on to the upper part of the 
back of the fire. The air thus brought into contact with the 
heated coal is received at a high temperature, in consequence 
of passing through the heated fire-lump, and is forced into 
contact with the gases from the coal by means of the piece of 

fire-lump which projects over the fire at the back of the grate ; 
and thus a more perfect combustion of the fuel is effected than 
with an ordinary grate, and the creation of smoke is pre- 
vented : in fact, with care, almost perfect combustion of the 
fuel, and consequent utilisation of the heat, can be obtained. 

Whilst the incandescent fuel and flame are kept away from 
actual contact with the iron back of the stove, the heated 

on Ventilation. 127 

gases from combustion, and such small amount of smoke as 
exists, are compelled, by the form of the back of the grate, 
and the iron part of the smoke-flue, to impinge upon a large 
heating surface, so that as much heat as possible may be 
extracted from the gases before they pass into the chim- 
ney ; the heat thus extracted is employed to warm air taken 
directly from the outer air. The air is warmed by the iron 
back of the stove and smoke-flue, upon both of which several 
broad flanges are cast, so as to obtain a large surface of metal 
to give off" the heat. This giving-off" surface (amounting in the 
case of No. i grate to about 18 square feet) is sufficient to 
prevent the fire from ever rendering the back of the grate so 
hot as to injure the air which it is employed to heat. The 
fresh air, after it has been warmed, is passed into the room 
near the ceiling by the flue shown in the woodcut. 

The flue which has been adopted for barracks is carried up 
by the side of the smoke-flue in the chimney-breast. 

In order to afford facilities for the occasional cleansing of 
the air-chamber, and those parts of the air-channels connected 
with it, the front of the stove is secured by screws, so that 
it can be easily removed, thus rendering the air-chambers 

The stove was designed with tte object of being applied 
to existing chimney-openings. In so applying it, the air- 
chamber is to be left as large as possible, thoroughly cleansed 
from old soot, rendered with cement, and lime-whited. Should 
the fireplace be deeper than i foot 6 inches, which is the depth 
required for the curved iron smoke-flue, then a lining of brick- 
work is to be built up at the back to reduce it to that 
dimension. The chimney-bars, if too high, must be lowered 
to suit the height of the stove, or to a height above the hearth 
of 3 feet 3 inches ; they must also be straightened, to receive 
the covering of the air-chambers. These coverings should be 
of 3-inch York or other flagging, cut out to receive the curved 
iron smoke-flue, and also to form the bottom of the warm-air 

128 Action of Open Fireplaces 

flue in the chimney-breast ; or the covering may be formed 
of a brick arch. In new buildings the air-chambers may be 
rectangular; they must be 4 inches narrower than the ex- 
treme dimensions of the moulded frame of the stove, so as to 
give a margin of 1 inches in width all round for a bedding of 
hair mortar. In existing buildings, the recess in which an 
ordinary firegrate would be fixed forms the chamber in 
which the air is warmed. Great care must be taken in 
bedding the several joints to prevent smoke from the flue 
passing into the fresh-air chamber or fresh-air flue. 

The mode of admitting external air into this chamber must 
depend upon the locality. If the fireplace be built in an 
external wall, the opening for fresh air can be made in the 
back ; but if in an internal wall, it will be necessary to con- 
struct a channel from the outside, either between the floor- 
ing of the room and the ceiling-joists of the room below (if 
there be independent ceiling-joists), or between the floor- 
boards and the plaster ceiling of the room below, in the 
spaces between the joists; or by a tube or hollow beam carried 
either below the ceiling of the room altogether ; or, as is often 
more convenient, behind the skirting of the room in which the 
firegrate is placed. 

In this country these horizontal ducts should contain a sec- 
tional area, for the large size grates of 84 square inches, for the 
medium size of 60 square inches, for the smaller size of 36 square 
inches ; the clear area through the grating covering the opening 
to the outer air should be equal in area with that of the flue. 

If the flues are of considerable length, and with bends, the 
sectional area should be rather more than that mentioned, 
to allow for friction ; but if there be a direct communication 
with the outer air, the sectional area may be rather less than 
that recommended. In exceptionally cold weather it may be 
advisable to reduce temporarily the area of inlet. 

The amount of air delivered through the fresh-air flue 
varies somewhat with the direction of the wind. The inlet 

on Ventilation, 129 

shaft acts best when the windows, doors, and other inlets to 
the room are closed, as it then becomes the sole inlet for fresh 
air for the room. 

In the ventilation of barrack-rooms or hospitals, it was not 
intended that the fresh air warmed by the grate should be 
the whole supply of fresh air, nor that the chimney should be 
the sole means employed for the removal of the air to be 
extracted. In ordinary houses, however, the grate, if adopted, 
might be used in such a manner as to perform the whole 
functions of ventilation. In this case it is of course necessary 
to remember that the ventilating power is a fixed quantity, 
and that in originally settling the size of a grate for a par- 
ticular room it will be necessary to bear in mind the general 
object for which the room is to be employed, and the number 
of persons by whom it is required to be occupied with 
efficient ventilation, because all experiments show that no 
room can be considered even tolerably ventilated as a per- 
manent arrangement unless at least icoo cubic feet of air per 
occupant are renewed per hour ; consequently a room lo feet 
long by 15 feet wide and 10 feet high (i. e. with 3000 cubic 
feet of space), with three people in it, would not require the 
air to be changed much more than once an hour; whilst, if 
occupied by twelve or fourteen people, it would require to be 
changed five times in an hour. If the normal use of the 
room was for three people it would not be worth while to 
provide for the extra number by which it might on occasions 
be occupied, as their wants in such a temporary case could 
be met by opening the windows slightly at the top. 

There is one point connected with the fresh-air flue which 
must be carefully attended to — viz. the fresh air should be 
taken from places where impurities cannot affect it, and the 
flue must be so arranged and constructed as to afford easy 
means of being periodically thoroughly examined and cleaned. 
In barracks the rule is that such cleansing should take place 
at least once a year. 


1 30 Aclion of Open Fireplaces 

The area of the grate of No. i stove is 84 square inches, of 
which 58 are solid, and 36 
afford space in the centre 
forthepassingofair. The . 
front is open, and air is 
passed on to the coal from 
the back in the manner 
already described. The 
grate will contain about 
i8to2olbs.ofcoal; when 
the fire is maintained for 
fig. ija. from twelve to fifteen 

hours, a total consumption of about %-^ lbs. per hour, or 
40 lbs. for sixteen hours, will suffice to maintain a good fire. 
In new buildings it 
would be possible, and 
in some cases it might 
be desirable, to extend 
this heating surface con- 
siderably by carrying 
up the smoke flueinside 
the warm-air flue. But 
care must be taken not 
to overheat the air. This 
plan has been adopted 
in the fireplaces for the 
wards of the Herbert 
Hospital, where the fire- 
place is in the centre of 
the ward, and the chim- 
ney consequently passes 
under the floor, and is 
f'g- >5*- placed in the centre of 

the flue which brings in the fresh air to be warmed by 
the fireplace; by this means more than 36 superficial feet 

on Ventilation. 131 

of heating surface have been obtained for warming the fresh 
air in addition to the heating surface afforded by the air flues 
in the fireplace. (Figs. ^^ a^ %^ b.) 

The fire stands in an iron cradle fitted to the fire- 
clay back and sides, and a current of air is brought through 
the fire-clay at the back, where it becomes heated, on to 
the top of the fire to assist the combustion, and thus pre- 
vent smoke. The top of the stove is coved inside, to lead 
the smoke easily into the chimney. The main body of the 
stove is a mass of fire-clay, with flues cast in it, up which 
the fresh air passes from the horizontal air flue already men- 
tioned, in which the chimney flue is laid. Thus all parts of 
the stove employed to warm the fresh air with which the fire 
has direct contact, are of fire-clay. This is especially essen- 
tial in hospitals, where every element of possible impurity 
of air should be avoided. The sectional area of the fresh- 
air flue with this arrangement of grate may be i square inch 
for every 100 feet of cubic contents of the spaces to be 
warmed, for favourable situation ; but in cold or exposed 
localities a less area may be allowed. 

The horizontal chimney flue in the Herbert Hospital fire- 
places is formed of two layers of sheet iron, separated by a 
thin layer of fire-clay, so as to prevent overheating of the 
surface, and it is about no square inches in area. The hori- 
zontal chimney flue terminates in a vertical flue in the side 
wall, which should be rather larger in area than the horizontal 
flue. This vertical flue is carried in the upper floors to a 
height of double the length of the horizontal flue, and is 
carried down to the basement, whence it can be swept. The 
points of connection between the horizontal chimney with the 
descending flue from the fireplace, and with the ascending flue 
in the wall, are very carefully rounded, as this is essential to 
assist the passage of the smoke. The horizontal flue is swept 
from an opening, to which access is obtained by taking up 
a moveable board in the floor, and by pushing a brush along 


Action of Open Fireplaces 

the flue, and thus forcing the soot into the vertical flue, 
whence it falls down and is removed at the opening in the 

There is placed a spare flue by the side of the vertical flue, 
terminating in a fireplace in the basement, which enables the 
vertical flue to be warmed, so as either to make it draw when 
the fire is first lighted, or to enable a current to be main- 
tained for ventilating purposes through the fireplace when the 
fire is not lighted. The portion of floor over the horizontal 
flue should be so constructed as to be taken up in order to 
enable the air flue to be easily and thoroughly cleaned 


A. Fire-lump with warm air flue through 


B. Warm-air pipe to fit into socket on hob, 

in lengths of x ft. 3 in. each. 

C. Bend to fit socket of the above pipe. 

D. Mouthpiece with louvre front to fit on 

bend. One of these, 6 in. long, is 
supplied with each range. 
Increased heating surface for warming 
the fresh air is provided by means of 
a grating inside the socket at £. 



Fig. 26. 

The principle of these arrangements for utilising to some 
extent the heat in the chimney has been adopted for barracks 
in the case of grates for married soldiers ; these would be 
useful as cottage grates. They have a small oven, and an 

on Ventilation, 133 

open fire ; warmed air is introduced into the room by means 
of an iron flue carried up from the fire-brick lining of the stove 
inside the chimney, and introduced into the room near the 
ceiling through a louvred opening ; by this means the heat of 
the smoke is utilised to warm fresh air admitted to the room, 
and prevent draughts. 

This description of grate was devised for the purpose of 
combining a sufficient power of cooking for a cottage with 
great compulsory economy of fuel (see Fig. a6). 

It must however be observed, that in proportion as the 
heat is removed from the chimney, so is the draught, i.e. the 
effect of the chimney as a pumping-engine to remove the air, 
diminished, and the combustion of the fuel to some extent 

The limit to which the heat from the fire can be utilised 
will be the point at which a sufficient amount of heat is left in 
the chimney, to cause an adequate draught, so as to ensure 
the combustion of the fuel. 

With these ventilating fireplaces in temperate weather, 
when windows or other inlets are opened direct from the 
fresh air, the entrance of warmed air will be checked, whilst 
the ventilation of the room would be continued ; and if 
desired, the inlets for warmed air may be provided with 
valves to be closed when the fireplace is wanted rather for 
ventilation than for warming. 

Numerous experiments have been made on the ventilating 
fireplace for barracks. These experiments show that the air is 
generally admitted into the rooms at a temperature of about 
ao°, or from that to 30° Fahr. above that of the outer air. The 
design of the grate wa,s intended to preclude the possibility of 
such a temperature as would in any way injure the air intro- 
duced ; and the experiments made by the late Dr. Parkes in a 
hospital ward at Chatham, in April 1864, illustrate the hygro- 
metric effect with the grate in use. The difference between 
the dry and wet bulbs in the ward varied from 8° to 5°j viz. 

134 Action of Open Fireplaces 

on the 17th of April, 8-5°; on the i8th, 6° ; on the 19th, ^-^ \ 
on the aoth, 6*5° ; on the 21st, 5«o°. On examining the record 
of the dry and wet bulbs, no evidence was seen at any time 
of any unusual or improper dryness of the atmosphere. The 
difference between the two bulbs was certainly always greater 
in the ward than in the outer air, but the difference was not 
material. The temperature of the rooms was invariably found 
to be so equable, that when the grate was in full action, and 
the windows and other means of ventilation closed, thermo- 
meters placed in different parts of the room, near the ceiling 
and floor, in corners furthest from the fire, and on the side 
nearest to it, but sheltered from the radiating effect of the 
fire, did not vary more than about i°Fahr. The variation of 
temperature in different parts of a room warmed by a fire, by 
radiation, without the action of warmed air, will be found to 
be from 4° to 6° Fahr., and sometimes even much more in 
cold weather. 

General Morin, with the object of utilising the grate as the 
sole means of ventilation for a room, lays down the principle 
that the whole of the air shall be renewed five times in the 
hour. To perform this effectually, it is necessary that the 
area of the chimney outlet shall afford about one square inch 
of area for every lOO cubic feet of content of the room, and 
that the area of the fresh air inlet at its opening into the 
room should be enlarged to afford about 14 square inches for 
every 100 cubic feet of content of the room ; so as to prevent 
the air entering with a rapid current. But on an average this 
quantity of air is more than is necessary. The Barrack and 
Hospital Improvement Committee's proposal would resolve 
itself into this — viz. that the air in barrack-rooms should be 
completely changed about twice in an hour (inasmuch as they 
required a cubic space of 600 cubic feet per man), and for all 
ordinary purposes this would probably suffice; as, however, 
this proposal was based on a limited number of occupants, 
vith a more crowded room the amount must be increased. 

on Ventilation, 135 

General Morin's experiments in 1864-5-6, with fireplaces 
constructed in the ordinary manner, and with others on the 
plan above described, in which the chimney was utilised for 
warming the air, showed that whilst with an ordinary fireplace 
the heat which is utilised in a room a,mounts only to one eighth 
of the heat given off by the coal, or '125, in these fireplaces 
the heat utilised in the room was '355 of the heat given off 
by the coal, or one third ; therefore, to produce the same 
degree of warmth in a room, this grate requires little more 
than one third of the quantity of coal required by an ordin- 
ary grate. The ventilation of the rooms was at the same 
time effected by passing a volume of air through the room 
in one hour equal to five times the cubic contents of the 
room. An equable temperature was maintained during the 
experiment. There was no perceptible draught, and although 
the doors fitted badly, scarcely any air was drawn in through 
the crevices. 

In conclusion, the merits which are claimed for this venti- 
lating fireplace are : — 

1. That it ventilates the room. 

2. That it maintains an equable temperature in all parts of 
the room, and prevents draughts. 

3. That the heat from radiation is thrown into the room * 
better than from other grates. 

4. That the fire-brick lining prevents the fire from going 
out, even when left untouched for a long time, and prevents 
the rapid changes of temperature which occur in rooms in cold 
weather from that cause. 

5. That it economises fuel partly by making use of the 
spare heat, which otherwise would all pass up the chimney, and 
partly by ensuring by its construction a more complete com- 
bustion, and thereby diminishing smoke. 

6. That it prevents smoky chimneys, by the ample supply of 
warmed air to the room, and by the draught created in the 
neck of the chimney. 



The open fireplace is convenient in ordinary living-rooms 
in this climate, because, the weather is generally so temperate 
that it is only exceptionally necessary to provide much 
artificial warmth. In climates with colder winters the open 
fireplace becomes expensive, and is used rather as a luxury, 
because every room and corridor requires to be warmed, and 
if this is done by open fireplaces the carrying of the fuel alone 
in such cases becomes a serious inconvenience. 

In order to ensure an adequate change of air in cold 
climates when there is no open fireplace, and when the 
weather requires closed windows, and in hot climates when 
the movement of air is very small, the fresh air ought to be 
^ introduced or extracted by air-flues connected with fans, 
pumps, or heated extraction-shafts. 

This, however, has not hitherto always been the practice. 
In many cases, rooms are warmed by close stoves on the 
French or German plan, or by hot water pipes, without any 
arrangements for introducing fresh air. On the other hand, 
when the warming is effected by an inflow of warm air, the 
air is generally brought in by special channels from a heating 
apparatus placed in a central part of the building. This latter 
plan is much used in the United States, and in Canada, where 
the winters are very cold. The warmed air is supplied (gener- 
ally at a high temperature) from a heating apparatus which is 
usually placed in the basement or lower part of the house. 

Ventilation combined with Wai^med Air. 137 

The effect of introducing warm air at the lower part of a ward 
on each side, and allowing it to escape at the top, was illustrated 
by a series of careful experiments, by Professor E. Wood of 
Harvard University, upon the various currents prevailing at 
the same time in a hospital ward at Boston, U. S. (Fig. 27.) 

The ward was 96 feet long by 26 feet 3 inches wide, with 
seven opposite windows, the tops of which were 16 feet above 
the floor-level, and 14 beds on a side. The ceiling was arched, 
20 feet high in the centre, and averaged a height of about 18 
feet, without any obstructing cornices. The floor-space of the 
beds was 88»65 square feet, and the cubic space per bed 1629 
feet. The fresh air was admitted warmed at a temperature 
of 90° by openings i foot square each, under each window, 
just above the floor-level, 14 openings in all, and the vitiated 
air was extracted by live openings, each ^x6 feet in the 
length of the ward in the centre of the ceiling. The hourly 
supply per bed was 9000 cubic feet. The high temperature 
of the inflowing air caused it to enter with velocity. This 
velocity was soon lost. There was an upward movement of 
air throughout the ward, but a comparative stagnation in the 
centre, for a height of about 8 feet from the floor-level — and 
also a comparative stagnation commencing over the bed-heads 
and extending to the upper part of the ward, as shown in the 
diagram. The respiratory impurity, as shown by excess of 
COji over outer air, at the lower level was '00131, and at the 
upper part '00240 — thus showing, that with this construc- 
tion of ward and system of ventilation, there was a want of 
adequate circulation of air, as well as a disadvantage in the 
height above the top of the windows, which were 14 feet in 
height ; openings at the sides higher up would have per- 
mitted the foul air to escape more rapidly from the room, 
instead of occupying the waste space above that level at the 
risk of cooling and falling again and remixing with the air of 
the room. This example shows that extraction-shafts should 
be so placed as to cause a circulation of air such as is effected 

138 Ventilation combined with Warmed Air 

External Air 33° Fahr. 
Air Entering 91° to 93°. 

at Floor 67° 
„ at Head of bed 68^. 
„ at Ceiling 75^. 
,, at Ventilating Chamber 
81 -6. 


Hot water pipes to warm in- 
flowing Air. 

Inlets for inflowing Air. 

Outlet for Vitiated Air. 

Velocity in feet per second. 

The shade shows the appar- 
ent course of the Fresh Air. 

Fig. 27. Experiment in a hospital ward at Boston, U. S., by Professor E. Wood 

of Harvard University, 

in Rooms without Open Fireplaces. 139 

by a fireplace, which acts to cause a circulation, and to extract 
the air. In England the system is often resorted to in asylums 
and large houses, of warming the air, and supplying the warmed 
air by flues to different parts of the building ; this should be 
combined with some plan for the extraction of the vitiated air. 

The method of warming rooms by close stoves or by hot 
water pipes in the room, do not of themselves necessarily 
entail any change of air. 

In Germany and the northern parts of Europe, where close 
stoves are used, the stove is generally filled with fuel once 
a day at an opening often outside the room, and no removal 
of air takes place by means of the stove. Therefore, where 
there are no special means for changing the air, the rooms 
become close and unhealthy ; but in very cold weather, when 
there is a great difference between the temperature indoors 
and out, a considerable change of air is effected through 
crevices of doors and windows, and through every available 
aperture ; moreover a spontaneous change of air will take 
place through the walls, when the inside temperature greatly 
exceeds that out of doors. 

Dr. Bohm (than whom no one has better studied ventila- 
tion) has adopted for some years the following system in the 
Rudolf Hospital at Vienna, where the wards are heated by 
close stoves. He there warms fresh air by means of passages 
constructed in the fire-clay stoves, placed within the ward, and 
the fresh warmed air passes into the ward from the top of the 
stove. He provides flues of a large size, and proportioned to 
the size of the ward, carried from the level of the ward floor 
to above the roof, inside which the flue from the stove is 
carried, and in cold weather the difference of temperature 
between the air in the flue and the outer air causes a suf- 
ficient current in these flues to ventilate adequately the ward. 
By this means the fresh warmed air, instead of passing off 
to the upper part of the ward, and thence away by flues, 
is made to circulate towards the floor of the ward, thus 

140 Ventilation combined with Warmed Air 

bringing into action one of the conditions of the open fire- 
place ; each ward is self-contained in respect of its warming 
and ventilation, as is the case when an open fireplace is 
used. This is for winter ventilation. 

In warm weather, when the stove is not in use, an outlet 
into the extraction-shaft is opened by means of a valve near 
the ceiling, so as to allow of the escape of the air from the 
upper part of the room, in the manner already mentioned ; 
the flues in the stove remain as inlets, and other inlets direct 
from the open air are also provided at the floor-level and 
elsewhere, and the windows can be opened if desired. 

There have been numerous close stoves invented of late years 
to supply fresh warmed air ; in France the system of hot air 
stoves prevails extensively, and it is being gradually adopted 
in this country. These provide for the admission of fresh air ; 
but care is not always taken to provide both for the removal 
of the vitiated air and the supply of fresh air ; and without 
provision for both good ventilation is impossible. 

In this climate sufficient ventilation can generally be 
obtained in private houses, in hospital wards, in barrack- 
rooms, and in workhouses, when the proportion of floor-space 
allotted to each occupant is comparatively large, by providing 
extraction flues and inlets as already described for the pur- 
pose of taking advantage of the difference between the tem- 
perature indoors and the temperature outside, and by a 
judicious use of open windows. But in prisons, asylums, and 
buildings occupied by persons under strict rule, as well as in 
buildings where large numbers are congregated together for 
a limited time, such as theatres, churches, legislative assem- 
blies, public meetings, crowded dining rooms or smoking 
rooms, and otherwise, cidequate ventilation can as a rule 
only be secured by the adoption of some means for extract- 
ing the air, beyond that which would be provided by the 
simple form of outlet and inlet above described. 

This may arise from various causes. For instance, in densely 

in Rooms witkouf Open Fireplaces. 141 

occupied buildings, where the volume of air to be removed is 
large, and the building complicated in shape, dependence on 
the difference of temperature and the action of the atmosphere 
alone might require inlets and outlets of an inconvenient size, 
and therefore some power of extraction to accelerate the 
current is necessary. There are also many other cases in 
which it is necessary to resort to mechanical means for 
changing the air. In all such cases, ventilation must be com- 
bined with warming, on the score of convenience as well as 
of economy. 

In hot climates the conditions differ materially from those 
already alluded to. The difference of temperature indoors and 
out is rarely sufficient to produce an adequate change of air. 
The object to be attained is the reverse of that sought in this 
climate. The fresh air to be supplied should be cooled down 
below its outside temperature instead of being raised above it. 

In hot climates it will therefore generally be necessary to 
provide some artificial means of removing the air, in order 
to secure an adequate change of air in rooms or halls occupied 
by many people. 

Ventilation by mechanical means may be effected either by 
propulsion or by extraction. By propulsion fresh air would 
be forced into the building to drive out the vitiated air. To 
effect this, Dr. Arnott proposed an apparatus on the principle 
of the gas-holder, by which he could carefully regulate the 
quantity of inflowing air. In the House of Commons there 
is an apparatus on the principle of a large pump, for oc- 
casional use, by which in the summer a given quantity of 
cooled air can be forced into the House. At the Hospital 
Necker at Paris, fans were used for forcing a given quantity 
of air into the wards. 

It is probable that a system of propulsion might be found 
to be the more convenient system for freshening the air in 
rooms in hot climates, combined with openings in the upper 
part of the room. The air might be compressed in the course 

142 Ventilation combined with Warmed Air 

of the propulsion in impervious underground channels, in 
which it would be retained sufficiently long to be cooled down 
by the lower underground temperature, and in passing into 
the rooms or wards it would be still further cooled down by 
the expansion. In towns where the subsoil is liable to be 
very polluted greater care must be exercised to prevent the 
ground air from mixing with the air for ventilation. 

But, as has been already observed, experience shows that a 
system of propulsion has not acted satisfactorily in hospitals 
in this climate, unless it has been combined with a system for 
the extraction of the vitiated air. And under ordinary cir- 
cumstances, where extraction is resorted to, an adequate 
quantity of fresh air can be drawn in by the operation of 
the extraction-shafts, without the additional expense and 
trouble of propulsicJn. 

In the case of ventilation by extraction, it is generally 
found convenient and economical to provide one principal 
extraction-shaft, to which the extraction-flues from the sepa- 
rate apartments are brought; and to warm the fresh air at 
some central apparatus before it is drawn into the rooms. 
The effect to be sought for in a room from the use of 
extraction-shafts is as follows : — 

Currents created by outlets placed near the floor-level and 
led into extraction-shafts must be arranged to move only at a 
low velocity into the outlet, in order to prevent draughts : the 
temperature of air as it enters the outlet is not affected, as is 
the case with the open fireplaces ; therefore the tendency to 
create a circulation of air in the room does not prevail to the 
same extent as with an open fireplace ; and if a room with- 
out an open fireplace is to be equally efficiently ventilated by 
extraction-shafts alone, the advantages which an open fire 
presents must be compensated for by placing the outlets and 
inlets in such positions and in such numbers as will prevent 
stagnation of air in any part of the room. In cold weather 
the extraction should generally take place from the lower 

in Rooms without Open Fireplaces, 143 

part of the room, so as to draw off the cold air from near 
the floor, and cause the warm air above to circulate. 

In this case the orifices for extracting the air should be 
raised above the floor sufficiently to prevent dirt falling into 
them, and it is advisable if possible not to place them near 
the feet of persons sitting down in the room, because the con- 
tinual flow of air, even at a low velocity, produces an evapora- 
tion which in time may create a feeling of cold. It may be 
generally said that the larger the area through which the air is 
drawn off the less will any unpleasant effects be experienced. 

Thus in a lecture-room the vertical risers of the seats may 
be advantageously made entirely of open grating, so that the 
whole area is available for drawing off the air. 

On the other hand, in rooms with many occupants, where 
much gas is used, in hospital wards, in smoking rooms and 
dining rooms, or other places under similar conditions of the 
vitiation of air, the heated air at the top of the room should 
be rapidly removed by special outlets near or in the ceiling, 
in addition to extraction at the floor-level ; care being taken 
that both sets of outlets shall draw efficiently. The heat 
which results from gas affords of itself a means of extrac- 
tion, if judiciously applied. 

The supply of fresh air should be placed above the heads 
of the occupants of the room, and admitted through inlets 
sufficiently far from the extracting outlets to prevent its being 
carried away at once. 

The construction of the large cathedrals lends itself to 
ventilation ; because the lofty nave with clerestory windows, 
and the towers, are convenient means for the removal and 
admission of air, and if the windows were all opened after a 
service, very little inconvenience would be experienced even 
from the largest congregation. But there are many churches 
in this country where large congregations are crowded into 
buildings of inadequate size and height, and these require 
special care. 

144 Ventilation combifted with Warmed Air 

In churches with galleries, densely occupied, and in halls 
for meetings, the air should be drawn off from the top as well 
as from below. The upper air, in cases of crowded meetings, 
is filled with much moisture from the breath ; it is therefore 
advantageous to warm the air thus drawn off as it enters the 
extraction-shafts, in order to increase its capacity for mois- 
ture, so that the moisture shall not be deposited before it 
passes into the outer air. 

Theatres present peculiar difficulties of ventilation, from 
their • complex arrangement. The entrance, staircases, and 
corridors should be warmed independently of the part oc- 
cupied by spectators. These require but little ventilation. 
The part occupied by spectators is under different conditions 
when the curtain is down and when it is raised ; therefore the 
stage should be warmed and ventilated independently of the 
part reserved for spectators. As the latter part requires a 
large amount of light, and as light is inseparable from heat, 
the light should be utilised for promoting ventilation. The 
heated air should be drawn off above, and the fresh air duly 
warmed in winter should be admitted at levels on which 
spectators are placed, without draughts. Adequate venti- 
lation in a theatre will, as a rule, be more satisfactory, if some 
mechanical appliance be provided to force in the fresh air as 
well as to remove the vitiated air. In summer the ventilation 
requires to be modified so as to supply air cooled either by 
compression or otherwise, as may be found most advanta- 
geous. The effectual ventilation of a theatre requires the 
presence of a special attendant to watch continuously the 
condition of the air in the theatre, and modify the arrange- 
ments from time to time. 

The object of this treatise is not to define closely special 
methods of ventilation; but rather to bring into view the 
principles which should guide the architect or engineer in 
considering the system most applicable to the requirements of 
his particular case ; because the enunciation of specific rules 

in Rooms without Open Fireplaces. 145 

in so complex a subject might fetter him and prevent advance- 
ment in sanitary science. 

The following instances of some of the systems of ventila- 
tion which have been adopted in various buildings will afford 
a general idea of the principal points which have to be con- 

The old Roman plan, by which the source of warmth was 
uniformly distributed by heating the floor and walls of a 
room, gave an equable temperature; and by warming the 
floor and walls rather than the air partially provided one of 
the advantages arising from the open fire. 

If in connection with this system fresh cold air were ad- 
mitted near the ceiling, and if the warmed vitiated air were 
removed by outlets near the ceiling, and if each occupant had 
a sufficient floor space, it is probable that this would form a 
very comfortable arrangement. 

The Roman floor was a floor of tiles laid hollow with flues 
underneath. The furnace was formed at one end, and the 
smoke from the furnace was carried backwards and forwards 
along the flues under the floor, and finally led up the walls or 
to a chimney. This chimney was generally so placed as to 
act as a means of keeping the store of wood dry. 

But such a plan could only be adopted in a specially 
constructed building. It would not be convenient in ordinary 

An experiment was made to compare the economical effect 
of warming by means of a heated floor with the effect of heat- 
ing by means of a ventilating fireplace; the experiment lasted, 
with each mode of warming, for two days. It showed that, in 
the case of the warmed floor, the room was maintained at a 
temperature of about 18 degrees above the temperature of the 
outer air with an expenditure of 56 lbs. of coal and iialbs. of 
coke, whilst with the ventilating fireplace the expenditure was 
only 75 lbs. of coal ; the cost being 3^. 4^. for the warmed floor 
as compared with i^. 4^. for the ventilating fireplace. It is 


146 VepJilatton combined with Wanned Air 

probable however that in a house designed and constructed' 
specially for this mode of heating, better results would have 
been obtained. 

The ventilation of Derby Infirmary, designed and carried 
out more than sixty years ago is an instance of ventilation 
by natural extraction and propulsion alone. 

Flues were led from each ward to a central shaft, at the top 
of which was placed a cap with a vane arranged so as always 
to cause the opening to face away from the wind. Fresh 
air was introduced by means of another shaft, also furnished 
with a cap with a vane arranged to make the opening 
always turn towards the wind like a wind-sail in a ship. The 
fresh air entering this latter shaft was carried down under- 
ground along a channel 70 feet long to an iron stove covered 
with flanges, where it was heated in cold weather ; and thence 
it passed up by separate flues to the wards. The extraction- 
flues were led from apertures near the ceiling, and the wards 
were also provided with open fireplaces near the floor-level. 
Thus the action of the wind was employed to extract the 
vitiated and to propel the fresh air. The amount of air 
which was thus passed through the wards with a velocity of 
wind of about 3 miles per hour, was recorded as 400,000 
cubic feet per hour. But of course on perfectly calm days the 
effect of the extraction-shaft was limited to the extraction 
force resulting from the diflference of temperature and the 
height of the shaft; this force was necessarily much modified 
by the friction from the length of the flues in addition to the 
delays caused by the horizontal portions of the flues. Straight 
flues carried up separately from each vvard would have been 
more efficient for extraction. The underground channel had 
the effect of partially warming the fresh air in winter and 
cooling It in summer.^ 

The Great Opera at Vienna is ventilated by extraction- 

* This system was invented by G. W. Sylvester and William Strutt, but recent 
alterations have abolished it. 

in Rooms without Open Fireplaces. 147 

shafts carried from round the ceiling, from above the central 
chandelier and from the upper part of the boxes and galleries. 
These are brought into action by the heat generated by hot 
water boiliers and by the products of the gas chandelier ; the 
admission of air is regulated by propulsion by a fan from 
below. The air thus supplied is partly passed over pipes 
heated by hot water in winter, and is partly left cold. 
Channels for hot air and channels for cold air are provided 
to all parts of the theatre, under each occupant of the pit 
and stalls, and in the corridors for the supply of the boxes, 
in addition to a large volume of air admitted near the stagey 
so as to keep the ventilation for the spectators separate from 
that for the actors. In hot weather the ait can be cooled by 
being passed through spray, tn order to nlaintain the venti- 
lation in efficient action, metal thermometers are placed in 
different parts of the theatre, all connecting by electric wires 
with a small office in the basement, and these indicate the 
temperature in different parts of the building at each moment 
on a board in the office. When the observer sees that the tem- 
perature is raised or falls unduly, in any part of the house, he 
can open or close valves, which are under his hand, and which 
regulate the quantity of warm or cool air admitted to each part 
of the house, and thus alter the temperature as desired. 

The New Opera at Paris is warmed by hot water apparatus 
for all the parts behind the stage, on the assumption that this 
method does not dry the air so much, whilst the part of the 
theatre occupied by the public is warmed by stoves, as being 
more rapidly put in action. 

The fresh air is introduced through flues in the floor of the 
several tiers of boxes, and the vitiated air is extracted partly 
from the back of the boxes and partly from the floor of the 
pit and stalls. The heat from the lights is not applied to 

Figure 38, on the next page, shows the system in use in the 

Houses of Parliament. 

L % 

148 Ventilation combined with Warmed Air 

Fig. 28. Venlilalion of the Housi! of Commons. 

in Rooms without Open Fireplaces, 149 

The Houses of Parliament are warmed and ventilated by a 
combined arrangement. For the House of Lords and the 
House of Commons the arrangements are identical. 

The fresh air is admitted into a lower chamber, where it 
is warmed by pipes. Through these pipes steam circulates ; 
broad vertical flanges are cast at intervals on the pipes 
in order to increase the heating surface. Each of these 
flanged portions of pipes has therefore a similar heating 
power. The frequent alteration in the number of occupants 
of the House of Commons makes it necessary frequently to 
alter the temperature of the air. This alteration is effected 
by an assistant, who watches the thermometer and covers with 
pieces of woollen fabric, or uncovers, a greater or less number 
of these flanged portions of the pipes, so as to diminish or 
increase the heating surface as may be required. 

The fresh air is supplied to this chamber from the adjacent 
courtyards, which are covered with asphalte and kept tolerably 

The air, after having been warmed in the lower chamber, is 


passed through four large circular openings of about three feet 
six inches diameter each, into a chamber above : this chamber 
is practically a portion of the House of Commons, and is 
separated from it only by means of an iron grated floor 
covered with open matting. ' The grated floor and the grated 
risers of the steps on which the seats are placed are all avail- 
able as inlets. The glass ceiling of fhe House of Commons 
has openings into the space in the roof, from which a channel 
leads down under the basement to the foot of the clock-tower, 
where a large fire is maintained ; and this forms the exhaust 
by which the air is drawn through the house ; as much as 
1,500,000 cubic feet have been passed through per hour. If 
the house were full, this would represent somewhat under 
aoGO cubic feet per occupant per hour, including members, 
attendants, and strangers. The lighting is effected by gas- 
lights placed above the glazed ceiling. 

150 Ventilation combined with Warmed Air 

In summer an arrangement has been adopted, by which air, 
made cool by passing over ice, has been forced into the house. 
This is an expensive process for cooling the air. Equally 
good results at a much less cost might be obtained by utilis- 
ing the underground temperature in summer, or by using 
compressed air. The air is drawn from the ground-level, 
which is an undesirable arrangement. Much purer air could 
be obtained by bringing the air down from the lofty towers 
which form part of the structure. 

The French Legislative Chambers were ventilated by 
General Morin on a plai> the reverse of that adopted in the 
Houses of Parliament. The Legislative Chamber has a cubic 
content of about 400,000 cubic feet ; the arrangements are 
intended to provide for 1,060,000 cubic f^et of air to pass 
through the chamber in an hour. The air is extracted 
through openings at the floor-level, and others placed in the 
vertical risers under the seats, in the body of the hall and in 
the galleries ; there are also outlets in the staircases and 
corridors of approach. The united area of the extracting 
outlets is nearly 160 square feet, which would allow of a 
velocity of i-8 feet per second in supplying the regulated 
quantity of air. 

The air is admitted at the ceiling-level through openings in 
the cornice all round the chamber and in the capitals of the 
columns. The total area of inlet is about 195 square feet, 
which would allow of a velocity of about i«.5 feet per second 
for the inflowing air, with the maximum supply, which how- 
ever is rarely attained. 

The fresh air is brought down by a flue from a height of 60 
metres into a chamber at its base, where it is warmed to the 
necessary temperature by four (or fewer) stoves of brick 
provided with air-passages. The adjacent committee-rooms, 
and the rooms and corridors for the use of members, are 
warmed and ventilated in connection with the chamber, or on 
the same principle — the total amount of air being 1,600,000 

in Rooms without Open Fireplaces. 


Cubic feet per hour. The difficulty of maintaining a uniform 
temperature in the chamber arises from the continual varia- 
tion in the number of occupants. The temperature main- 
tained in the winter is about 64^° Fahrenheit, with a variation 
seldom exceeding 3° or 4°. In spring, with an outside tem- 
perature of 57°, the inside temperature has rarely varied 

Fig. 29. Sir Joshua Jebb*s system of Ventilation for Prisons. 

more than from 64® to 68°. In summer, with an external 
temperature of 82°, a temperature has been maintained in 
the chamber of from 73° to 77°. 

The air being brought first into the vaults below the ground, 
its temperature is raised to begin with by the temperature 

152 Ventilation combined with Warmed Air 

of the vaults ; in winter as much as from 9° to 13° above the 
outside air; in summer it could be cooled down as much 
as from 16° to 18° below the outer air, but this would be 
beyond what would be endurable in practice. 

The temperature in the main extraction-shaft averages from 
36° to 45° above that of the outer air ; but in summer, with 
an out-door temperature of 83°, it was found necessary to 
raise that of the main extraction-shaft to about 142°, or a 
difference of 59°. Experiments were made to ascertain 
whether, in order to gool down the chamber in the great 
summer heats, it was desirable to continue the ventilation 
through the night, but the results did not correspond with the 
extra expense. 

The system of ventilation of cells in prisons adopted by 
Sir Joshua Jebb was practically the same as the system 
originally proposed by Sylvester and General Morin, viz. the 
extraction of the air at the lower part 
of the cell and its admission near the 
ceiling (Fig. 29). In cold weather this 
t method of admitting the warmed air 
1 keeps the temperature of the cell uni- 
form. When the extraction- current is 
sluggish, as may be the case in warm 
weather unless much extra fire is used, 
this plan of removing the air from below 
does not always sufficiently relieve the 
cell unless the window is opened. A 
, better effect will be produced by the 
1 removal of air from the upper part of 
the cell. The most convenient arrange- 
ment would be to have a shaft as shown 
in the diagram (Fig. 30), with an opening 
■S' l°- tQ be closed alternately either at the top 

or the bottom according to the weather. In lai^e prisons the 
most economical arrangement is to warm the fresh air at some 

in Rooms without Open Fireplaces. 153 

central source of heat and distribute it thence over the build- 
ing. The extraction must either be mechanical, or by means of 
a heated shaft, because these prisons have large central halls 
into which the cells all open, and which would disturb the ven- 
tilation in the absence of mechanical extraction. Care should be 
taken that the flue from each cell shall have a sufficient length 
before it enters the main extraction-shaft, to ensure that there 
are no reverse currents, for these would bring impure air back 
into the cells. This arrangement requires that the fire for the 
extraction be kept up day and night. For cells on one floor, 
such as barrack or police cells, where the number is small, the 
simplest and most efficient arrangement is to provide a small 
shaft for each cell direct to the open air, with an opening near 
the floor for winter and near the ceiling for summer, one valve 
being arranged to be closed when the other is open, the cells in 
this case being warmed by hot-water pipes in coils, or flanged 
pipes, behind which fresh air is admitted into the cell so as to 
come in warmed when the pipes are in use. Direct openings 
from such cells into a general extraction-shaft carried along 
the ceiling of the several cells are objectionable, because under 
such an arrangement the impure air from one cell frequently 
passes into the adjacent cells. If such a general extraction- 
shaft be provided, the least dangerous plan would be to place 
it at .a height of at least 6 feet above tlie ceiling of the cells; 
and to carry a small shaft from each cell vertically up for a 
few feet in length into it ; a current being maintained by 
mechanical means, or by heat in the shaft. 

In the Herbert Hospital the warming of the wards is 
eff*ected by open fireplaces, whilst the subsidiary accom- 
modation is warmed by hot-water pipes. The administrative 
offices are in a separate building from those which contain 
the sick. 

The wards are on the pavilion principle, with windows on 
opposite sides, the beds being placed between the windows. 
Each ward has its nurses' room and ward-scullery at one 

154 Ventilation combined with Warmed Air 

end, and its ablution- and bath-room and water-closets at the 
other. Every water-closet has a separate window, and the 
ventilation of these ward offices is most carefully cut off from 
that of the wards themselves. The walls are built hollow, 
and the windows are glazed with plate-glass, to save the heat, 
which the large extent of wall and window surface would 
otherwise allow to escape. The ward walls are of parian 
cement, the ward floors of oak, to be kept polished with 
beeswax. The ventilation of all the wards is as fol- 
lows :— 

First the windows open at top and bottom on both sides, 
then shafts, 14 by 14, pass up at each corner of the ward, to 
above the roof, to allow of the escape of foul air. Sherring- 
ham's ventilators are placed in each wall-space, between the 
windows, to admit fresh air when the windows are closed. 
For cold weather the principal engine of ventilation is the 
fireplace. There are two in each ward, placed in the centre 
line of the wards. The flue passes horizontally under the 
floor to a vertical flue in the wall. Fresh air is admitted by 
the side of the horizontal flue, through openings in the fire- 
clay sides of the fireplace, into the ward, by which means it 
becomes warmed ; so that each fireplace, when the fire is 
lighted, pours a continuous supply of fresh warmed air into 
the ward, and a great portion of the heat, which would other- 
wise pass into the chimney, is saved. The ablution-rooms 
and water-closets, as well as the lobbies which separate the 
ablution-rooms and water-closets from the wards, are each 
warmed by a separate service of coils heated by hot water 
from central boilers, and ventilated by fresh air admitted 
through the coils of these hot-water pipes. The vitiated air 
escapes by means of a separate shaft carried up through the 
roof from each ablution-room, water-closet, and lobby. The 
staircases, as well as the corridors separating the pavilions, are 
each warmed independently by hot-water coils, and ventilated 
by fresh air admitted through the coils. Each part of the hos* 

in Rooms without Open Fireplaces. 155 

pital is thus independent of every other part for warming and 

These instances sufficiently elucidate the general principles 
affecting combined warming and ventilation. 

In all systems in which the walls derive their heat from the 
air of the room only they are necessarily somewhat colder 
than the air of the room. It has been mentioned that one of 
the main causes of the comfort of the open fireplace is that 
with it the air of the room derives its warmth from the walls 
which are warmed by the rays of the fire, and therefore that 
the walls are at least of the same temperature as the air ; 
stoves and hot-water and steam pipes in a room also radiate 
^ portion of their heat to the walls, those heated to a low 
temperature less effectively than those heated to a high 
temperature. Heating effected by warmed air can only be 
thoroughly comfortable when its use is combined with some 
plc^n of warming the floors, walls, and ceiling, so that their 
temperature may not be dependent on that of the air after it 
has entered the room. 

In conclusion, no system of ventilation and warming in a 
large building or establishment can be satisfactorily conducted 
unless some person is charged with the duty of seeing that it 
is maintained at all times in effective action, on the principle 
of that adopted at the Opera in Vienna. 




The laws regulating the movement of air should govern the 
form of buildings. 

Plans for houses, barracks, asylums, schools, and hospitals 
necessarily vary either with the wants of the individual or 
with public requirements, with the site, the aspect, and the 
cost. It is therefore impossible to do moire than sum up the 
general principles which should be observed in a design. 'But 
in both private and public buildings arcliitects should conform 
their architectural design to the internal requirements, and 
not, as is too often the case, make the internal arrangements 
conform to the design of the fa§ade. 

Fire-proof buildings are desirable; a planter covering to 
woodwork is a very good protection against fire. In the 
Communist riots in Taris, wood covered with plaster was 
charred, not consumed. 

Fires frequently spread with rapidity in English houses 
because currents of air generated by a fire are enabled to pass 
up from one -floor to another behind the boding of windows, 
the battening on walls, the lath and plaster partitions, &c. 
The free passage of air-currents should be stopped between 
every room and the room under it and over it, by means of 
some substance difficult of combustion. There should also be 
a course of fire-clay slabs (Fig. 31), projecting at least 

lolid Ebb of Fii 
Fig. 31- 

The Internal Arrangements of Buildings. 157 

6 inches all round the walla of a room to receive the beams 
of the floor above, and against which the plaster ceiling 
should abut. Such arrangements 
would greatly check the passage of 
fire from story to story in a house. 

There should invariably be circu- 
lation of air round every building. 

Back-to-back dwellings, as well as 
houses and cottages built with in- 
terlocking walls, are inadmissible, for 
reasons already explained. 

Dwellings should not be built over 
stables, because it is impossible to keep the air of a stable 
as pure as that of a living-room ; nor should dwellings be 
placed over stores or shops, where matters liable to putrefy 
are kept. 

To guard against the liability of impurity in the ground on 
which the house is built, when danger is apprehended from the 
presence of decaying organic matter, the most healthy plan 
would be to build the house over an arched floor raised above 
the ground-level, the space under the floor being open to the 
air on all sides. 

Basements are sometimes used in towns as dwellings : they 
are undesirable. There should always be an area between 
the basement and the street in houses situated in towns, to 
prevent emanations passing into the houses through the earth 
from defective gas-mains and sewers. Where cellars are pro- 
vided undei^ound, means should be taken to cut them off 
from the ground-air. 

Pure, dry air of the required temperature should pervade, 
every part of a dwelling. The, form of the interior of a 
dwelling should be such as to ensure this, and at the satne 
time prevent stagnation of air. In cold and temperate climates 
when the weather is warm, and in hot climates always, change 
of air largely depends on open windows and doors ; conse- 


Conditions affecting 

quently the relative position of windows and doors should be 
selected so as to enable the air of a room to be thoroughly 
changed when they are opened. 

Lofty rooms are advisable, because the height facilitates the 
change of air without draught, provided the windows or other 
means of outlet for air be arranged so -as to prevent stagnation 
near the ceiling, and so that impure and heated air which has 
risen to the upper part of the room may be removed rapidly. 
If time be afforded for its cooling, the impurities will fall and 
mix again with the air of the room. A lofty room, if warmed 
and provided with adequate means of change of air at the 
upper part, is more comfortable and healthy than a low room. 
In new houses or cottages, a height of ten feet should be the 
minimum height for the best rooms, and no room in a cottage 
should be under an average of eight feet high. 

An abundance of light, and in this country direct sunshine, 
is always necessary for maintaining purity of air. In a hot 
climate means must exist for intercepting the direct rays of 
the sun. 

A dark house is an unhealthy house, an ill-aired house, 
and a dirty house. Therefore light should penetrate to every 

part. There should be no dark 
staircases, corridors, corners, or 
closets. Direct light by means 
of windows easily opened to 
the outer air is required to 
ensure the frequent renewal of 
the air. Staircases, if lighted 
by skylights, should have the 
skylights made of the lantern 
form, with side-lights to admit 
fresh air without admitting rain. 

A hall or staircase in the centre of a house, carried up the 
whole height of the building, with light and ample ventilation 
by large windows at the top, forms a reservoir of air which 

Fig. 32. 

the Internal Arrangements of Buildings. 159 

may be kept fresh, and which materially assists in keeping 
pure the air of a house. 

Every room in a building should have access to light and 
air, by means of a window in an outside wall. The rooms 
appropriated to removal of refuse, such as housemaid's closets, 
and water-closets, require more light, and therefore propor- 
tionately larger windows, than other rooms. Store-closets 
should also have direct access to fresh air. 

The larger the proportion which the area of surface occupied 
by a house bears to the number of occupants the better. 

In towns where land is dear, and where a large number of 
persons are crowded on a given area, better ventilation and cir- 
culation of air may be obtained by placing dwellings on stories 
one above the other, and leaving spaces between the buildings, 
instead of in one-storied buildings which would be too close 
together to allow of circulation of air round the building. 

Under these circumstances, the height of a dwelling must 
be regulated with respect to its surroundings. That is to say, 
in the case of ordinary dwellings adjacent to each other, the 
distance apart of the dwellings should at least equal the 
height of the dwelling, so as to ensure adequate light and air 
in the lower floors. These considerations limit the number of 
stories in houses of the better classes in towns, or of an aggre- 
gation of cottages, each with its separate family, such as is 
created by a model lodging-house, where the arrangements 
prevent a community of air throughout the building. 

Basements should never be used for sleeping- rooms, nor 
indeed for human dwellings. They are always more or less 
liable to damp, stagnation of air, and deficiency of sunlight, 
and are well-known nurseries of disease. 

The number of stories in the case of barracks, workhouses, 
asj'lums, schools, and hospitals, where the conditions of the 
occupation entail a community of air throughout the building, 
must be more closely limited. 

i6o Condilions affecting 

The following conditions afford a general idea of the con- 
siderations which affect these buildings. 

In all such buildings, where a large number of human 
beings are to be lodged together, it is especially advisable, 
as a general principle, not to place anything which might 
injuriously affect the purity of the air in the same building 
with the inhabitants. 

Kitchens, latrines, ablution-rooms, and baths, should there- 
fore as far as possible be built away from them. 

Buildings should be arranged in the simplest manner. 

Squares with closed angles should be as far as possible 
avoided. The great object to be aimed at is to have free 
external ventilation all round the buildings ; in temperate and 
cold climates to have as much sunlight as possible, and to 
avoid a purely northern exposure for living-rooms. These 
conditions are essential to health. Free access of sunlight to 
a square is best obtained by placing two opposite angles of 
the square north and south. 

If the administrative conditions allow of it, the simplest 
arrangement for such buildings is in a single line, lying north 
and south if possible, to allow the sun to shine on both sides 
of the range every day. The line may be divided into separate 
blocks for facility of passing across it at different points. 

No part of any asylum, workhouse or barrack, whether for 
sick or healthy men, should be placed close to the boundary 
walls. There should be always intervening space sufficient to 
ensure thorough ventilation round the buildings between them 
and the wall, and to prevent the ventilation from being 
injuriously affected by buildings belonging to the adjacent 
population placed close" to the walls. Latrines, cook-houses, 
stores, and other similar buildings, can be placed between the 
building and the wall, but the arrangement should be such as 
not to interfere with the external ventilation of the building. 

All populous buildings where there is an intercommunity of 
air throughout the building, such as barracks, schools, work- 

the Internal Arrangefnents of Buildings, i6i 

houses, and asylums, are best constructed of only two stories 
of living-rooms. Three stories are not objectionable for healthy 
people, though very undesirable for sick. Four stories should 
only be resorted to when, from restricted dimensions or from 
the form of the ground, it is absolutely necessary to adopt this 
number of floors. 

- Rooms for dry stores or for administration, or rooms occa- 
sionally occupied, such as libraries and reading-rooms, may be 
placed without detriment on the ground-floor, the sleeping- 
rooms being placed over them, when necessary. Barracks and 
hospitals have this in common, viz. that the sleeping-rooms 
are mainly also living-rooms ; they differ in this respect from 
asylums, schools, and workhouses. The following rule should 
apply uniformly to all buildings of the nature of an asylum, 
barrack, hospital, and workhouse, viz. such buildings should 
be subdivided into separate houses, without direct communi- 
cation between the adjoining houses. To ensure this, the party 
walls between the houses should be carried above the roof. 

Each house should further be divided up the middle by 
a wide roomy staircase, extending from the ground to the top 
floor, with a free ventilation through the roof. The staircase 
and passages should extend across the house from front to 
back, with windows on opposite sides for through light and 
ventilation. Besides affording means of access, the stairs and 
passages should be so constructed as to afford ventilation 
upwards between the two halves of the house, suflScient to 
prevent the atmosphere in rooms on opposite sides of the stair- 
case and passages from intermingling. 

There should be a unit of size for all asylums, workhouses, 
and barrack-rooms, containing the principal rooms and theii" 
appurtenances, so that a building of any size, which has a 
definite object, may be constructed by simply increasing the 
number of such units. 

The unit of construction for workhouses and asylums has 
not been laid down by any distinct authority; for barracks 


1 62 Conditions affecting 

and hospitals the conditions have been laid down by the 
Barrack and Hospital Improvement Commission ; and for 
prisons they have been stated by the late Sir J. Jebb. 

The unit of construction for a school necessarily depends 
upon the nature of the school, and the number of classes into 
which it is divided. 

These conditions vary in almost every case. It may how- 
ever be assumed, that for school-rooms or lecture-rooms 
which are occupied for limited periods in the day-time, and 
thoroughly purified between times, a superficial area per 
pupil in the class-room of from 15 to 20 feet as a minimum, 
should be afforded, and a cubic space of not less than from 
180 to 250 cubic feet. This, at 1200 cubic feet per hour per 
occupant, would imply a renewal of the air of the room from 
5 to 7 times in the hour. But for a schoolroom occupied for 
limited periods, the windows being opened between times, 750 
cubic feet per hour per occupant during school hours would 
probably suffice ; reckoning after dark each candle as an occu- 
pant, and each gaslight as two candles. Windows in rooms 
where drawing or writing is carried on are best placed so as 
to be on the left hand of the student. 

The conditions of school dormitories would follow those in 
barracks and workhouse wards which have been already 
alluded to in a previous chapter. 

A short account of how the unit of accommodation far 
barracks and hospitals was arrived at will best explain the 
application of the general principles to special cases. 

For barracks, from 20 to 30 beds was fixed as a convenient 
unit of number for each barrack-room, the beds being arranged 
v^ ith their heads to the walls on opposite sides of the room. 

There should be about half as many windows as there are 
beds in the room ; they should be on opposite sides of the 
room ; they should be carried up to within a few inches of the 
ceiling, and be hung so that both upper and lower sashes can 
be opened or shut. 

the Internal Arrangements of Buildings. 163 

In no case should there be more than two rows of beds 
between the opposite windows. This rule holds good in all 
climates, but more especially in hot climates. 

Assuming a fixed floor-space, the width of the room must 
be such as to secure an adequate distance between the beds. 

There is a minimum limit of width which depends on the 
question of convenience. Nineteen or twenty feet would be a 
good width for a barrack^room in this climate, but in hot 
climates a larger floor-space would be required, which entails 
additional width and length. The width above mentioned 
would allow space for tables and forms when the beds are 
down, and would allow about 11 or 12 feet between the oppo- 
site beds during the day when the bedsteads are turned up. 

Barrack-rooms should never be less than la feet high. 

A room 20 feet wide and la feet high, with 5-feet bed- 
spaces along the walls, would give the regulation amount of 
600 cubic feet per bed. If the height of the room is less than 
I a feet, it would be better to make up the unit of cubic space 
by increasing the bed-space along the walls, rather than by 
increasing the width of the room. 

All sleeping-rooms should have ceilings. The space in the 
slope of the roof should not be taken into barrack-rooms any 
more than into the rooms of ordinary dwelling-houses. That 
space if in the room should always be ventilated ; otherwise it 
affords facilities for the impure air to stagnate, cool, and remix 
with the air of the room. 

The fireplace should be placed in the side wall in the 
centre of the length of one side of the room, and should be 
constructed to warm part of the air admitted for ventilation^ 
If the room were constructed for 30 beds, two fireplaces 
would probably be required ; in which case they should be 
placed on the same side of the room, to diminish the proba- 
bility of their smoking, and to assist ventilation. 

Each barrack-room should have a room for a non-com- 
missioned officer opening from the landing or passage, and 

M 1 

164 Conditions affecting 

connected with the barrack-room ; and either at the entrance, 
or at the further end opposite the entrance, there should be 
a well-lighted and well-ventilated room, with ablution-basins, 
and a bath. There should be in addition a water-closet and 
urinal for night use, with windows to the outer air. 

Hospitals require especial care, because the sick are more 
easily affected by insanitary conditions than persons in health. 

The unit of hospital construction is the ward, with its ward 

The area of the ward depends on the floor-space allotted to 
the patients, to which allusion has already been made, and 
varies with the climate, and with the object of the hospital. 
The breadth of the ward to some extent regulates the other 
dimensions. It is essential that the windows should be oppo- 
site, and, in order that they may act efficiently for changing 
the air, that they shall not be too great a distance apart. The 
breadth which combines convenience with this condition is 
from 24 to 28 feet, but 30 feet would not be too wide. With 
this form of ward, the beds would stand between the windows. 

The ward offices are of two kinds. 

1. Those which are necessary for facilitating the nursing 
and administration of the wards, as the nurse's room and 
ward scullery. 

a. Those which are required for the direct use of the sick, 
so as to prevent any unnecessary processes of the patients 
taking place in the ward ; as, for instance, the ablution-room, 
the bath-room, the water-closets, urinals, and sinks for empty- 
ing foul slops. There should, in addition to the bath-room 
here mentioned, be a general bathing establishment attached 
to every hospital, with hot, cold, vapour, sulphur, medicated, 
shower, and douche baths. Separate water-closets are re- 
quired for the nursing staff. 

Hot and cold water should be laid on to all ward offices in 
which the use of either is constantly required, in order to 
economise labour in the current working of the hospital. 

the Internal A rrangemenls of Buildings. 165 

The nurse^s room should be sufficiently large to contain a 
bed, and to be the nurse's sitting-room. It should be light, 
airy, and well ventilated. It should be close to the ward door, 
and it should have a window looking into the ward. 

There should be a ward scullery attached to each ward, 
and adjacent or opposite to the nurse's room, so as to be 
under her eye. 

There should be no dark corners in the scullery, and it 
should have ample window-space. 

There should be provided, adjacent to the scullery or nurse's 
room, well lighted rooms for linen, stores, or patients' clothes, 
and a hot closet for airing clean towels and sheets. 

The ward offices of the second class ought to be as near as 
possible to the ward, but cut off from it by a lobby, with doors 
at each end, and windows on each side, and with separate 
ventilation and warming, so as to prevent the possibility of 
foul air passing from the ward offices into the wards. These 
offices are therefore most conveniently placed at the end of 
the ward, furthest from the entrance and nurse's room, and 
distributed at each side, so as to enable the ward to have an 
end window. 

The ablution-room should contain a small bath-room with 
one fixed bath of copper, supplied with hot and cold water. 
A terra-cotta bath when once warmed has the advantage of 
retaining the heat longer than a bath of almost any other 
material, and of being always cleanly, but it absorbs a great 
deal of heat at first. Hence, when the bath is frequently 
used, it is the best material ; but if the bath is seldom used, 
then copper is better, or polished French metal. 

The water-closets should never be placed against an inner 
wall, but always against an outer wall of the compartment in 
which they are situated, and the soil-pipe should be carried 
down outside. 

Walls of ablution-rooms and water-closets should be 
covered with white glazed tile, slate enamelled or plain, or 

1 66 Conditions affecting 

Parian cement ; plaster is not a good covering, for them on 
account of their liability to be splashed, and of the necessity 
for the walls to be frequently washed down. 

The ablution-room and water-closets should have plenty of 
windows opening to the outer air. They should have shafts 
carried up to above the roof, to carry off the foul air, and 
ventilated openings to admit fresh air independently of the 
windows, and warmed air should be supplied to them inde- 
pendently both of the wards and of the lobbies which cut 
them off from the wards, which latter should also be carefully 
ventilated and warmed separately. 

These ward offices will vary but little with the size of the 
ward ; that is to say, a ward of twenty beds will require 
nearly as large ward offices as a ward of thirty-two beds. 


'-^■■f...f ."> f^ .*> ^ .«> fio .^ .^ ^ ,»oo " 

Fig 33. 

For instance, three water-closets per ward will suffice for a 
ward of thirty-two beds, but two at least will be required for 
even a twelve-bed ward. The superficial area to be added in 
the wards of thirty-two beds for these appliances would be 
about thirty square feet per bed, whereas in wards of twenty 
beds each it would come to nearly fifty square feet per bed. 

The ward with its ward offices is a small hospital, which 
may be increased to any required size by the addition of 
similar units. 

The principles upon which these units of ward construction. 

the hiternal A rrangements of Buildings, 167 

or, as they are generally termed, pavilions, should be added, 
are as follows : — 

I. There should be free circulation of air between the 

%, The space between the pavilions should be exposed to 
sunshine, and the sunshine should fall on all the windows in 
the course of the day, for which purpose it is desirable that 
the pavilions should be placed on a north and south line. 

3. The distance between adjacent pavilions should not be 
less than twice the height of the pavilion reckoned from the 
floors of the ground-floor ward. This is the smallest width 
between pavilions which will prevent the lower wards from 
being gloomy in this climate; and where from local con- 
ditions there is not a free movement of air round the buildings 
this distance should be increased. 

4. The arrangement of the pavilions should be such as 
to allow of convenient covered communication between the 
wards, without interfering with the light and ventilation ; 
and therefore the connection between adjacent pavilions 
should be on the ground-floor only, or in the basement, and 
the top of the covered corridor uniting the ends of pavilions 
should not be carried above the ceiling of the ground-floor 
wards. Indeed, whilst it is necessary to make the ground-floor 
wards from twelve to fourteen or fifteen feet high, it would be 
unnecessary for purposes of communication to give the cor- 
ridor a greater height than nine or ten feet ; there is how- 
ever this consideration, that if the top of the corridor is made 
level with the ward floors of upstairs wards, it affords a 
convenient terrace on to which the beds of patients can be 
wheeled, so as to allow them to lie in the open air. Each 
block of wards — that is, each pavilion — should have its own 

5. No ward should be so placed as to form a passage-room 
to other wards. 

6. As a general rule, there should not be more than two 

1 68 Conditions affecting 

floors of wards in a pavilion. If there are three floors or 
more, the distances between the pavilions become very con- 
siderable, because of the rule, which ought to be absolutely 
observed, of placing the pavilions at a distance apart equal to 
at least twice the height of the pavilion, measured from the 
floor-level of the ward nearest to the ground. Besides, when 
two wards open into a common staircase, there is, with every 
care, to some extent a community of ventilation. When 
there are as many as four wards one over the other, the stair- 
case becomes a powerful shaft for drawing up to its upper 
part the impure air of the lower wards, which is then liable to 
penetrate into the upper wards. Similarly, heated impure air 
from the windows of the lower wards has occasionally a ten- 
dency to pass into the windows of the wards above. For 
these reasons, no hospital should have more than two floors 
of wards, one over the other ; and if there is a basement under 
sick wards, it should not be used for' any purpose, such as 
cooking, from which smells could penetrate into the wards ; 
and, when possible, it is best not to continue the staircase 
into the basement. 

7. There is ^ limit to the numbers which should be congre- 
gated under one roof. This limit will depend very much on 
the nature of the cases. A careful consideration of the experi- 
ence of military hospitals, into which many slight cases are 
received, led to the conclusion that 136 cases should be the 
largest number placed in one double pavilion, divided into two 
equal halves in such a way as to cut off* by through ventila- 
tion the communication between the two halves. In town 
hospitals, where the cases are of a more severe character, a 
similar double pavilion should probably not contain above 
80 to 100 beds. 

The passage lobbies and staircases connecting the two 
halves of a double pavilion should be well lighted by large 
windows, and provided with ample ventilation direct from the 
open air; in cold weather they should be supplied with 

the Internal Arrangements of Buildings. 1 69 

warmth and fresh warm air independently of the wards and 
ward offices. 

The size of any given hospital ought not to be determined 
by increasing the number of beds in any one building, but by 
increasing the number of units, each containing the numbers 
of beds mentioned ; and the extent to which these units 
should be multiplied might, if the units have been properly 
constructed and arranged, be determined not so much by 
the number of patients as by considerations of economy in 
administering the hospital. 

In addition to the larger wards, it is necessary to have a few 
wards of one or two beds each for special cases ; but these 
should be as few as possible, so as to economise labour in 
nursing ; and their position must be adapted in each hospital 
to suit the arrangements of the principal wards, so as to afford 
easy supervision by the nurses. 

It is moreover desirable that if convalescent patients remain 
in the hospital they should have rooms in which they can 
dine and spend the day apart from the other sick ; the situa- 
tion of these rooms should be such as not to interfere with 
the light and air of the wards. This class of patients also 
requires a chapel. It is, however, a subject for consideration 
whether, as a rule, patients who are able to move about in 
this way should be retained in hospitals. 

Arrangements for the several requirements above described 
must all be made subservient to the broad general principle of 
giving air and light to the wards. 

The corridors connecting the units may, in a warm climate, 
consist of an open arcade ; in our climate in cold weather a 
closed corridor may be necessary : closed corridors should be 
lighted by Windows on both sides, capable of opening wide, or 
of being removed altogether in warm weather ; the corridors 
should be cut off from the adjacent pavilions by swing doors 
and be provided with separate means of ventilation, as well as 
with an independent supply of fresh warmed air in cold weather. 

170 Conditions affecting 

These arrangements prevent draughts, and cause the 
corridors, lobbies, and staircases to be the means of effectually 
cutting off the ventilation of one pavilion from that of another. 

The staircases for patients should be broad and easy ; the 
rise of each step should not exceed four inches in height, and 
the tread should be at least one foot in width ; there should 
be a handrail on each side, and a landing after every six or 
eight steps. 

The considerations which these data suggest are equally 
applicable, with modifications suited to the special case, to all 
buildings in which large numbers are congregated. 

The sanitary condition of dwellings which are built for a 
special purpose, such as barracks, workhouses, hospitals, or 
asylums, can be easily controlled. There the occupation is 
continually of the same description. But in ordinary resi- 
dences, the conditions are continually subject to change from 
the necessities or caprice of the occupants. In the houses of 
the rich, inconveniences arise from rooms built for every-day 
life being occasionally used for the reception of large numbers 
of people. Thus a dining-room which would be supplied with 
adequate fresh air for twelve people is sometimes used to contain 
thirty to dine, in addition to servants and numerous lights. 
Drawing-rooms adapted for a small number are often filled 
so full as barely to afford three square feet of surface per occu- 
pant, and are lighted by numerous gaslights or candles. 

Rooms intended to be frequently applied to receptions or 
dinners, either in private houses, hotels, or municipal build- 
ings, should be provided with permanent arrangements for the 
renewal of the air on a sufficient scale. 

The heat which is generated by persons and lights will 
suffice in this country for affording a sufficiently strong 
upward current in shafts to ensure the removal of the 
necessary quantity of air, provided the shafts are properly 

the Internal Arrangements of Buildings. 171 

proportioned, and provided adequate inlets for fresh air be 
supplied. Dinners and receptions are limited in duration, 
hence it may be assumed that 750 cubic feet of air per hour 
supplied for each guest, servant, and candle, reckoning a gas- 
burner as two candles, would suffice. When arrangenjents for 
the removal of vitiated air and the inflow of fresh air have not 
been provided in the building, temporary arrangements may 
be resorted to, such as have been proposed by M. Joly of 
Paris and Mr. Verity in England ; viz. by means of a small 
fan worked by hand or by water power, which forces air into 
the rooms through a perforated india-rubber or other pipe laid 
temporarily along the cornice in a- convenient situation. But 
such an arrangement is only a makeshift to remedy defects 
in the original construction of the building. 

Artificial Lighting. 

The impurity of the air caused by artificial light is very 
serious. Lamps, candles, and gas-lights, each consume the 
oxygen and produce carbonic acid. 

An oil-lamp with a moderately good wick burns about i54 
grains of oil per hour, consumes the oxygen of about 3*2 cubic 
feet of air, and produces a little more than \ a cubic foot of 
carbonic acid ; i lb. of oil demands from 140 to 160 cubic 
feet of air for complete combustion. 

A tallow candle of 6 to the lb., burns about 170 grains per 
hour, consuming the oxygen of about 4 cubic feet of air; 
I lb. of tallow requires about 170 feet of air for combustion. 

Coal-gas consists of olefiant gas and analogous hydro- 
carbons and hydrocarbon vapours, all of which contribute to 
its illuminating properties. It also contains hydrogen and 
marsh-gas, and in addition carbonic oxide, carbonic anhydride, 
sulphuretted hydrogen and other sulphur compounds ; these 
latter are impurities. 

The following table (from Dr. Tidy's Handbook of Chemis- 


Conditions affecting 

try) shows the composition of gas from cannel and from 
common coal. 

Illuminating power com- 
pared to sperm candle 
burning 120 grains per 
hour, the ^as burning 
5 cubic feet. 

Composition in xoo volumes. 












Heavy Hydrocar- 
bons, (C„H2)«. 

Equal to Olefiant 
gas, CaH*. 

Nitrogen, Oxygen, 
and Carbonic Acid. 

Cannel gas . 
Coal gas 








• • • 

I cubic foot of coal-gas will (according to the quality of the 
gas) unite with from -9 to \*6\ cubic feet of oxygen, and pro- 
duces on an average 3 cubic feet of carbonic acid, and from -2 
to '5 grains of sulphurous acid. In other words, i* cubic foot of 
gas will destroy the entire oxygen of about 8 cubic feet of air. 

The presence of i per cent, of carbonic acid in gas is said 
to decrease the light 6 per cent. To obtain a maximum light 
from any flame, the supply of air must not be excessive, 
otherwise the carbon particles are consumed before they are 
sufficiently heated to emit light, and the excess of atmospheric 
nitrogen serves to cool the flame and decrease its illuminating 
power. If, on the other hand, the supply of air is too limited, 
the carbon passes off unburnt, and the flame becomes smoky. 

When gas is only partly burnt in a room, nitrogen, water, 
carbonic acid, carbonic oxide, sulphurous acid and other im- 
purities, may escape into and vitiate the air. 

It is most important to the purity of the air of a room that 
the products of the combustion of gas should not mingle with 
the air. Several forms of lights have been designed for this 
purpose, but as a rule they do not entirely fulfil their object, 
for in many cases they conflict with the other arrangements 
for ventilation. Thus, if a sunlight is placed in the ceiling, its 
proper action is to carry off" a large volume of air from near the 

the Internal Arrangements of Buildings. 173 

ceiling ; an open fireplace in the room draws a large volume 
of air towards itself. The currents conHict, and the fumes 
from the sun-burner are liable to be drawn into the room. 

In the case of a gas globe lamp suspended in the middle of 
the room, supplied with fresh air from the room, with a pipe to 
carry oflf the fumes leading from the light up to the ceiling, 
and along the ceiling into the chimney, the smallness of the 

Fig. 34. Ventilated Gaslight. 

pipe, the bend, and the horizontal length, all contribute a large 
amount of friction to diminish the draught in the tube : in 
rooms where there is no arrangement for replacing the air 
removed by an open fire, cross currents will sometimes prevail 
in the chimney-flue, especially if it is large. Whenever the 
draught in the tube is slu^ish, the Are in the chimney draws 
down some of the fumes directly towards the fire; conse- 
quently it is very rare that, even with this class of burner, the 
fumes of the gas are removed. 

The only safe plan is to place the gas-burners in a globe 
entirely cut off from the room, supplied with fresh air directly 

1 74 ' Conditions affecting 

from the open air, the fumes being also carried directly into 
the open air. Such an arrangement is very simple in the case 
of an outside wall, and this is the only system which will keep 
the air of a room free from the fumes of gas. By the arrange- 
ment in Fig. 34 the fresh air is supplied through the grating 
C, passing along the outer tube to the globe,. and the heated 
fumes pass up and away through the inner tube. When wind 
blows on the opening the pressure is equal on both tubes, and 
if the joints in the room are all air-tight the flame is not 
affected by wind. 

The electric light, if it be ever perfected for domestic use, 
may, it is to be hoped, relieve us from some of the sanitary 
defects experienced with other forms of lighting; but the 
electric light disengages nitric acid, and it will therefore require 
to be kept in a receptacle cut off from the air of the room. 

. Workshops. 

In workshops the purity of air should be maintained by an 
adequate removal of vitiated air, and a supply of fresh air at 
the temperature which may be either necessary for comfort or 
required in the processes carried on in the workshop. With 
adequate renewal of air a temperature of from 75° to 78"* or 
higher can be easily .borne, whereas without such renewal of 
air lower temperatures soon become oppressive. In work- 
shops where the processes carried on occasion fumes, steam, 
or a large amount of dust, special extraction should be 
arranged by means of rapid currents of air to carry off the 
steam, fumes, or dust as soon as it is created, so as to prevent 
its mixing with the air of the room. 


It has not yet been ascertained how much fresh air 
is required to keep a horse in health. Such an inquiry, 
although of great value when warmth has to be combined 
with ventilation, is of little comparative importance as applied 

the Internul Arrangements of Buildings. 175 

to stables, because the horse is not an exotic animal requiring 
artificial warmth. He is taken from a perfectly open-air life, 
with its vicissitudes of weather and temperature, to be confined, 
more or less, in a stable for purposes quite apart from his 
health. The question is, how to subject the horse to the 
captivity he has to undergo in serving man, without injuring 
him ih his health a;nd strength. 

Animal life is most perfectly developed, and its functions 
are most perfectly performed, under the conditions of free 
diffusion of the atmosphere, including absence of stagnation, 
abundance of light, good drainage, absence of nuisance, and 
sufficient space to live in. 

These are the conditions (besides of course food and drink) 
which nature requires for the horse. 

Good stable ventilation includes the other conditions, because 
if the stable is filthy or ill-drained, or the ground saturated 
with putrid urine, it must be obvious that no amount of fresh 
air passing through the stable will keep it sweet and whole- 
some. Any amount of fresh air coming in will immediately be 
tainted by filth which has already collected there. 

Again, if a stable be ever so clean or well drained, it will 
never be well ventilated without perfect freedom of movement 
of air through every part of it, together with free ingress and 
egress of air, so provided as to prevent hurtful blasts falling on 
the horses. 

A fundamental requirement in all stables is paving of such 
a character as to wear well, not to become slippery, to be 
water-tight, and to be easily cleansed. 

Another fundamental requirement is good stable drainage. 

Surface drainage is the only kind of drainage applicable 
to the interior of stables. 

The drains, like the stable floors, should be impervious to 
moisture. They should always be made of smooth material, 
with as few joints as possible, be carefully laid, having a shallow 
saucer-shaped section, and with as rapid an incline as it is 


Conditions affecting 

YatttUbmr to UuSi 


h-ieh »X» 

Section of a Loose Box 1 7ft. x i aft. 


^ Om/iWMf . "I 
mtfcj* tanw 

" '- car»« 


Section of Stable. 

Elevation of Stable. 

Windxrw^ orer trejy Stall ia-s? jtAv-zjiUt jetween f-^rj two atalls 

^heJlcm SK^fatB jbn^jy 

Fig. 35. plan of Stable. 

the Internal Arrangements of Buildings. 177 

possible to obtain. They should pass behind the line of stalls, 
and be conducted in as straight a line and by as short a course 
as possible to the outside of the stable, where they should be 
discharged into an underground drain, over a trapped gulley- 
grating, placed at a distance of some feet from the stable wall, 
so as to prevent effluvia returning, and to prevent dung and 
straw from entering the drain. 

The mast scrupulous cleanliness of the surface of the stable 
should be enforced. 

The great principle which ought to be kept in view in 
stables is to have the air moving freely through every part of 
them, above and around the horses when they are standing, 
and in all the angles between the floor and walls when the 
horses are lying down, and every horse should have sufficient 
ventilation for himself without being obliged to breathe the 
foul air of his neighbours. These conditions would be most 
completely obtained in an open shed, such as is used for 
stabling horses in warm climates, and the nearer we can 
approach to this construction, keeping in view the necessity 
for protecting horses in this climate, while at rest, from 
extreme cold and cold blasts of wind, the healthier will be the 

That form of construction which affords the maximum of 
facility for obtaining a free moving atmosphere throughout the 
body of the stable is the open roof with ridge ventilation 
carried all the way along. 

Where the roof of the stable is not open, but flat and 
impervious, the distance between the effective ventilating 
openings, whether windows or other apertures, corresponds 
of course to the breadth of the stable. But with an open roof 
and ridge ventilation the distance is reduced to one-half, while 
the difference of height above the ground between the ridge 
opening and the side windows ensures a far more certain and 
continuous movement of the air than could by possib*ility take 
plaee with side windows, unless a high wind were blowing. 


178 Conditions affecting 

Therefore a stable with ridge ventilation is the most healthy 

A flat impervious roof, a hay-loft, or a barrack-room over 
a stable increases the difficulties of ventilation. 

In so far then as concerns the general movement and 
renewal of the mass of air in a stable, the form of construction 
which effects this most easily and efficiently is an open-roofed 
stable, with ventilation along the ridge, swing windows along 
the sides, and a continuous inlet for fresh air under the eaves 
made of perforated brick, so arranged as to throw the entering 
currents up towards the roof 

A great incidental advantage of the open roof should not 
be overlooked, and that is the facility with which it enables 
the stable to be thoroughly well lighted. Light is in its place 
as essential to health as air, and moreover, when introduced 
vertically from the roof, it enables the state of cleanliness of 
the stable to be seen at once. 

Besides providing for free movement of the mass of air 
within the stable, it is necessary in all stables, but in some 
much more than in others, to supply fresh air near the ground- 
level at the head of each stall, so that the horse may have 
fresh air to breathe when he is lying down. 

The reason of this necessity is that in all stables the stratum 
of air next the floor-level is the most impure, and will always 
be the most impure under any improved conditions of drain- 
age and paving. Besides this, the horse, in lying down, places 
his head close to the angle between the floor and the wall 
where the air is stagnant. 

It follows from what has been said that the easiest and 
most efficiently ventilated stable is the open roof partially 
glazed, with ridge ventilation all along, ventilation at the 
eaves, a swing window for every stall, and the horses* heads 
turned outwards, with a proper air-brick in the outer wall, 
introduced 6 inches from the ground between every two stalls. 

With these conditions of ventilation the Barrack and 

the Internal Arrangements of Buildings. 1 79 

Hospital Improvement Commission stated that each cavalry 
horse should have 1600 cubic feet and 100 superficial feet of 

General Morin states that the cubic space allowed in French 
cavalry stables is 1750 cubic feet per horse, and he lays it 
down that 7000 cubic feet is the volume of fresh air which 
should be supplied, or of vitiated air which should be removed, 
per horse per hour. 





In every dwelling dryness is an essential of health. In 
this climate it is necessary to provide for warmth. In a 
hot climate coolness is sought. In a hot climate, or for hot 
weather, the roofs and walls should be of such construction 
as to prevent the temperature inside a house being raised 
by the heat of the sun. For cold weather, on the other hand, 
the roof and walls must keep in the heat, so as to main- 
tain the air of the house at a higher temperature than the 
outside air. 

The materials and constructional arrangements best 
adapted to keep warmth in a house will to a considerable 
extent be effective for keeping out the heat. 

The temperature which can be maintained in a house 
will greatly depend on the construction of the walls, and 
on the materials of which they are composed. Materials 
differ greatly in their power of allowing heat to pass through 

The following examples, showing for different materials 
the units of heat transmitted per square foot per hour by a 
plate I inch thick, the two surfaces differing in temperature 
1° Fahrenheit, illustrate their relative advantages in this 

Conditions affecting Materials^ &c. i8r 

Marble, grey fine-grained , . 

Do. white, coarse-grained 
Stone — ordinary freestone 

. . 28 
• . 22 
. . 13-68 
. . 6.6 

Brickwork ....... 

. . 483 

Plaster 3.86 

Brickdnst i'33 

Chalk powdered •Sy 

Fir planks .•..,.... i<37 

Increased conductivity of heat in a material must be 
counteracted by increased thickness. All ordinary wall 
materials admit of a greater or less change of air through the 
material itself, depending upon the extent to which the 
material is porous. 

The porosity of a material is shown by its power to absorb 
water. The following ^ is the percentage of its own weight of 
water which each of the materials mentioned below has been 
found to absorb — 



Malm Cutters 22 

Malm Bright Stock .... 22 

Malm Seconds 20 

Brown Paviors 17 

Hard Paviors 9J 

Common Grey Stock ... 10 J 

Hard Do 7} 

Washed Hard Stocks ... 4 J 

Common Blue 6*5 

Dressed Do 2.3 

Brown glazed brick .... 8*6 


Good Granite 

Indifferent Do. ..... 

Bad specimen Do 

Trap and Basalt .... 
Sandstone — 




Hassock (very bad quality) 
Limestone — 






Kent Rag 

Ransome Artificial stone . . 


• 1 


. 3 
a trace 




a trace 




From this it appears that brick and stone walls, being 
always more or less porous, must admit, as already mentioned, 
of a considerable spontaneous change of air when dry. 

' See Notes on Building Construction, Science and Art Department, 1879. 

£82 Conditions affecting Materials 

Porous walls tend to absorb the moisture given out in 
breathing, or in the combustion of lights. In this process 
walls absorb organic and other impurities, which after a time 
decay, and the wall may become a source of impurity for the 
air. When a wall is damp, change of air can no longer go on 
through it. The evaporation from the wall cools down the 
temperature. Damp in walls is considered to be a cause of 
fever, especially in warm climates. The damp wall, whilst it 
checks the passage of the air, is cold, and consequently 
occasions a rapid radiation of heat from persons sitting within 
its influence. On the other hand, in hot and dry climates, 
wet mats are often hung over all openings, as a means of 
cooling the air without injury to health. These represent 
absolutely pervious walls, admitting of a rapid change of air. 

In this climate, damp walls, besides being unhealthy, are 
uneconomical. They cause a great absorption of heat by the 
evaporation of the moisture from the surface. New walls are 
always damp. The quantity of water which will be contained 
in a new wall is very remarkable. Suppose that 100,000 
bricks are used for a building, each weighing seven pounds ; a 
good brick can suck up from 10 to i^o per cent, of its weight 
in water, but assume 7 per cent, as what gets into it by the 
manipulations of the bricklayer. Also assume that the same 
amount of water is contained in the mortar, a quantity 
certainly much understated ; the mortar forms about one-fifth . 
of the walls ; thus nearly 100,000 pounds of water, equal to 
10,000 gallons, may be assumed to be put in the walls in the 
process of building, and which must be removed from the 
walls of the house before it becomes habitable. This water 
must be removed by evaporation into and by the air. The 
capacity of the air for receiving water depends on the different 
tension of the vapour at different temperatures, on the quantity 
of water already contained in the air as it flows over this 
moist surface, and finally on the velocity of that air. Assume 
the average temperature of the year to be about 50° Fahren- 

and Details of Construction. 183 

heit, and the average hygrometric condition of the air to be 
75 per cent, of its full saturation. At the temperature named 
one cubic foot of air can take up four grains of water in the 
shape of vapour, but as it contains already 75 per cent, of these 
four grains, which amounts to three grains, it can only take up 
one additional grain. As often then as one grain is contained 
in the 10,000 gallons of water mentioned above, as many 
cubic feet of air must come in contact with the new walls, and 
become saturated v/ith the water contained in them ; or, about 
700,000,000 cubic feet of air are required to dry the building 
in question. I'herefore the drying of a building will be best 
effected by passing a large volume of air through it, and air 
at a higher temperature, and therefore of a greater hygro- 
metric capacity than the outer air, will effect this object mfT^t 

Until this damp has been expelled by fires or by time, the 
building should not be occupied ; when the walls have been 
dried inside, it is the proper function of the walls to prevent 
damp from entering from the outside. Damp should be 
precluded from rising into the wall from the ground by means 
of a damp course carried round in all the walls below the 
level of the lowest floor. The damp course may be of slate, 
asphalte, or glazed perforated bricks, this latter form of damp 
course has the advantage of allowing air to penetrate on all 
sides uniformly under the basement floor. 

Damp should be prevented from descending into the wall 
from above by an impervious coping or by eaves. Improperly 
laid copings will act as conductors of wet into the wall, rather 
than as protectors against it. Projecting eaves are advan- 
tageous; the more they project, the greater protection do 
they afford to the surface of the walls. The rain which beats 
against a wall will be partly evaporated out again by the 
action of the air and sun, and partly drawn through it by 
capillary attraction ; and if the material is very porous, or the 
wall very thin, it may saturate the wall. The capillary action 

184 Conditions affecting Materials 

will be checked by joints in brick or stone work, and arrested 
by air-spaces in the wall or by the use of a hollow wall. 

The heat generated in rooms passes away in cold weather 
through the floor, ceiling, and walls and windows into the 
open air or adjacent colder parts of the house, and the heat 
of the house is similarly constantly passing away into the 
open air through the walls and the windows and the roof. 

The temperature of the basement floor when below the 
level of the adjacent surface will be practically that of the 
mean annual temperature, and will therefore occasion little 
loss of heat. 

The loss of heat by walls varies in a direct ratio with the 
conducting power of the material of the walls, and with the 
diff"erence of temperature between the inner and the outer 
surface of the wall, and it varies inversely with the thickness 
of the wall. The actual temperature of the surface of the wall 
is troublesome to ascertain ; if instead of the temperature of 
the surface of the wall, the temperature of the air inside the 
building and outside be taken into account, the formula be- 
comes somewhat more complicated, and varies in a direct ratio 
with the conducting and radiant power of the material of the 
wall, the loss from contact with air, the difference of tempera- 
ture between the air inside and that outside, and in an inverse 
ratio with the thickness of the wall.^ 

* If T" «s temperature of internal air, 
7"«= temperature of external air, 
R a radiant power of the material, 
A s loss by contact of air, 
C — conducting power of the material, 

27 8s units of heat per hour, 

E « thickness of the wall in inches, 

the rule is — 

^_ (AxC)((l)x(T-r) 

{Cx[2XA + R]}+ {ExAxQ^}' 

From Box on Heat. 

and Details of Construction. 185 

The loss of heat by a vertical wall from contact of cold air 
per square foot of area will be greater with a low wall than 
with a high wall. The reason is obvious; the cold air in 
immediate contact with the lower part of the warmer wall 
receives heat from it, and ascends, and at each successive 
gradation in height the difference of temperature between the 
air and the wall is decreased by these successive increments 
of heat, and the amount of heat given out by the wall to the 
air is thus progressively diminished as the air reaches the 
upper part of the wall, in a proportion dependent on the 
square root of the height. 

Thus whilst a vertical plane i foot high would lose -5945 
units of heat per square foot per hour for each 1° Fahrenheit 
difference in the temperature between the surface of the plane 
and the adjacent air, a vertical plane 10 feet high would lose 
•4350 units of heat per square foot, one 40 feet high would 
lose "3980, and one 100 feet high would lose '3843 units per 
square foot. 

The problem connected with the loss of heat by walls 
requires more space for its full discussion than a treatise of 
the nature of this one, limited to the general enunciation of 
principles, will admit of. For the special study of the 
question P^clet, Balfour Stewart, Box on Heat, and other 
writers may be advantageously consulted. From the latter 
work, which is eminently practical in its character, the follow- 
ing table is extracted. The units of heat transmitted per 
square foot per hour by a plate i inch thick, the two surfaces 
differing in temperature 1° Fahrenheit, being as shown by 
P^clet's experiments previously alluded to — 

for ordinary stone = 13*68, 
for brick-work = 4*83. 

The table shows the loss of heat per square foot per hour by 
brick and stone walls, 40 feet high, in buildings where only 
one face is exposed, and for 1° difference of internal and 
external temperature. 

1 86 

Conditions affecting Materials 




Units of Heat. 


Units of Heat. ! 

brick, inches. 

i= 4i 




I = 9 




ij« 14 




2 » 18 




3 = 27 




4 =36 



'2 28 

The number of units of heat lost through a hollow wall, or 
wall with an air-space in the centre, is less than that through a 
solid wall. The units of heat, dissipated by the outer air, will 
be in a direct ratio to the difference between the temperature 
of the air-space in the wall and that of the outer air ; whilst 
the effect of diminished thickness in the wall follows an 
inverse ratio somewhat less than that of the actual diminution 
of thickness of the wall. For instance, if the difference of 
temperature between the room and the outer air = t^ and the 
temperature of the air-space be a mean between that of the 
room and the outer air, then the difference of temperature 
between the air-space and the outer air will equal \ /, and if 
further the thickness of the wall be jE", and the air-space be in 
the middle of the wall, so that the thickness on each side of 
the air-space = i -£", the formula in the preceding note for 
ascertaining the value of the units of heat lost in each case, 
expressed briefly would assume the general form, in the case 

of the wall without an air-space, of -^ — ^rvv and in the case 

of the wall with an air-space, of — ,^ y^ ^r ' or, if we assume 

from the previous table a wall of a room 9 inches thick, 
with a temperature in the room a° above that outside, the 
loss of heat would be -550 units per square foot per hour. If, 
on the other hand, two walls half a brick (or 4^ inches) thick 

and Details of Construction. 187 

each, were used on each side of an air-space instead, and the 
temperature of the intermediate air-space was 1° above the 
outer air, and 1° below the temperature of the air of the 
rooms, the loss of heat would be '371 per square foot per 
hour, or about two-thirds of the heat lost by the solid wall. 

The loss of heat, as well as the porosity of a wall, is in- 
fluenced by wall coverings, such as stucco outside ; or plaster, 
cement, or papering inside. 

Independently of the advantages which may be derived from 
the smaller conductivity, if any, of the covering material, each 
layer forms an additional break in the continuity of any one 
material, and lessens both the porosity and the loss of heat. 

The best internal wall-surface for a dwelling would be an 
impervious polished surface, which would not absorb moisture 
from breathing, and which, on being washed with soap arid 
water, and dried, would be made quite clean. This wall would 
not absorb organic matter, but when colder than the air of the 
room, the moisture from the breath, &c., would be condensed 
upon the wall. To prevent this, either the temperature of the 
wall should exceed that of the air, or a larger volume of air 
would require to be passed through a room with impervious 
walls than a room with pervious walls. In conveying warmth 
to rooms with impervious walls by an agency different from an 
open fire, it would be preferable to convey it in such a manner 
as to warm the walls, and warm the air of the room through 
their means. An enamelled metallic wall surface, with a space 
between the surface and the brick or stone wall for the 
passage of warmed air, would effect this object. 

Plaster, wood, paint, and varnish, all absorb 'the organic 
impurities given off by the body, and any plastered or papered 
room, after long occupation, acquires a peculiar smell. 

In a discussion, in 1862, in the French Academy of Medicine, 
a case was mentioned in which an analysis had been made of 
the plaster of a hospital wall, and 46 per cent, of organic 
matter was found in the plaster. No doubt the expensive 

1 88 Conditions affecting Materials 

process which is sometimes termed enamelling the walls, 
which consists of painting and varnishing with repeated coats, 
somewhat in the manner adopted for painting the panels of 
carriages, would probably prove impervious for some time, but 
it would be expensive, and very liable to be scratched and 

Parian cement polished is practically an impervious material, 
but It is costly; unless carefully applied, its appearance is 
unsatisfactory, and it can only be applied on brick or stone 
walls, and not on wood-work or partitions, because, being 
inelastic, it is liable to crack. The want of elasticity in Parian 
cement is unfavourable to its use in ceilings. 

The numerous joints required for glazed bricks, or tiles, 
render the use of these questionable as wall linings. The 
cement of the joint being more or less porous is sooner or later 
discoloured. Moreover, cracks and joints are objectionable, as 
they get filled with impurities, and may even harbour insects. 

In default of any impervious covering, the safest arrange- 
ment in hospital wards is plaster lime-whited or painted, 
which should be periodically scraped so as to remove the 
tainted surface, and be then again lime-whited or painted. In 
hospitals, of course these arrangements require the wards to 
be periodically vacated ; but this is of itself an advantage, be- 
cause every ward should be left empty annually for a period. 

When plaster is used, it is essential, for the reasons before 
mentioned, that at the expiration of a number of years, 
dependent upon the degree to which the room has been 
occupied, the whole outer coat of plaster should be removed 
from the walls and ceilings, and new plaster substituted. 
When walls are re-papered, the old paper should be invariably 
removed, as it is saturated with organic matter. In all places 
occupied by many persons, such as hospital wards, barracks, or 
asylums, the walls and ceilings should be quite plain, and free 
from all projections, angles, or ornaments which could catch or 
accumulate dust. 

and Details of Construction. 189 

In connection with wall coverings, it is necessary to allude 
to the danger of poison in wall-paper or paint. Arsenic is the 
substance from which this danger is generally found to arise. 

Green paper, as a rule, contains more arsenic than others ; 
but colour in paper is no guarantee of freedom from arsenic. 
If not in the colour itself, it may still be in the mordant dyes 
or other material used in the manufacture of the paper. 
Arsenic in various combinations more or less dangerous is 
used in a great variety of colours. Many French greys and 
neutral tints, and some white papers, are as heavily loaded as 
green. Nothing but an examination of the individual sample 
of paper or colour will afford security against its presence. 

The danger is in proportion to the quantity of arsenic or 
mineral poison used in the colouring matter of the paper ; and 
in proportion to the facility with which it may be removed 
from the fabric, either as dust or gas. 

Danger from arsenic as a colouring matter seems to depend 
in part upon the presence of size. Dr. Fleck showed by 
experiment that a mixture of arsenious acid and starch paste 
or other organic substance gives rise to the formation of 
arseniuretted hydrogen, but no arsenic could be detected in 
air which had been in contact with a mixture of arsenious 
acid and water without the presence of organic matter. 

Arsenic is frequently present in distemper, which being 
mixed with size forms a direct combination of arsenic and 
organic matter, liable to give off arsenic under many circum- 
stances ; and in the case of damp walls it is there ready for 
the development of arseniuretted hydrogen. 

Arsenic is a powerful antiseptic, and hence the danger of its 
being introduced into glue and size, because it is so effectual in 
preventing decomposition, and is free from smell ; moreover, 
size is largely used for fixing colours. In purchasing wall- 
papers a guarantee should be required from the seller ; but it 
is also desirable to have the specimen analysed. 

190 Conditions affecting Materials 


One of the most important conditions to be observed in the 
materials for floors in this climate is that they should not 
be cold to the feet, consequently wood floors are desirable, 
unless the tiled floor is arranged so as to be warmed on the old 
Roman plan — v\z, by means of tiles laid with a hollow space 
or flues underneath, warmed by the smoke or heated gases 
from a furnace. 

Concrete, cement, and stone and brick more or less permit 
the passage of damp ; therefore the floor, of whatever material 
it be, should always have an air-space under it so as to ensure 

The basement floor of a house may have an important 
influence on the purity of air. The heat of the basement floor 
will be substantially that of the ground under it, and that 
temperature differs but little from the mean annual tem- 
perature of the ground. The basement will thus be cooler in 
summer and warmer in winter than the outside air. This 
warmth tends to cause the air to rise through the house, and 
hence in cold weather, if warmed fresh air is not otherwise 
provided in a house, and if there is a basement, the air will 
be liable to pass up from the basement and pervade the 
house. Consequently, in such a case, the purity of the air in 
the house will depend upon the purity of air in the basement. 

The purity of air in the basement will depend upon the 
arrangements for keeping all refuse and objectionable things 
out of the basement itself, but it depends especially upon the 
ground-air in the subsoil under and around the house being 
entirely cut off from the basement. To secure this, a con- 
tinuous bed of concrete, or a layer of asphalte should be laid 
over the whole surface covered by the house, and sufficient 
areas should be carried below the level of the basement so 
a§ to cut the ground-air off* by an open airspace from the 
adjacent soil. 

and Details of Construction, 191 

In order further to diminish the liability of ground-air to 
penetrate into the basement, there should be an air-space 
between the ground and the floor of the basement. 

The floor of the basement, whether it be of wood, of stone, 
or of tile, should be from one foot to eighteen inches above the 
level of the surface of the ground around and under it : joists 
supporting the floor should not be laid on the ground, but 
should have a space underneath, to which a free circulation of 
air should be admitted, by means of gratings communicating 
with the outer air. 

Drains should not be carried under the basement-floor. 

The only case in which it is advisable to dispense with an 
air-space under the floor would be when the floor is of tiles or 
of wooden blocks embedded in asphalte, to ensure dryness. 
A floor of wooden blocks laid on and bedded in asphalte 
combines dryness with warmth for the feet. 

With wooden huts care should be taken to level the ground 
under the floor, and to allow of free circulation of air, either 
between each pair of joists, or, what is preferable, to raise the 
joists so as to allow free circulation under the whole floor. 

Floors should be laid with close joints, and wooden floors 
should be tongued, so as to prevent dirt from falling through and 
accumulating under the floor, as such dirt is liable to putrefy. 

The frequent saturation of wooden floors with water to keep 
them clean diff"uses damp ; consequently a closely laid polished 
floor of hard wood possesses great sanitary advantages. 

For hospitals the floor requires special considerations. 
Ward-floors should be as non-absorbent as possible, and for 
the sake of warmth to the feet they must in this country be of 
wood ; oak, or other close hard wood, with close joints, oiled 
and beeswaxed, and rubbed to a polish, makes a very good 
floor, and absorbs very little moisture. It is impossible to pay 
too much attention to the joints ; they should be like those of 
the best parqueterie, aff'ording no inlet for dirt to lodge ; because 
the impurities which become lodged in the cracks of a hospital 

192 Conditions affecting Materials 

floor are eminently objectionable. There should be no saw- 
dust, or other organic matter subject to decay, under the 
floor. When one ward is placed above another, it is essential 
that the floor should be non-conducting of sound, and that it 
should be so formed as to prevent emanations from patients 
in the lower ward from passing into the upper wards. The 
floors of the Herbert Hospital are formed of concrete, sup- 
ported by iron joists, over which the oak boards are laid. 

An economical and non-absorbent surface for the floor can 
be obtained by first laying rough deal boards and covering 
them with thin, closely-laid oak boards oiled and beeswaxed. 
These oak floors must be treated like the French parquet, by 
occasional frottage. A very good hospital floor is that used 
at Berlin, which is oiled, lacquered, and polished, so as to 
resemble French polish. It is damp-rubbed and dry-rubbed 
every morning, which removes the dust. This wet and dry 
rubbing process of cleaning is far less laborious than either 
frottage or scrubbing, and completely removes the dust and 
freshens the ward in the morning. The only objection to this 
surface is its want of durability, and consequent necessity for 
periodical renewal. Both of the processes above mentioned 
render the floor non-absorbent, and both processes do away 
with the necessity of frequent scouring, which is objectionable 
from the quantity of damp it introduces into the ward. The 
French floor stands the most wear and tear, but must be 
rubbed periodically by a frotteur, which cleaning is more 
laborious than scrubbing. The daily cleaning of a bees- 
waxed floor and the removal of dust may be effected by 
wiping with a damp cloth wetted with warm water, and 
carefully rubbing with a dry cloth. Practically, with care, 
a well-laid oak floor, with a good beeswaxed surface, can 
always be kept clean and polished in this manner assisted 
by periodical frottage. 

and Details of Construction. 193 


The outer covering of a house should be impervious to 
moisture from without; but experience shews that it is un- 
healthy to live under a ceiling impervious to air. Air heated 
by contact with the human body carries up emanations which, 
when they rest upon a pervious ceiling, are retained there, 
whilst the moisture passes off through the ceiling; on the 
other hand, if these emanations come in contact with an 
impervious ceiling, they are not absorbed, and may be again 
brought into circulation in the air of the room. Consequently, 
if circumstances render it necessary to have a ceiling imper- 
vious to air and moisture, this must be discounted by pro- 
viding under such impervious covering additional facilities for 
change of air in the upper part of the room. 

The ordinary lath and plaster ceiling, although a good non- 
conductor of heat, allows of a considerable passage of air and 
moisture. With a pressure obtained by a difference of tem- 
perature of 72° inside and 40° outside, the quantity of air 
which was found to pass through ordinary plaster was about 
1*5 cubic feet per square foot of area per hour. 

The nature of the roof must depend on the materials avail- 
able. Thus cement, clay, tiles, wooden shingles, slate, iron, 
copper, lead, are used according to circumstances. If good 
conductors of heat are used to keep out wet, such as copper, 
iron, or slates, some non-conducting material is required 
underneath ; for instance, in the case of a slated roof, the 
slates should be laid on close boards covered with felt, to 
ensure the best protection against heat or cold. 

The loss of heat through the roof will depend upon whether 
the rooms are ceiled, and upon the form and nature of the roof- 
covering. If there is a lath and plaster ceiling to the upper 
rooms, and an air-space between the ceiling and the roof, 
closed from the outer air, so as to prevent any rapid circula- 
tion of air, and if the roof be formed in the most approved 


194 Conditions affecting Materials 

manner in this country, viz. with close boards covered with 
felt under the slates or tiles, the loss of heat in winter, and the 
effect of heat in summer in raising the inner temperature, will 
be comparatively small. 

If there is no ceiling, and if the roof be not carefully con- 
structed, as above mentioned, the loss of heat will be very 
considerable. The loss of heat from glazed roofs, ceilings, and 
skylights, and from metal roofs, such as are used in railway 
stations and markets, is very considerable. 

The air in contact with metal or thin glass exposed to cool- 
ing influences is under the most favourable condition for being 
cooled. Each layer as it is cooled falls down and is replaced 
by warm air, which undergoes the same process. This renders 
a space covered with a metal or glass roof without inter- 
mediate ceiling very difficult to warm. Therefore in halls or 
rooms lighted by a glass roof, or staircases lighted by sky- 
lights, it is essential for preserving the heat that there should 
be a second glass ceiling below the one exposed to the outer 
air ; and in cold weather it may be advisable to adopt special 
means to warm the intermediate space, if an equable tempera- 
ture is sought to be maintained in the room at all times. 
Where glazed ceilings are lighted by gas-lights above the 
lower ceiling of glass, the heat from the gas when lighted is 
sufficient to keep up the temperature. 

In very hot weather, when it is desired to cool down the 
temperature of an iron or glass roof, it may be watered by 
jets from 8 or 9 o'clock in the morning till about 5 o'clock 
in the evening. The quantity of water would however be 
considerable, probably not less than about 25 gallons per hour 
per square of 100 feet. The practice of lime-whiting the roof, 
which is largely resorted to in some places, is a great protec- 
tion against heat. 

Windows should be carried up as near the ceiling as possi- 
ble. In a low room this is essential, in order to freshen the 
air. In a lofty room it is equally necessary, in order to pre- 

and Details of Construction. 195 

vent the warm impure air which has risen to the top from 
cooling, falling, and remixing with the air in the lower part of 
the room. 

Therefore if the tops of the windows are much below the 
ceiling there should be openings above the windows f6r change 
of air. Moreover, whilst it is an element of cheerfulness in 
a room for the upper part of the windows to be near the ceil- 
ing, it is equally important that the sills of the windows should 
be brought down near to the floor. 

Similarly, it is an element of cheerfulness to splay the sides 
of window openings. 

The proportion of window surface to the cubic contents of 
the room must to some extent depend on the climate and 

In England, adequate light will generally not be obtained 
with less than one square foot of window-surface to about 100 
cubic feet of the contents of the room. In hospital wards, one 
square foot to 60 cubic feet of content is desirable. This 
should be a minimum allowance, and assumes that the win- 
dows in all cases are of clear glass ; but greater cheerfulness 
will be secured by more light. In climates with bright sun- 
shine less light may be found sufficient. Where light is abun- 
dant it can always be modified, when desirable, by sunshades 
or blinds ; but if window openings are small, light cannot be 
increased at will. 

The amount of light afforded by a window is considerably 
modified by the quality of glass. 

In some recent experiments :— 

Polished British plate glass, \ inch thick, intercepted 13 per cent of the light, 

36 oz. sheet glass , 2% 

Cast plate glass, \ inch thick, .... „ 30 „ 

Rolled plate glass, 4 corrugations in an inch „ 53 „ 

»» *> 

- »t 

Clear glass is thus of great importance, and the thicker it is, 
consistent with clearness, the better, because thin glass allows 
of a more rapid loss of heat. 


196 Conditions affecting Materials 

Good glass is desirable, because dust adheres easily to bad 
glass, but not to good glass ; and the surface of bad glass is 
more or less rapidly rendered uneven by exposure to the 
atmosphere, whilst good glass is unaltered by long exposure. 
The quality of glass depends upon the admixture of the 
ingredients. The percentage composition of window glass 
is 66'37 of silica, i4'23 of soda, ii«86 of lime, 8*i6 of alu- 
mina; that of plate glass 73-5 of silica, 5-5 of potash, 12 of 
soda, 5'5 of lime, y^ of alumina. A soda glass is more 
fusible and more brilliant than a potash glass. But soda 
imparts a slightly green tinge to glass, which does not occur 
with potash. Lime diminishes the fusibility of the glass, 
imparts no colour, and increases its hardness and lustre. 

The most convenient form of window for ventilation in 
ordinary dwelling-rooms in this country is the sash-window, 
opening top and bottom. This mode of construction assists 
ventilation in the manner already described by enabling a 
current to be maintained, fresh air passing in below whilst 
the vitiated air of the room passes out at the top. Ventilation 
should not however be dependent only on windows. In 
hospitals, where the wards, and consequently the windows, are 
lofty, the lower part of the windows may be advantageously 
constructed of tlie sash form, whilst the upper part is hinged 
on a transom so as to open inwards, and thus facilitate the 
inflow and outflow of air, on the plan of a hopper ventilator. 

In hot climates the French window is advantageous, as it 
enables the whole window opening to be utilised for supplying 
fresh air. 

The loss of heat through windows amounts to that lost by 
radiation added to the loss of heat by contact with air. 

It may be assumed with thin glass that the temperature of 
the outer surface of the glass is a mean between the tempera- 
ture inside the room and that of the outer air. 

For thin glass, adopting the previous notation in the note, 
page 184, U=(,R^A){T--P). 

and Details of Construction. 197 

With thick glass the conducting power of the material must 
also be taken into account, as in the case of a wall. 

The loss of heat with double windows is much less than 
that with single windows, and they have the advantage not 
only of transmitting less heat, but, from the temperature of 
the inside glass being greater, less radiant heat is absorbed 
from the occupants of the room. P^clet found that the loss of 
heat in double windows increased somewhat with the distance 
apart of the inner and outer glass, owing probably to the 
greater facility for currents of air in the wider space between 
the glass. 

Thus with an intermediate space between the windows of 
•8 of an inch, the loss of heat of the single window to that of 
the double window was in the proportion of i : -47 ; with a 
distance apart of % inches the proportion was as i : '55 ; 
with a distance apart 2-8 inches, which is nearly what exists 
in practice, the proportion would probably be as i : -6. 

The proportionate loss of heat by walls, as compared with 
the loss of heat by windows, varies to some extent with the 
conditions of the room, 1. e. the proportionate extent of wall 
exposed to the outer air ; but with 14-inch brick walls, and an 
assumed internal temperature of 60° in the room and an out- 
side temperature of 30°, the proportion of loss of heat from 
wall-surface to loss of heat from window-surface may be 
approximately taken to be about 1 : 2*5. 



After air, water is the first requirement for existence ; and 
impure water is as fertile a source of disease as impure air. 

AH available water comes from the rainfall ; if rain falls on 
an impervious surface, it runs off in surface streams and rivers ; 
if it falls on porous formations, it runs off in underground 
streams and rivers. 

A cubic foot of water weighs looo ounces, or 62-5 lbs. An 
imperial gallon weighs 10 lbs. avoirdupois at 69° Fahrenheit. 
Fresh water in cooling becomes denser until the temperature 
reaches 39° Fahrenheit ; after this it again expands until it 
reaches 31*°, when it solidifies into ice. In the act of freezing, 
water expands considerably, and with sufficient force to burst 
iron pipes. This is so well known a fact that it is quitp 
astonishing how builders and architects continue year after 
year to leave water-pipes unprotected from frost. 

There are few things which water does not dissolve to some 
extent. Its solvent powers are generally increased by rise of 
temperature, but there are some exceptions to this. Chloride 
of sodium (e.g. salt) is dissolved to the same extent whatever 
the temperature ; sulphate of lime is less soluble in hot than 
in cold water. 

Water absorbs gases in variable amounts — it dissolves 
ammonia and hydrochloric acid in large quantities; but it 
dissolves the gases of the atmosphere, oxygen, nitrogen, and 

Purity of Water. 199 

carbonic acid only in small quantities, as will be seen from the 
following table : — 

Quantity of different gases absorbed by i volume of water at 1 5*C 
(yfF) and 30 inches Barometric Pressure, 

Volume of gas dissolved 
by I voL of water. 





Ammonia .... 
Hydrochloric acid 
Sulphurous acid (SO,) 
Carbonic acid (CO,) 


Nitrogen .... 

Water usually absorbs more gas the lower the temperature 
and the greater the pressure : when water is warmed it gives 
out the gases ; when it freezes the dissolved gases are usually 
liberated. This is a point of importance in the question of 
storage of water. 

Rain, as it leaves the clouds, is pure, but in its passage 
through the air it absorbs certain gases, and carries with it 
particles of matter which may be floating in the air. 

The gases it absorbs are oxygen, nitrogen, carbonic acid, a 
little ammonia, and nitric acid. The particles floating in the 
air are for the most part organic. 

Rain water collected from a rocky surface sparsely occupied 
by population is the purest attainable water supply. 

Near large towns other impurities are found in rain water. 
But notwithstanding this, such rain water is generally far 
purer than river water. The water found in rivers has either 
drained into the rivers from land, or, having fallen on porous 
strata, is given out from them in springs. For this reason all 
water in rivers or streams contains more or less of matters 
taken up from the soil. Thus rivers fed from a granite country 
will be comparatively pure, though frequently peaty. There 
is no evidence that a little peaty discolouration is injurious to 

Rain which falls on cultivated land is liable to pollution 

200 Purity of Water. 

from the manure intended for the crops. When rain falls on 
a closely inhabited surface, or passes into a subsoil saturated 
with impurities^ it will be contaminated. 

The most agreeable waters are generally those containing 
nitrates and chlorides ; such waters should be avoided. 

Pure water acts rapidly on unprotected iron pipes — cast 
iron is less affected than wrought iron — and a coating of 
asphalte or pitch will protect the iron, provided the coating 
be perfect. If pure rain water is allowed to come in contact 
with lead it will dissolve a part of the metal. 

The action of water on lead appears to depend on the 
quantity of oxygen and carbonic acid. Where there is a 
large quantity of oxygen, the lead is rapidly oxydised, and 
the oxide of lead is to a certain extent soluble in pure water ; 
but if the water contains a sufficient quantity of carbonic acid 
to convert the oxide into carbonate of lead, which is only 
slightly soluble, the water will be comparatively safe from 
dangerous contamination. 

A lead top and lead fittings in a pump of a well have been 
found to be injurious, from the evaporated water, which is 
pure, condensing on the lead, dissolving some of the metal, and 
dripping back into the well. Lead cisterns are subject to the 
same effect, and therefore, lead cisterns for storing water 
should be avoided. 

Peaty matter in water forms a coating which protects the 
surface of the lead ; water which has been for some time in 
contact with unprotected iron pipes, and has been deprived of 
its free oxygen, is less liable to action on lead. 

But, in point of fact, the conditions under which water acts 
on lead are so intricate that it is preferable, where any doubt 
exists, to avoid such water for domestic purposes, if possible ; 
or if no other supply is available, to use iron or earthenware 
pipes for its conveyance, and slate or earthenware cisterns for 
storing it. 

Hard water never acts on lead ; it forms a protecting surface 

Purity of Water. 201 

by the deposition mostly of sulphates, and carbonates of lime 
and magnesia. 

The effect of hardness of water on health is a somewhat 
intricate question. 

A comparison made on an average of five years between the 
death-rates of twelve towns furnished with soft water, as com- 
pared with twelve towns furnished with hard water, showed a 
marked preponderance in favour of the towns supplied with 
hard water as compared with those supplied with soft water, 
10 degrees of hardness being the standard. But the sanitary 
condition of the several towns in other respects would require 
special examination before this evidence could be admitted as 

Of the conscripts taken in France, a larger number were 
found to be rejected on medical examination from soft water 
districts than those taken from hard water districts. On the 
other hand, Highlanders are a stalwart race, and the water 
they have is mostly soft water. 

Hard water is alleged to produce in some individuals 
certain diseases, such as stone ; but, on the other hand, in 
valleys in mountainous districts, where the water is practically 
pure rain water, the inhabitants suffer from goitre. 

The economical advantages for domestic use of soft water are 
undoubted, arising especially from the saving of soap. Each 
degree of hardness destroys 2 J ounces of soap in each 100 gal- 
Ions of water ; therefore, soft water is commercially of more value 
than hard water, in the proportion of 5 ounces of soap to each 
aoo gallons for each degree of hardness. For many manufactur- 
ing purposes soft water is preferable. Thus, dyeing requires soft 
water, — on the other hand, ale-brewing requires hard water. 

Water of 10 degrees of hardness would probably satisfy the 
general requirements of a town supply. 

The terms * hard ' and * soft * refer to the soap-destroying 
power of a water. Soap is an alkaline stearate. The addition 
to it of lime and magnesia decompose it, forming a calcic or 

202 Purity of Water. 

magnesic stearate. Hence the reason why it is difficult to 
obtain *a lather' with hard water (i.e. a water containing 
lime and magnesian salts), viz. because a certain quantity of 
the soap is required to decompose the calcic and magnesian 
salts before a lather can be obtained, whilst, conversely, a 
lather is at once formed with a soft water, because of the 
absence of calcic and magnesic salts. 

Two kinds of hardness are usually described — 

Temporary hardness is due to calcic or magnesic carbon- 
ates. These salts are almost insoluble in pure water, but 
are freely soluble in water containing carbonic acid. On 
boiling, the carbonic acid is expelled, and the carbonates are 
precipitated. Hence temporary hardness is that hardness 
which may be got rid of by boiling the water. 

Permanent hardness is chiefly due to the presence of calcic 
and magnesic sulphates, and is not got rid of by boiling. 

In expressing the hardness of a water in degrees, it is to be 
understood that every degree, theoretically, represents one 
grain of calcic carbonate, or its equivalent in soap-destroying 
power, in one gallon of water. 

Hence, the universally-employed test for hardness is the 
soap test, originally suggested by Dr. Clarke. This test 
consists in employing a solution of soap of known strength, 
and ascertaining how much of this solution is required to form 
a lather which will last a certain time. 

For the purpose of this test, in the first place, i6 grains 
of neutral chloride of calcium (prepared by solution of car- 
bonate of lime in hydrochloric acid, and repeated evaporation 
to dryness, until all the excess of acid is driven off") are 
dissolved in a gallon of distilled water, which is then said to 
be of 16 degrees of hardness, a solution of soap of a given 
degree of strength is then formed. The former of these liquids 
is used to graduate the latter. The degree of hardness of the 
water to be tested is estimated from the number of measures 
of soap solution required to form a permanent lather. 

Purity of Water. 203 

Chalk waters are notoriously hard waters. They are partially 
softened by boiling, and deposit a large amount of fur. 

In selecting a water for use it is necessary to determine — 

I. The initial hardness. 

a. The permanent or irremoveable hardness. 

3. The removeable or temporary hardness. 

Chalk water may be softened to a considerable extent by 
another process invented by Dr. Clarke. 

To explain this process it is necessary to consider the com- 
position of chalk. 

A pound of chalk consists of — lime 9 oz., carbonic acid 
(CO2) 7 oz. 

The 9 oz. of lime may be obtained separately from CO2 by 
burning, e.g. driving off the 7 oz. of carbonic acid by heat. 
When so separated the 9 oz. of lime may be dissolved in not 
less than 40 gallons of water, and what is called lime-water 

The 16 oz. of chalk would probably require 5000 gallons of 
water to dissolve them — so sparingly soluble is chalk. 

But if 7 oz. more of carbonic acid, in addition to the 7 oz. 
of carbonic acid in the chalk, be added to the 16 oz. of 
chalk in water, then it becomes readily soluble, and bicar- 
bonate of lime is formed. If the proportion of such a solution 
were 16 oz. of chalk and 7 oz. of carbonic acid to 400 gallons 
of water, then the water would resemble well water from the 
chalk strata. 

Now if a solution of lime-water, e. g. water containing 9 oz. 
of lime, be united to a solution of water containing 16 oz. of 
chalk and 7 oz. of carbonic acid, then the solutions will so act 
on each other as to form % lbs. of chalk. Thus : — 

Bicarbonate of lime ( lib. chalk =.,6oz.f Lime 9 oz. 1 . ,5 ^z ) 

in 400 gallons. ) ^^^^ iCarbonic acid 7 oz. / * f = 2lbs. 

( Carbonic acid 7 oz.l s= 16 oz 1 ^^ Chalk. 

Burnt lime in 40 gallons of water 9 oz. J ' ' 

In point of fact, only W of the chalk would be separated — 

204 Purity of Water, 

thus the water of 17^ degrees of hardness would remain of \\ 
degrees of hardness. 

In practice, however, chalk water contains other hardening 
matters besides chalk ; and consequently the softening can 
rarely be effected below 2^ or 3 degrees. But the process 
also appears to remove mechanically a great portion of any 
organic matter in the water. 

Where water has to be stored, or to remain exposed, as in 
reservoirs, for any length of time, there is on the one hand 
the great liability of soft water to absorb noxious gases, and 
on the other the tendency with hard water to foster vege- 
tation of various kinds. Waters from the new red sandstone 
are worse even than the chalk waters in this latter respect. 
Water softened by Clarke's process is comparatively free from 
this objectionable tendency. 

The general principles of water-supply may be stated briefly 
as follows : — 

1. To select the purest available source after careful analysis. 
a. To filter the water, if necessary, in order to free it from 

suspended matter and from dissolved organic matter. 
3. To store it in covered tanks, and to raise it a sufficient 

height for distribution by gravitation. 
Water may be obtained from — 

mountain ranges, which act as condensers, 
rivers and streams ; 
natural springs ; 
wells artificially formed ; 
impounding reservoirs ; 

a combination of two or more of the sources named. 
And water may be conveyed for distribution by means of — 
open conduits (before filtration) ; 
covered conduits, always, after filtration ; 
cast-iron pipes under pressure, — where a district is to 
be supplied with water. 
^ Water containing much suspended matter is generally to be 

Purity of Water. 205 

avoided, but frequently the suspended matter may be removed 
by filtration or simple straining. 

As regards organic matter, it is the kind of organic matter 
which is of importance. 

The mere existence of organic matter either in solution or 
in suspension is no proof of impurity. If water contains the 
fresh juices of inoffensive plants, it would not be unwholesome. 
But when these juices putrefy , or when water contains organic 
matter, and especially animal matter, ready to putrefy, it 
should be. avoided. "" 

The question of analysis of water is beyond the scope of 
this treatise, but a few remarks on the simplest facts connected 
with examination of water may be useful. 

The quality of a water can only be judged of from its con-* 
stituents as a whole. The values of these constituents are not 
their values absolutely, but their values relatively, and in con- 
nection with the history of the water. Fifty grains of salt per 
gallon may mean nothing ; five grains per gallon may mean 

Care must be taken that the specimen of water to be 
examined is a normal sample of the river or spring to be 
tested, that the vessel it is placed in is cleaned and rinsed out 
with some of the water to be tested. The water should 
be collected from a spot where it is not subject to arti- 
ficial disturbance, and in filling a bottle its mouth should 
be immersed, if possible, from one to two feet below the 

The bottle should be kept stoppered, and allowed to stand 
for a day or two, exposed to light but not to evaporation, to 
see whether vegetation or putrefaction is developed. It should 
be in a temperature suited to vegetation. 

If animalculae appear, they are an indication of nitrogenous 
matter, and are one proof of the presence of substances 
capable of putrefaction • a microscope is necessary to detect 
the smaller forms of life. 

2o6 Purity of Water. 

Clearness and absence of colour in water is no certain indi- 
cation of its being wholesome. The worst waters may be 
clear, bright, and colourless. Colour as a companion to 
chemical analysis may give to an experienced observer a clue 
to the kind and quantity of organic matter present. Its 
colour is best judged of through a tube 2 feet long, % inches 

If a water exhibits a bluish tint, or, say, appears nearly 
colourless in the 2-foot tube ; if it, moreover, uses up very 
little oxygen after standing for three hours in contact with 
permanganate, the freedom of that water from organic im- 
purity may be relied upon as certain. 

If a water exhibits but little colour, or at most a slightly 
yellow, or a greenish-yellow tint, but if the oxygen it uses up 
is found to be large, such a water as a potable water is 

If a water exhibits in the a-foot tube a decided peaty tint ; 
if by experiment it is found to need a large quantity of 
oxygen after standing for three hours ; knowing that peaty 
matter acts as a reducing agent on permanganate, the quantity 
of oxygen required, although far in excess of what was used 
in the former case where there was an absence of colour in the 
water, is not to be regarded with the same suspicion, peaty 
matter not being injurious to health. 

Water that will not bear the test of standing will, in most 
cases, be rejected at once. If no other water can be obtained, 
it ought to be used before putrefaction has set in, but this is 
a great risk; the next best method is to wait until after 
putrefaction has terminated. 

An indication of the nature of organic matter in water may 
be obtained by evaporating a small quantity of the water, 
weighing the residue, and then burning it to drive off the 
volatile matter, and comparing its weight before and after 
burning. If the residue blackens in burning, it indicates 
animal organic matter. 

Ptirity of Water. 207 

In considering whether organic matter comes from animals 
or vegetables, the presence of common salt may in many 
cases, with proper precautions, be found to be a nearly certain 

We consume not less than 100 grains of salt a day. This 
salt is thrown off daily. From all animals there is a large 
outflow of salt. It is the constant accompaniment of the 
animal living, or decomposing after death. 

If much salt is found in water containing organic matter, 
nitric acid will generally be found, and if not nitric acid, then 
animal matter unoxidised. 

In the case of dead animals organic matter is destroyed or 
retained in the soil ; phosphates and other organic substances 
are also retained. Salt is removed by water. Sewage 
contains chlorides, and the amount of salt is the most certain 
method of ascertaining the quantity of sewage which is or 
has been present — assuming that there are no disturbing 

Of course this test must be used with caution. Near the 
sea the spray is driven many miles inland. In many districts, 
where deposits of salt exist, wells and springs are saline. 
Chlorides are also given out from manufactures. 

Water containing chlorides to a great extent ought not to 
be used without careful examination as to the source—three 
grains per gallon of common salt coexisting with an excess of 
nitrates is a cause of suspicion. 

The average of the chlorides in the water of a district 
having been obtained, any increase of common salt above the 
average in a well situated in a camp or city, or near habita- 
tions, is an almost sure test of impure drainage. 

Although caution must be exercised in drawing conclusions 
from the presence of chlorides, their absence is conclusive 
against the presence of decomposed animal matter or ex- 

Nitrates are common in small quantities. The amount 

2o8 Purity of Water. 

from atmospheric causes is very minute, but they are found 
in water from manured land, in gardens, in wells near houses, 
in towns, and in great abundance near churchyards. 

If chlorides and nitrates are found together in water, it may 
be assumed tRat animal matter has existed, or does exist, in 
the water. 

If there is much nitrous acid, it indicates recent organic 
matter or oxidation going on. 

The examination thus suggested would show the presence 
or not of : — u. 

I. Organic matter decomposed or putrid. 

3. Organic matter readily decomposed and probably ready 
to become putrid. 

3. Organic matter slow to decompose, but still capable of 

becoming putrid. 

4. From th-e nitrites recent organic matter. 

5. From the nitrates old organic matter. 

6. Vegetable organic matter. 

7. Animal organic matter. 

In deciding on the quality of water the relationship which 
each of its constituents bears to the others must be con- 
sidered as well as the natural history and the physical 
condition of the water. Any classification of waters founded 
on a single factor in the chemical analysis of water is to be 
accepted with great caution, however important that single 
factor may be. 

The general line of examination of waters can only be 
indicated here. 

The engineer in selecting a source of water-supply is so 
deeply concerned in the question as to what is good water, 
that these few practical hints on the subject may assist in the 
consideration of the subject. 

But whilst much knowledge is derived from a water 
analysis, especially as to the waters which should be avoided, 
it is necessary in the case of waters reported by analysis to 

Purity of Water. 209 


be good, to be very careful to keep up a continuous examina- 
tion of their sources, and to see that these are not liable to 

The sources of contamination of water arise at every step. 
That which this country suffers from most largely is the 
contamination of rivers by sewage. 

With an increasing population over the districts which form 
the water-sheds of rivers, this is inevitable. 

The fact that sewage or other contamination is poured into 
a river at a point high up on its banks is, however, no bar to 
the water of that river being used at some spot lower down 
for domestic purposes — ^the distance between the spot at 
which contamination has been poured in and that at which 
the water can be again used for domestic purposes, will 
depend upon the nature of the bed of the river, the rapidity 
of the current, and the temperature ; the effect varying with 
the season, being much more rapid at a high than at a low 

This cannot be better illustrated than by extracts from the 
Report of the Commission which reported on the pollution of 
the River Seine at Paris in 1875 and 1876. 

Above Paris the Seine presents a satisfactory appearance. 

The sewers of Paris discharge into the river black foetid 
streams, covered with layers of greasy matter, which ac- 
cumulates on the sides of the river. These streams transport 
particles of organic matter and debris, which are deposited as 
gray and black mud on the banks, or else form shoals. This 
mud is the seat of an active fermentation, throwing up in- 
numerable bubbles of gas, which burst at the surface of the 
water; the bubbles attaining sometimes in hot weather a 
diameter of li metres (nearly 5 feet), dragging up the black 
mud with them to the surface. Fish and plants cannot exist. 
But as the river leaves these sources of pollution, it gradually 

From Epinay to Argenteuil the water is still of a deep 



Purity of Water. 

colour, mud has disappeared, fish make their appearance. 
Below Bezons a most abundant vegetation clothes both 
banks, and large sheets of water-plants partly impede the 
course of the river. At Meulan all visible sign of pollution 
has disappeared, and the river is chemically pure. 

The following table extracted from the report of the 
Commissioners shows the chemical condition of the water : 


Nitrogen not yet 
transformed into vola- 
tile ammoniacal salts, 
or organic nitrogen. 

Total nitrogen, 
including volatile am- 
moniacal salts. 

Dissolved oxygen in 

cubic centimetres per 

litre of water. 

Grammes per cubic 
metres or zooo litres. 

per cubic metre. 

Above Paris — 
Bridge of Asnieres . . . 

Tn Paris— 

Clichy, below inter- 
cepting sewer ... 

St, Denis, below inter- 
cepting sewer ... 

Below Paris — 









Cubic centimetres. 





The table explains the process which takes place. 

The organic matters change into carbonic acid, water, 
ammonia, sulphuretted hydrogen, and different mineral sub- 
stances. This change implies an absorption of oxygen from 
the gases dissolved in the water, and a production of mineral 
nitrogenous bodies. 

As long as the water contains such matters susceptible of 
fermentation, it is unfit for use. When fermentation is ac- 
complished, and the organic matter has passed into the 
state of mineral matters inoffensive in themselves, the water 
presents ,a disappearancQ of nitrogenous organic matter. 

Purity of Water. 2 1 r 

replaced by nitrogenous mineral matter, by ammonia. The 
dissolved oxygen in the water is used up, but may be 
restored by movement, such as is caused by the flow of 
the stream, or more rapidly by agitation, as for instance in 
passing over a weir, and thus the water can be rendered 
fit for drinking, • 

It will be seen from this that although a river may have 
received sewage at one part of its course, there is some point 
below at which it will still become fit for use. 

There are other and more occult sources of contamination. 

All water proceeds from evaporation and rainfall. The 
rain which falls on impervious strata passes from the surface 
into rivers, whilst water which falls on pervious strata passes; 
into the ground. 

When this water, in its passage through pervious strata, 
meets with an impervious bed, it is arrested in its course ; 
and if the impervious bed dips down and forms a basin, then 
the water will remain in a subterranean reservoir accessible 
only by wells. 

Where the impervious strata which underlie the pervious 
strata crop out at the surface, the water flows gradually 
down as an underground river, to pass out at the lowest point 
in springs. 

An examination of the water-level of different districts 
seems to show that this general water-level forms an inclined 
plane, rising from the natural vent or outfall. 

But the course of this underground river depends on the 
inclination of the water bearing and underlying impervious 
strata, and not on the form of the surface. 

Taking the water-levels of a line of wells at right angles to 
the outfall, be it in a river or in springs, it has been found 
that in the chalk the inclined plane of the surface of under- 
ground water rises at a rate of not less than ten feet per 
mile^; in the boulder strata of Norfolk it appears to vary 

* Clutterbuck. 
P % 


sPurtty of Water, 

frorii three to a hundred feet per mile ; and in certain of the 
tertiary beds it is about five feet per mile ^. 

Of course this inclination varies with the head of water 
and the porosity or permeability to water of the material, 

Mr. Roberts, of Liverpool, found by experiments in the 
new red sandstone of average coarseness of that district, that 
the following quantities of water passed per hour through a 
square foot of the sandstone loi inches in thickness : — 
With a pressure of lo lbs. to square inch, 4i gallons 

^o « « 7i 

4<5 „ „ 19 

which showed the increase to vary in a direct ratio with the 

The level of the subterranean sheet of water, moreover, 
rises and falls with the quantity supplied by rainfall ; thus, 
in wet seasons, the water-level approaches near to the sur- 
face, and in dry seasons it recedes. Mr. Baldwin Latham 
has shown the effect of this change in contaminating the 
water supply at Croydon. (Fig. 36.) 





± 8 9 S S e 7 6 8 JVfi2M 

Fig. 36, Section of water-level in chalk near Croydon. 

Croydon has been drained effectually, and the water-supply 
IS obtained from springs at some distance off in the chalk. 

^ Baldwin Latham. 

Purity of Water. 2 1 3 

In the dry seasons the supply was excellent; in wet 
seasons fever was found to prevail, which was attributed to 
the water. The cause of this was explained by the fact 
that in dry seasons the level of the water was far below 
any subsoil contamination; but that, in 3 wet season, the 
water-level of the underground river rose up to the level 
of cesspits which were not impervious, manure heaps, and 
other sources of contamination, and thus became contami- 
nated with the contents. 


^S^^^^^^S^F^^^^S^^&L^- » ''" ,< ^^ --r^= 

There are, however, many causes of contamination of water 
which are much more obvious. 

In a porous soil, the vicinity of a cesspit is a frequent cause 
of danger. Figs. 37 and 38 show how fluctuating may be 
the danger in such cases. 

In one case {Fig. 37) the flow of underground water is 
from the cesspit to the well. 

Whilst the ordinary level of underground water prevails. 


Puriiy of JVater. 

there would be nothing to indicate the danger; but upon 
an excess of rainfall occurring, by which the current of 
underground water was altered, then the well would become 
polluted, and sickness break out, to cease probably on a 
fresh change of conditions. 

In the second case (Fig. 38) the flow of underground 
water is from the well to the cesspit, and in this case there is 
not the same imminent danger of pollution as in the former 

The pollution of the water may occur in its transit from 
the source of supply to the consumer. The London Water 
Companies have sometimes had the water polluted by pass- 
ing it through iron mains calked with hemp, when the hemp 
has become a cause of injury to the water. 

The question of water-supply would require a treatise to 
itself ; a few practical hints are all that can be given here. 

The quantity of water for the civil population cannot be 
assumed at less than 15 gallons per head per day ; for infantry 
ii2 gallons per head per day, for cavalry 20 gallons per head 
per day, in barracks. 

An Artesian well (Fig. 39) is a well sunk through imper- 
vious strata, in which there is no water, into pervious strata 


f77» m J' J' . . 4 yi . -^ 3 . , 

Fig. 39. Artesian Well. 

which derive their supply from a water-shed area at a dis- 
tance. Thus, in London, wells are sunk through the London 
clay to derive their supply from the chalk. The water in the 

Purity of Water. 215 

chalk comes from the water which falls on the chalk hills 
surrounding the London basin. Deep well water is generally- 
very good and palatable water. 

In porous soils water must generally be obtained froni 
surface wells. 

A spring is the lowest point or lip in the stratification of an 
underground reservoir 'of water. A well sunk in such strata 
will most probably furnish, besides the volume of the spring, 
an additional supply of water. 

Well water will vary in purity according to the nature 
and amount of the soluble matters contained in the ground 
from which the well derives its supply. 

Shallow wells are always liable to pollution from vegetable 
matter, or animal matter, in or on the surface soil. 

In deep wells differences have been observed in the purity 
of well water, according as the water was taken from the 
surface by dipping, or was pumped from the bottom of the 
well. In one case, where the depth of water was over 50 feet, 
the surface water was found to be yellowish in colour, much 
harder, and more contaminated with organic impurities than 
that from the bottom of the well, as if a layer of lighter water 
from the surface drainage floated on the spring water below. 
In another case the solids per gallon amounted to 66'%2^ grains 
in the bottom water of a well and to 3 grains only in the 
surface water. 

Such differences are often constant. Exhaustive pumping 
in wells may be injurious to the water ; especially in wells in 
the vicinity of the sea, when the brackish water usually kept 
back by the lan^ water may be drawn in. 

Norton's tubes are useful for ascertaining the quantity of 
subsoil water, and fof drawing the supply from a certain 
depth. They are easily applied ; but the supply from each 
has necessarily a limit. 

In camps it may frequently be necessary to resort to a 
surface supply of Water. 

2i6 Purity of Water. 

The following are a few temporary expedients for camp 
purposes of water supply. ■ 

Where the water was near the surface, one system pursued 
by the Turks in the Crimean war was to select a site near their 
camp where the surface was clean, and dig a hole and place 
in it a barrel {Fig. 40), in the bottom of which holes had been 
bored, so as to ensure that water frdm the deepest part of 
the hole should alone come into the 
barrel. If they wanted further fil- 
tration, they got a smaller barrel, 
and bored holes in the sides as 
high as was desired, and they then 
placed the smaller barrel in the larger 
one, ramming sand in below and all 
^"'''^Fig."^! ^*"' '°""d so as to bring its top to the 

level of the top of the first barrel, 
and thus forming an upward filter of sand through which the 
water passed. 

The Russian plan, where brushwood was obtainable, was to 
make a large gabion four or five feet in diameter, some 10 or 
12 feet long, and then to dig the well and drop this gabion 
into it. This formed a temporary casing, which answered its 
immediate object. 

For wells of any degree of permanence, care is necessary to 
ensure that the surface water for a certain distance round the 
well should not pass back into it, for the surface near a well 
is always liable to pollution. 

In deep wells, i. e. wells in which the water of the requisite 
purity is only found at a considerable depth, the surface soil 
water should be cut off from the deep water by carefully 
casing the well and puddling behind the casing above the 
water level, so as to prevent surface water from trickling 
down the sides. 

Surface wells in porous strata should be lined with puddle 
behind the steyning to the full depth, or in wells beyond la 

Purity of Water. 2 1 7 

to 15 feet in depth, then to that depth at least, so as to 
ensure complete filtration of the surface water which passes 
into the well. The surface round the well should also be 
puddled, sloped away from the well, and paved, so as to 
prevent the dirty water which is thrown down near the well 
from finding its way back again. (Fig. 41.) 

The distance to which this preparation of the surface 
should be carried round the well depends on the depth of 
the well ; with a deep well, in 
which there would be a con- 
siderable depth of soil through 
which the surface water would 
have to pass before it reached 
the well, an extensive surface 
covering would not be necessary; 
whilst with a shallow well, by 
extending the area of surface 

puddling, the amount of fil- 

. .... J- Fig. 41. ShflUowWell. 

tration to which the surface 

water would be subjected before it reached the well may be 

materially extended. 

The tube lining of the well should be carried up to at least 
two feet abovethe surface, so as to prevent surface impurities 
from falling in. The well should be covered as a protection 
against dead leaves, &c. 

For keeping well-water clean, it is preferable that it should 
be drawn up by a pump. In the use of buckets impurities fre- 
quently fall in. If not a pump, then an iron chain and bucket, 
because they can be kept cleaner than rope and wood. 

In some cases there is advantage in the aeration which the 
water obtains by being exposed to the air. In such cases, 
closing the well and drawing water by pump has rendered 
the water undrinkable. 

Where a river flows through a valley over porous substrata, 
sinking a well or wells in the strata within the influence of 

2i8 Purity of Water. 

the river filtration is a cheap and ready niethod of obtaining 
river water naturally filtered. The tops of wells so situated 
must be carried above the level of extreme floods. 

If a single well on a river bank does not produce sufficient 
water, or if the; subsoil is clay, impervious to water, trenches 
may be excavated parallel to the river or stream, in which 
trenches perforated earthenware pipes may be laid, leading to 
a well or wells. The trenches above such pipes should be 
filled in with fine assorted gravel, charcoal, and sand, so as to 
form a filtering medium within the reach of the dry weather 
flow of such stream or river. These trenches should not be 
less than six feet deep to the top of the pipes. 

At Florence the new water-supply for the town is derived 
from channels nearly a mile long, carried by the side of and 
below the level of the Arno. 

Natural springs may be utilised by storing the water in a 
reservoir which will contain the flow of one entire day, or 
of a longer period. 

All reservoirs and tanks should be constructed to prevent 
infiltration of water from the soil, the sides and bottom should 
be lined with puddle, faced with brick or masonry walls, and 
coated with cement. Reservoirs may be covered in to 
protect the water from contamination. In case of peaty 
waters aeration is beneficial ; and in such cases it may be 
advisable to draw off" the upper film of water for use. 

In positions where there is a difficulty as to wells, springs, 
or streams, dependence must be placed on rainfall. Where 
water is to be collected from rainfall, great care is requisite 
in order to keep perfectly clean the surfaces on which it falls, 
as well as the conduits to the reservoirs and tanks. 

When rain water is collected from roofs, the first washings 
of the roof should not be collected. 

Rain water is not agreeable to the taste until it has been 
stored in the ground, where it absorbs carbonic acid gas. 

The rain water from roofs should be first collected in a 

Purity of Water. 219 

settling tank, of a sufficient area to allow the heavier sedi- 
ment to be deposited; it should then be passed through a 
gravel or coarse sand strainer into the tank or cistern for 
use. It is customary in parts of Italy to provide three 
receptacles. The rain water is received and allowed to settle 
in the first, it is strained off into the second, where it is 
retained if possible for a year, and then passed into the third 
for use. The settling tank and the strainer should be cleaned 
out frequently. 

In the case of porous strata of clean sand or gravel, where 
sufficient ground can be set apart, and where military or 
cheap labour is available, water may be stored by excavating 
an area, of a size calculated on the minimum rainfall, to a 
depth of five or six feet, and paving the bottom with tiles, or 
impervious material such as puddle, cement, or asphalte, 
sloped down towards one comer, and lining the sides with 
puddle, and then filling in the whole again with the porous 
material of the ground. A pump could be fixed in the 
lowest corner. 

The surface should be covered with grass or shrubs, and kept 
free from impurities. By these means a valuable reservoir 
would be formed, which would keep the water cool. 

All water, before being stored in tanks, from which it is to 
be pumped direct for use, should be passed through a filter ; 
every rain water storage tank should have its filter. 

For filters on a large scale, sand is the simplest material. 
It not only operates as a strainer, but the surfaces of the 
particles of sand serve more or less to attract solid matter 
brought within range of their attraction. It has been estimated 
that a cubic yard of sand would contain an area of surface in 
the particles of about ^2,500 square yards. 

A sand filtering-bed should have a thickness of about 
two feet of sand at the top, under which should be placed four 
layers of gravel, each six inches thick, the first of the size of 
shot, the next of peas, the third of beans, the fourth the size 

220 Purity of Water. 

of walnuts; the rate of filtration should not exceed two 
gallons per superficial foot per hour. 

The surface soon becomes clogged, and the rate of filtration 
decreases ; the upper film of sand should then be removed and 
a fresh layer put on. 

Filtering beds of sand are stated to filter the water less 
efTectually when first constructed ; the filtration afterwards 
becomes more perfect, and remains so until the lower stratum 
of sand is filled up with organic matter. The filtering bed 
should then be renewed. The duration of a filtering bed in 
fairly constant use should not exceed two years. If the bed 
were left dry at intervals, a somewhat longer period might 
be afforded before renewal. 

The sand filter acts not only mechanically, but it will to 
some extent remove matters in solution. 

Filters for domestic purposes were experimented on by 
Dr. Parkes. He showed that a sand filter has considerable 
purifying power when first used, but that it clogs rapidly. The 

sediment stops mainly on the surface, and thus diminishes its 


filtering power; this can be restored for a certain time by 
scraping off a small film of the top surface of the sand. 

Magnetic oxide of iron may be usefully added to a sand 
filter where there is much organic matter. It increases the 
oxidizing power of the filter, and renders it more effective for 
destroying organic matter. 

Sponges are useful with sedimentary waters; they arrest 
more than their own weight of solid matter, and are very 
easily cleaned. 

One of the best filtering mediums is animal charcoal. The 
charcoal, in consequence of the very large surface arising from 
its porosity, undoubtedly has great power as a mechanical 
filter. It has been contended by some chemists that the 
organic matter which passes into it becomes oxidised. On 
this point, however, the evidence is not conclusive. 

Spongy iron appears to be an active agent, not only in 

Purity of Water. 221 

removing organic matter from water, but in reducing its 
hardness and altering its character when water is filtered 
through the spongy material. 

Filters retain the impurities which are present in the water, 
and at first the outflowing stream is materially purified ; but 
the retention of the impurities in the filter becomes very soon 
a source of danger of itself. 

The pores become filled with the dead bodies of the 
organisms which were living in the water, and the water 
passing through may become after a time more dangerously 
polluted than it was before it entered the filter. 

Filters must, therefore, be frequently and thoroughly 
cleansed by washing the materials of which they are com- 
posed, and by exposing these materials to the air, so as to 
cause any remaining impurities to be oxidised. 

A filter with sponge to arrest the sediment, so arranged 
that the sponge can be frequently cleaned, and a porous 
filtering medium so arranged as to have a free exposure to 
the air when water is not actually passing through, would be 
the best form of domestic filter. 

After filtration water should be kept in covered receptacles 
for use. 

The storage and distribution of water is the next con- 

The best position for keeping water is in tanks under- 
ground, assuming always that it is not in proximity with 
sources of impurity. In the ground it will retain an even 
temperature, and will take up carbonic acid gas, which makes 
it pleasant to the palate. When stored in cisterns above the 
ground-level, where its temperature is necessarily variable, 
much danger may ensue from its power of absorbing gases, 

For instance, on a warm day, the water becoming warm, 
gives out the oxygen or other gases it contains; at night, 
when it cools down, its capacity for absorbing gases is much 
increased ; and if there are impure gases near, such as sewer 

222 Purity of Water. 

gas^ or ammonia from dung heaps or manure heaps, it will 
absorb them in large quantities, and thus become dangerously- 

The system of water-supply which minimises this source of 
danger is to be preferred. Thus well-water was observed to 
have a maximum temperature of S^*^^ and a minimum tem- 
perature of 5o«5°, or a mean of 51° during the year in the same 
locality. The water supplied from mains had a maximum 
temperature during the year of 64-8°, a minimum temperature 
of 39*7°, and a mean temperature of ^%*%1^ ; and the water in 
cisterns above ground varied from 7i'5® maximum to 33'6o*' 
minimum, with a mean of 5iZ'55°. Thus whilst the range of 
temperature of the well-water was only to 1°, the range of the 
water in the cistern was nearly 38**, 

There is danger from the use of cisterns unless frequently 
cleaned, for the cistern becomes a place of deposit for any 
impurities which there may be in the water ; especially if the 
cistern is left unused for a time ; and thus, like the filter when 
not cleaned, cisterns may render the fresh water which comes 
into them impure. 

Moreover, the supply of water to cisterns is necessarily 
intermittent, and when cisterns have been filled, and the pipes 
are emptied of water, cases have occurred in which impure 
gases have been drawn into the pipes, and have been taken up 
by the water when it was again turned on into the pipes, 

A case occurred in which a town council, in order to econo- 
mise the laying of pipes, carried them in the sewers. A 
severe epidemic of typhoid fever occurred. This was accounted 
for by the assumption that sewer gas was drawn into the 
pipes, either through the joints or by diffusion through the 
pores of the metal. 

On these grounds the constant service, in which the pipes 
are always filled, is preferable to the intermittent supply. 

As water will readily absorb foul gases, and may become 
poisonous, the possibility of contamination from foul air should 


Purity of Water. 223 

be prevented. Hence any overflow or waste-water pipe from 
a service reservoir or tank should deliver the water at an open 
end into a channel, and be thence conducted into the covered 
sewer, or drain, so as to prevent gases from the sewer rising 
back through such overflow or waste-water pipe to the water 
in the reservoir or tank. 

Water conduits should be laid at such depth and be so 
covered with earth as to prevent the water becoming heated 
unduly by the rays of the sun, or being injuriously affected by 
frost in winter. This depth may be considered to be not less 
than two or three feet. 

All covered reservoirs and tanks should be ventilated, and 
so situated £is to be easily emptied, inspected, and cleaned. 

All supply-pipes should be arranged in such manner as to 
allow of easy inspection and subsequent repairs. Stop-taps 
should be placed betwixt the water main and the building in 
all cases, so as to allow of the isolation of any line of service- 
pipe for repairs. 

All house service-tanks and service-pipes should be fixed in 
such manner as to be carefully protected from frost, and so 
that the best rooms shall not be flooded on the occurrence of 
leaks or overflows. When protected by wooden casings, the 
front of the casing should be made to open easily. 

The distribution of water in camps should invariably be 
placed under inspection, and under the charge of some person 
who should be responsible for not allowing waste or pollution. 
With this object the general dipping of buckets into a well 
should be prevented. 

If a pump is not fixed, there should be one man made 
responsible for raising the water out of the well. He should 
pour it into a box or reservoir in front provided with taps, out 
of which all who come for water should draw without confusion 
or fouling of the well with dirty buckets. 

The surface of the ground near any place where water is 
obtained should, if possible, be paved, or else it will soon 

224 Purity of Water. 

become a mass of mud. This is especially desirable with 

In making troughs for watering horses care should be taken 
to supply each trough independently of the neighbouring 
troughs. It has been often the practice to place the troughs 
in a line, and let the overflow of the first fill the second, and 
so on ; but the water of a trough becomes fouled by the horses 
drinking in it, and consequently the overflow from the upper 
troughs conveys nothing but polluted water into the lower 

No vessel to convey water for drinking purposes for human 
beings should under any circumstances be allowed to be 
dipped into a trough used for watering horses. 



A THEORETICALLY perfect system of refuse removal would 
be one where a large volume of rapidly flowing water received 
the whole refuse, and carried it away, before it had time to 
decompose, to a large river not used for drinking purposes, and 
thence to the sea ; but this is generally unattainable, and it 
would leave the waste matter unutilised. 

Refuse from dwellings falls under the heads of ; — 

1. Ashes. 

2. Kitchen refuse. 

3. Stable manure. 

4. Solid or liquid ejections. 

5. Rain water and washings of passages, stables, 

yards, and pavements. 

The first three of these should be removed by hand labour ; 
it is only to the two last that the term sewage is applied. 

To arrive at a clear conception of what the problem is, it 
must be remembered first, that even where a water carriage 
system prevails in any town, some plan of removing house 
refuse, ashes, and dust — some 'dry system* of collection — 
must also be in force, and secondly, that an ordinary house- 
hold of, say, six persons, enjoying a proper water supply, will 
consume for all purposes from 15 to ao gallons per head per 


226 Removal of Refuse. 

diem, or a total daily quantity of from 90 to \%o gallons, which 
must necessarily be fouled in its use, and pass away from the 
house, laden with dangerous impurities. If earth-closets are 
used instead of water-closets, a certain quantity of water, 
perhaps as much as four or five gallons per head may be 
saved, but there will be still from 10 to 15 gallons of water 
fouled with grease, soapsuds, vegetable refuse, and other 
materials of a highly putrescible kind, which must necessarily 
pass away and be disposed of at the outfall. 

Thus some system for the removal of foul water is absolutely 
necessary in every town, and the Rivers Pollution Com- 
missioners, in their First Report, after a careful comparison of 
thirty-one towns, in fifteen of which the midden system 
prevailed, while sixteen were water-closet towns, came to the 
conclusion that it mattered little, as regards the degree of 
pollution in the sewer water, whether a system of interception 
of the excreta was practised or not. 

Hence, with the dry conservancy system, there must be the 
removal of sewer water. 

With the water carriage system the removal of ashes, refuse, 
and stable-yard manure has still to be provided for ; but the 
removal of ashes and stable yard manure is a much simpler 
matter than the removal of excreta. 

The great principle to be observed in removing the solid 
refuse is that every decomposable substance should be taken 
away at once. 

This is especially necessary in warm climates. 

The principle may, however, be applied in various ways to 
suit local convenience. In open situations, exposed to cool 
winds, there is less danger of injury to health from decom- 
posing matters than there would be in hot, moist, or close 
positions. In towns in warm climates all refuse should be 
removed daily from both the house and its enclosures. 

In the country, generally, there is less risk of injury than in 
the close parts of towns. These considerations show that the 

Removal of Refuse. 227 

same stringency is not necessarily required everywhere. 
Position by itself affords a certain degree of protection from 
nuisance. The amount of decomposing matter usually pro- 
duced is also another point to be considered. A small daily 
product is not, of course, so injurious as a large product. 
Even the manner of accumulating decomposing substances 
influences their effect on health. There is less risk from a 
dung-heap to the leeward than to the windward of an in- 
habited building. 

The receptacles in which refuse is temporarily placed should 
never be below the level of the ground. 

If a deep pit is dug in the ground, into which the refuse is 
thrown in the intervals between times of removal, rain and 
surface water will mix with the refuse and hasten its decom- 
position, and generally the lowest part of the filth will not be 
removed, but will be left to ferment or putrefy, and the 
presence of this fermenting mass will tend to promote the 
decomposition of the fresh refuse added to it. 

Even the daily removal of refuse entails the necessity of 
places for the deposit of refuse. In the selection of these the 
following conditions should be attended to : — 

I. That the places of deposit be sufficiently removed from 
inhabited buildings to prevent any smell being perceived by 
the occupants. 

%. That the places of deposit be above the level of the 
ground— never sunk in the ground. The floor of an ash- 
pit, or dung-pit, should be at least three inches above the 

3. That they be lined with non-porous material. That the 
floor be paved with square setts, or flagged, and drained. 

4. That they be protected from rain and s,un, but open to 
the air. 

5. That a space be paved in front, so as to prevent the 
traffic, which takes place in depositing the refuse or in 
removing it, from cutting up and polluting the surface. 

228 Removal of Refuse. 

A moveable iron box is a good receptacle for kitchen waste. 
For stable-yard refuse, the mcJveable wire-work grating in use 
in London is safer than any form of dung-pit. 

In camps the disposal of manure and offal requires great 
care ; burning, unless carefully done, may give rise to much 
nuisance ; burying at a safe distance is generally the most 
advisable method. 

The disposal of the solid or liquid ejections, which constitute 
what is generally termed sewage must next be considered; 
and before entering upon the general question, a few of the 
rules which govern temporary emergencies in camps may be 

A camp unprovided with latrines is always in a state of 
danger from epidemic disease. 

One of the most frequent causes of an unhealthy condition 
of the air of a camp is, either neglecting to provide latrines, so 
that the ground outside the camp becomes covered with filth, 
or constructing the latrines too shallow, and exposing too 
large a surface to rain, sun, and air. Latrines should be so 
managed that no smell from them should ever reach the 
men's tents ; to ensure this, very simple precautions only are 

1. The latrines should be placed to leeward with respect to 
prevailing winds, and at as great a distance from the tents as 
is compatible with convenience. 

2. They should be dug narrow and deep, and their contents 
covered over every evening with at least a foot of fresh earth. 
A certain bulk and thickness of earth are required to absorb 
the putrescent gas, otherwise it will disperse itself and pollute 
the air to a considerable distance round. 

3. When the latrine is filled to within two feet six inches 
or three feet of the surface, earth should be thrown into it, 
and heaped over it like a grave to mark its site. 

4. Great care should be taken not to place latrines near 
existing wells, nor to dig wells near where latrines have been 

Removal of Refuse. 229 

placed. The necessity of these precautions to prevent wells 
becoming polluted is obvious. 

Screens made out of any available material are, of course, 
required for latrines. 

This arrangement applies to a temporary camp, and is only 
admissible under such conditions. 

A deep trench saves labour, and places the refuse in the 
most immediately safe position ; but a buried mass of refuse 
will take a long time to decay ; it should not be disturbed, 
and will taint the adjacent soil for a long time. This is of 
less consequence in a merely temporary encampment, whilst 
it might entail serious evils in localities continuously in- 

The following plan of trench has been adopted as a more 
permanent arrangement in Indian villages, with the object 
of checking the frightful evil of surface pollution of the whole 
country, from the people habitually fouling the fields, roads, 
streets, and watercourses. 

Long trenches are dug, at about one foot or less in depth, 
on a spot set apart, about rzoo or 300 yards from dwellings. 
Matting screens are placed round for decency. Each day the 
trench which has received the excreta of the preceding day is 
filled up, the excreta being covered with fresh earth obtained 
by digging a new trench adjoining, which when it has been 
used is treated in the same manner. Thus the trenches are 
gradually extended, until sufficient ground has been utilised, 
when they are ploughed up and the site used for cultivation. 

The Indian plough does not penetrate more than eight 
inches, consequently if the trench is too deep, the lower 
stratum is left unmixed with earth, forming a permanent 
cesspool, and becomes a source of future trouble. 

It is to be observed, however, that in the wet season these 
trenches cannot be used ; and in sandy soil they do not answer. 

This system, although it is preferable to what formerly 
prevailed, viz. the surface defilement of the ground all round 

230 Removal of Refuse. 

villages, and of the adjacent water courses, is fraught with 
danger, unless subsequent cultivation of the site be strictly 
enforced, because it would otherwise retain large and increas- 
ing masses of putrefying matter in the soil, in a condition 
somewhat unfavourable to rapid absorption. 

These arrangements are applicable only to very rough life 
or very poor communities. 

In permanent camps and barracks, and in villages where 
the houses or cottages are close together, and where therefore 
sufficient garden ground cannot be allotted to each to allow of 
the refuse being absorbed, which is especially the case with 
large Indian villages, some one of the methods adopted for 
town conservancy hereinafter described would usually be found 

Midden, Pail^ and Dry Earth systems. 

Moveable apparatus, such as middens, pails, or dry earth 
closets all involve the retention of the excreta for some period 
of greater or less duration, as contradistinguished from the 
immediate removal which takes place when water carriage is 
used. Moreover the weight of the tub, pail, or receptacle in 
which the excreta are removed has to be added to the weight 
of the refuse to be carried away, whilst in the water carriage 
system the water itself is the vehicle of transport. 

There are numerous towns, and innumerable villages and 
cottages, where the midden closet or privy with a fixed re- 
ceptacle still prevails. 

In its old form the receptacle consists of a pit with sides of 
porous materials, permitting free soakage of filth into the 
surrounding soil, capable of maintaining the entire dejections 
from one or more houses for months or years, uncovered and 
open to wind and rain and sun, undrained and difficult of 
access for cleansing. 

The first improvement was to provide a cover to keep out 
the wet ; the next, to provide a drain for excess of liquid, and 

Removal of Refuse. 231 

to make the sides and bottom of the pit impervious ; the next, 
to deodorise the contents with ashes, or some other cheap 
deodoriser. The next important improvement was to reduce 
the size to a mere space behind the seat, formed of some non- 
porous material, such as glazed brick, and arranged for easy 
cleaning from the back. The small size prevents the retention 
in the vicinity of the dwelling of a large mass of putrescent 
matter. Privies of this class are, however, inadmissible except 
in detached positions away from dwellings, as for instance at 
the end of a cottage garden in the country ; but where used, 
the floor of the privy should be raised, so that the seat may 
be high enough to admit of the floor of the receptacle being 
on or a little above the level of the ground, with a slightly 
raised edge to prevent any liquid flowing over. The recep- 
tacle should also receive dry refuse and ashes, which help to 
deodorise the contents and soak up the liquid ; and the 
receptacle should have a cover to prevent rain falling into it, but 
so placed as to allow of the circulation of air under the cover. 

The next transition from this was to a moveable receptacle. 
Of this type the simplest arrangement is a box placed under 
the seat, which is taken out, the contents emptied into the 
scavenger's cart, and the box cleansed and replaced. 

A further alteration is the separation of the solid and liquid 
excreta ; a system which has not hitherto attained to much 
practical application. 

The difficulty of cleaning the angles of the boxes led to the 
adoption of oval or round pails. (Figs. 4fl and 43.) The pail 
is placed under the seat, and removed at stated intervals, or 
when full ; some form of deodorising material is thrown in daily. 
In the north of England the arrangement generally is that the 
ashes shall be passed through a shoot, on which they are 
sifted — the finer fall into the pail to deodorise it, the coarser 
pass into a box, whence they can be taken to be again burned 
— whilst a separate shoot is provided for kitchen refuse, which 
falls into another pail adjacent. (Figs. 44, 45, 46, and 47.) 

233 Removal of Refuse. 

Excrtmeol Paitfiia. ^"'tl' li^". 

ExcrtmtKt Pail tinffy, rtadyfor ust. 

Fig. 41. 43. Hie Rochdale Pail. 

EltMlk^i n rant. 

Sictional EUvatieK- 

Flan aiDoi Privy Smi. Plan rfPrinf Fleer. 

Fig. 44. 45, 46. 47. Closet with pail and refuse box. 

Removal of He/use, 233 

The general arrangement is for the floor of the privies to be 
somewhat raised, so that the space underneath the seat in 
which the tub is placed may be at or somewhat above the 
level of the ground, so as to be easily cleaned. 

A door gives access to the space under the seat, and when 
the tub is removed it is at once covered with a lid and placed 
in a van, while a clean tub, having in it a small supply of 
disinfecting fluid, is substituted for the full one taken away. 
The van holds twenty-four tubs ; each van making several 
journeys a day. 

The Goux system, which has been in use at Halifax, 
consists in lining the pail with a composition formed from 
the ashes and all the dry refuse which can be conveniently 
collected, together with some clay to give it adhesion. The 
lining is adjusted and kept in position by means of a core or 
mould, which is allowed to remain in the pails until just before 
they are about to be placed under the seat j the core is then 
withdrawn, and the pail is left ready for use. 

Ijued Pail used for Godi System. 

The liquid which passes into the pail soaks into this lining, 
which thus forms the deodorizing medium. 

The proportion of absorbents, in a lining three inches thick, 
to the central space in a tub of the above dimensions, would 
be about two to one ; but unless the absorbents are dry, this 
proportion would be insufficient to produce a dry mass in the 
tubs when used for a week, and experience has shown that 
after being in use for several days the absorbing power of the 

234 Removal of Refuse. 

lining is already exceeded, and the contents have remained 

There would appear to be little gain by the use of the 
Goux lining as regards freedom from nuisance, and though it 
removes the risk of splashing and does away with much of 
the unsightliness of the contents, the absorbent, inasmuch as 
it adds extra weight which has to be carried to and from 
the houses, is rather a disadvantage than otherwise from 
the manurial point of view. " 

The great superiority of all the pail or pan systems over 
fixed middens, such as formerly prevailed, is due in the first 
place to the fact that the interval of collection is reduced to 
a minimum, the changing or emptying of the receptacles 
being sometimes effected daily, the period never exceeding 
a week ; and secondly to the receptacles for the refuse being 
independent of the structure, and therefore capable of easy 
periodical renewal when the material becomes saturated. 

In the ordinary pail system, the plan of deodorising the 
material whilst the pail is in use is scarcely resorted to, from 
the trouble and expense it occasions ; and probably the best 
known contrivance for deodorising the excreta immediately 
after they fill into the receptacle, is Moule's Earth-closet, in 
which advantage is taken of the deodorising properties of 
earth. Many modifications of this system are in use. 

Dry earth is a good deodoriser; li lbs. of dry earth, of 
good garden ground or clay, will deodorise each excretion. 
A larger quantity is required of sand or gravel. If the earth 
after use is dried, it can be applied again, and it is stated that 
the deodorising powers of earth are not destroyed until it has 
been used ten or twelve times. 

This system requires close attention, or the dry earth closet 
will get out of order, and the best dry closet is liable to a 
peculiar smell, which becomes very objectionable when the 
closet is not properly cleansed. 

As compared with water-closets, the earth-closet is cheaper 

Removal of Refuse. 235 

in first construction, and is not liable to injury by frost ; and 
it has this advantage over any form of cess-pit, that it ne- 
cessitates the daily removal of refuse. 

On the other hand, the dry earth system is much more 
expensive than the pail system. 

The dry earth system, though applicable to separate houses, 
or to institutions where much attention can be given to it, is 
inapplicable to large towns from the practical difficulties 
connected with procuring, carting, and storing the dry earth. 

The pail system, which is in use in various ways in the 
northern towns of England, and in the permanent camps to 
some extent at least, is more convenient as well as more 
economical than the dry earth system, where the removal of 
a large amount of refuse has to be accomplished. 

The value of the manure depends on its bulk in relation to 
the distance it has to be carried. 

If the manure is highly concentrated, like guano, it can 
stand a high carriage. If the manuring elements are diffused 
through a large bulk of passive substances, the cost of the 
carriage of the extra, or non-manuring elements, absorbs all 
profit If a town, therefore, by adding deodorants to the 
contents of pails, produces a large quantity of manure, con- 
taining much besides the actual manuring elements — such as 
is generally the case with dry earth — as soon as the districts 
immediately around have been fully supplied, a point is soon 
reached at which it is impossible to continue to find pur- 
chasers in consequence of the high cost of carriage. 

According to theory, the manurial value of dejections per 
head per annum ought to be from 8j. to loj. And if the pail 
system is to be commercially advantageous, the liquid and 
solid dejections must be collected without admixture of 
foreign substances. 

General Scott has devised a plan by which he hopes to 
effect this. 

He mixes the contents of the pails with lime, arid distils 

236 Removal of Refuse. 

the mixture in a close boiler, so as to draw off four-fifths of 
the ammonia. 

The ammonia thus evaporated is passed into a tank filled 
with sulphuric acid, and thus sulphate of ammonia is obtained 
in a crystallised state fit for the market. 

The residue, after being mixed with superphosphate and 
dried, is ready for sale as manure. 

In a sanitary point of view the pail or tub system has an 
enormous advantage over the old midden or cess-pit system, 
in the facility with which the excrementitious matter is 
removed, without soaking into the ground or putrefying in 
the midst of a population. 

There is no doubt that in some parts of India, where the 
water supply is often not plentiful, and the question of sewage 
removal presents many difficulties, a systematised pail system 
would afford great advantages. 

Pneumatic System for the Removal of Excreta. 

Instead of the removal of the excreta, as just described, by 
the pail system, there is in Paris, in Florence, in Philadelphia, 
and in other foreign towns, a pneumatic system in operation. 
The water-closet refuse is passed into close cess-pits, from 
which the contents are removed at frequent intervals. To effect 
this, in Paris a large cylinder, in Philadelphia a series of barrels, 
are brought on a cart ; a tube from them is connected with the 
cess-pit, the air is then exhausted in the barrel or cylinder 
by means of an air-pump, and the contents of the cess-pit 
drawn through the tube by the atmospheric pressure into the 
cylinder or barrels. 

A plan which is practically an extension of this system has 
been introduced by Captain Liernur, in Holland. 

He removes the foecal matter from closets, and the sedimen- 
tary products of kitchen sinks by pneumatic agency. He places 
large air-tight tanks in a suitable part of the town, to which 

Removal of Refuse. 237 

he leads pipes from all {he houses. He creates a vacuum in 
the tanks, and thus sucks into one centre the foecal matter 
from all the houses. 

The mode of action is briefly as follows : — These tanks are 
in communication, by means of pipes, with small tanks in 
each street, so arranged that the vacuum in the central tanks 
can at will be extended to any given street reservoir. 

Each of these street reservoirs is the centre of a small 
drainage system of houses, independent of all others, and the 
foecal matter out of those houses is drawn into it by means of 
the vacuum, created in the manner described, after which the 
matter is at once despatched to the main building by means 
of the same pipe that first conveyed the air. The closets are 
connected with a main pipe lying in the street, by means of 
branch pipes, somewhat like the mains and branches of water- 
works. Each main has, however, only one stop-cOck, and 
when this is turned all the closets connected with it empty 
themselves at once through the main pipe into the street 

The manipulation is as follows : — The air-pump engines 
maintain throughout the day a vacuum in the central tanks, 
and hence also in the tubes, extending the vacuum into the 
street reservoirs. A couple of men perambulating the town, 
visit each reservoir in turn once a day, and discharge into 
them, by opening the stop-cocks of the mains, the closet 
matter of all the houses belonging to the particular section 
connected with the reservoir ; after which, before leaving, they 
dispatch the whole to the central building by simply opening 
the stop-cock of the communicating pipe. 

This system is extremely ingenious ; it is in operation at 
Dordrecht and Leyden, and has been experimented on at 
Amsterdam ; but from recent reports it does not appear 
probable that it will be adopted by other towns. 

In a sanitary point of view it would seem primd facie to 
present some questionable features. 

538 Removal of Refuse. 

No current of air, however rapid, can be so rapid as the rate 
at which gases diffuse themselves ; there must be always some 
film of filth left on the sides of the pipes, however perfect 
the vacuum ; it is moreover stated that in practice filth hangs 
about the pans of the closets. Consequently, if any putrefying 
action went on in the pipes, decomposing matter or injurious 
gases might be liable to pass into the houses, if not in spite 
of the vacuum in the pipes and reservoirs, at all events when 
the vacuum is not in operation. 

This method of removal of excreta necessarily requires 
a system of sewers, in addition, to carry off the water from 
baths, washing, kitchens, &c., and rain water. 

It follows from these observations that except in cases where 
some special conditions prevail, it will be more convenient in 
towns, that the drains, which must be provided to remove 
waste water and water fouled in daily use, should also be 
employed to carry off the excreta and similar refuse which 
requires immediate removal from the houses. 



The removal from a dwelling of the water which has been 
polluted in the processes of daily life is a separate question 
from that as to how the refuse water is to be dealt with after 

The chief objects of a perfect system of house drainage 
are — 

1. The immediate and complete removal from the house of 
all foul and effete matter directly it is produced. 

2. The prevention of any back current of foul air into the 
house through the pipes or drains which are used for removing 
the foul matter. 

The first object, viz. removal of foul matter, can commonly 
be best attained by the water-closet system when carried out 
in its integrity, though in certain cases a system of earth, or 
other form of dry closet, may be preferable. Cesspools in a 
house do not fulfil this condition of immediate removal. They 
serve for the retention of excremental and other matters, 
and they are inadmissible where complete removal can be 
effected. Where this is not possible, and cesspools must be 
had recourse to, they must be carefully designed ; they must 
be placed outside, and as far removed from the immediate 
neighbourhood of the dwelling as circumstances will allow. 


Drains in a Dwelling. 

It has been the practice to cany 
the overflow from cisterns into the 
soil-pipe from the water-closet, with 
'ater-trap between. , 
lorecver, it has also been largely 
practice to carry the soil-pipe into 
rain, and to connect that drain 
[i a sewer; providing as the only 
ins for preventing the gases from 
sewer from entering into the house, 
t a 3ap at the end of the drain at 
:ommunication with the sewer, and 
:t,trapsbelow the water-closet basin. 
Such an arrangement is extremely 
igerous. The flaps, even if they 
ed well, would allow sewer-gas to 
s into the house drain each time 
t a flap was opened by the pressure 
the water in its effort to pass from 
house drain into the sewer. 
!n inspecting sewers it will be fre- 
;ntly noticed that a bit of straw or 
)er will lodge under the flap, and 
;p it slightly opened. 
?ewer gas, thus entering, would pass 
the drain to the soil-pipe, where, 

each time that a Volume 

of water was thrown 
JOT^-. down, some of it would 

a '■!.: necessarily be forced 

through one or other 
KA^^___of the traps, to make 
^ way for the inflowing 

Fig- 50- 

Drains in a Dwelling. 241 

Eut independently of this, the capacity of water to absorb 
sewer-gas is very great, consequently the water in the trap 
would absorb this gas. When the water became warm from 
increase of temperature it would give out the gas into the 
house ; when it cooled down at night it would again absorb 
more gas from the soil-pipe : and frequent change of tem- 
perature would cause it to give out and re-absorb the gas 

Hence if the trapped waste-pipe from a cistern is passed 
into the soil pipe it will allow sewer-gas to pass through, and 
be absorbed by the water in the cistern ; but many cisterns 
have a waste-pipe directly communicating with the soil-pipe, 
without any trap between. 

Thus a water-trap without other precautions is but a frail 
protection against this very insidious and dangerous enemy, 
sewer-gas, especially in a concentrated condition. 

There is, moreover, another element of danger in the 
water-trap. When a body of water is thrown down a closed 
soil-pipe it tends to draw in air, and is followed by the water 
from all the traps communicating with the soil-pipe, and thus 
the trap is frequently left without water. 

To prevent these evils the followii^ points should be 
attended to in house drainage : — 

I. General arrangement and position of drains. 

The first object to be attained in house drainage is 
to prevent the sewer-gas from pass- 
ing from the main sewer into the 
house drains. The flap will not pre- 
vent it, and therefore a water-trap 
should be interposed. This water- 
trap should have as little surface as 
possible. ^^-S'- 

In large houses the break of connection with the sewer 
may be by a disconnecting manhole (see Fig. 51). 

In small houses it may be arranged to take a branch 

242 Drains in a Dwelling. 

pipe off the house drain, on the house side of the syphon, 
and to bring this branch up to the surface of the ground, 
as shewn in Fig. 50; in cases where this branch can be 
laid at an angle of about 45^ it will not only act as a 
ventilator, but the syphon can be cleaned from it in case of 

The house drain if closely sealed by water traps will become 
a reservoir for the sewer-gas which may pass through this 
water-trap. It is therefore necessary in the next place to 
oxidise or aerate this gas. To do this we must ensure that 
a current of air shall be continually passing through the 
drains ; both an inlet and an outlet for fresh air must be 
provided in the portions of the house drain which are cut off 
from the main sewer, for without an inlet and outlet there 
can be no efficient ventilation. This outlet and inlet can 
be obtained in the following manner. In the first place, an 
outlet may be formed by prolonging the soil-pipe at its full 
diameter and with an open top to above the roof, in a 
position away from windows, skylights, or chimneys ; and 
every branch pipe connected with a soil-pipe should be 
similarly carried up, and be left open at the top for ventila- 
tion. And secondly, an inlet may be obtained by an opening 
into the house drain, on the dwelling side of and close to 
the trap, by means of the disconnecting manhole or branch 
pipe before mentioned, or where necessary by carrying up 
the inlet by means of a ventilating pipe to above the roof. 
The inlet should be equal in area to the drain-pipe, and 
not in any case less than four inches in diameter. In an 
open situation, and at a distance from a house, the pipe 
might be dispensed with. If the opening is low down 
and near openings into the house, it may be covered by 
a charcoal air-filter of an area at least six times that of the 
pipe, which should be carefully protected from wet, so that 
its action may not be impeded. This is advisable to prevent 
smell in case there should be deposit in the pipes, and in 

Drains in a Dwelling. 243 

case circumstances should arise to reverse the ventilation and 
convert the inlet temporarily into an outlet, or vice versA. 

Much of the value of these arrangements is lost, if the 
house drains be laid so as to allow of deposit ; because if 
deposits occur in the drains or traps, they will putrefy and 
develop sewer-gas. But if the house drain and soil-pipe 
have a sufficient uniform fall to remove the refuse at once, 
and if they be well ventilated — and the more the openings 
the better — ^very little smell will arise. Indeed soil-pipes and 
drains cannot be too open or too much ventilated, provided 
always that they are cut off by a water-trap from the main 
sewer, which should always be ventilated also. 

Flues in walls should not be used for sewer or drain venti- 
lation ; the sewer-gas would be liable to permeate the wall 
and pass into the house. All pipes or openings for ventilation 
should be external. Ventilating pipes should be carried 
upwards without angles or horizontal lengths, and with air- 
tight joints. The, upper end should not be near windows or 
ventilating openings, and in confined spaces should be carried 
above the level of the ridge of the roof. 

The best position for a soil-pipe is outside the house; 
because any escape of sewer-gas resulting from defective joints, 
corrosion, or otherwise would take place in the open air. 

In very cold climates, or very exposed localities, pipes so 
placed would require protection from the frost. 

The mean temperature of the internal air in town sewers in 
this country from a year's experiments was found to be about 
I2&° higher than the external air in winter, and 3° cooler than 
the outer air in summer. The temperature of sewage is always 
above freezing point, and in this country danger from frost 
might easily be guarded against. 

It is unnecessary for waste-pipes from baths, lavatories, or 
sinks (except slop sinks for urine, or water-closet sinks, such 
as are used in hospitals) to pass direct into the drain. 

Similarly, it is unnecessary and undesirable for waste-pipes 

R 2, 

244 Drains in a Dwelling. 

from cisterns or rain-water tanks to communicate with drains. 
Ih a well-arranged set of cisterns in a house, the overflow 
from the upper cistern will fill the lower ones consecutively, 
so that one waste from the lower cistern would suffice, and 
this one should deliver into the open air. Moreover, to prevent 
the possibility of the drinking water becoming polluted by 
sewer-gas, which might find its way up the pipe which supplies 
water to the closet-pan, each water-closet should have its own 
special cistern, supplied by a ball-cock from the principal 
cistern, and it should not be possible to draw water from any 
cistern supplying a water-closet for any other purpose than 
the supply of such water-closet. 

It should therefore be an absolute rule that no overflow or 
waste-pipe from any cistern or rain-water tank^ or from any 
sink other than a slop-sink for urine, or slop-sinks for hospital 
use, or from any bath or lavatory, or safe of a bath^ or of a 
water-closet, or of a lavatory, should pass directly to any 
drain, soil-pipe, or trap of a water-closet ; but every such pipe 
should pass through the wall to the outside of the house and 
discharge near or over trapped gullies with an end open to 
the air, or deliver into a pipe which so passes and discharges. 
But all waste-pipes should be properly trapped close to the 
sink or lavatory or bath, because any pipe through which 
fouled water passes at intervals will have its interior coated 
with deposit, which will occasion a smell. This smell will 
be removed by fresh air, therefore a current of air should be 
maintained in all waste-pipes which are open at the bottom, 
by means of an opening at the upper part as far away from 
windows as can be conveniently managed. Care must be 
taken to protect waste-pipes against frost. 

Sinks and water-closets should be placed against external 
walls, so that the pipes conveying refuse water or soil may 
be more immediately passed outside the main wall, and so 
that the rooms or closets so appropriated may have windows 
opening directly to the air. 

Drains in a Dwelling. 545 

Water-closets or sinks situated within houses which have 
no means of direct daylight and external air-ventilation, are 
liable to become nuisances, and may be injurious to health ; 
and if such sinks and water-closets cannot be ventilated in an 
efficient manner they had better be removed. 

%, Constructional requirements in laying house drains. 

Every drain should be laid in a straight line with proper 
falls and true gradients. All drains and soil-pipes should be 
round and not angular. Round pipes are more self-cleansing 
than angular pipes, and the circumference being smaller in 
proportion to the area, they cause less friction in the passage 
of the fluid. 

Drain-pipes as distinguished from soil-pipes and other 
down pipes, are generally (i) o£ glazed stone-ware, with 
asphalte joints for inside work and cement joints for outside 
work ; (a) of concrete cemented inside and with cement joints ; 
or (3) of cast iron pipes coated inside with tar, with lead 
water-tight joints. Lead joints can only be made in a strong 
iron pipe; and the use of these joints is to some extent a 
guarantee of soundness, but every iron pipe should be care- 
fully tested by water pressure to see that there are no holes 
or flaws. 

The use of cast iron for house drains, if the Cast iron is 
solid, sound, and free from porosity, will prevent leakage and 
subsoil tainting beneath the house, and will be as cheap as 
earthenware pipes in many cases. 

In laying pipes in the ground, sufficient space should be cut 
out in the ground at the joints to allow of a good joint being 
made, and examined before being closed in ; to ensure 
accuracy in laying, the pipes for house drains should always 
be bedded in concrete. No right-angled junction should be 
allowed, except in the case of a drain discharging into a 
vertical shaft. It is advisable to construct access channels 
with manholes at every change of direction, to admit of 
examination of the drain. 

246 Drains in a Dtvelling. 

No drain should pass through and under a dwelling-house, 
except where absolutely necessary. When it must so pass, it 
should be laid in a straight line under the house, between 
disconnecting manholes placed outside the house, arranged 
for cleansing and examination ; the portion of the drain 
which is under the house should be laid in an air drain, 
so constructed as to be as impermeable as possible, open 
to the air at each end, and allowing access to the drain- 
pipes when necessary. Where possible, all connections with 
drains should be made outside a house. 

In laying the drain-pipes proper junctions must be pro- 
vided in all drains for sink and water-closet drainage, and 
for drain ventilation; because communications with drains 
without proper junctions lead to deposit and leakage. 

For an ordinary house a 4-inch drain-pipe is ample — a 6- 
inch pipe would be required for a large hotel — a 9-inch pipe 
would suffice for a very large building. To ensure freedom 
from deposit pipes for moderate sized houses should never be 
larger than six inches. 

A fall of I in 40 is a good fall for a 4-inch drain, 
„ I in 60 „ „ 6 

I in 80 „ „ 9 

If from insufficient fall or other causes the drain cannot 
be kept clear of deposit without flushing, special flushing 
arrangements must be provided. Leaving taps open, and 
propping the handles of water-closets, will not flush drains, but 
will only waste water. A large flush of water is required 
for flushing drains. Where water is plentiful, a discharge 
direct from a large cistern provided for the purpose may be 
obtained. Where water is limited, the drains may be flushed 
by introducing small paddles in grooves formed at the access 
or junction chambers, and suddenly releasing the pent-up 
water ; or where the flow of sewage is small, a tumbler flush- 
ing box or Field's self-acting flush-tank may be used. 

Mr. Roger Field's flush-tank is designed for flushing 

Drains in a Dwelling, 


Fig. 52. 

house drains, but may of course also be used for flushing 
sewers. The syphon is so constructed as to be put in action 
by a very small constant flow of water. This is efiected by 
making the discharging limb of the syphon and its connection 
with the bend of the syphon in such a way that when a small 
quantity of water flows over the bend, it falls clear of the 
sides, instead of running down along the sides of the dis- 
charging limb. 

By this arrangement, the water as it 
drops carries away the air in the dis- 
charging limb, and thereby starts the 
syphon. The longer limb of the syphon 
must be dipped into water below the 
tank. A convenient arrangement for this 
purpose is an annular form of syphon, as 
shown in the drawing. By this means 
a large syphon can be put in action by 
a very small flow of water, so that a 
large flushing tank may be fed, for 
instance, by the constant flow from a 
small water-tap which would take a day 
or two to fill it, and yet the tank will 
discharge itself automatically as soon as 
it is full. 

The syphon is capable of being easily 
taken to pieces in case of any stoppage 
occurring. The outer case can be 
readily lifted off", and as readily replaced 
for the purpose of examining or cleansing 
the syphon. 

Rain-water pipes should deliver into 
an open channel or over a gully, or in 
some other way so as to have the dis- 
charging end open to the air. In special cases they may be 
connected with the drain so as to be used as ventilating 

Fig. 53. 

248 Drains in a Dwelling. 

pipes, provided their upper extremities are situated at such 
a distance from windows, openings, or projecting eaves, as 
to ensure that there is no danger of escape of foul air into the 
interior of the house from the pipes. When so used the joints 
should be air-tight. 

All inlets to drains should be properly trapped, except 
where left open for ventilation of the drains. 

The gullies to receive discharge from waste-pipes or rain- 
water pipes, surface drainage or otherwise, should be trapped 
(see Fig. 54, $5) : they should be placed at a distance of a 
yard at least from windows. The waste-pipe from a sink 
may discharge under a grating for appearance sake. 

Fig. 54- Fig. 55- 

The surface of water exposed in a trapped gully should be 
as small as possible, compatible with facilities of cleansing, so 
as to limit the surface for evaporation. The gully at which 
the waste-pipe from a scullery sink discharges should be pro- 
vided with a grease-trap outside the house ; otherwise grease 
may choke the drains. 

Gullies should not be placed inside a house, in cellars, base- 
ments, or otherwise, unless absolutely necessary. Where such 
gully cannot be avoided, it should be properly trapped, and 
the outlet pipes should not pass directly to any drain or 
sewer, but should be disconnected therefrom by passing 
through the wall to the outside of the house, and there de- 
livering with an end open to the air over a suitable trap or in 

Drains in a Dwelling. 249 

a disconnecting manhole (Fig. 56). Where the basement is so 
deep that this could not be arranged, it would be advisable to 
run the drain to a small brick and cemented well-hole to be 
pumped out daily, or when occasion might require, so as to 
avoid direct communication with a drain. 

When from the damp- 
ness of a site it is neces- 
sary to lay subsoil or 
land drains, and to dis- 
charge the water into a 
sewage drain, the sub- 
soil or land drain should 
not pass directly to the 
sewage drain or sewer, 
but should deliver into 
a disconnecting man- 
hole, as in Fig. 51, 56. 
■ All inlets to the 
drains and all openings 
for ventilation should be 
efficiently protected by 
gratings, to prevent the 
introduction of improper 
substances. When not 
so protected the gully 

should be arranged (Fig. y\%, s6. 

51) so that stones, sand, 

or other heavy matters falling in will drop to the bottom clear 
of the trap and entrance to the drain ; and the gratings 
should be moveable, to allow of such heavy matters being 
easily removed from the bottom of the gully, and they should 
be cleaned out periodically. 

Gratings to inlets, or openings used for ventilation or dis- 
connection, should have a free air-space at least equal in area 
to that of the drain-pipe or gully at the place. 

250 Drains in a Dwelling. 

Gratings should be made of a form to allow the water to 
pass freely, and to prevent dirt from lodging and choking 
the passage for the water. Circular holes are least adapted 
of any form for allowing water to pass. 
3. Soil and waste pipes. 

Soil pipes carry away the discharge from water-closets and 
from slop-sinks. 

A 4-inch diameter soil-pipe is sufficient for ordinary sized 
houses ; if many closets are on the same pipe then a 4|-inch 
or 5-inch may be necessary, very rarely a larger size. Solid 
drawn lead pipes are preferable. 6 lbs. lead is the least 
thickness admissible. 7 or 8 lbs. lead, and more, with the 
larger pipes, should be used in all cases where hot water is 
occasionally thrown down, being better calculated to resist 
the alternate expansion and contraction. 

Iron pipes are sometimes used, but the making of a perfect 
joint is more certain with lead than with iron ; and cast iron, 
if used, must be sound, and free from porosity. Joints in iron 
pipes should be made with lead. Cement joints cannot be 
relied on, either for lead or iron pipes, as they may crack. 
The down pipe, whether lead or iron, must be properly sup- 
ported at intervals, so that its weight or the jar of the flow of 
water shall not tend to move it and open the joints, either 
at the junction with the closets or at the bottom, and thus 
allow of a leak of sewer-gas. 

Lead soil-pipes are usually supported by blocks at every ten 
feet, but intermediate supports by means of a flange or a tack 
soldered to the pipes are desirable. Soil or down pipes placed 
inside a house should be placed in a space built in the wall, of 
sufficient size to allow of the joints being well made and 
examined. If placed in chases cut in the wall, the sides of 
the chase should be carefully cemented and smooth. Casings 
made to open should be provided to the chases. But it is far 
preferable that where possible the soil-pipes should be entirely 
outside the house ; so that if a leak should occur, it will not 

Drains in a Dwelling. ^ 251 

do much harm ; because there is always danger of concealed 
lead pipes being injured by nails being driven into them 

The connection with the drain should be outside, because 
the junction of the vertical soil-pipe with the house drain is 
a place at which there is always risk of cracks, which would 
allow of the leakage of sewer-gas. On this account it is 
advisable when a lead or an iron soil-pipe must terminate in 
an earthenware drain-pipe inside a house that the bottom of 
the soil-pipe should discharge into a trap in the earthenware 
pipe. In such a case the horizontal drain and the soil-pipe 
should each have independent adequate ventilation by an 
inlet and outlet for air. 

The waste from a cistern should be of a size capable of 
emptying the cistern very rapidly from its lowest point, so 
that the cistern may be easily cleaned. Cisterns should be 
throughly cleaned out at least once a month. The water 
from the cistern should not discharge directly over or into a 
gully, for fear the trap should be dry and foul air pass up the 
waste into the cistern. 

Waste-pipes from baths, as well as the valves, should be at 
least 2 inches diameter, and from basins at least i^ inches 
diameter for comfort. Lead waste-pipes carrying away hot 
water, should be of at least 8 lbs. lead. Where branch wastes 
ce^mmifeate with a main waste-pipe, bends in the branch 
waste are desiraBleTo allow for the expansion and contraction 
of the wastes. 

4. Fittings for water-closets. 

Many kinds of water-closet apparatus and of so-called 
' traps ' have a tendency to retain foul matter in the house, 
and therefore in reality partake more or less of the nature of 
small cesspools ; and nuisances are frequently attributed to 
the ingress of * sewer-gas' which have nothing whatever to do 
with the sewers, but arise from foul air generated in the house 
drains and internal fittings. 

252 Drains in a Dwelling. 

The ordinary form is a basin with a hole below, which is 
sealed by a moveable lower basin, which holds the water. A 
lever movement is required to. drop the second basin, so that 
its contents may pass into the drain ; but, in order to prevent 
the sewer-gas from passing through the hole in the first basin 
at the moment of discharge of the contents, and up into the 
closet, and also in order to prevent the sewer-gas from pass- 
ing, as it would do continuously, round the lever apparatus 
into the closet, the contents of the moveable or second basin 
are dropped into a receptacle with a D trap. 

This trap is out of sight, and it is necessarily rather large, 
and from its shape it is liable to accumulate much foul putres- 
cent matter. To remedy this hundreds of forms of water- 
closet pans have been introduced, of more or less merit. 

In the selection of a pan, the object to be attained is to 
take that form in which all the parts of the trap can be easily 
examined and cleaned, in which both the pan and the trap 
will be washed clean by the water at each discharge, and in 
which the lever movement of the handle will not allow of the 
passage of sewer-gas. 

The trap is, however, useless if any escape of gas can take 
place at its junction with the waste-pipe ; consequently the 
trap should always form a portion of the soil or waste-pipe, 
instead of a part of the fitting, and the water-closet or sink 
should be capable of being moved without disturbing the trap. 
Lead is the best material in a house for traps and waste-pipes, 
because it is smooth, durable, free from corrosion, and joints 
can be made easily and safely in lead. For outside work 
stone-ware may be used. The best form of lead trap is a 
smooth cast lead trap without corners. When such a volume 
of water is thrown into a trap as completely to fill it, the trap 
will act like a syphon and empty the pipe, leaving the trap 
without water ; consequently it is desirable to have a small 
ventilating pipe at the bend beyond the trap carried into the 
main ventilating pipe or into the open air, to prevent the 

Drains in a Dwelling. 253 

syphoning action. Lead traps through which hot water passes 
should be of 8 lbs. lead. No trap should be of less than 6 lbs. 

The depth of dip of a trap varies from half an inch to as 
much as 3^ inches. It should depend on the frequency of use. 
When rarely used more depth of water is required, to prevent 
the trap failing from evaporation. Openings for cleaning 
traps should be below the water-level of the trap, then they 
show by a leak if they are not tight. 
5. Cesspits. 

The arrangements described have special reference to the 
case of sewers ; but in certain cases, such as private houses, or 
small detached barracks, it is convenient, and indeed often 
necessary to provide a cesspit, to receive and retain the 
refuse for a certain time. 

The old form of cesspit was to dig a hole in the ground ; 
if porous ground, so much the better ; and to pour into this 
the whole foul refuse of the house or barracks, until its over- 
flow compelled its cleaning out. Numerous cesspits of this 
sort used to exist in barracks, which the porous soil had so 


favoured that there had been no necessity to empty them out 
for many years. These cesspits consequently saturated the 
soil and polluted the adjacent wells. 

From what has been said of the movement of ground air, it 
is apparent how dangerous this mode of dealing with refuse 
must be, especially if the cesspit is near the house. Of 
course the risk of danger to a house from this pollution of 
ground air from the cesspit will increase with the depth of the 
cesspit. But in addition to the risk of pollution of the ground 
air, there is the risk of pollution of water in adjacent wells. 

If it is necessary at any place to turn the refuse into cess- 
pits, the cesspits should be made impervious by puddling 
outside, and lining with brickwork in cement ; they should be 
as small in size as circumstances allow, so as to necessitate 
frequent cleansing. 

254 Drains in a Dwelling. 

It is essential that the gases geaerated in the cesspit 
should not pass into the house drain. 

The cesspit must be covered, but if necessarily near a 
dwelling, it should have ventilation through a lai^e charcoal 
filtering screen, carefully protected from wet ; If away from 
dwellings, large openings protected from rain would suffice ; 
and there should be a water-trap between the cesspit and 
the house ; and between this trap and the house, the ventila- 
tion of the house drains should be arranged on the principles 
already described, so as to ensure that any gases which pass 
the trap may have an opportunity of being diluted with fresh 

When cesspits, or sewage filters have to be provided for 
barracks, as frequently occurs in the new Depot Brigade 

Fig. 57. VentiEator for Cesspool. 

Barracks, they should invariably be outside the barrack en- 
closure, and removed from the barracks as far as the ground 
in possession of the Government will allow. 

In these suggestions it is assumed that the sewers or cess- 
pits from which the house drain is cut off, are themselves ade- 
quately ventilated, for otherwise the precautions suggested 
would be much less effective. 
6, Examination of drains. 

It will be seen from these few remarks that in the question 
of house drainage safety depends quite as much on the work- 
man who executes the work as on the architect or engineer 

Drains in a Dwelling. 255 

who designs it ; and this part of the subject would not be 
complete without a few observations on the examination of 
house drains. 

In examining a drain the main drain should be opened at 
the point where it leaves the house, and by pouring water 
down the various closets, sinks, bell-traps, and rain-water 
pipes, the various pipes may be traced, and an opinion formed 
as to the state of the drains. The velocity of the water flow- 
ing in the drain may be ascertained by pouring water down 
at any point, and accurately noting the time it takes to reach 
the opening. After examining the basement, every closet, 
sink, bath, &c., should be examined in detail. To do this it 
is necessary to have the wooden casings, seats of closets, Sc, 
removed, to trace how each closet, sink, bath, &c., is supplied 
with water, how the soil-pipes from the closets, and the waste- 
pipes from sinks, baths, overflows of cisterns, waste to safes, 
&c., discharge. It can be determined whether the drains are 
trapped off* from the main sewer by ascertaining if there is 
any draught from the bell-traps in the sinks. 

To test the drains, the fumes of ether or of sulphur may be 
used. If ether is poured down a soil-pipe the fumes will be 
perceptible in the house at any leaks in the soil-pipe or 
failures in the traps. Sulphur fumes may be applied by 
putting into an opening made in the lowest part of the drain 
an iron pan containing a few live coals, and throwing one or 
more handfuls of sulphur upon the coals, and closing up the 
opening to the drain with clay or otherwise. The fumes will 
soon be very perceptible at any leaks or rat-holes in the soil- 
pipe, drains, or traps. 




Whatever may be the system adopted for the removal of 
the excreta, there will always be a large amount of liquid 
refuse to be dealt with which is liable to putrefy, and conse- 
quently; in providing for the removal of this, it is generally 
more economical to arrange at the same time for removing 
the excreta, than to institute a separate service for each. 

In the removal of the foul water, the question is often 
mooted of providing for the water used in households only, 
and making a separate service for rainfall. 

This question is one which has obtained considerable 
importance from the theoretical views which have sometimes 
been permitted to decide it. 

A strict rule cannot be laid down. In the case of a detached 
house or of a barrack, or of a small collection of houses, the 
conditions are very different from those in large or even 
moderate sized towns. 

In the case of detached houses and barracks, it is frequently 
found advisable to save the water which falls on roofs, and to 
use it for washing and other purposes requiring soft water. 
But when, as in a large town, the ground is closely occupied 
by houses, the rain-water, as it falls, is fouled by soot ; and it 

Conditions to be observed in Main Sewerage Works. 257 

moreover takes up many other impurities from the atmo* 
sphere, and is unfit for domestic use. 

In a town area, whilst the soot and the impurities of the 
atmosphere which the rain-water takes up in falling frequently 
render it unfit for domestic purposes, that which falls on 
streets and yards is made quite foul by the large quantities 
of horse-dung and other impurities which are the chief consti* 
tuents of the street dust in towns. 

Experiments made by Professor Way some years ago on 
the rain-water which passed into sewers, showed that whilst a 
small part of the solid matter from paved streets consisted of 
granite, "the bulk was horse-dung ; and he showed that the 
value as manure of street water taken in rainy weather on its 
way to the sewers was as great, or even greater, than the 
ordinary contents of the dry weather sewage. 

It is on these grounds necessary to provide in the drainage 
of a large town, as far as possible, for the removal of the foul 
street and yard water, as well as the house sewage proper. 

If provision were made for the very large amount of rainfall 
which may be expected to occur exceptionally and at rare 
intervals, other elements of inconvenience and even danger to 
health would be introduced. For instance, the small amount 
of the dry weather flow of sewage would necessarily only form 
a trickling and stagnating stream in a sewer calculated to re- 
move the maximum volume of rainfall, and this dry weather flow 
would be liable to create deposits. A large sewer, in which 
deposits can occur, would forni a great reservoir for sewer-gas. 

Where provision is made for rainfall, the nature of the 
area which has to be drained must regulate the proportion 
of rainfall to be removed. For instance, in a town area, 
closely built over, the rain-water will run off rapidly into 
the drain ; in a suburban area, that which falls on garden 
ground will pass off more slowly. The rain-water which 
passes from streets and courts into a sewer during a storm 
immediately succeeding a dry season, is generally richer in 


258 Conditions to be observed in 

manurial value than that which passes in after the rain has 
lasted some time. 

But in any town or district sewered effectively, in which 
the admission of surface water is limited, and in which the 
subsoil waters are excluded from the sewers, the volume of 
sewage will be reduced, and be thus more easily dealt with, 
either by pumping or by gravitating to land for irrigation. 

The excess of rainfall not provided for in sewers must be 
carried off by carefully formed surface drains ; or else by 
storm overflow channels formed in the sewers to pass the 
water into natural watercourses : to effect this object many 
ingenious devices have been designed. 

The flow of water from a drainage area during continued 
heavy rain, as in a wet season, or during a thunderstorm, may 
swell the rivers and natural .streams so as to swamp and 
water-log the lower parts of the sewers. When the outfall is in 
a tidal estuary or in the sea, tides are liable to backwater the 
outlet-sewers. Some tidal rivers in England rise vertically in 
flood as much as 20 feet. In some of these cases pumping is 
resorted to ; in other cases the sewage is retained in tanks, or 
in tank-sewers, until the ebbing of the tide, and no absolute 
rule can be given when one or the other mode shall or shall 
not be adopted. Where an outlet sewer is liable to be back- 
watered, it may sometimes be advisable to keep the invert up 
as much as possible, even at the expense of the gradient, and 
to depend on flushing. 

The question as to whether this plan should be adopted, or 
whether on the other hand the invert should be kept down, 
although liable to be backwatered daily, will depend upon 
whether a sufficient velocity of flow can be obtained during 
the period of maximum discharge to remove any silt which 
may have have been deposited during the stagnation arising 
from the backwater. 

The velocity of flow of water which will move materials is 
regulated by the following considerations : — (i) For objects of 

Main Sewerage Works, 259 

the same character, the velocity required to start them in- 
creases with the mass of the object ; {%) for different objects, 
the velocity required increases with the specific gravity ; 
(3) according as the object assumes a form approaching a 
sphere, the less velocity it takes to move it ; whereas a flat 
object, like slate, requires a considerable current before it 
becomes disturbed from its position ; (4) the object moves at 
a less rate than the current, but when the velocity of current 
increases after an object is in motion, the velocity of such 
object increases in a progressive ratio. 

A velocity which will start an object will (when constantly 
maintained, and no accidental circumstance occurs to prevent 
it) never allow such object to deposit in the stream. 

A velocity of from 2 feet to 2 feet 6 inches per second with 
a continuous flow will remove all objects of the nature of those 
likely to be found in sewers ; but if the flow is intermittent, or 
if a part of the sewer remains occasionally stagnant, a velocity 
of 3 feet would be advisable during a portion at least of the 
period of maximum flow. 

Sewage should not be allowed to acquire a velocity of 
more than four feet per second. Six feet per second will 
move grit and other solids along the sewer invert, with a 
cutting action rapidly destructive of the material of the sewer. 

The sewage of a town or village will consist of waste-water 
and excreta from the houses, and the volume, in round figures, 
may range from 100 to 2^0 gallons per day from each house. 

This flow of sewage is not uniform throughout the day. It 
varies with different places, according to the habits of the 
people. In London, 46 per cent, of the daily dry weather 
flow reaches the outfalls between 11 a.m. and 8 p.m., whilst 
only 21 per cent, flows off between % a.m. and 9 a.m. The 
largest flow per hour amounts to 6 per cent., and the smallest 
to v6 per cent, of the whole daily dry weather discharge. The 
volume of maximum flow regulates the size of the sewer, so 
far as sewage is concerned. 

S % 

26o Conditions to be observed in 

In addition to this, the provision for rainfall has been 
assumed in London at \ of an inch of rainfall to be carried off 
in 24 hours from the urban districts, and | of an inch in Z4. 
hours for the suburban districts. 

As all rainfall cannot be excluded from the sewers, it has 
been suggested to provide for a total amount of not less than 
1000 gallons from each house ; that is to say, for a town of 
1000 houses (5500 population) the sewers should have a deli- 
vering capacity of about 1,000,000 gallons. An outlet-sewer of 
% feet diameter, laid with a fall of 5 feet per mile, will deliver 
upwards of 2,000,000 gallons, flowing a little more than half 
full ; and as provision should be made for increase of popula- 
tion, a sewer of 2 feet diameter may be provided for each 5500 
persons, where no better fall than i in 1000 can be obtained. 
Lesser diameters will answer where there are greater falls. 

A comparatively long line of outlet-sewer may be necessary 
to intercept and take the sewage of several lines of sewers 
to some common outlet, where there are sewage tanks, or 
* sewage-farms ' ; this outlet may, however, be confined to, say, 
three times the ordinary flow of sewage, if storm-water over- 
flows can be provided ; because it would not as a rule be 
desirable to construct the intercepting sewer of sufllicient 
capacity to remove storm-water, .as neither sewage-tanks 
nor sewage-farms can deal satisfactorily with such excessive 
volumes of water. In some cases it may be necessary to 
provide a storm-water sewage-tank ; for this purpose an area 
of low-lying land may be deep drained and embanked, and 
the excess of sewage during rain be discharged on to it and 
left to filter away in the intervals. Such an area would-be 
a rough sewage-filter, to be used during wet seasons. A 
portion of land on the margin of a watercourse may, in some 
cases, be embanked above flood-level for this purpose. 

In laying out the sewerage of a town, it is of importance 
that the water from the highest levels should be carried off* 
independently of the lower levels, when it can be done. 

Main. Sewerage Works. 261 

Gibraltar afforded a striking instance of this (Fig. 58). The 
town is at the foot of the rock — part is situated on the steep 
side of the rock, and part on a flat piece of land, which is 
separated from the sea by the fortifications. 

The sewers were carried down the steep streets, which ran 
from the upper part of the town, across the flat portion of the 
town, and through the line wall into the sea. One of the main 
sewers commenced at a height of about 138 feet above h^h 
water mark, and reached the sea-level after a course of about 
1000 feet. 

Another commenced at a height of 264 feet, and descended 
to the sea-level after a course of about 1500 feet, 500 feet of 
the lower end being nearly horizontal. A third had a lall of 

Fig. 53. Former Sewers at Gibraltar, 

264 feet in 2200 feet. A fourth discharged its water into a 
nearly horizontal main, after a fall of 232 feet in 1550 feet. 

Therefore the rainfall on the surface of the town, and on 
the roofs of the houses, together with the deluge of surface 
water which descended from the highly inclined slopes of the 
rock above, ran with extreme rapidity from all the higher 
levels into the lower and flatter districts, where it used to 
burst the sewers, saturate the subsoil, and interfere with the 
eflicient drainage of all the lower parts of the town. 

The rapid current down the steep part became a slow 
stream in the flat portion, deposited a large sediment, and 
in dry weather became a source of intolerable nuisance. This 

262 Condi ttofis to be observed in 

has now been remedied by an intercepting sewer, which 
carries the water from the upper part of the town to the sea, 
independently of the lower levels. 

Sewers liable to be affected by the rise of tides or land 
floods, as on the sea-shore or on a river, must be so arranged 
that any backing of the sewage shall not injuriously affect the 
sewers and drains within the town. 

The lower districts of London used to suffer from periodical 
flooding in consequence of the water from the high districts 
coming down upon the lower districts ; with the aggravation 
that the sewers delivered through the low level districts only 
at low water. 

The intercepting sewers now laid down in London, carry 
off the water which falls on the upper districts by gravitation, 
whilst the water carried off by the sewers in the lower levels 
has to be pumped. 

The lower portion of any system of sewers below the level 
of high water of the sea or the land floods of an inland river, 
must be cut off from the upper portions, which should have 
an entirely independent outlet, and both systems must be so 
abundantly ventilated that any sewage gases which may be 
formed may pass out at points specially provided for the 
purpose, and not through the drains and into the houses. 

The intercepting system is especially needful in towns 
where the levels vary much, because a sewer which is carried 
down a steep hill becomes a chimney up which the sewer-gas 
flows, and may pass through the drains into the houses at the 
upper part. Consequently, whdh a sewer is necessarily placed 
in such a position, the highest part should be kept open, and 
numerous openings should be provided in its course to allow 
of its complete aeration. 

Means for full and permanent ventilation of town sewers 
and house drains are required to prevent stagnation or con- 
centration of sewage gases within sewers and drains. Ventila- 
tion requires both inlets and outlets : these will be adequately 

Main Sewerage Works. 263 

aflforded by making numerous openings from the sewers to 
the external air, the object being to cause unceasing motion 
and interchange betwixt the outer air and the inner sewer air, 
which will bring about and maintain extreme dilution and 
dispersion of any sewage gas so soon as generated. 

It has been found by experiment that in unventilated 
sewers the concentrated gas becomes deadly, whilst in fully 
ventilated sewers with continued flow without deposit, the 
sewer air is purer than that of stables, or even than that in 
a public room when fully occupied. 

Ordinary main sewer ventilation should be provided for all 
sewers, at intervals not greater than 100 yards ; that is to say, 
not fewer than 18 fixed openings for ventilation should exist 
on each mile of main sewer. 

If sewer air' at any sewer ventilator, or at any other point, 
should be offensive, additional means for prevention of deposit 
and for ventilation on this sewer are required, and should, as 
soon as possible, be supplied. 

Although in properly constructed sewers the ventilation 
should prevent any smell, in old sewers of defective construction 
charcoal filters for the ventilations may sometimes be advisable. 
Where charcoal is used in sewer ventilation it must be under- 
stood that it retards motion, and provision should be made to 
meet this. Charcoal trays, or boxes, for sewer ventilation, 
should never have less than 1000 square inches of surface 
exposed for the passage of sewage gas to each 50 square 
inches of free opening to the outer air. The meshes of a 
charcoal tray may be about Jth of an inch. The charcoal 
(wood) may be about the size of coffee beans, clean sifted, and 
placed carefully in a layer of ^ or 3 inches. It should be 
carefully protected from wet. The vibration of the traffic in 
a street is said to consolidate it so as to check its action. 
Charcoal in a dry state acts the best, but its disinfecting 
property is only diminished by damp — it is not entirely 
destroyed. The charcoal may require, in some places, to be 

264 Conditions to be observed in 

renewed at intervals of six months. It should only be looked 
on as a makeshift; in ordinary cases of sewer ventilation 
charcoal need not be used, as the more readily and freely 
the interchange can take place from the sewer or drain to the 
outer air the better will the ventilation be. 

The upper, or * dead ends,' of all sewers and drains should 
have means provided for full ventilation continued beyond the 
junction of the last house drain. 

For detached houses, villa residences, or larger establish- 
ments, drains should never end at the house which has to be 
drained, but should be continued beyond and above to some 
higher point or ventilating shaft where means for full and 
permanent vehtilation can be provided, so as effectively to 
relieve the house from any chance of sewage gas contamination. 

With the abundant means for ventilation suggested, the air 
within the sewers will be made comparatively pure by dilution 
and oxidation, and the further dilution and dispersion which 
would occur when the gas passes into the open air will 
generally dissipate remaining traces of taint and danger. 

Manholes should have moveable covers at the surface 
of the ground. There should be a side chamber for ventila- 
tion, 'step irons,' to give access to the invert, and a groove 
in the invert and sides to allow of a flushing board being 
inserted at will for flushing purposes. The side chamber 
might be arranged for a charcoal screen or filter, where 
from the defective construction of the sewer this is desirable. 
Details for manhole and side chamber for sewer ventilation 
are given in Figs. 59, 60. 

The ends of all sewers and drains at the lowest outlets must 
be so protected that the wind cannot blow in and force any 
sewage gases back to the streets and houses. Flap-valves, or 
other contrivances, may be provided to cover and protect 
outlet ends of sewers and drains, and so prevent the wind 
blowing in ; or the mouth can be turned down into the 

Main Sewerage Works. 


Where the flow in the sewer is uniform in quantity, or where 
the minimum flow is equal to half the maximum flow, a circular 
form of sewer is best if made of such a size that the depth of 
water m the sewer at the time of maximum flow shall afford 

Live qfR ad 

^.59. Flaoattc 

Fig. 60. Plan at AA. 

the maximum hydraulic mean depth (i.e. the quotient afforded 
by the area of section divided by the wetted perimeter) ; but 
when the flow in a sewer varies very considerably at different 
times of the day, the egg-shaped form of sewer is preferred as 
affording better results with a small flow of sewage. Figs. 61 

266 Conditions to be observed in 

and 6a show the form now generally adopted. Fig. 65 shows 
the section of the river at Brussels, where it is covered over, 
and the intercepting sewer which runs parallel to the river. 
The sewer in this case is formed with a smaller trough at the 
lower part and expanded at the upper part, by which means 
a higher velocity is obtained with the dry weather flow of 

Fig, 63. River and Sewer Conduits at Brussels. 

sewage than would otherwise prevail, and the larger area 
provides for carrying off rainfall ; during the dry weather 
flow a footpath is reserved on each side. When the sewer 
is quite full the overflow passes to the river through side 
flaps, which the pressure forces open. 

Steep gradients in sewers must be modified by forming 
vertical steps, or falls, to prevent the sewj^e, during heavy 
rains, from acquiring such a velocity as shall not only wear 

Main Sewerage Works. 267 

out the invert and blow the joints, but also burst the sewers. 
The steps, or falls, should be so formed as tp prevent any 
accumulation of deposit. Earthenware pipe sewers, when 
laid down steep gradients, should be bedded and jointed with 

Sewers should be watertight. If the water finds its way- 
out of a sewer, the diminished flow causes sediment to be 
deposited, and obstructions to be created, and the subsoil 
round the sewer becomes polluted with sewage matter, con- 
sequently in dry porous subsoils the trench cut for the sewer 
should be made watertight with clay puddle. 

Brick sewers should never be set dry to be grouted — they 
should be set in hydraulic mortar. If sewers are to be water- 
tight so as to exclude subsoil water, the bricks must be set in 
Portland lime mortar, with a lining or coating of mortar 
outside the sewer, and betwixt each 4i-in. course of bricks. 
Wet subsoils may require special land drains. Cast iron pipes 
may be used for main sewers with economy and advantage 
through quick-sands, or where the strata is full of water ; also 
in narrow streets where deep trenches have to be excavated. 
A cast iron sewer may be two-thirds the diameter of an 
earthenware pipe or a brick sewer, as the cast iron pipe 
may work full and even under pressure. 

Main sewers require to be provided with storm overflows in 
case of heavy rains. 

Sewers should not join at right angles, unless the curve and 
extra fall is provided in the manhole. 

Tributary sewers should deliver sewage in the direction of 
the mainflow. Fig. 64. 

Sewers and drains at junctions and curves should have 
extra fall to compensate for friction. 

Sewers of unequal sectional diameters should not join with 
level inverts, but the lesser, or tributary sewer, should have 
a fall into the main sewer at least equal to the difference in 
the sectional diameter. If the inverts of tributary sewers are 

Steiian CD 

Fig. 64. Sewer-junction 3' o" X i' o" into 4' 6" x 3' o" dtowing ade entntnce. 

Conditions to be observed in Main Sewerage Works. 269 

not above, or, at least, are not up to the level of the ordinary 
flow of sewage in the main sewer, such tributary sewers, or 
drains, will be liable to be backwatered, in which case deposit 
will take place in the length of submerged invert, and so the 
tributary sewer, or drain, will become choked with its own silt. 

Pipe drains should have the joints made with cement. In 
bad soils the joints should also be bedded in concrete, a gap 
being cut to receive the concrete so that the pipes may bed 

Sewer pipes of more than eighteen inches diameter are 
troublesome to handle and difficult to joint, and do not 
make such good work as a brick drain. 

Earthenware pipes of equal diameter should not be laid as 
tributaries, but a lesser pipe should be joined on to a greater. 

Side openings for house drains should be provided in the 
original construction of all sewers wherever it is probable they 
will be required. Every effort should be made by careful 
construction to prevent sewage and sewer-gas from passing 
out of the sewer into the subsoil. Where subsoil drainage is 
necessary in connection with sewers, either the subsoil water 
should be carried off by separate drains or the mouth of the 
subsoil drain should terminate in a ventilated disconnecting 
manhole, whence it should pass into the sewer. 

Flushing Sewers. 

Flushing a sewer means accumulating water at some point 
in the sewer in sufficient quantity to pass down and along 
the portion of the sewer below with a rush when suddenly 
liberated, in order that the water may loosen and carry away 
all sediment. 

In a perfect system of sewers, flushing should be unnecessary ; 
but perfection in a sewer depends as much on the excellence 
of workmanship as on the design ; and any imperfections lead 
to deposit and its corresponding evils. 

Every care should be taken to prevent heavy substances 

270 Conditions to be observed in 

entering the sewers by properly formed gullies to catch and 
retain the material ; especially where road washings pass into 
the sewers ; and it may in some exceptional cases be desirable 
to form catch-tanks in the sewer. These catch-tanks should 
be wider than the sewer, so as to cause a check in the flow, 
and thus cause heavy materials to be deposited, and the 
bottom should be below the invert of the sewer so as to retain 
the materials when dropped. The materials thus dropped 
should be dredged out, at intervals, as found necessary. 

As a general rule, however, sewers when true in line and 
gradient, and sound in construction, and provided with side- 
entrances, manholes, and flushing-chambers, at proper inter- 
vals, will permit of silt being flushed through to the outlet. 

As it is difficult to provide a perfect system of sewers, or to 
prevent foreign heavy matters passing in, it is always safe to 
provide a means of flushing. In a system of sewers every 
manhole may be a flushing chamber, so managed as to be 
charged with water for flushing purposes. There should also 
be a flushing chamber at the head of each sewer and drain, 
and every flushing chamber should be permanently ventilated. 

Mr. Roger Field's flush tank, with his automatic syphon 
already described, affords a convenient means of periodically 
flushing drains or moderate sized sewers by the flow of water 
taps or from the overflow of a drinking fountain, or other 
small continuous flow which is generally wasted. 

It is possible to injure sewers by overflushing them. 

In towns where there is a regular water supply the flushing 
chambers may be filled by the flow in the drains, otherwise 
it may be necessary to resort to water carts. 

Springs of water, and the water from canals, reservoirs, 
rivers, and streams, may occasionally be so situated as to be 
easily made available for purposes of sewer-flushing. 

In the London sewers the flushing is effected by flushing- 
gates provided at certain points. In the Paris and Brussels 
sewers a moveable flushing apparatus is provided. 

Main Sewerage Works. 271 

The sewer has a footpath on each side, along which a truck 
is arranged to run as on a railway, as shown in Fig. 6^, The 
truck is provided with a board of the exact section of the 
sewer, which can be raised or lowered by a screw apparatus 
fitted to the truck. When the board is lowered, the water in 

Fig. 65. Moveable flushing apparatus. 

the sewer is dammed back. Two small doors are provided in 
the board near the bottom, and by opening these when the 
sewi^e is dammed back, a strong current of water is passed 
through so as to wash away any accumulation of debris. The 
truck can be moved on to any spot where the action is required. 


^^—- — 

-jg; ^ 


— ^1 


i -i 

Fig. 66. Sewage syphon 

IS under the Seine at Paris. 

The sewers at Paris are carried under the Seine from one 
bank to another by means of pipes laid in the form of an 
inverted syphon. (Fig, 66.) Each pipe is about one metre in 
diameter ; to clean out the pipes a wooden ball of a diameter 

272 Conditions to be observed in Main Sewerage Works. 

a little less than that of the tube is periodically caused to roll 
through the pipe. It delays the flow, and thus accumulates 
a head of water behind it, sufficient to force away any im- 
pediment from the bottom of the syphon. 

There cannot be any strict rule by which to ascertain the 
weight of mud any special town sewage will deposit during a 
day, week, month, or year, as the weather necessarily affects 
the condition of road and street surfaces, and it is (for the 
most part) from dirty yards and street-surfaces that sewage- 
mud is derived. A wet season will, as a rule, give from any 
area more mud than a dry season. The relative weight and 
character of the mud will also be determined by the character 
of the street-surfaces, and the weight and rapidity of the 
traffic. Paved streets produce less mud than macadamised 
roads, whilst good scavenging prevents an excess of mud 
being washed from the street surfaces into and through the 
drains and sewers. 

With respect to wear of streets and roads, rapid traffic 
is destructive, heavy and rapid traffic being most destructive. 
A good foundation for a road or street is of the first im- 
portance, because much of the mud found on pavements 
has worked up from below. Consequently all street surfaces 
should be laid on a bed of concrete : by this means not only 
will surface mud be diminished but the road will last longer ; 
a well-formed road, though more costly to make, will be more 
cheaply maintained. It is also of great importance to uSe 
all proper means to prevent grit and mud being washed into 
drains and sewers. For this purpose all street gullies instead 
of opening directly into the sewers should be provided with 
catch-pits constructed to catch and retain the detritus from 
the street washings. 



The Rivers Pollution Prevention Act, 1876, forbids the 
pollution of rivers and streams by the admission of crude 

At first sight, the simplest mode of getting rid of sewage is 
to place it in the sea. 

There are, however, towns both on the sea-shore and on 
tidal estuaries in which a nuisance is caused by the discharge 
of crude sewage too near to roads, houses, or bathing-places. 

Moreover there are many places where this method of dis- 
posing of sewage is inapplicable, and it would in any case be 
singularly wasteful. 

There is, however, no doubt that water carriage for the 
removal of excreta is the cheapest and most convenient 
system, so far as the transport to a distance is itself con- 
cerned. It is only when the sewage thus removed has to 
be finally disposed of without creating a nuisance that the 
difficulties arise. 

To dispose of sewage therefore either the outlets must be 
extended into the sea, and the sewage so poured iiuthat it shall 
not be returned on to the shore : or else it must be pumped 
inland for irrigation, or the solids must be precipitated, and the 
clarified sewage disinfected, and allowed to flow into a stream 
or river. 

The degree of purity to be required in clarified sewage 


274 Disposal of 

before it is admitted into a river, depends to some extent on 
the degree of purity of the river water into which it is to flow. 
For instance, a different standard of purity might reasonably 
be admitted for water turned into the Thames at Gravesend 
from that required at Oxford or Reading. 

The cost of irrigation or disinfecting and clarifying sewage 
so as to prepare it for passing into a stream must be regarded 
as expenditure of money for cleansing the town to whose 
sewage it is applied. 

The disposal of sewage is a question on which the Sanitary 
Engineer has expended a great share of attention, without 
attaining a result which all parties have been willing to accept 
as a final settlement of that question. One reason of this 
arises from the fact that the question of utilisation of sewage 
has been mixed up with that of removal of sewage. 

The desire to make a profit has, in many cases, become 
paramount, and people have been unwilling to recognise that 
whilst theory shows that the utilisation of sewage in ma- 
nuring the land should repay the cost, in practice there are 
various barriers to financial success. Moreover, the popu- 
lation in different localities is living under very various 
conditions. In one part of the country the population is 
scattered — and plenty of land unoccupied by dwellings can 
be found for sewage irrigation. In other parts the population 
is so dense that there is little room for sewage farms at any 
reasonable distance from towns. In some towns chemical 
products have been allowed to find their way into sewers, and 
this has rendered the sewage unfit to apply to land. 

The inventors of many of the systems for clarifying sewage- 
water have generally advocated their projects on the ground 
that profit will be derived from the manure which is to be 
extracted from the sewage. If there had been no idea of profit, 
the number of inventors would have been more limited, and 
the subject would have received less consideration ; and, there- 
fore, looking at it from this point of view, it has been 

Water-carried Sewage. 275 

advantaigeous for science that the idea of profit has so largely 

On the other hand, the various corporations stayed their 
hands for many years in providing drainage and water supply, 
in order to wait for the discovery of this modern philosopher's 
stone, and thus much sickness, death, and misery have continued 
in the country, which more energetic measures would have 

It is no doubt true that sewage contains matter of enor- 
mous value. The value of the manurial constituents of the 
sewage of London was estimated by Dr. Hofmann in 1857 at 
nearly ;^i,5oo,ooo per annum ; but he added : — * An enquiry 
into the nature of the valuable matter carried off in our 
sewers, an attentive examination of the chemical properties 
of the constituents, together with the consideration of the 
extraordinary and constantly-increasing degree of dilution in 
which they exist, cannot fail to impress the chemist, on purely 
theoretical grounds, with the magnitude of the difficulties 
which oppose themselves to the successful accomplishment 
of the task.' 

Nor do these difficulties diminish, if the question be sub- 
mitted to the test of experiment. 

The valuable constituents of sewage are like the gold in the 
sand of the Rhine ; its aggregate value must be immense, but 
no company has yet succeeded in raising the treasure. 

The character of the sewage of different towns varies with 
the industrial pursuits in those towns; in towns, however, 
where no special trades exist, the sewage is of a uniform 

The limits of this chapter merely permit of a brief allusion 
to some of the processes which have been proposed. 

The schemes for defecating sewage are based on pre- 
cipitating the solid constituents. The object of precipitation 
is to remove in a solid, dry, or semi-dry state the putrescible 
constituents of the sewage, and to render the filtrate or effluent 

T % 

276 Disposal of 

water sufficiently pure to mingle with streams, or to be 
employed for purposes of irrigation. 

There are two points connected with the effluent from 
precipitation processes which ought to be borne in mind. In 
the first place an alkaline effluent must be avoided, because 
wherever there is an alkalinity there is a tendency to putre- 
faction. Where the effluent can be kept acid it is safe. Another 
point, urged many years ago by Professor Heisch, is that 
there must not be phosphoric acid in the effluent, or there 
will be a tendency to produce the low conifervoid growth 
commonly called sewage fungus. That is a difficulty with 
any process which employs phosphates. 

The lime process is the simplest, and was the first to attain 
prominence. It consists of mixing cream of lime with sewage 
in the proportion of about sixteen grains of lime to the gallon. 
The mixture is then allowed to subside, and the clear water 
flows off. With recent sewage, and on a small scale, the 
deodorisation is tolerably complete, but it leaves in the effluent 
water at least four-fifths of the soluble organic matter of 
the sewage. A mixture of lime and sulphate of alumina is 
capable of removing a larger quantity of matter from sewage 
than lime alone, and much more quickly. This process not 
only removes the whole of the suspended matter, but also a 
considerable quantity of dissolved matter, both mineral and 
organic ; while with the lime process a minute quantity even 
of suspended mineral matter is left unremoved. 

The ABC process consists of using alum, blood, clay, and 
charcoal. The clay and charcoal, and where necessary a 
little lime, are finely ground up with water to form an 
emulsion, and mixed with the sewage ; a solution of sulphate 
of alumina is then added. 

The sewage being a slightly alkaline liquid charged with 
nitrogenous organic matter, the sulphate of alumina is decom- 
posed ; the alumina is separated in a flocculent state, and, by 
virtue of its affinity for dissolved organic matter, when it sinks 

Water-carried Sewage. 277 

it drags down with it a corresponding amount of nitrogenous 
impurity : the blood is a liquid charged with albumen ; albu- 
men is coagulated in the presence of alum, and in the same 
way as this coagulability of albumen is utilised in fining wine 
and coffee, so it is made use of in this process to join with 
the alumina in its precipitation, and to assist in the further 
removal of the putrescible constituents out of the solution. 

The immediate action of the sulphate of alumina depends 
on the presence of natural alkali in the sewage produced 
by the decomposition of the urea into carbonate of ammonia ; 
but a great deal of the sulphate of alumina is sometimes 
wasted because the sewage is not sufficiently alkaline to 
precipitate the whole of it : when this is the case lime must 
be added. 

The effluent water is stated to be sufficiently purified for 
admission into a running stream without being a nuisance or 
injurious to public health ; but the question remains as to 
how far the cost of the process can be covered by the value 
of the residuum. 

The phosphate of alumina process consists of employing 
prepared phosphates, the action of which is to curdle, and 
separate the fcecal matter in the sewage, and to add lime to 
separate the soluble phosphates. The solid matter is col- 
lected in precipitation tanks. The effluent is said to be 
sufficiently pure for admission into a river of the purity of 
the lower part of the Thames, where the water is not used 
for drinking purposes. 

The system of the General Sewage and Manure Company, 
or sulphate of alumina process, is first to pass the sewage 
through strainers, the solid matter thus retained is applied 
directly as manure ; the strained sewage is mixed with a solu- 
tion of sulphate of alumina ; after which it receives an addition 
of cream of lime, the mixture being thoroughly agitated after 
the addition of each chemical. 

It is then passed into precipitation tanks, of which it passes 

278 Disposal of 

through three in succession. The effluent water flows in a 
thin sheet from the surface to filtering beds, from which it 
percolates through a depth of five feet of earth and then 
passes into the river in a clear bright condition. 

The filter beds are used intermittently and are planted with 
osiers and rye grass. 

The sludge or deposit recovered from the precipitating 
tanks has to be dried. 

In all the systems of precipitation a large portion of the 
fertilising matter passes off in the effluent water, and the dis- 
posal of the sludge is a great difficulty. In proportion as the 
quantity of the precipitating material is increased, the value of 
the sludge or manure is diminished. It must be dried, and 
unless heat is applied, the space required for its stowage 
would be enormous. General Scott proposed to utilise the 
organic matter in the sludge as fuel, and to produce lime com- 
pounds to be used as cement, or for agricultural purposes. 
He adds to the sewage a sufficient quantity of slaked lime to 
produce a complete precipitation of the acids present in 
sewage, and enough clay to form, with the silica and alumina 
already present in the sewer water, about 20 per cent, of the 
bulk of the calcined sewage sludge. When the sludge is dry 
it is introduced into a kiln with a small quantity of fuel to 
commence the combustion, the organic matter in the sludge 
furnishes the rest of the fuel required. A million gallons 
of ordinary town sewage treated with i^ tons of slaked lime 
yield % tons of Portland cement. 

This brief account of a few of the methods which have been 
adopted for clarifying sewage, sufficiently illustrate this branch 
of the subject. 

Dr. Frankland states that of all the processes which have 
been proposed for the purification of sewage, there is not one 
which is sufficiently effective for rendering the water which 
has been contaminated fit for domestic purposes. But all 
these processes more or less purify the sewage, and render it 

Water-carried Sewage. 279 

more or less fit for turning into a river, or for further puri- 
fication by application to land. 

There is one way, and one way only, in which the fertilising 
matter in sewage can be effectually removed and utilised, and 
by which the effluent water can be rendered sufficiently pure 
to pass into streams. 

That method is by passing the fresh crude sewage through 
the soil by irrigation in thin films, under proper and favour- 
able conditions of land covered with growing crops. The soil 
acts mechanically as a filter, whilst the oxidising action of 
the air in the soil, and the growth of vegetation on the surface 
decomposes and assimilates the organic and other compounds 
in the sewage which may be available as fertilising ingredients. 
By this means the trouble, nuisance and cost of precipitating, 
screening and filtering are avoided, and the whole of the 
fresh sewage, sludge and fluid, is incorporated with the land. 
When the amount of land available is not sufficient to allow of 
this being done, then concentrated land filtration, combined 
possibly with some one of the best forms of chemical treatment 
would have to be resorted to. 

The difficulty of sewage irrigation lies in the impossibility 
of always securing the necessary favourable conditions. 

The best land for a sewage-farm would have a free loamy 
soil, and open subsoil, with a sufficient proportion of clay to 
moderate the percolative powers of its other constituents ; the 
surface tolerably even, having a southern aspect, and gently 
sloping to the south. 

Where clay soils must be resorted to, the drains must be 
frequent, in order to overcome the natural retentiveness of the 
land ; and they should be so laid out as to allow the sewage 
if possible to be applied twice — that is to say, the underdrains 
of the land to which the sewage is first applied should dis- 
charge their effluent upon a lower surface of land, by which 
any impurity retained after the first application may be re- 
moved by the second. 

28o Disposal of 

Clay land requires deep draining, and the surface should 
be well broken up, either by spade labour or by deep steam- 
ploughing. The surface should be deeply and perfectly 
trenched, so as to secure that disintegration of the soil which 
will prevent its cracking, and give it a uniform filtering power ; 
the drains must be so laid and protected as to remove subsoil- 
water, after filtration through the soil ; they should not allow 
either unfiltered surface-water or sewage to pass through 
cracks direct to the drains. 

Clay soil, properly subsoiled and underdrained, would pro- 
duce better crops than light sandy soils. The reason is 
simple. In clay soils there are found in great abundance all 
those mineral matters which plants take up as food ; whereas, 
there is a deficiency of mineral plant food in light soils, and 
the excess of nitrogenous matters in sewage produces over 
luxuriance of growth in the plants grown on light sandy soils. 
In clay soils the excess of mineral matters counterbalances the 
excess of nitrogenous compounds in sewage, and more healthy 
and heavier crops can be obtained, by means of sewage irriga- 
tion, on well drained and well worked clay soils than on light 
sandy soils. 

Under-drainage and surface preparation are necessary, the 
one to prevent wetness and the other to secure an even distri- 
bution of the sewage over the surface, and each must be regu- 
lated by the character of the soil. 

Drainage can never do harm, while if the subsoil partakes 
of a retentive character drainage is indispensable. In soils 
comparatively free in character but few drains are required. 

The inclination which it is necessary to give to the surface 
of land prepared for irrigation will entirely depend upon the 
character of the soil and subsoil. The area of a sewage farm 
must be laid out in slopes, regulated in size and position by 
the configuration of the surface and the degree of porosity the 
soil possesses. With very free soils, a gradient of i in %^^ or 
as much as i in ao, may be necessary to gain an overflow 

Water-carried Sewage, 281 

which will cover the surface and prevent excessive absorption, 
whilst on very suitable soils a flat slope of i in 150 may be 
a workable gradient ; the sewage-carriers should be shallow 
trenches carried along the ridge nearly level, and made to 
overflow evenly over the surface by damming up at intervals. 
Whenever their position is likely to be permanent, they should 
be made of bricks, stone, or concrete. 

Every area, however rough or uneven, may have level 
contour lines set out over its entire surface, so that by forming 
conduits on these contour lines the surface may be irrigated. 

Land having an irregular and steeply sloping surface may 
have sewage-intercepting drains and carriers so arranged as to 
intercept the sewage from the upper areas and bring it over 
the lower areas a second or a third time, by such means more 
eff"ectually purifying the sewage. 

Roads over a sewage farm should, where practicable, be along 
the fences ;. costly permanent roads are not required. When 
land has been properly prepared for the reception of sewage, 
it may be irrigated in all weathers, so as to purify the sewage. 

A wet season does not necessarily injure a sewage farm if 
the means of removing and consuming the produce are equal 
to the growth of the crops. 

One gallon of sewage weighs 10 lbs.; %%^ gallons, or 3^240 
lbs. are one ton ; 22,400 gallons, or 224,000 lbs. are 100 tons 
— equal to nearly one inch in depth over one acre of land. 

A dressing of liquid sewage half an inch deep amounts to 
nearly 50 tons per acre, a quantity more than sufficient to 
fertilise any growing crop. 

Ten inches equals 1000 tons, and 12,000 tons per acre per 
annum equals 120 inches in depth, and this volume may be 
used on well-prepared land without swamping it, as land will 
filter several inches in depth per day when the sewage is 
equally and evenly distributed. Looked at from a productive 
point of view, this is a wasteful application of sewage, but it 
answers as a means of purification. 

282 Disposal of 

Italian rye grass will dispose of a larger quantity of sewage 
than almost any crop, and give heavy crops if the roots are 
young. The greatest producing power will be in the first 
year's growth. A second year is probably the utmost length 
of time it should be in the ground. 

No larger area of Italian rye grass should be sown than 
will admit of the grass upon it being disposed of in the 
district, as it will not keep, nor will it bear distant carriage. 
Sewage-grown grass will however make good and wholesome 
hay if the season will permit, or if the grass can be artificially 

To give a sewage farm the chance of paying, the land must 
be obtained at a reasonable price, and the cost of preparation 
must be moderate; there must also be reasonable skill in 
cropping, in cultivation, and in management; under such 
conditions land irrigated with sewage ought to pay a reason- 
able rent. If steam power has to be used for pumping the 
sewage, this of course must be paid for in addition. 

Sewage has been valued as a manure at from \d. to id, per 
ton. The same sewage will, however, be worth 2d, in a dry 
summer, which may not be worth even a halfpenny to the 
farmer in a wet season and through the winter. 

In estimating the quantity of land to be allotted to a sewage 
farm it would be advisable as a rule to take one acre for 100 

The sewage has to be applied carefully, so as not to injure the 
growing plants ; it cannot be continually applied, as the matter 
in suspension has a tendency to choke the pores of the soil. 
It must be stored somewhere, to be applied at proper times ; 
and when not applied at once the suspended matter should 
be, as far as possible, removed. Moreover, it should not be 
applied in proximity to dwellings. 

These limitations have led to the adoption, in places where 
land free from building is difficult to be obtained, to the other 
system of direct application of sewage to land, viz. that of 

Water-carried Sewage. 283 

intermittent filtration. In this case one acre is allotted to 
1000 persons. 

Whilst irrigation may be described as the distribution of 
the sewage without supersaturation of the land, having in view 
the production of a maximum growth of vegetation, inter- 
mittent filtration is the concentration of the sewage at short 
intervals on as few acres of land as will absorb and cleanse it, 
without excluding the production of vegetation at the same 

The best soil for this purpose is a finely comminuted soil, 
with a great power of seizing on the fertilising substances in 
the sewage. The drains should afford an effective aeration of 
the soil to a depth of at least six feet. To secure this the 
drains should be somewhat deeper than that — say a foot if 
possible. Then under every square yard of surface there will 
exist two cubic yards of aerated filtering material giving 9680 
cubic yards per acre— say 10,000 yards. This arrangement 
secures the best result ; and at the same time it facilitates 
calculation, every cubic yard having a cleansing power vary- 
ing from 4 to 1 12'4 gallons of sewage per diem. Hence it 
follows that a single acre drained so that the sewage shall pass 
through a thickness of six feet of aerated soil, will purify 
'sewage proper' in quantities varying from a minimum of 
40,000 gallons up to a maximum of 124,000 gallons. 

Where the depth of drainage is reduced to less than six feet, 
the superficial extent must be increased in proportion as the 
depth of under-drainage is diminished, in order to secure the 
necessary quantity of filtering material for purification. 

This brief description conveys a general idea of the system, 
which has been fully explained by Mr. Bailey Denton in his 
Lectures to the Royal Engineers. 

One third of the land is in use for a year at a time. The 
sewage is strained, so as to remove the grosser particles ; it is 
distributed uniformly by carriers over the area, and allowed to 
sink through. 

284 Disposal of 

During the two years when the filtration is not in use the 
land is cropped. 

Present Aspect of the Sewage Question. 

The following are the general conclusions to which a con- 
sideration of the several methods of disposing of sewage 
leads : — 

1. No one system of sewage disposal could be adopted 
universally: the peculiarities of different localities require 
different methods. 

2. With dry systems, where collection at short intervals is 
properly carried out, the result, as regards health, appears to 
be satisfactory. 

3. It is evident that by some of the various processes, based 
upon subsidence, precipitation, or filtration, a sufficiently puri- 
fied effluent can be produced for discharge, without injurious 
result, into water-courses and rivers, provided they are of 
sufficient magnitude to effect a considerable dilution ; and 
that in the case of many towns, where land is not readily 
obtained at a moderate price, those particular processes afford 
the most suitable means of disposing of water-carried sewage. 
It appears, further, that the sludge in a manurial point of view 
is of low and uncertain commercial value ; that the cost of its 
conversion into a valuable manure will preclude the attain- 
ment of any adequate return on the outlay, and even, it may 
be, of the working expenses connected therewith ; that means 
must therefore be found for getting rid of it without reference 
to possible profit. 

4. In certain localities, where land at a reasonable price can 
be procured, with favourable natural gradients, with soil of a 
suitable quality, and in a sufficient quantity, a sewage farm, if 
properly conducted, is apparently the best method of dis- 
posing of water-carried sewage. It is essential, however, to 
bear in mind that a profit should not be looked for by the 

Water-carried Sewage. 285 

locality establishing the sewage farm, and only a moderate 
profit by the farmer. 

5. As a rule no profit can be derived at present from sewage 

6. For health's sake, without consideration of commercial 
profit, sewage and excreta must be got rid of at any cost. 



The conditions which should govern the healthy construction 
of dwellings are embodied in pure air and pure water. Five 
hundred years ago the population of the kingdom was only 
equal to the present population of the metropolis. When the 
first census was taken in 1801, the population of England and 
Wales was less than 9,000,000 ; it has now reached nearly 
25,000,000. We are crowded together as we were never 
crowded before ; our pursuits are more sedentary, our habits 
more luxurious ; houses increase in number ; land is more 
valuable, the green fields more remote; our children are 
reared among bricks and paving-stones; the public health 
can only be maintained by special sanitary appliances and 

It has therefore become impossible in the question of health 
for any one member of a community to separate his interest 
from that of his neighbours. If he places his house away 
from others, the air which he breathes may receive contamin- 
ation from the neighbouring district ; the dirty water which 
he throws away may pollute the stream from which his 
neighbours draw their supply ; and when a population con- 
gregates into towns, the influence of the proceedings of each 
individual on his neighbour becomes strongly apparent. 

Conclusion. 287 

In places where many dwellings are congregated together, 
the requirements for health may be classed under the heads, 
first, of those that are common to the community, such as 
the supply of good water, the removal of foul water, and the 
removal of refuse matter; and secondly, of those which 
immediately concern the individual householder, such as 
the condition of his house and the circumstances of its 

But the existence of some danger to health in a house in 
a town or village may be a source of danger to the houses 
around. Thus it is the interest of every person in a community 
of houses, that every other member of the community should 
live under conditions favourable to health. 

Each year, as civilisation and population increase, so do 
these considerations increase in importance. So long as 
preventible disease exists in this country, we must not delude 
ourselves with the idea that we have done more than touch 
the borders of sanitary improvement. 

Laws alone can do little to remove or prevent sanitary 
defects. A central department of the government may assist 
in spreading through the community the knowledge of what 
is necessary, and may publish practical methods of applying 
that knowledge. To that extent government may usefully 
interfere. But real and permanent sanitary progress can only 
be obtained by the efforts of the people themselves ; by the 
education of the nation in a knowledge of the laws of health ; 
and by the creation of an efficient local administration en- 
dowed with adequate power and responsibility. 

The first step therefore towards further progress is to imbue 
the owners and occupiers of houses and cottages with a know- 
ledge of the laws of health : they are the laws of common 
sense : simple and economical methods of applying them are 
generally those found most effectual. 

The health, intelligence, morality, and general wellbeing of 
a community depends upon the condition of the dwellings. 

288 Conclusion. 

The local authorities in towns and villages may direct un- 
healthy dwellings to be pulled down, but the removal of 
insanitary dwellings should not be accompanied by over- 
crowding in other dwellings. When therefore in such cases 
private enterprise fails to supply new houses, it should be the 
duty of the local authorities to build healthy dwellings out 
of the rates to replace the dwellings so removed. 

A supply of pure drinking water within the reach of all the 
inhabitants should be provided ; as well as such a degree of 
drainage as the local conditions show to be necessary for 
ensuring that all fouled water would be removed rapidly, and 
not be allowed to stagnate in ditches or on the surface, nor to 
pass into streams until it has been clarified by passing over 
land or otherwise. 

All refuse should be rapidly removed from the immediate 
vicinity of dwellings. 

The plans of all new dwellings and important alterations of 
existing dwellings over the whole country should be subject 
to a general Building Act containing provisions based on the 
principles suggested in this volume, to be enforced by the 
local authorities ; and whenever the local medical officer has 
reason to suspect that a cause of disease exists in any house 
in a town or village, there should be a power to enter and 
inspect the premises, and to require the removal, at the 
expense of the owner or occupier, of any cause of disease 
found to exist. Such arrangements would require that all 
congregations of houses in the country districts should be 
subjected to the jurisdiction of a local surveyor, such as is the 
case generally in towns. 

It is the function of the sanitary engineer or local surveyor 
to adopt measures to prevent or to remove those sources of 
danger to health which the medical officer is called upon to 

The community does not permit any man to practise 
medicine without having satisfied a careful and responsible 

Conctusidfu 289 

board of examiners that he has educated himself for his 
position : and education Jn the principles of sanitary science 
is just as necessary to ensure the efficient fulfilment of the 
duties of a sanitary engineer or local surveyor, as is the study 
of medicine to the medical man. Sanitary science is somewhat 
new, it rests on the attentive observation of facts ; its true 
principles have been slowly and painfully collected during 
a long period, from a careful observation of human disease 
and misery, by those who have preferred the study of facts to 
the more enticing but more fallacious creation of theories ; 
hence sanitary science is built up from details ; wherever the 
details have been carefully and intelligently applied success 
has invariably followed their application. When the public 
realise that the progress of the nation in healthiness is to be 
attained by a careful attention to these details, they will insist 
that the local surveyor and sanitary engineer shall have a 
complete education in the science of the healthy construction 
of buildings, and in the arrangements for health to be adopted 
in towns and villages ; that is to say, in the conditions neces- 
sary for the prevention of disease ; just as at the present time 
they require education in those who minister to the cure of 

The various processes which are necessarily incidental to 
life, -especially to life in crowded communities, contribute 
largely to that deterioration of air and water which is a 
principal cause of preventible disease. But the operation of 
a free atmosphere and of running streams, provides a ready 
means of purifying the air and water thus contaminated. 

The evils which have arisen from this deterioration have 
been gigantic, in consequence of the apathy of the community : 
an apathy which results from an ignorance of the cause of 
these evils, and of the means of remedying them. But the 
remedy for these evils would be comparatively easy if each 
member of the community were induced to perform his part in 
their prevention or removal. In order to attain this end, every 


290 ConcluHon. 

member of the community should be taught the principles of 
sanitary knowledge. When this has been done, and when the 
co-operation of every individual in a community has been 
enlisted to aid in enforcing attention to sanitary details, we 
may hope for practical progress in the diminution of preven- 
tible disease, and for a general improvement in the health, and 
therefore in the happiness, as well as in the wealth producing 
powerj of the community. 



Ablution rooms in barracks, p. 164. 

— in hospitals, 165. 

ventilation of, 166. 

Absorption of gases by water, 198, aaa. 

— of neat by different bodies, loa. 
Aeration of soil, 15. 

Air, circulation of round buildings, 31. 
by means of open fireplace, 122, 

— composition of, 37, 95. 

— condition of indoors, 43-55. 

— density of, 63. 

— dilation and compression of, 62, 98. 

— expansion of, 97. 

— Alters, 82. 

— flues, importance of cleansing. 1 29. 

— London, impurities in, 43. 

— movement of, 44, 62, 64. 

— organic matter in, 39. 

— permeability of materials to, 17, 1 93. 

— propulsion of, 7a. 

— sources whence taken, 55, 80. 

— space between ceiling and roof, 193. 

— space under basement floors, 191* 

— suspended matters in, 38. 

— warmed ; capacity for moisture, 118. 

— weight of, dry, 63, 95. 

— — saturated with moisture, 63, 96, 
Anemometer, 67, 68. 

Areas to basements, 18, 157. 

Arsenic in wall paper and paint, 189. 

Artesian well, 214. 

Artificial lighting, 171. 

Asylums, building arrangements, 160. 


Barracks, ablution rooms in, 164. 

— arrangement of, 56, 160, 163. 

— fireplaces in, 124, 133, 163. 

— height of rooms in, 163. 

— unit of construction for, i6a. 

— windows in, i6a. 

Barracks and hospitals, impurity of 

air in, 45. 
Basement, air space under floor, 191, 

— purity of air in, 190, 

— temperature, 190. 

— areas, 18, 157. 

— occupation of, undesirable, 18, 157, 


Baths, waste pipes from, 251, 


Beams, ventilating, 86. 

Buildings, arrangement oft 50, 1 56, 160. 

— fire- proof, 156. 

-^ numoer of stories in, 33, 161. 


Camps, densities of population in, 33. 

— removal of refuse from, 328. 
Candles, combustion of, ^4, 171. 
Carbonic acid (CO,) in air, 37, 
Carbonic oxide, formation of, in ordi- 
nary grate, loi. 

-^ in air heated by iron stoves, 108. 
Cavalry stables, cubic space per horse, 

Ceilings, impervious to air, 192. 

— lath and plaster, 193. 

— pervious, 193. 

— passage of air through, 54, 193. 
Cellars, objections to, as dwellings, 18, 

— ground air in, 157. 

Cells in prisons, ventilation and warming 

of, 152. 
Cess-pits, 253. 
— - for barracks, 254. 

— near wells, danger of, 213. 

— pollution of ground air and of water, 

2i3» 253- 

— ventilation of, a 54. 

Chalk water, Dr. Clarke's process, 203. 
Channels for admission of air, 81, 117. 

— shape of, 82. 

— material, 82. 

— underground. 82. 
Charcoal filters for sewers, 264. 
Chimneys, movement of air in, 65,,! 21. 

— resistance to passage of air, 66. 

— smoky, 79. 

— temperature in, 101. 
Circulation of air in room with open 

fireplace, 122. 

round buildings, 31. 

under basement floor, 191. 

— of water in pipes for heating, 112. 
Cisterns, cleansing of, 222. 

— for storing water, aoo. 

— waste pipes from, a44, 251. 
Clarke's soap test for hardness of water, 


— process for softening chalk water, 203. 



Climate, classification of, 4. 
Coal, sulphurous acid from, 4?. 

— consumption of, in open fireplace, 1 20. 
Colour of water, ao6. 

Combustion of fuel (coal, charcoal, coke, 
wood), 100, 121. 

— of oil, candles, gas, 43, 171. 
Conduction, heat by, 104. 
Conductivity of heat of different mate- 
rials, 181. 

Convalescent patients, 169. 

Copings, 183. 

Cottage grates, with open fire and oven, 

Cowls, 76. 

Cubic space, 48. 

— in barrack rooms, 58. 

— in hospitals, 59. 

— in schools, 61. 

— in workhouses, 58, 60. 


Decomposing matter, danger to health, 

Density of air, 63. 

Densities of population for camps, 33. 
Derby Infirmary, ventilation of, 146. 
Dilatation of air, 62. 
Dilation of air, cold produced by, 98. 
Distribution of water in camps, 223. 
Down draughts in ventilating shafts, 90. 
Drainage, combined with irrigation, 26, 


— of stables, 175. 
Drains in a dwelling, 239. 

— arrangement and position of, 241. 

— constructional requirements in laying, 


— examination of, 254. 

— flushing of, 246. 

— gratings to inlets to, 249. 

— joints, 245, 267. 

— materials, 245. 

— shape, 245. 

— ventilation of, 242, 264. 
Drains, subsoil or land, 249, 269. 
Draughts in a room, prevention of, 122. 
Dry earth closets, 234. 

Dry earth, midden, and pail systems for 

removing refuse, 230. 
Dryness of air, 81, 118. 
Dwellings, arrangement on a site, 30. 

— distance apart of, 159. 

— height of, 159. 


Earth closets, 234. 
Eaves, for protection to walb, 183. 
Electric light in rooms, 174. 
Emanations, production and removal of, 

Emission of heat from bodies by radia- 
tion, 102. 
Evaporation, 10. 
Exhaust steam for heating, 113. 
Expansion of gases and air, 62, 97. 

— of water, 112, iq8. 
Extraction of air, ^5, 142. 
Extraction shafts, form of, 84. 

— position and arrangement of, 83. 


Field^s self-acting flush-tank, 246, 270. 
Filters for air, 82. 
Filters, 219. 

— animal charcoal for, 220. 

— cleansing of, 221. 

— sand for, 219. 
-- sponges for, 220. 

Fireplaces (open), action on ventilation, 

— in barracks, 163. 

— circulation of air in a room with open, 

— consumption of coal in open, 1 20. 
— General Morin's experiments with, 135. 

— Herbert Hospital, 130. 

— materials for, 121. 

— two in a room, position of, 1 23. 
Fire-proofing, T56. 

Fittings for water-closets, 251. 
Fletcher's anemometer, 68. 
Floors, 190. 

— Herbert Hospital, 192. 

— hospital wards, 191, 192. 
' — joints in, 191. 

— warmed, 145. 

— wooden, 191. 
Floor space, 48. 

— in barracks, 58, 163. 

— hospitals, 59. 

— prison cells, 61. 

— schools, 61. 

— workhouses, 58, 60. 
Florence, water from Amo, 218. 
Flow of water in hot-water pipes, no. 

— in sewers (see Sewers), 
Flue-pipes, heat given out by, 106. 

— horizontal and vertical, 107. 

— sloped, 107. 

— vertical, 107. 

Flues, brick and iron, 106. 
Flushing drains and sewers, 246, 269. 
Foul water, removal of, 226, 256. 
French Legislative Chambers, warming 

and ventilation of, 150, 
Friction of air in a chimney or shaft, 

Fuel, combustion of, 100, lai. 

— constituents of, 99. 

— evaporative powers of, 99. 



Fuel, heating powerof different kinds of, 


Gas burners, ventilated, 173. 

Gas, products of combustion of, 43, 171. 

Gases absorbed by water, 198, 22a. 

— expansion of, 97, 
Gibraltar, sewers at, 260. 

Glass for windows, quality of, 1 95. 
Glazed roofs, 194. 

Goux system for removal of refuse, 233. 
Grates, materials for, 121. 

— ventilating (see Ventilating fireplaces). 
Ground air, 15. 

Gullies to receive discharge from waste 
pipes, 248. 

— position of, 248. 

— trapped gratings to, 248. 

Hall or staircase in centre of house, 158. 
Healthiness of buildings, 35. 

— of site, 4. 

effect of adjacent land on, 24. 

Heat, absorption of, by different mate* 
rials, 102. 

— of different substances, 96. 

— conduction and transmission, 104. 

— given out by cast and wrought iron 
pipes, 116. 

— loss of, by walls, 184. 
by windows, 196. 

— produced by compression of air, 98. 

— radiation of, 102. 
Heating, by an open fire, 104. 

— by a close stove, 104, 136, 139. 

— by hot-water pipes, Perkins', 112.. 

Easton and Andereon's, 112. 

approximate rules for, 117. 

— by steam and exhaust steam, 113. 
at Lockport, U.S., 114. 

— by warmed air, experiments by Prof. 
E. Wood, 136. 

Heating power of different kinds of fuel, 

Heating water under pressure, 109. 
Height of rooms, 158. 
Herbage, effect on moisture, 13. 

— effect on temperature, 8. 
Herbert Hospital, floors of, 192. 

— warming and ventilation of, 130, 153. 
Horse troughs, 224. 

Hospitals and barracks, impurity of air 

in, 45. 
Hospitals, ablution rooms and watex^ 

closets in, 165. 

— building arrangements, 160. 

— cubic space allowed in, 59-61. 

— floors in, 191, 192. 

— nurses* rooms, 164. 

— staircases in, 168, 170. 

— unit of construction, 164. 

Hospitals, ward of!ices, i64» 165. 

— wards, construction of, 164. 
Hospital wards, windows in, 164. 
Houses for residential purposes, 1 70* 
Houses of Parliament, warming and ven- 
tilation of, 141, 147. 

Humidity of air, 46, 53. 
Hygiene, definition of, i. 


Intercepting sewers, 262. 

Iron pipes, action of water on, 200. 

Irrigation, 25, 279. 


Joints in floors, 191. 


Latrines in temporary camps, 228. 
Lead, action of water on, 200. 
Lightme;, 171. 
Lime-whiting roofs, 194. 


Manure, value of, 235, 275, 282, 284. 
Materialsarid details of construction, 180. 

— for fireplaces, 121. 

— for floors, 190. 

— for roofs, 193. 

— for walls, 1 8a. 

— for wall coverings, 187. 

— permeability to air, 17, 54. 
Metal roofs, 194. 

Midden, pail, and dry earth systems for 

removal of refuse, 250. 
Model lodging-houses, 33. 
Moisture, effect of herbage and trees on, 


— in air, 46, 53, 1x8. 

— in soils, 14. 

Mortality in towns and country districts, 

Moule's earth closet, 234. 
Movement of air, 44, 55, 62, 64, 74. 

Nitrates in water, 207. 
Norton's tubes, 215. 


Oil, combustion of, 44, 171. 
Organic matter in air, 39. 

— in water, 205. 

Overcrowding, effect on health, 32. 
Ozone in the atmosphere, 39. 


Pail system for removal of refuse, 231. 
Parian cement for walls, 188. 
Paving of stables, 175. 
Pavilions in hospitals, 164. 



Pavilions in hospitals, distance between, 

— height of corridors between, 167. 

— number of beds in, 168. 

— number of floors of wards in, 167. 

— position of, 167. 
Percolation, 10-15. 

Permeability of materials to air, 17, 54. 
Pipes for conveyance of water, 300. 

— for heating, 103. 
Plaster for walls, 188. 

Pneumatic system of sewage removal* 

Poison in wall-paper or paint, 189. 
Pollution of river Seine at Paris, 209. 

— of surface round Indian village, 229. 

— of water in its transit from source, 214. 

— ^- underground, 213. 
Population, densities of, for camps, 33. 

— of different districts of metropolis, 32. 

— in towns and country parishes, 30. 

— increase of, 286. 

Porosity of different materials, 181. 
Porous walls, 182. 
Pressure of air and water, 72. 
Prison cells, floor space, 61. 
Privies, 230. 


Radiation of heat, 103. 
Rainfall, 9, 211. 

— evaporation of, 10. 

— percolation from, 10. 

— removal of, 357. 
Rain-water, 199, 218. 

— collection of, 218. 

— impurities in, 199, 257. 
Rain-water pipes, 200, 247. 
Refuse from dwellings, 225, 226. 

— from camps, 228. 

— Goux system, 233. 

— pail system, 331. 

— privies, 230. 

— receptacles for, 227, 329. 

— solid 226. 

Reservoirs for storing water, 204, 218. 
Ridge ventilation for stables, 177. 
Rivers, contamination of, 209. 

— underground, 211. 
River-water, 199. 
Roads, 272. 

Roofs,, construction of, 180. 

— glass, 194. 

— impervious, 192. 

— materials for, 193. 

— metal, 194. 

— slated, 193. 

— to stables, 177. 

Sand for filters, 219. 

Sandstone, permeability to water, 212. 
Sanitary science, education in, 288. 
Sash windows, 196. 
Schools, 160. 

— unit of construction for, 163. 
Sewage, A. B. C. process, 276. 
^- cluiracter of, 375. 

— disposal of, 373. 

— farms, 360, 379, 284. 

— intermittent filtration, 283. 

— irrigation, 279. 

— lime process, 276. 

— phosphate of alumina process, 277. 

— pneumatic system, 236. 

— precipitation, 375. 

— question, present aspect of, 284. 

— removal of, 225. 

— systems for clarif)dng, 274, 278. 

— utilisation of, 274, 278. 

— value of, as manure, 235, 275, 283, 284. 
Sewer-gas in drains, 340. 

— in water, 341, 344. 
Sewerage works (see Seweri), 
Sewers, brick, 367. 

— Brussels, 366. 

—catch tanks for heavy substances in, 369. 

— charcoal filters for ventilation of, 364. 

— flushing of, 346, 369. 

— Gibraltar, 360. 

— gradients for, 366. 

— intercepting system of, 363. 

— junctions at, 367. 

— mud in, 272. 

— outlets, 358. 

— Paris, 371. 

— protection of ends of, 364. 

— provision for rainfall, 359. 

— shape of, 265. 

— size of, 360, 367, 

— temperature, 343. 

— tributary, 267. 

— velocity of flow in, 358, 
London, 359. 

— ventilation of» 362, 264. 

Sinks and water-closets, position of, 244. 
Sites, healthiness of, 4, 27. 

— position of, 24, 37. 
Skylights, 158. 
Slated roofs, 193. 
Soil, aeration of, 15, 
-— carbonic acid in, 16. 

— conductivity of, 7. 

— moisture in, 14. 

— nature of, as affecting health, 30. 

— variation in temperature of, 6. 

— water-level in, 18. 
Soil-pipes, 350. 

— connection with drain, 351. 

— material for, 350. 

— position o^ 350. 




Soil-pipes, size of, 350. 

— ventilation of, 242, 251. 
Solid refuse, removal of, 226. 
Solvent power of water, 108. 
Specific neat of bodies, 90. 
Sponges for filters, 220. 
Springs, 211. 

— temperature of water in, 7. 
Stables, cubic space per horse, 179. 

— drainage of, 175. 

— paving of, 175. 

— roofs of, 177. 

— ventilation of, 1 74* 

— windows of, 1 77* 
Staircases in hospitals, 168, 170. 

— ventilation of, 80. 

Steam boiler, temperature of air in chim- 
ney of, lOI. 

— pipes for heating (see Heating). 
Storage of water, 204, 218. 
Stove-pipes, amount of heat given out 

by, 106. 
Stoves, brick and iron, 106. 

— cast and wrought iron, 108. 
lining of, 109, 

— close, 104, 136, 139, 140. 

— supply of air to, lOO, 

Strata, pervious or impervious, 211. 
Subsoil contamination of water, 213. 

— drains, 249, 269. 

Sulphurous acid from coal, 42. ^ 


Tanks for reception of rain-water, 218. 

— ventilation of, 223. 
Temperature, effect of changes of, on 

movement of air in a room, 55. 

— effect of herbage and trees on, 8. 

-r- obtained from water by heating under 
pressure, 109. 

— of air, 4, 46, 81. 

■— — as it leaves a coal fire, loi. 

effect of propulsion on, 71. 

effect on death-rate, 5. 

in chimney of steam boiler, loi. 

in flue pipe, 106. 

in sewers, 243. 

variation with altitude, 4. 

latitude, 5. 

local conditions, 5. 

— of iron or glass roof, 194. 

— of rooms, 95. 

coal required to raise the, 120. 

with ventilating fireplaces, 133. 

•» of soil, variation due to depth, 6. 

— water in springs, 7. 

— of water stored under and above 
ground, 221. 

Tents, arrangement of, 34. 
Trap, lead, dip of, 253. 

Trees, effect on moisture, 13. 

— — temperature, 8. 
Troughs for watering horses, 224. 


Uniform diffusion of poison in air of 

room, 50. 
Unit of construction for asylumst 161. 

— barracks, 162. 

— hospitals, 164. 

— schools, 162. 

— workhouses, 161. 


Vegetable organic matter in water, 207. 
Vegetation, effect on health, 19. 
Velocity of air in a chimney, 65, loi, 121. 
measurement of, 67. 

— flow of water in sewers, 258. 
Ventilating beams, 86. 
Ventilating fireplaces, 1 24. 

— advantages of, 135. 

— application to existing chinmeys, 127. 

— area of fresh-air channels for, 128. 

— at Herbert Hospital, 130. 
•— constmiption of coal in, 130. 

— experiments on temperature, 133. 

— heating surface, 130. 

— size of, according to ventilation re- 
quired, 129. 

— supply of external air to air chambers, 

— temperature of rooms, 133. 
Ventilation, action of open fireplaces on, 


— and warming of churches, 143. 
corridors between wards in hos- 
pitals, 169. 

— ' — densely occupied buildings, 140. 

— — Derby Infirmary (Sylvester), 146. 
French Legislative Chambers 

(Morin), 150. 

halls for meetings, 144. 

Herbert Hospital, 130, 153. 

Houses of Parliament, 141, 147. 

in rooms without open fireplaces, 


— — new opera at Paris, 147. 

— — opera at Vienna, 146. 
ordinary rooms, 140. 

— — prisons, asylums, &c , 140, 152. 
theatres, 144. 

— by compressed air, 73. 

— by extraction, 65, 142. 

— — in rooms, 143. 
position of flues, 142. 

— by propulsion, 71. 

in House of Conmions, 141. 

— in hot climates, 141. 

— of ablution rooms and water-closets 
in hospitals, 166. 



Ventilation of rooms, 49, 77, 85, 
for dinners and receptions, 170. 

— of sewers, 262, 264. 

stables, 174. 

staircases, 80. 

workshops, 1 74. 

— Roman plan, 145. 

— through shafts, 65-76, 86, 90. 

— warmed air, 117. 

— without warmed air, 85. 
Ventilators, Amott's, 89. 

— the Hopper, 85. 

— Mackinnel's, 93. 

— Moore's louvred panes, 86. 

— Muir's, 93. 

— Sherringham*s, 87. 

— Vertical tubes, 87, 88. 

— Watson's, 92. 


Wall and window surface, proportionate 

loss of heat, 197. 
Wall coverings, 187. 

— paper and paint, poison in, 189. 
Walls, passage of air through, 54. 

— and roofs of houses, construction of, 1 80. 

— damp in, 182. 

— enamelling, 187. 

— loss of heat by, 184, 

r— materials for, conductivity of beat 
and porosity, 181. 

— new, quantity of water in, 182. 

— Parian cement for, 188, 
— '- plaster for, 188. 

— prevention of damp in, 183. 

— thickness of, 186. 
Wards for special cases, 169. 
Warmed air, capacity for moisture, 118. 

— supply to a room, 124. 
Warming £^ir by hot water or steani 

pipes, 116. 

— by in-flow of warmed air, 136. 

— observations on, 95. 
Waste pipes, 243, 250. 

Water, action on iron and lead, 200. 
carried sewage (see Sewage), 

— chlorides in, 207. 

— cisterns for storing, 200. 
closets, 239. 

fittings for, 251. 

pans for, 252. 

position of, 244. 

traps for, 252. 

ventilation of, 244. 

— colour as test of, 206. 

— compression of, 112. 

— contamination of, 209, 213. 

Water, distribution of, in camps, 223. 
— Dr. Clarke's process for softening, 203. 

— expansion of, 198. 

— filtering of, 219. 

— flow of, in hot-water pipes, 110. 

— gases absorbed by, 198, 222. 

— hardness, Clarke's test of, 202. 

— hard and soft, commercial value of, 

— ^hardness of, temporary and permanent, 

— high temperature obtained from, by 
heating under pressure, 109. 

— -level in soil, 18. 

of different districts, 211. 

— organic matter in, 205. 

— pipes for conveyance of, 200. 
impure gases in, 198, 222. 

— pollution of, in its transit from source, 

— pressure of, 72. 

— purity of» 198. 

— quality of, 205. 

— quantity to be supplied, 214. 

— rain- and river-, 199, 218. 

— solvent power of, 198. 

— storage of, 404, 218, 221. 

— subsoil contamination of, 213. 

— supply, camps, 216. 

Croydon, 21 2( 

Florence, 218. 

— — general principles of, 204. 
pipes for, 206, 223. 

— temperature of, in springs, 7. 
-^ -traps to waste pipes, 240, 252, 
'. — underground, 211. 

— weight of, 198. 
Weight of the air, 63, 95. 
Wells, in London, 214. 

— permanent,. 216. 
lining of, 216. 

preparation of surface round^ 2 1 7^ 

— in camps, 215, 216. 
Well-water, purity of, 215. 
Windows, barracks, 16 a. 

— best form for ventilation, 196. 

— double, 196. ' 

— French, 196. 

glass, quality of, 195. 

— hospitals, 164. 

— loss of heat through, 196. 

— rooms, store-closets, &c., 158, 194. 

— stables, 177. 
Window surface, 195. 
Workhouses, 160. 

— cubic space allowed in, 58, 60. 
Workshops, ventilation of, 1 74,