106 
paratively few towns are supplied from polluted 
rivers. Another reason may be that in this 
country there is still a popular prejudice against 
the chemical treatment of a drinking water 
which is probably not nearly so marked in 
America. ; 
As is usual with works on water purification, 
the book opens with a chapter dealing with the 
sources of supply and impurities likely to be found 
in water from different sources. The next chap- 
ter, and by far the longest, deals with types of 
purification plants, and describes in some detail 
the works of several large American cities, which 
may be taken as typical of the others. 
Chapters follow on physical and chemical tests, 
bacterial tests, and the interpretation of these 
tests. In these chapters the author goes into very 
minute detail on the methods of carrying out these 
tests. It is worthy of note that no mention is 
made of chemical tests for organic impurities, the 
author confining his attention to tests for such 
substances as affect the dose of chemicals to be 
added—e.g. alkalinity and CO,—and the estima- 
tion of any excess of the chemicals added. This 
is certainly at variance with the usual practice of 
English water analysts. Another point is the 
method suggested for the detection of B. coli in 
water, which is quite different from that usually 
adopted in this country, and scarcely seems to be 
capable of being used with any great degree of 
certainty, depending as it does on the relative pro- 
portions of CO, and hydrogen formed by the fer- 
mentation of dextrose broth. One cannot but 
feel that it is not altogether desirable that persons 
other than qualified chemists and bacteriologists 
should be entrusted with the carrying out of 
these tests on water, especially when, as in 
this case, the descriptions are given by an 
engineer. 
The best part of the book is what follows in 
chapters dealing with coagulation and _ sterilisa- 
tion, water softening, and sedimentation. These 
subjects are clearly and fully treated, and the 
chapters contain a host of valuable information. 
The last chapter consists of notes on filtration 
and general operation. Next follow some very 
ingenious charts for computing the results of ana- 
lyses and the amounts of chemicals required 
according to the analysis of the water. These 
charts, however, are on rather too small a scale to 
be of much practical use. The work concludes 
with. appendices dealing with the analysis of co- 
agulants, standard solutions, specifications for 
coagulanfs, and one or two useful tables. 
The book is decidedly limited in its scope, and 
deals with only a few of the modern methods of 
purification, and, indeed, entirely omits to mention 
several important processes introduced in this 
country during the last decade. For example, no 
mention is made of either Houston’s Excess Lime 
method of sterilisation and softening or the use of 
such substances as Permutit. 
The volume is clearly printed and profusely illus- 
trated with diagrams and photographs. 
Denison B. BYLES. 
NO. 2450, VOL. 98] 
NATURE 
[OcToBER 12, 1916 
THEORY AND EXPERIMENT. 
A Text-Book of Practical Physics. By Dr. H. Ss: 
Allen and H. Moore. Pp. xv+622. (London: 
Macmillan and Co., Ltd., 1916.) Price 8s. 6d- 
net. 
. 
HE distrust which the “practical man” feels 
for science is well known, and it must be con- 
fessed that he is by no means alone to blame for 
his attitude in the matter. The fault lies to some 
extent with the teacher of science, who too often 
lays undue stress on the theory, and, if an experi- 
ment which is supposed to illustrate the theory 
does so but indifferently, the student is apt to get 
the impression that theory and fact are somewhat 
distantly related. For example, a student who, 
haying been rashly told that Atwood’s machine. is. 
used for measuring “g,” finds by careful experi- 
ment that the value obtained differs more or less 
widely from that given in text-books, generally 
concludes that the experiment is ‘ wrong.” He 
rarely has sufficient confidence in his work to know 
that the experiment cannot be wrong, and that 
it is the theory which is at fault. Of course, the 
trouble lies in an insufficient realisation of the 
assumptions made in the theory. If the experi- 
ment does not agree with theory, the student 
should be taught to find the cause of the dis- 
crepancy and to estimate the degree of concord- 
ance which the limitations of the theory and the 
accuracy of his measurements may lead him to 
expect. It is of the utmost importance that such 
discrepancies should not be passed over; and in 
a book like that before us, in which theory and 
experiment are brought together, the valuable 
introductory note on the accuracy of observations: 
might with advantage have been amplified. 
The course covered is a very complete one (up 
to about Pass B.Sc. standard), and we are glad 
to see that a fair share of attention has beem 
devoted to mechanics. The electrical section, 
also, is very full, and has a valuable chapter of 
notes on electrical apparatus. An appendix con- 
tains mathematical tables and an assortment of 
useful data which are constantly being required in 
a physics laboratory. The book is convenient in 
size, well arranged, well illustrated, and well pro- 
duced, and there are remarkably few misprints. 
Each experiment is preceded by a brief theoreti- 
cal account of the principles involved, which 
should enable the student to acquire an intelligent 
appreciation of the experiment to be performed ; 
but in some cases scarcely sufficient attention is 
directed to the possible experimental errors. 
There are occasionally statements which might 
prove misleading. We cannot, for example, agree 
that “the truth of Archimedes’ principle can be 
demonstrated “readily by purely theoretical 
methods.” Such a statement tends to conceal the 
purely experimental foundation of the principle. 
In the optical section “pin methods” receive 
perhaps more attention than they deserve, and in 
some cases they might with advantage be replaced 
by the more accurate and quicker “line methods.” 
i We should also like to have seen a description of 
