





























Aprit 28, 1923] 
and measuring apparatus—is not very helpful; in 
fact, it is difficult to understand why this title was 
chosen, except for its attractive alliteration, for it 
certainly would not lead us to expect articles on 
earthquakes, oceanography, radiation, and many other 
_ subjects which are not meteorology, or metrology, or 
_ Measuring apparatus. 
_ As a matter of fact, most of the subjects treated 
appear to fall into some such classification as the 
following : 
_Measurement—Theory and practice of measure- 
ents of length, mass, time, and their derivatives ; 
alcoholometry ; saccharometry; drawing _ instru- 
‘ments; calculating instruments; combination of 
observations ; and allied subjects. 
_ Geophysics—Form and mass of the earth ; meteoro- 
logy, including atmospheric electricity ; oceanography ; 
seismology and tides. 
and magnetic measurements, and thermometry are not 
treated in this volume. 
One of the chief values of the articles lies in the fact 
that they are not compilations from text-books and 
scientific journals, but each is written by a man whose 
life is engaged on the work he describes. This is 
early seen from the following list of the institutions 
which have provided writers from their staffs, past or 
present. There are 43 main articles in the volume, 
of which 12 are provided by the National Physical 
Laboratory, 7 by the Meteorological Office, 2 each by 
the Royal Geographical Society and the Survey of 
India, 1 each by the Bureau of Standards, U.S.A., and 
the Ordnance Survey. The writers of the 18 remaining 
articles include such well-known names as Profs. Boys, 
Knott, Sampson, Skinner, and Turner; Sir Horace 
Darwin, and Mr. C. T. R. Wilson. 
Before dealing with the articles themselves, it may 
be worth while to remark on a few points connected 
with the general arrangement of the book which have 
struck us very forcibly while reading the 800 or so 
pages of which it consists. The arrangement is 
obviously a compromise, and a compromise can never 
give entire satisfaction. It would appear that the 
first idea of the work was that of a dictionary with 
probably the alphabetic arrangement throughout, as 
in the “Encyclopedia Britannica.” But that idea 
has been modified, and a series of volumes each dealing 
with more or less allied subjects has been adopted. 
We cannot be too grateful for this decision ; for in 
these hard times a single volume may be within the 
means of many who could not afford the whole set. 
We cannot help regretting, however, that the whole 
dictionary idea was not abandoned at the same time. 
Whatever the intention may have been, the volume 
NO. 2791, VOL. 111] 
NATURE 
On the other hand, terrestrial magnetism, electrical. 
wae 
before us is practically a collection of 43 articles and 
an index to them ; but instead of the latter being placed 
at the end, it is embodied by the dictionary method 
throughout the whole book, and the articles are strung 
on to it like large beads on a necklace. This method 
has two great drawbacks: in the first place, the long 
breaks in the sequence of the words, due to the inter- 
polation of the long articles, make it difficult to turn 
up a word quickly. Then technical difficulties of 
printing have made it impossible to give references to 
pages, and the reader is referred to sections and para- 
graphs of the main articles, the title of the article 
being set out in full in each case. This entails a great 
deal of unnecessary printing, and it is not easy to find 
a specified paragraph, as the sections in some cases 
extend over several pages. A simple index with refer- 
ences to pages would have served the same purpose, 
and would not only have been easier to use but also 
probably have reduced the size of the volume by many 
pages, with great convenience to the reader and a 
reduction in the cost of printing. 
While we are discussing the convenience of the 
reader, it may be as well to direct attention to the 
want of system with regard to references to literature. 
Some of the articles have very full references while 
others have practically none, but the method of making 
the references varies from article to article. In some 
the references are given in the text, in others in foot- 
notes, while in a few they are collected together in a 
bibliography at the end of the article, the numbered 
items of which are indicated in the text by the use of 
numbers in brackets. In fact, this book reflects the 
chaos in general scientific literature in this matter. 
Nothing is more disturbing when reading a difficult 
article than to have the attention constantly dis- 
tracted by frequent references to footnotes, some of 
which may be of importance to the argument, and 
therefore must be read, while others are only refer- 
ences to literature. For this reason it is surely desir- 
able that there should be some distinction between the 
two kinds of references. The method which seems the 
most reasonable is to use figures in brackets in the 
text to connect with references to literature collected 
together at the end of the article—or chapter in the 
case of a text-book—while notes necessary to the 
argument should be given, if they cannot be avoided, 
at the foot of the page, and attention directed to them 
by an asterisk or other conventional sign used to 
indicate a footnote. In this way a reader would 
almost unconsciously pass over the literature refer- 
ences and yet never miss a footnote. The advantages 
of this method are so obvious when many references 
are given that it is surprising it is so little used. 
There is still one more point of arrangement which 
