148 
NATURE 
[JUNE 12, 1902 
most serious consequences as regards the construction 
and the use of these tables. 
In a work of such magnitude and diversity, it is futile 
to think that errors can have been completely avoided, | 
but from frequent consultation of the earliér volumes we 
can testify to their remarkable freedom from printers’ 
errors. We think a short account of each of the principal 
natural orders should have been given, and the space so 
allotted might have been saved, in part at least, by thus 
obviating the necessity of some amount of repetition in 
dealing with the several genera. 
Prof. Bailey has availed himself of the resources | 
| must know that, when testing a faulty cable under water, 
of Cornell University, of the “ Dictionary of Garden- 
ing” by Nicholson and of the numerous standard 
publications issued from Kew, and, amongst other sources 
of information, has consulted and compared some 
hundred or more catalogues of nurserymen. This latter 
procedure needs to be followed with the utmost caution 
and is one to which, perhaps, the omission of the genus 
Trochodendron is to be attributed. After all, the plants 
that have special interest for commercial purposes are | 
few in number as compared with those which appeal 
primarily to the lover of plants or to the scientific 
botanist. 
We might extend our notice of this book to a much | 
greater length than the editor could allow space for. 
We can only add that the illustrations are very 
numerous, uniform in treatment, often very useful, but, 
on the whole, not equal in value to the text. Further, 
that although expressly compiled to meet American 
conditions, it will, with the necessary modifications, be of 
great value in all English-speaking countries. 
THE MANUFACTURE OF SUBMARINE 
CABLES. 
Les Cables Sous-Marins. Fabrication. Par Alfred Gay. 
Pp. 203. (Paris: Gauthier Villars et Fils, n.d.) 
HE author of this little book, as we are informed on 
the title-page, is an engineer in the employment of 
the Société industrielle des Téléphones, the leading 
French firm for the manufacture of submarine cables. 
The volume is one of a series appearing under the name 
of “ Encyclopédie scientifique des Aide-memoire,” edited 
by M. Léauté, who is also, we understand, connected 
with the Société industrielle des Téléphones. From’ the 
title of the series we gather that this publication is 
designed to serve as a pocket text-book for submarine 
cable engineers, though the style in which it is written 
and the absence of an index—a fatal omission for any 
work of reference—make it resemble a popular treatise on 
the subject of cable manufacture rather than a scientific 
handbook. One example will serve to justify this view. 
In his reference to the Wheatstone Bridge—the most 
usual form of testing the conductor resistance of a cable 
—the author makes no attempt to explain the theory of 
the test, but merely gives the connections and the 
formula for obtaining the result. A book on cable testing 
which evades an explanation of the Pont de Wheatstone 
is as great a curiosity as a treatise on Euclid which 
omits all reference to the Poms Asinorum. 
One or two other points call for comment. With regard 
NO. 1702, VOL. 66] 
‘ é ; 4 5 
to the testing of the dielectric resistance, M. Gay ob- 
| serves that some physicists have expressed the opinion 
that, if sufficient time were allowed, the “ spot ” would 
return to zero and remain there. This could only happen 
in the case of a material which possessed an absolute 
dielectric resistance, and through which, consequently, 
no current could escape. Manufacturers have hitherto 
failed to discover this material. Further on, the author 
asks why the negative current is always the first to be 
applied to the cable, and answers his question by saying 
that he believes that there is no good reason for using 
one current in preference to the other. But M. Gay 
the chemical action of the zinc current tends to clean the 
fault and make it more apparent, while the copper 
current throws a deposit on the exposed surface and 
masks the fault. Thus the reason for using the zinc 
current first is to discover at once any fault that may 
exist. 
Throughout his book the author pays too little attention 
to the question of capacity in connection with the manu- 
facture of cables. On p. 14, in enumerating a long list 
of the conditions which a good dielectric must satisfy, 
he does not mention the desirability of a low capacity. 
In fact, on p. 107 he goes out of his way to lay stress on 
the superior importance of insulation tests to capacity 
tests, ignoring the fact that, cae¢eris paribus, the work to 
be got out of a cable depends on its capacity, its insula- 
tion being purely a secondary matter. Finally, on p. 
145 M. Gay says that the engineer is not master of the 
capacity of a core, the dimensions of which are given 
him, as though the capacity could not be varied by the 
selection and mixture of the gutta-percha used, inde- 
pendently of its relative weight to the conductor. 
For the rest, it may be sufficient to point out that, in 
connection with the table of coefficients, given on p. 85, 
for reducing the D.R. of the cable at the temperature at 
which it is tested to its equivalent at 75°, one must dvide 
and not muz/¢ip/y (as instructed on pp. 147 and 149) by 
the coefficient given, for the D.R. at 75° is, of course, 
less than at a lower temperature and more than ata 
higher temperature. With regard to the brazing of a 
joint, M. Gay would find it difficult to scarf the two ends 
of the conductor, if he omits, as he does in the directions 
on p. 173, to solder them first. 
Enough has been said to show that the book is not 
likely to prove of great value as a work of reference for 
cable engineers. But as a popular treatise on a process 
of manufacture of which the public knows little, and may 
like to know more, it deserves very favourable notice. 
The chapter on the composition and properties of gutta- 
percha is specially good, and on pp. 90 and 91 the author 
sums up very clearly and succinctly the reasons for the 
various conditions which specifications require the di- 
electric to satisfy. 
“Voici, en deux mots, sur quels motifs est basée 
Vintroduction de chacune de ces régles : on impose une 
limite inférieure @isolement pour se garantir contre les 
défauts de fabrication ; on impose une limite supérieure 
disolement pour empécher lemploi des guttas trés 
résineuses qui, en général, s’altérent vite avec le temps ; 
| on impose un résidu maximum dans le chloroforme ou le 
toluéne pour s’assurer que le mélange a été bien nettoyé 
