October 5, 191 1] 



NATURE 



447 



involved are, however, fundamental to all electrical 

 •engineering, and thus the usefulness of the book is 

 t>y no means limited to the class of readers for which, 

 to judge by the title, it has been written. The 

 phenomena of electric wave propagation play an im- 

 portant part in long power-lines, and much of what 

 the author has to tell us about the waves in telephone 

 •cables may, with some obvious modifications, be 

 directly translated into the domain of heavy electrical 

 work. Problems connected with propagation are 

 most easily treated by the use of the symbolic method, 

 and although Heaviside, Steimetz, Pupin, Kennelly, 

 and others have for some years used this method in 

 their publications, the majority of electrical engineer- 

 ing text-books still ignore it. Electrical engineers 

 will therefore feel grateful to the author for having 

 given them in his book a very clear and readable 

 exposition of the treatment of electric problems by 

 complex quantities. This is done in the first chapter. 

 Then follows a chapter on wave propagation generally. 

 The subject is introduced by the investigation of 

 sound waves in air. In this way the main principles 

 of such an investigation are established by reference 

 to a problem with which all engineers are more or 

 less familiar, and this is a material help to the more 

 complicated problems of magnetic and electric waves, 

 which are treated next. 



In the third chapter we come to the general case 

 of an infinitely long cable having at one end impressed 

 on it an alternating e.m.f. It is this case which is 

 of interest not only to the telephone engineer, but also 

 to the designer of a power-line. By making use of 

 the symbolic methods outlined in the first chapter, the 

 author shows how the current gets weaker as we 

 proceed from the home end, and how at the same 

 time the phase angle between e.m.f. and current 

 increases. He distinguishes thus between an "at- 

 tenuation factor," which applies to the real part, and 

 a "phase factor," which applies to the imaginary part 

 of the complex quantity. In telephony the attenua- 

 tion factor is not of paramount importance, since the 

 ear is able to appreciate even very weak sounds, pro- 

 vided their general character as determined by the 

 sequence of the waves of different frequency remains 

 the same. But this is just the condition which in 

 an ordinary telephone cable only exists if its length 

 is moderate. The phase factor is different for each 

 wave-length, and thus the longer the line the greater 

 is the distortion in the arriving waves. A moderate 

 amount of distortion the human ear is able to analyse, 

 in the same way as we are able to recognise a per- 

 son's face from a caricature if the distortion of the 

 true features is moderate, but on very long lines the 

 attenuation and phase factors have so altered the 

 character of the waves that the ear is no longer able 

 to analvse them, and telephony becomes impossible. 



This refers to an ordinary cable in which there is 

 little inductance, but much capacity. The author 

 shows that the old rule according to which the product 

 of capacity and resistance was considered, the im- 

 portant item on which clearness of speech depended, 

 is wrong, and that, as was first pointed out by Heavi- 

 side, the condition for perfect transmission is equality 

 NO. 2l88. VOL. 87] 



between two products, namely, that of capacity and 

 resistance per mile and that of inductance and leakage 

 per mile. In a cable of this kind the velocity of pro- 

 pagation is the same for all frequencies, and conse- 

 quently all parts of a composite wave travel at the 

 same speed and arrive without distortion, although, 

 of course, attenuated. Heaviside called a cable of 

 this kind a "distortionless cable." 



By reference to Pupin 's theory it is next shown 

 how an approach to the perfection of such a cable 

 may be obtained by "loading," that is, putting in- 

 ductances at intervals along the line in series with 

 thi- conductor. This approach to perfection will 

 obvii iusly be the closer the more it approaches the 

 condition of a uniformly loaded line. This means 

 that the impedance coils must not be too far apart. 

 Eight to nine coils per wave-length is the number 

 theoretically found in an example given of a 90 ohms 

 per mile line, where coils of 02 henrv every two 

 miles satisfy this condition. In the Anglo-French 

 telephone cable laid by the British Post Office last 

 year the impedance coils have each an inductance of 

 o'i henry, and are spaced 1*53 miles apart. The con- 

 struction of this cable is fullv described. There is also 

 a chapter devoted to submarine telegraphy, and 

 another to the study of the propagation of waves of 

 very high frequency along wires. Here, again, the 

 power engineer will find much useful information. 



GlSBERT KAPP. 



UARI.XE REFRIGERATION. 

 Cold Storage, Heating, and Ventilating on Board 

 Ship. By Sydney F. Walker, R.N. Pp. vi + 269. 

 iLondon: Constable and Co., Ltd., 1911.) Price 

 8s. net. 



THE increasing luxury of modern sea-travel makes 

 the use of cold storage almost a necessity for 

 any but the smallest passenger boats. In the large 

 liners the most elaborate arrangements are used to 

 preserve the food and to keep a continual flow of air 

 at the proper degrees of temperature and dr; 

 through the passenger regions. In this book the 

 author has brought together the general principles 

 which must be used to get such a result, and in the 

 section on cold storage considers the principal methods 

 used to convey the enormous quantity of food now 

 brought to this country in cold storage. 



An improvement would be an index, especially to 

 designers and freezer engineers, for whom the book 

 is primarily intended, as it is not easy to find if a 

 reference is made to any particular fitting or arrange- 

 ment. It is well known, for instance, that there are 

 two distinct methods of treatment suitable for different 

 kinds of food. In one, so soon as the natural heat 

 has gone out, the food is frozen to some degrees below 

 freezing-point, and maintained frozen — but at no par- 

 ticular temperature — until ultimately thawed for use; 

 in the other, the food is never frozen, but is maintained 

 at as constant a temperature as possible a few degrees 

 above freezing-point, and in an atmosphere with a 

 definite degree of humidity. There is no difficulty 

 about the first method, which is used, for instance, 



