May 27, 1922] 



NA TURE 



675 



followed will so limit the position or nature of the 

 fault that its detection becomes easy. Prof. Miles 

 Walker has had exceptional experience with machinery 

 during the last thirty years and so lays the greatest 

 stress on those faults which occur most often ; in some 

 cases the faults are due to abstruse causes which make 

 great demands on the expert's knowledge of physical 

 .science. 



The subjects are well divided, chapters being devoted 

 to break-down of insulation, over-heating, low efficiency, 

 sparking at the brushes, etc. The method of deter- 

 mining the efficiency of an electric generator by air 

 calorimetry — a method first devised and put into 

 practice by Sir Richard Threlfall — is commended and 

 the various methods of measuring the velocity of the 

 stream of cooling air are described. The author attri- 

 butes to Kennelly the discovery that the heat con- 

 vection from a thin wire increases as the square root 

 of the air velocity. It is true that Kennelly discovered 

 this law experimentally, but the complete law had 

 been deduced mathematically from physical principles 

 several years previously by the French mathematician 

 Boussinesq. He also proved theoretically Newton's 

 law of cooling ; namely, that the heat convected is 

 proportional to the difference of temperature between 

 the wire and the cooling fluid. This law, generally 

 assumed as obvious by engineers, has been verified 

 in the most satisfactory way by physicists. 



In chapter 5 the practical application of vector 

 diagrams is described, but the author does not dis- 

 tinguish clearly between the various kinds of vectors 

 which are in everyday use. He begins by considering 

 the vectors of two alternating functions which do 

 not follow the harmonic law. The cosine of the angle 

 between the two vectors is defined as the ratio of the 

 mean value of their product to the product of the 

 effective or root-mean-square value of each. It is 

 therefore a highly complicated function. It can be 

 shown that in accordance with this definition, when 

 we Jiave three alternating functions, their vectors can 

 be represented by three lines drawn from a point in 

 space. Hence, contrary to what the author says, a 

 knowledge of the angles between the first and third 

 and the second and third vectors does not enable us 

 to give the angle between the first and second. The 

 author then proceeds to describe rotating vectors, but 

 he does not state explicitly that he is now making 

 sine curve assumptions. The diagram representing 

 as vectors the fluctuating part of the electrical power 

 taken from the various mains requires more explanation. 



The comments made on balanced loads are of 



interest, but are not very practical. It seems to the 



writer that a polyphase load is only balanced when 



the magnitudes of the v^olt-amperes taken from each 



NO. 2743, VOL. 109] 



main and the phase differences between the volts and 

 amperes of each main are the same. 



In a few cases where the author gives formulae, as, 

 for instance, the formulae for eddy current losses, it 

 would be a help to many if the proof were indicated. 

 The limitations also of the formulae should have been 

 stated. A few new words are introduced. The 

 phrase " wattful load " is used to indicate the power 

 expended on the load. It is complementary to " watt- 

 less load," a phrase used almost universally by engineers 

 to denote the magnetising power required by the load. 

 Although many diagrams are given showing ripples 

 in waves, harmonic analysis is barely mentioned. The 

 causes of these ripples, however, are described, and 

 n\any ingenious remedies are suggested. 



The Inner Impulse. 



La Forme et le Moiivement : essai de dynamique de la 

 vie. By Georges Bohn. (Bibliotheque de Culture 

 generate.) Pp. xii-i-175. (Paris: Ernst Flam- 

 marion, 192 1.) 4 fr. 50 net. 



The title leads one to expect a discussion of the effect 

 of movement on the form of animals, perhaps new 

 evidence for Lamarck. Nothing could be further from 

 the author's intention. He seeks to show that the laws 

 regulating the reproduction and growth of living 

 creatures are the same as those which govern their 

 movements, and that these are the laws of chemical 

 physics. The conception of an organism as a whirlpool 

 is at least as old as Cuvier, but remained little more 

 than a useful analogy till F. Houssay superimposed on 

 it the idea of vibration. But for him both the vortex 

 and its vibrations expressed the effect on the creature 

 of its environment. Prof. Bohn starts with molecules 

 of living substance, the inherent vibrations of which 

 produce a system not merely vortical but polarised, 

 manifesting its internal forces through oscillations in 

 space and in time. That sentence, so far as possible in 

 Prof. Bohn's own words, will scarcely be intelligible to 

 one who has not read the book. Nor, we are warned, 

 will perusal of the book profit an inquirer unacquainted 

 with the fundamental ideas of physics and mechanics. 

 It may therefore be due to some gap in our knowledge 

 that we rise from a second reading provoked but 

 puzzled, interested but unconvinced. 



It needs no very close acquaintance with modem 

 biological research to realise the importance of chemical 

 constitution. We have learned that every species has 

 a chemical character of its own ; we are familiar with 

 the part that catalysers, hormones, and other more 

 hypothetical substances, play or are supposed to play 

 in all the functions of the body, in growth, and in 

 the hereditary transmission of form. The study of 



