Notices respecting New Boohs. 437 



1872. Since that time (in 1884) a Committee of the British 

 Association, of which the author was Secretary, has collected all 

 the results of various observers relating to the spectra of the 

 elements, and has printed them in the Reports of the Association. 

 In the present volume all these results are reprinted together with 

 others published since the Committee finished its work. It does 

 not, however, contain the most recent measurements made with 

 the large diffraction-gratings at the Johns Hopkins University by 

 Prof. Rowland and his colleagues. Indeed the numbers given in 

 the tables were all obtained from prismatic spectra, and corrected, 

 of course, for the irregularities in the dispersion of the prism. In 

 the preface Dr. "Watts gives a table of corrections for reducing 

 Angstrom's and Cornu's measurements to the standard of Rowland's 

 map. This is presumably the map issued in 1888, which, if we 

 remember rightly, Prof. Rowland stated to be in some respects 

 erroneous ; at any rate not so accurate as the more recent one 

 (1889). The author has introduced a new feature into this Index 

 in the shape of a column of oscillation frequencies, these being the 

 reciprocals of the wave-lengths. This column ought to be very 

 useful in ascertaining any harmonic relations which may exist be- 

 tween the various lines in the spectrum. The Index contains a 

 very large amount of information concentrated into a small com- 

 pass. By its means it should be possible to identify with com- 

 parative ease any unknown element which happens to be under 

 examination. We are glad to learn that Dr. Watts contemplates 

 issuing an Atlas of Spectra as a companion to the book before us. 

 The Atlas is to be constructed so that the intervals between the 

 lines will be proportional to their differences of oscillation fre- 

 quency. This will differ considerably from Rowland's map, where 

 the distance between two lines is proportional to their difference 

 in wave-length. Dr. Watts's map should show at a glance what 

 relations, if any, exist between the various lines given by any 

 element. J. L. H. 



The Phonopore and the Simplex Phonopore Telegraph. By C. 

 Laxgdon-Davies. (London: Kegan Paul, Trench, Triibner, 

 &Co.) 



If two insulated wires be wound side by side on a bobbin so that 

 a current passed through one acts inductively on the other, it is 

 well known that, if one end of the secondary coil be insulated and 

 the other joined up to a telephone and a galvanometer whose 

 second terminal is likewise insulated, whenever a current is made 

 or broken in the primary coil the telephone of the secondary coil 

 gives a click, but the galvanometer shows no effect. By periodically 

 repeating the making and breaking of the primary current, we can 

 produce a musical note in the telephone, but the galvanometer is 

 still undisturbed. The author has made use of this fact in the 

 phonopore, which consists of two coils, as described above, and 

 which he proposes to apply to a new system of duplex telegraphy. 



