534 TRANSACTIONS OF SECTION A. 



tube. While the glass is hot a small piece of heated steel tube surrounding the 

 wire is pushed a few millimetres into the glass. After cooling the tube is 

 soldered to the wire. The vacuum-tight seal is produced between the inner 

 surface of the elastic steel tube, which on cooling is put under tension, and the 

 glass which comes under compression. Seals with wires of 1 mm. diameter have 

 been produced in this way. 



2. The Relation between Density and Refractive Index. 

 By Dr. T. H. Havelock. 



3. A Complete Apparatus for the Measurement of Sound. 

 By Professor Arthur Gordon Webster. 



This research involves three parts, all necessary for the measurement of 

 sound intensity in absolute measure. First, a constant source of harmonic 

 vibrations, termed the -phone ; second, an instrument capable of measuring 

 the intensity of the sound thus generated, and called a phonometer; third, the 

 study of the propagation of the sound from the former and the latter, involving 

 the determination of the reflecting power of the flat surface of the earth. The 

 phone consists of a steel diaphragm rigidly driven by an electrically maintained 

 tuning-fork, and made the back of a resonator, of the form of a small hollow 

 chamber or of a tube of variable length. The reacting of the sound upon the 

 amplitude of the fork enables the constants of the resonator to be accurately 

 determined, so that the emission may be measured in watts. The phonometer 

 is a glass diaphragm, made the base of a resonator of either type, and bearing 

 a light mirror which constitutes one mirror of a Michelson interferometer. The 

 displacement is measured stroboscopically by a moving telescope, and the 

 amplitude of the pressure change is read off on a scale in dynes /cm 2 . On account 

 of the magnification due to resonance, the instrument is as sensitive as the ear 

 for pitch 256 vibrations per second. The law of propagation is confirmed and 

 the reflection of the earth determined by measuring at different distances from 

 the phone the sound given by interference with its image in the ground. All 

 points in the theory involved have been experimentally verified, and it seems 

 possible to have an over-all accuracy of within 10 per cent. 



4. On the Relation of Spectra to the Periodic Series of the Elements. 

 By Professor W. M. Hicks, D.Sc, F.R.S. 



The author described some results recently obtained by him in a critical 

 study of the spectral series of the second and third groups of the periodic table 

 of elements, more especially their dependence on the atomic volume of the element. 

 Formulae constants obtained from the wave numbers of three lines of the Sharp 

 series are definite functions of atomic weight and atomic volume. The exact 

 form of the function of atomic weight had not been yet determined, but when 

 it is known the measurement of the wave-lengths of a spectrum should give one 

 of the most accurate methods of determining the atomic weight of an element. 

 The function of the atomic volume was so far determined as to give deductions 

 of atomic volume, or of density, very close to observational values in the case of 

 the first three groups of the periodic series. Applying the method to the spec- 

 trum of Europium as given by Exner and Haschek, a density of 13.1 was predicted 

 for that element. 



5. The Series Spectrum of the Mercury Arc. By S. R. Milner, D.Sc. 



With the arc in air there is always a continuous background to the spectrum 

 present. This sets a limit to the faintness of the lines which can be observed, 

 since attempts to photograph faint lines by increasing the time of exposure will 

 simply result in the fogging of the plate by the background. The spectrum of 

 the arc in vacuo is characterised by an almost entire absence of a continuous 



