TRANSACTIONS OF SECTION A. 645 



only the circular shape of the coin ; as the pressure increased towards the normal 

 the pictures became more perfect. 



III. The output of the transformer was regulated by resistances put into the 

 primary circuit, also by the rate at which the reversing commutator was driven ; 

 the reversing commutator is similar to those used in the early Siemens machine, 

 and driven from a belt in the laboratory. 



IV. The temperature varied between 5° 0. and 100° 0. Perhaps the effects 

 may have been produced more quickly at the higher temperature, but no very 

 significant change was apparent. 



V. Similar experiments are about to be performed in many other gases, and 

 under considerable pressure, up to 10 atmospheres. 



VI. In addition to pictures on photo-plates, good impressions have been 

 obtained on bromide paper and other papers direct. Pictures can easily be got 

 from woodcuts after they have been well covered with plumbago and other 

 conducting substances. 



VII. Impressions can be produced upon photo-plates which have been exposed 

 to full sunlight previous to an inducto-script experiment. The development of the 

 electrical picture appears to be independent of the condition of the plate due to 

 light. Thus an ordinary photograph and an inductograph may be produced on the 

 same plate. 



VIII. In some cases the plate, on which a good image was formed, was very 

 carefully insulated, so that it was as far as possible only subject to a quickly 

 changing potential difference. 



IX. In one experiment a metal pointer was attached to the coil of a siphon 

 recorder, and nearly touched a ribbon of bromide paper ; on development the 

 trace was evident. The experiment is now being repeated, under somewhat altered 

 conditions, to determine the speed at which such signals can be made perfectly. 



7. On a Periodic Effect which the Size of Bubbles has on the Velocity of their 

 Ascent in Vertical Tubes containing Liquid. By Fred. T. Trouton, 

 M.A., D.Sc. 



An account was given of some determinations made of the velocity with which 

 bubbles of air of difierent sizes ascend vertical tubes containing water. The chief 

 peculiarity brought out by these experiments is that the velocity of ascent is a 

 periodic function of the size of the bubbles. The form of the curve, obtained by 

 plotting the size, i.e., the volume of the different bubbles used as abscissae, and the 

 corresponding values of the velocity of ascent as ordinates, was exhibited. This 

 showed that at first, as would be expected, increase in size diminished the velocity, 

 but that on reaching a certain size the velocity begins to increase in value, then 

 comes to a maximum (in value about twice that of the minimum velocity), after 

 which it again diminishes, and so on two or more times, depending on the diameter 

 of the tube employed. The oscillations in the curve die out in much the same 

 fashion as that for a pendulum swinging in a viscous medium. The form taken by 

 the bubbles was also described, being almost spherical at the first minimum ; 

 beyond this ' the bubble is pointed at top, and is so until reaching the second 

 minimum, when it is again rounded at top, and has a waist-like appearance, 

 having thus a somewhat dumbbell shape, and so on, presenting in this way 

 similarities to the breaking up of a liquid column through surface tension. 



Experiments were described where liquids which do not mix with water were 

 employed instead of air — such affording more simple conditions. Also some 

 experiments similar to those with water, but made with other liquids. 



8. Experiments on Flame Spectra. 

 By Professor Arthur Smithells, B.Sc. 



The author exhibited his apparatus for separating the two cones of combustion 

 which constitute a Bunsen flame, and showed how it afl'orded a means of ascer- 



