Aprils, 1889] 



NATURE 



549 



To prevent the direct action a small uninsulated metal screen 

 can be placed between the vibrator and the tube, or the tube 

 can be hung by a long and fine wire, in order to be removed 

 from the sphere of the direct action of the vibrator. The best 

 plan is, however, to surround the tube by wire gauze, which 

 stops the direct action of the vibrator on the tube, and yet per- 

 mits the tube to be observed. The absence of direct action can 

 be ascertained in the different positions which the tube takes by 

 insulating it from the resonator without changing its position, 

 and noticing if it becomes quite dark. 



If we place before the vibrator a resonator, consisting of a 

 straight wire 220 centimetres long (I used copper wire No. 6), 

 we find, by the tubes, that nodes exist in the middle and the two 

 ends of the wire, consequently two segments at 55 centimetres 

 from each end of the wire. 



(The above-described circular resonator can be likened to a 

 linear one which is curved to a circle and its two ends soldered 

 tc^ether, thus the two nodes of the ends becoming one single 

 node. ) 



If this straight wire is cut in the middle, a torrent of sparks passes 

 between the separated ends, even if they are removed several 

 millimetres apart. If then we examine each half of the wire, 

 we find that it possesses a node in the middle and two segments, 

 each at one end, but the node is not so well defined as in the case 

 of the uncut wire ; there is no single point the potential of which 

 remains continually equal to zero, but a line in which the 

 difference of potential from zero is a minimum. This com- 

 plicated form of oscillation is produced by the fact, that the forces 

 acting in this resonator are not equal at all points or symme- 

 trically distributed with respect to it, as in the case of the long 

 resonator. The oscillations of the short resonator may be com- 

 pared to those of a rod which is not firmly fixed by its middle. 

 The state of these oscillations is not stable. If one or both ends 

 of this resonator are touched by the fin ger, they become nodes, 

 and a well-defined segment appears at the middle of the resona- 

 tor. If the Geissler tube be connected with this middle point, 

 it begins to light up when the ends of the resonator are touched, 

 and ceases to light the moment the fingers are removed ; the 

 contrary takes place if the tube be connected with one end of 

 the resonator. This phenomenon is analogous to the change of 

 the form of the vibration of a rod when fixed by its middle or 

 by its ends. 



Quite similar is the mode of oscillation of a resonator 220 

 centimetres long disposed on one side of the vibrator ; it 

 possesses also a node at the middle not well defined, and a 

 segment at each end. 



To conclude, I will describe the mode of oscillation of a 

 resonator, no centimetres long, disposed parallel and symme- 

 trically to the vibrator. This resonator possesses one node in 

 the middle, very clearly defined, and a segment at either end. 

 This form of oscillation is the same as would occur in the long 

 resonator if one-fourth of its length from each end were cut off. 



In the case of the latter resonator and of the long one, which 

 is also symmetrically disposed to the vibrator, the oscillations 

 are very stable, and much stronger than in the case of the 

 resonators placed on one side of the vibrator. The mode of the 

 oscillations of these symmetrical resonators ir> not disturbed by 

 touching them by the fingers at any point, although the 

 mode of oscillation is disturbed if we touch the unsymmetrical 

 resonators at any point whatever. 



The experiments described must be performed in a dark room, 

 and much care be used in the choice of the proper Geissler 

 tubes. Tubes containing mercury are very sensitive, and they 

 become more so if the mercury be allowed to flow several times 

 from one end of the tube to the other. 



I-iverpool, March. E. J. Dragoumis. 



SOCIETIES AND ACADEMIES. 

 London. 



Royal Society, March 21. — "An Experimental Investigation 

 of the Circumstances under which a Change of the Velocity in the 

 Propagation of the Ignition of an Explosive Gaseous Mixture 

 takes place in Closed and Open Vessels. Part I. Chronographic 

 Measurements." By Frederick J. Smith, M.A., Millard Lec- 

 turer on Mechanics, Trinity College, Oxford, Communicated 

 by A. G. Vernon Harcourt, F. R. S. 



It has been noticed by several investigators, viz. MM. 

 Herthelot and Vielle, MM, Mallard and Le Chatelier, and 

 Prof. H. B. Dixon, F.R.S., that explosive gaseous mixtures 



after ignition do not reach their maximum velocity of propaga- 

 tion at once, but that a certain maximum velocity is attained 

 soon after initial ignition. 



In order to investigate this period, which may be called the 

 acceleration period of an explosion, chronographic measurements 

 of a peculiar nature were found necessary. 



It was at once evident that but little advance in this branch 

 of the subject of explosions could be made unless exceedingly 

 minute periods of time could be measured with certainty. 



A new form of chronograph has been devised to meet as far 

 as possible all the requirements of the case, by means of the 

 instrument. The following results have been obtained : — 



(i) The irijJffff of a second can be measured with ease, and 

 periods of time differing from iV of a second to ij^finr of a 

 second can be recorded on the same moving surface. 



(2) The surface which receives the record moves at a velocity 

 which is practically constant during the traverse of 5 cm. ; also 

 its velocity can be varied between wide limits. 



(3) A large number of time records can be made side by side, 

 all records being made in straight lines. 



(4) Fractions of recorded vibrations of a fork can be sub- 

 divided by means of a micrometer microscope. This is not the 

 case with vibrations recorded on a surface attached to a pen- 

 dulum, where the velocity varies from zero up to a maximum at 

 the middle of the swing. 



The electro-magnetic styli, by means of which events 'are 

 marked, are so constructed that their period of "latency" is 

 almost absolutely constant, and their electro-magnets are so 

 wound that no sparking takes place on breaking the circuit. 



A moving surface is carried on a carriage, which is propelled 

 by means of a falling weight, which after a certain velocity has 

 been attained is removed : the surface then moves with a 

 velocity which is found to be practically constant for the limits 

 between which a time record is made. 



The chronograph is used in conjunction with a steel tube in 

 which the explosions take place. At even distances along the 

 axis of the tube, conducting bridges, eight to ten in number, of 

 Dutch metal insulated from the tube, are placed ; each bridge 

 is connected el ectrically with a recording stylus, so that as each 

 bridge is broken by the explosion, a mark is made on the 

 surface of the chronograph ; these markings when duly inter- 

 preted provide data for constructing a curve, which indicates the 

 rate at which the velocity .of the explosion is changing during its 

 propagation. 



The rest of the paper treats of the methods by means of 

 which the errors due to the use of electro-magnets in chrono- 

 graphic work have been dealt with and reduced as far as possible. 



Chemical Society, March 7.— Mr. W. Crookes, F.R.S., in 

 the chair. — The following papers were read : — The decomposition 

 of carbon disulphide by shock ; a lecture experiment, by Prof. 

 T, E. Thorpe, F. R.S. The author, in studying the action of 

 the fluid alloy of potassium and sodium on carbon disulphide, 

 obtained a yellowish-brown solid substance which exploded with 

 great violence when subjected to pressure or friction. If the 

 explosion occurred in contact with carbon disulphide, that sub- 

 stance was resolved into its elements. A similar decomposition 

 of carbon disulphide into carbon and sulphur can readily be 

 effected by exploding a charge of 0*05 gramme of fulminate 

 within a stout glass tube containing carbon disulphide vapour, 

 and the experiment affords a good illustration for class purposes 

 of the resolution of an endothermic compound into its elements 

 by sudden shock. — The determination of the constitution of the 

 heteronucleal a/3- and ;3)3-di-derivatives of naphthalene, by Prof. 

 H. E. Armstrong and Mr. W. P. Wynne. A preliminary note 

 on the constitution of the three chloramidonaphthalenesulphonic 

 acids obtained by sulphonating o-chloro-/8-amidonaphtha!ene 

 hydrochloride with weakly fuming sulphuric acid. — The action 

 of chloroform and alcoholic potash on phenylhydrazine, by Dr. 

 S. Ruhemann. 



March 28. — Annual General Meeting. — The [following is an 

 abstract of the Annual Report, read by the President, Mr. 

 W. Crookes, F.R.S. : — The pleasant duty again devolves on me 

 to present to you the annual report on the state of the Chemical 

 Society during the twelve months just past. The following 

 statement shows the numerical position of the Society — 



Number of Fellows {March 28, 1888) 

 Present number of Fellows 



Increase .^ 



••• 1534 

 ... 1614 



80 



