132 



NA TURE 



{yttne 9, 1887 



would make the instrument inconveniently large, and a 

 rapidly rotating spiral tube would probably break, from 

 centrifugal force acting on those parts of the tube that 

 were not on the axis of rotation. Hence, in the latest 

 form of secohmmeter, Fig. 3, we have been led to employ 

 a stationary spiral glass tube, G, with its end cemented 

 into a stationary hollow steel conical plug fitting mer- 

 cury-tight in the collar of the rotating metal box, B, with 

 its weighted elastic sides. This arrangement simplifies 

 the tap mechanism, as the tap now is not rotating, also 

 many small improvements have been introduced into this 

 last form : for example, at all the joints there is mercury 

 under pressure, so that there is no tendency for air to be 

 drawn into the apparatus at the joints, a fault which some- 

 times occurred with the earlier form of the apparatus, and 

 led to irregularities in the readings from a bubble of air 

 in the box acting on an air spring, or from air in the glass 

 tube altering the length of the thread of mercury. The 

 temperature adjustment in this last form of the secohm- 

 meter is made by screwing a screw in or out, which 

 slightly alters the volume of the stationary portion of the 

 mercury vessel. The fly-wheel F has been made to have 



Fig. 3. — Imprcved Secohmmeter. 



a much larger moment of inertia, and the box B is placed 

 inside it, so as to be screened from damage. 



Mutual Induction. 



If we wish not merely to determine the coefficient of 

 self-induction, L, of a coil, S, but also the coefficient of 

 mutual induction, M, between it and any other coil, we 

 first exclude the other coil from the battery circuit and 

 determine L in the manner already described. We next 

 include the other coil in the battery circuit, and repeat 

 the experiment with the secohmmeter ; then, as shown by 

 one of our students, Mr. Sumpner (to whom our thanks 

 are due for the most able assistance that he has rendered 

 in this investigation) — • 



p-Vr\ N 



or if Nj and o-i are the speeds and apparent increase of 

 resistance in a first experiment, and Nj and o-g in a second, 

 we have 



-0. 



L 



M 



_ P 



6ol(^-Il \ 

 P + r VN, N, /' 



where p is the resistance of the arm of the bridge 



opposite the coil S, and r the resistance of the arm 

 joining /J and S. 



Capacity. 



We have also shown that, if, instead of placing a coil 

 with self-induction in one of the arms of the bridge, 

 the arm be shunted with a condenser, there will be an 

 apparent diminution of the resistance of that arm, since 

 such a shunted condenser acts as if it had a negative self- 

 induction. This apparent diminution divided by the 

 product of the square of the actual resistance of the arm 

 into the reading of the scale of the instrument corre- 

 sponding with the speed at which balance is obtained 

 gives the capacity absolutely in farads. The formula is, 

 therefore, far simpler than that given by Clerk Maxwell 

 for the absolute measurement of the capacity of a con- 

 denser, by placing it on one of the arms of the bridge 

 and rapidly reversing the connexions of the condenser 

 with the bridge. 



Secohmmeter without Speed Indicator. 



Lastly, all known zero galvanometric methods of com- 

 paring the coefficients of self or mutual induction with 

 one another, or with the capacity of a condenser, can be 

 increased enormously in sensibility by the use of the 

 secohmmeter, and, since in such cases the speed of rotation 

 need not be known, a very simple form of secohmmeter 

 without speed indicator can be employed. The com- 

 parison of the coefficients of self or mutual induction with 

 one another, or with the capacity of a condenser, is 

 usually effected by tests that are completed during the 

 growth or the dying away of a current, since it is only 

 during the variation of a current that self or mutual in- 

 duction, or the electro-static capacity of a condenser, 

 evidence themselves. The effect of an error in the 

 balance only lasts for a very short time, and therefore is 

 very small if the error be small, that is, the tests are not 

 sensitive. But by the use of the secohmmeter it is now 

 possible not merely to measure the coefficients of self and 

 mutual induction and the capacities of condensers abso- 

 lutely, but, in addition, to secure the same high degree of 

 sensibility with comparison tests that have hitherto had 

 to be completed during the growth or dying away of a 

 current that it is customary to obtain in the use of the 

 Wheatstone bridge for measuring resistances with steady 

 currents. " W. E. Ayrton. 



John Perry. 



THE FOSSIL FISHES OF MOUNT LEBANON. 



THE last published part of the Transactions of the 

 Royal Dublin Society (May 1887) contains a memoir 

 on the fossil fishes of the chalk of Mount Lebanon, in 

 Syria, by James W. Davis, which is an important con- 

 tribution to a very interesting subject. 



The existence of fossil fishes in the chalk of Mount 

 Lebanon has been known from remote antiquity ; Hero- 

 dotus refers to them, and various statements about them 

 are recorded in writings scattered over the period between 

 the fourteenth and eighteenth centuries. In our own 

 century, Louis Agassiz, Pictet, Haeckel, Costa, Botta, 

 Fraas, and others, have added greatly to our knowledge 

 of the various species met with, and now this memoir of 

 Mr. Davis, illustrated as it is by twenty-four excellent 

 plates, several of which are folding plates, brings up our 

 knowledge of these remains to the most recent date. 



For the chief material on which this memoir is based 

 the author is indebted to the zeal and energy of the Rev. 

 Prof. Lewis, who, during his residence in the American 

 College at Beyrout, availed himself of every opportunity 

 of collecting specimens of these fossils, and succeeded in 

 accumulating a very large series of new forms. Many of 

 these have been acquired for the British Museum Natural 



