218 Mr. G. M. Whipple on Temperature-correction [May 3, 



II. "On the Temperature- correction and Induction-coefficients 

 of Magnets." By G. M. Whipple, B.Sc, Superintendent of 

 Kew Observatory. Communicated by Robert H. Scott, 

 F.R.S. Received April 12, 1877. 



It has been the practice at the Kew Observatory since 1856 to deter- 

 mine, for the use of persons about to make observations upon terrestrial 

 magnetism, the various constants of the magnets and instruments with 

 which they intend to observe ; and a considerable number of these con- 

 stants are now recorded in the books of the Observatory. 



Having been frequently applied to for information respecting the con- 

 stants of magnets and the method of determining them, I have extracted 

 from our registers the values of the two most important constants, viz. 

 that of the variation of the power of the magnet under changes of tem- 

 perature, and that showing the effect of the inductive action of the earth 

 upon the magnets, as found by experiment. The results are given in the 

 following paper. 



The magnets experimented upon may be classed under six heads (ap- 

 proximate dimensions) : — 



a. Solid cylinders, 3*6 inches long, 0*3 inch diameter. 



b. Hollow cylinders (collimators), 3*65 inches long, 0-28 inch internal 



diameter. 



c. Parallelopipeds (bars), 5*4 inches long, 0*8 inch wide, 0*12 inch 



thick. 



cl. Rhomboidal plates (dip-needles), 5*7 inches long, 0*05 inch thick. 

 e. Thin cylinders (enclosed in brass tubes, deflectors), 3*75 inches long, 



0*30 inch diameter. 

 /. Various. 



All of the magnets were made of the best steel, and rendered as hard 

 as possible before magnetization, which operation has been in every case 

 carried to saturation. 



The method employed in determining the temperature-coefficient is as 

 follows : — The magnet being firmly fixed in a water-tight wooden box, 

 provided with a thermometer, is placed upon a frame in such a manner 

 that its axis shall lie in the same horizontal plane with the needle of a 

 unifilar magnetometer, at a short distance away, and adjusted until the 

 axes of the two needles are approximately at right angles to each other, 

 the position of the deflecting magnet being that which is designated 

 by the Astronomer Royal as end on. Warm water at 85° F. is then 

 poured into the box, and as soon as the magnet has become heated to 

 this temperature the unifilar magnet is brought to rest and its position 

 accurately read off. The water at 85° F. is then removed, and the appa- 

 ratus being cooled duwn to 60° F., water of that temperature is placed 

 in the box j another observation of deflection is then made, and finally 



