TRANSACTIONS OF SECTION A. 577 



hypothesis was correct. The comparison was made with very satisfactory results, 

 -the form of the curves being exactly the same in both cases. 



From the hypothesis and the curves connecting P.D. and length of arc for 

 constant currents it was deduced that — 



1. JVTien a coi-ed positive carbo7i is used and a constant current is Jlowing the 

 area of the crater must increase as the length of the arc increases. 



2. The change that takes place in the ratio of the soft carbon to the total amount 

 of carbon in the surface of the crater ivith a given change of length must diminish 

 ■as the arc increases in length. 



3. The change that takes place in the ratio of the soft carbon to the total amount 

 of carbon in the surface of the crater with a given change of length must be smaller, 

 and the rate of change must becmne more nearly constant for all lengths of arc as 

 the value of the constant current increases. 



Remembering that the ratio of the area of soft carbon to the area of the crater 

 Is called the ' soft crater ratio,' these three conditions may be put thus : — 



1. With a cored positive carbon, and with a constant current flowing, the area 

 of the crater must increase, and consequently the soft crater ratio must diminish as 

 the length of the arc increases. 



2. The change of soft crater ratio with change of length must diminish as the 

 length of the arc increases. 



3. The change of soft crater ratio with change of length must be the smaller, 

 •and the rate of change must become the more nearly constant the larger the 

 current. 



To test the accuracy of these conclusions, and therefore of the hypothesis upon 

 -which they were founded, measurements of the crater, made on the enlarged image 

 of the arc in 1893, were used. It was found that straight line laws were obtained 

 in two ways : (1) by plotting the P.D. for each length of arc, with the correspond- 

 ing area of the crater with various constant currents ; (2) by plotting the current 

 with the corresponding area of the crater with various constant lengths of arc. 

 From these two sets of straight lines corrected areas of crater were obtained, from 

 which the laws connecting the area of the crater and the soft crater ratio with the 

 length of the arc could be seen more clearly than with the uncorrected areas of 

 crater. 



These laws were exactly what had been predicted. It was found that with a 

 cored positive and solid negative carbon, and with a constant current flowing — 



1. The area of the crater did increase, and consequently the soft crater ratio 

 diminished as the length of the arc increased ; and 



2. The change of soft crater ratio with a given change of length did diminish 

 as the length of the arc increased ; and 



3. The change of soft crater ratio with a given change of length was smaller, 

 and the rate of change was more nearly constant the larger the current. 



From the parallel straight lines connecting the area of the crater with the 

 current for constant lengths of arc three facts were deduced, viz — 



1. That with constant lengths of arc the area of the crater, minus a constant 

 depending on the length of the arc, is proportional to the current. 



2. That the change of area of crater with a given change of length of arc is 

 'Independent of the value of the current flowing. 



3. That the change of area of crater vrith. a given change of current is inde- 

 pendent of the length of the arc. 



10. On the Source of Luminosity in the Electric Arc. 

 By H. Crew and O. H. Basquin. 



Three possible causes of luminosity were considered, viz. heat alone, chemical 

 oction, the electric current. The problem set was to determine the parts which 

 thermal, chemical, and electric cause, respectively, plays in the electric arc. 



1897. pp 



