1916.] The Third Indian Science Congress. cxix 



The south-west monsoon current was found to be on the average about 

 3| miles deep at Madras, 4| to 5 miles at Vizagapatam and Calcutta and 

 about 5 miles over north-west India. Above it there appears to be a 

 return current towards the equator ; so that the air approaches India fron 

 south-west at the surface, rises and deposits its moisture m India, and 

 then return southwards. The winter or north-east monsoon was shown to 

 form part of the circulation of the same character, but in the reverse 

 direction, and its strong resemblance to the north-east trade wind of the 

 Atlantic was pointed out. 



( The effect of Tensile Stress on the Coefficient of Expansion 



of Nickel. — By E. P. Harrison. 



The instrument used in making these measurements was the same 

 as that recently devised by the author for investigating Young's Modulu> 

 at high temperatures and exhibited to the Science Congress at the Madra- 

 meeting last year. 



By deduction from the experiments a series of graphs is obtained 

 showing the relation between the elongation of the wire and its tempera 

 ture, each graph corresponding to a different tension. 



From the whole series of expansion temperature graphs the average 

 coefficient of expansion over different ranges of temperature and corres- 

 ponding to different tensions can be deduced. These coefficients are 

 plotted against the tension. ' 



The following results emerge : — 



(1) At temperatures below about 150°C the coefficient of expansion 

 increases very slowly as a linear function of the tension in accordance 

 with the known equation 



d a 1_ dE_ 



where a is the coefficient of expansion 



T the tension in the wire in kilos per w* 2 , 



E the value at 0° of Young's [Modulus for Nickel and 



dE 



-^t the temperature coefficient of E supplied by the author's previou 



experiments (Lond. Phys. Soc. Proa, Dec. 1914). This may be called the 

 normal change in a with tension, and its value is about 2 x 10 deg 



kilos mm*. 



(2) For temperatures above about 150°C the coefficient of expansion 

 increases linearly (as in case (1) ) until a certain tension is reached, after 

 which a very rapid increase of a takes place. This may be called the 

 abnormal increase in a with tension, and begins to occur when the tension 



exceeds 15 kilos per sq. mm. 



The conclusion drawn from the present research so far as it goes is 

 that during the measurement of expansion coefficients especially at high 

 temperature careful regard must always be paid to the stress conditions 

 even though the latter are far below what is necessary to cause a per- 

 manent set. Only when the stress on the material is below a certain mini- 

 mum value, depending (no doubt) on the material, can the true or normal 

 coefficient be obtained. 



Potential Difference ard current in De la Rive (vacuum) 



tube.— By D. N. Mallik and A. B. Das. 



In such a tube, so long as, and only when, there is a band discharge 

 which alone rotates under the influence of Tran verse Magnetic Fields), 



