Page:Proceedings of the Royal Society of London Vol 69.djvu/503

Rh selecting that specimen of iron which has about the same carbon or other impurity as that contained in the particular alloy and subtracting the resistance of the iron. The result is shown in the next table.

Table I. Approximate Increase in Electric Resistivity (in microhms per c.c. at 18 C.) of Annealed Iron Alloys produced by adding to Iron Different Percentages of the Elements named in the first column.

Percentage of added element. Allovs of iron, *.

with 2 per cent. 3 per cent. 5 per cent.

Tungsten 4*0 5-0 6*0 microhms

Nickel 7-0 l>-0 13-0

Chromium 10 -0 11- 5 14-0

Manganese 16-0 18-0 24-5

Silicon 2G-0 ;U'0 49-0

Aluminium 2S'0 36'0 54'0

Dividing these values by the percentage at the head of each column, it will be seen that as the alloy becomes richer in the added element the increase of electric resistivity becomes less and less for every <>nr per cent, of the element added. The comparative effect of carbon is difficult to ascertain owing to various causes, (") the impossibility of making homogeneous alloys of iron with large percentages of carbon, the highest carbon in our specimens being 1 } per cent. ; (It) the different conditions in which the carbon may exist in the alloy ; and (<} the difficulty of excluding impurities such as manganese and silicon, minute amounts of which exert a serious effect on the conductivity. However, by comparing the conductivity of alloys of iron having nearly similar quantities of Si and Mn, but containing different quantities of carbon, a close approximation of the effect of this element on the conductivity of iron is arrived at. The result shows that the increase of one-tenth of one per cent, of carbon in iron containing 0*03 carbon causes an increase in the specific resistance of nearly 2 microhms per c.c., but when the carbon is raised from 0'13 to 1*13 the increase in specific resistance is at the rate of 5 microhms for 1 per cent, of added carl on, about one-fourth of the rate of increase at the smaller per- centages. The position of carbon on Table I therefore appears to lie near to chromium. From specimens of two impure alloys containing cobalt with iron, the increase of resistance produced by cobalt was estimated to be somewhat less than nickel but greater than tungsten ; its position in Table I would therefore appear to lie near to nickel.

If now we compare the increase of resistance in iron produced by alloying it with these elements, with the atomic weights of the added elements, we find, as a rule, that those elements with the highest