540 



PROFESSOR KNOTT ON THE STRAINS PRODUCED IN 



fields No. III. lies lowest and No. V. highest; but in the highest fields No. V. lies 

 highest and No. IV. lowest. The curves, in fact, all cross one another in the intermediate 

 fields. 



The curves on Plate V. are plotted to a vertical scale only one-tenth of the scale used 

 for the steel curves. In other words, the changes of volume in the nickel tubes are on 

 the whole much greater than the corresponding changes in the iron and steel tubes. 

 This is in accord with the well-known fact, that the changes of length of nickel rods 

 when magnetised longitudinally are much greater than the changes of length of iron rods. 



Again, the generally opposite characters of the strains in iron and nickel are worthy 

 of note. In low fields there is decrease of volume in the one and increase in the other ; 

 in high fields this is simply reversed. In this respect also the results are in accord with 

 well-known results in allied fields of research. 



The dotted curve shown on Plate V. is the curve of No. VII. plotted in terms of in- 

 ductions instead of fields. As with the corresponding curve on Plate I., the early 

 portion is drawn out along the horizontal. The dotted curve meets No. VII. at Field 

 500 in the latter. This corresponds to induction 3500 across the middle section of the 

 tube. The average induction throughout the whole tube is of course considerably 

 smaller. 



§ 11. The Other Permanent Tubes. — The following tables contain the reduced 

 observations for the other tubes, which are distinguished by the Arabic numerals : — 



Volume Changes in 10" 5 Cub. Cm. 



Field. 



Iron. 



Steel. 



Nickel. 



No. 3. 



No. 5. 



No. 7. 



No. 3. 



No. 5. 



No. 7. 



No. 1. 



No. 4. 



No. 7. 



15 













- -8 





_ 



+ -4 



25 



- 



- -9 



- -5 



+ 



+ 



- 2-5 





+ '2 



•4 



50 



- 1 



- 3-0 



- 6-1 



+ -8 



+ -7 



- 4-4 



- -1 



- 2-6 



- 21 



75 



- -3 



- 8-9 



-14-4 



+ 1-6 



+ 1-4 



- 7-5 



- 2-3 



-15-2 



- 91 



100 



- 5-3 



- 18-8 



-15-9 



+ 2-2 



+ 2-6 



- 6-7 



- 8-2 



-35-8 



-123 



150 



-14-1 



-21-3 



-10-2 



+ 1-0 



+ 8-7 



+ 4-8 



- 24-5 



-63-9 



-192 



200 



-15-7 



-18-0 



- 8-9 



+ 2-0 



+ 16-0 



+ 11-8 



-38-3 



-79-4 



-239 



300 



-18-0 



-18-8 



- 7-4 



+ 3-6 



+ 27-4 



+ 17-2 



-50-9 



-92-4 



-273 



400 



-22-3 



-20-7 



- 6-9 



+ 3-3 



+ 33-7 



4-21-9 



-60-5 



-96-7 



-280 



500 



-25-0 



-22-6 



- 7-5 



+ 2-4 



+ 36-8 



+ 23'8 



-63-4 



-97-5 



-287 



In all these cases the cap was of the same material as the tube. The results for 

 the same tubes with the brass cap differ in detail from these in a manner very similar to 

 what was found with the other tubes. 



Consider, first, the group of iron tubes, and compare them with the corresponding tubes 

 in the previous table (§ 8). The differences are very striking. With Nos. 3, 5, and 7, 

 there is always diminution of volume ; whereas in the previous cases there was increase 

 in certain fields, decrease in others. Nos. III. and 3 present least divergence in type, 

 but the change of volume is numerically much greater in the latter. Nos. 5 and 7, 



