198 MR. HENRY M. SAYERS 



separated, to be rather vague. This photograph suggests that com- 

 plete separation has not taken place in the laminated form, and 

 occurs only in the spheroidised form; e.g., Fig. 2. 



The appearance of sub-laminations in Fig. 1 is not a false effect 

 due to excessive cutting down of the iris diaphragm, as no dia- 

 phragm was used in this instance. A slightly false effect, due to 

 this cause, is evident in Fig. 5, which was taken expressly for this 

 contribution. 



Fig. 3 shows a subject in which the laminations are too fine to 

 be clearly photographed at a niagnification of 1,500 diameters. The 

 photograph demonstrates that even air cooling a rod about \ in. 

 diameter is not sufficiently rapid to arrest the partial production 

 of ** pearlite.*' It will be noticed that the constituent which we 

 have called ** sorbitic pearlite " is partly cellular. 



The difference between these structures is not apparent at lower 

 powers. 



Fig. 4 is to be compared with Fig. 5. In spite of the slightly 

 false effect in the latter, caused by the iris diaphragm, it draws 

 attention to the existence of a feature which might easily escape 

 notice, but which is apparent on closer examination, in Fig. 4, 

 namely, the sub-laminations. 



Mr. Henry M. Sayers : These photomicrographs of steel at 5,000 

 and 8,000 diameters are very fine, and testify to the skill and patience 

 of the authors. They confirm the accepted theory that no new details 

 can be revealed by magnifications incommensurate with the N.A. of 

 the objective. With an N.A. of 1.4, 1,000 diameters shows all that 

 can be seen, but, of course, greater amplification may be useful for 

 diagrams to be displayed to large numbers of people, at once, just 

 as lantern slides are magnified by projection. 



The authors state that the illuminant used was a 20 ampere alter- 

 nating current arc, the arc being focussed on the stop or aperture of 

 the vertical illuminator. Presumably one or other of the carbons 

 was so focussed. This adds to the merit of the work, for certainly 

 an A.C. arc is less satisfactory in intensity and form of the radiant 

 than a C.C. arc crater. 



It will probably be found that one of the larger tungsten arc 

 lajnps is better than an A.C. arc. It is true that the tungsten arc 

 requires continuous current, but this can be got from an A.C. supply 

 with a simple form of auto-transformer and rectifier. A nominal 

 100 c.p. Pointolite, taking about \\ amperes, gives satisfactory 

 negatives of steel with 5 minutes* exposure, at 1,200 diameters, 

 using a light filter denominated " 5 times," and Wellington "Anti- 

 Screen *' plates. 



The Pointolite lamp is somewhat more convenient for the neces- 

 sary source emplifying lens system than an arc. The exposures above 

 mentioned were taken with a combination which magnified the source 

 about three diameters, giving a field of 3 in. diameter, i.e., com- 

 fortably filling a quarter plate. With no amplifier the field on 

 the plate was only about 1 in. diameter. Greater amplification can 

 be obtained if required by varying the lens distance of the combina- 

 tion. 



