394 Notes. 



dense medium, such as realgar, are more difficult to resolve than 

 diatoms of equal fineness mounted in quinidine, and much more 

 difficult than diatoms mounted dry on cover. 



Before a lens fails to resolve a band, the next lower band appears 

 to be irregularly ruled, but these apparent irregularities of ruling 

 disappear when the band is examined by a lens of greater aperture. 

 When, therefore, some lines in a band appear strongly ruled, and 

 others faint or missing, the observer knows that the limit has been 

 reached. He may, by improving his illuminating conditions, or 

 by altering his lens-adjustment, remove all appearance of these 

 irregularities of ruling, but then they are sure to reassert themselves 

 in some higher band. The best screen for microscopical work of 

 this kind is made by dissolving acetate of copper crystals in 

 distilled water, until a saturated solution has been obtained. This 

 solution must be filtered many times, then to it should be added a 

 very small quantity of methyl blue. It appears that acetate of 

 copper will hold only a certain amount of blue, the surplus, there- 

 fore, will be thrown down. Sunlight with a heliostat was used, 

 and the light made oblique, in one azimuth, by an achromatic oil- 

 immersion condenser N.A. 1 ■ 4. 



The theoretical resolving limit for oblique light may roughly 

 be taken to be 100,000 times the N.A. of the objective, and the 

 practical at something a trifle short of this ; but 100,000 N. A. is a 

 convenient form for a " memoria technica." 



In conclusion, might I be allowed to suggest that the labour of 

 ruling these bands would be considerably lightened if fewer lines 

 were ruled in each band ? 



It is, perhaps, a mistake to have all the bands ruled upon one 

 plate ; for a complete set of test bands four plates would be 

 preferable. The first might contain bands from 1000 to 10,000 

 per inch; the second, 12, 14, 16, 18, 20, 25, 30, 35, 40, 45 thousand 

 per inch; the third, 50 to 100 thousand; and the fourth, 90, 110, 

 120, 130, 140, 150, 160 thousand per inch. 



The first plate would be suitable for loups and very low-power 

 lenses, the second for lenses up to §, the third would carry us on 

 from that point up to a dry lens of maximum aperture, and the 

 fourth from thence to the limiting power of the Microscope, as at 

 present constituted. , 



In the fourth plate, the 90,000 band is inserted as a strong 

 band to adjust the lens upon. 



This band-plate is a beautiful specimen of Mr. Grayson's un- 

 rivalled ruling's. 



o- 



Since the above Note was written, I have received from Mr. 

 Grayson two more examples of his rulings. The lines in these 

 plates are ruled with marvellous accuracy. Just at this time a 

 new screen was obtained, giving results superior to any as yet 



