150 CORRESPONDENCE. 



particles that I have yet been able to detect with any degree of cer- 

 tainty are estimated at ^^owiro*^ ^^ ^^ inch diameter. The observa- 

 tions were made with a new -V^h immersion of 165° angle aperture, 

 made by E. B. Tolles, Boston, Mass. 



The method used by me " of arriving at an estimate " is a slight 

 modification of the old double-sight mode, I paste a i^iece of paper 

 of suitable colour on the end of the object slide, on which a series of 

 dots are made of various known sizes. The magnified molecules in 

 the instrument are readily compared with these dots or circles by 

 double sight, and the magnifying power being known, by a simple 

 calculation, their sizes are approximately ascertained. This is not 

 given as the best possible method, but as one having some advan- 

 tages. 



In regard to the third question, viz. " Have the most recently con- 

 structed microscope objectives, such as the g^oth or ^V^^^) ^^7 advan- 

 tages over the i\th. or y^th inch objectives in the determination of 

 the data above referred to ? And have immersion lenses any advantage 

 in this resj)ect ? " I would suggest that the great uncertainty in 

 determining the size of extremely minute particles renders any com- 

 parison based on such observations alone, almost, yea altogether, 

 worthless. It seems to me therefore that it would be better to de- 

 pend on the recognized tests of definition, at least for the present, and 

 draw the legitimate inference that an objective that gives the finest 

 definition on these will also do the best work on the " particles." 



The Jyth above referred to has in my hands done better work on 

 the most difiicult test scales and diatoms than has up to this time been 

 done by any " yVth or ^^th," as far as is known to me through the 

 published performance of other lenses or otherwise. First-class l^ths 

 to ^V^h are showing the transverse striae of Ampliipleura pellucida, 

 Navicula crassenervis, Frustulia saxonica, and Nitzschia curvula. The 

 ^th reveals longitudinal lines on all these, much finer than the trans- 

 verse, and evidently genuine. Under favourable conditions the reso- 

 lution into the so-called beading is distinctly effected on the first three 

 named. The diameter of each dot on Amjihipleura j)eUucida is pro- 

 bably not over the 270^0 oijth of an inch. JByalodiscus suhtilis Bailey 

 is instantly seen covered, throughout the hyaline portion, with nicely 

 defined hexagons in place of the fine " rulings." 



This is done with sunlight and the ammonio-sulphate of copper 

 cell. With ordinary day or lamp illumination, either central or 

 oblique, on scales or Bacteria the performance is also excellent. 



This objective works either dry or wet by turning the adjusting 

 collar, but its immersion work is preferred. 



The preceding facts appear to me to demonstrate the superiority 

 of the best g^th over the medium powers, for the kind of investiga- 

 tion H. H. is engaged in. 



Yours respectfully, 



G. W. Morehouse. 



