134 



TESTING THE MICROSCOPE. 



attention, on account of the great uniformity of the scales, (b) 

 Those which by reflected light appear light brown, and by trans- 

 mitted light grey or dusky. Longitudinal stria 1 , as in the previous 

 case ; transverse strut, although closer together, somewhat less 

 difficult, though not perceptible with a lower power than 300. (c) 

 Scales of oval form, which appear yellow by reflected and trans- 

 mitted light. With rows of dark points about 2J 3 mic. distant 

 from one another, and which, therefore, ought to be easily seen 

 sharply denned with medium powers. 



The markings of the diatom-valves are much more varied than 

 is the case with the butterfly-scales. As test-objects those are 

 generally used which admit of two or three series of lines being 

 observed ; in the first case the lines intersect at a right angle, in 

 the latter obliquely. Whether these series of lines are due to 

 ridge-like projections, or to slight elevations or 

 depressions, disposed in rows, or to differences of 

 density, is in most cases undecided ; the various 

 views here and there met with, and at times ex- 

 pressed with great confidence, need confirmation. 

 We leave this an open question, and confine our- 

 selves to the characteristics of the microscopic 

 image, which is the sole matter entering into con- 

 sideration in testing. 



4. Pinnularia viridis. A diatom nearly related 

 to Navicula, with transverse striae of about 1'5 

 mic. These stria? ought to be resolved even by 

 objectives of 10 15 mm. focal length. 



5. Pleurosigma angulat'iim (Fig. 82). Two series 

 of striae equally strong and equally inclined to- 

 wards the median-line, intersect at angles of about 

 53 58 (not, as is usually said, at 60) ; there 

 is a third series, somewhat less distinct, at right 

 angles to the median-line, forming, with the two 



oblique series, angles of 61 63 J. With oblique illumination 

 incident at a right angle, each of these series of lines ought to be 

 distinctly seen alone with medium objectives, such as Hartnack's 

 No. 7, or Beneche's Xo. 9 that is, with a magnification of about 

 250 300 linear 1 so that on revolving the object, the three series 



1 These magnifications necessitate an eye-piece amplification of about 4^ to 

 5 linear, <v/., Beneche's No. 2 eye-piece, and Hartnack's No. 3 measured 



FIG. 82. 



