DIFFE ACTION SHADOW. 69 



stance be suspended in a dark room, and a pencil of homo- 

 geneous light be allowed to fall upon it, so that its shadow 

 may be received on a screen, it will be found that a bright 

 spot will appear in the middle of the shadow, outside of 

 which will be a dark circle, and beyond this a bright 

 circle, and so on, the circles corresponding successively 

 to the interference of the rays, by which their brilliancy 

 is either doubled or extinguished ; the colour of the bright 

 circles corresponds to that of the light. If common com- 

 pound light be used, the central spot will be white, and 

 w^ill be surrounded by a series of coloured fringes. 



No rule can be laid down for the avoidance of errors 

 such as we have been dwelling upon, but the practised 

 microscopist will, however, in time learn, almost as it were 

 insti actively, to overcome the difficulties of diffraction, as 

 well as those due to oblique illumination at certain angles, 

 particularly in that which results in the production of a 

 double image or overlying shadow, which we have seen 

 nearly as strougly marked as the real image, This kind 

 of shadow is sometimes called the " diffraction spectrum,^' 

 although it must be evident to anyone who will be at the 

 trouble to study the subject, that it is due to a different 

 cause. It cannot be denied that errors of interpretation 

 are not unfrequently due, as Mr. Brooke has shown, to 

 the increasing desire there is to produce object-glasses with 

 large angles of aperture. Mr. Brooke observes : — " The 

 superiority in resolving power possessed by objectives of 

 large angular aperture is obtained at the expense of other 

 advantages." And Messrs. Sullivant and Wormlej'-, in their 

 paper on " JSTobert's Test-Plate and the Striae of Diatoms," 

 American Journal of Science, 1861, are convinced "that 

 w^hen the resolving power of an objective is near its limit, 

 * spectral ' or spurious lines are to be seen, and these are 

 only to be distinguished from the true by a practised eye.' 

 Hence they think " that the mere exhibition of lines is 

 not always conclusive evidence of their ultimate resolu- 

 tion." * For the same reason we have not the amount 

 of confidence in the higher powers that some observers 

 seem to have ; we know, indeed, with every increase in 

 this direction, how liable we are to encounter unforeseen 

 errors and exaggerations, and we still prefer the -^th or 



