METHODS OF STUDYING VISION IN ANIMALS 65 



whole spectrum in terms of the divisions on the scale, in exactly 

 the same way as the curve for the scale of every spectrometer 

 is plotted. The mercury arc gives very great assistance in this 

 calibration. The wave-lengths admitted by any given opening 

 can be immediately read off without the use of a spectrometer. 

 Since several of the solar lines are also clearly visible upon the 

 polished jaws of the slit, all one needs to do when the solar 

 spectrum is used is to make the D-line of the solar spectrum 

 coincide with the D-line of the spectrum obtained by the use 

 of the brine-soaked positive carbon. The distribution of wave- 

 lengths in the solar spectrum will coincide with that of the arc. 

 Since the D-line can be so conveniently used, the apparatus 

 should be tested every week or so, to see that the D-line always 

 falls upon the same scale division. 



In the actual setting of the jaws so as to admit the passage 

 of the monochromatic beams through the slit, the following 

 method must be used: In fig. lo the jaw J, as shown by its 

 index, stands at 3.65. Suppose we desire to admit some other 

 portion of the spectrum, say that which would be admitted 

 when the scale should read 6 for J and 5.9 for Jj respectively; 

 we bring the jaw J back until the index reads 6 or more. The 

 small jaw Ji is then moved up by hand to touch J. J is then 

 screwed forward by the micrometer screw-head Cal until the 

 index reads 5.9. J is then brought back until its index reads 6. 

 An optically perfect slit is thus formed, i mm. wide. After 

 J 3 and J, have been set in a similar way, an opening, O, is left 

 between J^ and J 3. This opening must be closed with black 

 oiled skin, thin soft black leather or blackened tinfoil, or by 

 some other pliable opaque material. Four little projections 

 are to be found upon Ji and J 2 for this purpose. They are 

 marked P, Pj, P2, and P3. This strip should be fastened on 

 before Ji and J 3 are withdrawn from J and J 2 respectively. 

 In the diagram the two slit openings are shown as Sg and Sr. 



The jaws Jj and J 2 are each 5 mm. wide; consequently, with 

 these two jaws two places in the spectrum lying closer than i 

 cm. cannot be obtained. A series of single strips, knife-edged 

 on both their vertical surfaces and beveled at their ends, should 

 be made up. These strips should vary in width, starting at i 

 mm. and increasing by .5 mm. up to i cm. Each should be pro- 

 vided with a spring clip at the back narrower than the front 



