32 STUDIES IN PHYSIOLOGY. 



the actual injrease in distance between any two or more lines. 

 In all cases mark the part under examinatiou with good India- 

 ink, making clear, narrow lines. To avoid any possible error 

 caused by influence of lines marked only on one side, make 

 lines on both sides of a part whenever possible. To measure 

 the growth of leaves, use the method spoken of on page 156. 



2. Measitrement hij a micrometer eye-j:iece. With the tube 

 of the microscope kept perfectly horizontal, examine the position 

 of a line of India-ink, upon a perianth leaf of Crocus, or upon 

 the root-cap of Windsor bean. Observe the space which the 

 image of the line appears to pass through in a given time, and 

 refer this to the previously determined values of the spaces of 

 the micrometer. 



3. Measurement by an index, (a) On a simple arc. For 

 this use the simple and admirable modification of Sachs's aux- 

 anometer, devised by Bessej- (American Naturalist). 



{b) On a recording drum. A slender brass or steel shaft is 

 attached to the hour-spindle of a cheap clock, and from the shaft 

 is suspended firmly a stiff pasteboard drum of about the same 

 size. This revolves with the spindle, and if well made is 

 carried without any^ appreciable vibration. A piece of glazed 

 paper of the size of the drum is moistened, and a little mucilage 

 placed on one edge, so that when the paper is rolled around the 

 drum, its edges can be firmly fastened together. Be careful to 

 have the seam in the paper so placed as to avoid any catching 

 of the needle index attached to the plant. When the paper on 

 the drum is dry, it is smoked lightlj' and evenly over a smoky 

 turpentine flame. The needle at the tip of the index is now 

 placed against the smoked paper so as to press lightly upon it, 

 and, as the drum revolves, leave a clean mark. When a suffi- 

 ciently long record has been registered, the paper is carefully 

 removed and dipped in (not brushed with) a solution of common 

 rosin in alcohol, which upon drj-ing prevents any of the lamp- 

 black IVom coming oflT. 



Two corrections are necessary with this simple apparatus : 

 (1) for the cnrve of the descending needle at the end of the 

 radius ; (2) for anj' changes in the position of the needle caused 

 by the varying amount of moisture in the air. 



For recording temperature, it is possible to use a metallic 

 thermometer with a long index, and have the two records side 

 by side. It is well, however, to liave the needle for the ther- 

 mometer give a different mark in order to prevent any subsequent 

 confusion. 



