2 MISS A. BATESON AND MR. F. DARWIN ON THE EFFECT OF 
METHOD. 
The plants used were the Jerusalem Artichoke (Helianthus 
tuberosus) and the Sunflower (H. annuus). 
Both of these grew close at hand in the Laboratory garden, and 
could therefore be used in a perfectly fresh condition. The 
internodes were freed from the external tissue, so that pieces of 
pith 12 to 17 centim. in length and about 5 to 10 millim. in thick- 
ness were obtained. It is essential that the vascular bundles 
should be, as far as possible, removed ; unless this is carefully 
done, the specimen, when placed in water, bends, owing to 
unequal elongation, and is useless for experimental purposes. 
The increase in length was measured by means of an auxano- 
meter-lever. One end of the pith was attached to the bottom of 
a narrow glass jar, the upper end being connected, by means of 
a thread of plaited silk, with the short arm of the lever. 
The form of lever used is a convenient one for growth-experi- 
ments. Jt is made of light wood, 628 millim. in length, and 
turns on a knife-edge. The thread connected with the plant is 
also attached to the short arm of the lever by means of a knife- 
edge-suspension, in the same way as the scale-pans are attached 
to the beam of a balance. Great freedom is thus ensured in 
the movement of the lever, à condition not easy to ensure in 
all forms of auxanometer. 
It will be seen that a theoretical error is introduced by the use 
of a lever in place of a pulley; but when the short arm is of 
reasonable length, 45 millim., and especially when the thread 
connecting it with the plant is fairly long in our experiments 
(38 centim.), the error introduced by the end of the short arm 
describing an arc instead of a vertical line is so small as to dis- 
appear in comparison with other unavoidable errors. 
If a strip of pith is fitted up as here described, and the jar is 
then filled with water, the elongation as exbibited by the descent 
of the long arm of the lever is most striking ; indeed, so rapid is 
its movement, that it travels in many cases over 10 millim. in 
one minute. This rapid movement necessitates an occasional 
alteration in the position of the support on which the lever turns. 
When the long end of the lever has descended from an angle of 
30° or so above the horizon toa corresponding angle below it, the 
support on which the knife-edge of the lever turns has to be 
raised so as to bring the long end of the lever once more above 
