Balls. Temperature and Growth. 567 
circumstances it is possible to read to -0003 mm. with this apparatus, but 
readings to less than -coi mm. are usually to be regarded as doubtful 1 . 
After some time has elapsed the hyphae resume the growth which 
had been interrupted by their transference from the test-tube to the 
chamber ; a suitable one is then selected for measurement. The require- 
ments are— rapid growth in a horizontal direction, and the possession of some 
mark near the apex which may be used to check any general movement. 
When the growth at a constant low temperature has been determined, 
the small flame is put under the copper plate, and the following readings 
are taken. 
Readings. During the progress of an experiment the following readings 
are taken once every minute - 
(a) Time. 
(b) Position of control-point of hypha on micrometer scale. 
(c) Position of end of hypha on micrometer scale. 
(d) Temperature of control-bath. 
(e) Deflexion of galvanometer. 
In practice the two latter are taken fifteen seconds earlier than the 
others, and a correction made, if necessary, when the curve is plotted. 
When the rise in temperature of the chamber has carried the galvano- 
meter to zero, the hot-water syphon is opened and the bath temperature 
raised about i° C. To ensure thorough mixing of the newly-added hot 
water with that already in the bath, it is stirred up by blowing air through 
with a small hand-pump, the tube from this pump being weighted with 
lead to prevent it from jumping about in the bath and injuring the couple. 
The pump is worked with the left hand while the other is writing the note 
of the last reading. 
One galvanometer reading is omitted after thus adding water, in order 
to make sure that the bath-couple, thermometer, and water are all at the 
same temperature. The amount added is adjusted by rule of thumb so 
that the galvanometer shall return to zero after two or three minutes. 
The fungus may move slightly in the field ; the adjustment to correct 
this is made by the mechanical stage. It is fortunate that the later stages 
of the experiment, when the fungus is near to the stopping-temperature, 
are rarely troubled by this movement, which seems to be due to the 
expansion of the copper plate. 
The process might appear to be rather feverish from this outline, but 
in practice there is always time to spare. After some experience one 
experiment can be completed in about two or three hours, the last hour 
See curve ‘ Growth per Minute’ in Curve r. 
