994. E. H. HARPER 
EXPERIMENTS WITH SUBSTANCES OF HIGHER SPECIFIC GRAVITY 
The experiments were carried out as follows: The water used 
was ordinary tap water boiled to drive away gases. <A control 
experiment was in all cases conducted side by side with the treated 
specimens. Ordinary test tubes were employed for the experi- 
ments. ‘Iron by alcohol’ was used, being ground up fine in an 
agate mortar. In order that the number of Paramoecia might 
be practically equal in the two test tubes, the water containing 
the animals was measured into equal portions. Large numbers 
were used in order to facilitate observation and comparison 
of aggregations. In the control test tube was placed a quantity 
of finely ground iron and after this had settled to the bottom, 
which occurs very quickly, the measured quantity of Paramoecia 
were introduced into the tube. The others were put in the agate 
mortar with finely divided iron and stirred for a definite time so 
as to keep the particles in suspension. In some cases the control 
animals were similarly stirred before placing them in the test 
tube in order that mechanical agitation should affect the results 
equally if at all, since this is known to change the reaction, some- 
times, to positive. 
The amount of iron ingested is readily observed with the micro- 
scope, and a short treatment may be sufficient to cause the inges- 
tion of a considerable quantity. They were allowed in different 
experiments to ingest the iron for intervals varying from fifteen 
seconds to five minutes before transferring to the test tube The 
control animals and the treated ones were placed in the tubes at 
practically the same time, so that there would be no difference 
in time interval to allow for in comparing the movements and 
aggregations of the animals in the two tubes. As a method of 
recording the observations, which were made with the naked eye 
and hand lens, the approximate distribution was indicated on a 
diagram. This could be made sufficiently accurate to indicate 
any decided difference that might be noted in the regions of 
greatest aggregation as well as the general distribution in the two 
tubes. 
