14 The Philippine Journal of Science iw» 
The growth shown in Table VII was very spindling and more- 
over did not keep pace with the death of the plants at the cut 
ends, so that they were all actually shorter at the end than at 
the beginning of the experiment. It was due only to the differ- 
ence in the appearance of the growth before and after the plants 
were placed in the jars that the amount of growth could be 
measured. The dying at the cut end did not occur in any of 
the plants in the jars through which CO2 was passed. The 
growth of the plants in this experiment, run during April and 
May, was much greater than that made during January and 
February by any of the plants in jars in which COg was ob- 
tained only from the air. The plants in this experiment were 
alive at the end of twenty-four days whereas in the previous 
experiments growth ceased in less than three days and death 
ensued in less than twenty. This increased gi'owth was observed 
in all experiments run without the addition of CO. during April 
and May and was probably connected with the greater intensity 
of the sunlight. However, in all cases, the plants were dying at 
a faster rate than they were growing so that, if kept under the 
same conditions, they would probably have died in a compar- 
atively short time. 
The results given in Table VII show that the addition of nu- 
trient salts increased the growth of the plants when CO2 was 
not passed through the water, the plants in the jar containing 
four times the standard strength of Knop's solution growing 
215.7 per cent as much as those in tap water. This difference 
is, however, small when compared with that produced by passing 
CO2 through the water. A comparison of Tables VI and VII 
shows that the growth of the plants, in tap water through which 
CO2 was passed, was in twenty-five days 766.7 per cent as much 
as the growth of similar plants for twenty-four days in water 
to which CO2 was not added. This, moreover, does not take 
into account the unhealthy condition of the latter plants. 
The water used in these experiments came from a small stream 
near Baltimore and may be considered as an average soft water 
under natural conditions. Using it as a basis the experiments 
so far reported would seem to show that the addition of nutrient 
salts is nothing like so important for the growth of Elodea as 
the addition of CO2 other than that absorbed from the air. The 
comparatively small variations produced by keeping the plants 
in solutions containing very different concentrations of nutrient 
salts would indicate, moreover, that under natural conditions 
