he copper." This letter was written over three
years after Dr. Moore made his experiment in these cress beds.

Satisfied with the results attained in exterminating algal growth in
water-cress beds, attention was next given to reservoirs. Some fifty
water supplies were treated during the summer of 1904, and in every
case success attended the copper cure. In one respect the results were
surprising. It was found that in practice the copper-sulphate method
worked better than in theoretic experimentation; results in large
reservoirs were more pronounced than in the laboratory. In fact, it
developed that the solution necessary to kill algae in the laboratory
must contain from five to twenty times as much copper as that
contained in a solution which will exterminate algal growth in its
natural habitat. This is not easily explained, if it can be explained
at all. The test reason advanced is that only the most resistant
organisms stand transplanting to an artificial environment. But, after
all, the important point is that the new method works better in
practice than was expected.

_A Prescription for the Copper Cure_

Thus the department is able to announce that the process is no longer
in the experimental stage, and also to say what conditions must be
known in determining the proper quantity of copper sulphate for
destroying algae, together with a prescription for the copper cure.
Here it is, for the benefit of careful persons who will use the method
with proper intelligence: "The importance of knowing the temperature
of the contaminated water is second only to the necessity of knowing
the organism present. With increase of temperature the toxicity of a
given dilution increases, and _vice versa_. Assuming that 59 deg. F. is
the average temperature of reservoirs during the seasons when
treatment is demanded, the quantity of copper should be increased or
decreased approximately 2.5 per cent for each degree below or above
59 deg. F.

"Similar scales should be arranged for the organic content and the
temporary hardness of the water. With the limited data at hand it is
impracticable to determine these figures, but an increase of 2 per
cent in the quantity of copper for each part per 100,000 of organic
matter and an increase of 0.5 to 5 per cent in the proportion of
copper for each part per 100,000 of temporary hardness will possibly
be found correct. The proper variation in the increase due to hardness
will depend upon the am