17, 4 Lee: Fungicides and the Citrus-canker Organism 337 
copper, in most of the tests showed no bactericidal value what- 
soever. In a few tests slight killing was effected with the un- 
diluted mixture at the longer exposures. When the results with 
the Bordeaux mixtures made it apparent that it was excess of 
lime which contributed the bactericidal value, the Burgundy 
mixture was prepared with an excess of sodium carbonate. In 
a comparative test, Burgundy mixture 3-4—50 showed a stronger 
toxic action against Pseudomonas citri than did Burgundy mix- 
ture 3-3-50. The toxic action was far from being as complete 
as in the case of Bordeaux mixture 4—6-50, or even Bordeaux 
mixture 4—4—50. 
A comparison of these results with copper sprays and with 
lime leads to the conclusion that in none of the copper precip- 
itate sprays, where there is no excess of either of the precip- 
itants, is there any bactericidal action. This of course, on 
retrospection, was to be expected from a fungicide which con- 
sists of an insoluble precipitate in suspension. 
The criticism of the value of these tests will be raised, that 
the fungicidal value of copper precipitate sprays depends upon 
the action of the products of metabolism of the fungus itself 
on the copper precipitate; these metabolism products liberate 
the copper in a soluble form which is then toxic to the fungus. 
It will be said then that, although the bacteria are not instantly 
killed in the foregoing copper precipitate tests, upon the tree 
their metabolism products will also liberate the copper from 
the copper precipitate as do the fungi, and that the bacterial cells 
would then be coagulated. 
To prove or disprove this, organisms from a 3-day-old Pseu- 
domonas citri culture were exposed to neutral Bordeaux mixture 
for 60 minutes. At the end of the exposure P. citri was easily 
recovered. P. citri from a 3-day-old culture was then inoculated 
into bouillon tubes containing cubie centimeter of neutral 
Bordeaux mixture to 10 cubic centimeters of bouillon. Growth 
appeared as rapidly in the presence of the Bordeaux mixture as 
in the control tubes containing no P. citri; a copious ring char- 
acteristic of P. citri in vigorous cultures formed at the surfaces 
of all bouillon tubes, both in those containing neutral Bordeaux 
mixture and in the controls. The tests were repeated a number 
of times. 
This is taken as showing that the metabolism products of 
Pseudomonas citri do not cause the liberation of copper from 
the copper precipitate formed in Bordeaux mixture, at least not 
