284 
JOURNAL OF ECONOMIC ENTOMOLOGY 
IVol. 17 
of numbers. This substantiates the work of Rudolf 1 . The reason the 
coefficient (V 13 . 24 ) is not larger is, perhaps, due to the fact that convection 
currents are greater and humidity less with rise in temperature, the 
former making envelopment of the tree more difficult. 
This study illustrates the danger of using zero order coefficients alone 
in drawing conclusions and shows the necessity of measuring as many 
simultaneous variables as possible and investigating by means of partial 
correlation before drawing conclusions. In the experiments discussed, 
it must be concluded that the results are indicative but do not have as 
great an element of certainty as would have been possible with a larger 
number of observations. However, the method of analyzing the data 
rests upon a substantial foundation and one purpose of the paper is to 
show this need. To use partial correlation, two conditions are necessary; 
( 1 ) all zero order correlation must have linear regressions, and ( 2 ) the 
number of observations must be large. 
Determination of Dosage 
The determination of dosage is now a simple matter since it has been 
shown that, to a certain extent at least, the influence of temperature,when 
above 60° Fahr., can be safely ignored in calculating dosage. In 
making recommendation, on the other hand, 65 degrees Fahr., should 
be set as a minimum for orchard operation. However, because the 
number of observations is rather limited, and since dosage plays so 
important a part in success, it is suggested that growers take account 
of the amount of open space, at least, until further investigation has 
definately proven that this is not a factor. If all the trees where the 
efficiency is 80 per cent or more be assumed as successes, it will be noted 
that there were no failures when the number of cubic feet of tree 
volume was 1,500 or less, and, on the basis of tree space, 1,900 cubic 
feet or less. 
In order that a grower may determine the amount of dust to use,several 
average sized trees should be measured and the amount of material for 
each calculated on the basis of tree volume and then on the basis of tree 
space. The larger amount is the correct dosage for each tree. Add 
these together and take the average, multiplying by the number of trees 
in the orchard will give the amount of dust for one application. It may 
be of interest to note that on the basis of these studies, the requirements 
for 72 trees, which were the average in a number of orchards, varied from 
0.86 lb., to 4.09 lbs., with a mean of 2.53 pounds per tree. 
Ttudolfs, Willem. Nicotine Delivery from Dust Carriers. Jour. Econ. Ent., 
15:421-424, 1922. 
