8 BULLETIN 813, U. S. DEPARTMENT OF AGRICULTURE. 
order that the results of the individual-tree studies might be truly 
comparative. 
As a result of experience in similar work with other varieties, the 
performance-record plats of Eureka lemons were located so that they 
included representative trees of several important strains. The 
records of the performance of these typical trees constitute the basis 
for comparing the production of the different strains and their value 
for commercial orcharding. 
The original plat included 105 trees. Later, additional trees 
representing important strains were added, in order to secure as 
complete information as possible concerning the variability of the 
strains. By the selection of plats in uniform soil areas and in sections 
with the same altitude, like exposure, and comparative drainage 
conditions, it is believed that the results obtained are less affected by 
fluctuations due to variations in soil fertility, soil moisture, and other 
environmental factors than would be the case if the records of all of 
the trees in the orchards were used for comparison. 
METHODS OF KEEPING PERFORMANCE RECORDS. 
The methods of keeping the individual-tree performance records 
in the lemon plats were adopted as a result of the experience gained 
during the preceding similar study of orange and grapefruit trees. 
These methods involve more detailed observations than is usually 
considered necessary in commercial-orchard practice. A modification 
of these methods, adapting them for regular lemon-tree record keep- 
ing, is described in Farmers' Bulletin 794, entitled "Citrus-Fruit 
Improvement: How to Secure and Use Tree-Performance Records." 
TREE NUMBERS. 
The individual lemon trees were given numbers so that the data 
obtained from the study of each tree could be kept separate during 
the entire period of observation. Each number consists of three parts 
— the number of the block in which the tree occurs, the number of 
the row in the block, counting from some fixed point (as, for example, 
the irrigation head), and the number of trees in the row. Where 
several different orchards are under observation, the tree number is 
preceded by the name of the orchard or an abbreviation indicating it. 
In the case of bearing trees it is most desirable to paint the number 
on the tree trunk or on one of the main limbs. The figures should be 
large and plain, so as to be readily identified from some distance. 
Ordinary white-lead paint and a small brush are the materials used 
for tree numbering. The numbers, when properly applied, have re- 
mained in a satisfactory condition for six or seven years. Where it 
is desirable to number small trees, the number may be stamped on a 
metal tag, or painted on a wooden label, and attached to one of the 
branches. The numbers should always be placed in the same relative 
position on all of the trees, so that they can easily be found. 
