8 BULLETIN" 697, U. S. DEPARTMENT OF AGRICULTURE. 
mercial inferiority of the product from such, trees constitutes one of 
the important reasons for the systematic selection of bud wood from 
trees of the right character. 
The diversity of strains of the Marsh variety represented in estab- 
lished groves is of great commercial importance to the grapefruit 
industry, for the reputation and value of the crop depends on the 
maintenance of a uniform quality. For this reason, the identification 
of the Marsh strain, which is the most productive and valuable strain 
of the variety, and the maintenance of the purity of the strain by 
bud selection are of fundamental importance in the development of 
the industry. This does not entail any increase in the cost of pro- 
ducing, handling, or distributing the crop. On the contrary, from 
the standpoint of cost of production based on the unit cost per box, the 
production of the best strains can be effected more cheaply than is 
the case when inferior and less productive strains are present. 
Fruits containing many seeds, such as those of the Seedy strain, are 
of much less value than those which are commercially seedless. On 
account of the fact that some of the seedy fruits can not be readily 
separated after picking from those of the Marsh or other strains, the 
presence of trees of this strain in established groves is detrimental to 
the reputation and value of the crop as a whole. 
The late-ripening characteristic of fruits of the Bell strain is also 
detrimental to the effort to build up a superior commercial reputation 
for this variety. The fruits of this strain are difficult to eliminate from 
the general pick and pack of the groves. Because of their unripe con- 
dition, if included with fruits of the Marsh strain they tend to detract 
from the quality of the whole package and to injure the reputation of 
the variety. 
OBJECTS OF THE INVESTIGATIONS. 
The objects of these investigations are (1) to determine the fre- 
quency of occurrence and the nature of bud variations in the Marsh 
variety of grapefruit, (2) to secure definite information concerning the 
commercial value of the different strains of this variety by means of 
individual- tree performance records, (3) to originate and introduce 
methods for isolating the best strains and controlling variation within 
them through bud selection based on tree-performance records and 
definite tree knowledge, (4) to eliminate the trees of inferior strains 
in established orchards by top-working them with select buds, and (5) 
to demonstrate the value and importance of the use of improved 
methods of propagation based on definite individual-tree knowledge 
for the improvement of the production of established orchards and 
for the conservation and improvement of the variety. 
