BUREAU OF PLANT INDUSTRY. 143 



is conducive to fruit production. Incident to the bud-selection work 

 it appears evident that trees propagated from those having a high 

 yielding record of desirable fi-uit require less pruning to keep them 

 in good form and in good productive condition than do trees prop- 

 agated according to ordinary nursery practices. The bud-selection 

 work Avhicli has been in progress with citrus fruits in California 

 for a number of years has been continued, though on a somewliat 

 restricted scale. 



From a practical commercial standpoint, the performance records 

 of individual trees which have accumulated up to the present time 

 are serving an exceedingly important purpose, in that they are used 

 as a basis for choosing the trees selected as a source of bud wood by 

 the California Fruit Growers' Exchange, throngh its bud-selection 

 dej^artment. This department Avas organized for the purpose of fur- 

 nishing buds cut from trees of known bearing habits to nurserymen 

 or citrus growers who desired to nse them for top-budding trees in 

 their groves which were known to be relatively unproductive or 

 which produced fruit of undesirable characteristics. The individual 

 tree-perforniiince records of the Euby variety in a 10-acre orchard at 

 Corona. Calif., have been conti]iued, as during the t'wo precedhig 

 years. The information thus obtained emphasizes the striking fre- 

 quency of bud variation in this variety. This, in turn, emphasizes 

 the importance of selecting buds for propagation not only from trees 

 of known bearing proclivities, but even from limbs of known bearing 

 procliA'ities, because trees are exceedingly common which in the 

 main bear fruit of desirable type but have some limbs on which fruit 

 that varies from the accepted standard is produced. 



The records which are now being accumulated confirm more and 

 more emphatically the previously drawn conclusion that the charac- 

 teristics of the parent tree are perpetuated in the progeny with no 

 important exception or variation. The progenies in experimental 

 plantings are producing fruit comparable in every particular with 

 that borne b}^ the parent trees of large-yielding or other desirable 

 characteristics, or, like the parent trees, they are barren or relatively 

 bari'en. 



In order that the standard strains of the different kinds and vari- 

 eties which have been established may he fully maintained and the 

 strains accurately recorded, additional descriptive notes and photo- 

 graphs of selected citrus trees and their fruits have been secured 

 during the past year. 



A large number of commercial citrus-performance records arc 

 being made, the interest of the growers having increased each year. 

 The utility of these records is practical rather than scientific. They 

 enable the grower to determine inferior tr^es for removal or top- 

 working, as well as make possible the selection of superior trees from 

 which buds for use in top-working may be obtained. It has been 

 found that these commercial performance records can be made in 

 Washington Navel and Valencia orange and in Marsh grapefruit 

 orcliards at an average cost of about $1 an acre each year, while 

 similar records in lemon groves cost about $10. the difference being 

 due to the greater number of pickings in the case of tlie lemons. 



Considerable time has been devoted during the year to assisting 

 growers in introducing individual-tree records and to the utilization 

 of the data thus secured. 



