16 



all. This work can be accomplished (1) by means of individual plant perform- 

 ance records whereby the best individual parent plants for propagation can be 

 found, and (2) through the propagation of the superior plants selected on the 

 basis of performance records and progeny tests. 



INDIVIDUAL PLANT PERFORMANCE RECORDS. 



Individual plant performance records for plant improvement work include 

 systematic records, for as long a period as is found to be necessary in order to 

 determine the inherent plant characteristics, of the quantity and commercial 

 quality of the crop, the variability of the product, the season of production, the 

 habit of growth of the plant, and any other data likely to be of value in judging 

 the behavior and value of the plant. In order to illustrate this work the method 

 of securing individual citrus tree performance records as carried on by the 

 writer and his associates in southern California for the past eleven years will be 

 briefly described (48, 49, 51, 64). There have been two phases in the evolution 

 of this work first, the investigational, and, second, the commercial. In the in- 

 vestigational work carefully-selected plats of about one hundred full-bearing trees 

 each were located in several of the best commercial orchards of the Washington 

 Navel orange, the Valencia orange, the Marsh grapefruit, the Eureka and Lisbon 

 lemon varieties. These locations were decided upon with particular regard to 

 environmental influences and cultural conditions, so that reliable data of inherent 

 tree characteristics could be secured. The location of two orchards in which the 

 investigational tree performance records were secured is shown in Plate 33. 



Each individual tree in these plats was given a number so arranged as to 

 both identify and locate it. This number was painted on the tree trunks with 

 pure white lead paint, always in the same relative position and so as to be easily 

 legible and conveniently found when desired. 



The fruits from each tree were picked by or under the immediate supervision 

 of the writer or his associates by expert pickers for the entire period of the 

 investigation. The fruits from each tree were studied immediately after picking, 

 as shown in Plate 34. They were assorted into the standard commercial grades 

 for the variety. The fruits of each grade were then assorted into the standard 

 commercial sizes for the variety. The weight and number of fruits of each size 

 in each grade were then carefully determined by the writer or his associates. So 

 far as possible, the same man conducted this work in each plat every year in 

 order to avoid differences due to the personal equation. After weighing and 

 counting each lot, the fruits were examined for variations and the number of 

 variable fruits of each class was recorded, as shown in Tables 1 and 2. During 

 the course of this performance record work descriptions of tree and fruit char- 

 acteristics were recorded. These studies were conducted on a total of 858 Wash- 

 ington Navel orange trees, 179 Valencia orange trees, 253 Eureka lemon trees, 

 128 Lisbon lemon trees, 104 Marsh grapefruit trees, and a number of trees of sev- 

 eral citrus varieties of minor commercial importance. These records covered periods 

 of 4, 6, and 8 successive years, as the circumstances warranted, in order to carry 

 out the purposes of this investigation. The work entailed a large amount of 



