298 



The Journal of Heredity 



The buds of each \ariety were selected 

 from a single tree in each case, thus 

 largeK' eliminating any possible dif- 

 ference in growth of scion due to bud 

 mutation. The growth of different 

 varieties on different stocks was not 

 found to be the primary cause of 

 correlation l)etween size of nursery 

 tree and the same tree in the orchard. 

 We ha\'e found that the trees which 

 are small when set in the orchard are 

 also the small trees in succeeding 

 years. The correlation between cir- 

 cumference in 1916, two years after 

 the trees were set, and the circum- 

 ference in 1921, was found to be Aery 

 high (r = .68). We ma>' conclude then 

 that small trees in the nursery, in 

 general result in small trees in the 

 orchard, due to a considerable extent 

 to the effect of the root stock. 



In citrus trees Webber (j) has 

 found great variability of root stocks, 

 both in size and morphological charac- 

 ters. The A'ariable root stocks are 

 considered one of the primary causes 

 of variability in nursery trees budded 

 on such stock. Webber has shown 

 that large, medium and small nursery 

 trees of Washington navel and Val- 

 encia oranges and Marsh grapefruit 

 retain, to a considerable extent, their 

 relative sizes when grown in the 

 orchard for se\eral years. The growth 

 of the different sizes of nursery trees 

 is not gi\en in detail and it may be 

 (|uestioned if the differences in size of 

 orcliard trees are significant. Webber 

 attributes many of the irregularities 

 in size and fruitfulness of orchard trees 

 to the u.se of the seedling root-stocks. 

 Since root stocks grown from seeds 

 are extremely \ariable and ma\' often 

 be weak and dwarfed it is not .sur- 

 prising that clonal varieties grafted 

 on such stock varies greatly in growth 

 and prf)ducti\ity. The presence of 

 many unpnxluctive trees in our Ben 

 Davis orchard may be attributed, in 

 part at least, to the effect of weak or 

 incompatii)le root stocks. 



V.AKIAHILITV IN IKKI-: TYPK AND I'KO- 



DUCTivrrv die to nrn variation 



In recent years much of the \aria- 

 IiilitN' in performance of fruit trees 



within a clonal \ariety has been 

 attributed to bud mutation. This 

 belief has been strengthened by the 

 work of Shamel and his colleagues 

 with citrus fruits in California. As a 

 result of Shamel's work there has been 

 a general acceptance of the idea that 

 bud mutation may cause increased 

 or decreased \ields, not only in citrus 

 varieties but in other fruits as well. 



In a paper, the publication of which 

 has been somewhat delayed, we have 

 analyzed in considerable detail the 

 data pertaining to bud variation in 

 citrus and apple \'arieties. We will 

 therefore limit the present discussion 

 of bud \'ariation to the more important 

 phases of the work. 



Experiments on bud \ariation in 

 apples have been conducted by Whit- 

 ten in Missouri, Macoun in Canada, 

 and Cummings in Vermont. Whitten 

 (k) selected scions from an exception- 

 ally poor Ben Davis tree and others 

 from the best Ben Davis tree in the 

 orchard. There was no significant 

 difference in the performance of the 

 progeny of the two extreme types. 



Macoun (d) selected scions from a 

 heavy bearing, a regular bearing, and 

 a very unproductive tree in a row 

 of 18 Wealthy trees. A comparison 

 of four years yield of the progeny of 

 the aboA'e trees shows no significant 

 difference in the performance of the 

 progeny of the three parental types. 

 In actual amount of fruit the trees 

 resulting from the most producti\'e 

 parent produced slightly less fruit 

 than either the progeny of the regular 

 bearing or unproductiA'e trees. The 

 differences were in no case statistically 

 significant. 



RecentK' Cinnmings (a) has reported 

 the results of a rather extensi\e bud 

 .selection experiment in Vermont. 

 Scions were selected from productive 

 and unproductive trees of seven va- 

 rieties. Two hundred and forty-eight 

 .scions were irsed, 120 scions from pro- 

 ducti\e trees and 128 from imjiroduc- 

 ti\e ones. At the end of ten years 

 86 of the "|)r()ductive" and 82 of the 

 "uiipro(lucti\e" scions had borne fruit. 

 Ill man>' cases the two classes of .scions 

 were grafted into the same \ariel\- 



