Crown-Gall of Apple and Peach 17 



general considerations 



Field experimentation is very unsatisfactory at best, because so many 

 uncontrollable factors are involved which interfere with a correct inter- 

 pretation of results. This condition has come to be rather generally 

 understood, the recent analyses by Harris 10 of results from field plot tests, 

 and by Anthony and Waring 1 L for orchard tests, only serving to indicate 

 how generally unreliable such data are. In addition to the usual vicissi- 

 tudes encountered in plot testing, an additional set of variables is added 

 when one attempts to compare plants in health and in disease. Means 

 have been devised for overcoming heterogeneity in field plots, but not 

 enough is known about plant responses under conditions of parasitism 

 to make suitable corrections. This is particularly true in the case of the 

 disease under consideration, because Smith has shown that the response 

 to the stimulus of Bacterium tumejaciens depends a great deal on the 

 particular tissue affected. This individual variation can be eliminated 

 by increasing the number of plants under test. The number of plants 

 involved in a wholly satisfactory experiment obviously should be 

 considerably greater than was included in this experiment, and the 

 arrangement should be different. 



The general indication, however, is that under the conditions of this 

 experiment, crown-gall is not a factor in the growth and development 

 of either apple or peach trees. 



The agitation among fruit growers in regard to crown-gall has practically 

 disappeared, doubtless due to the inspection service which causes rejection 

 of trees bearing galls. The writers, however, in 1914 received a consign- 

 ment of 400 apple trees, of which 28 (7 per cent) bore definite lesions of 

 crown-gall. All of these were planted on private land and their exact 

 location was recorded. While it is likely that many growers treat galled 

 trees just as the writers did, it is to be recalled that the amount of crown- 

 gall on nursery stock has not been large, and the rejection of trees on 

 account of crown-gall has not been a serious loss to nurserymen. 



The role of crop rotation in reducing losses from crown-gall in the nursery 

 is an interesting phase of the problem, on which, unfortunately, experi- 

 mental data are too meager to warrant any conclusions. 



••Harris, J. Arthur. Practical universality of field heterogeneity as a factor influencing plot yields. 

 Journ. agr. res. 19:279-314. L920 , , 



"Anthony. H. D.. and Waring. J. H. Methods of interpreting yield records in apple fertilization 

 experiments. Pennsylvania Aicr. Exp. Sta. Bui. 173:1-42. 1922. 



