SErnNTEENTH ANNUAL MEETING. 65 



elled. imperfectly developed (not premature) fruit. The trees 

 usually die, root and Ijranch, the same season, altlious^ii in 

 jMissouri and Arkansas a few cases have been observed where 

 the tree was only half affected and the good half survived 

 until the second year. I believe this takes i)lace on plums 

 more frequently than on the peach. This rapid death of the 

 trees affected b)- Rosette is an advantage to the orchardist, as 

 the disease is mostly self-eradicatory. 



Host Plants. 

 Peach yellows occurs mainly on the peach, but it attacks 

 also the Japanese group of plimis. No other plums are affected 

 by it, apparently. It also attacks the apricot, almond and nec- 

 tarine. The little peach is known only on the peach and the 

 Japanese group of plums. The rosette occurs on the peach 

 and the native Chickasaw plum and probably also on the Jap- 

 anese group of plums. Apparently peach yellows and rosette 

 are native American diseases. Little peach may possibly be, 

 but I doubt it. It did not appear until a few years ago, the 

 date of appearance corresponding about with the introduction 

 of the Japanese plums. My supposition has been, therefore, 

 that it was introduced with the Japanese plum, but further 

 investigations in the native home of these fruits will be neces- 

 sary to determine this. 



Characteristics and Behavior. 

 Coinnniiiicabilify. Since we have not been able to as- 

 certain the cause of the yellows, it is not surprising that the 

 method of infection also remains to be discovered. So far, 

 yellows has only been reproduced artificially by actual bud- 

 ding or grafting from diseased trees. By this method yellows 

 has been reproduced quite frequently in experimental work, 

 particularly in the experiments of Dr. Erwin F. Smith. Re- 

 cently in making some tests in comj^arison with the little peach 

 disease, both little peach and yellows w-ere propagated by 

 budding. Dr. Smith also freely reproduced the rosette in 

 Georgia ])y budding in nursery stock. In this case it is not 

 neces.sary for the bud to live, provided the bud shield or a 



