PEACH YELLOWS AND PEACH ROSETTE. 169 



II. INOCULATIONS. 



Experiment 1. — The trees selected for this series of inoculations were 

 grown from Smock seed, procured in Kent county, Maryland. They form- 

 ed a part of a large nursery owned by Norris Barnard, Still Pond, Mary- 

 land. When first seen the seedlings were about five months old, and were 

 being worked for commercial purposes. They numbered more than 100,- 

 000, and all presented a very healthy, thrifty appearance. The trees de- 

 voted to the experiment were in one corner, and not different in appear- 

 ance from the rest. 



The buds for inoculation were cut in an orchard on the Bay farm of 

 James S. Harris, Still Pond, Maryland. They came from diseased shoots 

 on fifteen or twenty vigorous, four year old trees. These trees had shown 

 no symptoms of disease until that summer, when they bore the red-spotted 

 prematurely ripened fruit as well as the characteristic shoots. 



The buds were cut August 12, 1887, and inserted the same day. The 

 manner of insertion was like that ordinarily practiced in reproduction by 

 budding, i. e., the bud, with a portion of the surrounding bark and often 

 some of the underlying wood, was inserted under the bark of the seedling 

 about six inches from the ground, by means of a T-shaped slit. The insert 

 was then bound into place securely by strings, which were cut or loosened 

 at the expiration of ten days. According to their size the trees received 

 one or two buds, none more than two. 



In most instances the wood and bark which were inserted healed on 

 quickly and retained their vitality over winter, but there was no growth 

 from the buds that autumn, nor any symptoms of disease in the stocks. 



The seedling tops were removed in the spring. 



One year from budding these trees were re-examined. Three quarters of 

 the inserted buds had failed to push. Of the rest, some had grown into 

 diseased shoots; others (a few) had grown into shoots which did not yet 

 show the characteristic symptoms of yellows. The effect on the stocks 

 was marked. About thirty-four per cent, of the whole number (202) had 

 become diseased beyond question, while only twenty-three per cent, were 

 entirely healthy. Some of the trees had died during the summer, evi- 

 dently from the effects of the disease. A few yet living were badly 

 affected, but most of them showed only slight symptoms. Nevertheless, 

 here were the feeble shoots and the winter buds germinating six months in 

 advance of the proper time. 



The evidence seemed to be overwhelming and complete, for while the 

 experiment was performed in a region where the disease occurs, it was on a 

 large scale, and moreover the remainder of this nursery, and other nurser- 

 ies in the same region, subject so far as known to the same influences, were 

 examined in vain for anything which would correspond. The 542 trees in 

 adjoining rows, which were examined critically for comparison, showed no 

 traces of the disease, although all other conditions appeared to be identi- 

 cal. The fact that peach yellows in that locality was comparatively rare in 

 trees under three years of age is also opposed to the view that the 

 proximity of diseased orchards had anything to do with this special case. 

 With a full knowledge of the facts, the conclusion was irresistible that the 

 infection came only from the inserted buds. 



An examination made in November, i. e., three months later, showed 

 plainly that the disease was progressing — -more trees were dead and fewer 

 were healthy. At that date the number of stock s clearly affected amounted 

 22 



