48 PEACH ROSETTE. 



flavor of these fruits was natural and they were ripening only a week 

 or ten days in advance of the proper time. 



(d) Roots and rootlets. Superficially the roots appear to be normal, 

 but the rootlets are dead and shriveled as in peach yellows. 



(e) Gummosis. The green and shriveling fruits from resetted limbs 

 were full of minute gum-pockets, but this symptom was afterwards 

 observed quite frequently in imperfect fruits taken from healthy trees. 

 The" June-drop " from Elbertatrtes was very copious at Griffin, Georgia, 

 in 1891, and most of these fruits were smooth externally but gummy 

 within, the entire seed-cavity being changed into a gum-pocket in many 

 cases. These fruits were an inch by an inch and a half in diameter 

 at time of dropping. 



Many of the roots of resetted trees were honeycombed by gum-pockets 

 (June, 1891). This symptom was found to be nearly constant j i. e., it 

 occurred on some roots of nearly every affected tree. But it was also ob- 

 served to a slight extent on the roots of two trees which stand very near 

 diseased ones, but which have not yet developed any symptoms of the 

 rosette. Many of the roots were so gummy as to feel sticky when 

 bruised. By making a clean cut at right angles to the main axis of the 

 root the gum-pockets were frequently visible without the aid of a trip- 

 let, especially when the oozing of gum was copious. These pockets were 

 confined generally to the outer part of the wood cylinder, and in several 

 cases, known to have been healthy in November, 1890, the cylinder of 

 gum-cavities was confined to the wood laid down in 1891, or to the 

 dividing line between this wood cylinder and the next inner one, and 

 did not extend into the wood formed in 1890. 



These closed gum-pockets involve few to many wood fibers, and fre- 

 quently include the very resistant cells of the medullary rays. On 

 cross-section they appear like an interrupted ring, or like a necklace of 

 triangular beads, in which each member corresponds to a pocket. The 

 base of these triangular cavities was always nearest the pith. Occa- 

 sionally two cylinders of gum-pockets were observed, one in wood formed 

 early this spring when the disease first appeared in the trees and the 

 other formed in wood recently laid down. In a few cases they occurred 

 also in the bark. The walls were brownish and irregular, with modified 

 and frequently hyper trophied and projecting cells. These cavities ap- 

 peared to be free from mycelium and bacteria, and cultures from the 

 walls and the exuding gum gave no positive results. 



The roots of about fifty trees were examined for this symptom, but 

 lack of time prevented the examinations from being as extensive and 

 complete as desirable. 



(3) First appearance and present distribution. So far as I can learn 

 by diligent inquiry this disease first appeared in Georgia about 10 

 years ago. It may have existed longer, but satisfactory evidence of 

 this is wanting. It occurs in the middle and north part of Georgia 

 quite generally, but I have not observed it in South Georgia or in Flor- 

 ida, nor in South Carolina or States to the north. 



