MISCELLANEOUS PAPERS. 305 



80 modified by climatic influence and crossing of the different types 

 that it is difficult to devise a system which will include all varieties and 

 not be subject to some criticism. But if a system can be devised by 

 which one can distinguish a large number of varieties and predict 

 within a reasonable degree of certainty which types are likely to suc- 

 ceed in a given section, a long step is made in the direction of most 

 valuable knowledge. 



There are types coming from China, Japan and Java, and others 

 which are now being built up in the southern part of the United States, 

 which the above points do not distinguish with clearness. We have 

 seedlings from the Honey peach which came from China in 1854; seed- 

 lings from the Peen-to, which came from Australia in 1869; then, again, 

 there are seedlings of the Spanish or Indian type, all of which are 

 adapted to more southern latitudes than those of the Persian strain, 

 which constitute the bulk of northern orchards. These new types are 

 proving valuable along the Gulf states, where the peach has not been 

 grown as successfully as it has farther north. It is important now 

 that the botanical characteristics of each type (which we shall call 

 race for^a better term) should be recorded with clearness, and the cli- 

 mate in which each grows best be pointed out. 



From what I have thus learned it seems to me that what is known 

 as the " Onderdonk classification " is the best. An outline of this 

 scheme appears in the report of the United States Department of Ag- 

 riculture for 1887, page 648. Some of the distinctions made in this 

 classification cannot be noticed with decisive clearness a few hundred 

 miles farther north, but in the semi-tropical climate of the coast region 

 the characters are striking. This at once indicates that the differ- 

 ent races originated in different degrees of latitude and at different 

 altitudes. These facts are further substantiated by botanical charac- 

 teristics. We divide the peaches now cultivated in the United States 

 into five races: (1) Peen-to, (2) South China, (3) Spanish or Indian, 

 (4) North China and (5) Persian. By race is meant "a variety so 

 fixed as to reproduce itself with considerable certainty by seed." 



Seed characteristics The shape, size and corrugations of the seed 



are so well marked in the more distinct representatives of the differ- 

 ent races that after a little practice one can distinguish them apart by 

 this means alone. 



The seed of the Peen-to is nearly round, much compressed at the 

 ends, corrugations small and somewhat rounded. The seed of the 

 Honey is oval, with apex slightly recurved, corrugations slight, promi- 

 nent flange on one side. Seed of Spanish is large, oval, nearly flat, 

 apex prominent, corrugations very long and wide, at the base they run 



