BUREAU OF PLANT INDUSTRY. 59 



Some difficulty is experienced in determining truly resistant plants, 

 as the percentage of alkali in the soil may differ materially within a 

 distance of a few feet, and a plant apparently resistant may owe its 

 success to a lower percentage of alkali in the soil under it and not to 

 any inherent resistance. Before a plant is finally selected for breed- 

 ing purposes, therefore, it is necessary to make a careful test of the 

 soil immediately beneath it. 



Crossbreeding such resistant plants with resistant varieties imported 

 from other countries also furnishes an important field for experiment. 

 The results already obtained are very encouraging, and if the neces- 

 sary funds are provided the work will be extended and vigorously 

 prosecuted. The work will include the selection of seed from resist- 

 ant plants, the growing and testing of such seed, the crossbreeding of 

 resistant strains, the selection and adaptation of imported resistant 

 strains, etc. The work is of the greatest importance and should 

 receive commensurate attention. 



The orange. — The work of breeding a hardy orange by crossing the 

 hardy Japanese trifoliate with the different varieties of the ordinary 

 sweet orange has not yet reached a final conclusion. It was thought 

 that some of the trees would fruit this year, but they have not done 

 so, owing probably to the fact that the bud wood was taken from the 

 hybrid seedlings while very young. Budded trees of each of the 

 hybrids have been distributed to various experiment stations in the 

 Southern States to be grown and tested and also to furnish bud wood 

 for distribution of any sorts that may prove valuable. Several of the 

 hybrids having trifoliate leaves like the trifoliate orange parent, but 

 much larger, are of evergreen habit, and have shown themselves quite 

 hardy. Should these produce good fruit, they will be sufficiently 

 hardy to endure the cold of ordinary winters in the Gnlf States. As 

 stated in the previous report, some of these hybrids will make valuable 

 hedge plants and will be of great value for this purpose alone. The 

 hybrids which fruit will be carefully tested and studied and those 

 which prove promising will be propagated as rapidly as possible to 

 secure buds for distribution should the results of a second year's fruit- 

 ing show the fruit to hold up. The value of a variety can be determined 

 only after several years' growth under a wide range of conditions, 

 and only the growing of a variety on a commercial scale would fur- 

 nish such a test. Of course, it is manifestly impossible for the Depart- 

 ment to make such a test; but if after two years' fruiting a sort seems 

 valuable it will be grown and distributed. 



Grape fruit and tangerines. — The hybrids of different varieties of 

 grape fruit and tangerines, made particularly to obtain new fruits of 

 better or distinct quality, will fruit this fall for the first time, and it 

 is confidently expected that some of them will show improvements of 

 value. 



Grapes. — All breeding work done with grapes was carried on \>y the 

 assistant in charge of the Pacific coast laboratory, and is discussed 

 under the heading of "Work on the Pacific coast." 



The pear, apple, and peach. — During the year quite a large series of 

 hand pollinations were made with the Kieffer pear in the vicinity of 

 Washington, D. C, in order to get more exact knowledge about this 

 important commercial variety and also to secure new hybrids between 

 the Kieffer and choice dessert pears, such as Seckel, Anjou, Mannings, 

 etc. A fine lot of cross-pollinated Kieffers has resulted from the sea- 



