82 EXPERIMENTS IN BLUEBERRY CULTURE. 



The largest berry measured on this bush was 14.02 mm. (0.552 of an 

 inch) in diameter. 



Three quarts of berries were picked from the bush; all those less 

 than 10 mm. in diameter were discarded, and the remainder, about 2 

 quarts, were carried to Washington for seed purposes. 



(38) There is every reason to beiieve that tjie blueberry can be improved 



• BY breeding and BY SELECTION. 



The swamp blueberry {Vaccininm corymhosum) is an exceedingly 

 variable bush. There are three especially well-marked forms, called 

 V. cnnoenum, V. atrococcum^ and V. follkhnn^ by some authors 

 regarded as distinct species, by others as forms of F. cori/mbosiim. 

 "Within the limits of these forms variation is also extensive. There is 

 great opportunity for selection among wild varieties in the size, color, 

 flavor, and time of ripening of the berries and in the productiveness 

 and vigor of the bushes. 



That types possessing desirable qualities can be crossed there is no 

 question. A method of pollination has already been described (see 

 p. 78), Avhich, supplemented by the removal of the stamens on the 

 female parent before they have matured their pollen and also by the 

 protection of the pollinated flowers from insects, would insure a 

 genuine cross. 



The possibility of securing valuable varieties is accentuated by the 

 marked variation observed in the character of the offspring of the 

 large-berried bush from which the seedlings of 1908 were grown. Be- 

 sides minor variations, these seedlings show three forms which may 

 be regarded as types. One of these, characterized by its low stature 

 and leaves tending to be conduplicate and by its long persistence into 

 the winter in a green state, is perhaps the result of some pathological 

 difficulty. Two of the types, however, appear in every way to be 

 normal. One has its leaves large, obovate-elliptical, glaucous on the 

 back, and with entire margins, such as are possessed by the parent 

 and are typical of true Yacclniwn corymhosam^ and it develops only 

 a few though very robust stpms, with few flowering buds. The other 

 has smaller, narrower leaves, green on both surfaces, and with mar- 

 gins closely and evenly serrulate. It produces many stems smaller 

 than those of the other, and more numerous flowering buds. It is 

 strongly suggestive of the plant called Vaccinium amoenum. It is 

 much larger and more robust than V. peniisylvanicwn, and may pos- 

 sibly be a hybrid between that species and V. corymhosum. 



The characters of bush and foliage in these two types have not yet 

 been correlated with any ditferences the}' may show in flower and 

 fruit. It is, however, of great interest that these same two types 

 occur among the seedlings of 1907, as well as those of 1908, which 

 came from a different though similar bush growing about 2 miles 

 from the other. 



193 



