28 



M. haccata; three bags enclosing 24, 35, and 58 flowers, a 

 total of 117 flowers; in each of two bags one fruit developed, in 

 the other there were two fruits. 



In another tree labelled M. baccota, obtained from a difl'erent 

 source, of 199 flowers enclosed in li\e bags a total of 50 fruits 

 developed. ApparciUly tliis plant was more than feebl}' self- 

 compatible. Difl'erence such as this is frequently seen for dif- 

 ferent individuals of a single species. 



.1/. . Inioldiona; Ave bags enclosed from 29 to 50 flowers each 

 with a total of 190 flowers but only three fruits matured. 



iM. Sieboldii; 5 flowers, i fruit ; 35 flowers, 2 fruits ; 38 flow- 

 ers, 2 fruits; 42 flowers, 2 fruits; 54 flowers, 5 fniits. For this 

 plant, of the 174 flowers enclosed, only twelve gave fruits. 



The examination of pollen under microscope and tests for its 

 germination in artificial media were made for each of the plants. 

 For the pollen of M. coronaria the percentage of " good " or 

 germinating grains was low (about 10%) but those that did 

 germinate made a vigorous growth. In all other cases the pollen 

 was excellent with vigorous germination of from 75% to at 

 least 90% of all grains. Pollen aborti(^n is hence not in any 

 way accountable for the self-fruitlessness of these apples as it 

 is in such plants as the grape in which intersexes are present. 

 Of the trees tested, only the tree of M. haccata and the tree of 

 M. syivcstris bore lightly, but for these there was a light bloom, 

 so the trees in question were able to produce fruit. 



It seems evident from these results that self-incompatibility 

 is present in at least some individuals of certain species of apples. 

 The flower parts are perfect and able to function in certain cross 

 relations but there is failure to ferlili/.e and to set fruit to self- 

 pollination. The tests indicate that some individuals were fully 

 self-compatible, others feebly self-compatible and others com- 

 pletely self-incompatible. This type of sterility with such in- 

 dividuar*differences as here observed is freciuent even within 

 many species of plants both wild and cuhi\;ited. It should be 

 noted that the tests do not show the extent to which the many 

 individuals grown from seed in each species are self-incomi)ali- 

 ble. Possibl}' the \arious grades exist in many or ah the species 

 noted above. Some individuals may be fully self-compatible 

 and others fu!l\ self-inconi|)atibK' in llu' same species. Tests 



