86 



literature, concerning the behavior of the flowers of unimproved 

 trees in the tropics. Since such trees seem closer to the ancestral 

 wild stock than the highly selected varieties chosen by commercial 

 orchardists, a record of the behavior of a few of them might con- 

 tain certain points of interest. 



Stout's careful studies of the flower behavior and the pollina- 

 tion of the avocado (2, 3), begun in southern California during the 

 winter of 1922-23, are the basis of our knowledge of the subject. 

 He discovered that each flower has two distinct periods of opening, 

 separated by a period of variable duration in which it is closed. Dur- 

 ing the first opening the anthers remain closed and shed no pollen, 

 but the pistil is receptive to pollen brought from other trees. In the 

 intervening closed period the stigma withers, with the result that 

 at the second opening of the flower on the following day, when the 

 anthers dehisce and shed their pollen, self-pollination is no longer 

 possible. Since all flowers of a set on the same tree behave in the 

 same manner, opening and closing simultaneously, agreeing in the 

 condition of their anthers and stigmas, with slight or no overlap- 

 ping of the first and second opening of different flowers, close-polli- 

 nation is very unlikely to occur. 



In respect to their periods of dianthesis, all avocados so far 

 studied fall into two distinct groups. In the one, which Stout has 

 designated as Class A, the flowers open first with receptive stigmas 

 in the morning, when we may for conciseness term them "function- 

 ally pistillate," close some time during the middle of the day, re- 

 main closed for about twenty-four hours, and open the second 

 time to shed their pollen during the afternoon of the succeeding 

 day. After persisting a few hours as functionally staminate flowers, 

 they close in the late afternoon and never again expand. Trees of 

 Class B are characterized by the fact that their flowers open first 

 in the afternoon, — like the former functionally pistillate during the 

 first anthesis, — and close again late in the day, but on the basis of 

 the time of the second expansion their flowers fall into three groups. 

 In the first (Bj). all the flowers of the tree open for the second 

 time and shed their pollen over a period of several hours the fol- 

 lowing morning. The life of these flowers is roughly 24 hours. In 

 the second (Bg), after closing at the end of their first period of 

 expansion, the flowers remain closed during two nights and a day, 

 and open again only on the second morning following, when they 

 shed their pollen as in the case of the preceding group. Their life 



