COdONUT PLANTER^ MANUAL. 155 



The results of investigations have convinced the writers : — 



(1) That tlie inflorescences on the same tree follow each other after 

 an average interval of four days. 



(2) That in two cases only out of forty-three, did overlapping of inflores- 

 cences occur. 



(3) That the average duration of the male phase was 2L days, with a 

 variation of from 15-24 days. 



(4; That the average duration of the fenale phase was S days. This 

 average seems independent of the number of female flowers on the infl >res- 

 cence, hut might be reduced over a longer period as the weather was un- 

 usually showery, though sunny, during most of the period covered by these 

 records. 



(5) TUat iu forty out of the forty-three cases, the female phase began 

 before the end of the male phase. 



(6 That in forty out of the forty-three inflorescences the female phase 

 ended at the time of, or earlier than, the male phase, thus making self-polli 

 nation not only possible but very highly probable. 



(7) That the end of the female phase was never more than three days 

 behind the end of the male phase, making self-pollination still possible in 

 dry weather, since coconut pollen-grains appear to maintain their viability 

 for several days under fairly dry conditions. 



(8) That overlapping of inflorescences only occurred twice out of forty- 

 three cases. 



The trees on which the observations were taken were approximately 

 eight years old and fairly widely planted, but not well cared. However, 

 shortly before the observations commenced a space round each tree was 

 cleared and cultivated. In support of the facts revealed in the above table, 

 three unopened inflorescences were bagged in muslin bags and it was found 

 that in each case, self-pollination was effected naturally and fruits, which 

 still remained on the trees, were formed. 



Again, three other inflorescences were em isculated immediately on 

 opening and it was found that, though the female flowers behaved normally, 

 no pollination took place and no fruits were formed, although male flowers 

 on adjacent trees were in full bloom. In a similar connection, it has been 

 observed that odd isolated coconut trees growing even under bad conditions 

 produce fruit so that self-pollination must have taken place. When coconut 

 (lowers are in full bloom, at about 10 a.m., when the dew has dried up and 

 when gentle breezes frequently begin, clouds of pollen can be seen lloating 

 away in the sunlight. In a very slight breeze these pollen clouds do not 

 travel far owing to the weight of the pollen, but it is highly probable that 

 with the strengthening of the breeze as the day advances the pollen clouds 

 are carried to a considerable distance and thus cross-polliuation is effected. 



