1890.] NATURAL SCIENCKS OF PHILADELPHIA. 275 



In all cases where the branches of the previous year had matured 

 in partial shade, the flowers on the spike proceeding therefrom were 

 wholly male. This is in accord with the views heretofore developed 

 by me that conditions unfavorable to perfect nutrition are unfavor- 

 able to the development of female flowers. In the branches that 

 had been developed under conditions favorable to nutrition, some 

 would have spikes less vigorous than others, ])ut so far as I could 

 see, tliis had no eflect on the production of the sexes. As many per- 

 fect flowers would be found on a comparatively weak as on a strong- 

 spike. 



What I have placed on record in relation to growth -rhythms 

 in Acer dasycarpum, Acer ruhruvi and some other plants, oc- 

 curs in these flower buds also. At an early stage in the bud, the 

 anthers are fully formed and thegyna?cium reaches a certain stage, 

 when a resting time occurs. When growth is renewed the ovary 

 with its pistil proceeds to move forward rapidly, while the filaments 

 proceed slowly ; or the latter move rapidly and the gynoicium 

 continues wholly at rest. In the former case we have the perfect, 

 in the latter the male only or barren one. 



It is evident that up to the time when this last growth-rhythm 

 begins, the flower may be male or female (hermaphrodite in this 

 case.) It is male or female simpl}' as the renewed growth-force is 

 directed towards the gyna?cium, or to the stamens alone. 



The development of the stamens presents points of interest. In 

 the hermaphrodite flower, the point of the stigma and points of the 

 anthers appear simultaneously bursting through the petals, then rest 

 for some twenty-four hours. The growth of these parts is then very 

 rajjid, the stamens and pistils extending to nearly an inch beyond the 

 calyx. This renewed growth commences soon after sun-rise, and 

 ceases with sun-down. The anthers do not burst their cells on the 

 day this growth is in progress. The following day, however, one 

 anther sheds its pollen, the next day another, the third day another 

 and on the fourth day all the remaining burst nearly simultaneously. 



It has become the habit in observations of this kind, to suspect 

 that the whole behavior has some relation to the visits of insects, and 

 to cross-fertilization. But though endeavoring to trace such a rela- 

 tion here, it could not be ascertained. If the division of the sexes 

 is with a view to economy in energy, the enormous disproportion of 

 male to female flowers strikes the thought unfavorably. If we watch 

 the flowers closelv we find bees and other insects visitinsr them 



