1886.] NATURAL SCIENCES OF PHILADELPHIA. 293 



will fall dow from sheer maturity, but thinks insects in the search 

 for honey may anticipate this natural development. The Rev. 

 Geo. Henslow, in the ix vol. (1868) of the Journal of the Lin- 

 nsean Society^ pfige 357, commenting on these statements, suggests 

 that though self-impregnation ma}^ be possible, the jerkings of the 

 stamens when the carina falls away, causes the pollen to scatter 

 over the insects, which then carry the pollen to other flowers. In 

 the X vol. (1869), p. 468, he notes that when the carina is liberated 

 it falls back with a jerk, as noted by Don, and that the stamens 

 fly up and closely press the vexillum. But no opportunity oc- 

 curred to him of observing what insects effected this process 

 which he accomplished artificially. In Mr. Meehan's grounds the 

 East Indian Indigofera Dosiia is a hardy shrub, and for the past 

 two years, he said, he had watched the behavior of the flowers, 

 and the insects that visited them. The carina is arched and 

 hooded, and extends so far over the apex of the stigma, against 

 which it presses, that both the stigma and column of stamens are 

 borne down to a right angle with the vexillum. The effort of the 

 stamens and pistil to rise, and of the carina to recurve, keeps 

 these portions in an exactly horizontal position. The anthers 

 burst while thus enclosed in the carina, but the pollen does not 

 escape, nor does it reach the stigma, for the pistil extends beyond 

 the anthers, and its apex is pressed strongly against tlie carina, 

 and no pollen can possibly reach the apex. The two wings of 

 the flower are caught in the subulate spurs, and are borne down 

 to the horizontal plane with the carina, and together form a level 

 platform on which insects in search of honey alight. Humble- 

 bees, hone3''-bees, and numerous species of sand wasps visit the 

 flowers, but in no instance was an insect's visit found to be effectual 

 in liberating the stamens and pistil from the gi'asp of the carina. 

 This was only accomplished by slitting the upper portion of the 

 carina with a penknife. Then the divided carina would instan- 

 taneously fall back, and the stamens and pistil jerk upwards, 

 scattering a little cloud of pollen in every direction. In some 

 cases pollen so scattered would light on a stigma, but in many 

 cases the pollen would be so completely projected that none 

 could be ti'aced to the stigma, and these probably received 

 pollen from the upper flowers where they in turn projected their 

 pollen. Possibly many were by this time too mature to profit by 

 the pollen they received, at any rate only a very small pro- 

 portion of the flowers matured seed vessels. In the older 

 flowers, the carina evidently became separated from the stamens 

 and pistil only when withering away. 



Mr. Meehan remai'ked in conclusion, that, as noted by authors 

 quoted, so far as his observations in this part of the world 

 (Philadelphia) with this single species goes, the behavior of the 

 flowers were neither favorable to self-fertilization nor cross- 

 fertilization. If insects had pollen scattered on them, there is 

 no way by which it could be communicated to the pistils of 



