I am glad you have drawn attention to the difference in the bud-state of the perfect and imperfect flowers; for I remember, many years ago, objecting to Asa Gray that he considered the imperfect flowers (not, I think, in the case of Impatiens) arrested buds, and I maintained that their structure had been specially modified for their functions.² From observations by myself in 1863, I find I was struck with the small size of the anthers, and the very small quantity of pollen. The grains are of the same diameter as in the perfect flowers, but they appeared to be more unequal in size. I distinctly saw pollen-grains protruding from the grains whilst within the anthers, and penetrating the stigma. I cannot believe that I could have overlooked the facts of the anthers not dehiscing. I do not mention in my notes that the pollen-grains are tied together by threads, as I do in the case of the pollen of the perfect flowers. I speak of the nectary in the cleistogenous flowers as a mere rudiment.³ From the fact of the nectary in the perfect flowers containing nectar, and from the pollen-grains being tied together by threads, I cannot doubt that they are crossed by insects, and I am almost certain that they are frequently visited by humble-bees. The structure of the flowers seems to me so well adapted for crossing, that I expected that the perfect flowers would be sterile without the aid of insects. In this I was quite wrong, as the perfect flowers, when protected, produced pods.⁴ Eleven such pods from perfect flowers, spontaneously self-fertilized, yielded on an average 3.45 seeds. I carefully brushed away the pollen from some of the perfect flowers, and fertilized them with pollen from a distinct plant, but got only three pods, containing, to my surprise, only 2, 2, and 1 seed. I attributed this poverty at the time to this plant probably requiring repeated doses of pollen, as is certainly sometimes the case.