104 M. E. Regel on Parthenogenesis. 



experimental plants. Even with the most careful observation, 

 an absolutely conclusive result could scarcely be obtained with 

 this plant ; for the male flowers are only detected after they have 

 opened, and therefore may have scattered pollen. I used my 

 utmost endeavours to suppress the male flowers at the right 

 time ; and in fact hitherto neither of the experimental plants 

 have set fruit, all the earliest developed female flowers having 

 withered up. But if these plants should still set fruit, this must 

 be attributed to pollen received from some of the male flowers. 



Spinacia. — Difficult as it is in Mercurialis to neutralize the 

 influence of pollen from adventitiously developed male flowers, it 

 is still more difficult with Spinacia. All the experimental plants 

 were cut in. I observed at first, in the axillary tufts of female 

 flowers, solitary normally developed anthers, which projected 

 over the female flowers. I removed them, and placed the plants 

 on which I had noticed them in a diff'erent locality. All my 

 experimental plants appeared inclined to set seeds. I therefore 

 placed all except one, on which the first flowers were beginning 

 to unfold, in another situation, and continued the examination 

 of this plant with redoubled attention, allowing in all only ten 

 axillary tufts of blossom to come to perfection. All newly- 

 produced lateral branches were necessarily broken off", as these 

 at once developed new blossoms. First of all, I observed on this 

 plant two stamens with anthers containing abundance of pollen. 

 Placed under the microscope, this exactly resembled normal 

 pollen. These stamens, however, did not arise (as I observed in 

 Chamarops last year) from female flowers ; but among the female 

 flowers were scattered solitary stunted male flowers, which 

 brought only one stamen, seldom more, to perfection. This 

 fixed my attention. With the help of the lens, I soon saw, in 

 the tufts of female flowers, isolated gland-like bodies, which I 

 had taken at first for mis-shapen bracts. When I had dissected 

 them out, I found that they were sessile anthers, developed in 

 scattered abortive male flowers. These contained perfect pollen, 

 as the above-mentioned gentlemen as well as myself can testify. 

 These anthers are seldom perfectly seen, but are almost always 

 partly covered up by the involucral scales of the flowers in which 

 they arise, so that they may be easily overlooked or be taken for 

 transformed bracts. In the isolated male flowers I usually found 

 one sessile perfect anther, with several abortive; more rarely 

 several perfectly developed anthers filled with pollen (all, how- 

 ever, sessile) exist in one flower. From one single axillary in- 

 florescence I dissected out ten such male flowers with sessile 

 perfect anthers. But as this had to be done on living plants 

 under a lens, it could seldom be effected without injuring the 

 anthers, by which pollen was always scattered. In such cases. 



