256 Mr. J. Lowe on Salix Andersoniana. 



union of the styles ; a pollen-mass still remains on the side of 

 the united ovaries. 



Fig. 8. A complete and well-formed ovary. 



I ought to remark here that these figures are not intended to 

 represent the progressive development of the ovary as shown by 

 any individual floret, but simply the different stages which may 

 be traced in a number of florets and which may be reasonably 

 regarded as successive. 



We may now consider the cause of these pheenomena and the 

 laws which govern formations of this nature. 



The generally received opinion regarding the production of 

 dioecious flowers is that each flower is rendered unisexual by the 

 suppression of the other sexual whorl, and though this may be 

 mainly true of many dioecious plants, it does not appear to ex- 

 press the whole truth with respect to dioecious Amentiferee. 



Dr. Braun in his ' Rejuvenescence of Nature/ states that both 

 kinds of sexual organs are derived from the same leaf, or as he 

 expresses it, ^' the same leaves appear in the male as stamens and 

 in the female as carpels." In other words, the leaf which fails 

 to produce a male will give rise to a female organ, and vice versa. 

 Hence, though it is perfectly correct to say, that there is an 

 arrest of development when stamens are alone produced, it is 

 otherwise with respect to female organs, since there is here not 

 an arrest but an exaltation of development. 



These specimens illustrate also the parts of the leaf which 

 give origin to the different parts of the essential organs ; thus, the 

 anther gradually merging into the carpel shows that it is derived 

 from the lamina of the leaf (the fact of the carpel being formed 

 by the lamina of the leaf being ascertained by morphological 

 changes in other plants). The pollen observed on the edge of 

 the ovary in fig. 5, would encourage the idea that pollen is 

 merely a gemmiferous condition of the lamina of the leaf. And, 

 lastly, we may allude to the gland, which, although not pre- 

 senting any peculiarity in the present specimens, I have never- 

 theless found in others assuming a very interesting form. I am 

 not aware that the question has been mooted as to what is its 

 real morphological value. It might be assumed to represent an 

 abortive stamen, but that we find it present in those Salices 

 which may be regarded as having their staminal whorl complete, 

 Salix pentandra for instance. Moreover it is found to be placed 

 opposite the interval in flowers which have only two stamens, 

 thus having an alternate arrangement. I have little doubt, 

 especially since meeting with the specimen shown in fig. 9, where 

 there are two glands alternating with the stamens, that they re- 

 present the corolla. Regarding the scale as the calyx, we have 

 thus the various whorls of the flower complete. 



