370 



The Journal of Heredity 



seems to be no single underlying princi- 

 ple governing their appearance. Their 

 secretions also vary widely, although 

 in general they give oft" merely a kind of 

 sugar-water, containing varying propor- 

 tions of dextrose, levulose and sucrose. 



When, only a few years ago, natural- 

 ists generally began to realize that there 

 might conceivably be parts of an organ- 

 ism which had no particular use, and 

 that the appearance of such parts 

 cotdd be perfectly explained in various 

 ways, without resorting to the principle 

 of natural selection, and that their 

 useless existence, further, would not 

 cause the whole structure of organic 

 evolution to totter on its foundation, 

 the extra-floral nectaries began to be 

 examined in a more critical way. 



Numerous plants with extra-floral 

 nectaries have been deprived of these 

 organs, but it has been found that they 

 produce the normal amount of seed 

 just the same, and that none of their 

 activities appears to be altered. 



Further, it has appeared that plants 

 which attract ants by extra-floral nec- 

 taries .sometimes have their flowers 

 more robbed by ants, than would 

 l)robably be the case if they lacked 

 extra-floral nectaries. The idea that 

 the presence of such nectaries is a " sop 

 to divert the attention of ants from the 

 greater treasures in the flowers, there- 

 fore, appears to have little basis in fact, 

 as far as observation goes. 



Again, in some of the broad beans 

 (Vicia), it has been found that bees 

 \'isit the extra-floral nectaries, in prefer- 

 ence to those in the flower. In such 

 cases, the flower may fail altogether to 

 be cross-pollinated; evidently, then, the 

 presence of extra-floral nectaries is in 

 these cases a distinct disadvantage 

 rather than an advantage to the plant. 



Finally, the theory that nectaries 

 outside the flower are intended to 

 jjrotect those inside the flower from 

 unwelcome visitors, would seem to 

 d.-mand that the two sets of nectaries 

 be functioning at the .same time. But 

 in point of fact, it has been observed 

 that there is rarely exact correspondence 

 in tim", and that in some cases the chief 

 secretion of extra-floral nectar occurs 

 l)efore, in others after, the |)eri()d when 



NECTARIKS ON LKAF STEM 



I'lioto-microj^rapli of cherry ])cliolc, 

 showinj^ two lyi)iral (.'xlra-floral 

 nectaries. (Fi^- I''-' 



