NECTAR AND ITS USES. 325 



all after the first few possess the outer set, though it is not till the cot- 

 ton has been blooming about a month that the inner set appear. Con- 

 fining our observations to flowers which possess both sets fully devel- 

 oped, we find that a number of days before a flower-bud opens all of its 

 iuvolucral glands are visited more or less frequently by ants, and occa- 

 sionally a wasp or hive bee may be seen about them, although to our 

 eyes they are dry. Evidently, then, they secrete a thin sugary film. 

 The evening before such a bud opens, its visitors increase in number, 

 and we may, perhaps, see a little nectar in its glands. But during the 

 night preceding its unfolding, its cups fill out with the sweet fluid which 

 is collected by large numbers of ants, and early the next morning a 

 large drop may be seen suspended from the lower margin of each, or in 

 some cases running down the bract ; and throughout this, which may be 

 called the day of blooming, bees, wasps, and ants of many species may 

 be found in constant attendance on the glands. Though drawn so close 

 to the flowers, these insects never enter them, so they can have no direct 

 influence on their fertilization. Perhaps the strangest thing about these 

 glands is, that during the night, when this abundance of nectar is col- 

 lecting, they are visited by thousands of the moths of Aletia argillacea 

 and HeUothis armigera whenever these moths are flying and laying their 

 eggs. This appears to be a strange paradox. Xectar is secreted ap- 

 parently to attract insects to a plant ; and some of the insects so at- 

 tracted have the instinct to oviposit on the plant, on the foliage, flowers, 

 and fruit of which their larvae feed. How could this secretion have 

 been acquired by natural selection ? It looks as if such an acquisition 

 must imply the survival of the untittest ! As has been shown elsewhere, 

 the flowers of the cotton plant suffer from the attacks of the larvae of 

 both these moths ; but most of the eggs of both species are laid on other 

 parts of the plant than the flowers or floral appendages, consequently 

 a larva to reach the flower must ascend the peduncle, and run the gaunt- 

 let of ants, wasps, and bees found at its summit; though I never saw 

 one ascending when these insects were at their post, and therefore never 

 had an opportunity to see what would happen then. I found that when 

 these larvae are on the leaves of the plant they are sometimes attacked 

 and killed by the ants without any provocation. So it appears that the 

 secretion of these glands first attracts the worst enemies of the plant, 

 and then attracts their enemies, which afford it partial relief fr6m the 

 misfortune that it has brought on itself. 



An example of nectar secreted on the flower-stalk is found in the cow- 

 pea. At the summit of each peduncle are several small, crater- shaped, 

 circumvallate glands, which secrete until the fruit is well advanced 

 toward maturity, as well as during the flowering period. Occupying, as 

 they do, the very end of the peduncle, they are beyond the clustered 

 flowers and seed-vessels. In Alabama I found that they are much fre- 

 quented by ante of several species. Like the cotton plant, the cow-pea 

 is visited by the moths of both Aletia and HeliothiSj but only the latter 



