1894.] NATURAL. SCIENCES OF PHILADELPHIA. 55 



They bring pollen from other Howers, and intermediate forms neces- 

 sarily result. Quite early in the modern discussions of this subject' 

 I showed that variation must first occur, and that the insect visitor was 

 rather a conservative agent than a factor in variation, — bringing back 

 towards the original that which had departed. I have published 

 many papers showing that variation is at least as great in monotypic 

 species as where there are numerous closely related species or varie- 

 ties to cross with. Iwpatieiis fiilra is another illustration. It is not 

 necessary to recount the character of the variations. One may 

 fairly say that no one character can be named that does not show 

 variation in some individuals. Even the glaucous leaves are some- 

 times bright green, at others almost silvery. The flowers not only 

 vary in color and markings, but the several parts of the flowers are 

 changeable. The spur particularly is sometimes of remarkable 

 lengtli, at others well warranting the remark of Dr. Gray "spur 

 rarely wanting." Aside from the fact that there is no other species 

 near to get any aid to variation in that way, many of the 

 colonies on my grounds are from last years' seeds. The obser- 

 vations on tliis plant confirm records I have made during the past 

 quarter of a century that there is an innate power to vary co-existant 

 with the species itself, independent of any conditions of environment. 

 This may be granted without prejudice to the proposition tliat 

 changes can and do occur at times by tiie influence of environment, 

 for which there is abundant evidence. It seems proper to present 

 the strong facts on the former side, because of the modern tendency 

 to exalt the latter as the prime motor in evolution. 



The amount of nectar secreted at the base of the spur is consider- 

 able. By cutting off the apex and stripping down the tube as in 

 milking, a globule as large as a pin's head will form at the cut. A 

 large proportion of flowers, especially in the latter part of the season, 

 are cleistogeue. But even in these cleistogamous flowers spurs are 

 formed. An interesting fact Avas that the nectar formed as 

 freely in the spurs of the cleistogene flowers as in those which we 

 should call normal and fully "adapted" to insect visitors! 



Is there any necessary relation between the nectariferous 

 structures of flowers and the visits of insects? The cases in which 

 the relationship appears close are numerous, but equally numerous 



' On the Agency of Insects in Obstructing Evolution. Proc. Ac. Nat. Sciences. 

 1872, pp. 235-237. 



