NATURAL SCIENCES OF PHILADELPHIA. 23? 



And now in regard to the bearing of all these facts on the great 

 scientific questions of the day, we have to note first, that the 

 plant is an introduced weed, with nothing allied to it anywhere, 

 in the localities where we usually find it, with which it can 

 possibly hybridize. The variations must be from some natural 

 law of evolution inherent in the plant itself. Varieties of course 

 may cross-fertilize as well as species ; and some of these varia- 

 tions may be owing to one form fertilizing another form ; but 

 there can be no avoiding the fact, that at least the first pair of 

 varying forms must have originated by simple evolution. 



Now going back to our florists' experience the question occurs, 

 that as varieties once evolved will reproduce themselves from 

 seed, why does not some one of these Linarias, which has been 

 struck off into some distinct mould, reproduce itself from seed, 

 and establish, in, a state of nature, a new race, as it would do under 

 the florist's care ? Why, for instance, is there not a spurless race? 

 It is scarcely probable that the solitary plant, found on this after- 

 noon's walk, is the only one ever produced. Dr. Darrach's 

 recollection shows it is not a solitary case. The bumblebee 

 furnishes the answer. They, so far as I have been able to see, 

 are the only insects which visit these flowers. They seem very 

 fond of them, and enter regularly at the mouth, and stretch down 

 deep into the spur for the sweets gathered there. The pollen 

 is collected on the thorax, and of course is carried to the next 

 flower. The florist, to " fix" the form, carefully isolates the plant . 

 but in the wild state a spurless form has no chance. The bee 

 from the neighboring flower of course fertilizing it with the 

 pollen from any of the other forms. 



If there were no bees, no agenc}^ whatever for cross fertilization, 

 nothing but the plant's own pollen to depend on, there would 

 undoubtedly be races of this linaria, which, again, by natural 

 evolution at times changing, would produce other races ; and in 

 time the difference might be as great as to be even though 

 generic. But we see that by the agency of the bumblebee the 

 progress of the newly evolved form is checked. The pollen of 

 the original form is again introduced to the offspring, and it is 

 brought back at least half a degree to its starting point. 



The conclusion seems to me inevitable, that insects in their 

 fertilizing agencies, are not always abettors, but rather at times 

 conservators of advancing evolution. 



