18 WILLIS : EVIDENCE AGAINST THE ORIGIN OF 



small trees. Not only so, but, as already pointed out in the 

 previous paper, the two species of Dillenia grow side by side, 

 and are equally common. 



The Dipterocarpaceae, so common in Ceylon, afford a still 

 better case. Dipterocarpus itself is divided into subgenera 

 on characters of the fruiting calyx, and can it for one moment 

 be imagined that there is any struggle for existence between 

 trees 40 years old, in which it matters whether the calyx is 

 smooth or 5-ribbed. These trees are very much alike in their 

 vegetative characters, and their flowers do not appear for a 

 very long time, by which time the struggle for existence 

 between members of the same species must be long over. 



Innumerable examples of this kind might be quoted from the 

 flora of Ceylon or any tropical country, but these will suffice 

 to indicate the point in view. If the characters of the species 

 were characters of the seeds , one might perhaps say that these 

 had such an effect upon the seedlings that they enabled them 

 to survive, but in most cases the characters are not so, but of 

 petals or sepals, of size of flower, and what not, that cannot 

 be supposed to have any effect upon the capability of the seed 

 to germinate or to preserve its life ; in most cases they cannot 

 even have an effect in causing more seed to be set. 



Further, in most cases allied species live near together, and 

 why should this be the case, if infinitesimal variation is the 

 rule ? Why, if one species with a stalk to the cai"pel of 4 

 inches long is living in a given place, should it break up into 

 two with stalks 1 inch and 7 inches long, from the point of 

 view of advantage to the species ? This fact, that nearly 

 allied species usually live very closely together, is a strong 

 general argument against advantage having anything to do 

 with the origin of species. 



Another strong argument against infinitesimal variations is 

 that for selection among them to produce any great effect, it 

 should be between a large number. But in actual fact any 

 given plant can only have, on the average, six like itself 

 around it, and the competition can therefore only be between 



