HETEROSTYLISM. 207 



otlier two are practically inappreciable. On the other hand, 

 considering every difference under 5 as inappreciable, there 

 are four cases (§) of the short-styled in which it is consider- 

 able ; and of these it was only 3 in the case of P. Sinensis with 

 Hildebrand ; consequently one cannot confidently say that the 

 short-styled is more feminine than the long-styled — at least, 

 to any well-marked extent. 



With the corresponding column in the second table, one 

 notices nine cases whei'e the difference is great ; while in all 

 of those marked (f) it is inappreciable. Hence the con- 

 clusion is much more pronounced in favour of the greater 

 fertility of the long-styled forms when illegitimately crossed. 



Miiller accounts for " the greater productiveness of 

 illegitimate crossings in the case of the long-styled form 

 of Sottonia than in short-styled flowers, to the fact that the 

 former kind of illegitimate crossings occur frequently in 

 nature ; as these flowers are visited by pollen-seeking flies 

 which have no need to thriist their heads into the flower 

 of the short-styled form," which is, therefore, presumably 

 neglected.* 



The table I have here drawn up shows that the greater 

 fertility of the long-styled form when illegitimately fertilised, 

 is a general feature of hetei^ostyled plants, and not peculiar 

 to Hottonia palustris ; hence we must look to a more general 

 cause. 



As another hypothesis, it may perhaps be suggested that, 

 as the homomorphic condition of sJiort stamens with a short 

 style seems to have been the primitive form, then in the 



* If Miiller's idea be true, Hottonia furnishes another instance of 

 the disadvantage of great differentiations, and is only one degree better 

 ofE than the Cowshp. In either case, one is inclined to ask what has 

 become of the proper insects (whatever they may be) required for the 

 perfect intercrossing of these flowers. 



