96 POLLINATION AND FERTILIZATION 



where self-pollination would otherwise commonly result, and it may 

 be assumed to have been the first step toward the uni-sexual state, so 

 common among flowers of that class. Careful notice should be taken 

 of the fact that in dichogamy the retarded state observed in androecium 

 or gynaecium is but temporary, and that the finally develc^ed form is 

 the same, whether the flower be proterandrous or proterogynous. 



Dimorphism. — A far more profound modification is that in which there 

 is a permanent change in the androecium (Fig. 273) of one flower and 

 a similar change in the gynaecium (Fig. 274) of another, by which a 

 similar result is obtained to that proceeding from dichogamy. Such 

 a provision constitutes Dimorphism. By a modification of it, a third 

 form of flower, intermediate between the other two, is produced, con- 

 stituting Trimorphism. The explanation of the case of dimorphism 

 here exhibited is as follows : An insect visiting flower No. 1 and thrusting 



his proboscis deeply into the corolla- 

 tube in search of nectar, brings his 

 body into contact with the stamens, 

 and pollen is deposited upon it. The 

 next flower visited may be one like 

 2/^4^ No. 2, having a long style. The por- 

 tion of the body which is now covered 

 with pollen will then be brought into 



Fig. 273. Long-styled form of dimorphous COUtact with the StigUia, UpOH wllich 

 aoy^er of Houstonia. 274. The same, long- ^j^^ jj^^^ j^ dcpOsitcd. At the SamC 



staminate form. ^ 



time a different part of the body is 

 being laden with pollen from the short stamens of flower No. 2, to be 

 deposited upon the short pistil of still another flower, similar to No. 1. 

 If perchance two flowers of the same form are visited in succession, the 

 result is that an additional deposit of pollen is secured, or at most a 

 portion of the pollen already being carried is left upon the stamens of 

 the visited flower. 



As will be seen by a consideration of typical examples of each, dimor- 

 phism is more intimately connected with the transferring of the pollen 

 than is dichogamy, though the latter is rarely without some special 

 provision for thus supplementing the effect which it produces in 

 excluding the pollen from the stigma of its own flower. 



In conclusion, it may be said that even if, by some failure in the 

 provision here described, the flower should become self-pollinated, we 

 have excellent reasons for believing that pollen from a different flower 

 which might be deposited at the same time would find an advantage 



