OF THE SEPALS IX SEKAPIAS. 491 



of the lateral sepals remained of tlie ordinary sepal structure. One 

 of these flowers is represented of the natural size at fig. 1, and its 

 parts magnified below ; and in this one the column was perfect, 

 while in the other four flowers it was reduced to a mere rudi- 

 mentary process above the callus or guiding-plates, neither anther 

 nor stigmatic surface being traceable. In all the ovary was less 

 than half its proper length ; but otherwise the remaining parts 

 of each flower were normal, though rather small, except that the 

 sepals and petals Avere free, not forming a hood. 



During the past spring (April 20-23, 1870) I have again 

 come across this curious form of monstrosity at Mentone, but 

 this time in Serapias cordigeray L. I examined no less than six 

 specimens of S. cordigera, L,, all of which faithfully reproduced 

 the kind of modification described and figured above in S. lingua^ L., 

 the column being perfect in all; while in one other specimen, 

 having six expanded flow^ers, one flower corresponded wuth that 

 figured in its abnormal details, one was perfectly formed, and 

 the four remaining flowers exhibited different stages of the par- 

 tial conversion of the lateral petals. It is instructive to note 

 that in all the complete specimens it was the same part that was 

 affected, and that both the kind and amount of change was the 

 same in all : thus, both in S. lingua, L., and >S'. cordigera^ L., it 

 was always the anterior * half of the lateral sepals that became 

 labelliform ; and so exactly was the line of demarcation observed 

 between the changed and unchanged halves of these sepals, that 

 in every case only one half of the callus or guiding-plate was deve- 

 loped (figs. 1 h and 1 c, Plate III.). In all these abnormal flowers, 

 also, the ovary was reduced to less than one half its ordinary 

 length ; and in four out of the five flowers on the spike of S. 



im 



The exact reproduction of these semilabelliform sepals in eight 

 distinct plants belonging to two species of Serapias surely indi- 

 cates that this is no mere passing change brought upon the 

 plant by the action of some temporary condition, but rather a 

 deep-seated tendency forming a part of the constitution of these 

 species, or perhaps of the genus, a tendency which may be 

 always present in each individual, though usually in a latent 



condition. 



In order to realize the extent and the detailed nature of this 



* Throughout I refer to the position of tlic parts of the flower as found in 

 expanded blooms. 



