FORKING AND FASOJATLON. 



49 



subtending bracts ; but half-way clown the scape was 

 a petaloid elongated bract with a tiny blue petal (sole 

 vestige of a flower) in its axil. 



This last fact seems to show that the disappearance 

 of the lateral flowers is due, not to synanthy, i. e. their 

 fusion together to form a single compound apical flower, 

 but to the fact that the developmental energies and 

 nourishment have been concentrated at the apex of 

 the whole inflorescence, i. e. in the uppermost flower, 

 thereby preventing the proper, or any, development 

 of the lateral flowers and their bracts. 



If this abnormality is to be explained by supposing 

 that all the lateral flowers have fused with the terminal 

 one of the raceme, then the presence of the tiny blue 

 petal half-way down the scape must be accounted for 

 by supposing that part of the flower originally in that 

 position has been carried up and become fused with the 

 terminal one, and part of it has been left behind. But 

 this would be a far-fetched conception of a case without 

 known parallel. 



A very similar abnormality occurs in the foxglove 

 {Digitalis purpurea) . The large, terminal, peloric flower 

 is the equivalent, and is due to the "absorption" into it, 

 as it were, of several of the uppermost flowers of the 

 raceme (PI. XXXIV) ; as in the Scilla, rudimentary 

 flowers in the guise of single, tubular petals, occur in 

 the axils of some of the uppermost bracts. 



Hence neither of these two phenomena can be 

 ascribed to " synanthy " ; it seems best to regard the 

 large terminal flower as a single, not a compound, 

 structure. There is a tendency, in the upper part of 

 the raceme, towards the rapid and vigorous develop- 

 ment of a large terminal flower, and, in the foxglove, 

 the lateral flowers are imperfectly formed and con- 

 gested towards the top, while in the Scilla they are 

 not formed at all save for a single isolated petal, all 

 the nutriment being carried up to the apex of the 

 inflorescence. There is no real evidence in either case 

 of actual fusion of two or more distinct floivers. 



VOL. II. 4 



