PKOLIFE RATION. 21 



Proliferation in the axils of petals is illustrated by 

 the following cases. 



A pink [Dianthus plumarius) has been seen with 

 flowers in this position. Masters figures a flower of 

 Dianthus sp. with small inflorescences occurring in a 

 similar position ; they were formed at the expense of 

 the androecium which was completely suppressed. In 

 double flowers of Spirsea prnnifoUa rudimentary 

 flowers occur in the axils of the petals. 



Masters describes a cucumber in the male flowers 

 of which long shoots occurred axillary to the petals 

 which produced stalked secondary flowers. 



Turpin figures a flower of Gelastrus in which every 

 petal has a flower in its axil (PI. XXIX, fig. 5). 



Proliferations axillary to 'Stamens are seen in the 

 following. 



Masters figures a long-stalked flower arising from 

 the axil of an ordinary stamen in Nymphsea Lotus. In 

 Begonia Pearcei inflorescences were observed in the 

 axils of several of the stamens of a flower ; in this case 

 the phenomenon is less surprising inasmuch as the 

 stamens were semi-foliaceous. 



As an instance of flowers produced in the axils of 

 carpels there may be mentioned the case of a wild 

 mignonette {Reseda lutea) of which Celakovsky gives 

 an interesting figure. 



Axillary vegetative proliferation will now be con- 

 sidered, and firstly, in relation to bracts. 



Abnormal axillary branching is very frequent in the 

 Coniferge. Parlatore illustrates a case in Finns Le- 

 tnoiiiana in which, in place of the ovuliferous scales, 

 2-needled short-shoots were situated in the axils of 

 the " bracts." This seems to represent the extreme 

 vegetative condition of an axillary production of rudi- 

 mentary shoots, which has given rise to a wide discus- 

 sion as to the nature and origin of the ovuliferous 

 scale in certaiji groups of the Coniferae and has led to 

 an interpretation thereof which hitherto has had the 

 greatest number of adherents. 



