162 



PRINCIPLES 



OP 



PLANT-TERATOLOGY. 



same, and Wirtgen observed very interesting cases in 

 abnormal plants of G. arvensis in which, they appeared 

 in the position which normally flowers occupy in the 

 axils of bracts at the base of the inflorescence-axis, 

 and occasionally replacing flowers in the smaller rami- 

 fications of the inflorescence ; in the same positions on 

 different plants he found normal flowers, more or less 

 abortive flowers, and bulbils. In one case he observed 

 a bulbil bearing terminally a stalked flower at the 

 base of the inflorescence-axis. Thirdly, the fact that 

 the bulbils of L. tigrinum var. Fortunei became, without 

 any unusual or apparent stimulus, metamorphosed 

 directly into flowers (the comparatively late stage at 

 which this took place alone prevented the lowermost 

 bulbil-scales becoming changed into perianth -leaves), 

 is quite incompatible with the view that they represent 

 vegetative shoots, and before the true nature of bulbils 

 was realized these abnormal ones presented an alto- 

 gether anomalous phenomenon. 



It follows, from the above considerations, and from 

 the fact of the gradual transformation (as shown by 

 the transitional structures above-mentioned) of the 

 bulbil of L. tigrinum var. Fortunei into a flower, that 

 the bulbil as found in the Liliacese represents a modi- 

 fied flower, specially changed for purposes of vegeta- 

 tive propagation. And inasmuch as onty isolated 

 members of the order produce bulbils, this modification 

 must be a comparatively recent one. The fact that 

 bulbils, on being placed in very damp situations, as 

 on the soil, grow out into vegetative shoots, is no 

 argument against this conclusion, for under unusual 

 stimuli of moisture and nourishment even typical 

 flowers will often proliferate into vegetative shoots. 



That the bulbil of the Liliacese is a modified flower 

 would seem to be a new discovery, for the slightest 

 hint of such an idea cannot be found in any of the 

 text-books on morphology. Hesselman, who has 

 described the same abnormality as the above in L. 

 bidbiferum, considered the bulbil to be a vegetative 



