80 PHILOSOPHICAL NOTES ON 



The latter are the marked segments. I have often observed, and 

 others have done so also, that on the same spike some of the 

 flo-^vers have one labellum, and others two labella* 



No amount of twisting will account for this curious feature, 

 for a twist would bring all the three marked petals uppermost^ 

 as we have them in the Hippeastrum. This never does occur ; 

 moreover the stamens and style of the Gladiolus are never twisted, 

 but are always arched under the upper one or two petals, as the 

 case may be. 



Transposition of whorls is what I wish to prove, and therefore 

 I shall not discuss how the three loioer segments of Gladiolus 

 became the marked ones. Suffice it to note that, among the fine 

 lot of Griadioli exhibited by Messrs. Kelway and Son at the 

 meeting of the Royal Horticultural Society on the 25th August 

 1891, there were spikes which had an occasional regular or peloric 

 flower, and one new variety more prominent than the restf had all 

 three inner segments marked. If in this regular Gladiolus the 

 inner whorl were to become outer by transposition, we should have 

 an Iris much like Rohinsoniana. 



Then how is it that in these hydrids two varieties of flower 

 occur in the same spike ? The solution of the puzzle is in the 

 transpositio?i of the outer to the inner whorl, and in no other 

 way. The segments of the two whorls alternate, and it is plain 

 that by transposition sometimes one and sometimes two segments 

 will be the inner lower ones. 



If the reader will consider that all the six segments of the 

 Glacholus are cellular nipples, when the flower is in its first for- 

 mation in the bud, he will perceive how readily this transposition 

 may occur by a mere infinitesimal displacement of the nipples. 

 The reader might say, if this were possible it would surely some- 

 times occur that not all the three segments belonging to a whorl 

 would be displaced at the same time. True, and this is exactly 

 what sometimes occurs, for I have before me a Gladiolus flower 

 with one of its segments neither in nor out, that is, one half of it 

 belongs to the outer whorl and one half to the inner whorl ! It is 



* I call them labella for brevity sake ; in the one case there is one lower 

 and two upper inner segments ; in the other there are two lower and one 

 upper segment in the inner whorl. 



t Called " Thales." 



