150 PEINOIPIiBS OP PLANT-TE&A-TOLOGy. 



greatly reduced in development, and more or less yellow 

 or greenish, as in a variety of tlie feverfew {Pi/re- 

 thravi FartheAviuvi var. diacoideum), and of the ox-eye 

 daisy {Ghri/santheuium Levcantliemum var. discoideum) ; 

 in this last genus we have an instance of a character 

 appearing as an abnormality in one species which in 

 another species, viz., G. floscitlosnm {Plagius virgiitus) 

 is a perfectly normal feature. There are also the 

 varieties discoidea of the chamomiles {Mafricaria 

 Chaynoinilla, M. inodora, aiid Anthemis riiontann), and 

 of the ragwort {Senccio Jacuhsea) ; this last represents 

 a condition which is normal in the common groundsel 

 {S. vulgaris). Robbins describes pelory of the 3-lobed 

 corolla of the ray-florets of Oaillardia aristata which 

 became tubular and 4-lobed ; this case thus exhibited 

 partial dialysis of the ligula : the fifth lobe still 

 remained unindividualized. 



General Gonchisiovs on Pelory. — The change from 

 a zygomorphic to an actinomorphic condition it is safe 

 to regard as a reversion, for the former is certainly 

 secondary and not primary. This conclusion applies, 

 however, to the mere change as such. But the par- 

 ticular form which the flower or some of its parts 

 assumes under the change cannot by any means in all, 

 or even most, cases be regnrded as partaking in the 

 reversion. Possibly the spurless form of perianth- 

 leaves in certain peloric flowers may be, and probably 

 it is, ancestral, for the spurred perianth-leaf is a 

 secondary structure. This will apply to other cases 

 of simplification of perianth-leaves. On the other 

 hand the acquisition of spurs by the four normally 

 spurless petals in Linariu and Viola must be regarded 

 as a progressive and not a reversionary phenomenon. 

 The form of the petals in vexillai'- and alar-pelory of 

 Leguminous flowers must Ije envisag-ed in the same 

 way, as also the " liemi-peloric " flowers of LabiatiB, 

 and the regular ray-florets of Compositfe. 



It may be concluded then, that in the peloric flowers 

 here treated of there is a most evident and interesting 



