76 



PRINCIPLES OF PLANT-TERATOLOGY. 



and petaloid enations almost equalling the petals in 

 size.* The tubular enations are comparable to the 

 similar objects described by Celakovsky in Narcissus.- 



It is possible that this explanation may apply, not 

 only to the Primula, but to all doubling where this is 

 due to serial dedoublement, for it is obvious that each 

 well-developed petal-like enation may itself repeat the 

 phenomenon by which it arose, and so on, until a large 

 number of such structures would be formed. Masters 

 figures in his paper such a series of enations. 



C. de Candolle observed a unique case in the edel- 

 weiss {Leontojpodium aljpinum (5 nivale) ; in some female 

 and hermaphrodite florets there was an extra inner set 

 of pappus-hairs united mostly in bundles, and attached 

 along the sutures (/. e. alternating with the petals) of 

 the corolla for varying heights, some adnate to its 

 whole length (fig. 91) ; in the hermaphrodite florets 

 the extra pappus-hairs had the character, not of the 



* An explanation of the dorsal enations of Primula, etc., along the lines 

 of extrorse anther-structure, is given elsewhere. 



Fig. 91. — Gnaphalium Leontopodiimi (Edelweiss). Female flower 

 showing extra outer corolla (co') in form of a pappus. (After C. 

 de Candolle.) 



