METAMORPHOSIS. 



127 



former case into two to eight, in the latter into two to 

 five narrow green leaves (PI. XL, figs. 2 and 8), a very 

 interesting reversion to the ancestral condition . It was 

 due to the capitula being infested by a mite (Phytoptus). 



Magnus describes flowers of Anemone ranunculoides 

 attacked by the Uredineous fungus (Ecidiam punctatum 

 in which the perianth-leaves had become foliaceous or 

 virescent. 



In the primitive order Ranunculacese some of the 

 genera normally possess petal oid calyces, e. g. Helle- 

 borus sp., Anemone. Velenovsky figures an abnormal 

 flower of the wood-anemone (A. nemorosa) with three 

 inner coloured leaves (best regarded as petals) and 

 three outermost, perfectly green, and smaller sepals. 

 In another flower figured by him there is an inner 

 whorl of six narrower and an outer of six broader leaves, 

 but all are similarly coloured. Here we get a distinct 

 differentiation, in the first case greater, in the second 

 smaller, into two divisions of the perianth ; in the first- 

 mentioned instance the three outer green leaves may 

 be compared to the trifoliate green calyx of the lesser 

 celandine (Ranunculus Ficaria). Such a case throws 

 a distinct light upon the origin of the calyx in flowers 

 generally. Here we have, in an abnormal example, a 

 green calyx which has been derived directly from a 

 petaloid one. 



Prantl, Goebel, Velenovsky, and others hold that 

 the calyx is phylogenetically bracteal in origin, a view 

 derived probably from the fact that the calyx is usually 

 green and leaf -like, and often differs from the corolla 

 in displaying a spiral arrangement of its sepals. 

 Velenovsky brings forward the examples of Camellia, 

 Gryptandra, etc., to show that the sepals here belong- 

 to the same spiral series as the numerous bracts below, 

 them and therefore must have been derived from them. 

 This phenomenon is also very obvious in the Japanese 

 allspice (Galycanthus). 



The view of Celakovsky, however, here supported, 

 is that the calyx is in all cases a derivative, like the 



