116 Prof. Keeble, Dr. Armstrong, and Mr. Jones. 



The waning of the yellow plastid pigment coincides with the disappearance 

 from the petals of an inhibitor of oxydase. Thus, if petals of various stages 

 are tested by means of benzidine and hydrogen peroxide, only the mature 

 cream-white petals, from which the yellow pigment has disappeared, give 

 an oxydase reaction. If, however, the petals are first treated for five minutes 

 with absolute alcohol, and then with benzidine and hydrogen peroxide, they 

 give a good oxydase reaction. 



Inasmuch as we have shown already (Part IV) that absolute alcohol serves 

 to remove the inhibitor of oxydase which occurs in the flowers of dominant 

 white Primula sinensis, we conclude that there exists in this " primrose " 

 strain of wallflower an oxydase inhibitor which is closely bound up with the 

 yellow plastid pigment. As the latter is destroyed so is the former, and 

 after their disappearance the oxydase contained in the cells in which they 

 were present may be demonstrated by means of the ordinary oxydase 

 reagents. 



These observations, together with the fact that red and purple anthocyan 

 only make their appearance in cells in which the plastids are degenerating, 

 may, perhaps, offer a clue to the significance of the curious phenomenon of 

 striping which is so common in wallflowers. In striped wallflowers the red 

 or brown colour is broken by wider or narrower yellow bands. We believe 

 that the phenomenon is to be interpreted in terms, first of the persistence 

 of the plastids and of the inhibitor of oxydase, and, second, of failure of 

 supply of chromogen. 



That the former of these factors is concerned in the phenomenon is 

 probable from the foregoing observations, and that lack of chromogen is also 

 concerned is rendered probable by observations which we have made on the 

 striped flowers of Honesty {Lunaria annua). 



The garden varieties of this plant bear flowers showing all degrees of 

 striping. In some the petals are. self (uniformly) coloured ; in others there 

 is a small colourless area at the junction of limb and claw, and in others 

 this white area may extend so far as to leave but a narrow border of magenta 

 at the edge of the petal. The white regions of the flower are well supplied 

 with peroxydase, and therefore their lack of colour -is to be ascribed to lack 

 of chromogen. That this is actually the case may be demonstrated by 

 removing all the open flowers and some of the buds from a head of flowers. 

 As a result of thus reducing the number of flowers of the inflorescence, those 

 that are left assume in what should be the white region of the flowers a 

 dark magenta colour, so dark as to look like a deep magenta splash on a pale 

 ground. Into the physiological interpretation of this effect of the operation 

 we need not enter now, but we may take it as demonstrating that streaking 



