216 ANTHOCYANINS AND GENETICS 



of a colourless isomer of the pigment. If, however, a neutral salt, 

 such as sodium nitrate or chloride, is added to the solution of the blue 

 pigment, the latter remains stable, and the isomerisation does not 

 take place, or at any rate takes place in a lesser degree. The neutral 

 purple pigment, too, isomerises in solution ; also the red acid salt, 

 unless the acidity of the solution is maintained. The colour returns 

 again, on acidification, to a colourless solution containing the isomer. 

 It has been suggested recently by Schiemann (262) that in these 

 reactions lies the explanation of the "dominant white' phenomenon, 

 and one may add also the phenomenon of pigmented plants with 

 white flowers. The pigment is supposed to be present, though 

 inhibited, in these varieties. Applying Willstatter's results we may 

 consider the pigment to be developed in the flower as the isomer, and 

 that lack of salts in solution in the cell-sap, or diminution in acidity 

 of the cell-sap, prevents the pigment from existing in its coloured 

 state. This may possibly be the cause also of the lack of colour in 

 tinged forms, and increase in concentration of the neutral salts in the 

 cell-sap may increase the intensity of the colour. But it is difficult 

 to conceive of any factor, or process, which would lead to this result. 

 It has been observed that treating tissues with concentrated acid may 

 bring about development of red colour. There is no real evidence in 

 this case that the reddening is due to anthocyanin ; yet the fact is signi- 

 ficant, and it may indicate that the isomer is present in the tissues, 

 and that acidification will cause isomerisation to the coloured form. 



The bearing of the chemical interpretation of colour factors on the 

 phenomenon of striping is worth a brief consideration. The curious 

 limitation of the condition of striping to certain factors is well exem- 

 plified in Antirrhinum. In this species we never find, apparently, 

 striping of yellow or ivory on a white ground ; nor do we find striping 

 in the red anthocyanin. The magenta anthocyanin may show stripes, 

 though not in the tinged varieties, but only when the full-colour factor 

 is present (see p. 159). 



Much of what has been said in this section is purely speculative; 

 it may however be of value in suggesting material and openings for the 

 additional research which is much needed to place the whole subject 

 on a sound basis. As we have seen, very little is known of the causes 

 and mechanism of colour variation. Yet we can recognise, among 

 the factors concerned, three types: (1) those which control the forma- 

 tion of some definite substance as in the Y factor of Antirrhinum, 

 (2) those which modify the substance when it is formed, as the I and 



