54 SECTIONAL ADDRESSES 
that the cyanin anion is present in a complex form, giving a stable aggre- 
gate with a negative charge ; in some way the strength of cyanin colour- 
base as an acid must be vastly increased. 
Some form of colloidal solution was considered most likely to fulfil the 
necessary conditions, and Dr. Conmar Robinson, of the Chemistry 
Department, University College, London, kindly examined a filtered, 
distilled-water extract of blue cornflowers and reported as follows : 
‘'The solution contains ultramicrons easily visible in the slit ultra- 
microscope, but small enough to be in fairly rapid Brownian movement. 
Microcataphoresis showed them to be negatively charged. Without 
more quantitative work it is impossible to say if these particles can repre- 
sent the bulk of the material present, but this seems probable if the 
solution is very dilute ; the possibility of observing a colloidal impurity 
is always a trap. he visibility of the ultramicrons suggests a lyo- 
phobic colloid. It is, however, not precipitated even by 2N NaCl, 
which indicates that a protective colloid is also present.’ 
Our next step was to attempt the production of blue cyanin sols stable 
in neutral or weakly acid solution, and some measure of success was 
achieved, although the solutions are by no means so stable as those from 
the blue cornflower. 
If a little crystalline cyanin chloride is added to boiling tap-water 
(pH 8-0) then the usual violet solution results (see above), the colour 
being what we consider ‘ normal.’ If, however, the cyanin is triturated 
in the cold for a minute with the water and gradually heated to boiling 
with shaking, then a beautiful blue solution results. The fact that the 
same materials can be used to produce two entirely different results shows 
that it can only be the state of aggregation of the cyanin which can have 
stabilised the anionic charge and hence produced a blue colour under 
the conditions that normally produce a violet solution. If very small 
quantities of cyanin chloride are employed, this phenomenon can be repro- 
duced using distilled water. Willstatter and Everest found that their corn- 
flower extracts contained xylan and other polysaccharides, and we have 
attempted to produce blue acid cyanin solutions in the presence of various 
polysaccharides. The addition of dispersed xylan and various kinds of 
starch, also Agar-Agar, makes the preparation of blue solutions of pH about 
7:5 a very simple matter (demonstration), but we have not yet found a way 
of imitating the cornflower solution in respect to its stability at pH 5-0. 
Probably these colloid associations are much more readily formed at 
values of pH between 5-5 and 6-5, and on the whole the blue flowers have 
less acid cell-saps than the red flowers. The petals of the rose in contrast 
with the cornflower constitute an exception (pH 5-6), and the following 
further provisional results may be quoted although no great accuracy 
can be claimed for a method which involves the destruction of the petals. 
The pigment of the orange-red polyantha rose ‘ Gloria Mundi’ is found 
to be pelargonin and the pH was 5-5. On the same plant some flowers 
had reverted to the cyanin type. The red-flowered hydrangea had petal 
PH 3°75, whilst the blue flowers gave pH 4:9; similarly the red-flowered 
linum (anthocyanin based on delphinidin) gave petal pH 4-6, and the’ 
blue variety pH 5-9. Blue anchusa, 6:2; Meconopsis Baileyi, 5-3 
(Miss R. Scott-Moncrieff found that blue and violet flowers had the 
