James Small. 
23 
It should be noted that colloidal chlorophyll is an electronega¬ 
tive suspensoid, and that it is a very excellent substance for 
demonstrating the properties of such colloids. It might even be 
worth while for this purpose to prepare a small quantity of pure 
chlorophyll (a -i- b) (see section D of this chapter). 
In most suspensoids it is difficult to see when the precipitation 
has taken place, but here where any precipitated chlorophyll can 
immediately be extracted by ether, the method is open for quantita¬ 
tive work of high accuracy. 
By the use of chlorophyll can be demonstrated such properties 
of colloids as the salt concentration required for precipitation, the 
effect on this of the valency of the precipitating ion, the stabilising 
effect of alkali and the reverse effect of acids, the action of protective 
colloids, etc. 
Crude colloidal chlorophyll contains, of course, a good many 
accompanying substances which vary in composition and quantity, 
so comparable data are not obtainable by the use of such 
chlorophyll. 
(To be continued). 
NOTES ON THE COROLLA IN THE COMPOSITE. 
By James Small, B.Sc. (Lond.), Ph.C. 
Demonstrator in Botany , Armstrong College , Neivcastle-on-Tyne. 
[With Thirty-Three Figures in the Text]. 
I N a previous contribution on the Compositae (29) the writer 
explained that any phylogenetic scheme must take into 
account, among other data, “ the form, development and colour of 
the corolla,” and the present paper is an attempt to show that 
these characters in the corolla show a development which confirms 
the hypothesis that the course of evolution in the Compositae has 
followed the lines indicated by the study of the pollen-presentation 
mechanism (29, p. 466). 
So much has been done in the actual observation of the form, 
development and colour of the corolla, that there is little room for 
fundamental research on this part of the flower, except along the 
lines of casual morphology as indicated by Lang (19). There 
remains, therefore, the critical consideration of known facts in the 
light of the phylogenetic suggestions obtained from the study of 
the stamens and styles. I have been on the lookout for facts 
which would prove the changes suggested in the relations of the 
tribes (29) unsatisfactory or contrary to the true history of triba 
