572 Holden and Bcxon.—On the Seedling 
utilized as rapidly as they are produced in addition to the stored starch of 
the embryo during the early phases of rapid growth. Following on the 
expansion of the cotyledons, which apparently coincides with the advent of 
full assimilatory activity, the formation of storage starch commences. This 
appears to be closely associated with the developing vascular structures, the 
major portion being situated in the outer part of the pith and the inner part 
of the cortex. As the secondary xylem of the hypocotyl develops, abundant 
starch also occurs in the xylem parenchyma and in the medullary rays. 
In addition to the starch reserves there appear to be at least two proteins. 
During the early stages of the investigation free-hand sections were cut from 
embryos which had been preserved in methylated spirit, and it was noticed 
Fig. I. Cell from the cotyledonary parenchyma mounted ‘ dry ’• 
Fig. 2 . The same cell irrigated with tap-water, showing the emulsoid phase of precipitation. 
Figs. 3-5. Cells showing variations in the final phase of precipitation. The starch grains are 
indicated as black granules throughout. 
that the cell-contents were obscured by numbers of globules which stained 
brown with iodine (Fig. 3). Subsequent investigation showed that the 
globules were also present in sections of material fixed in formalin or 
chromacetic fixatives, and of jresh material mounted in tap-water. It was 
not until the precaution was taken of mounting sections of fresh material in 
tap-water and examining at once that the clue to the formation of these 
globules was obtained. It was then discovered that they were the result of 
precipitation, presumably from colloidal solution. The first stage in precipi¬ 
tation is the development of a fine haziness in the cell-contents, succeeded 
by the formation of minute granules showing active Brownian movement 
(Fig. 2). The coalescence of these granules produces the globules (Figs. 3 -5). 
Sections of fresh material cut and mounted in distilled water developed 
a haziness of the cell-contents similar to that of the preliminary stages of 
precipitation produced by tap-water, but no further precipitation was 
observed. 
Sections were also mounted in dilute glycerol, and here precipitation in 
the form of globules was immediate, there being no emulsoid phase. After 
some time the cell-contents became clear owing to the solution of the 
globules, and this suggested the probability that, with higher concentrations 
of glycerol, no precipitation would occur. Owing to the difficulty of 
obtaining anhydrous glycerol the ‘ pure glycerine ’ of the British Pharma- 
