RESERVE MATERIALS OF PLANTS. 115 



by a gradual change of a chemical nature. At first the de- 

 posit or secretion stained homogeneously much more deeply 

 than the protoplasm ; the grains dissolved in dilute caustic 

 potash, but not completely, a membrane always being left. 

 At this early stage they were not dissolved by sodium chloride 

 either in ten per cent, or saturated solution, though they were 

 soluble in these fluids when mature. As they advanced in 

 development they became vacuolated, remaining always 

 denser on their periphery. Finally they were again homo- 

 geneous and then showed the reactions of the grains of the 

 ripe seed. Intermediate conditions could be observed as 

 maturity was gradually approached. 



Rendle thus argues that the process of aleurone grain 

 formation is one of true secretion and not of a mere 

 mechanical precipitation. He opposes Pfeffer's view of the 

 part played by the mineral matter, observing that in Lupinus 

 at any rate the cells contain no crystalline bodies at all at 

 the stage when the grains begin to be formed. 



Another view of the mode of their formation has been 

 advanced. Wakker writing in 1888 (3) describes what he 

 believes to take place in the endosperm of the castor oil 

 plant (Ricinus communis). He notes that the protoplasm 

 of the cells is greatly vacuolated, and suggests from his 

 observations that each vacuole corresponds to an aleurone 

 grain. This vacuole becomes filled little by little with the 

 albuminoid matter as the seed ripens. Werminski (4), 

 working on the same seed, in the main confirms his obser- 

 vations and supports his views. Van Tiegham has sug- 

 gested from the firmness of the protoplasm round the 

 vacuoles that the latter are not mere cavities in the living 

 substance, but that they really correspond to plastids, and 

 as they are filled generally with sap only they may be called 

 hydroplasts. As the aleurone grains, according to Wakker's 

 observations, are always formed in these vacuoles, we have a 

 kind of plastid formation, something like the work of the 

 amyloplasts. 



This view has been criticised adversely by Ludtke (5), 

 who in the main agrees with Rendle in regarding the 

 process rather as one of intraprotoplasmic secretion. His 



