THE RESERVE MATERIALS OE PLANTS. 77 



pollen is placed in a nutritive fluid (27) very speedily small 

 granules of starch appear in this homogeneous protoplasm, 

 and the number continually increases as germination begins. 

 In the pollen tubes of the lily the same thing is seen, the 

 number of small grains being enormous. In the style of 

 the same plant, in the cells containing again quite trans- 

 parent protoplasm, the same formation is found, the cells 

 in some cases becoming quite filled with starch. It is 

 rather hard to conceive in these cases of a simultaneous 

 formation of plastids to carry out the deposition. 



The same considerations apply to the deposition of 

 starch grains in young developing embryos. Some obser- 

 vations, published by Belzung (28), made upon the young 

 embryos of various Leguminosae, particularly Lupinus 

 albus and Phaseolus vulgaris, point to the non-occurrence 

 of the leucoplast as an antecedent to starch deposition. 

 Indeed, Belzung goes so far as to advance the view, in 

 opposition to Schimper, that the leucoplast theory is based 

 upon an erroneous interpretation of the appearances seen. 



His conclusions may be briefly stated as follows : In 

 Lupinus albtis and L. mutabilis there are no plastids in the 

 tissue of the developing embryo, yet starch is deposited 

 there during the first phase of its development, which 

 starch later disappears during the ripening of the seed. 

 To a less extent, the same thing may be observed in the 

 embryo of Phaseolus vulgaris. The protoplasm of the 

 cells of such embryos in course of formation may be seen 

 to have the form of a network with many small meshes, 

 which are empty spaces, or contain only cell-sap. There is 

 no leucoplast inside them, nor anything comparable to one. 

 The starch grains originate in these meshes at some point 

 in contact with the protoplasm, and gradually increase in 

 size till they fill them, growing in fact like crystals, and 

 forming the simple grains of starch. Later these dis- 

 appear, and in cases where starch appears further on in the 

 development in the form of compound grains, the formation 

 is due to the vacuole or mesh being partly filled by a 

 smaller and much finer network, giving still smaller meshes. 

 In these smaller meshes, forming part of the larger original 



