MICROSCOPIC CHANGES. 15 



readily ))e comprohendod by ii coinpurison of PI. Ill, li^s. i and 2. 

 Fig. 2 is from n phc)t()<>iapii of an unstained section taken from the 

 region of brown wood at the outermost edi>e. 



It is rather diflicult to determine the origin of this decomposition 

 ])roduet. It is possil)ly the hist product of a change in the starch 

 grains, possibly also a substance derived from wood cells farther 

 inward, which infiltrates into the nu^dullary ray cells and wood paivn- 

 chyma in advance of the fungus hy})hie. The latter is the })robable 

 explanation, for one finds the hunuis compound in the suumier wood 

 cells, which had very little starch originally. The hunuis compound 

 appears to form in many of the wood cells, however, as ji product of 

 the walls. Figs, -i and 5 of PI. IV show various stages of this change. 

 The cells tt are sound wood cells, which have very thick walls and a 

 ver}'^ small Imuen. The walls of cells marked // are ver}' much thinner, 

 and at these points they are coated with the humus com})ound. Such 

 walls when stained with })hloroglucin show no \"ery sharp dividing line 

 between the yellow hunuis compound and the a})parently sound ligni- 

 lied wall. Cell c is completely tilled with the hunuis mass. This evi- 

 dence that th«; wall actually changes into the yellow mass is not very 

 conclusive. The humus compound does not seem to be formed from 

 the walls of the medullary ray cells, where it is found ultimately, for 

 no signs of change are evident in the walls of these cells. Tlu' local- 

 ized distribution of the hunuis substance is very striking. It is always 

 absent from the wood cells of the spring wood (PI. Ill, tig. 2) and 

 from the large vessels. In the cells it appears to be as a solid mass, 

 sometimes completely tilling the lumen (PI. IV, figs. 2 and 5), or in 

 globules or plates adhering to the walls (PI. IV, fig. 2). It is this su]> 

 stance which gives the brown color to the earlv stage of diseased wood. 



The next stage in the dissolution of the wood cells takes place 

 abruptly, and is rapid after it has once set in. The liypha of the 

 fungus first evident in the medullary ra3's spread through the wood 

 of both the spring and summer bands, branching in all directions. 

 They give off an enzyme which attacks the inner parts of the wood 

 cells, extracting the lignin. A transverse section of wood in this stage 

 (PI, IV, tig. 1) stained with phloroglucin presents a most striking 

 picture. Here and there, in irregular groups and in all stages, one 

 finds wood cells from which the hadromal has been removed; the 

 extracted parts remain white and stand out in sharp contrast to the 

 unaffected parts of the walls. In the figure the unaffected parts are 

 shaded. The white parts represent delignified walls. The middle 

 lamella is dissolved last and then the individual cells fall apart. When 

 this takes place throughout larger areas, for instance, one or more 

 wood rings become separated from one another, and this gives rise to 

 the plates spoken of above. The white areas which are evident in the 

 figures on PI. I represent wood thus destroj^ed. The individual fibers 



