130 THE AMERICAN MONTHLY [June, 



where the cell walls are thin, and with a view to finding which 

 jDlants of those accessible to most students in our country are 

 most suitable for the purpose, we began the work by examining 

 the cortical layer of one or more species of about seventy-five 

 genera of native and cultivated exotic trees and shrubs. With- 

 out going into details in regard to all the various methods which 

 have been tried by Sachs, Gardiner, and others, we will merely 

 state that at first beginning, a thin fresh tangential section was 

 made with a razor and at once a drop of strong sulphuric acid 

 from a glass rod was placed upon it. After one to three min- 

 utes, depending on the thickness of the cell walls of the specimen 

 examined, the sections were rapidly washed in water, and 

 stained with iodine or analine blue. 



Chloriodide of zinc was also tried. But the most satisfactory re- 

 sults were obtained by placing a thin section in a solution of 

 iodine, until it turned brown. Then wash, and then run strong 

 sulphuric acid under the cover-glass, and soon, again, draw out 

 the acid, replacing it by water. After washing well, mount in 

 glycerine and color with green or blue analine. 



In these experiments, a great deal of time and labor for weeks 

 may here be passed over by a mere mention of the fact, calling 

 it " dead work." Now for the results : The large masses of pro- 

 toplasm in all the cells properly treated were very pronounced, 

 and from each mass extended arms or strands in every direction. 

 Each arm in a cell exactly approached and met or seemed to 

 alfjiosf meet one from an adjoining cell. In some places, in every 

 species examined, a continuous strand seemed to extend from 

 one cell to the next one, but this was very likely due to some im- 

 perfect work in making the experiment. In most instances there 

 was a light line, through which the connection could not be 

 successfully traced. At this place the tips of the strands were 

 swollen and between them a light, hazy look was very common. 



The question to be demonstrated was the presence of delicate 

 threads of protoplasm passing from the swollen tip of one strand 

 to the swollen tip of a contiguous strand in the next cell. In the 

 cells of the endosperm of palms the walls are very thick, and the 

 space between the tips of contiguous strands rather prominent ; 

 but in the cortical layer of our woody plants the cell wall is rather 

 thin, and the space between the tips of the strands is a very nar- 

 row one. To trace threads across this narrow passage is difficult 

 and often unsatisfactory, as the threads are ver}' short. We em- 

 ployed a .Spencer's one-eighteenth-inch immersion objective. 



Among the most satisfactory of the many specimens of cortex 

 examined was that of the common lilac, where the sections were 

 taken from large-sized twigs two to three years old. Viburnum 

 opulus (snow ball) was much the same. In specimens of our 

 hawthorns the strands were slender and longer, but the threads 

 joining their tips showed very well. Perhaps the best one of all 

 was taken from the cortex of Alnus irlntlnosa. Here the 



