526 THE AMERICAN NATURALIST. [Vol. XXXVIII. 



the characteristics of which are well known and described by 

 Wittstein (D. 53, p. 63). Tests were applied to thin radial and 

 tangential sections, employing for this purpose ( i ) various essen- 

 tial and fixed oils, (2) ether, (3) alcohol, (4) ammonia, (5) potas- 

 sium hydrate in one and one half percentage solution. The 

 plates were found to be very refractory with respect to both the 

 fixed and essential oils, as well as towards ether, alcohol, ammo- 

 nia and xylol, and in all of these cases no change was to be 

 observed, even after an action extending over several weeks. A 

 partial exception applies to alcohol and ether. In the latter case 



there did 



rp. 





rp. 





n 



L 























nution in volume, apparently 

 through solution, when the 

 reagent was first applied, 

 but after that there was no 

 further alteration. The 

 application of alcohol, both 

 in the hot and in the cold, 

 showed that while the resin 

 contained in the medullary 

 rays was all dissolved, the 

 plates were only partially 

 affected. The 



the reagent was chiefly manifested in the 

 development of strong curvature, often 

 accompanied by fracture. This was evi- 

 dently due to an increase in volume, and 

 a tendency toward solution, and it gave 

 the first definite evidence that the plates 

 could not be of a cellulose character. Be- 

 yond this no further change was brought 

 about, even after several weeks of action. 

 The potassium hydrate gave the most 

 positive results. At first there was no 

 apparent change, but after an interval of 

 about ten days or two weeks the plates 

 were found to have completely disappeared, 

 leaving a perfectly clear channel in the cell 

 cavity. A further proof of the resinous 

 character of these plates is to be found in 



