14 JOURNAL OF THE [January,. 



greedy absorbence of the anilin dye by the protoplasmic sarcode ; 

 but the overcoloration may be controlled by proper dilution of 

 the stain. The infusorians are then mounted in balsam, Wickers- 

 heimer's fluid, or other preservative, in the usual way. 



NOTE ON THE STRUCTURE OF THE ENDOSPERM 



OF PHYTELEPHAS MACROCARPA RUIZ AND 



PAVON, AND OF SMILACINA RACEMOSA 



DESF. 



BY J. L. ZABRISKIE. 

 (^Presented December 15//;, i?930 



I. Fhytelephas macrocarpa. — The seed of this plant, popularly- 

 known as the ivory nut, is an irregular ovate body about one and. 

 one-half inches iYi diameter by two inches in length. When it is- 

 cut through the middle transversely and longitudinally the endo- 

 sperm discloses a ''grain," easily seen with a hand lens, resem- 

 bling in miniature the "grain" of the wood of an exogenous tiee. 

 The transverse section shows concentric circles sweeping around 

 the longitudinal axis of the nut. The longitudinal section, through 

 the middle portion of the nut excluding the poles, shows minute 

 longitudinal bands nearly parallel with the axis. Close inspection 

 shows why the poles must be excluded in referring to the bands as. 

 parallel with the axis. Rings and bands are found to conform 

 quite accurately with the dark outer surface of the nut, so that 

 when the poles are approached the bands sweep around the latter. 

 It is as if the entire substance of the endosperm were composed of 

 tenuous films or shells of " vegetable ivory," these shells conform- 

 ing to the outer surface, closely fitting one within the other, and 

 therefore gradually decreasing in size as they approach the axis. 



The ''grain" is caused by a remarkably uniform arrangement 

 of similar, lengthened cells composing the entire endosperm. But 

 the arrangement is diametrically opposite that which causes the 

 "grain" in the tree. In the tree the cells of the fibre generally 

 overlap each other, and all lie parallel with the axis, and of course 

 parallel with the bark. And a ring or band is caused by the 

 assembling of some of these cells, which are comparatively smalL 



