EMBRYO, SEED AND CARPEL IN THE DATE. 125 



with difficulty recognizable as of similar origin to the iso- 

 diametric idioplasts themselves. 



The inner mesocarp may, for our present purpose, be de- 

 scribed as a mixture of collapsed, chiefly elongated, thin- 

 walled elements penetrated by a few displaced and distorted 

 vascular strands. Here and there, isolated or in small 

 groups, are very long cells rich in tannin. Sclerosed cells 

 of similar form, having dimensions upward of 10 mm. in 

 length and .2 mm. in diameter, their walls penetrated by 

 simple oval or circular pores, and vaguely obliquely striated, 

 are also to be found, more especially opposite each lateral 

 angle of the seed, where a group readily visible to the naked 

 eye, may occur. In transverse section, these stereids are 

 circular or oval and sometimes contain tannin (f. 34). 



Additional tannin is occasionally observable also through- 

 out the mesocarp elements and bundles, in both hadrome and 

 leptome. Frequently vessels filled with tannin have been 

 seen. I am unable, however, to make a sufficiently detailed 

 account of this phase of the matter to warrant any definite 

 conclusions. This would require a more particular study 

 of the vascular tissues than I have been able to give them. 



There is some evidence that plastic tannin occurs in small 

 amounts in the parenchyma of the mesocarp in certain re- 

 gions. Copper acetate and ethyl nitrite material frequently 

 displays a copper-red or orange coloration respectively, which 

 blackens with iron salts. As maturity is approached, this 

 behavior is not to be observed. Fresh material was not 



may 



seem 



— _ 



to be described in the endosperm, occurs in small amounts 

 in the mesocarp during development. The reaction upon 

 which this conclusion is based is more marked in the sutural 

 sector -than elsewhere, and in the inner than outer mesocarp. 

 In the latter, however, the amount, as previously stated, is 

 exceedingly small at any time. For some time previous to 

 ripening, as attested by Thornber, Vinson and myself, no 

 tannin occurs in the major portion of the mesocarp. There 



is, therefore no evidence in this instance that tannin gives 



