3°° 



BOTANICAL GAZETTE 



[OCTOBER 



vascular or conductive elements of any kind ended near it, and it 

 usually originated from a single epidermal cell. Although large 

 and solid papillate structures were most numerous, every transition 

 was found between that type and the simplest form of stoma. 

 Thus some papillae consisted of only a few cells in addition to the 

 stomatal guard cells (fig. 26), and there was every gradation in 

 size from this up to a papilla of dozens of cells. Some of the 

 papillae were hollow also, and it appeared that the cavity had 

 formed by the breaking down of the internal cells into a mass 

 resembling mucilage. The simplest stomata consisted of guard 

 cells with no differentiated auxiliary cells (figs. 32, 33), and often 

 with neither intercellular space nor air chamber below. A few 

 normal stomata had mucilage masses just underneath the stomatal 

 opening. 



Another peculiar situation was suggested by the occasional 

 occurrence of a perfectly developed stoma lying almost under the 

 margin of decorticating patches, and it was finally disclosed after 

 a careful search. Three stomata were found developing under 

 several layers of cork cells, one of which had guard cells just 

 beginning to split apart (fig. 30). So far as can be discovered, 

 the subcortical formation of stomata has never been reported. 

 Perhaps it does not occur outside of the cacti. 



In numbers the stomata of the etiolated shoot ran far below 

 the normal plant, 12-16 being the maximal numbers found per 

 sq. mm., even in the lower part of the shoot where they would be 

 expected to be most numerous, in agreement with the nature of 

 the epidermis which here most closely resembled that of the nor- 

 mal shoot. The stomata were mostly open. 



The regions of a cross-section of the etiolated shoot also con- 

 trasted sharply with those of the normal stem (figs. 1, 3, 11, 13, 

 17). That the cuticle was absent has already been mentioned. 

 The epidermis consisted of cells much broader than normal and 

 much better supplied with protoplasm. Large budding chloro- 

 plasts were present which resembled chains of yeast cells in form, 

 and varied in size from 30 /x to granules too small to be studied 

 with the highest available power of the microscope (fig. 17). 

 Below the epidermis there was neither a crystal-containing layer 



