F. C. Miles 207 



if the specimens were first killed and then prepared according to the 

 paraffin method. 



Experiments were made with several different killing reagents, but 

 the most satisfactory method was to allow the leaf sections to remain 

 for about two and one-half hours in a modification of Carnoy's killing 

 fluid composed of absolute alcohol 50 per cent., chloroform 25 per cent., 

 and glacial acetic acid 25 per cent. After killing, the specimens were 

 washed in absolute alcohol, cleared in xylol, infiltrated and embedded 

 in paraffin. 



Cross sections of the leaves were made of from three to fifteen 

 microns in thickness, but those sections which were either ten or twelve 

 microns in thickness proved the best for a study of the plastids. Of the 

 stains employed Lichtgriln appeared to be most satisfactory, yet some 

 good results were obtained from Delafield's haematoxylin, and also from 

 acid fuchsin. Camera lucida drawings were made from the prepared 

 slides by Miss Lucille Goodloe, Washington, D.C., to whom the writer 

 desires to express his thanks at this time. 



Examination of leaves from the pure white plants showed that the 

 plastids apparently were almost, if not entirely, lacking (Fig. L). Even 

 leucoplasts could not be differentiated with any of the stains which 

 were used. This is an extreme condition, yet the conclusion that no 



Fig. 1. Cross section through a leaf of a 

 pure white seedling, such as is illustrated 

 in PI. VIII, fig. 1. No plastids could be 

 differentiated, x 360. 



plastids are present is substantiated by the fact that these pure white 

 plants have never been known to turn green, or even greenish, in colour, 

 but always die as soon as the young seedlings have exhausted the food 

 stored in the kernel planted. 



The condition found, however, in the leaves of the yellowish-white 



