SPECIMEN PREPARATION 



The paraffin -embedding method was employed in preparing the microscopic slides from 

 the specimens. Small wedges were cut from the specimens, aspirated, dehydrated through a 

 tertiary butyl alcohol series, and embedded in Paraplast. 2 After soaking in an alcohol-glycerine 

 solution of Aerosol OT 100, the embedded specimens were sectioned in thicknesses of from six 

 to 20 micra on a rotary microtome; for most of the material a thickness of eight to 10 micra 

 was optimum. The paraffin ribbons were affixed to slides with Haupt's adhesive. After re- 

 moval of the paraffin by xylene, the sections were run through an ethyl-alcohol series into 

 water, stained by means of the Feulgen reaction, dehydrated in a second alcohol series, and 

 counterstained with fast green. In the first alcohol series, a celloidon solution was interposed 

 to improve adhesion of sections. The mounting medium was Harleco's Synthetic Resin. Cross, 

 radial, and tangential sections were cut for each species. 



PREPARATION OF KEY 



The key was first outlined from photomicrographs of the sections and then checked with 

 the slides and with published accounts of wood anatomy. Because in many cases the original 

 photographs did not show the characters sought, the slides for each species had to be examined 

 several times. Each character used to distinguish a species was checked in all specimens 

 representing that species . 



The characters used to distinguish among the species are not in all cases the same; ves- 

 sel distribution, perforation plates, rays, and axial parenchyma are the characters most often 

 utilized. Where information about wood anatomy for a species was available, it was consulted 

 as a check against the specimens used in this study. These accounts, with few exceptions, 

 confirmed the conclusions outlined. 



DISCUSSION 



The premise upon which this study was based is that species of shrubs are identifiable on 

 the basis of characteristics of the xylem . However, evidence indicates that such distinction 

 cannot be made. The results of this study confirm that in many genera the species produce 

 xylem that is so similar as to be valueless for identification; e.g. , the two species of Ceanothus 

 included in this key. Consequently, wood traits alone may not be adequate to identify a shrub. 



Before attempting to use this key, the reader should recognize that identification of shrubs 

 by means of microscopic characteristics of the wood is difficult. Careful preparation of spec- 

 imens and adequate laboratory facilities are necessary. Moreover, error may result from two 

 sources. The first is that the unknown shrub may be one not included in this study. The second 

 is that species variability has not been assessed. Identification of unknown shrubs by xylem 

 should be attempted only when all other means have been exhausted. 



Use of trade names herein is for identification only, and does not imply endorsement by 

 the U.S. Forest Service. 



