162 Univcrsify of California. Puhlications in Botany \yoh.7 



This is believed to provide for all of the principal forms, and it is 

 quite certain tliat eacli of the described varieties is a natural unit, 

 althoufrh in most cases it is itself made up of still smaller variants. 



Altliougii tlu' relationships between the varieties accepted in this 

 paper are expressed somewhat in the arranj^ement, and in the form 

 of brief notes from time to time, yet our actual knowledge of these 

 matters is so slight tliat it seems unwise to attempt an expression of 

 it through an elaborate system of polynomials. Some of the varieties 

 are based upon characters which are at least partially heritabl(% while 

 others are doubtless ecologic forms. In some cases two varieties 

 occupying the same territory may be easily distinguished by a number 

 of characters which at that locality do not overlap in their variations; 

 yet as forms from other places are brought into comparison, it is 

 discovered that these characters are so variable that they cannot be 

 used for the recognition of species in the broad sense. It seems almost 

 certain that after all of the forms shall have been assembled it will 

 be found that no two of the characters thus far proposed for the 

 separation of "species" in this group wall be found to vary in unison, 

 and furthermore that there is not a single one of them that does not 

 vary by imperceptible degrees. Therefore we need not be surprised 

 ultimately to find as man}- forms as there are possible combinations 

 of characters and to have, in addition, numerous forms which cannot 

 be satisfactorily^ placed because of the intermediate nature of one 

 or more of the characters. It is thus seen to be impossible to accord 

 specific rank to any of the units of this exceedingly complex and 

 variable assemblage of forms unless one is willing to accept a species 

 concept so narrow that its usefulness would be lost because of the 

 impossibility of definite application. 



If some readers object to the use of even trinomials we woidd 

 suggest that the binomial indicating the inclusive major species, 

 that is, Chrysothamniis nauscosus in this case, is all that need be 

 used in most instances, while in any special paper on the group the 

 varietal name alone may be used without confusion, since the generic 

 and specific names are understood. This practice will be followed 

 in the report on Chrysothamnits as a possible source of rubber, in 

 which paper the actual w^orking out of the method may be observed. 



In preparing the following synopsis, the presence or absence of 

 hairs on the involucre is taken as the principal character, notwith- 

 standing its variable nature. This is done because it comes the nearest 

 to assembling the forms into what seem to bi' natural groups, or sub- 



