I9I0.] TRELEASE— SPECIES IN AGAVE. 235 



specimens a generation ago, prove, within limits, entirely depend- 

 able. A few illustrations — from very many that might have been 

 selected — will render this clear : 



If the end-spine of Agave americana — as it occurs in our 

 gardens everywhere, green, striped, yellow-margined or yellow- 

 centered — had been attentively studied years ago, as it has been 

 recently, in comparison with that of the now almost equally common 

 yellow-margined A. picta, the former species would never have 

 been made to include the latter (PI. XXXII.). The spines of gray 

 henequen (A. fourcroydes) and green sisal {A. sisalana) supple- 

 ment other characters in segregating these constituents of what is 

 still too often called A. rigida; and in this respect A. angustifolia 

 differs so greatly from either as to make one who knows the differ- 

 ences wonder how, under whichever of its aliases it was encountered, 

 it ever could have entered into this same modern complex called 

 rigida (PI. XXXII.). Three groups — superspecies, they might be 

 called — of the now economically interesting zapupe agaves are dis- 

 tinguishable from one another, even to the touch, in this same char- 

 acter (PI. XXXIII. ),^ and each group falls into species on its mar- 

 ginal arming. The likewise important group of mezcals grown for 

 the production of Tequila spirits, known to science in one com- 

 positely described species (A. tequilana), shows a similar differen- 

 tiation into an even greater number of forms (PI. XXXIII.) ; and 

 many of the great maguey forms grown all over the Mexican table- 

 land for the production of pulque are unmistakably distinguishable 

 on their spine and prickle characters. 



These examples, I trust, may justify the devotion of a some- 

 what lengthy prologue to the argument that small things are not 

 to be despised ; or to a short epilogue drawing the conclusion that 

 the arming of an Agave is no less significant in species discrimina- 

 tion than the disarticulation of the sepals of an apple, and that, in 

 fact, neither stands alone. 



Until within very recent years, few herbaria have possessed more 

 than one or two leaf fragments of a given Agave, and where more 

 than one occurred the chances were good that they were not co- 



^ Trans. Acad. Sci. of St. Louis, XVIII., no. 3, May, 1909. 



