I I 



ROSE MEXICAN AND CENTRAL AMERICAN PLANTS. 87 



LILIACEAE. 



* -^ 



DASYLIRION AND ITS ALLIES. 



After passing in review the specimens and descriptions of Dasylirion 

 and Xolina I have become convinced that certain species should be 

 excluded from both and l)rou<;-ht tofrother into a third genus. Some 

 of these species have long been known as lieaucarnea and are an 

 attractive sight in every large conservatory. 



The discovery of female flowers and mature fruits of the so-called 

 i\dsylirw?i liooherl furnishes data vrhich justify its segregation also. 



The following key will point out the essential differences in these 



o'cnera: 



Ovary 1-celled; fruit strongly S-wintj^ed. 



Stems strongly l)ulbose at base; leaves never saw-toothed; in- 

 florescence a broad open ]»anicle - - - BjiArcAKXKA. 



Stems never bulbopo at base; leaves saw-tootbu<l (except in <»ne 



species); inflorescence an elongated spike-like panicle -- Dasvlikion. 



Ovary 3-celIcd; fruit not at all \\ iii^^(Ml. 



Fruit not 3-lobed,its walls very thick and woody, not dropping 

 awav; trunk a swollen ^rlobular hodv Calibanus. 



Fruit strongly 3-lobed, its Avail very thin, t^oon dri)pj»ing away; 



trunk if present never swollen 2soliaa. 



]5KAL'CAKNEA. 



a 



The genus Beaucarn(ni was described by L<unaire in 18G1 with B^ 

 recuTvata as the type. 'I'wo other species {B, striata and IL gracilis) 

 \v(^re also described bv him. In 1872 and also in 1881 J. G. Baker 

 monographed the genus, coiiibiiiing- with it the imieh ohUu- genus 



Nolina. He describes twelve species and several varieties. In ISTo 

 S. Watson monographed the United States species of Kolina, stating 

 that the Mexican species of Beaucarnca described by Mr. Raker should 



doubtless be referred to Nolina. 



last 



paper Mr. Hemsley again takes up the name Nolina, and so the two 

 names have been alternating, first one and then .the other receiving the 

 sanction of botanists. Jn Anu^rica the name Nolina, being nmch the 

 older, has generally been ac(^epted by botanists, while gardeners 

 throughout the world have, as a rule, clung to Beaucarnca. 



After a careful study of the species along with those of Dasylirion 

 it seems clear that both genera should l)e retained. For at least two 

 accepted species of Dasylirion with several referred to Beaucarnca 

 and Nolina form a very natural genus al)undantly distinct from both 

 Nolina and Dasylirion. The genus Beaucarnca has the inflorescence 

 and foliage similar to those of Nolina while the fruit is much like 

 that of Dasylirion; hence heretofore those species of Beaucarnca 



aEEAUCAKXEA Lomaire, Illust Ilortie. 8: Misc. ^7. pi. MS. 1861. Type species 7? 

 recurvata. 



4153— VOL X, I'T 3—06 2 



