C 66 ] 



itself in the fact that A. sisalana Pem'ne sends up suckers 

 freely and continuously, whereas Miller says of his A. rigida, 

 " This sort never puts out suckers from the root, nor have I seen 

 any plants of this Irind in flower^ although there are many of 

 them in the English Gardens some of which are of considerable 

 age." 



As to A. Ixtli Karwinski, and A. decipiens Baker, the 

 peculiar apple-green of the rather narrow channelled leaves 

 is a striking mark, as remarked by both authors, whereas 

 Miller's plant was glaucous. Moreover A. rigida came from the 

 Vera Cruz country, while A. Ixtli is, like sisalana, from the 

 Yucatan peninsula, which has a Flora, as Grisebach has pointed 

 out, more akin to that of the Caribean region than to that of 

 Central Mexico. Miller's A. rigida resembles sisalana in one 

 point it is true> that it was spineless (though this is by no 

 means always so with sisalana) but this very character divides 

 it from A. Ixtli, which is well armed. 



Specimens of what we believe to be Baker's A. decipiens 

 are growing at Sibpur and Saharanpur, also with the Agri- 

 Hort. Society at Lahore, from which garden the Calcutta 

 specimen is derived, the Lahore name being tl A. rigida" 

 and the stock received from Kew originally. The Saha- 

 ranpur supply came from a Nursery in Florida labelled 

 " A. rigida, type" referring doubtless to the work of 

 Engelmann. These plants exactly match Miss Mulford's 

 illustration (from a, photograph) of A. decipiens (Missouri Bot. 

 Garden Report, Vol. 7, 1896, p. 67), and Miss G. E. Johnson's 

 outline drawings (pi. 68-69), also the sketch in U. S. A. 

 Department of Agriculture Report No. 9 Washington 1897, 

 p. 5, figure 9, except that the trunk in Mr. Dodge's sketch! is 

 more conspicuous, the subject having manifestly been an old 

 and large specimen from which the outer leaves have been cut 

 or dropped away extensively. 



"A. Ixtli" of Qartenflora 1883 (p. 149) may be=.4. 

 ivtlioides of Bot. Mag. 5893. We cannot think that it is= A. 

 decipiens Baker, and it is widely removed from A. sisalana, 

 Perrine. The synonymy is very intricate. 



We have discussed this Agave (A. decipiens Baker) at some 

 length, partly because of its supposed resemblance to A. 

 sisalana, a resemblance which we are bound at the same time 

 to sav we are unable to discover, and partly in the hope of 



