370 TRANS. ST. LOUIS ACAD. SCIENCE. 



beds of sandstone. I have also observed the sandstone resting on 

 porphyry, and have also seen beds of sandstone, or conglomerate,^ 

 resting on porphyry, or on granite. 



We feel ourselves inclined to regard the granite as the older,^ 

 the porphyries as of more recent formation. 



Pleasant Hill, Mo., March, 1876. 



Addition to Dr. Engehnann^ s Article on Agave, 

 p. 291-322. 



Additions and corrections have accumulated in the interval 

 betw^een printing and publication. 



Page 293. The words of the first note, line 3, "while" to the 

 end of the sentence "liliaceae" should be carried up into the text 

 and added to the first paragraph, line 8, ending with the word 

 "stigmas." 



Page 294. It requires further and extensive observations in the 

 field and in the garden to ascertain the limits of variability of the 

 edges of the leaf and its aculeate-toothed margins. Cultivators 

 have already discovered considerable latitude in this respect in 

 plants raised from seeds from the same parent. 



Page 302. Var. tigrina does not grow in salt-marshes, but 

 was found by Dr. Mellichamp, "in one spot only, on a tongue of 

 partly brackish land, extending out into the salt-mud and marsh, 

 under dwarfed live-oaks, Cassine, and saw-palmetto, on the de- 

 cayed shells, mixed with sand and earth, of what seems to be an 

 old Indian oyster-heap" ! 



Page 304. A. falcata. The lately introduced A. Hystrix of 

 the Belgian nurseries may have to be referred to this species ; na- 

 tive country and flowers, as usual, unknown. 



Page 305. A. ScJiottii (better than Schotii, as printed). It 

 is proper to add that A. geminijlora^ Gavvl., the Bonapartea 

 juncea of the gardens, is a very different plant, and has nothing 

 to do with Schottii^ except that both belong to the same section 

 of the genus. 



Page 310. A.deserti. In characterizing the foliage, I was led 

 into the too common mistake of adopting the individual charac- 



