36 TRANS. ST. LOUIS ACAD. SCIENCE. 



found shorter and broader than usually (18-21 by I2-2I inches), 

 and it was here that the three-branched plants were observed. 

 In cultivation the leaves are 12-21 inches long, and 1— 1 i wide. 

 Dr. Mellichamp has sometimes, in plants growing close to the 

 beach, seen the upper surface of leaves incrusted with a white 

 deposit, which might be taken for saline efflorescence, but proves 

 to be carbonate of lime, with a very delicate film of organic 

 matter representing a cast of the epidermis cells ; probably an 

 exudation from these cells of oxalate of lime altered by oxydation. 



The flowers open in July and August, and in the evening 

 expand 3-4 inches, while in daytime they are 12-if inches 

 deep. I find the stamens, in native as well as in cultivated 

 plants, as long as the ovary, often as long as the whole pistil and 

 occasionally even overtopping it. The unusually stout ovary with 

 the short stigma is 9-1 1 lines long, the ovules I found 0.35-0.38 

 mm. thick. The pod is 2J-3 times as long as it is thick (3-4 by 

 ii-if inches), six-sided, the sides corresponding to the carpels 

 more elevated, the alternate ones sharply depressed and turning 

 purple before the rest of the fruit, which at last assumes a deep 

 purple color inside and outside, has a sweet, not unpleasant, 

 taste, and is much eaten under the name of Banana. It is always 

 acutish but never rostrate, distinctly tipped with the 3-lobed 

 stigma preserving its tube, whence the fruit is described as " per- 

 forated at the apex." The seeds, 6-7 mm. in diameter by 2^-3 

 in thickness, are very similar to those of all other Sarcoyuccas 

 examined by me. 



Var. Draconis I cannot distinguish from the last except by the 

 leaves being said to be less crowded, longer, softer, less pun- 

 gent, and somewhat flaccid and curved. It is said to come 

 from the same region where my specimens, above described, were 

 obtained, and may perhaps be the form with very long leaves, 

 grown in the shade, described above. The plants, cultivated here 

 and there under that name, may in part be T. Guatemalensis, de- 

 scribed below. 



Var. consflicua, or at least the plant cultivated under that name 

 in the botanic garden at Naples, differs from the type by its softer, 

 though not pendulous, leaves, with a green scarcely pungent point. 

 It there makes large bushes, over 20 feet high, branching abun- 

 dantly from or near the base, the thickest trunks 6-9 inches in 

 diameter. I notice that the panicles, sometimes three feet long, 



