NOTES ON ALABAMA PLANTS I53 



is not necessarily very reliable, and, not infrequently, must be 

 accepted with caution. Such a character, however, may become 

 very valuable in the course of time if it proves constant under 

 further observation. 



Robinson and Fernald's Gray's New Manual confirms the con- 

 stancy of the limit as stated in Gray's older manual. The same 

 limit — 15-20 — is rnentioned in Small's Flora of the South Eastern 

 United States,'' tut with the restricting remark "usually." It is 

 not clear whether this "usually" refers to an occasionally lower or 

 higher number, but -the former seems more probable when we 

 refer to the key, which reads, "stamens less than 25." In Britton's 

 works, the Manual of the Flora of the Northern States, s and the 

 Illustrated Flora, the limit is extended both ways, viz., 10-30, 

 against Gray's 15-20. Another good character is mentioned in 

 these works, viz., the length of the style as compared with that of 

 the stamens. This character with that regarding the number of 

 stamens unquestionably makes the plant from the Cliffs on the 

 lyittle River in CuUman County a species distinct from Talinum 

 tereiifoUum Pursh. 



The species is 'easily recognized by: stamens more than 40; 

 style protruding beyond the stamens about as much as their 

 respective length; petals 9-13 mm. long. It becomes then 



Talinum Mengesii, spec. nov. 



Type specimen in Herbarium St. Bernard College, no. 1668. 

 Plant succulent, glabrous, 2-4 dm. tall, perennial by a short 

 horizontal rootstock which is sometimes branched, or. the innova- 

 tions arise from a very short vertical caudex; roots slender, fibrous. 

 Stems tufted, generally few or single, 3-10, or rarely up to 15 cm. 

 high, unbranched for almost the whole length, the one to five 

 br9,nches. arising from near the top just below the base of the long 

 wiry peduncle by which the stem is terminated. The branches, 

 from a few to 10 cm. long, are often branched again at their ends 

 from below the secondary peduncles, but these secondary branches 

 generally do not develope beyond the phase of fascicled leaves, in 

 other cases, however, grow to normal size and terminate into 

 peduncles. The leaves are quite numerous but not crowded on the 

 stem and branches, terete, 2-5.5 cm. long, 1.5-3 mm. thick, not 



^ Small's Fl. S. E. U. S. p. 440. 1903. 



5 Brit's. Man. N. U. S. & C. p. 385. 1905. 



