found by Dr. Wight, probably not unfrequently in the 

 Madras Peninsula ; and I possess numerous specimens 

 from Colonel and Mrs. Walker, gathered in Ceylon. It is 

 allied to E. strictum; but abundantly distinct in the very 

 different bracteas, larger size, in the colour of the flower, 

 and the much longer tube. Nees described four varie- 

 ties, chiefly differing in the nature of the bracteas, and in the 

 hairiness about them and the calyx : the stem also seems to 

 vary. Nees von Esenbeck describes it as " teretiusculus." 

 Roxburgh says that the young shoots are four-sided. In 

 ours the branches are acutely tetragonal. It flowers copi- 

 ously in the stove in April and May. 



Descr. Stems weak. Branches four-sided, erect. Leaves 

 petioled, ovato-lanceolate, acuminate, membranaceous, four 

 to six inches long, entire, or, sometimes, according to 

 Nees, repando-crenate, glabrous, strongly nerved. Panicle 

 trifid, much longer than the leaves. Bracteas linear-lance- 

 olate, alternate, more or less glanduloso-ciliate. Calyx 

 five-partite, clothed with patent hairs, glandular at their 

 apices ; segments subulate, erect. Corolla lilac, or rose- 

 purple. Tube very long, curved. Limb of five deep, nearly 

 regular, patent, obcordate, waved lobes, striated and reti- 

 culated with deeper blue lines on one side at the faux. 

 Stamens two, slightly exserted. Style very long, slender, 

 filiform, exserted. Stigma of two very unequal, subulate 



segments. 



Fig. 1. Calyx and Pistil. 2. Ovary. 3. Immature Fruit : — magnified. 



