long, very dense, cylindrical, the flowers closely appressed"), 

 bracts lanceolate the lower ones about as long as the ovary ; 

 sepals broadly ovate subobtuse, 7 mill, long, the side ones 

 spreading, the odd one deflexed ; petals erect-spreading oblong 

 narrower and a little shorter than the sepals ; lip lanceolate 

 erect, with two parallel coloured stripes ; column erect stipitate, 

 shorter than the perianth, furnished behind with two horn-like 

 processes (staminodes ?) as long as the rostellum ; rostellum 

 terminal erect thick triangular truncate, wider than the column ; 

 cells of the anther placed one on each side of the stigma, parallel 

 distant ascending but scarcely erect ; glands affixed to the base 

 of the rostellum, naked, distant ; stigma in front rather large, 

 oblong or somewhat horse-shoe-shaped depressed margined ; 

 ovary subcylindrical channelled. 



Described and drawn from a living specimen sent by Mr. 

 Schlechter, as above. Colour of the perianth lilac, the lip 

 bearing two longitudinal parallel stripes, the anther a little 

 darker, outer margins of the rostellum red. Eespecting this 

 most interesting and rare plant, the reader is referred to the 

 remarks under Plate No. 26 (Pachites Bodkini) in the first part 

 of this work. Burchell's type specimen in herb. Kew is so 

 much larger and more densely-flowered than that figured that 

 one might hesitate to identify them, especially as the flowers in 

 that are rather too far advanced to give entirely satisfactory 

 results. However, a careful dissection and drawing have left 

 little doubt in my mind of the correctness of the identification. 

 Lindley speaks of the anther as "resupinate"; and it appeared 

 to me that the rostellum had fallen forward, age having altered 

 its position, as is often the case in the Ophrydeae. The genus 

 seems to connect Disa with Satyrium, but has very distinct 

 characters of its own. — I may in this place ask the reader to 

 correct an error in Plate 26, where the numbers 3 and 4 of the 

 analytical figures, by an unfortunate mistake, have been trans- 

 posed. 



