268 THE GENUS CROCUS. 



Perianth- Tube from two to two and a half inches (0.050-0.063 metre) in length from the ovary to the 

 throat. Throat unbearded, orange. Segments from an inch to an inch and a quarter (0.025-0.032 

 metre) long, and from a quarter to a third of an inch (0.0063—0.0084 metre) broad, white; the 

 outer surface of the outer segments either self-coloured white, or suffused or feathered with purple. 



Stamens from half an inch to five-eighths of an inch (0.013-0 016 metre) high, reaching to the level 

 of the stigmata; the orange Anthers a little longer than the orange Filament. Pollen Grains ^ 

 of an inch (0.00005 metre) in diameter, channelled with sinuous depressions. 



Pistil from half an inch to five-eighths of an inch (0.013-0.016 metre) high from the throat; the Style 

 dividing at the level of the base of the anthers and produced into about a dozen capillary divisions 

 of the orange stigmata, which reach nearly to the level of the summit of the anthers. 



Scape about an inch and a quarter (0.032 metre) high at the flowering-time, and produced in April to 

 a height of four or five inches (o. 100— 0.125 metre). 



Capsule purple, three-quarters of an inch (0.019 metre) high, and one-third of an inch (0.084 metre) 

 broad. 



Seed bright crimson, glabrous, oblong, one-fifth of an inch (0.005 metre) high, and °^ te . n ^ of a , n 

 inch (0.0025 metre) broad with a prominent caruncle of the same colour as the body of the seed. 



The history of this species is somewhat remarkable. It is a native of The 

 Troad, 26^° east longitude, and 39^° north latitude, and was discovered there by 

 Dr. E. D. Clarke, the English traveller, in the spring of 1806, on the flanks of Mount 

 Gargarus, (Kaz Dagh,) one of the heights of Mount Ida. Clarke's specimens how- 

 ever had been lost sight of for more than half a century; and in their absence 

 Monsieur Boissier assumed the identity of Clarke's C. Candidas with C. Fleisckeri 



of Gay. 



In the autumn of 1879, I had the good fortune to find them without a name, 

 and merely labelled "Troas" by Dr. Clarke, in the Webb Herbarium at Florence, 

 —unquestionably the plant described in a foot note at p. 145, vol. 2, of his Travels. 

 By a curious coincidence, in the spring of the same year I met with specimens in 

 the Herbarium of the Edinburgh Botanic Gardens, collected by Dr. Kirk in March 

 and April, 1856, at Renkioi, near the Dardanelles, from the Limestone Hills about 

 two miles inland Dr. Kirk tells me. These were also without name, and believing 

 at the time they belonged to a new species, I described them in the Gardeners' 

 Chronicle as C. Kirki, a subsequent comparison with the Florence specimens however 

 identified them beyond doubt with Clarke's C. candidus. 



The corm tunic, leaf structure, and spathes point to its affinity with C. aureus, 

 but the fine capillary stigmata, and flower-colouring clearly separate it as a distinct 

 species; and I must dissent from Herbert's view of its being merely a variety of 

 C. lagenaflorus of Salisbury (C. aureus, Sibth. and Smith.) I am indebted to Mr. 

 Maling, Her Majesty's Vice-Consul at the Dardanelles, for having sent several 

 expeditions to Renkioi in search of it, though without result. I have received from 

 Herr Ascherson, of the Royal Botanical Museum, Berlin, specimens of a Crocus in 



