IIX THE GENUS CROCUS. 



Stamens barely an inch (0.025 metre) in height; the Anther orange, a little shorter than the yellow 



Filament, which is about half an inch (0.013 metre) in length. 

 Pistil about an inch and a quarter (0.032 metre) in height from the throat, higher than the stamens; 



the Style dividing high up into spreading orange stigmata, which are generally entire, but occasionally 



divided. . 



Scape at the flowering-time barely an eighth of an inch (0.0032 metre) high, the ovary being almost 



sessile. 

 Capsule and Seed unknown. 



Crocus lazicus is not yet in cultivation; and is only known from the herbarium 

 specimens distributed by Monsieur Balansa, who discovered it in the year 1866, 

 in moist meadows in the alpine region of Lazistan, above the villages of Djimel 

 south-east of Trebizond, latitude 40° 40 north, longitude 40 45' east, at an altitude 

 of 8500 feet. Some doubt exists as to the month in which Monsieur Balansa 

 gathered it in flower, though August is named on his labels. In a communication 

 with which he has favoured me, he casually mentions June as the time of his visit 

 to Djimel. C. lazicus has the aspect of a vernal species; and no other early autumnal 

 species developes leaves till the spring. 



M. Balansa informs me that he reached Djimel by way of Rizas, a small sea- 

 port, twenty-five or thirty miles east of Trebizond; and after going inland to Andon, 

 ascended through forests of Abies Nordmanniana, and thickets of Rhododendron 

 caucasicum, across a bare alpine ridge, at a height of between 10,000 and 11,000 

 feet, covered with snow. On the southern side of this ridge he descended on Djimel, 

 and found C. lazicus by the side of a zigzag path high above the villages, and about 

 1800 feet lower than the ridge he had crossed from the north. 



C. lazicus has no near ally; and is remarkable for its almost sessile ovary, the 

 scape at the flowering- time being scarcely one-eighth of an inch in height. It is 

 the only species in which the almost abortive sheathing leaves are shorter than 

 the basal spathe, which is exposed; and with the exception of C. nudiprus, it is 

 the only Crocus in which the corm is stoloniferous. 



REFERENCES TO PLATE XII. 



Fig. 1. Flowering-state, actual size, August (June?), from specimens in Monsieur Boissier's Herbarium. 



Fig. 2. Pistil, magnified six-fold. 



Fig. 3. Stamens and Pistil, magnified two-fold. 



Fig. 4. Diagrammatic dissection of scape, ovary, and spathes, actual size. 



Fig. 5. Section of leaf, magnified six-fold. 



