AND THEIR CULTUEE. 



91 



duck's egg or larger, and formed of five or six (in very large bulbs 

 more) thick, rounded, whitish scales, which become green, suffused 

 with brown, if exposed to the light and air. These scales are semi- 

 circular or lunate in section, the cut part or interior being formed of 

 rich green cellular tissue, among which are embedded numerous lon- 

 gitudinal fibres, which terminate suddenly at the articulation between 

 the petioles of the decayed leaves and the apex of the bulb scales, 

 leaving a clean scar, as shown in the engraving. 



L. Giganteum has larger and more elongated scales of a rough 

 leathery texture, and often fibrous externally, and the articulation 

 between the petiole and the top of the scale is not a clean one, as in Cor- 

 difolium, but each scale is terminated by an irregular tuft of fibres, as 

 shown in the figure. The bulbs of both these plants have a cgespi- 

 tose (?)* or clustered habit of growth, as in many other species, but their 

 extremely thick scales, formed of 

 green cellular tissues and longitu- 

 dinal fibres enclosed in an indurated 

 or leathery, glossy or fibrous coat, 

 serve to distinguish them at a glance 

 from their allies. 



SUB-GENUS II. 



EULIRION. 



L. Longiflorum . This is a highly 

 variable plant in aboveground 

 growth and flower, but the bulb 

 structure is remarkably constant in 

 form and colour, the main points of 

 difference being the relative size 

 and thickness of the scales and 

 bulbs. The bulb structure of this 

 plant closely resembles that of the 

 common white Candidum, being 

 of the typical ovoid type, with lance- 

 shaped imbricated scales of a white 

 colour changing to pale yellow on 

 exposure. The illust ration, t on page 



L. Giganteum (India) ; one-third natural 



92, shows the contour of these better size ; from a cultivated bulb ; colour,, 

 than any description could possibly brownish. 



do. The old flowering bulb of July, 1875, which yet bears the remains 

 of the flower-stem, has, as will be seen, beeji succeeded by a 



* Hardly the correct term, the fact being that after flowering, the old bulb is absorbed, 

 leaving five or m6re offsets at its base, and these, if left alone, grow up in a clump. 



t The contour of Candidum is much more like the figure given, ptige 94, of Wallich* 

 ianum than that of Longiflorum, being longer vertically, with scales more closely 

 appressed and overlapping. 



