50. Hookeri. 51. Roseum.
We said that in all divisions of natural objects there were “aberrant” types which refused to be located in any one group and remained sitting on the wall between several of the divisions. And in this group there are likewise some which stick upon the wall. The chief reasons we have for not using this classification will be apparent from the following criticism of it.
The first section, Cardiocrinum, forms a very natural group. L. Cordifolium and L. Giganteum, though distinct species, are yet very near akin and are totally different from any other lilies.
Of the Eulirion section, the first or Longiflorum group, containing L. Longiflorum, L. Wallichianum, L. Philippinense, L. Neilgherrense, and the new L. Formosanum, forms as natural a division as is Cardiocrinum. But the other members of the Eulirion group are by no means so easy to classify.
L. Japonicum Odorum and L. Brownii are very nearly allied. L. Krameri, with the new L. Rubellum, more nearly resemble the Archelirion than the present group.
In certain characters L. Nepaulense nearly resembles L. Monadelphum, a member of the Martagon group.
L. Candidum bears but little resemblance to the other Eulirions. Its flowers are short and numerous, and the bulb sends up an autumn crop of leaves. In the last characteristic it differs very markedly from every known lily.
L. Belladonna is unknown to us.
L. Washingtonianum and its varieties resemble L. Candidum in bearing numerous small short flowers. Its bulb is very similar to that of L. Humboldti in being an oblique, almost rhizomatose, structure.
L. Parryi resembles L. Nepaulense in some particulars, and L. Washingtonianum in others. Its bulb, however, is more like that of L. Pardalinum than that of any other species.