Having had the above Water Lilies growing side by side for five or six years, the differences appear to be so great and so constant that it is very difficult to believe the former to be a natural variety or sport from the other. The differences consist in the size and shape of the leaf, the size, shape, and number of the petals, and the colour of both leaves and flowers. These differences are as great as, if not greater than, those between most of the recognised species of the genus Nymphæa.

    I will briefly enumerate these differences. The leaves of N. alba are somewhat longer than broad; those of N. rosea are broader than long. A full-sized leaf of each chosen at random measured as follows: N. alba, 10 1/4 inches long, 8 3/4 inches broad; N. rosea, 6 inches long, 6 5/8 inches broad. Of course, both vary somewhat in proportions, but this difference seems fairly constant and gives a distinct character to the two groups of leaves. Accompanying this difference of proportion is a marked difference of outline, especially of the base of the leaf. In the white species the lobes usually overlap for more than half their length, and the lower termination of each lobe is a very slightly pointed oval. In the red form the lobes open at a very wide angle, and the termination is an oblique unequal-sided blunt point. Out of more than 100 leaves on my red Lily I can see none that approach the overlapping form of the white species, of which it is a specific character, Babington indicating it by the words "notch in the leaves with parallel sides." A Siberian species, N. nitida, is described in Hemsley's "Hardy Plants" as having "deeply lobed leaves with spreading lobes," which exactly describes those of the red-flowered form. Equally distinct is the colouring of the two leaves, which differ as follows: N. alba, above, clear bright green, below, yellowish green; N. rosea, above, an olive-green, very distinct; below, a full dull red, faintly tinged with green.

    Coming to the flowers, the differences are equally great. The botanical books give from sixteen to twenty-four as the number of petals in N. alba, but in N. rosea they are much more numerous--usually about forty. They are also broader and blunter; hence the flowers of N. rosea are smaller, with a more regular rosette of petals, often as perfect and compact as in a well- formed double Dahlia. Now it appears to me that these various and well-marked differences in form, size, proportions, and coloration of the leaves, as well as in the size, shape, and number of the petals and the coloration of the complete flower--differences which in their entirety are as great as between any two species of the genus inhabiting the same continent--do in themselves constitute primâ facie evidence that the two forms are distinct species. Of course such evidence must give way if there is direct proof that the one form is produced from the other by seed without the possibility of any crossing having taken place. The words in italics indicate the essential point needed to constitute such a proof, and I send these notes in the hope of eliciting from some of your readers what is the nature of the evidence that has led botanists, apparently without any hesitation, to class two such very different plants as belonging to the same species.

    I may add that I am now growing and flowering in a small tank out of doors, but warmed by a lamp, two blue Water Lilies--Nymphæa scutata, from Port Elizabeth, and what appears to be Nymphæa stellata, the roots of which were obtained from Diep River, about ten miles from Capetown, where they were probably introduced, though now apparently growing wild.

ALFRED R. WALLACE, Dorset.