74 JOURNAL OF THE ROYAL HORTICULTURAL SOCIETY. 
modification of such ordinal types as NaiadacecB, Ccratophyllacece, and 
Podostemacece makes it a matter of the very greatest difficulty to the 
systematist to place them satisfactorily in his scheme of classifica- 
tion ; so, though it is probable that they are descended from marsh 
plants which were gradually driven by the struggle for existence to adopt 
a purely aquatic mode of life, it would seem that this was at a far more 
remote date in these cases than in others. There are perhaps barely fifty 
species that discharge their pollen under water. The well-known case of 
Vallisneria spiralis, in which the pistillate flowers are carried up to the 
surface by long rapidly-growing and spirally-coiled peduncles while the 
staminate ones break off at the bottom and rise to the surface before un- 
folding, is an obvious adaptation of a plant of terrestrial ancestry not 
only to water, but to water of varying depth. When, as is so often the 
case, w^e find aquatic genera or sub-genera, such as Hottonia and 
Batracliium, in orders otherwise terrestrial, we should a priori assume 
that their aquatic adaptation is of a more recent date ; and the Water 
Crowfoots are a striking example of the puzzling variability or instability 
of character that distinguishes such modern evolutions. That great 
modification in response to changed conditions may occur within the 
limits of a single species we have already seen in several instances. In 
the case of the Brooklime {Veronica Beccabunga, L.) the leaves when 
submerged form only four or five layers of mesophyll cells instead of ten or 
twelve : the Flote-grass {Glyceria fluitans) produces submerged leaves three 
times as long as, and rather narrower than, those grown on land : similar 
differences are recorded in the cases of Elatine, Castalia, and others;* 
and the Bur-marigold {Bidens), a water- side plant, when growing in the 
water is stated to lose all woodiness and to form air-canals in its cortical 
tissue. The most rapid recorded case of adaptation, however, is, perhaps^ 
that of a Common Daisy {Bellis perennis, L.) recorded by Mr. G. 
Leslie, E. A., in that charming work "Letters to Marco." This plant, 
which normally has its leaves spreading horizontally on the ground,, 
during a few days of river flood which covered Mr. Leslie's lawn at 
Wallingford raised them into an almost erect position, as if emulating a 
submerged Stratiotes or Lobelia. 
That we can show such adaptation to be progressive, I will only 
illustrate by the case of Potamogeton cited by Schenck. P. natans, L.,, 
the least modified or most ancestral, has its upper leaves ovate, leathery, 
and floating, and the lower ones submerged and sometimes linear. It. 
dies down in winter to the rhizome. P. heterophylhts, Schreb., has- 
all its submerged leaves narrow. P. lucens, L., P. crispus, L., &c., have 
all the leaves submerged and lanceolate, and form winter buds. P. 
obtusifolius, Mert. and Koch., P. pusillus, L., &c., have long, narrow,, 
ribbon-like leaves, all submerged ; and P. trichoides, Cham, and Schlecht.,. 
with a thread-like stem and bristle-like leaves, is the most highly modified 
of all. 
This example, or the genus as a whole, serves to illustrate the varying 
directions which the adaptation takes, according to the conditions. The 
leaves of aquatics must, in fact, be grouped under four very distinct types 
— the floating, ribbon, much-divided, and subulate types. Of these the 
Kerner, op. cif. ii. 505. 
