912 
THE GARDENING WORLD 
October 31, 1903. 
of Tea and Indian Roses at nearly 1,400, and that the trade 
has produced 6,400 Roses up to the present. 
It would be interesting to ascertain, if that were possible, 
how many hybrids or improved forms of Roses are in actual 
cultivation in this country at the present day. There can, be 
little doubt, however, that thousands of Roses which have been 
raised at one time or other have disappeared from cultivation. 
The same would apply to the hundreds of Chrysanthemums 
that have made their appearance and been hailed with greater 
or less delight by gardeners' at exhibitions during the past 
twenty years 1 . A few of the greatest favourites continue to 
linger on in the gardens of their admirers. It may be that 
some of these old sorts will be resuscitated after many years, 
either under their old names, or, what is more likely, under 
new ones, when the present-day cultivators have forgotten 
them. 
The practice of propagating the above subjects by bulbs, 
tubers, and cuttings has enabled the various forms to be 
perpetuated, and it may be taken for granted that hundreds 
of these plants disappear soon, after the cultivators cease to 
propagate them. Another way by which hundreds have be¬ 
come lost occurs in the case of those things which are raised 
from seeds. Some of the hybrids soon revert to their parents, 
and in other cases hybridity would gradually be eliminated 
during a few generations of the' plants that have been raised by 
hybridisation. 
Some very extraordinary cases of reputed plant raising are 
here exploded by the author, a,s well as many foolish notions 
and ideas that have been entertained by different' raisers for 
many years past. No doubt, many of these statements have 
been made in mistake, while others have been unfounded. One 
very common opinion is that the pollen, obtained from another 
plant is supposed to determine' the colour of the flowers in the 
seedlings that are raised, either as a, result, of cross-breeding 
or hybridising, and that the foliage and habit are supplied by 
the parent. This may partly result from the repetition of 
that opinion by different writers, cr the conclusion may first 
have been made as a, result of imperfect observation. In the 
case of a few crosses certain results might, happen which might 
not be corroborated by results, obtained in other cases. 
When a large number of cases have been critically examined 
it. will be seen that both the seed and the pollen, parent may 
affect the progeny in equal or in different degrees. Some 
may show a large amount of evidence of the seed parent, while 
others may resemble the pollen parent both in foliage and 
habit. All these divergent forms may 'actually come from a 
batch of seedlings raised from one seed pod. Among these, in 
the case of fruits, one might arise having qualities that would 
make it. a desirable acquisition, while the rest may be entirely 
worthless. From these facts we should conclude that the 
raiser, however scientifically and exactly he may work, cannot 
depend upon getting exactly what he wants. 
In the case of first crosses it mostly always happens, however, 
that a number of the seedlings are exactly intermediate be¬ 
tween the parents, and the raiser, judging from that case 
alone, might say it was possible to obtain the exact results for 
which he was working. The aim of many in the use of two 
parents is to combine the good characters of both, but, although 
this may happen occasionally, the case is rare compared with 
those instances in which the seedlings would seem to combine 
the> bad or undesirable properties of either or both the parents. 
A large number of experiments are here described that have 
little but scientific interest for the gardener, but, nevertheless, 
a study of the facts here disclosed shows what can be expected 
when working upon, scientific lines. It is evident that those 
who wish to be veiy successful in the raising of useful garden 
plants must work "cm a sufficiently large scale to grow the 
seedlings extensively, to give the greater chance of obtaining 
the desired improvement amongst so many variations. Of 
course, when the seedlings are exactly intermediate between 
the parents, it may be said that the raiser has obtained what 
he has been working for. When, the seeds obtained are scanty 
the chances of a good cross are greatly limited, and after 
having obtained them it is necessary to. make provision for 
growing all of them to the flowering or fruiting stage. 
The subject of graft hybrids has also been tackled by Keraer, 
but after reviewing all of the cases available be is left in 
the Siam pi uncertain position as to whether graft hybrids do 
actually occur, or have ever actually occurred. The difficulty 
lies, in the fact that no. one has ever been able to demonstrate 
that unions have been formed by the method suggested. The 
best authenticated case is that of Laburnum Adami, which was 
produced by a, grower named Adam at Vitry, near Paris, in 
1826, who inserted a. bud of Cytisus purpureus on a stock of 
Laburnum vulgare. Several experimenters, at various times 
since then have made repeated attempts to corroborate or refute 
the idea, that the peculiar Laburnum originated in this way. 
No one has succeeded in repeating the production of Laburnum 
Adami in this way, so that tlie theory remains to be proved. 
Our readers may remember that the supposed graft hybrid 
produces buff-purple flowers as the result of the combination 
of the .purple and yellow of the original parents. This plant 
produces no seeds. Frequently, however, a bud upon it reverts 
to Laburnum vulgare pure and simple, with yellow flowers. 
This produces seeds abundantly, and is so strong in constitution 
that if allowed to have its own way it will soon form the 
bulk of the tree, and the buff-purple flowers of Adami will soon 
disappear. Other buds may revert to Cytisus purpureus, and 
these also produce seeds. In various other forms of reversion 
upon Cytisus Adami, other minor and less distinct forms of 
combination or dissociation occur, which make the plant, 
altogether an interesting case which has not yet. been equalled 
in the plant world, so far a.s we know. 
Aquatics at Gunnersbury House. 
('See Supplement.) 
Aquatics are extensively cultivated at Gunnersbury House, 
Acton, London, W., one of the seats of Leopold de Rothschild, 
Esq. It is not our intention on the present occasion to review 
the whole of them, but. only some of the more interesting ones, 
particularly those which are in bloom at the present time. 
Some years ago a. small tank was constructed outside, but 
heated with a hot-water pipe running through it from one 
of the neighbouring glasshouses. This was intended for the 
cultivation of some of the tropical Water Lilies, with the idea 
that they would succeed better under this form of treatment 
than in hothouses. The' tank is covered over late in the 
autumn, and kept in that, condition until the weather is suffi¬ 
ciently warm in early summer for the exposure of the plants. 
At the present time two fine varieties grown in the tank 
are still in bloom. These are Nymphaea stellata Berlin 
variety and N. puloherrima, another form of the blue African 
Water Lily. The flowers of this latter form are larger, and 
the sepals mottled with dark brown spots on the back. The 
leaf stalks are also, shorter. It stands outside better than 
N. stellata., and is the hardiest of all the blue Water Lilies, 
because it does not get cut about by the wind. 
The same two varieties are also grown in a more recently 
constructed tank covered with a frame. The Berlin variety 
commences to flower in April. William Stone is an American 
variety with dark blue flowers. Nymphaea gigantea Hudsoni, 
of which we gave a supplement some time ago, is still in bloom, 
though the flowers are now only about 64 inches across. In 
the height of its season the blooms measured 104 inches 
across, making a veiy distinct and very handsome Lily indeed. 
It is the only Nymphaea which produces seeds here. Large 
numbers of seeds have been saved, and seedlings are already 
furnished with a few leaves. These seedlings will commence 
to bloom in the course of twelve months from the time of 
sowing. 
Like William Stone, Nymphaea Husteri is also an American 
variety, and belongs to the .tropical section. It has deep rose 
flowers and rounded or blunt petals faintly tinged with violet 
on the apices. In dull weather the flowers remain open all 
