Mirch 2(', 1890. J 
JOURNAL OF HORTICULTURE AND COTTAGE GARDENER 
235 
i;erm was, however, at one time applied to the reverse crossing 
already noted, and hybridising is a sufficiently general term to be 
fairly applicable to either case. In Primula sinensis and its 
numerous forms we have examples of extreme variation artificially 
•produced without the aid of another species, but the variations 
obtained under cultivation are so pronounced that crossing has been 
^resorted to with surprising results in increasing the number of 
forms and tints. One of the largest cultivators of these and some 
similar plants assures me in making crosses between two varieties 
■of Primula sinensis he always expects the seed parent to give three 
out of every four characters, and it is this fact which renders it 
difficult to produce distinct alterations in any type, as it is only 
either by chance or by a long series of experiments that a desired 
result can be secured. Another hybridiser who has been engaged 
for some years in the production of Primula seed, has found that 
the seed parent generally gives the foliage and habit, and the 
vpollen parent the floral colour. With regard to this I proved from 
experience a few years since that in crosses between the “ Fern¬ 
leaved ” and ‘‘ Palm-leaved ” varieties, when the former was made 
f he seed parent there was a much larger per-centage of seedlings 
with “Fern leaves ” than if the pollen was used on the “Palm¬ 
leaved” varieties, thus supporting the opinion just given. It has 
also been observed with regard to Primula sinensis, that the pollen 
from a double variety placed on a single variety was more likely to 
■give a good proportion of double flowered seedlings in the progeny 
fthan if the parentage were reversed. As an example of that the 
pollen of an excellent double variety with flowers of a moderately 
good crimson colour was employed to fertilise a single variety of 
«.n exceptionally bright tint, and the result was a double variety 
with flowers equal to the pollen parent in form and substance and 
■as bright as those of the seed parent. As exceptions to this, how¬ 
ever, it must be noted that in two instances where doubles have 
I)een crossed on singles, all the seedlings came single, and in the 
reverse cross about 10 per cent, were double. 
In experiments with Balsams I have found somewhat similar 
results to those described—namely, pollen from a single flower 
applied to a double flower produced a much larger per-centage of 
-double-flowered seedlings than where the single variety was made 
'the seed parent ; but when pollen could be obtained from another 
plant of a double variety for application to the double seed parent 
the result was still more satisfactory in the number of double- 
flowered plants produced, though the colours were often less bright 
'than in the parents, and this loss of colour was not so noticeable 
when a single-flowered pollen plant was employed. 
With Zonal Pelargoniums I made a number of experiments 
■some years ago, partly with the object of testing the relative effects 
•of the respective parents, and though the results were somewhat 
'diverse, yet amongst hundreds of seedlings raised quite 80 per cent, 
more closely resembled the seed parent in habit and foliage, and 
■the pollen parent in flowers. Mr. Peter Grieve, who performed 
good service amongst the variegated Pelargoniums, has stated that 
dn his opinion the constitution of a plant is chiefly due to the seed 
plant, and he attributes the delicacy of many of the Pelargoniums 
named to the employment of the plants with variegated leaves as 
seed hearers. He further says, “The most vigorous and in all 
fespects the best I have raised have been from green-leaved seed 
parents.” This we can easily understand, but my observations 
toach me that considerably more depends upon what may be termed 
the individual characters of the seed plant than upon the specific 
or varietal distinctions. Thus the selection of a healthy plant for 
•seed-bearing is likely to give better results than a weakly individual 
of the same species, whatever the cross may be. 
It has been stated with good reason that where so much 
difference results from reverse crossing it is in a large measure due 
to diversities of form or constitution of the pollen, and in some of 
'the most variable cultivated plants the microscope reveals a sur¬ 
prising variation in the form of the pollen grains. Certain plants, 
however, possess an inherent capacity for variation through a 
greater range than others, and in two that have been noted— i.e., 
Primula sinensis and the Balsam, all the forms have been originated 
by intercrossing without the aid of other species. Phlox Drum¬ 
mond! might be cited as another instance of a similar kind, all the 
variations having been formed within the limits of one species. 
It appears that under cultivation the constitution of a plant gradu¬ 
ally becomes disturbed ; different soils and conditions, with perhaps 
uncreased vigour, produce effects that are most likely to be felt in the 
•delicate floral organs, though probably more evident in the foliage 
and growth. To increase the tendency to variation, therefore, 
plants should be grown under as diverse conditions as possible con¬ 
sistent with their health, and when the crossing is confined co the 
forms of one species the most distinct should be selected, except 
where some particular character is desired, when close in-breeding 
will sometimes produce the wished for result. 
P.\RTI.\.L FeRTILI.S.'VTIOX AND STERILITY. 
Hitherto attention has only been directed to plants in which 
there has been little difficulty in securing seed ; the progeny have 
presented a fusion of characters more or less complete, and have 
been equally fertile, or nearly so, with their parents. Next to these 
we have the hybrids or crosses, which are themselves either partially 
or wholly sterile, except when fertilised with pollen fromanothir 
variety or species. Examples of this occur in nearly every family 
of plants where hybrids have been produced, and even amongst the 
descendants of one species, as in the highly bred forms of Primulas 
previously mentioned. Cyclamens, and other plants. In some of 
the brightly coloured single Primulas much difficulty is experienced 
in securing seed for this reason, as even when abundance of pollen 
is produced, as happens occasionally, it seems useless except when 
employed on other varieties. Close intercrossing often induces 
sterility in the seedlings, and is commonly attended by a greatly 
increased vegetative growth ; a consequence as some suppose of the 
sterility, but probably in certain cases it is the cause, for it is well 
known thtit excessively luxuriant plants are rarely such good 
seed bearers as those of a medium or even stunted growth. There 
is another form of partial sterility, this time in the parent plants, 
when perhaps it is possible to make one plant the seed parent in 
a cross, but not to reverse the cross and secure seed. Something 
after this kind took place in the production of the hybrid Rhodo¬ 
dendron Princess Alice, which was obtained by two cultivators 
fromR. ciliatum fertilised with pollen from R. Edgeworth', yet 
Mr. Anderson-Henry has stated that he repeatedly tried to secure 
seed from R. Edgeworth! by employing pollen of R. ciliatum, but 
failed every time. 
Sometimes in the results of hybridising one species with another 
some of the seedlings show altered characters partaking of the two 
parents, and others closely resemble one only of the parents. A 
peculiar instance of this kind has been recorded, in which Begonia 
manicata was crossed with B. coccinea, and one pod of seed was 
produced. Seedlings were raised, and amongst them were two 
distinct forms that were afterwards named B. prolifera and 
hybrida, but a large proportion were the true B. manicata. In this 
case it would seem that either the pollen of B. coccinea had 
sufficed to impart vitality to the ovules without altering tlmir 
character, or that some of the ovules had been self-fertilised, either 
being equally remarkable. 
Imperfect Fertilisation. 
The last form of fertilisation that need be considered here is 
that in which the pollen only influences the ovary to induce an 
enlargement, the ovules not being affected in any degree, or not 
sufficiently to render them fertile. Dean Herbert recorded many 
instances of this kind, and amongst them he mentioned that when 
Alstroemeria aurea is fertilised with pollen from other species full 
sized capsules are produced, but no perfect seeds. Fruits have 
been obtained on Passifloras in the same way, but seedless, and 
numbers of other instances could be given. 
Turning to the Orchids we have as regards the Odontoglossums 
two curious facts to consider. One is that numbers of Odonto¬ 
glossums are found in a wild state so exactly intermediate in floral 
form and colouring between other well known species as to leave no 
reasonable doubt that they are natural hybrids ; they have been 
introduced freely, and have puzzled botanists and horticulturists 
not a little. The other fact is, that notwithstanding this evident 
variability and tendency to intercrossing in a wild state, the Odontc- 
irlossums have not yielded to the cultivators’ attempts to obtain 
hybrids under artificial conditions, indeed they have stubbornly 
resisted all efforts, and have caused more disappointment than any 
other Orchids. Seedlings have been raised in a few instances, but 
invariably lost, and the strangeness of this circumstance induced 
me a few years ago to make some experiments with a view to 
testing the matter. I had several healthy plants of Odontoglossums, 
and amongst them were six each of two favourite species—0. Pes- 
catorei and 0. triumphans. They were in good condition for 
experiment, flowered well, and if a cross could be effected they 
gave a promise of something worth securing. Yellow forms of 
the 0. Pescatorei type had been introduced, and one charming 
variety, for which Mr. Biownlow D. Knox obtained a first-class 
certificate from the Royal Horticultural Society on April 13th, 
1886, impressed me greatly, and the impression was intensified 
when the same plant realised £165 at a public auction the next day. 
There was little doubt that 0. triumphans or 0. tripudians, most 
likely the former, had conferred the yellow colour, and it did not 
seem too much to expect similar results from an artificial cross. 
Three of the strongest plants were selected of 0. Pescatorei and 
of 0. triumphans, and two flowers of each were reciprocally fer- 
tilised-'i e , the pollen of 0. Pescatorei was placed on 0. triumphans, 
and vice versa, thus twelve flowers were fertilised, and to my great 
