February 2,1882.] JOURNAL OF HORTICULTURE AND COTTAGE GARDENER. 87 
house, and the leaves of plante situated there are frozen. 
It will be remembered that in the autumn season about 
two years ago the Vines in the large house at Chiswick 
and several other places lost their foliage while it was 
still green, and caused no little consternation and sur¬ 
prise because the thermometers in some of the houses 
thus affected were known not to have recorded a lower 
temperature than 40°. The matter appeared simple 
enough to me. The temperature in the houses under my 
charge probably fell as low as in those where the mis¬ 
chief was done, but the Vines did not suffer. Those 
who would have solved the problem should have placed 
their registering thermometers on a frosty night as 
close to the glass as they allowed their Vine leaves to 
be, and the mystery would have been less. It was 
well known at the time that if there -was little or no 
frost registered in the air at 4 feet above the ground, 
the temperature on the grass fell considerably below 
the freezing point (there is often as much as 8° differ¬ 
ence between the surface of the ground and 4 feet above 
it), glass having very little'power to prevent radia¬ 
tion, the Vine leaves at such times are comparatively 
little better off than the grass under our feet. 
The fourth advantage I claim for my trellis is that 
the Vine shoots can be allowed to grow in a natural 
position till after flowering. Probably the flowers set 
better for this, and it is certain that tying down be¬ 
comes a very easy matter. Everyone knows how 
difficult it is to tie a young Vine shoot down, and how 
very often wflien a shoot is tied down in the day and 
seems all right we find it snapped off next morning at the 
junction with the old wood ; and not only do we lose the 
shoot that season, but there is often left a permanent 
blank. This never happens when the tying is deferred 
till after flowering, the shoots are then comparatively 
wflry and can be bent to a horizontal position with ease. 
There has been several contrivances for remedying the 
evil of shoots breaking off in this way, such as employ¬ 
ing V-sliaped trellises, slinging the rods below the 
trellis, tying down a little at a time, &c., but the 
simplest plan is the best, and that is to have the trellis 
at least 2 feet from the glass and tie the growths down 
at the same time the fruit is thinned.— Wm. Taylor. 
CHOICE IRIDS. 
LAPEYROUSIA. 
In this genus we seem to lose the prevailing characters of the 
Irid family, at least as regards their general aspect. In the place of 
a perianth with the inner and outer series of divisions differing con¬ 
siderably in appearance, the petals and sepals are similar in form, 
colour, and direction. This imparts a most distinctive effect to the 
flowers ; and .a casual observer would scarcely suspect them to be 
allies of the Irises. Though the flowers are smaller than those of 
many others in the order, they are brightly coloured in some forms, 
and are produced in sufficient numbers to render them equally as 
attractive and pleasing as many of more pretentions. Seventeen or 
eighteen species are known, but few are in cultivation, and to these 
the preceding remarks chiefly apply, as, judging by figures that 
have been published, some are far from ornamental either in form 
or colouring. They are all natives of the Cape, with small bulbs, 
and hearing racemes or corymbs of flowers in spring—May and 
onwards. They are best grown in the cool house, though L. corym- 
bosa is occasionally grown out-ide in warm sheltered positions. 
L. corymbosa. —Trobably this is the most widely known species, 
and either under the names of Ixia or Ovioda it may be found in 
many collections both in England and on the continent. This is 
partly due to its own attractions, and partly to the fact that it has been 
longest in cultivation. It is one among the numerous plants discovered 
by Thun berg at the Cape of Good Hope, where it was chiefly found 
in sandy positions in Swartland, and subsequently introduced a few 
years before the close of the eighteenth century. The specific name 
was applied in reference to the inflorescence, which is corymbose in 
form—a good distinguishing mark, as in most of the others it is 
spicate or racemose. The flowers are composed of six ovate 
segments bright blue in colour, with a white and dark blue-angled 
band encircling the central portion of the flower, and near the base 
of the segments. The angles are acute, and extend about half the 
length of the petals, thus giving a peculiar starlike appearance to 
the flowers. These are clustered on lax slender stems, the leaves 
being very narrow and tapering. The woodcut (fig. 18) conveys a 
fair idea of a corymb of the flowers. This form varies considerably 
in the depth of the colouring, and it is said there is a variety with 
fine white flowers, but I have not seen it. 
L. fissifolia. —As an example of a distinct section of the genus, 
and one which is scarcely represented in gardens, this species 
deserves a few words of description. It rarely exceeds 7 or 8 inches 
in height, with tapering stem-clasping leaves, which towards the 
Fig. 18.—Lapeyrousia corymbosa. 
upper part of the stem become small, much like foliaceous bracts. 
The flowers are produced from the axils of the leaves, chiefly near 
the apex of the stem; they have very narrow tubes 2 inches or 
more in length, expanding into six small ovate lobes, varying in 
colour from very pale pink to bright rose. They are also pleasantly 
fragrant—a quality the first-named species does not possess—and 
they are remarkable for their brittleness. It is a delicate little 
plant, far from imposing, yet pretty, and w-ell deserves a place in 
a collection of Cape bulbs. I believe it is still in cultivation in a 
few gardens, but it is rarely seen, which is to be regretted, as many 
less interesting plants have been preserved and hold a place in 
gardens. Bulbs were introduced by Messrs. Lee & Kennedy in 
1809, and an excellent coloured figure was published in the 
“ Botanical Magazine ” of the same year. 
ANOMATHECA. 
The two species which constitute this genus are nearly re¬ 
lated to the Lapcyrousias, differing chiefly in the arrangement 
