Ds

i

ae Be
ay AG it

eh

nba

cha

mhiak ANNALS OF
APPLIED BIOLOGY

tak OFFICIAL ORGAN OF THE ASSOCIATION
OF ECONOMIC BIOLOGISTS

EDITED BY
E. E. GREEN, Way’s End, Camberley (late Government Entomologist, Ceylon)

AND
Proressor B. T. P. BARKER, National Fruit and Cider Institute, Bristol
Dr S. E. CHANDLER, Imperial Institute, London
F. J. CHITTENDEN, Royal Horticultural Society’s Gardens, Wisley
J. C. F. FRYER, Board of Agriculture and Fisheries, London
Proressor F. W. GAMBLE, The University, Birmingham
Proressor PERCY GROOM, Imperial College of Science and Technology, London
Dr A. D. IMMS, The University, Manchester

ProFressor R. NEWSTEAD, The University, Liverpool
Proressor J. H. PRIESTLEY, The University, Leeds

Volume III 1916-17

CAMBRIDGE UNIVERSITY PRESS
C. F. CLAY, Manacer
LONDON: FETTER LANE E.C. 4
EDINBURGH: I00, PRINCES STREET
also
H. K. LEWIS & CO., LTD., 136, GOWER STREET, LONDON, W.C. f
WILLIAM WESLEY & SON, 28, ESSEX STREET, LONDON, W.C. 2
PARIS: LIBRAIRIE HACHETTE & CIE.

CHICAGO: THE UNIVERSITY OF CHICAGO PRESS
BOMBAY, CALCUTTA AND MADRAS: MACMILLAN & CO., LTD
TORONTO: J. M. DENT & SONS, LTD.

TOKYO: THE MARUZEN-KABUSHIKI-KAISHA

VV uf Q ? iy

Cambridge:
PRINTED BY J. B. PEACE, M.A.,
AT THE UNIVERSITY PRESS

CON TEN TS
No. 1 (June, 1916)

The Fig “Canker,” caused by Phoma Cinerescens Sacc. By
K. 8. SatmMon and H. Wormatp. (With Plates I and II,
and 1 Text-figure.) :

Shrinkage, Swelling and Warping of Grose: tad Woods: No. L
Yang (Dipterocarpus sp.). By Percy Groom. (With a
Diagram.) : ; : : . 5

The Dalby Profile eee ‘By W. E. Datsy, M.Inst.C.E.,
F.R.S. (With 7 Text-figures.) : : : : :

The Action of Enchytraeid Worms. By the Rev. Hiuperic
FrienD, F.R.MS.

The Food of Slugs and the Development of Wnelocepialides
By Proressor A. RatLurer : : 2

Nos. 2 and 3 (January, 1917)

A Bacterial Se of Citrus. By Eruet M. Dore, D.Sc., F.L.S.
(With Plates III—XIIL.) ;

Report on a Trial of Tarred Felt “ eee i ie Patostiig
Cabbages and Cauliflowers from attacks of the Cabbage-root
Fly. By J. T. WapswortH. (With Plate XIV.)

On the Resistance to Fungicides shown by the Hop-mildew
(Sphaerotheca humuli (D.C.) Burr.) in different stages of
Development. By E. 8. Satmon. (With Plate XV.)

Observations on the Larval and Pupal Stages of Agriotes
obscurus, Linnaeus. By Grorce H. Forp, M.Sc. (Vict.).
(With Plates XVI and XVII and 1 Text-figure.)

On the Biology and Economic significance of Tipula paludosa.
By Joun Renniz, D.Sc., F.R.S.E. Part II. Hatching,
Growth and Habits of the Larva. (With Plates XVITI—XX
and 3 Text-figures.) ‘ :

Note on Attacks of Phyllotreta vittula on ope hee

49

52

53

82

93

97

116

iV

1:

2.

3.

Contents

No. 4 (April, 1917)

Accessory Wetting Substances with Special Reference to
Paraffin Emulsions. By A. H. Less, M.A. d

Empusa Muscae versus Musca domestica L. By H. T. Gissow.
(With Plate X XI.) . P : ;

A Blossom Wilt and Canker of pon Teed By H. WorMaALD,
M.Sc. (Lond.), A.R.C.Sc. (With Plates XXII—XXIV.)

PAGE

141

150

159

VoutumeE III JUNE, 1916 No. 1

THE FIG “CANKER,” CAUSED BY PHOMA
CINERESCENS Sacc.

By E. 8. SALMON anno H. WORMALD.

Mycological Department, South-Eustern Agricultural College,
Wye, Kent.

(With Plates I and II, and 1 Text-figure.)

In 1914 complaints were received by the South-Eastern Agricultural
College at Wye of serious diseases affecting the plantations of fig-trees
in the district of Sompting, Sussex. On a visit being paid in May
1914 to the affected plantations, fig-trees of all ages were found to be
suffering from fungous attacks of two kinds. A species of Botrytis
was found on the tips of the branches, apparently gradually killing them
back. This disease is still under observation and will be the subject
of a further communication.

The second disease which occurred was a ‘‘canker”’ of the branches.
In some cases, where old trees were badly attacked, numerous “cankers”’
were found, both on the younger branches, and on the old, main branches,
often quite close to the ground. With the progress of the disease, the
“canker” area enlarges, until the tissues extending through the branch
are killed, with the result that the parts above die. With the successive
removal of these ““cankered”’ branches the productiveness of the tree
is soon seriously impaired. The majority of the trees in these plantations
are of the variety known as “Brown Turkey,” but a few are of the
“White Marseilles” variety. It was obvious from the effects produced
that this “canker” disease was of serious economic importance; it was
the opinion of the farmer of the largest fig-plantations that unless the
cause of the disease could be discovered and a remedy devised the whole
future of Fig-growing in that district was threatened.

The constant occurrence of a fungus with pyenidial fructifications
on the cankered area was noted in the field, and the following notes
were made from the examination of the material collected.

Ann. Biol. 1 1

2 Fig “Canker, caused by Phoma cinerescens Sace.

One of the “‘cankers” which may be taken as typical of those found
on the larger branches is shown in the photograph at Fig. 1, Plate I,
where the characteristic cracking of the bark in the older portions is

seen. It will be observed that near the centre of
the “canker” is the old base of a smaller branch
and it is probable that that was the place where
the fungus made an entrance. This particular
canker was evidently the result of three distinct
periods of, activity of the causal organism. The
oldest portion, recognized by the fact that the
bark is very much cracked and separating from
the wood, was elliptical and measured 16-5 x 5em.
At each end of this ellipse the canker extended
5 em. upward and 2-5 cm. downward, these ex-
tensions corresponding to the second period of
activity. The check to the normal growth in ©
thickness of the branch induced by the presence
of the organism caused these affected areas to be
depressed below the general surface of the branch.
The youngest portions of the canker extended
still further upward and downward and though
they were but indistinctly marked off from the
adjoining healthy portions of the branch they bore
numerous pycnidia. The accompanying diagram
of the canker shows the areas, affected during the
three periods of activity, numbered respectively 1,
2 and 3; this diagram should be compared with
Fig. 1, Plate I, which is a photograph of the same
canker.

The portion of the branch bearing the “canker”
was placed in a damp chamber and in twenty-four
hours “tendrils” of conidia were issuing from
the youngest portions of the canker but not from
the rest. No conidia were obtainable from the
older portions and there the fungus was either
dead or had ceased to produce conidia, although
these areas were still dotted over with old pyenidial
pustules. The tendrils were at first of a pale
orange colour but when fully protruded were almost
or quite white. When a tendril or a portion of one is placed in a drop

E. S. SauMon AND H. WoRMALD 3

of water the mucilaginous matrix binding the conidia together is dissolved
and the conidia themselves stream apart and become diffused through
the water; a slight “Brownian movement” is to be detected when the
conidia are mounted in water.

Transverse sections through the bark show the fungal fructifications
to be pycnidia. These bodies are more or less circular in outline and
are from 250» to 600 u in diameter; they are somewhat flattened and
are therefore lenticular, appearing elliptical in transverse section. The
pyenidia are produced a little below the surface, but on approaching
maturity each develops a short neck which ruptures the outer layers
of the bark and the tendril emerges through an apical pore. The wall
of the pyenidium is lined with the conidiophores which are from 15
to 25 in length and which abstrict the conidia from their apices.

The conidia are continuous, ellipsoidal to fusiform, often with one
extremity more rounded and broader than the other. Their dimensions
are 6-5-13 x 2:4-3-6 4. Usually they are about 9 x 3 and as a rule
an increase of length is associated with a decrease in the width; thus
the following dimensions are typical: 6-5 x 3-6, 9-2 x 3, 11-5 x 2-5,
13 x 2-4. With medium magnification they appear to be biguttulate,
but by employing an oil-immersion objective this appearance is seen to
be due to two groups of minute guttules. The two groups may merge
into one another but usually they are quite distinct and situated at
opposite ends of the conidium (see Fig. 10, Plate II).

The conidia are capable of germinating in water. When placed
in a hanging-drop of distilled water and examined after remaining
three days at the temperature of the laboratory (about 18°C.) a few
of the conidia had germinated. They showed no appreciable increase
in size before the protrusion of the germ tubes. The latter were at
this stage one to four times the length of a conidium and frequently
were more or less geniculate; they emerged laterally, sometimes at
or near the middle of the conidium, at others more towards one or the
other extremity, but none was seen truly polar.

In June of the same year (1914) Mr F. J. Chittenden sent us a
specimen of the Fig “canker” from the glass-houses at Wisley; this
bore pycnidia and conidia similar to those described above.

The pycnidial form of fructification and the spore-characters referred
the fungus to the genus Phoma, and reference to Saccardo’s Sylloge
Fungorum enabled us to identify it with P. cinerescens Sacc. in
Mich. 1, p. 521 (1879). The specific diagnosis given is as follows: “ Peri-
theciis gregariis, globoso-depressis, atrolivaceis, subcutaneis; sporulis

1—2

4 Fig “ Canker,’ caused by Phoma cinerescens Sace.

fusoideis, biguttulatis, 6-8 x 2-2-5, hyalinis. Hab. in ramis corticatis v.
demum decorticatis Fici Caricae, in Italia et Gallia. Spermogonium
Diaporthes cinerescentis Sacc.”

To make certain of its specific identification, an example was sent
to Prof. P. A. Saccardo, who replied, “J’ai examiné votre specimen.
C’est sans doute mon Phomopsis cinerescens. Ma diagnose est incom-
pléte, car les sporophora (comme il se fait trés-souvent) étaient déja
disparus; aussi les gultulae dans votre specimen sont peu distinctes,
mais cela est aussi variable.” The author who transferred the present
fungus from Phoma to Phomopsis was J. B. Traverso (in his Flora
Italica cryptogama, vol. 11, fase. 1, p. 278 (1906)), who after giving a
description of Diaporthe cinerescens Sacc. merely remarks: “Status
pyenidicus verisimiliter Phomopsis cinerescens (Sacc.) Trav. sporulis
fusoideis, hyalinis, biguttulatis, 8 x 2.” In general appearance and
structure the present fungus agrees so well with other parasitic species
which are still named Phoma that we retain it in that genus.

In searching for records of the occurrence of a Fig-tree “canker” in
England, we met with the description given by Mr Massee first in the
Gardeners’ Magazine, for July 23 (1898), and later in his Teazt-Book
of Plant Diseases, p. 431 (1903). In the former place Mr Massee wrote:
“A disease presumably of old standing has of late years proved very
injurious to fig-trees, and one remarkable feature in connection with
this disease is the fact that it is most prevalent and destructive in those
cases where the trees have received the greatest amount of attention,
pruning more especially favouring its extension. The most usual
symptoms of its presence are a cankered or ulcerated appearance of
the bark, which frequently becomes eaten away in large patches, or
variously cracked. In the majority of cases it is very evident that
the canker first starts at a pomt where a branch has been cut away
or accidentally broken off, and in all instances it appears that a broken
surface of the bark is absolutely necessary to enable the fungus causing
the disease to gain a foothold.” While the general description of the
disease given here agrees exactly with what we have found, the descrip-
tion of the fungus which Mr Massee gives as the causal organism is
quite different from that of Phoma cinerescens. Mr Massee places his
fungus in the Melanconiaceae (“perithecia absent; conidia produced
on a more or less developed cushion or stroma formed beneath the
surface of the matrix, and becoming erumpent’’) as a new species of
the genus Libertella, viz. L. ulcerata Mass. In the Gardeners’ Magazine,
L.c., it is stated: “Finally the fruit of the fungus, which is formed in

E. S. SALMoN AND H. WorMALD 5

minute pustules, below the epidermis, bursts through to the surface,
the exceedingly minute spores, only about 1-3,000 of an inch in length,
oozing out through minute cracks in the epidermis, under the form of
very slender, white, gelatinous threads or tendrils.” The figure given
of these spores is reproduced at Fig. 11, Plate II. It will be seen
from Fig. 11, traced from Mr Massee’s figure of the Fig canker Libertella
in the Gardeners’ Magazine, that the length of the conidia is about
1-3,000 of an inch (7.e. 8-9 1), or approximately that of the conidia of
Phoma cinerescens (compare Figs. 10 and 11, allowing for the difference
in magnification). In A Teat-Book of Plant Diseases! Mr Massee in
describing his species under the name Libertella ulcerata Massee (sp. nov.)
says, “Conidia fusiform, ends acute, continuous, curved, hyaline,
55-60 x 4 py,” and this description is repeated in his Diseases of Cultivated
Plants and Trees®. This discrepancy in the accounts of the dimension
of the conidia of Libertella ulcerata we are unable to account for. Our
fungus, z.e. Phoma cinerescens, belongs to the Sphaeropsidiaceae (“ peri-
thecia containing conidia borne at the tips of slender conidiophores ”’)
and its conidia are ellipsoidal to fusiform, straight or slightly curved,
ends usually rounded, averaging 9 x 3, and never exceeding 13 pw in
length.

The Herbarium at Kew was found not to contain the type specimen
of L. ulcerata Mass. In March 1915, Mr Massee kindly forwarded to
us an example of his fungus, writing “Enclosed is a fragment of the
type specimen,” and adding as a postscript, “I am almost certain that
the Libertella is followed by a Phoma stage, but have not been able to
get one from the other in cultures.” The portion of the type specimen
sent consisted of some hundreds of fructifications, mostly with dried-up
tendrils of conidia still attached, and proved to be entirely Phoma
cinerescens, with all the characters as described above. A thorough
search of this type material showed no conidia resembling those described
by Mr Massee, or any fructification of the type found in Libertella.

With reference to the parasitic nature of his fungus, Mr Massee
states: “Experiments conducted on healthy young fig-trees show that
the spores of the fungus—a species of Libertella—-will not cause the
disease when placed on the unbroken surface of even very young
branches, whereas when the spores are placed on the end of a cut branch
or on injured bark, inoculation always followed, and the mycelium was
found in abundance at the expiration of 10 days. In one experiment

1 Massee, G., A Text-Book of Plant Diseases, p. 431 (1903).
2 Idem, Diseases of Cultivated Plants and Trees, p. 448 (1910).

6 ig ‘“ Canker,” caused by Phoma cinerescens Sace.

a badly diseased branch, showing numerous threads of spores, was
cut through; immediately afterwards the same knife was used for
making an incision in the bark of a branch of a healthy young plant of
Ficus religiosa. At the expiration of 10 days the wound showed un-
doubted symptoms of disease and at the end of five weeks the white
threads of spores were found. This experiment proves that the disease
may be imparted to healthy plants by using a knife that had previously
been used for pruning diseased plants, and an examination of various
diseased plants suggests the idea that this method of spreading the
disease is not an unusual one.”

Since the above work was not conducted with a pure culture of
the fungus, and especially since we had found associated with the
Fig “canker” disease a quite distinct fungus from that described
by Mr Massee, it was important to obtain scientific evidence of the
parasitism of Phoma cinerescens.

It was found that pure cultures could be obtained by two methods:

(1) <A portion of a tendril obtained from the canker shown in the
illustration (Plate I, fig. 1) was placed in a drop of sterilized distilled
water; after dilution a small drop containing conidia was transferred
to a tube of melted agar (prepared with an extract of prunes as nutrient)
and a “poured plate” made. One such plate produced six sporelings ;
growth was rather slow and after eleven days the mycelial masses only
measured from 4 to 7 mm. in diameter. From each of two of these a
tube culture on an agar slant was prepared. In the tube cultures growth
continued with more vigour and in seven days each showed a rather
dense, flat, hyaline disc of mycelium about 2 cm. in diameter with
a raised white ring of aerial hyphae midway between the point of origin
and the periphery. Later these slant cultures consisted of a rich
growth of dark brown mycelium with a few whitish tufts.

(2) An alternative method of obtaining cultures of the fungus
was also adopted. The canker referred to above was cut across and
transverse sections made through the discoloured wood. In this wood
the vessels were found to be blocked with hyphae, tyloses and “ wound-
gum.” Two small fragments of such sections made with a sterile
razor were each transferred by means of a flamed platinum wire to an
agar slant. The resulting growth resembled that obtained under
similar conditions with mycelium produced from the poured plate of
conidia, the white ring of aerial mycelium at a short distance from the
point of inoculation again being a characteristic feature. The inference
is that the mycelium in the affected tissues of the xylem is that of the

E. S. Saumon and H. WorMALD v6

fungus seen on the exterior and indicates that it is possible to obtain
a pure culture directly by transferring fragments of infected wood to
a culture medium.

The medium used in these primary cultures was a decoction of
prunes containing 1} per cent. agar and although it was found to be
quite suitable for vegetative growth, no reproductive bodies were
produced under those conditions. Sub-cultures made on sterilized
cylinders of potato in Roux’s tubes and on slabs of wood (obtained
by making transverse slices? of a branch from a fig-tree) placed in Petri
dishes and autoclaved at 115°C. for 30 minutes, readily produced
pycnidia and “tendrils” of conidia. On potato the pycnidia were
numerous and in some cases well-developed “tendrils” emerged, while
in others the conidia accumulated at the mouth of the pyenidium in
a globular mass, the latter condition obtaining when the moisture within
the tube was too great for the typical tendrils to retain their characteristic
form. The “tendrils” and globules of conidia were at first of an orange
colour changing later to a dark red. The conidia were similar to those
obtained originally directly from the canker in their size, shape and

euttulation.
On the slices of fig wood the fungus grew vigorously as a white
mycelium and eventually produced a few scattered “tendrils”: the

latter were larger (stouter and usually longer) than those normally
produced under natural conditions and were dark red in colour, the
conidia however were similar to those obtained from the canker itself.
It is interesting to note in this connexion that the “tendrils” were
more numerous on the under surface of the wood, 7.e. where it came in
contact with the imner surface of the Petri dish, than on the upper
free surface.

INOCULATION EXPERIMENTS ON FIG-TREES.

Young fig-trees in pots were obtained for inoculation with pure
cultures of the Phoma and were kept in a greenhouse throughout the
experiments.

Experiment I. In this preliminary experiment a plate culture of
the fungus was started on June 8, 1914 as a sub-culture from one of the
tube-cultures mentioned above as obtained from a “poured plate”
of conidia. At the end of eight days (June 16) there was sufficient
mycelium for use in the following inoculations:

1 These slices were 4—5 cm. diam. and about 1 cm. thick.

8 ig “ Canker,’ caused by Phoma cinerescens Sace.

(1) A V-shaped cut was made through the bark of a branch of one
of the young fig-trees; the triangular tongue of bark was turned back
and a little agar bearing mycelium was removed from the plate culture
and inserted between the bark and the wood. The free portion of bark
was then pressed back into its place and the wound was covered up
with cotton wool and tinfoil to prevent possible infection from air-borne
spores. This precaution was perhaps unnecessary for as shown in
Exp. II exposed wounds remained uninfected.

(2) From another branch a small portion of the bark (about 1 em. x
0-5 cm.) was entirely cut away and agar with mycelium placed on the
exposed tissues, after which the wound was bound up as in the previous
case.

On September 16 the coverings were removed and the following
results observed :

(1) Round the inoculated wound was a depressed very dark elliptical
portion of bark 2-5 x 1-2 em.; the tissue bordering the canker was pale
in colour and thus showed strong contrast with the sunken canker.
The latter bore a few pycnidia which appeared on the surface as small
protuberances with incipient “tendrils” consisting of conidia which
on microscopic examination were indistinguishable from those obtained
from natural cankers.

(2) The result of the second inoculation was similar to the one just
described except that the canker was smaller, its dimensions being
1:8 x 0-8 em.; tendrils with typical conidia were again produced.

In both cases the infected branch was seen in September 1914 to
be slightly swollen when viewed in a direction perpendicular to the
surface of the canker. No further increase in the size of these cankers
was subsequently observed and in the following season both branches
bore fruit and leaves. In February of the present year (1916) leaves
and fruit again appeared but by the middle of March the secondary
branch immediately above the canker of the second inoculation, and
on the same side of the main branch as the canker itself, showed signs
of wilting for the leaves began to droop and turn yellow, indicating
that the transpiration current was interrupted in the neighbourhood
of the canker.

Experiment II. Semicylinders of potato were placed in Roux’s
tubes and sterilized in an autoclave by steaming at a temperature of
115° C. for 30 minutes; they were inoculated from an agar slant culture
on June 22,1914. The fungus grew vigorously on the potato; pyenidia
were readily produced and by July 15 “tendrils” of conidia began to

E. S. Saumon AND H. WorRMALD 9

appear. Later the latter were further developed and on September 29
were used for a second series of inoculations on fig-trees. Two of the
“tendrils” were removed on a sterile hooked platinum wire and placed
in a small tube of sterile distilled water which was then agitated in order
to diffuse the conidia throughout the liquid. Wounds are made on
branches as in Exp. I, but instead of inserting agar and mycelium, a
drop of the water containing the conidia was transferred to the wound
by means of a pipette (previously sterilized) and allowed to flow in
between the bark and the wood. In the case of control wounds a
drop of sterile water only was inserted in a similar manner. Some of
the wounds were bound round with damp cotton wool (previously
sterilized in autoclave) and tinfoil secured by raffia in order to ensure
favourable conditions for the germination of the conidia: the other
wounds were left uncovered. In other cases again water containing
conidia and fragments of tendrils was dropped on damp cotton wool
which was then bound round certain internodes with tinfoil and raffa
as before, but without injuring the bark.

Hight of the largest branches on two young fig-trees, var. White
Marseilles, were selected and treated as follows, the branches being
from 0-8 cm. to 2 cm. in thickness.

No. 1. Inoculated; wound covered with cotton wool and tinfoil.
As in 1.

Inoculated; wound left uncovered.
Control; not inoculated but branch cut and wound covered
with cotton wool and tinfoil.
» 0. Control as in 4.
6. Control; wound left uncovered.
,, 7. Spores placed on damp cotton wool which was then wrapped
round an internode.
= One AS in ie

Kight branches were similarly selected on two young trees of
the Brown Turkey variety and respectively treated as those of
the White Marseilles. Thus in this experiment six branches were
wounded and inoculated, with an equal number of control branches
also wounded, and in four cases conidia were applied to uninjured
branches.

To test the viability of the conidia used in the experiment a drop
of the conidia-containing water (as used in the inoculations) was used
in preparing a “poured plate’’ which was then placed in the greenhouse
in close proximity to the trees. No growth was to be observed in the

99

= wo be

10 Fig ‘“ Canker,” caused by Phoma cinerescens Sace.

plate for several days but on October 9 (7.e. when ten days old) numerous
small mycelial masses were conspicuous.

Positive results were obtained in each case where the conidia had
been inserted in the wound. ‘The rest all yielded negative results.

The trees were examined from time to time and the progress of the
disease in the inoculated branches noted as follows.

White Marseilles.

No. 1. (Branch about 1 em. diam.)

Jan. 15, 1915. No change to be observed.

Feb. 16. No visible canker: no discoloration of the bark but “tendrils” were
seen in the neighbourhood of the wound: they were arranged in the form of an
irregular ellipse 2-2 em. x 1-4 cm., elongated in a direction parallel with the axis of
the branch.

Jan. 25, 1916. Distinct sunken canker 4 cm. long and extending laterally
three-quarters round the branch.

No. 2. (Branch about 2 cm. diam.)

Jan. 15, 1915. No change to be observed.

Feb. 16. ‘5 es As

March 25. Asingle tendril of typical conidia was seen at 2 mm. above the wound ;
otherwise there was no change externally.

April 20. The bark round the wound was slightly depressed below the general
surface; the affected area measured 1-5 x 1-5 cm.

Jan. 25, 1916. Distinct sunken canker 1:5 cm. long and extending one-third
round the branch.

No. 3. (Branch 1 em. diam.)

Jan. 15, 1915. No distinct canker (i.e. no sinking or cracking of the bark) but
round the wound and extending 1 cm. above and below and 0-5 cm. laterally were
numerous small protuberances; one of these pustules removed and mounted in
water proved to be a pycnidium containing conidia as described above.

Feb. 16. Tendrils were issuing from the pustules which now covered an area
3°5 x 1-8 em.

Jan. 25, 1916. Distinct sunken canker 5 cm. long and extending three-quarters
round the branch.

Brown Turkey.
No. 1. (Branch 1 em. diam.)

Jan. 15, 1915. No change to be observed.

Feb. 16. 5 55 -

Mar. 3. A rather indistinct zone of discoloured bark (darker than the rest)
was observed at 3-5 ecm. from the wound.

EK. 8S. Sanmon AND H. WorMALD ti

Mar. 25. A single tendril of typical conidia was found at 2 mm. from the point

of inoculation.
Jan. 25, 1916. Distinct sunken canker 2cm. long and one-third round the

branch.

No. 2. (Branch 0-8 em. diam.)

Jan. 15, 1915. No change to be observed.

Feb. 16. The bark was discoloured for 2-3 mm. round the wound; at 4mm.
from the point of inoculation was a small group of pustules.

Mar. 3. A tendril was observed projecting from one of the pustules ; it was
removed and microscopic examination showed conidia as before.

Jan. 25, 1916. The affected area, now 2 cm. long and extending three-quarters
round the branch, was only slightly depressed below the general surface but was
distinguished from the rest by its darker colour.

Feb. 24. Branch dead above the canker, as shown by absence of any growth,
the rest bearing leaves and young fruit.

No. 3. (Branch | cm. diam.)

Jan. 15, 1915. No external change.

Feb. 16. No canker and no discoloration noticeable but round the wound was
an irregular circle (about 1 em. diam.) of tendrils.

Jan. 25, 1916. The affected area appeared as a slightly sunken canker 2 em.
long and extending half-way round the branch.

Thus in the two series of experiments eight inoculations through
wounds were made in all, and in every case not only was there a “‘canker”’
formed round the point of inoculation but the fruiting fungus appeared
in the affected tissues and produced tendrils of conidia resembling those
found on “cankers” arising from natural infections. Since ‘“‘cankers”
were so readily induced when the fungus, either in the form of conidia
or of mycelium, was introduced into the internal tissues through wounds,
the negative results in those cases in which conidia were applied to the
uninjured bark indicate that in all probability the fungus is solely a
wound-parasite.

ee

Preventive Measures.

From our knowledge of the biology of the parasite, it is very unlikely
that spraying with any fungicides will hold the disease in check. The
direct measures should consist of a search for all “‘cankers,’’ and the
cutting of them out, down to the sound tissues, at a time when the
disease is dormant; it will be advisable to paint over the wounds made
with Stockholm tar. All dead branches must be cut out and prompily
burned, as otherwise the fungus producing conidia on the dead wood
might constitute a dangerous source of infection. The indirect measures

12 Fig “Canker,” caused by Phoma cinerescens Sace.

which should be employed—and these will be no less important than
the direct measures—must consist of the prevention as far as possible
of wounds to the bark. The making of wounds by the boots of men
climbing in the branches when gathering the figs must be rigorously
prevented, as well as those caused by careless hoeing or digging round
the trees or due to the attacks of animals. The minimum amount
of pruning to the older branches should be employed until the disease
has been stamped out or reduced to a small amount.

DESCRIPTION OF PLATES I], Il
PLATE |

1. Canker on a large branch of a fig-tree. 2 natural size.

2. Lower half of the same canker, natural size, showing the small wart-like protu-
berances produced by the pycnidia.

3. Section through the bark of a canker; the pyenidial form of the fructifications
is evident. x60.

PLATE Il

4. Portion of a canker with “tendrils.” x 4.

5. Transverse section, natural size, through the middle of the canker shown in Fig. 1;
the affected tissues were dark brown in colour.

6. Phoma cinerescens in pure culture on sterilized potato, natural size.

7. Two “tendrils” of Phoma cinerescens produced in pure culture on a slice of sterilized
fig wood. x4.

8. Branch of fig-tree after inoculation with conidia taken from a pure culture, natural
size; an irregular ring of “tendrils” at some distance from the wound is shown.

9. Another branch after inoculation with conidia; the protuberances produced by
the pycnidia are seen around the wound.

10. A camera-lucida drawing of conidia of Phoma cinerescens; each conidium is
provided with two groups of guttules. x 1000.

11. Drawing traced from Mr Massee’s illustrations, in the Gardeners’ Magazine, of
the conidia of the Fig canker Libertella. ~ 500.

PLATE

eNO:

VOL

THE ANNALS OF APPLIED BIOLOGY.

THE ANNALS OF APPLIED BIOLOGY. VOL. III, NO. | PLATE Il

PLATE

Il, NO.

VOE

THE ANNALS OF APPLIED BIOLOGY.

*
’
is
a
.
7
pa
'

.

as? 7 to i:

13

SHRINKAGE, SWELLING AND WARPING OF CROSS-
GRAINED WOODS: No. 1, YANG (DIPTERO-
CARPUS sv.).

By PERCY GROOM,

Professor of the Technology of Woods and Fibres,
Imperial College, London.

(With a Diagram.)

No detailed investigations appear to have been conducted on the
changes of dimensions and shape occurring during the wetting and
drying of cross-grained woods. Yet such research bears upon problems
that are of scientific as well as industrial importance; for many timbers
growing in hot countries are cross-grained, that is to say, their fibres
and other constituents in successive concentric layers round the
longitudinal axis are directed alternately in right-handed and _left-
handed spirals. Moreover, a number of such timbers are of com-
mercial importance.

The statistics available show that when straight-grained woods dry
from their fresh (green) condition to an air-dried (ordinary seasoned)
state the percentage shrinkages of their rods are as follows when cut:
longitudinally, usually -1 (-01—-8); radially, usually 3-5 (1-1-6);
tangentially, usually 6-10 (1-8—10-5).

At first thought it might be anticipated that a cross-grained timber
would show increased longitudinal but decreased transverse shrinkage
or swelling, because of the existence of transverse and longitudinal
components respectively. The results obtained do not bear out such
an anticipation.

In this investigation the Indo-Malayan cross-grained commercial
timber known as Yang, Dipterocarpus sp. (possibly D. twberculatus),
was used. When first imported into England it was tested as a
substitute for teak (in window-cills, doors, etc.), and was sometimes

14 Shrinkage, Swelling, Warping of Cross-grained Woods
sold as “yang teak,” but the results were not satisfactory. In-doors
the wood twisted as it dried; while out-of-doors when it absorbed
moisture yang was described as speedily showing on its surface a “ white
growth” and as undergoing early decay. During the progress of this
investigation the practical problems of preventing these defects were
more or less solved, and a digression on the cause of the disfigurement
and decay out-of-doors may not be out of place.

Submersion of yang boards under water causes quantities of white
substance to appear at intervals over the surface and especially at the
ends. This substance is mainly resin that is forced out by the swelling
of the woody tissue. I examined a few minute quantities of the sub-
stance dissolved in methylated spirit, and in each there occurred a
microscopic fungal mycelium. When the sodden yang wood was
allowed to dry the white substance remained as a disfiguring chalk-like
exudation. When, in turn, a very little methylated spirit was rubbed
over the surface of such dry wood, the disfigurement vanished as the
resin dissolved and covered the surface with a film of varnish (recalling
French polish in its effect). Practical trial alone can show the extent
to which this treatment will preserve this wood from decay, or be of
service in connexion with other resinous dipterocarp woods.

In the investigations the pieces of yang wood used were in the form
of radially and tangentially cut boards, about 48-61 cm. in length,
12-13 cm. in width, and 1°35 cm. in thickness. In order to induce
and measure the swelling they were submerged under water and
measured at intervals. Thereafter they were allowed to dry at ordinary
temperatures in the laboratory, and measured at intervals. The first
measurements were made by means of a wooden rod graduated only
to millimetres, but subsequently use was made exclusively of vernier
callipers accurate to -02 mm. The changes of curvature were recorded
graphically by an instrument which Professor Dalby was so kind as
to design for the purpose. The instrument is described in a separate
paper, in the present issue.

GENERAL RESULTS.

The most interesting results concerned the changes in length of the
wood during the absorption and emission of water. In both cases
a shortening and a lengthening took place. During the absorption
of water the radial and tangential boards showed successively :

First, an increase of length together with relatively small increase
of width. This latter, in the one case tested, was most marked near

P. Groom 15

the ends of the board, because the water was entering mainly through
these.

Second, the boards actually shortened while at the same time they
continued to widen.

Third, once more lengthening took place and was accompanied by
widening.

During drying the reverse phenomena wholly or in part were
exhibited; for the shortening at the beginning was succeeded by
elongation, which was final in one case (yang 2) but was followed
by lengthening in the case of yang 1; in all these cases the board
continued throughout to shrink in width (excepting for occasional slight
deviations).

Yet throughout these contrary changes in length the board
continued to increase in surface during the absorption of water, and
to decrease in surface during drying. And, in connexion with the
explanation of the phenomenon, it is worthy of note that the board
(tangential, yang 2) which showed the smallest ultimate decrease in
length and most persistent elongation during drying yet experienced
the greatest decrease in surface.

The most probable explanation of the reversal in change of length
during the absorption of water appears to be one akin to that offered
by Professor Brereton Baker, who suggested the analogy of a net-work
or lattice-work in which swelling takes place more rapidly along the
length than at right angles to this. In this case the meshes formed
by the crossing vessels, fibres, and so forth are lozenge-shaped, narrow,
and elongated along the axis of the tree-trunk (and board).

If we imagine two fibres or tracheae ABC and DBE intersecting
at B, and consider the two sides BC and BE of the triangle CBE, then
as water is absorbed by the board it is taken in most rapidly at the
ends and travels more rapidly (in the vessels) along ABC and DBE
than in any other direction. The result is relatively considerable
elongation of BC and BE, with relatively inconsiderable increase in
the angle CBE. Increased distribution of the water at right angles
to BC and BE next causes more rapid expansion in those directions,
tending to cause the two sides to rotate outwards about the centre B
and thus cause a widening of the base of the triangle but a shortening
of the perpendicular from B to that base. This would also take place
if the sides elongated to a disproportionately small extent. Thereafter
elongation of the two sides associated with disproportionately less
expansion at right angles would cause a reversal of the immediately

16 Shrinkage, Swelling, Warping of Cross-grained Woods

preceding phase. Through these three phases the board named yang 2
passed, but during the final stage of elongation there were fluctuations,
which possibly were due to the successive transverse belts or zones of
the board at different distances from the ends being at different phases,
and to irregularities in the rate of distribution of moisture in the board.
Dotted lines in Diagram 1 illustrate, on a very exaggerated scale, the
paths traversed by the points C and E during the absorption of water.

Diagram 1

When the board is drying and shrinking the reverse changes tend
to take place, as drying is most rapid at the ends and presumably so
along the lines of vessels. Yang 1 (tangential) went through such
changes completely, but yang 2 (tangential) and the radial board
showed merely the initial decrease and the succeeding increase in length ;
both associated with decrease in width. Dotted lines in Diagram 1
represent on a very exaggerated scale the paths of the points D and A
during drying.

P. Groom 17

There is no absolute certainty that during soaking the boards had
attained their maximum dimensions, though it is probable that such
was at least approximately the case. The tangential board, yang 1,
had certainly attained its greatest length; and judging by the lengths
of the diagonals and widths so likewise had yang 2. When, as a
result of drying, the boards had shrunk to their minimal recorded
dimensions, these may be assumed to represent the true minimum
sizes in relation to the surrounding atmosphere (dry air of the
laboratory).

The changes in length were small or exceedingly small ultimately,
being only -012—-075 % of the maximum length. The changes in
width at the ends were: tangentially, 5-92-6-29 °% during drying,
and 6-98 °% (in the one case measured) during soaking; radially, during
drying 5% of the maximum width.

The subjoined Table gives the percentage changes of the (maximum)
linear dimensions.

Tangential Boards Radial Board
a SO S —— _ A rn
Yang 1 Yang 2
a OO ..
Water absorbed Waterexhaled Waterexhaled Water absorbed Water exhaled
313 grammes 307 grammes 209 grammes 316 grammes 331 grammes
In length:
maximum — ‘O75 -030 — -037
ultimate ‘079 “063 -012 —- -020
In width:
at ends 6-980 5-920 6-29 (5-5) 5:
middle 7-400 5-980 — = =

The shrinkage in both length and width was less in all cases than the
previous swelling, and this was so even when (in the radial board)
more water was lost than had originally been absorbed. This permanent
expansion of pieces of wood after soaking in water appears not to have
been recorded, and it remains to be seen whether or no straight-grained
woods show the same behaviour. That some change takes place in
wood by the prolonged action of water at high temperatures (steam)
is well known, and reveals itself in permanent decrease of elasticity
and strength.

Although during drying and soaking respectively lengthening and
shortening may take place, the area of the faces of the board always
decreases and increases respectively during those processes. So much
does the change in width correct any reverse change in length that the
board yang 2 when drying lost in area to a greater extent than did the
other boards despite of the fact that its ultimate decrease in length was

much the smallest.

bo

Ann. Biol. 111

18 Shrinkage, Swelling, Warping of Cross-grained Woods

One marked character of at least some cross-grained woods is their
lability to twist during wetting and drying, and the tangential boards
of yang displayed this character. The two tangential boards of
yang showed on their faces the grain running obliquely apparently
in the same sense in each board, but in opposite senses in the two
boards. Thus on each face of yang | the grain tended to run from the
right hand to the left hand down the board, whereas in yang 2 it
tended to run from left to right downwards. Considering either of
these boards on one face the one diagonal running rather with than
across the grain changes relatively slightly in length as the board becomes
wetter or drier, whereas the other diagonal on the same face changes
more in length becoming concave when drying or convex when wetting:
the diagonals superposed on these on the reverse face behave in an
opposite manner. The result is that each face of the board becomes
convex along one diagonal, and concave along the other; that is to say
a twisting movement results from the establishment of a couple. That
this explanation is correct is confirmed by the fact that the twist is in
the opposite sense during the swelling and drying of the same board;
and is in the opposite sense in the two boards yang 1 and yang 2 when
these are undergoing the same kind of change of water contents, whether
this change be drying or wetting.

Moreover if such be the true explanation of the twisting a radially
cut board should not exhibit this behaviour. The radially cut yang
board showed no trace of twisting either when wetted or dried. These
facts suggest, on the one hand, that possibly the twisting of so-called
straight-grained boards in general is due to a less marked and differential
obliquity on the two faces, and on the other hand that cross-grained
woods now excluded from commercial use by reason of their twisting
propensities could be utilized if they were cut into radial (“ quartered,”
“rift-sawn”’?) boards. The facts also emphasize the advisability of
seasoning cross-grained woods not only thoroughly, but also slowly,
the latter because the extent of the twist is largely influenced by the rate
of drying of the surface.

In a transverse direction the tangential boards at their ends
remained nearly straight, the very slight amount of curvature possibly
conforming with the moderate degree of tangential shrinkage or swelling.

OBSERVATIONS ON YANG 1.

This board was tangentially cut, its length and width slightly
exceeding 61 em. and 12-85 em. respectively.

P. Groom 19

The two faces were labelled y and Y respectively, Y and Ya, yand yx
denoting the four corners of what may be termed the top of the board,
y and Y being at the same edge or corner: similarly Yb and Ya,
yb and yxb denoted the four corners at the bottom end of the board.
Consequently the lengths of the board on the two faces are represented
by measurements y. yb, yw.yxb, Y.Yb, Yu. Yub, of which y. yb
and Y. Yb represent the same margin of the board on two different
faces, as do yx. yxzb and Yaz. Yab. The diagonals were thus Y . Y2b,
Ya. Yb, yx. yb, and y . yxb.

In this cross-grained board on each face the grain runs from the top
towards the bottom in a direction from right to left hand, though on
face Y the grain is to a considerable extent nearly straight on the lower
(bottom) half.

Observations. Set A (Table A).

The first measurements of yang 1 were made on a board, already
partially seasoned in the timber-yard, during its drying in the laboratory.
As a simple measuring rod was used the longitudinal changes in length
were too small for accurate estimation, but the changes in length of
the diagonals and in width were sufficiently considerable for approximate
measurement and are recorded in the accompanying Table A.

In accordance with the direction of the grain the shrinkages along
the two diagonals Y. Yxb and yx. yb were greater (-2 and -15) than
along the other two diagonals, so that the board became concave along
the former pair and convex along the latter pair. The two ends of the
board were thus twisted so as to lie in different planes, but they never-
theless retained their transverse straightness approximately.

TABLE A.
Diagonals in centimetres Width in centimetres
Weight -_——— “A ——_—_—_—_—_—_. > N —~
Date in grammes Y. Yab Yifides MAa) y .yxd yo. yb Top Bottom
April 28 761-4 60-7 59-7 61-1 61-1 12-9 12-85

s1130 751-2 —_ 59-65 61-05 61-05 12:8 12-7
May 3 — 60-6 59-6 61 61 12:7 12-6
Ese SU) 733-54 60-55 59-625 61-05 61 12-75 12-6
a IO 738-85 — —- —- — 12-75 12-6

June 1 725°35 60-5 59-6 61 60-95 12-7 12-525

Total
difference 36-05 0-2 0-1 0-1 0-15 0-2 0-325

20 Shrinkage, Swelling, Warping of Cross-grained Woods

Set B (Tables Br and B11.)

The next set of measurements (B) were made on the board as it was
compelled to swell by submersion under water: vernier callipers and
Professor Dalby’s instrument being used for the observations.

The main results as evidenced by the statistics given in Tables Br
and Bu, and as proved by the graphic records (unpublished) of
the curves, are:

The absorption of water at first causes a sudden relatively consider-
able increase in length, which is relatively more marked than the increase
of width even at the ends. This is succeeded by a decrease in length,
which is synchronous with a more rapid gain in width.

The water enters most largely by the ends; the gain in width is
marked here at first, and only after some time does the middle of the
board display a similar full rate of widening.

As water continues to be absorbed the board, after undergoing its
peculiar shortening, gradually lengthens and attains its maximum length
even before it has absorbed the maximum amount of water, after which
one or more times a shortening was noticeable. These contrary
fluctuations as regards length are probably to be attributed to the fact
that as the water penetrates from the ends the transverse belts at
increasing distances from the two ends successively go through the
shortening and lengthening phases, so that the measured result as regards
the whole board is the algebraic sum of a number of items that are
unequal and at times opposite in sense.

Widening of the board continued as long as water was absorbed,
but the slightness of the increase in width towards the conclusion of
the soaking strongly suggests that the maximum width had been very
nearly reached.

Table B 1 records the changes in linear dimensions of yang 1 during
the absorption of water. The results of these measurements are arranged
in Table B 11, by grouping together measurements made at short intervals
on the same day, and by taking the averages of the longitudinal, trans-
verse end, and transverse middle measurements.

In reference to these Tables it was conceivable that two sources
of misinterpretation were possible. On the one hand curvature or
straightening of the board might cause a spurious appearance of
shortening or lengthening respectively. On the other hand the interval
necessarily intervening between the weighing of the board and the
recording of the measurements and curves would lead to a certain degree

+ROOM

of

‘€.9+ pUe G-O8g oI” sIOqtUNU poyeUITyso OAWVARdUAOD oY, 4
*Sp1Od0d [VUISIIO OY} JO SUOTIRIO}[R 94} 9DUOY
'¢Q- Jo oovd url g. osvo YORI UT papsOdIAI T YVY} ONS Joo} [ ING HZE-09 PUe ‘FOC-09 ‘ZZE-09 *C-09 SB popazodod AT[UISIIO OOM SIOGUINU OSOYT, »

{ SuUOoIS
PL TL c8:9 L80- GLO0> cs “uetuIp [euy oy, Jo
( gouoIOyIp o5vjUI0I0T
000-1 G96° CE6: €90: 9F0- S6-G1E gouoloHtp [840],
900: SLE“ 800 og: “ €00-- &F9-“ 100-- g¢s80d-“ 0: GLO: SLT 9-FEOT 14 Oe pa
LGO: GLP“ FIO: F6F-“ 910° 879°“ GO00-- 980:“ 100:- GLO“ G08 T-LIOT SOIT ¢ “
ae | |
-o1 SJUOTHOINSveTT Udyy ‘*)LG'T 100:-— ctP“ ZOO--— OSt-“ ZO0O-- ZE9:‘° ZO0O:— 8sd:-‘ g00:-— EL0°“ §-6T 9-986 L Cals
38 poysiom ‘“eT‘OT 4e ate | |
Suryey-oAino pue sjusmoeInseeyy) 900- OFF: “* OT0- CSP: “* F00- Fe9g-** S00: 060: ** #00: 910: “* CLOL [WA
ih 0: OFF“ ZO00'- GLP“ 0: 0g9: “ 0: G80: Z00:- ZLO-“ ieee aes
ulese [
poumsvoul udyy ‘UeyeZ somo | |
usy} “peysiea oy} pornsvoyy) 900: OFF: £00: FLE“ 00: 0€9:““ ¥00:- 80°“ 200: FLO: G-6% 6-196) OZ IT 6% “
is GI 4" (
unseq SsJUsMIeINsRou pPoMoUoI | 0: Fer“ ZOO-- OLF:“ ZOO-— 829: 910--— 980: T00-—- ZLO>“ T-8&6 Alea
Udy} “CF TT 48 poysioa ‘eT ‘OT 4v | |
unSeq ‘0}9 soAdno ‘syuomommseoyy) FOT- FEF“ OFO- ZLE“ O10 O89“ 8ZO- ZOT-“ C10. 20°“ 9-9F STOI 83 “
900- 0&8“ 900-- 92%“ ZO0-- 029: ZOO: FL0-“ ZOO-- 8g0-“ T-OI- 168 Of et “ “
FI: = HEE“ 800- cE“ ~F00-— GZ9-“ GIO ZL0-“ G00- 090°" 0-82 9-106 ST'Il $3 “
0cO- O8T-“ 00: et“ ZOO: 969: 600-— 090°“ §00- seo-“ 4L-- 49-€L8 O¢e “ “
0€'€ 78 po

“YSTOM Ayjeuy pure ‘uaye} soammo
‘poamsvom (Ge'z Ajoyeutxoidde

ye) Woy} ‘podsromqns OT'ZI 1O9FV SCZ O9T-E1 FLO- GIF“ FO0-— #29“ TIO: 690°“ 0 ZO. FL- EFL8 OTZT “ “
a —= Gaines) — — 60: 890°" €10- c90°~ 1-68 L188 STII Fo “

sovjd yoo} Surysioa

usyM ‘ec, TUN ponuTyuoOD
pue CT'¢ 4B UNsoq syUoMIOIMsvoTY 9FO- G06: C20: SFE“ FOO. 829“ FOO FEO“ G00. 6e0-“ TT 9-278 STE “ “
F9G- = 98.“ EOF —OTE-ET GLI- #69: “ OFO-— O80-“ 9E0--— 980°“ GBFP F118 STIT & “
080: 69: C6I- L98-“ FI9- GSF-EI ZO +¥9L0-“ G20: %cL0-“ ST-69 SaL6L STS “ “

sanoy §Z 10F
10¥A UT pooeyd quowoInsevoul JoIJY F8O- ZO“ GZI- B99“ SST. BER“ ZO: 4FCO-“ FZO- 4090 CFOT T88L O1 2 “

OL “GG “IA
T4un ossnoy-joy wv jo are duep
oy} Ul 4jo] JUoMoINseoM JOIJyY — SLFZI — OFfZIT — OILZI- — 28019 — 920-19 — OS9-1GL O€'ZT To ‘TA
9oua SIPPIN =: 90a quh* qh 90u0 wht 9oud qh h 9oue Qh wh goua »=soulmIeIs UL IMOyT aye
“AO BIT “AOU UCT “AO HTT “AO BLT ~AOHT “AOTC. WUS19 AA
SS ey eo —+-———_——_—_—
jeguasuey, [eulpnylsuo0'T

‘Waon buiqsosqp ‘Ig AAV,

=

S
&
~~

~
Av)
~
~
~

ng of Cross-gra

Warpi

ig,

Shrinkage, Swell

. »)
—_

‘UOTJOOATP [LUIPNALSUOT & UT 4LY} SOUT] 1G SI UOTJOOITp JeIUISUeY v UT “yUd god SuToAs oY,

€1-0 90-0 080-928

L8§-0 992-0 6F8-SZ8

GE:0 GE-0 899-FE8

¢0-0 é10-0 PES‘ FES

10-1 1¥-0 8L0-FZ8

GS 960-0 EL8-618

62° ¥6:0 €68-F18

90-F &P O8§-F08

&% GP 90G-98L

LO-¥ 8-7 PLO-GLL

= = GL6-S9L
pavog ejoya = spula ygpra
ed ODVIOAB

JO YJPIA oseIdAv x x YUE]
yysue, Aq pozeut = Roly

-I}80 UoYA “poqrosqe
IOYVM JO SOTUMMLIS OT
rad wore JO asvaIOUT

L

62-89

110-668
€06-8¢8
680-828
FE9-LE8
0Z9-LE8
816-Sé8

TOL-Gé8

169-Gé8
L9€-€08
S1P-8LL
16F-OLL
spua
ayy jo

qapra
ISRIOAT

x Yysuory

SUOISUOUIIp [eu

PL ae 86:9 = 610: = _ ~ (oy} Jo edvquooi9g

001 — ST6- — 6F0- _ 1g — soUdIOYIP [VIOT,
900: SLF “* ZOO’ 29°“ T00-— 8Z0-“ GLI 94801 — | ade
LZ0- SLE“ P10: TLe-“ 800-—- 620°“° G08 TLI0I — Ge
GOO: Str“ 900° xz9¢-“ 00: Z80-“ €61 9-986 — I ‘UA
900: OFF: “ 100: ee“ Z00-— LLO-“ Z6S = $-L96 — Gar =:
LOI: 8%“ Gzo- oge-“ 10: 620°“ G:F 1-886 — sc.5
LST- LZe-“ 200: gece: 200° 990°“ ¢¢-2¢ ¢¢-968 — age:
16%: OLL-El 70. e209“ ¥20- 690-% S6Te PL8 Hoe +
L8z- —-6L8: “ GZe- 6LF-“ 680:- sEo- “ StF &0-ZFS8 — ez
0f0- 26S: OF PST-SI 320: FL0°“ ST6E GZ-L6L SIZ a
780: agg: “ onl: SOSh. — “20 cco: GP9L T-SEL (ie coes
— = SLF-T — 29:21 — 620-19 — ° &9-1ZL te Cura
9oueleyIp 9oua1ayIp 9soueIeyIp ules Inoy] aye

“wd UT “Wd Ul “UD UL YQoUeTT «= “SURI UT YYSIO A,

O[PPIU JO WPI SPue Fo WPI

“11g ATAVY,

P. GRoom 23

of inaccuracy as to the exact amount of water present in the board at
the time of the records. Both these objections are shown in the suc-
ceeding remarks to be groundless as regards the essential results obtained.

Rate of loss of water during measurement and recording of curves.
The board was first left in very damp warm air for nearly 24 hours
(vi. 21-22) ; on June 22nd during measurement and so forth from 10 a.m.
to 11.45 a.m. it lost water at the rate of 1-7 grammes per hour. In
this case the water was wholly hygroscopic and in the cell-walls. On
other days in summer in the dry laboratory air the board lost water
during measurement etc. in most cases at the rate of 7-2 grammes per
hour (vi. 22 from 2.15 to 3.5 p.m., and from 3.5 to 3.30 p.m.; VI. 23
from 11.15 a.m. to 12.5 p.m.; vi. 25 from 11.15 a.m. to 12.30 p.m.),
though on one day (vi. 24 from 11.15 a.m. to 12.15 p.m.) the rate was
8:1. In these cases it may be assumed that the loss of water was largely
at least at the expense of that present in the lumina of the wood, and
at the surface of the sides. These figures render possible an approxi-
mately accurate estimate of the amount of water present in the wood at
the middle of the time occupied in recording measurements and curves
on any day.

In regard to the estimates of the amount of water in the board and
the length the following cases may be discussed. On June 24th
observations were made in the following order: at 11.15 a.m. the board
was weighed (881-7 g.), then curves and measurements were taken
until 12.10 p.m., when the board was once more weighed (874-3) and
a second set of curves and measurements were immediately taken.
When these last were concluded the approximate weight would be
874-3 —7-3 = 867 g. The board was then submerged until 2.30 p.m., |
when curves and measurements were made before it was weighed
(873-69) at 3.30, so that at 2.30 p.m. its approximate weight was 880-9.
But whether these corrected or the recorded weights be accepted the
changes in linear dimensions are not proportional to the change in the
amount of water contained. During this phase these dimensions are
determined rather by the exact distribution than by absolute amount
of the moisture, probably because the water is largely passing from
the lumina into the cell-walls whether the wood be soaking or drying.

Again on June 28th, 29th, and July Ist, to begin with the measure-
ments and curves were taken, then the board was weighed, after which
a second set of measurements and curves were made. This second set
therefore represents the wood in a drier condition (11, 7-3, and 12-5
grammes of water respectively having been lost) so that the linear

24 Shrinkage, Swelling, Warping of Cross-grained Woods

dimensions are smaller than in the first set. Accordingly the average
between the two sets of measurements represents more accurately the
linear dimensions at the recorded weight.

The subjoined Table Bim reports the changes of curvature
lengthwise of yang 1 (during the absorption and exhalation of water)
along the two sides y. yb and yz. yxb.

= eT Oe
Absorbing water

Drying

TABLE B 111.
y . yb yx . yxb ]
a SSS SSS SS a a ee ar es I SS SS SSS =~
Date Endy Middle End yb Date End yz Middle Endyxb |!
Vi, 21 2 vs c (2) c Wise (vs) c (3) c |
22 11.40 Stocs c (3) Cc go 22 PLO US: torcs) 63) c
Be b222o c c (5) c » » 2.30 (curve defectively taken)
Poa) INE cs c (4) c » 23 11.30(cirea) (vs) oo cs
5 4.10 cs c (4-25) Cc sou iss eULo v vs vs to (cs)
24 11.46 Sn Sn Sn », 24 11.30 S S S
4 BSW) Sn Sn Sn Hy ates S S S
25) es Sn Sn Sn Ay ay NMA S S S
28 11 to 12 (vs) to(cs) cs(-75) — ,, 28 11.45 S S S
29) EDS S S S » 29 11.50 S S S |
1 11.10 v(-9) s s
5 11.40 v(-75) 3 MSU's 1s Ss S Ss |
7 1117 ~~ v (2:25) (2) v wy 11.265 S S S
9 12.53 (A wavy concave to » 9 12.45 S (vs) S S
straight line)
May PAG c c (1-5) Cc pe lowl 20 Cc c (3-25) Cc
15 11.57 c ¢ (1-5) c » 15 11.50 cc (3:25) c
LG 7, fe c (1:5) Cc cl Gmdes() c c (3-25) Cc

The first four columns refer to y.yb, the others to ya. yxb.
Columns 1 and 5 give the dates and times of observation. The other
columns denote the curvature: and as the curve of each side is often
not uniform along its whole length, each column headed “middle”
refers to the curvature at the middle of the length, while the remaining
columns headed “end” give the curvatures at the end of each of the
two halves of the length. In the columns: ec, cs, and (cs) respectively
mean “concave,” slightly so, or very slightly so; v, vs, (vs) respectively
mean “convex,” slightly so, or very slightly so; S and Sn mean straight
and nearly straight respectively; the number succeeding in brackets
in columns gives the number of millimetres of the middle above or
below the straight or base line.

The statistics given in Table Bim when compared with those of
Table B 1 show in particular that the unexpected shortenings of the

P. Groom 25

board during the absorption of water are true, and not due to increased

curvature.

and 23rd, and July Ist to 9th.

Changes in transverse curvature.

This is true of the changes in length between June 22nd

In like manner Table Biv shows

on comparison with Table Br that the measurements recorded of the
widths are not determined by the transverse curvatures; even when
there is a considerable difference in the amount of widening of the two
ends (June 25th to 28th) this is not associated with any corresponding
difference of curvature.

Date and times

VI.

21
99
od

10.30)
10.45 J

230)
2.50

11.50)
1229 i)

TABLE B tv.

yb . yxb
SEE EEE,
Change
Change of shape of width
Concave
Less concave (tends to straigh- 122
ten)
More concave (curves) -195
More concave throughout 463
(curves)
Concave throughout “104
Nearly straight) (straightens) 005
98s “043
as a whole
os - minutely -002
concave,
= zs nearly 007
straight
3 3° O15
Practically straight —-008
Slightly convex (curves) 586
Minutely concave (curves) “034
Definitely concave _,, ‘058

yx. Yy
—<—$$<$$<<$<—-—— ——_—$ —— ———
“ Change
Change of shape of width
Concave »)
Less concave (tends to straigh- 128
* ten)
Vigorously convex in middle 614
(curves equally)
Less convex (tends to straighten) -172

Practically straight

Still less convex (tends to 002
straighten)
Minutely convex\ (tends to —-005
straighten)
as a whole 009
Straight minutely
convex, 10)
Minutely convex| nearly |
straight 003 |
Nearly straight 013 |

Became concave (curves more) 631

Slightly more concave (curves -024
more)

Concave, about=vit. 13; hence “034

it straightens slightly

Absorbing Water

Drying

26 Shrinkage, Swelling, Warping of Cross-grained Woods

Set C (Tables C1 and C1).

The subjoined Table C1 records the linear changes of dimensions
of the same board (yang 1) during the process of drying from an
approximately saturated condition to a state of aqueous equilibrium
with the air of the laboratory. Table Cm gives the averages of the
results obtained.

In addition to details that will be discussed subsequently in
comparing swelling and shrinkage, two facts are worthy of note.

1. During shrinkage there is reached a certain phase at which
the board elongates while still decreasing in width. This phase, when
judged by the amount of water contained, is not exactly synchronous
with the corresponding reverse one during swelling. As in the latter
case this reversal in the change of length is temporary and is succeeded
by longitudinal shrinkage that brings the board down to its earlier
shortened condition.

2. Whereas the board when absorbing water at first widens more
rapidly at the ends than in the middle, when drying it contracts in
width more rapidly at the middle than at the ends. Two possible
explanations of this latter fact present themselves. (a) When drying
owing to the more rapid conduction along than across the grain there
is near the ends a rapid shortening which tends to cause the angle of
inclination of the crossing fibres to be increased with consequent tendency
towards widening; while in the middle of the board the loss of water
takes place more equally and more or less excludes this widening tendency.
(b) Or possibly at first during the drying the water near the ends is largely
in the lumina, as well as in the walls, so that the latter do not lose much,
while the evaporation draws water to the ends from the middle, which
probably contains less water in the lumina and becomes depleted of
this more rapidly than at the ends. Between these alternatives it is
impossible to decide until further researches are conducted on shrinkage
in general.

CHANGES OF DIMENSIONS OF YANG 1 DURING SWELLING
AND SHRINKAGE.

During swelling the percentage average increases in linear dimensions
were: in length ‘079; in width at the ends and middle 6-98 and 7-4
respectively. Thus the percentage tangential elongation was 91 times
as great as the longitudinal.

P. Groom

sv Jv oy} GT ANE ‘]JoF Uler Yonut FT Ayn

86°

908: 5
woTjORAqUOD eTppIa sTyM “FOS:
SI Lg: pue [6L- SuoTZoeIzU0D
pus 0M} oY} WeoMjoq Uvett OT, 908:

CEI: G

‘Suluoyvory} JoyyvoM puv 4slour

‘papp® oq JSNUT GG. S}YSTOAs quonbosqns [[v pus sty} 0} oouY] “YO TPF 6S: SUIYSIOM JUOUISVLT

Ajavou 98-G

L10:-. L08:“ 800°-

oro; 0

600: 0&8: +F€0-
G20: = 68°“ F80-
FOE: 198-el F£E-

L80- 91° F290:
O91: BGs“ 990:
gc0: . 31h 390:
porns
-vomt wey} (3xoopo,o Zt) urese | g00- OLF-“* 900:
poysioa uy} “Usyez Udy} SoAINO = SLE-El —
oy} ‘peysioa. uoy}, “pornsvo]il
aoue eIpPIAL = 990
“TOUT ~19 BIC

96F:
GOSS

qxh* qh

Ayjaeau 9

. 90u9
“10 BIC.

8z8-
696°"
966“
066-21
F10: “
gc0.
9gr
PIs: “
Z8g: “

OF9:
GP9:E1

wh fi

IQ wlavVy,

090-

90ud
“IOI

ZrO“
0c0- “
FLO“

9L0: “
¢80:19

qh hf

90°

CEO
ZE0: “
8z0: “*
ZE0: “
8z0: “
8z0: “
#0: “
OFO: “*
990: “*

890:
GLO-19

quh* wh

8z-LOE

89-1
GLL-—
GLE

1!

9-P

6:89

GST

6-89
GL-TF1

SLL

a0ud
“OHI

| UoTJOMS [[NF
| Z

yysuo]

JO 95¥}UIIIOgT

pouddoyfIp [2IOL

(83L)
G8:9GL
P-CPL
GC9-1FL
#9PL
G-LPL
TES
TI8
G98
1-S88
G8-9C01
9-FE0L

souIels UL
FUS19 A

— (01 ‘1x)
— 60 1x
Og Ol 6T “
Oc'OT 9T “
Geol ST “
Teel.
ioral
Oct e
crol 6 *
OF'OI 8 “
Gate ond
101 JA WAN

awry, aed

DH
Ss

S
=
~S

~
L
~
-~
~

gq of Cross-gra

a

28 Shrinkage, Swelling, Warp

UOTPOOATP [VUIPNFSUOT V Ul FLY} SB FVOIT Sv SOTUT} FG ST UOTJOOIIp [eIgUOSULy v UT “qUOd od SuTOAs OY,

(world) 9
€S-6P
OF L6:¥ I8F-6LL
= aa S§S-F8L
L0-9 SF:9 SE8-98L
(C6-01) (6-91) 829-682
9-F L¥-9 €8é-16L
OFS OL-€ 19€-F6L
GE GL:S 1é9-918
G1 SL-0 I18-0@8
$60 FZ-0 6&2-SE8
99-0 G0 S79-8E8
— — T10-6Z8
pavoq ofoyA. spue spuo
SS eee fw ORE
JO oDRIOAR x Wsue'T

“YApla x YQSUSy
Aq poyeurnysa “4807
IojZVM JO SOUITILID (CT
dod wore Jo osvaroa(T

86:9
908:
S€I: GLO: “
LI0O:— LO8: “
OFO: O6L: “
600- ogs: “
rer 68: “
FOE: 198-21
L80- cgT: “
O9T- Gas: “
890: SIP“
800- OLF: “
= SLP-E1
90U0
OHI
es
“uo Ul
Apis STppruA

C6S £90:
FOS: S80:
8éI: S691: “ £00: OFO- “
2200 Sues" Sot00e=9 ) rr0s
9F0: 068: “* — &Z00- G20: “
660: 966: “* 00: — SFO: “
6F0- S966. 100°- OF0- “
99§- FIO: “* 900: — 60: “
990: oss: “ G10: £80: “
L90: OFF “ G0: St0- “
90: S19“ 00: 010: “
¢¢00- soc: 9900: L0: “
a GELS-E1 a ¢8L0-19
eous §gahgh* ahh oouo gahxh: ghh
CEG eG
ee oe ee
“U0 UT “wo Ul

qIPTA puo uvoyy

IQ Wavy,

yQsuey UvoT

G
8E-LOE
89-91 GELEL
CLL — 6-CPL
GLE-bD SGCL-GPL
OT G:OFL
97 G-LPL
6:89 T-GSL
el II8
6:8¢ G:9E8
GL-I¥I ~—1:S88
GL-L g8-9C0T
a 9-FE01
ooue
“OHI
sae ey)
souUIeIs UI
ISIOM

WINUIIXeUL [VIY
“INL JO onRqyUODIOg

aoUdIOZIp [VIO],

O80
OS'01
cs'Or
Il
4
O08
SPOIL
OF OT
él
Il

owt],

6G “IX

9T “ce
cl “ce
€1 ““c

P. Groom 29

During shrinkage the board did not dry completely to its original
condition (it retained only about 5-5 g. more water), but making allowance
for this fact, its shrinkage was not relatively (nor absolutely) so great as
the swelling. The percentage average decreases in linear dimensions
were: in length -063; in width at the ends and middle 5-92 and 5-98
respectively. The percentage tangential shortening was 94 times as
great as the longitudinal.

Changes in area. During absorption and emission of water it has
been shown that synchronous longitudinal and tangential changes in
dimensions may be opposite in sense. The question arises therefore:
During absorption and emission of water does the volume or surface
as a whole always respectively increase and decrease? The statistics
available prove that a decrease and increase in the amount of water
respectively cause corresponding decrease and increase in area: thus
anomalous changes in length are overborne by the more normal changes
in width (the one apparent exception recorded in Table C 11 July 19th
is discussed later).

The subjoined Table D records the changes in surface for each
ten grammes of water gained or lost. In order to show the corresponding
conditions as regards the statistics concerning drying and soaking
respectively the data are ranged in columns in reverse order.

TABLE D.
Swelling Shrinking
oO a a t =a ee oS my
Increase of Decrease of
area per Weight area per Weight
10 grammes of board 10 grammes of board
Date and hour of water and water of water and water Date and hour
vi. 21 .
a 4:07 | 4:97 x1. 29
2.15 2°3 : vir. 16
‘ } 721-874 cea tae
| (10-95) 797 896 2° 15
3 23 4-06 | 4-6 cad Be ES
ae DA: 3°29 3:46 As 11
a5 2-2 874-896 3:2 » 9 4.30
Hy 4s) 1-01 896-938 1:23 826-885 » 9 10.45
53929 0-05 — — = —
vant, 1 0:22 967-986 0-24 885-1026 sae ks)
ayo 0-387 986-1017 0-66 1026-1034 hd d2
ee ei 0-13 1077-1034. — = —

Both sets of observations show that the greatest change in area (and,
judging by the statistics concerning the radial board, in volume) induced

30 Shrinkage, Swelling, Warping of Cross-grained Woods

by gain or loss of the same amount of water takes place when the wood
is poor in water, and that as the wood contains larger amounts of water
there is a decline in such a change in area. This is in part due to the
larger proportion of water present in the lumina of the wetter wood;
moreover if we assume that the same loss of water always causes the
same decrease in volume of the solid wood-substance (cell-wall) the
smaller diameters of the wood-constituents in the dry phase have to
be considered. But the irregularities in changes of area near the
beginning and conclusion of the drying and soaking require explanation.
One source of misinterpretation is brought out by the change between
July 16 and 19, when there was a slight hygroscopic absorption of
water and yet a decrease in area: probably this additional moisture
was superficially distributed to a considerable extent, and the main
mass of the wood was probably drier on the later date. During drying,
small changes in weight of the board may on the one hand be associated
with relatively greater changes in the distribution of moisture within
the board; moreover during such small changes a slight underestimation
of the amount of water lost will cause the increase in area to appear
excessive. And it will be noted that during drying apparently unduly
large results are associated with small changes of weight, viz. on July 7,
11 to 12 a.m., July 12-13, July 13-15, and July 15-16. On these dates
the maximum result (dividend) appears when the loss of water recorded
was the minimum (July 13-15). During soaking there were no such
small changes in weight, and, excepting at the beginning and end, no
sudden or great changes in the rate of dimensional increase. The
contrast between the results between June 21 and 22 at 10 a.m., and
June 22, 10 a.m. to 2.15 p.m., is possibly due to the fact that the increased
water-content during the former period was due to slow absorption of
aqueous vapour, and during the latter period to rapid intake of liquid
water much of which would be in the lumina and in operation on
swelling; but the unexplained smallness of the increase in width of
the middle of the board on June 22 is also partly responsible.

TWISTING OF YANG 1.

The warping along the length and transversely across the ends
have already been discussed; it remains to consider the warping along
the diagonals and the consequent twisting.

The deviations from the straight were measured in millimetres on
the curves recorded by Professor Dalby’s instrument; but as the

P. Groom 31

latter doubles the deviation the number recorded in the subjoined
Table EK must be halved in order to give the true facts. The measure-
ments recorded from June 21 to July 7 relate to the board during the
absorption of water, those subsequent to July 17 to the drying board.

The diagonal yx. yb runs rather across the grain exposed at the
surface and consequently underwent greater elongation and shortening
than did y . yxb, which tends more towards parallelism with that grain.
The lengths of the diagonals measured were approximately 56 centi-
metres.

In Table E:

Column | represents the time of observation.

Column 2 records the amount of water absorbed subsequent to the
immediately preceding period.

Columns 3 and 7 record the nature of the curve, whether concave (c),
or convex (v), or straight (s) of the whole board along the diagonal
and the number of millimetres that the middle is sunk or raised above
the true base line as recorded by the instrument.

TABLE E.
2 3 4 5 6 7 8 9 10
ya. yb y. yxb
a aS ore == Las 2a
Grammes Shape and deviation Shape and deviation
Date and of water = =—————_*__—_____, Twist Twist
hour (cirea) absorbed Whole yx half ybhalf deviation Whole y half yxbhalf deviation
vi. 21 — c 16-25 11-25 12-0 54 v 8-75 7-75 6-75 4-75
a ry gl il] 16-45 ¢ 15-5 11-75 9-25 51 v 8-25 6-5 6-25 3°75
Aso eh 3.25 59-15 c 14-25 6-5 12-5 53 v 9-75 7 6-75 6-75
oy a) dl 44-8 c 9-25 3°75 8-5 33:5 v 45 3 3:5 8
» 25 3.50 I:l ¢ 9-25 4-5 8-25 37 v 4-375 3 3 0
9» 24 12 32 s 0 0 0 1 oes: 5) 0 0
Somes 2.50 — fi sh O 0 0 0 s 0 0 0 0
6 25 12.5 18 (s) 0 0 0 0 s 0 0 0 0
so 26, 11-20 46°6) 1 (@) 0 dish 0 0 s 0 0 0 0
ap, AS) UP 29-2 » 1-5 1-5 “Hf 0 s 0 0 0 5
wary L135 Gye} cy aes: “75 5 Say (0) 0 0 0
ae) AL25 30°55 v4 2-5 2-0 1-0 ay (@) 5 5 ai! 1-0
soe fle 30 17-5 ¢ 2°75 2:5 1-25 0 (c) -5 “5 1 1-75
» 9 12.30 -2084 c 2:75 2 2°5 0 (c)) <1 375 1 0
ome l230) 75-4 6 13:5 9 10-0 A
lone UiSOmes — a t-O)) 7c: 15 11-25 11-5 51
WG) M45." = 4:4" ¢ 16 11-5 11-75 2
xm.10 12 —13-6 c 18-75 13:5 14-0 64

32 Shrinkage, Swelling, Warping of Cross-grained Woods

Columns 4, 5, 8, 9 similarly record the deviation in millimetres of
the middle point of each half of the respective diagonals; columns 4
and 8 denoting the half (top) towards the y. yx end, and columns 5
and 9 denoting the half (bottom) towards the yb . yxb end.

Columns 6 and 10 record the amount of twist by giving the number
of millimetres that the end point is raised above the base line, as recorded
by the instrument.

Table E shows that on June 21 the dry board was warped and
twisted, being on one face (y) concave along one diagonal yx . yb and
convex along the other, y.yxb. With the absorption of water these
diagonals by differential elongation lost their curvature, and were
almost or quite straight on June 24 and 25. As water continued to be
absorbed until July 5 the original curvatures of the diagonals were
gradually reversed, yx . yb becoming convex and y.yaxb concave, the
reversal taking place more obviously along yz . yb (crossing the grain).,
During the drying from July 7 onwards (and even before this, for some
unknown reason, in the case of yx . yb) the curvatures gradually reverted
to those originally present in the dry board.

[t is worthy of note that the amounts of warp and twist were not
exactly determined by the amount of water contained. This fact is
demonstrated by comparisons between the conditions of the board on
June 21 and on July 16 and December 10. The greater twist and warp
during the final drying are probably due to the fact that the loss of
water was more rapid in the warm dry air of the laboratory than during
the original industrial seasoning of the board. This consideration
emphasizes the special importance of carefully regulated seasoning of
cross-grained woods.

OBSERVATIONS ON YANG 2.

The board named yang 2 was cut tangentially and its faces were
lettered as in the case of yang 1; on one face, Y and Ya denoting the
top corners, Yb and Yab the bottom corners; and on the other face,
y and yx, yb and yxb denoting the corresponding points, but y was
superposed on Y, that is to say Y and y are at the same edge of the
board. Thus for instance Y Yb and yyb represent the same side of the
board on the two different faces and Y.Yab and y. yab represent the
same diagonal on the two faces.

P, Groom 33

Set F,

The first set of measurements, recorded in Table F, were made by
means of the measuring rod during the absorption of water. The changes
in length were too slight to be accurately estimated. On the other
hand measurements of the diagonals and across the two ends were
possible, and evidence in favour of their substantial accuracy is supplied
by the fact that the widening of the board at both ends was the same
on the average, viz. -6375: moreover on the face y it was -65 at each
end, and on face Y -625 at each end. This represents a widening of
5-3 per cent. of the full length attaimed. But it must be noted that
when the board was first placed in water it was not sufficiently dry to
be in equilibrium with the atmospheric humidity of the laboratory ;
this was proved in the second set of observations when the board was

dried.
TABLE F.
Diagonal Transverse (tangential)

Date Weel NC Veni) Vein Mo OOF FRAO MR ae Wii) Do Oo
April 29 588-4 44-5 44-3 44-3 44:3 11-5 11-55 11-25 11-25
May 3 672-6 44-6 44:5 44-5 44-4 12-08 12-1 11-75 11:8

m & 688 44-6 44-5 44-475 44-4 12-05 12-125 11:8 11-85
Sse 763-6 44-65 44-55 44-6 44-475 12-1 12-175 11-85 11-9
ase lia 787 44-6 44-5 44-55 44-475 12-125 12-2 11-875 11-9
Difference be-)
tween Apr. 29 ; — 0-1 0-2 0-25 0-175 0-625 0-65 0-625 0-65

and May 17

During soaking the board twisted, but the direction of the twist
was opposite to that of yang 1 in conformity with the fact that on the
two faces its exposed oblique grain ran in directions representing spirals
reverse in sense to those of yang 1. On face Y, the diagonal Y . Yab
ran more with the grain and therefore showed less elongation (-1) than
did Yx. Yb (-2) which was directed more across the grain and became
convex. For the same reason the elongation of yx. yb (-175) was less
than that of y . yxb (-25). Consequently of the two superposed diagonals
Yx.Yband yz. yb the former became convex and the latter concave:
while of the other two superposed diagonals y . yxb became convex and
Y . Yzb concave: the result was the above mentioned twist. Naturally
in subsequent drying these relations were reversed.

Ann. Biol. mt

ww

34 Shrinkage, Swelling, Warping of Cross-grained Woods

Set G.

The results of the measurements of yang 2 during its final process
of drying from a saturated condition are given in Table G.

The board at first decreased in length, but thereafter continued to
elongate. In this respect it differed from the tangential board yang 1,
but agreed with the radial board. The final loss of length was exceedingly
small, merely -012 per cent. of the maximum length, but the maximal
loss was -03 per cent. These figures are far less than those for yang 1,
and even less than for the radial board.

The shrinkage in width of the ends was 6-29 per cent. of the maximum
width, and was thus greater than those of yang 1 (5-92) and the radial
board (5).

When the shrinkage in area is assessed by the changes in length and
in width of the ends it will be seen that the loss of width atoned for the
shightness in the shortening: for the shrinkage of area was 41-266 cm.®,
compared with 49-53 cm.2 in the case of yang 1. Yang 2 was only
about 48 cm. long, whereas yang 1 was about 61. Had yang 2 been
of the latter length its contraction in area would have been (circa)
52-43 cm.”, thus slightly more than that of yang 1.

The changes in length and width from May 28, 1915 to February 4,
1916, are worthy of special note as they occurred when the board was
so far dry that it gained or lost water solely according as the air became
moister or drier; and during this phase the weight of the board varied
only between 570-075 and 570-48 g.

RADIALLY Cut Boarp.

The first set of observations were made while this board was absorbing
309 g. of water. As a measuring rod was employed for the purpose
the longitudinal measurements were found to be useless. As regards
the transverse measurements the ultimate increase in width was 5-5 per
cent. of the maximum attained. This statistic is not of value as showing
the full radial expansion, for it was subsequently found that when first
measured (and weighed) the board was not fully dry in relation to the
laboratory air. Its original weight was 719 g., whereas at the conclusion
of the second set of observations the board had dried down to a weight
of 706-5 g.

During the soaking, in contrast with the tangential boards, the
radial board remained flat and devoid of any twist.

»
”

P. Groom

‘sorqjowruoo “bs 99Z-[F ‘vor cs ce eynjosq V

6s:9 “qypra ee “6 ee e
"e0. 9z 09 LT dy £210: F Goq 09 LT Tad y + ygSue] wnturxeUt oy} Jo (asvoroop) doUoIOYIp 95kzUI010 J

UINUITX UL

098: StS: OLS: OLS: C10: S10: 910: S10: 910: SOG GHG |
i F 420
gcse: 098: OFS: TGS: ZLS- 900: 800: 900: LOO: £00: <= —LI Avy
| oOUdIO YLT
ZBL: “ z69: “ 889: “ 006: “ 87s: ¢z100- “ FL6: “ #20: “* GL6: “ Ze0: “ SF-OLE F “49
PSL: “ 969: “ 989: “ 006: #g8:“* G000-;8% ZL6:“ Tzo: “ LG: “ seo: “ CLO-OLG el Aine
8sL:“ ZOL: “ 969: “ 006: “ Fas: “ 866: “ OLG6: “ STO: “ ZLG: ZEO- CT-OLG ¢ oun
FSL-ZI FOL-ZI 90L-Z1 Z88-Z1 FPS-ZI £66: “ gg96-: “ 9TO- ‘ £96: “ 8zo-: “ L-OL¢ Gr
16g: “ C6F: “ CSF: “ GIL: “ GLO: “* 766°LF $96- “ FIO: “ OL6: “ 0z0- “ 8-699 6 “
8z9: seg: 6c: “ PPL: ZOL-“* GZL00- “ Z86: “ G0: “* 786: “ cgo: * €-19L ae
OF9-ET EOC FES-E1 COLT OZL-EL GL00-8F Z86-LF €0-8F Z86-LF 9€¢0-8F C-6LL LI Avy
oseioay .qrh* qh qxx qx ch: h aE’ X osveiay qvh: xh Gh NODS Hae Ok =A qYDIO AA oye
= ———_~—_—_—_—_———-——~ VW
So1joUTZUGD UT [IPI sorjouIIyUe. UT YySUN'T
‘‘K) ATAV I,

36 Shrinkage, Swelling, Warping of Cross-grained Woods

Set H,

In the second set of measurements, made on the radial board while
this was drying, vernier callipers were employed, and the results are
recorded in Table H.

During this thorough and at times rapid drying the board, in sharp
contrast with the tangential boards, remained flat and untwisted.

During drying the shortening was slight, only -025 or -04 per cent.
of the maximum length, being intermediate in this respect between
the two tangential boards. Reverse changes in length took place, as
during two phases elongation took place while at the same time decrease
in width was marked. Especially worthy of note is the fact that
elongation continued to the final period of drying (June 18 to November
29). Such synchronous elongation and narrowing took place not only
during the period June 14 to 16, when the board was drying, but actually
continued between June 16 and 17 during a slight absorption of atmo-
spheric moisture. This latter fact was probably caused by the smallness
of the amount of the water absorbed and its incomplete distribution
through the wood.

The decrease in width of the ends was 5 per cent. of the maximum
width, and thus less than in the tangential boards (5-92 and 6-28).

During the process of drying, even when the board was elongating,
the area of the faces (as assessed by length, and width of the two ends)
continued to decrease. The whole decrease in area was less (40-907)
than in the tangential boards.

There is evidence that soaking caused the board to increase in width
permanently, for the succeeding statistics show that for similar changes
in water-contents shrinkage in width during drying was less than
expansion in width during soaking. It must be noted in regard to these
statistics, that during soaking the width measured was not the whole
width of the board but the distance between two pencil points 11-325
to 12-15 em. apart, whereas during drying the whole width of the board,
13-258 to 12-802 em., was measured. Hence for perfectly similar
comparison the statistics of widening given below are too small.

Change of Gain or loss Change of width of
weight of water Y= Yoeand soe eva
Board absorbing water 722-1—1038 g. 315-9 g. *6 cm.
» losing 3 1038—723:3 g. —314-7¢. —-5 cm.
» absorbing ,, 719-1—1038 g. 318-9 g. -655 em.

>» losing a 1038—717-4 g. — 320-6 g. —-577 cm.

‘OUvS 94} SojOoUep USIS snuTU &e
ToyeM POUTeFUOS Jo JUNO oY} UT esuNY oY4 07 AreIZUOD OSes UT st SUOTSUOUUTP UT oSURYO oY} SOY Sosvo yrvur odAZ yory, ur poqyumd saunbr yp

"ygSug] Jo sso] osvqyusored x YOT=YYIPIA Jo ssoy oSvyuoor0d Ayyeury

‘g “UMUIxelm pure [eUug + yypra se 4 ss

‘F0-0 ‘(UF 09 496 SUN) TINTUIxeUT ws ee ss ue
"CZ0-0 “(ToqmIaAoN 07 ons) [BUY + 445u9] WNVATXeUT oYy Jo sso] oSeyUDDIOg

= a — — — CFO: ae == = = = S = a — — — | ‘IA-6‘IA

meee core = 199 — — eto) = cpr! G19" 299" Ga0: ZrO: G10. | B00. coon | ree eh oe
: QUINOA,

88-9 €9F-> ZO GLO: 819: 600-— ZEo-“ gs99-“ seg“ orc.“ ge“ ozo“ eF0-“ EsE0-“ ZEO“ Baz Ze-90L 63 “IX

ZOT-OLL
LEG C0: C9E-FLL €€0- €69-" 700: 820: Sci“ est“ oso.“ oLo-“ O10-“ 980°“ #20: Fo." 68 8-SIL Sine
S78 96-0 029-9LL F00-— 922°“ 000: ZLz0-“ FOL.“ B¥L“ 889.“ FOL“ LIO-“ geo“ HzO“ OF0-“ ‘elie L-LIL i
GPS L6I-€ EL8-9LL €&0- o€L-“ 300-- L9Z0:“ FOL“ OGL“ G69.“ OTL“ 910°“ 280°“ g9zo-“ 9zo-“ 6-¢ PLIL 91 “
€F-9 910-9 OL0-08L OOT- €8L-“ 300-- S¥z0-“ O18-“ os“ Zot“ OLL-“ ozo“ seo“ O10“ of0-“ GPE €-SSL Cla
Lé-G I8€6I 9FI-98L STE: 688-2 G10: 620: 9L8-ZT SL8-ZI LL8-ZI 006-ZI 810-“ zed.“ F20-“ 910°“ ecg GL-zeL ab &
5-0 ¢¢8-€ LzZG-G08 190: S61: O10: PEO" 98G° "SG EET." «SLT. «GTO. #0." 080° —oFO> T-€L 818 i
II'0 L29:I 88-608 92z0- 6¢c:“* — §00- HrO'~ 908: g8c-“ gle“ ose“ gzo:“ 090°“ seo“ ogo“ 6 9FI T-168 Ol.

600-TT8 C8C-€1 L¥O-19 (O8E-E1) OIE-EI1 SESE 8SZ-SI ZE0-19 (290-19) ZEO-19 Z&0-19 8éor 6 “IA
SOT Loves BdUBIABI Bly s9UaIayI YIPIAA suUeIepIq yYSueT quh qh Qqrx'qx ehh AOC RL gah h gh h Qo KE I9yeM pareoq Jo ued
JO 9°9 0OT SSS SS. OO Sa = sy Sa JOSSOT 44ST A,
tad Bare Jo (q}pTA pua Saajott}m99 $91qaUI1} 090 UT Sd1J9UITJU9O UI Spud JO TPT AL Sa1JOUIIUId UT Yysua'T SS
asvaioa(y ASvIOAG X 45197) ul spua Jo YSU] osR18A Wy SOURIS UL JY SION
seljetaiyuao arenbs Y}PIa ase10ay

Ul BIB DSVIDAY

‘HL Wavy,

38 Shrinkage, Swelling, Warping of Cross-grained Woods

While 323 (715-2 to 1038) g. of water were absorbed the average
measured width of the same face decreased by -612 cm., whereas while
the board underwent more thorough drying and lost 331-5 (1038 to
706-52) g. of water, the average whole width decreased by -673 cm.,
which represents a decrease of -606 on the width measured during
absorption.

I have pleasure in tendering thanks to Professor Dalby for his
extreme kindness in inventing and providing the instrument for
registering the changes in curvature, and to Mr Alexander Howard for
presenting the boards for investigation.

39

THE DALBY PROFILE RECORDER.

By W. E. DALBY, M. Inst.C.E., F.R.S.
Professor of Engineering, City and Guilds Engineering College.
(With 7 Text-figures.)

1. INTRODUCTION.

A sHoRT time ago Professor Groom asked me if a method could be
devised to enable him to measure the shape of the surface of a timber
sample with reasonable speed and accuracy so that measurements taken
from time to time could be compared in order to study questions relating
to the warping of timber.

In response to this request I designed a machine by means of which
the shape of the surface could be explored and recorded automatically
without the necessity of taking a single measurement directly.

The result of the exploration of the surface by the machine is a
drawing showing the shape of the surface along parallel lines spaced
at definite distances apart.

Such a drawing is seen in Fig. 1. It is the result of exploring an
artificially prepared surface of a piece of pine along five parallel lines.
The surface was made specially irregular in order to illustrate the
working of the machine.

The five datum lines numbered respectively 1 to 5 and the corre-
sponding profile curves were drawn in 3 minutes. The datum lines are
spaced 1 inch apart and the length of the record is about 2 feet. The
size of the sample used in the particular machine in which the record
was drawn is 30} inches long and 6 inches wide.

The curves on the drawing are really the profiles of five equidistant
sections of the timber sample taken normally to a reference plane.
The profiles recorded show the variations of shape to twice the actual
size for convenience of measurement. The machine can be designed
to give the record the actual size or any multiple of it.

2. THe REFERENCE PLANE.

The machine is so designed that the datum lines in the record
correspond to lines lying in a common plane in the timber sample.

40) The Dalby Profile Recorder

This plane is called the PLANE oF
REFERENCE. The plane of reference in the
timber sample corresponding to the lines on
the record is defined by any selected point on
the timber surface to be explored. Having
selected a suitable point the exploring roller
of the apparatus is set to the point and then
the plane through this point parallel to the
plane defined by the surfaces of three studs
to which the sample is clamped is the plane
of reference from which ordinates to the
surface are to be measured. The roller was
set on the sample to the point corresponding
to'Z in Fig, 1.

A reference to Fig. 2 will make this clear.
A, B, C are three supporting studs in the
machine. The upper surfaces of these studs
define a plane. Let a piece of timber be laid
on these three points. Let Z be the point
in the upper surface of this piece of timber
to which the exploring roller is set. Then
a plane through Z parallel to the plane
defined by the points A, B, C is the PLANE
OF REFERENCE from which all ordinates are
measured to the undulating surface. In all
forms of the machine there will be found
three supporting studs corresponding to the
points A, B,C. These studs may be regarded
as three blunted points, the upper surfaces of
which define what may be called the PLANE
OF THE MACHINE.

The plane of reference is parallel to the
plane of the machine at a distance from it
determined by the particular point in the
surface to which the roller is set at the
beginning of the process of taking a record.

The traces of these planes are shown in
the vertical elevation Fig. 3.

If the point Z is in the same plane as the
points A, B, C then the plane of the machine
and the plane of reference coincide. This
condition is practically fulfilled m some

4 1 2 3 4 5
Fig. 1. Profile curves from an
artificially prepared surface
of a piece of pine.
In the original record the da-
tum lines are 1 inch apart.

W. E. DALBy 41

forms of the machine. In other forms the three points A, B,C are in
the lower surface of the timber sample and the point Z is in the upper
surface so that the plane of reference is above the plane of the machine
by approximately the thickness of the sample.

Fig. 2

It is only when the sample is clamped at three points that a true
record of the surface can be made. The act of clamping at more
than three points distorts the timber so that the family of profile
curves correspond to a surface slightly strained, and the surface
will therefore change in shape immediately the sample is removed from
the machine. When however it is clamped at three points only the

Z

PLANE _OF REFERENCE _

sample is quite unstrained, and moreover it can be removed from the
machine and can be replaced in it again in the same position relatively
to the frame, providing always that it is clamped at the same three
points. The initial clamping points should therefore be carefully
marked on the sample by ringing them round or by any other con-
venient method.

3—)

42 The Dalby Profile Recorder

The position of the point Z must also be carefully marked on the
surface so that the exploring roller can be re-set to this point when
taking subsequent records of the surface. The re-setting to Z ensures
that the families of profile curves in all the series of records which may
be taken of the surface are recorded in relation to the same reference
plane. In other words by re-setting the roller to Z the position of the
reference plane in relation to the plane of the machine is maintained
substantially without change however often the sample is replaced in
the machine. In the time elapsing between two successive measure-
ments of a family of profile curves of the surface, warping slightly
changes the distances of the pomt Z on the surface from the clamping
points; but the change is quite negligible in its influence on the position
of the reference plane if the point Z is initially chosen close to a clamping
point. In Fig. 1 for example, Z is taken as close to the clamp A, Fig. 4,
as possible. It would probably be better in a definite series of experi-
ments to take Z as near to the central clamping point as possible (C
in Fig. 4). Then the warping of the surface would have the least effect
in changing the relative positions of the reference plane through Z and
the plane of the machine.

Warping and shrinking also change the relative positions of the three
clamping points themselves during lapse of time. The change is likely
to be slight and the effect on the relative position of the reference plane
and the plane of the machine negligible. To secure uniformity in
practice it is advisable to re-clamp a particular sample so that the central
clamp (like C Fig. 4) grips the timber at the same point in all the re-
settings of the series. The clamps A and 6 will then grip the sample
at points displaced from the original pomts by amounts due to shrinkage
in the linear dimension of the imaginary triangle formed by joining the
three points at which the sample was originally clamped.

In cases where great accuracy is required the hole-slot-plane method
of clamping may be used. Assuming the timber to be hard enough to
bear clamping without appreciably indenting the surface, a conical hole
is formed in the timber to receive the conical point C (Fig. 2); a
V-groove pointing towards C' is formed to receive the conical point B;
and the surface of the timber rests on the conical point A. The timber
sample when clamped down on to these conical points is then fixed
relatively to the frame of the machine in the most accurate manner
possible.

In cases where the timber is soft and therefore the conical sup-
porting points are likely to form indentations in the surface, metal

W. E. DALBy 43

screws may be screwed into the timber, the heads of these screws
being specially formed, the screw at C with a conical hole, the screw
at B with a V-slot, and the screw at A with a plane head.

The reference plane will stand at a fixed distance from the plane of
the machine defined by the hole-slot-and-plane clamping except for
the negligibly small error produced by warping and shrinking in the
distances of Z from the clamping points. This error is minimised by
selecting Z as close to the conical hole as possible.

3. GENERAL DESCRIPTION OF A RECORDER.

The particular form which the profile recorder takes depends upon
the purpose for which it is required and upon the sizes of the samples
which are to be tested.

Apparatus may be designed for drawing the profile curves of the
largest planks or for studying samples of moderate size or for measuring
the shape of the blades of an aircraft propeller.

One type (Mark 2) is shown by the photographs Figs. 4 and 5.
This apparatus takes samples 303” by 6” and any thickness up to 14”.
Variations in the shape of the surface are shown twice the actual size
in the record. This multiplication of the surface variations can be
carried to any extent desired but the scale once settled remains constant
for any particular machine.

Referring to Fig. 4 the timber sample 7' is clamped down by the
clamps A, B and C to the corresponding studs below it. The drawing
paper on which the records are to be taken is pinned down to the drawing
board at D. The angle iron framework is self-contained and supports
on the one side the timber sample and on the other side the drawing
board.

Lying on the bottom bars of the framework is a guide frame GG
consisting of two parallel rods secured in end pieces. It will be seen
that this frame can be lifted from its position and can then be placed
in another position defined by any one of four pairs of V-notches cut
in the lower bars of the framework. These notches are pitched | inch
apart.

The stock of the machine is supported by the guide frame. It is
shown separately in Fig. 5. Its base is made of cast iron and it is
srooved to slide along the guide bars GG@ seen in Fig. 4. It is pushed
along them by hand when a profile curve is being drawn.

The stock carries an exploring roller & on the end of an arm and a

+t The Dalby Profile Recorder

pencil P on the end of another arm. Mechanism connects the two arms
so that the vertical movement of Ff is changed into a horizontal movement
of the pencil in a direction at right angles to the direction of motion of
the stock.

To draw a profile curve the stock is placed on the guide frame GG
and the mechanism is locked in a zero position by turning the milled
head L seen at the lower part of Fig. 5. The roller is clamped clear of
the surface. The pencil is then lowered on to the paper and the stock

Fig. 4

is pushed along the guide rods by force applied to the handle U. The
pencil then draws a straight datum line. The roller is then lowered
into contact with the timber surface at a point of the path nearest to
a clamping point and this point is marked Z. This point defines the
position of the plane of reference, and the datum line already drawn lies
init. The lock is then released by turning the head ZL and the stock is
drawn or pushed along the guide bars. The roller now follows the
unevenness of the surface along the path which it is compelled to follow
and the pencil P draws automatically the curve giving the shape of

W. E. DALBy 45

this path twice full size. The curve is the profile of a section taken
through the roller path at right angles to the plane of reference.

The process of drawing a family of curves is similar. The datum
lines are first drawn one for each position of the guide frame in its

Fig. 5;

notches. Then the point Z is selected. Guide frame and stock are
adjusted to bring the roller over the selected point and then the roller
is lowered into contact with the surface and the roller post is securely

46 The Dalby Profile Recorder

clamped. The profile curve corresponding to each pair of notches
provided is then drawn.
The family of curves, Fig. 1, were drawn in this way.

4. INTERPRETATION OF THE PROFILE CURVES ON THE RECORD.

The datum lines lie in one plane, the plane of reference. The distance
from any point on any one of the profile curves to its datum line is
equal to twice the distance of the corresponding point in the timber
surface to the reference plane defined by the point Z.

For example the point Y, on the Record No. 1 (drawn again in Fig. 6)
is ~” above the reference plane containing the point Z on the surface of
the sample. Similarly the point Y, is 0-2” below this reference plane.

All the points on the timber surface corresponding with points of
intersection of the profile curves with their respective datum lines lie
in the reference plane containing Z.

For example the points on the surface corresponding to points

Zzzzz and yyyyy in the record all lie in the plane of reference. The
“lines joining these points are contour lines in the reference plane.

Points on the profile curves at equal distance from their respective
datum lines can be located. Curves through these points are contour
curves for the particular distance located.

The family of profile curves drawn by the machine can therefore
be used to find the contour lines of the surface explored. And these
contour lines can be drawn on the record.

A family of profile curves of sections at right angles to those drawn
automatically by the machine can be deduced. Suppose for example
that the profile is required across the section SS on the record shown
in Fig. 6. At each of the intersections of SS with the five datum lines
of the record, set up (or down) the intercepts on SS, cut off by the re-
spective datum lines and the corresponding profile curves. P,P,P,P,P;
are points obtained in this way and the curve through them is the profile
curve across the section SS.

Profile curves for a series of transverse sections at any assigned
interval apart can be deduced from the family of profile curves drawn
by the machine. Such a family is shown in Fig. 7 for transverse sections
taken about 3” apart on record No. 2, a record taken from an artificially
prepared surface of a sample of pine.

Lf /f,
A) fp fp

ALL LLL ll

Fig. 6.

Tr
WY VTA Poot
Y,
\
~\
\
\
\
\

Contour
line OQ
2) = Ss ee ee
PROFILE Curves. Redrawn Fig. 7. COS a fainily of profile
from Fig. 1. curves taken from an ‘artificially
: prepared surface of a_ pine

sample together with a family
of transverse profile curves de-
duced from them.

Each of these diagrams is a little larger than } full size. In the original

records the datum lines are 1 inch apart.

48 The Dalby Profile Recorder

5. Use oF THE RECORD TO PLOT A ContTOoUR LINE
ON THE TIMBER SURFACE.

The point on the timber surface corresponding to any point on the
record can be identified by placing guide frame and stock so that the
pencil is brought over the point on the record. The roller will then
mark the corresponding point on the timber surface.

The series of points lying on a contour line on the record can thus
be identified on the timber surface and an actual contour line can then
be drawn on the timber surface itself. The meaning of such a line is
that all points on it are equally distant from the reference plane con-
taining the point Z and parallel to the points A, B and C which have
been definitely imprinted on the timber surface by the three studs
against which the timber sample was clamped.

49

THE ACTION OF ENCHYTRAEID WORMS
By tHE Rev. HILDERIC FRIEND, F.R.MS.

Tue following paper is based on the Report which was presented
to the Board of Agriculture as the result of experiments carried out
in 1914-15 under the direction of Professor Gamble, F.R.8., at the
Birmingham University.

The main objects were twofold:

1. To determine the question of the injurious action of Enchytraeid
worms on living plants.

A long series of experiments was conducted at the Hdgbaston
Botanic Gardens with Asters, Antirrhinums and other plants, with a
view to ascertaining what effect, if any, was produced upon the living
plants by the presence of these worms, which had frequently been
charged with causing decay and death. Control plants were used,
and infections were made with various species of worms under a great
variety of conditions. The results showed conclusively that so long
as the plants are healthy and vigorous Enchytraeids do not attack them.
On the other hand, when living plants begin to decline from any cause
it is the usual thing for white worms to take up the work of destruction
and play the part of scavengers.

While the experiments were in progress visits were paid to Droitwich,
Kenilworth, Shrewsbury, Edgbaston and other places in which gardens
were reported to be suffering from injurious Enchytraeid attacks.
In no instance was the evidence sufficient to justify the opinion that
white worms were the original aggressors, though in several instances
they were busily engaged in clearing away the decaying plants. The
initial injury was due to a variety of causes, such as fungi, bacteria,
Julus and other pests.

2. To determine the réle of the red-blooded Enchytraeids.

During the year advantage was taken of many opportunities to
study both in the field and in the experiment house, the réle of that
group of minute worms belonging to the Enchytraeids which are charac-
terized by red blood. They are found in moist places among vegetable
debris, in sewage works, ponds, ditches, farmyards and especially by
the seaside, and are found to be invariably engaged as beneficent
scavengers. Though not infrequently met with in tap water their

50 The Action of Enchytraeid Worms

presence is never occasion for alarm, and even if they were inadvertently
swallowed no injurious effects need be feared. <A report will appear
in Science Progress for July.

The following details of observations and experiments have been
selected from a large number which have been carried out, as they
tend to the elucidation of the foregoing report.

1. Antirrhinum Cultures. (1) The object was to determine whether
or not Enchytraeid worms caused failure and death.

In December 1914, twelve plants were purchased from Messrs
Simpson, Florists, such as they were rearing for market. Six plants
were retained for control. The other six were inoculated with Enchy-
traeids which had been found associated with sickly plants and were
suspected of being inimical. The plants were kept in an Experiment
House at Edgbaston Botanic Gardens, and regularly watched and
examined. For a time the inoculated plants looked paler than the
controls, but eventually they showed themselves to be stronger. All
plants survived. In April 1915, the six plants were again inoculated.
Blossoms perfected at the same time on both sets of plants. On May 1,
the pots containing inoculated plants were examined, and found to
contain many worms, but the roots were all healthy. Plants were
then transferred to the Garden, and all treated alike with Enchytraeids,
but not one succumbed.

2. Antirrhinum Cultures. (2) It was reported that Antirrhinums
were suffering from root worms at Droitwich. A visit was paid to the
Experimental Gardens, the plants inspected, and specimens brought
away with soil. The soil contained many Enchytraeids, but was also
full of Julus larvae and other suspects. These were removed, and the
plants repotted with all the Enchytraeids as found. One plant died,
the evidence showing that it was too weak to survive removal. The
others rallied and became very vigorous in spite of the worms.

Slips taken from both these sets of Antirrhinum took root readily
and made healthy plants.

3. Antirrhinum pests. While these cultures and experiments were
in progress many plants which were diseased at Edgbaston, Kenilworth
and elsewhere were examined. In the case of seedlings, forced in frames
and greenhouses, fungoid growths were the chief cause of failure. Out-
of-door plants suffered from various causes, but in no instance could
evidence be found for convicting Enchytraeids.

H. FRIEND 51

4. Aster Cultures. December 2, 1914, Messrs Sutton and Sons,
Reading, kindly supplied three packets of seed for experiment. These
were A. sinensis (type) with the varieties Victoria and Ostrich Plume.
The Head Gardener at Edgbaston had ceased to grow Asters as they had
of recent years proved a complete failure.

The foreman sowed seeds of each kind (which were marked 1, 2
and 3) in the soil used in the gardens used for bringing on seedlings,
and these were allowed to grow till they were ready for repotting.
While a number of plants were reserved in the original large pots for
control, others were planted in (1) loam, (2) sterilized leaf mould, and
(3) Enchytraeus infected soil and manure. They grew side by side
and flourished.

Next, some of the repotted plants were inoculated with Enchytraeids
while the rest were retained as controls, but not a single plant fell a
victim to the treatment.

A second set of experiments yielded the same results. Seeds of
the same samples were sown in pots containing (1) loam, (2) sterilized
leaf mould, (3) Enchytraeus infected soil and (4) sand. These in due
course were infected with Enchytraeids, including species found in the
rubbish heaps in the Botanical Gardens; and again, all survived.

Some of the plants were transferred to the open border, and others
planted in soil containing Enchytraeids, but they made the most
perfect growth.

5. Miscellaneous investigations. During the year, every opportunity
which presented itself for investigating plant diseases was utilized.
Turnips near Coventry, Lilies at Shrewsbury, Peas at Erdington,
bowling green Grass on the Solway, and fairy rings at Cheltenham
were carefully examined, and all the evidence for and against the
Enchytraeids carefully weighed. In no single instance could it be
shown that white worms were the aggressors, but there were frequent
evidences that Enchytraeids came in as secondary agents when plants
were already weakened, and so played the part of scavengers.

Conctuston. Enchytraeid worms are abundant in leaf mould,
and are often found associated with decaying field and garden crops
on which account they have been suspected of causing their decay and
death. The evidence shows that they are a beneficent group of animals of
great economic value to the forester, gardener and farmer. They break
up decaying matter and prepare it for further service but it remains to be
shown that in any case they are the direct cause of the diseases of plants.

THE FOOD OF SLUGS AND THE DEVELOPMENT
OF ANOPLOCEPHALIDAE

By Proressor A. RAILLIET,
Ecole vétérinaire d Alfovrt,

Foreign Member of the Association of Economic Biologists.

I HAVE just read, a little tardily, but with much interest, the note
of Miss Marie V. Lebour, “Some feeding habits of slugs,” published in
The Annals of Applied Biology, Vol. 1, Nos. 3-4, January 1915, p. 393.

Miss Lebour has discovered that the slugs, Agriolimax agrestis and
Arion circumscriptus, ingest very greedily the proglottids of two
Anoplocephalidae,—Moniezia expansa of sheep and Cittotaenia pectinata
of rabbits; she has ascertained that the eggs of these Cestodes pass
through the intestine of the Molluscs without undergoing alteration,
the six-hooked embryos remaining alive; but she has not discovered
any trace of the larval Cestodes in the body of the slug.

It may not perhaps be without interest to record, in this connection,
the results of an experiment, very imperfect, but nevertheless important,
which I had occasion to make twenty years ago and about which I have
just found certain short notes.

On November 4, 1892, I collected from the faeces of a carrier pigeon
coming from Neufchatel-en-Bray (Normandy) some proglottids of
Bertiella delafondi; these proglottids, macerated in a little water, were
placed on a cabbage leaf with four grey slugs (Agriolimax agrestis).
The next morning, I noticed that almost the whole of the contaminated
leaf had been eaten.

On the 14th of November, I dissected one of these slugs, but could
only find some Nematodes and Trematodes.

On the 8th and 30th of November, I repeated this operation with
chicory leaves and a second lot of six grey slugs. On the 14th of
December, I fed two pigeons on these slugs, each receiving three.
Towards the middle of January, one of these pigeons began to evacuate
proglottides of Bertiella delafond:.

As can be seen, this experiment has been left unfinished, since I have
not observed in slugs the larval form of the parasite, and unfortunately
I have not had an opportunity of repeating the experiment.

VotumeE III JANUARY, 1917 Nos. 2 and 3

A BACTERIAL SPOT OF CITRUS.

By ETHEL M. DOIDGH, D-S8c., F.LS.,
Mycologist, Union Department of Agriculture.

(With Plates II[—XIII).

In October 1914, toward the close of the Citrus season, a grower in
the Western Province submitted for examination some lemons badly
disfigured by dark-coloured sunken spots; he stated that he had first
noticed this blemish two seasons back, but that it was rapidly on the
increase; as much as 30 per cent. of his lemons were affected, and the
trouble was spreading to the Washington Navel oranges.

No fungus mycelium could be found in the discoloured tissues, but
there were pockets of disintegrated cells filled with innumerable bacteria.
A pure culture was obtained at once from these tissues, and a few suc-
cessful inoculations carried out. It was too late in the season however
to enter on a complete investigation of the trouble, and further work
in connection with it had to be deferred until the present year, when
the matter was taken up as soon as favourable weather conditions
started the spread of infection to the new crop of fruit.

A detailed study of the causal organism has now been carried out,
and a survey made of orchards in the infected area with a view to pre-
venting if possible a further spread of infection.

LITERATURE.

Two bacterial diseases of citrus have been described, both occurring
in America; the so-called “citrus canker” of Florida need not be more
than mentioned here; the macroscopic characters of the “canker”
are entirely different from those of the spot under consideration, it 1s
characterised by a proliferation and subsequent suberisation of the
cells of the affected tissues (4).

A much more closely related trouble however is reported from
the lemon growing sections of Southern California (6); it has been

Ann. Biol. 1 4

54. A Bacterial Spot of Citrus

appearing occasionally for the past three years, has gradually increased
and is now assuming some economic importance. This disease has
been named “ Black Pit of Lemon” a term which aptly describes its
effects on the fruit, where it causes circular-oval discoloured spots
5 to 20 mm. in diameter, the tissue being depressed somewhat below
the bottom of the normal oil glands into the white portion of the rind.

The cause of the disease proved to be an actively motile rod with
a single polar flagellum, and it has been named Bacterium citripuleale.
The description and illustrations of fruit attacked by this organism
correspond very closely with those of spotted lemons from the Western
Province, and it was surmised that the same organism would prove
to be the source of the trouble. A thorough investigation however
has shown that the organism causing the disease in South Africa is
quite distinct both in morphology and in cultural characters.

DISTRIBUTION.

The Botanical Division of the Agricultural Department has no officer
stationed within easy reach of the citrus orchards of the Western Pro-
vince; arrangements were therefore made with the co-operation of
Mr C. W. Mally, the Cape Entomologist, to systematically inspect
the citrus trees in the districts with a winter rainfall, but unfortunately
the exceptionally rainy season prevented this plan from being carried
out with any degree of thoroughness. A number of orchards however,
have been inspected and a large number of specimens of suspected
fruit sent to Pretoria for examination; and a certain amount of infor-
mation has been gleaned as a result of circularising the principal citrus
growers of the Union.

The disease is now definitely known to occur on a number of farms
in the region of Simondium and Lower Paarl. It extends along the
Berg River Valley and has been found occurring at Wellington, Sir
Lowry Pass and French Hoek. It has also been reported as occurring
at Stellenbosch, but no specimens of supposedly intected fruit from
that locality have been examined.

So far as can be ascertained the Transvaal is entirely free from the
trouble; there is certainly no sign of it at Warmbaths or Nelspruit,
and growers at Barberton state that during a careful examination of
their trees they failed to find anything at all resembling the bacterial
spot. The Rustenburg area also seems to be clean; one box of
oranges and lemons sent from there in response to enquiries proved

E. M. DorpGE 55

to be attacked by anthracnose (Colletotrichum gloeosporioides) which
causes discoloured spots on the fruit not very widely differing in
appearance from those caused by the bacillus. They are different
in colour however, not conspicuously sunken, and can be readily
distinguished when the pinkish spore pustules begin to develop.

During November and December of this year (1915), after this
article was practically completed, Mr Pole Evans, the Chief of this
Division, made a personal inspection of the orchards in the infected
areas. At that time there was little or no fruit on the trees so that
his inspection resolved itself into a search for the disease on branches
of trees and nursery stock.

Branch infections were very plentiful in the orchard at Simondium
where the disease was first detected, in an orchard near Cillie’s Siding,
and one in Zuider Paarl: at Simondium the organism attacks stems } to
1 inch in diameter. At the Elsenburg Agricultural College, the nursery
stock has been severely attacked but no signs of the trouble were found
in any of the other nurseries which were inspected.

He saw no evidence in the orchards of the Clanwilliam, Piquetburg
or Montagu Districts. It is apparently confined at present to the
valleys of the Groot and Klein Drakenstein and the Berg River Valley ;
in this area the fruit has been attacked on a number of farms, but branch
infections have only been detected in the localities mentioned above.

From field observations it would appear that the leaf is attacked
first and that the disease is then communicated to the stem through
the leaf stalk. As soon as the petiole and neighbouring stem tissues
are invaded the leaf falls, and this would account for the fact that very
few leaf infections were found on branches sent for examination.

SIGNS OF THE DISEASE.

The disease has been found occurring in nature on lemons of the
Mediterranean type, navel oranges and naartjes, the term “naartje”’
including both Mandarins and Tangerines. It has been induced by
artificial inoculation in lemon, orange (three varieties), naartje, shaddock,
grape fruit, citron and sour and sweet limes.

In describing the appearance of affected parts, and the cultural
characters of the organism, the colours named have been compared as
accurately as possible with those in Ridgway’s ‘Color Standards and
Nomenclature (5).

It was first noticed in Geneva lemons, on which it forms discoloured
sunken spots varying from 1 to 3 mm. in diameter: more frequently

4—2

56 A Bacterial Spot of Citrus

they are 5—10 mm. in diameter and depressed 1—2 mm. below the
surface of the healthy rind. The spots are more or less round, but
when a number occur in close proximity to one another, they coalesce
and result in the formation of large irregular blotches (Plates III and IV).

A close examination of the spot usually reveals a small wound in
the centre which has been the starting point of infection; in a few
cases a large diseased area showed numerous scratches, possibly the
result of swinging against a thorn in the high winds.

Young infections vary considerably in colour; one spot, 28 mm.
in diameter, was maize yellow, the centre buckthorn brown, others
were chamois with margin and centre of Mikado brown. In a few
cases there were one or more concentric rings of brown on a lghter
ground. They become much darker with age, varying in colour from
Argus brown through light seal brown to blackish brown; frequently
the dark central portion is surrounded by a reddish rim.

It is not a soft rot, and the spots are either leathery in texture or
else quite hard. A number of affected lemons were kept for some days
in a moist chamber and at the end of that time, dirty yellowish drops
of a viscid substance were seen oozing from the point of infection.
A similar circumstance was observed in lemons artificially infected
and kept in a moist chamber.

By cutting through the discoloured spots it can be seen that there
is a very definite line between the sound and the diseased tissues, in
some cases only about half the thickness of the rind is affected; in
others the organism penetrates right through the rind and invades
the pulp. The pulp below the discoloured rind becomes rather dry
looking and has a peculiar taste and smell; in severe cases it is dis-
coloured to a cinnamon brown, only the outer parts are affected and
the disease does not appear to spread in a tangential direction.

On oranges the spots are slightly different in appearance; they
average 20—30 mm. in diameter and are not so deeply sunken as those
on the lemons; the colour is usually buckthorn brown to Dresden
brown; I have never observed spots on oranges with the conspicuous
almost black discoloration so common in lemons (Plate VIII).

Naartjes become very badly infected: a number of naartjes sent
direct from the orchard shewed as many as 20—30 spots of varying
size on a single fruit, they were round to irregular and 1—25 mm. in
diameter; slightly sunken and varying in colour from something
between Mikado and Sayal brown to blackish brown (Plate VII). In
some large spots with an evident thorn prick in the centre the disease

EK. M. Dower 57

had gone right through the skin and penetrated into the pulp. The
latter was not noticeably discoloured but looked rather dry and had
a peculiar taste and odour.

On shaddocks by artificial inoculation a number of small spots
were produced not exceeding 8mm. in diameter; they were very
shghtly sunken, and cinnamon brown rufous in colour. On cutting
through these spots it was found that the disease had not penetrated
far into the rind; the tissues were discoloured to a dirty brownish
yellow to a depth of 1:5 mm., 7.e., just below the oil ducts. This
discoloured portion was bounded by a thin scarlet line which in some
cases marked the boundary between healthy and diseased tissues.
The red line is not always present, and when present is not always
continuous. The red colourmg matter disappeared at once when
pieces of the rind were fixed in hot acetic alcohol.

Grape fruit artificially inoculated developed infected areas not unlike
those on the lemons. They were sunken, rather irregular in outline,
up to 1 cm. diameter, cinnamon colour in the centre with a darker
rim, of Mikado or Sayal brown. Sour and sweet limes are also affected
in a similar way to the lemons.

On citrons inoculation with a pure culture resulted in the formation
of numerous reddened areas up to 12 mm. diameter, some of these were
slightly sunken; the colour was at first zinc orange to apricot orange,
and later became cinnamon brown.

The only variety of citrus fruit which has so far proved resistant
is the Seville orange, which has not been found infected naturally,
and which has up to the present resisted infection by artificial means.

Affected fruit cannot be kept for any length of time; the discoloured
spots readily serve as a starting point for numerous fungus infections
and these complete the work of destruction. Almost every fruit
infected with the spot disease, if kept for any length of time, was
completely destroyed by Penicillium spp. or, less frequently, by
Colletotrichum gloeosporioides.

The disease not only disfigures and destroys the fruit, but affects
the branches to a considerable extent. The organism seems almost
invariably to attack the stems just round the leaf bases; an area of
about 6—10 mm. diameter becomes water-soaked in appearance, the
petiole becomes involved and sometimes the basal part of the leaf, and
the leaf falls. The infection then spreads to the branch in the leaf
axil, surrounds it at the base and it withers and dries up. The infected
area on the main part of the stem becomes up to 20 mm. long, and

58 A Bacterial Spot of Citrus

goes half way round the stem; after a time the central part loses its
water-soaked appearance and turns brown (chestnut brown to olive
brown) and becomes slightly sunken (Plate IX, a and 6b). The dis-
coloration spreads out to the edges of the infected area which ceases
to increase in size. In transverse sections the discoloration does not
appear to extend into the wood, but if the bark be peeled off the surface
of the woody cylinder appears slightly yellow.

Towards the end of the season, 7.e., in the spring, a considerable
amount of gummosis may be observed from the affected. stem tissues,
the gum oozing very largely from the severed leaf bases. In a number
of affected branches received at this time of the year the leaf subtending
the flowering shoot had been destroyed, and it was a question whether
the whole inflorescence would not have dropped before any fruit was set.

If the disease were confined to the fruit it would hardly be expected
that it would cause any considerable damage in the districts with
a summer rainfall and a dry winter, but the fact of its attacking the
branches and causing the destruction of the leaves causes the trees to
present a defohated appearance towards the end of the season. It
may be anticipated therefore, that even in a climate where the fruit
ripened during the dry season considerable damage would be done to
the trees during the summer, and their vitality considerably lowered.
An exceptional winter rain such as fell in the Transvaal during last
July would be sufficient to start the spread of infection amongst the
fruit.

Leaf infections are comparatively rare, occasionally a small dis-
coloured area develops in the leaf tissues, a thorn prick or small wound
serving as the starting point of infection. The dark brown spot is
surrounded by a yellowish zone and the injured tissues eventually
fall out leaving a hole. (Plate IX, c).

INOCULATION EXPERIMENTS.

An organism was plated out from lemons received from Simondium
in November, 1914; with a pure culture of this a number of lemons
were inoculated by needle pricks, and in four to five days these showed
a number of small sunken dark-coloured spots. Owing to the pressure
of work and the fact that the lemon season was nearly over, the work
was discontinued, and was not resumed until July, 1915.

At the beginning of this month the disease again became very
active, shortly after the first steady rains in the Western Province,
and specimens of infected fruit (both lemons and navel oranges) were

EK. M. Dormer 59

received from the same locality as before. From these a pure culture
was readily obtained, and a yellowish bacillus was isolated. With
these cultures an infection experiment was carried out; sound lemons
obtained from Warmbaths, where the disease does not occur, were
dipped in 1:1000 mercuric chloride and then washed in running
water. They were then divided into four lots, each lot placed in a
large glass jar with a drop-on lid, and treated as follows: in jar (1)
the lemons were pricked with a sterile needle; the fruit in (2) was
lightly pricked with a very fine needle then atomised with a suspension
of an agar culture of the bacillus; in (3) the lemons were inoculated
by needle pricks, and in (4) were injected with a hypodermic syringe.
At first the lemons were kept at room temperature which even in the
day time was only 13° C. After 48 hours there were slight indications
of infection, the tissues in the immediate neighbourhood of the needle
pricks were slightly discoloured.

After four days infections were quite marked, the oil glands in
particular were discoloured, and the tissues in the vicinity of the wound
were becoming slightly depressed. The fruit was then removed to
the incubator at 25° C. the spots subsequently developed much more
rapidly and after ten days all the inoculated fruit showed well developed
infections. The fruit in the four jars may be described as follows:

(1) Control, pricked with sterile needle, quite sound, wounds caused
by needle pricks barely visible on close inspection. (Plate V, a, right
hand figure.)

(2) Pricked, then atomised with suspension of culture. There was
an infection for almost every needle prick. As many as 100 spots
on one lemon, the colour of the spots cinnamon buff to snuff brown
and mummy brown; the diseased tissue sunken, and the spots small.
(Plate V, a and 0.)

(3) Inoculated by needle pricks, spots up to 6—8 mm. diameter,
colour as in (2), from 20—65 sunken spots on each lemon. There was
a viscid drop of dirty yellowish substance oozing from the centre of
each spot; on examination this proved to be composed of capsuled
bacteria, and on being plated out, yielded a pure culture of the organism
with which the fruit had been inoculated. (Plate VI, a.)

(4) Suspension of culture in distilled water injected with hypo-
dermic syringe; spots varied from 3 mm. to 30 mm. diameter; they
were sunk 1—14 mm. below the surface, 4—32 spots on each fruit,
some of which did not appear to be in connection with a wound. The
discoloration varied from Sayal brown to mummy brown. Viscid

60 A Bacterial Spot of Citrus

drops of bacterial slime were also oozing from the centre of these spots.
(Plate VI, b.)

An infection experiment on a larger scale was carried out with a
quantity of citrus fruit obtaimed from the Government Orchard at
Warmbaths; the fruit was picked from the trees in the writer’s pre-
sence, taken direct to the Laboratory in boxes and the inoculations
carried out two days later. There was no trace of the disease on any
of the fruit in the orchard.

The cultures used were from four different sources: from natural
infections on lemon, on orange and on naartje and from an artificial
infection onlemon. In the schedule the last is distinguished as lemon II,
the first as lemon I. In each case young cultures on nutrient agar were
employed for needle prick inoculations, and young cultures suspended
in sterile distilled water for atomising.

All controls were pricked with a sterile needle; inoculations were
carried out in four different ways, some were atomised with a “ glaseptic”’
sprayer without any previous wounding, others were lightly pricked with
a very fine steel needle and then atomised, into a third lot of fruit the
organism was introduced by means of needle pricks, and the bacillus
was injected into a fourth set by means of a hypodermic syringe.

No. of Source of :
Fruit Kind of Fruit Culture Method of Inoculation Result

6 Ripe lemons... Sh <= Sterile needle... ... No infection

7 = os aoe As ... Lemon! Pricked then atomised Positive

8 Green ,, abe Ht: sth ede ab fx a seis af

i Ripe 5, I Needle pricks ... nae *

8 Green ,, tL be iG ite au. $5

i Ripe ;; I Hypodermic syringe... »

7 Green il Pi o is 3

5 Ripe ,, $00 a: = Sterile needle... ... Noinfection

5. Ripe ;; se ane ... Orange Pricked then atomised Positive

5 Green ,, 35 — me os

6) Ripe” |; St aie ide r Needle pricks... x0 x.

5 Green ,, sf - fi: wae Ss

Upelvipe mess a0 ae oA nS Hypodermic syringe... +

7 Green ,, Sc sas a00 38 5 on ee BS

5 Navel oranges ... aa wee Sterile needle... .... No infection

5 PD x Aaa sue ... Lemon | Pricked then atomised Positive

6 5 Ab ae Rae ... Orange I ts 4 FA a

5 79 55 Sak “50 ... Lemon I Needle pricks... atts Pe

5 3 a ee Seti ... Orange se 3 Seid aes *

iD $ a 5c ae ... Lemon I Hypodermic syringe... 5

5 as Ss ae : . Orange 5 re Pr

8 Valencia late oranges ... oa Sterile needle... ... No infection

E. M. Dome@eE 61

No. of Source of
Fruit Kind of Fruit Culture Method of Inoculation Result
15 Valencia late oranges ... Lemon | Needle pricks Positive
8 a Pe x sll ... Orange | * ie at bets
10 Seedling oranges(Parson Brown) — Sterile needle No infection
11 - » LemonI Needle pricks Positive
10 a - Orange 53 x3 a5
3. Seville oranges -— Sterile needle No infection
3 : .. Orange Needle pricks Doubtful
4 : 74 Lemon | is * aes
3 5 . Orange Pricked then atomised EA
1 Citron — Sterile needle No infection
1 5 . Lemon II Needle pricks Positive
1 55 . Orange a An eae 33
i x . Lemon IL Pricked then atomised
1 sc dor . IL Hypodermic syringe 99
8 Naartjes , IL Atomised without wound Negative
4 Navel oranges ... . Orange 56 x 99
4 Sweet limes . Lemon II “3 a 35
4 Naartjes — Sterile needle No infection
5 3 . Lemon II Needle pricks Positive
5 = . Orange ” . ”
6 Sour limes — Sterile needle No infection
6 Ae . Lemon IL Needle pricks Positive
6 ae . Orange 33 55 Ses =
7 3 aA . Lemon II Pricked then atomised Ne
5 Navelencia oranges .» IL Atomised without wound Negative
4 Sweet limes Il Needle pricks Positive
4 pe ' . Orange er 5 ais 3
5 ; . . Lemon II Pricked then atomised .
4 a7 . Orange x Pf Pe os
5 ; ¥ . Lemon Il Hypodermic syringe x
7 Sour limes _ Sterile needle No infection
FS ; a . Lemon II Atomised without wound Negative
5 ns . Orange = oS ”»
5 py th . Lemon Il Hypodermic syringe Positive
12. Lemons .» IL Atomised without wound Negative
1 Shaddock ate = Sterile needle No infection
i Fe . Lemon II Needle pricks Positive
l 5 . Orange 29 os 2
4 Grape fruit — Sterile needle No infection
4 - c . Lemon II Needle pricks Positive
4 a . Orange ~ a ‘A
1 a es . Lemon Il Hypodermic syringe a
1 ‘ a S52 . Orange 35 fe .
16 Rough lemons ... . Naartje — Needle pricks

It will be observed from the above schedule that with the exception
of the Seville oranges all inoculations were positive on fruit which had

§2 A Bacterial Spot of Citrus

been pricked or otherwise slightly wounded. None of the lemons or
other citrus fruit which were atomised without wounding developed
any spots, but I do not regard this as conclusive evidence that the
organism cannot find its way through the stomata, for several reasons.
In the first place, the fruit which was used for the experiment had
been hanging on the trees in a very dry atmosphere, and the outer
tissues of the rind had become comparatively dry and leathery, and
it was noticed that even after wounding these infected less readily
than those used in the preliminary experiment which were cut after
an exceptional spell of rainy weather; also in the first experiment
a number of infections appeared on lemons which were injected with
a hypodermic syringe at points where no injections had been made,
and on naturally infected fruits on many spots where no wound can
be found, and sections through the tissues show the bacteria in the
substomatal cavity, but this will be alluded to later.

Further inoculations will be necessary before anything conclusive
can be arrived at with regard to the possibility of stomatal infection,
but obviously the most frequent method of infection is through wounds.

On the trees the spots are only found on ripe fruit, and two or three
farmers have reported it as occurring only on fallen fruit or developing
in the store room. The organism is very probably a soil bacillus which
first invaded fruit lying on the ground and has now taken on a parasitic
habit. It was found quite possible to infect green lemons, although
the organism attacked them less readily than the ripe fruit.

The organism loses its virulence rather rapidly on artificial media ;
this was clearly shown by inoculating two similar sets of lemons, one
with a freshly isolated culture, and the other with a culture isolated
about a month previously and transferred fifteen times. On the first,
signs of infection were visible after 48 hours and distinct spots in four
days; on the second, the discoloration was much delayed; it was not
visible till the fourth or fifth day and the infected areas comparatively
small and poorly developed.

The identity of the organism in fruit and branch was also tested.
A culture of this bacillus isolated from a leaf base infection was used
to inoculate a number of sound lemons with positive results.

A strain isolated from lemons sent from the Constantia Wine Farm
produced a few positive infections on some young trees growing in
tins in the Laboratory grounds. From these the organism was again
isolated and employed with positive results to infect a number of lemons.

EK. M. Dorper 65

PATHOLOGICAL HIsToLoGy.

The organism was most easily studied in the rind of affected fruit ;
satisfactory preparations were obtained by fixing in hot acetic alcohol
and staining with Ziehl’s carbol fuchsin and light green.

This is essentially a disease of the parenchyma, the lignified tissues
of the stem are not invaded, and the organism is not found in the small
fibro-vascular bundles in the rind. The bacillus does not cause hyper-
plasia,

As has been stated, in the majority of cases infection takes place
through a wound; the very slightest abrasion with a fine needle being
sufficient. No definite evidence has yet been obtained by inoculation,
that stomatal infections are possible, but in a number of sections,
particularly those through incipient spots on naartje rind, the location
of the bacilli strongly suggested that they had effected an entrance
through the stoma. (Plate X, a.) In such cases the stomatal
cavity was occupied by a mass of bacteria and the surrounding inter-
cellular spaces had also been invaded. The rods multiply very rapidly
in the intercellular spaces, until the latter become much distended
and bulge into the cell cavity. Considerable tension is thus set up
and eventually the wall gives way and the bacilli force an entrance
into the cell.

The cell cavity soon becomes entirely filled, but for a considerable
time the remaining walls are intact so that were it not for the absence
of any intercellular spaces it would appear that the organism was
intracellular. The cell contents become disorganised and disappear with
the exception of the protoplasmic lining of the cell wall which contracts
slightly away from the wall and stains deeply with fuchsin. The cells
at this stage consist of a central mass of bacteria surrounded by a
deeply stained sack of disorganised cell contents and then by the
partially destroyed cell wall. The staining reaction of the latter is
not altered. Finally the intervening walls give way and the tissues
become completely disorganised. (Plate X, c.)

The oil gland is often completely filled with a vast number of rods;
it was afterwards found that the organism grows extremely well in
the presence of orange oil and lemon oil. The cells surrounding the
oil glands are also very frequently attacked.

64 A Bacterial Spot of Citrus

MorPHOLOGY.

The organism is a rather slender rod with rounded ends; the living
rods diffusing from discoloured cells in the rind of a lemon measured
‘8 to 3-2 by -+5 to -7.; in a smear from the same source stained with
dilute fuchsin they measured 1-4 w by -45 to -7 w, the majority in each
case were about 1-5 x -5 to-6y. Similar measurements were recorded
for rods from diseased tissues of spots on oranges and naartjes. The
bacilh in the viscid mass oozing from affected fruit are slightly stouter
on the whole than those in the tissues, most of them are -6 to -65 4 in
thickness, in length there is not much variation.

Measurements of rods from a 24 hour culture on nutrient agar and
in nutrient broth coincided with those of rods taken direct from the
tissues of the host plant, but on gelatine they are considerably shorter,
the majority measuring -8 tol» by-5 to 6. In Uschinsky’s solution
short forms also predominate.

In old agar streaks the rods vary very much in length, some are
almost spherical, others attain a length of 5 before dividing; some
stain intensely, others faintly or unevenly.

No long filaments have been observed in the pellicle on sugar broth,
the longest unsegmented forms noticed in these media were about 8 uw
in length.

The sediment in three months old broth cultures consists for the
most part of rather short rods, some of which stain intensely with
gentian violet, but the majority stain faintly or unevenly. In old
cultures on steamed potato there are also a large number of individuals
which only stain faintly, and these are considerably swollen so as to
be ellipsoid or almost spherical in form.

Fission. The multiplication of the organism by fission was studied
on the agar hanging block; a smear was made from a 24 hour old
broth culture which had been incubated at a temperature of 35° C.,
the preparation was illuminated by a Nernst lamp, and the lenses used

uN”

were a Zeiss 4,” oil immersion objective and No. 12 compensating
ocular. With this magnification the alteration of the rods in size
and form could be easily followed, and drawings were made with the
camera lucida at frequent intervals. The temperature of the room
was approximately 26° C,

At 9.15a.m. a single rod was selected for study (Plate XII, a).
It at once elongated somewhat, became constricted and by 10 a.m.

had divided into subequal rods. The upper of the two new individuals

E. M. DoImpGE 65

thus formed had divided again by 10.45, the lower became constricted
and divided soon after. Subsequently each rod took approximately
20 minutes after separation to attain to its maximum size, and then
immediately became constricted and fission was complete in another
20 minutes. Each individual became slightly curved or bent when
it became constricted previous to fission, and when division was com-
pleted the resulting rods were not exactly equal. The newly formed
individuals at once became separated by an appreciable gap, their close
contact being prevented by the capsule which could be detected by a
careful adjustment of the illumination.

At 1 p.m. the young colony consisted of 16 rods; by 1.30 the
number had increased to 22 and it was no longer possible to make
accurate drawings as many of the rods had oriented themselves with
their long axes parallel to the line of vision and they were overlapping
in various directions.

No further observations were made until 9.15 on the following
morning; at that time the surface of the block was covered with highly
refractive moruloid masses of bacteria; the capsuled bacteria in each
mass were surrounded by a common envelope with a high refractive
index, and outside of this were a large number of rods lying free (Plate
XII, 6). When first examined the free rods round the masses near
the edge of the block which were probably better supplied with oxygen
than the rest were in active motion, and large numbers of them had
made their way from the agar to the surface of the cover glass beyond
which was covered with a thin film of moisture. After about 15
minutes’ exposure to brilliant illumination from the Nernst lamp all
these rods came to rest.

Grouping. The rods are most commonly single or in pairs; loose
chains are found in the pellicle and also in sediment in some sugar
broths, particularly on nutrient broth with 2 per cent. laevulose,
but these do not as a rule consist of more than 20 or 30 elements
(Plate XIII, a) and they lie amongst large numbers of single rods.
There are no chains in the ring above the liquid.

On solid media the organism almost invariably forms itself to a
greater or less extent into the capsuled, moruloid masses described
above, they are also found in the pellicle on ordinary nutrient broth.
If such masses be mounted in water they do not resolve themselves
into chains but isolated rods disentangle themselves and swim away
in a state of high activity.

Motility. The organism is often sluggishly motile when it is taken

66 A Bacterial Spot of Citrus

direct from the tissues of the host, and if a small portion of the viscid
mass oozing from affected areas on the fruit be mounted in water the
rods diffuse out and immediately become very active.

It is extremely active in young cultures; an examination under
the low power of the microscope of 18 to 24 hour old colonies on
nutrient agar shows the outer part of each colony as a swirling mass
of actively motile rods. In liquid media the bacillus retains its
activity for a considerable length of time; chains from the pellicle of
a 16 days old broth culture separated and immediately became very
active on being transferred to water.

The motility of the organism can be very well studied in prepara-
tions made from a young agar culture and observed with the dark
ground illumination. The rods move forward with a steady sinuous
motion which is frequently interrupted by rapid oscillating movements.

The flagella stain quite readily; in the living state they may be
stained by Straus’ method(). The position of the flagella can be well
made out, they are obviously peritrichous but are in too rapid motion
for any accurate estimate of their number. With Ellis’ modification
of Loeffler’s method very satisfactory preparations were obtained.
There are 5 to 10 long peritrichous flagella: these are several times the
length of the rod, measuring roughly 8 to 12. They are not distri-
buted evenly over the surface but frequently consist of one or two groups
of 3 to 5 which are crowded up towards the poles (Plate XII, c).

Capsules. The viscidity of the substance oozing from diseased
tissues and the leathery or mucilaginous texture of many cultures on
solid media at once suggest the presence of a capsule.

A capsule may almost always be detected and may be stained
either by one of the special capsule stains, by a dilute solution of fuchsin
or with carbol gentian violet. The viscid drops oozing from the fruit
consist of capsuled rods embedded in a tenuous slime which appears
granular when stained by MacConkey’s method (Plate XIII, d).

In an impression preparation of a young agar colony all the rods
were definitely capsuled (Plate XIII, c) and in the older parts of
the colony become crowded together into small tough masses which
are firmly embedded in the medium. In certain cases streak cultures
on agar take on a roughened or shagreen-like appearance, and the
streak becomes leathery and can be torn in a strip from the surface
of the medium. Old streaks of raised, shining, more homogeneous
appearance are usually mucilaginous; the difference in texture appears
to have some relation to the amount of water present in the medium.

Kk. M. Dower 67

No capsules have been observed around rods within the tissues
of the host. :

The capsule around each rod in an impression preparation stained
with fuchsin measures roughly 1-5 ~ in diameter and it exceeds the rod
in length by about -5 uw at each end. A few rods may be seen without
capsules and empty capsules are quite common, but it was not clear
whether the rods escape from the capsules or whether they had been
torn away in making the preparation.

Involution forms were observed in broth cultures containing high
percentages of sodium chloride. They are very common in 48 hour
old cultures in broth containing 9-5 per cent. to 10-5 per cent. NaCl.
The rods comparatively are very large, some grow out into filaments
as much as 100 long, and they are irregularly swollen and contorted
in every direction (Plate XIII, e). Quite a number are club-shaped.
They stain intensely with aniline gentian violet and when transferred
to ordinary nutrient broth rapidly divide and produce rods of normal
form.

Aberrant forms are also found in old potato cultures. A smear
made from a 12 days old potato culture consists of very minute intensely
staining capsuled forms, some of which are almost spherical and
slightly swollen rods of normal length which stain very feebly or
irregularly. The latter were only faintly stained in half-an-hour by
carbol gentian violet (Plate XIII, 6).

Staining reactions. The bacillus stains well by all the ordinary
methods, but varies considerably in its reaction to the various dyes.
In testing the action of the different stains smears were made from
a 24 hour old streak on nutrient agar. Dilute aqueous solutions were
allowed to act for two minutes; in this time the thionine and methylene
blue hardly stained the rods at all, fuchsin gave a distinct though
rather faint coloration and the rods staimed with gentian violet were
very intensely and quite sharply stained. Carbo] gentian violet also
gave a very strong intense stain in half a minute and brought out the
capsules which were not stained by the aqueous solution. Ziehl’s
carbol fuchsin gave a similar result; carbol methylene blue and carbol
thionine also stained very faintly in half a minute, but Loeffler’s blue
gave much better results in the same time though not nearly so intense
a stain as the gentian violet and fuchsin.

The organism is Gram-positive, and stains blue by the Ziehl Neelson
method, 7.¢., it is not acid fast.

Spores. No spores have been observed.

68 A Bacterial Spot of Citrus

CULTURAL CHARACTERS.

Nutrient agar colonies. The organism grows very rapidly on
nutrient agar and when first isolated from the tissues the speed with
which the colonies appear varies considerably; after preliminary
cultivation however, more even results are obtained. The following
is a description of the development of typical colonies at 25° C.

After 24 hours in thinly sown plates the colonies are 1—4 mm. in
diameter, they average 2 mm., and are round-irregular in shape; by
transmitted light translucent with a small opaque centre, pearly white
to bluish, yellowish by reflected light. Under the microscope the
colony is granular with a grumose centre, the granular margin being
composed of a swirling mass of actively motile rods. Submerged
colonies are minute opaque, mostly bi-convex.

After two days the average diameter of the colonies has increased
to 4mm., and some measure as much as 7 mm., they are coppery by
transmitted light with a bluish margin, straw yellow by reflected light.
There is a small opaque centre surrounded by a rather dense sub-opaque
ring and then by a translucent margin. Under the low power of the
microscope the central portion is grumose, the outer finely granular.

Submerged colonies have increased somewhat in size and now have
the appearance of two or more lenticular bodies placed at different
angles, or of a single lenticular body with sub-spherical outgrowths.

After three days the colonies average 5—6 mm. in diameter, the
largest are 10 mm., they are round-irregular, umbonate; Naples
yellow to buff vellow by reflected light, coppery throughout by trans-
mitted light; under the microscope they are much more dense than
before except at the edge, and the whole of the central part is grumose ;
there is no longer any motion visible with a low magnification. Sub-
merged colonies have increased in size up to } mm.

After four days the colour has become mustard yellow and long
feathery crystals have made their appearance in the medium.

There is usually no further change after the fifth day; the colonies
then measure up to 15mm. in diameter, the margin is entire or undulate,
and there are frequently two or three concentric rings. The surface
of the agar has become distinctly whitened especially in the immediate
vicinity of the colonies.

The growth in this case does not become tough or leathery, it is
always creamy in consistency.

EK. M. DoimpGE 69

The size of the colonies is very much affected by crowding. (Plate
XITI, b, ¢, d.)

Nutrient agar streak. A well developed agar streak appears in 24
hours at 25° C.: it is smooth, shining, about 3 mm. wide and grumose
in the centre. In three days this has become 3—5 mm. broad,
primuline yellow, the central part being pulvinate, and the margin
entire or slightly undulate.

After 12 days the surface of the agar has become very decidedly
whitened except for the parts covered by the bacterial growth; long
feathery crystals point downwards into the medium from the under
surface of the streak. The condensation water is very heavily clouded.
A second kind of crystal is formed on the surface of the agar under
the condensation water, very similar to those illustrated by Smith as
occurring in cultures of the olive tubercle organism (7).

The texture of the streaks in three weeks old culture varies very
much with the amount of water present in the medium and in the
atmosphere. Some remain creamy, but contain small tough moruloid
masses; others become very tough at an early stage, large strips may
be pulled away on the platinum needle and when this is done the growth
stretches somewhat and has the appearance of a number of spherical
colonies in a tough homogeneous matrix. Most commonly the growth
is raised and almost homogeneous to the naked eye. It is then very
mucilaginous, and in colour and texture is not unlike plastic sulphur.

If the surface of the sloped agar be bathed in lemon or orange oil
a very luxuriant growth is obtained which does not become tough or
mucilaginous.

Nutrient agar stab, the best growth is at the surface.

Glucose formate agar. Streaks were made on this medium as a
control for those kept under anaerobic conditions; the growth is very
similar to that on ordinary nutrient agar, but slightly more luxuriant.

Sulphindigotate agar. This medium was used for a similar purpose ;
a very luxuriant growth was obtained, on the upper drier portion of
the slope a number of small discrete colonies were produced which at
first absorbed the colouring matter from the medium and became
blue. The lower part of the slope was almost covered by a wet shining,
yellowish growth which was grumose in the centre. The colour gradu-
ally disappeared from the medium.

Lemon and Orange glucose agar. Colonies on these media are
similar to those on nutrient agar but considerably smaller, only 2-3 mm.
in diameter after three days; they are very markedly umbonate in

Ann. Biol. m1 5

70 A Bacterial Spot of Citrus

the centre, and concentric rings are visible by transmitted light;
under the low power of the microscope they are grumose in the centre
and coarsely granular near the edge; in colour they are dirty yellowish
white. After five days the diameter has increased to 5 mm., and the
colonies have a number of distinct concentric zones, by transmitted
light the two central zones are yellow and the remainder bluish white.

Lemon and orange glucose agar streaks! are dirty yellowish white,
glistening, very slightly raised, 3—10 mm. broad, and the condensation
water is rather heavily clouded. On the lemon glucose agar the streak
had a shagreen surface, on orange glucose agar the surface was smooth
but moruloid in the centre.

Orange laevulose agar. Streaks on this agar are similar in character
to those on orange glucose agar but are considerably more luxuriant.

Beerwort agar. The organism grows very well on this medium,
it forms a shining streak with a convex central portion and a flat
margin with a slightly undulate edge; the breadth of the central
portion is 4—5 mm., of the margin 1 mm. There is a copious yellow
growth in the condensation water.

As it dried up the surface of the raised central part becomes corru-
gated and it is decidedly viscous.

Bean agar is a fairly good medium, streaks are smooth, yellowish,
somewhat umbilicate, and about 3 mm. diameter.

Cabbage agar. A very characteristic growth forms on streaks of
cabbage agar. It is not very abundant and consists of a number of
spherical, opaque, yellowish colonies standing up above the surface
of the medium, and disposed irregularly in a flat translucent streak
1—4 mm. wide. The surface of the agar becomes conspicuously
whitened in the vicinity of the streak.

Loeffler’s blood serum is not liquefied, the organism forms a shining
yellow growth along the needle track.

Starch jelly. No growth.

Nutrient gelatine colonies are just visible to the naked eye after
24 hours at 20°C. as minute glistening points. In two days they
attain a diameter of 1—3 mm., and are round, glistening, slightly
convex, coppery by transmitted light, creamy by reflected light. Under
the microscope they are rather coarsely granular and inclined to become

1 Orange or Lemon glucose agar was made as follows: 50 grammes of fresh rind was
boiled for 30 minutes with 500 c.c. distilled water, filtered, then 10 ers. dextrose added
and solidified in the usual way with 1°5 agar powder.

In Orange laevulose agar, laevulose was substituted for glucose,

EK. M. DomweéE al

grumose in the centre. Submerged colonies are minute, white and sub-
spherical.

Seven days old colonies are up to 4 mm. diameter, apricot yellow,
they are beginning to sink in the centre owing to the softening of the
gelatine, the growth is shghtly viscous. In nine days the gelatine
in the immediate vicinity of the colonies is liquefied, and in 14 days
liquefaction is complete. The colonies are still almost intact and
floating in the liquefied medium.

Nutrient gelatine shake cultures develop very numerous minute
colonies, the majority of which are within 1 cm. of the surface of the
medium, which is slowly liquefied.

Nutrient gelatine stab. The organism grows comparatively slowly
at 20° C., the growth along the needle track being barely visible at
the end of 24 hours. After seven days there is a fairly thick line of
growth all along the needle track, the surface colony is slightly sunken
in a small crater of liquefied gelatine. In 14 days liquefaction is napi-
form, the surface growth sinks to the bottom of the liquefied gelatine,
continually re-forming and sinking, thus forming a considerable amount
of sediment. The liquefied portion is clouded, and there are very
numerous minute opaque yellowish flocculi in suspension throughout
it. (Plate XI, a.)

After 30 days the liquefaction is broadly saccate and extending to the
bottom of the tube, and in 6—7 weeks the gelatine is completely liquefied.

Nutrient gelatine streak. In three days there is a shining yellowish
streak about 2 mm. wide and slightly raised in the centre. Liquefac-
tion begins on the fourth day; a small pocket of gelatine liquefying
near the bottom of the streak. The following day there is a groove
along the surface of the slope, from which the liquefied gelatine has
run carrying the yellowish growth to the bottom of the tube. A fresh
streak forms in this groove, in its turn to be carried to the bottom of
the tube, and this process is repeated several times. The yellow growth
does not diffuse readily into the liquefied medium.

Potato. On potato the bacillus grows well, a thin spreading yellowish
growth almost covers the sloping face of the cylinder in 24 hours at
25° C.; after three days the growth is primuline yellow to mustard
yellow, glistening, spreading, covering all the lower part of the cylinder
except where it is in contact with the tube and the greater part of the
upper drier portions. No growth is visible in the liquid in the bulb.
The potato is distinctly browned, the discoloration being most marked
in the tubes containing glycerinated potato,

72 A Bacterial Spot of Citrus

Carrot. The water in the bulb of the Roux’ tube is distinctly
clouded, on the carrot there is a plentiful, colourless wet-looking, shining
growth spreading over the medium.

Turnip. Growth is much less abundant than on carrot, it is not
quite colourless but has a yellowish tinge.

Beet. Here the growth is very similar to that on carrot but is
distinctly yellow, and there is a yellowish ring at the surface of the
liquid in the bulb of the Roux tube.

Orange and lemon rind steamed show after three days glistening,
dirty yellowish masses of slightly viscid bacilli raised above the surface
of the rind. ‘

Nutrient bowllon (+15 Fuller). Nutrient broth clouds in six
hours at temperatures from 25° to 35° C., and in 24 hours there is a
distinct suggestion of pellicle formation. The pellicle is thin and falls
to the bottom of the tube if the latter is shaken slightly. Three months
old cultures are almost clear; there is a small amount of dried up
growth on the tube where the surface of the medium originally was
and a considerable amount of deposit. The rods in the deposit are
living and multiply readily when transferred to fresh tubes of broth.

Pellicle formation takes place most readily on broth with an acid
reaction, on alkaline broth little or no pellicle is formed. After stand-
ing for some days flasks of acid broth showed a thin pellicle in which
were suspended numerous more opaque portions which are shining,
yellowish and rather greasy looking. In some lights the pellicle is
distinctly iridescent.

The ring above the liquid does not become tough or mucilaginous,
and the pellicle is flocculent.

Sugar bouillons. The growth in dextrose broth is not particularly
good; a heavier clouding was obtained in broth containing maltose,
lactose and galactose, and the best growth of all in 2 per cent. laevulose
broth.

Intmus milk. In milk the bacillus grows comparatively slowly;
no change could be observed for seven days, after that period the milk
became slightly more acid in reaction and the litmus was then gradually
reduced. After 10 days at 35° C. the casein coagulated and there was
a gradual extrusion of whey. The whey was not abundant, almost
clear, and yellowish in colour, a sediment of yellow bacteria gradually
accumulated on the surface of the coagulum; the latter did not
become peptonised.

Milk serum, filtered through a porcelain candle, proved to be

E. M. Dormer 73

a favourable medium. ‘The liquid became turbid and a decided whitish
pellicle formed on the surface.

Dunham’s solution became only very slightly clouded; no pellicle
and very little sediment.

Cohn’s solution. No growth.

te Uschinsky’s solution at first is only slightly clouded but a thin tough
pellicle forms; this increases in size, remaining adherent to the sides
of the tube and becoming saccate and dipping below the surface of
the solution in the centre of the tube. This pellicle finally sinks to
the bottom and the liquid then becomes densely clouded and baryta
yellow to buff yellow in shade. There is a decided ring above the
liquid which becomes ochraceous tawny to cinnamon brown.

Cabbage broth. There is not much clouding and only a small
amount of sediment. The pellicle however is quite conspicuous, it
consists of vast numbers of minute, opaque, yellowish bodies and
presents a frosted appearance.

Lemon juice filtered through a porcelain candle and undiluted
proved to be too acid a medium for growth.

Nitrate broth is heavily clouded but no pellicle forms owing to the
vigorous evolution of gas. The broth is covered with froth to the depth
of 1 cm., for some days.

PHYSICAL AND BIOCHEMICAL FEATURES.
Enzyme production.

Proteolytic enzymes. Milk serum cultures tested at the end of five
days gave only negative results; after 10 days however, although
the culture gave negative reactions for proteose and peptone there
was a faint but unmistakable reaction for tyrosin which was absent
in tests made from the control flask.

Diastase. Potato cylinders on which the organism had been grown
gave a distinct red brown reaction with iodine indicating the presence
of amylodextrin. The same reaction was observed in peptone water
containing potato starch, but in neither case was there any reactior
for sugar with Fehling’s solution.

Nutrient bouillon cultures to which were added 2 per cent. thymo
and equal quantities of starch paste did not precipitate Fehling’s
solution after six to eight hours.

Invertase was also absent from broth cultures.

74 A Bacterial Spot of Citrus

Acid production. In nutrient broth containing 2 per cent. dextrose,
laevulose, saccharose, lactose and glycerine and in ordinary broth
without the addition of any carbohydrate there is no acid production,
but the medium in 10 days becomes slightly more alkaline, but only
by 10 degrees or less of Fuller’s scale.

In peptone water tinted with litmus, however, the organism produced
acid with several carbohydrates. The following table shows the
behaviour of the organism in tubes containing Dunham’s solution and
various carbohydrates and other substances.

Dextrose Slightly acid, litmus somewhat reduced
Laevulose Acid reaction marked

Galactose Slightly acid, litmus considerably reduced
Maltose Distinctly acid

Lactose No change in reaction

Saccharose Slightly acid, litmus somewhat reduced
Dextrin No change in reaction, litmus completely reduced
Starch No change in reaction

Glycerine No change in reaction

Mannite Distinctly acid

Sodium formate No change in reaction

Sodium citrate No change in reaction

The bacillus therefore produces acid in the presence of laevulose,
dextrose, galactose, maltose, saccharose and mannite, most marked
with laevulose.

Alcohol production. ‘The first distillate from a culture in 2 per cent.
laevulose broth was tested for alcohol, etc. The distillate was divided
into four portions, to the first was added Lugol’s iodine, then a little
NaOH solution; on heating the mixture there was a distinct smell
of iodoform. The second portion tested with Schifi’s reagent gave
a negative reaction for aldehyde. To 10 c.c. of a third portion was
added 2:5 ¢.c. 25 per cent. sulphuric acid and a crystal of potassium
bichromate. This was distilled, and in the process the bichromate
was reduced to a green colour, and the distillate smelled strongly of
acetaldehyde and reacted with Schiff’s reagent.

A similar test with distillate from control flask gave only negative
results. It is evident therefore that an appreciable amount of alcohol
is produced in sugar broth cultures.

Ammonia production. The distillate from cultures in nutrient
broth gave a positive reaction for ammonia with Nessler’s solution.

Indol is produced both in Dunham’s solution and in nutrient broth
incubated for 10 days at 25° C., but all tests for Phenol were negative.

EK. M. Dore 75

Pigment production has been described in various media in connection
with the cultural characters of the organism. The pigment is insoluble
in water (hot and cold), alcohol, ether, chloroform, carbon bisulphide
and dilute acids and alkalis.

Reducing agent formation.

Colour reduction. The reducing power of the bacillus is very slight;
it was tested in nutrient broth tinted with various coloured solutions.
Rosolic acid and indigo carmine are not reduced. Methylene blue
becomes peacock green in two days and in five days is completely
reduced. Neutral red which was at first flame scarlet in four days
was reduced to orange buff but was completely reduced only at the end
of 15 days.

Litmus showed no sign of reduction until after 15—21 days when
there was a small amount of colourless liquid at the bottom of the tube,
after a few days the colour again became even.

Reduction of nitrates. In nitrate broth the nitrates are reduced
with evolution of gas, the evolution being most vigorous at 25° C.
from the second to the fourth day, on the fifth day it ceased. The
tubes were tested at the end of the tenth day and gave no reaction
for nitrate or ammonia.

In nitrate peptone water no gas is evolved and after 10 days the
solution gives positive reactions for both nitrates and ammonia.

In control tubes both tests were negative.

Gas production. No gas is evolved in fermentation tubes containing
bouillon or peptone water and any of the substances mentioned in con-
nection with acid production. The organism grows in the closed end
of the fermentation tube but frequently the clouding is only slight.
In iron peptone and lead peptone solution there was a distinct blacken-
ing of the precipitate as compared with the control tubes; a small
amount of sulphuretted hydrogen is therefore present.

Almosphere. The organism is a facultative anaerobe, as suggested
by its ability to grow in the closed end of the fermentation tubes.

Four sets of cultivations were prepared, one of which was sealed
in Buchner’s tubes containing an alkaline solution of pyrogallic acid,
and the other three were each placed in a Bulloch’s apparatus.

From the first the air was exhausted as completely as possible so
that the organism was growing under reduced pressure; from the
second the oxygen was absorbed by pyrogallic acid, and the air in the
third was displaced by CO,.

76 A Bacterial Spot of Citrus

The bacillus grew almost as luxuriantly under reduced pressure as
in the control tubes. In the absence of oxygen, 7.e., practically in an
atmosphere of nitrogen, and in CO, growth was comparatively very
slow and restrained.

Temperature. The bacillus grows through a wide range of tempera-
ture, and very rapid multiplication takes place at all temperatures
from 25° C. to 38°C. The optimum temperature is 35°C. At 15—
18° C. the organism grows very slowly, and at 0° C. no clouding can
be observed in tubes of nutrient broth. The organism is not killed
by long exposures to this temperature, but rapidly clouded tubes trans-
ferred to a warmer temperature after 10 days. The upper limit for
growth is 43° C., broth is not clouded at 45° C. and the bacillus is
destroyed by several days exposure to this temperature.

The thermal death point (in standard 10 c¢.c. tubes 10 minutes
exposure) is 62° C.; dry on cover slips it is 110° C.

Reaction of medium. The organism is not very sensitive to the reac-
tion of the medium and grows almost equally well in broth with reactions
of + 15 to + 25, the optimum however is about + 20 of Fuller’s scale.
It grows through a wide range and can grow in a medium of fairly high
reaction if the acid used is malic, citric or tartaric.

Maximum of Amount to inhibit

Substance tested growth growth
Acetic acid +25 + 30
Oxalic acid +25 + 30
Citric acid +95 +100
Tartaric acid +95 +100
Malic acid +95 +100
Sodium hydrate —40 — 45

Toleration of NaCl. The percentage of salt which the organism can
tolerate is surprisingly high; some slight growth took place in tubes
containing 10-5 per cent. NaCl and those with 10 per cent. were decidedly
clouded. Involution forms are however produced in broth containing
high percentages.

Desiccation. A young agar culture was suspended in water and
smeared on cover slips. As soon as the smear was dry, these were
transferred to a desiccator; they were removed from time to time and
dropped into tubes of nutrient broth. The organism was still alive
after 80 days, more prolonged tests have not yet been carried out.

Sunlight. The test was carried out in very bright sunlight soon
after midday in the month of October. Thinly sown plates were

E. M. Dormer Ct

employed, one half being covered with black paper. To prevent the
effects of a rise in temperature, each plate was covered with a small
glass dish filled to a depth of 2 cm. with a 2 per cent. solution of potash
alum. The organism was killed by an exposure of 60 minutes and the
number considerably reduced in 45 minutes.

Chloroform. The bacillus will not grow in nutrient bouillon over
chloroform.

Germicides. The following table will serve to indicate the behaviour
of the organism toward the more common germicides:

Amount Amount to kill
Maximum for to inhibit organism in
Substance growth growth 30 minutes
Copper sulphate 1: LO000 1 : 5000 1: 1500
Carbolic acid 1: 1000 1 :500 1: 100
Mercurie chloride 1: LOOOO 1 : 5000 1: 4000
Formalin 1: 1000 1 3750 1:50

It will be observed from the above schedule that the organism is
very sensitive to copper sulphate, it is to be hoped therefore that good
results may be obtained by spraying the infected trees with Bordeaux
mixture.

PATHOGENICITY TO ANIMALS.

Through the courtesy of Sir Arnold Theiler, the pathogenicity of
the organism to animals was tested at the Veterinary Research Labora-
tories at Onderstepoort. I am indebted to Mr E. M. Robinson for the
following notes on experiments which he has carried out in this con-
nection.

To test the pathogenicity of the Bacillus of Lemon disease, two
rabbits were inoculated each with half an agar slope of a 24 hours’ growth
of the organism. One rabbit which will be called (A) was inoculated on
25th September, 1915, and the inoculation was given intraperitoneally.
This rabbit developed symptoms of extreme illness and on the two days
following the inoculation would not eat, refused to move and occasion-
ally made grinding movements with its teeth. On the third day after
the inoculation the rabbit appeared much less dull and at the end of
a week appeared quite well again. On the 7th October, 1915, the
rabbit received a further inoculation of half an agar slope (24 hours’
growth) subcutaneously, but showed no further symptoms of illness,
and is still healthy to date (14th November, 1915).

Another rabbit (B) was inoculated with half an agar slope of the
organism (24 hours’ growth) subcutaneously on the 25th September, 1915.

78 A Bacterial Spot of Citrus

It developed*no symptoms of illness and on 7th October, 1915, received
a further similar subcutaneous inoculation, the same dose being used.
On 11th October the rabbit showed symptoms of severe illness with
thick mucopurulent discharge from the eyes and extreme dullness. On
12th October, 1915, it was found dead, and the post mortem examina-
tion showed nothing but a slight hyperaemia of the lungs. Cultures
from the heart’s blood remained sterile. This rabbit’s death was
‘probably not due to the bacillus of Lemon disease as the inoculation
of the first rabbit (A) was a much more severe test of pathogenicity. °

Intravenous inoculations were not attempted as it has been found
that most non-pathogenic organisms will produce death from toxaemia
except in very small doses.

Rabbit (A) was bled on 12th November, 1915, and its serum tested
for agglutinins against the organism. Agglutination was obtained in
a dilution of 1—800 distinct, and partial in a dilution of 1—2000. The
serum of a normal rabbit gave no agglutination in as low a dilution as
1—10. Fresh orange juice would not agglutinate the bacillus in any
dilution.

The agglutination test should prove of use in the ultimate proof
of organisms causing plant diseases and is quite easy to apply.

Comparative table of characters of three organisms
causing diseases of Citrus.

Bacillus Bacterium Pseudomonas
citrimaculans citriputeale citri
Nature of disease produced Dark sunken spots Dark sunken spots Canker
Organs affected Fruit, stems, rarely Fruit Fruit, stems and
leaves leaves
Form of organism Rod with rounded Rod Rod with rounded
ends : ends
Dimensions 1—4 x -45—-7u 2—4 x -51—lp 1-5—2 x -5— 754
Flagella 5—10 peritrichous 1 polar 1 polar
Capsule Conspicuous None Not recorded
Colonies on nutrient agar Sub-circular, Greyish white to — Circular yellow
yellow pearl grey

Gelatine stahs

Milk

Test for indol

Liquefaction napi-
form

Becomes _ slightly
acid, casein pre-
cipitated with
eradual extru-
sion of whey

Positive

Liquefaction strati-
form

Becomes intensely
alkaline and
clear without
separation of
casein

Positive

Liquefaction _ fili-
form

Becomes alkaline,
casein is preci-
pitated, whey
clear

Negative

EK. M. DorGE 79

R&EsuME OF SALIENT CHARACTERS.

Bacillus citrimaculans, n.sp. Causes spot disease of citrus, attacking
fruit and branch, rarely found on leaves; attacks the parenchyma, first
occupying the intercellular spaces then breaking down the cell walls;
in nature attacks lemons, oranges and naartjes, and in addition has been
successfully inoculated into limes, shaddock, grape fruit, citron; seville
oranges are resistant.

A slender rod with rounded ends, 1—4 = -45—--7p, average 1:5 x -5
to -6u; motile by 5—10 peritrichous flagella; conspicuous capsule but
no spores; involution forms in broth containing high percentages NaCl;
stains readily by usual stains and exceptionally well with gentian violet
and by Gram’s method.

Forms showing sub-circular yellow colonies on nutrient agar with
dense, grumose centre; liquefies gelatine, clouds nutrient broth, forming
pellicle and sediment; does not liquefy blood serum; coagulates milk
with precipitation of casein and extrusion of whey, the coagulum is
not redissolved; no growth in Cohn’s solution; characteristic growth
in Uschinsky’s solution.

No gas in fermentation tubes, acid with dextrose, laevulose, galactose,
maltose, saccharose and mannite, but not with lactose, glycerine,
dextrin, or starch. Nitrates reduced with evolution of gas; tolerates
up to 10 per cent. NaCl; no growth in broth over chloroform, indol
produced in media containing peptone.

Aerobic, facultative anaerobe; killed by 60 minutes exposure to
sunlight; optimum reaction about 20 Fuller; T.D.P. 62° C., maximum
for growth 43° C,

Group number 221.2321523.

SUMMARY.

This paper gives an account of a citrus disease which is causing
considerable trouble in the Western Province of the Cape. It causes
dark sunken spots on fruit and shoot, and not only disfigures but pro-
vides an entrance for fungous parasites which completely destroy the
fruit.

The causal organism is described at some length, and as it appears
to be one hitherto undescribed has been named Bacillus citrimaculans.

The question of preventive measures is not discussed, but an

80 A Bacterial Spot of Citrus

improvement in the sanitation of the orchard would certainly prove
beneficial, and it is intended to carry out spraying experiments during
the ensuing season.

BoranicaL LABORATORIES OF THE UNION OF
SoutH AFRICA, PRETORIA.

LITERATURE CITED.

Besson, A. Practical Bacteriology, Microbiology and Serum Therapy. 1918.
Dowex, E. M. A Spot Disease of Citrus Fruits. South African Fruit Grower,
Sept. 1915, p. 43.

3. Dorper, E.M. A Bacterial Spot of Citrus Fruit. Agricultural Journal of South
Africa, Nov. 1915.

4. Hasse, CuraraA H. Pseudomonas Citri, the Cause of Citrus Canker. Journal
of Agricultural Research, vol. tv, No. 1. April, 1915.

5. Ripeway, R. Color Standards and Nomenclature. Washington, 1912.

6. Smrru, Clayton O. Black Pit of Lemon. Phytopathology, vol. m1, No. V1, p. 277.

7. Smira, Erwin F. Bacteria in relation to Plant Diseases, vol. 1. Washington,
1905.

8. Report of the College of Agriculture and the Agricultural Experiment Station of

the University of California, from July, 1913, to June 30, 1914, p. 68

wots

EXPLANATION OF PLATES.

All drawings were made with the aid of the camera lucida.
PLATE S

Ill. Lemons, natural infection from Simondium, C.P., (a) lemon with a single typical
spot, (6) the fruit on the right has been wounded by numerous scratches.

IV. Lemons, natural infection, from Simondium, (a) an early stage of infection, the
spots are still light coloured. (b) Numerous infections at and remote
from stalk coalesced to form irregular bodies.

V. Lemons, artificial infection, pricked with sterile needle, then three of them peice
with suspension of a pure culture of B. citrimaculans; the fruit to the right
in (a) is a control pricked with a sterile needle only.

VI. Lemons, artificial infection, ten days after inoculation, (a) inoculated by needle
pricks, (b) with hypodermic syringe.

VII. Naartjes (a) artificially infected by needle pricks, (b) natural infection from
Simondium.
VIII. (a) Navel oranges, natural infection from Simondium.
(b) “Parson Brown” seedling orange, artificial infection.

IX. (a) Lemon twigs showing infections round leaf bases, and petioles of fallen
leaves.

(6) Small branch from orange showing the same characteristic discolorations.
(c) Leaves with bacterial spot, these are apparently rare and associated with
tearing of tissues by thorns.

THE ANNALS OF APPLIED BIOLOGY. VOL. III, NOS. 2 and 3 PLATE III

(1)

(0)

THE ANNALS OF APPLIED BIOLOGY.

VORP

NOS. 2

and

3

PLATE

THE ANNALS OF APPLIED BIOLOGY. VOL. III, NOS. 2 and 38 PLATE V

(b)

THE ANNALS OF APPLIED BIOLOGY. VOL. III, NOS. 2 and 8 PLATE VI

THE ANNALS OF APPLIED BIOLOGY. VOL. Ill, NOS. 2 and 3 PLATE VII

(b)

THE ANNALS OF APPLIED BIOLOGY. VOL. Ill, NOS. 2 and 8 PLATE VIII

(D)

THE ANNALS OF APPLIED BIOLOGY. VOL. III, NOS. 2 and 3 PLATE IX

THE ANNALS OF APPLIED BIOLOGY. VOL. Ill, NOS. 2 and 3 PLATE X

PLATE XI

VOL.

THE ANNALS OF APPLIED BIOLOGY.

DEL Prensa

eon

(d)

PLATE XII

Ill, NOS. 2 and 38

VOL.

THE ANNALS OF APPLIED BIOLOGY.

10.15

10

9.45

9.30

9.15

J

10.30

11.30

11.15

ilal

10.45

12.30

12.1

11.45

Fy
we

ee?
Ley)

THE ANNALS OF APPLIED BIOLOGY. VOL. III, NOS. 2 and 3

PLATE XIII
BBE
o vy” \
Q U = z()
9 |
Q \ 74 Ok
S J Soe < } v BS
= 6
\ uy
Say gee 2S _S8,S0° S
LO G SS 253s
= CY yx . @ Sr 2D
ve 2 SAN Bs, eX Ci
G Ws See CS
(a) (b)

XII.

XIII.

E. M. Dorper SI

Section through naartje rind at an early stage of infection showing the
bacillus in sub-stomatal cavity and adjoining intercellular spaces. The size
of the bacilli is not indicated. Zeiss ;'; mm., comp. ocular No. 6*.

Section through lemon rind, bacillus has multiplied and distended inter-
cellular spaces: same magnification as (a)*.

Section through tissues of infected orange rind, showing cells completely dis-
integrated and occupied by masses of bacteria. Magnification as above*.
Stab cultures in nutrient gelatine after 11 days at 20° C., and 2 days at room
temperature (about 25°C.). (b), (c) and (d) colonies on nutrient agar at
25°C. (b) and (c) 3 days old, (d) 5 days old. Shows the effect of crowding
on the size of the colonies—somewhat reduced, diameter of petri dish, 34
inches.

Development of organism on agar hanging block: for full explanation see
text (Zeiss ='; oil immersion, comp. oc. No. 12. This magnification used for
all drawings in Plates XII and XIII unless otherwise stated)*.

24 hour colony on hanging block (comp. oc. No. 6).

Rods showing flagella stained by Ellis’ modification of Loeffler’s method.
18 hours at 25° C.

Chains from pellicle on laevulose broth, 16 days at 25° C.

Aberrant forms on potato cylinders, 12 days at 35° C.

Capsuled rods from edge of impression preparation of colonies on nutrient
agar, 18 hours at 25° C., stained with fuchsin.

Rods in slime oozing from fruit, stained by MacConkey’s method.
Involution forms from broth containing 7:5 per cent. NaCl, 48 hours at
25° C., stained with aniline gentian violet.

* Exigencies of space have necessitated the reduction of the following figures :

Pl. X (a) by 4: (0) by 4: (c) by #¢.
Pl. XII (c) by 4. ©
Pl. XIII (a) by 4: (ce) by d. [Ep.]

82

REPORT ON A TRIAL OF TARRED FELT “DISCS”
FOR PROTECTING CABBAGES AND CAULI-
FLOWERS FROM ATTACKS OF THE CABBAGE-
ROOT FEY.

By J. T. WADSWORTH,

Research Assistant in the Department of Agricultural Entomology,
Manchester University.

(With Plate XIV.)

CONTENTS

PAGE
(1) Introduction and Historical Remarks. ; F ; 82
(2) Description of the experiments . : : : : 83
(a) With Cabbages : : 2 é ‘ : 85
(6) With Cauliflowers . : ; : : : 86

(3) Precautions to be observed in applying the discs, and
other remarks . : 3 - é : 89
(4) Literature : ‘ : ; : ; : ; : 91
(5) Description of Plate > : : ; : 5 . 92

INTRODUCTION.

THE present investigation has been carried out at the suggestion
of Dr A. D. Imms. He has displayed a keen interest in the progress
of the experiments and I have had the benefit of discussing with him
any points that arose from time to time.

The expenses entailed during these experiments have been met
from the annual grant of the Dept. of Agricultural Entomology,
Manchester University.

It is a fact only too well known to most growers of cabbages,
cauliflowers and related vegetables that large proportions of their
crops are frequently lost owing to attacks of the Cabbage-root Maggot,
the larva of Chortophila brassicae. A very large number of remedies
have been devised with the object of either destroying the maggots

J. T. WapswortH 83

on infected roots, or preventing the deposition of eggs near the host
plants. With one or two exceptions, however, these remedies are of
little practical value since they require, as a rule, frequent application
to the plants in order to obtain any satisfactory measure of success.
For this reason, and also on the score of expense, most growers are
averse to their use. Gibson and Treherne(6), in their recently published
bulletin on the Cabbage-root Maggot, give a list of forty-eight
insecticidal mixtures or protective measures which they have experi-
mented with for several years. As a result of these experiments they
make the following statement: “It can be truly claimed that the only
protection to be relied upon for cabbages and cauliflowers and one
which is in every way practical, is in the use of tarred felt paper discs.”’

The idea of placing paper collars round the stems of cabbage plants,
and thus preventing egg deposition in the soil near plants so protected,
originated with Prof. W. W. Tracy, of Detroit, Mich. in 1887. The
material (manilla paper) which he used, however, was unsuitable for the
purpose and the results of his experiments were unsatisfactory. In
1889 Prof. E. 8. Goff of Wisconsin(7) substituted tarred paper for the
material used by Tracy, with complete success. Goff also devised an
efficient tool for cutting the discs expeditiously; the latter were about
three inches in diameter, hexagonal in shape, with a slit extending from
one angle to the centre and with a star shaped cut also in the centre.

Since Goff proved his method of protecting cabbages and cauliflowers
by means of tarred discs to be thoroughly practical it has been widely
adopted in America by commercial growers and others. The method
has also been tested at almost all the Agricultural Experiment Stations
in the United States and Canada in those districts where serious infesta-
tions of Cabbage-Maggots occur: with one or two exceptions the resulting
reports have been entirely favourable.

The following authors have published reports of successful trials of
tarred felt discs: Slingerland(1), Smithd2), Britton and Walden(),
Caesar (4), Schoene(s & 9), Britton and Lowry 2), Gibson and Treherne (6).

On the other hand Schéyen(i0), Blair”), and Washburn (15) obtained
unfavourable results in the trials which they made of tarred discs.

In these cases the lack of success is attributed either to placing the
discs on the plants too late in the season, or carelessly, or to using tarred
paper—which curls—instead of tarred felt.

The greater weight of opinion and evidence is distinctly in favour of
the tarred disc method of protection and the majority of the above
quoted observers agree that it is practical, economical and effective,

84 Report on a Trial of Tarred Felt “ Dises”

Apparently, however, there are no published records of any trials of

this method having ever been made in this country. References in °

British horticultural literature, and in reports on insect pests, to the
use of tarred felt discs are very meagre and not very encouraging.

As numerous complaints of damage and losses sustained by growers
owing to the depredations of the Cabbage-Maggot are continually being
received, together with requests for information how to deal with the
pest in question, the desirability of testing here the method which has
proved so successful in America thus becomes apparent.

During successive seasons In my own garden I have lost considerable
numbers of cabbages and cauliflowers owing to attacks of the Cabbage-
Maggot and I had already decided to test the American discs when
Dr A. D. Lmms suggested that I should test them on a fairly large scale
in a local market-garden. Arrangements were therefore made with
a market-gardener to rent from him a piece of land and the work was
commenced in the spring of 1916.

DESCRIPTION OF THE EXPERIMENTS.

The tarred felt discs were obtained from the United States where
they are regular articles of commerce. They were not the hexagonal
form usually figured and described but they were square, being 25 inches
each way with only two slits—a long slit extending from the middle of
one edge to a point half-an-inch beyond the centre of the disc, and
a short slit three-quarters of an inch long crossing the long slit at might
angles in the centre of the disc (Fig. 3).

Two separate pieces of land situated at Northenden, Cheshire, were
rented from Mr Chas. Heywood, a market-gardener, who undertook to
prepare the land and perform the necessary operations of cultivation,
planting, etc. The tests were made on both cabbages and cauliflowers
and the two plots of land were about sixty yards apart. The land on
which the cabbages were planted was under cauliflowers in 1915 and
the crop then suffered severely from maggot attack, from two to three-
fifths of the plants being lost owing to this cause. This piece of land
was selected in order to ensure, so far as possible, a heavy infestation
of maggots.

A heavy dressing of well-rotted farm-yard manure was spread over
the land towards the end of April; this was then ploughed under, the
land afterwards harrowed down and then rolled in order to consolidate
the soil—which is very light in texture—and to render conditions
favourable for placing the discs as flat as possible.

ee amie endian

le a ie
=

J. T. WADSWORTH 85

(a) Wath Cabbages.

The cabbages (variety Leeds Market) were planted out on May Ist
and the discs were placed in position on the following day. Altogether
816 cabbages were utilised for this experiment; they were planted in
eight rows, each row containing 102 plants: disced and undisced rows
alternating with each other. The spring was a particularly good
growing season and the plants made excellent growth. On June 18th,
some of the unprotected plants first showed signs of maggot attack and
on June 23rd a count of the plants was made. None of the protected
plants showed signs of attack but thirty-one of the unprotected plants
—equal to 7-6 per cent.—exhibited the usual signs of severe infestation.
The attack was very evenly spread over each of the four undisced rows:
the numbers of attacked plants in each row being nine, seven, seven,
and eight respectively.

On July 8th the cabbages were again counted when the percentage
of attack had increased to 13-2; fifty-four out of the total number of 408
unprotected plants being severely attacked, whilst only one of the
protected plants had succumbed to maggot attack, i.e. less than 0-25
per cent.

Several of the infected plants were taken up for examination and in
all cases the original fibrous roots were absent, having been destroyed
by maggots; in some cases the latter were still feeding on the roots
and in those cases where they were absent pupae were present in the
surrounding soil. In many plants new fibrous roots were developing
near the upper portion of the main root and it was evident that the
attack caused by the first generation of flies was ended.

The opinion of the market-gardener was that possibly one-third of
the infected plants would develop a fresh set of roots and recover from
the attack, although the resulting cabbages would be only small.

On July 13th the plants were counted again. No further attack was
evident; the cabbages were counted at intervals of seven days until
mid-August but no further losses were incurred.

As only one plant was lost owing to maggot attack out of 408 plants
protected by the discs it is very strikingly evident that the latter afforded
a very complete protection to the plants. The protected cabbages were
also larger on the average and slightly earlier in coming to maturity than
those in the unprotected rows. A probable explanation of these differ-
ences will be mentioned when discussing the results of the cauliflower
trials.

Ann. Biol. 1 6

86 Report on a Trial of Tarred Felt “ Dises”

Some alterations in the discs themselves were made and the method
of placing them on the plants was also varied, as follows. On one row
all the discs used were perforated with a small hole in the centre where
the two slits cross, in order to obtain a closer fit round the stems of
the plants. On another row two discs were applied to each plant, the
discs being crossed so that when the slits are forced open by the stem
thickening during growth, the soil is not exposed, as it is to some extent
when single discs alone are used. Where two discs were used it was
surmised that female flies would be absolutely prevented from gaining
access to the soil near the plants so protected. As will be seen from the
results obtained the single discs gave quite adequate protection and no
corresponding advantage was gained by using either perforated discs
or double ones to compensate for the additional expense and trouble
entailed.

(b) With Cauliflowers.

The land on which the cauliflowers were planted was similar in
character to that on which the cabbages were grown; it was prepared
in the same manner and was cropped with sage the previous season.
The seedlings (variety Autumn Giant) were planted out on May 27th
and the discs were placed round the plants immediately after planting.
The latter were planted in four rows, each row containing 233 plants,
Le. 932 plants altogether. The protected rows alternated with the
unprotected ones: rows I and III being protected, rows II and IV were
unprotected.

The first count was made on June 23rd with the following result.
On row I adjoining the field (see Fig. 1) no plants showed signs of maggot
infestation; row II, sixty-two plants attacked; row III one plant
infected; row IV, ten plants had succumbed.

Altogether seventy-two of the 466 unprotected plants were attacked,
equal to 15-4 per cent., whilst only one of the protected plants was
attacked up to that date.

On July 4th the plants were again counted when considerably further
losses had occurred; a photograph (Fig. 1) was taken on this date.

The results of the countings are given in tabular form; a table
showing the results obtained with the cabbages is also given for com-
parison.

No further cabbage plants were destroyed by maggots on this plot
alter July 8th. 768 cabbages, out of a total of 816 planted, had been

J. T. WADSWORTH 87

cut and marketed up to the end of September. This number agrees
closely with the number of plants (761) unaffected with maggots on
July 8th. <A few plants had evidently recovered and made marketable
plants by the end of September, as predicted by the market-gardener.

Table showing the numbers of Cauliflower plants destroyed by Cabbage-
root Maggots when protected by tarred felt discs, and when unprotected.
Each of the four rows contained 233 plants.

Results in percentages

——_— - OO

protected unprotected

Dates when Row I Row II Row III Row LV rows rows
counted protected unprotected protected unprotected (Land III) (II and IV)
June 23 0 62 l 10 — 15-4
July 4 0 132 7 56 1:5 40-3
July 15 5 141 10 85 3-2 48-5
Aug. 5 11 174 13 120 5-1 63-0

Results with Cabbages. Each row contained 102 plants.

Results in percentages

Rows Rows Rows Rows protected unprotected

Dates when ILand IV TandIIIl Vland VIII Vand VII rows II, rows I, III,
counted protected unprotected protected unprotected IV, VI, VIII V, VII

June 23 0 16 0 15 0 7-6
July 8 1 33 0 21 0-2 13-2

The cauliflower plants were also counted on July 8th but no further
losses had occurred since the previous count (July 4th). The effect of
the infestation produced by the first brood of flies had reached its
maximum about this date.

From July 15th onward the losses increased very considerably as
will be seen by reference to the table. This increase was attributed to
a secondary attack of maggots produced by the second generation of
flies. At this period also a spell of three weeks of very hot, rainless
weather prevailed; these weather conditions rendered the cauliflowers
more susceptible to maggot attack and less able to recover therefrom.

On August 12th the plants were again counted but no further losses
were observed; on August 16th some of the cauliflowers were ready for
| cutting and counting was discontinued.

The results obtained prove very conclusively that the tarred felt
discs are a very effective means of protection from cabbage maggot
attacks. Only twenty-four cauliflower plants were lost out of a total
| number of 466 protected by discs, whereas 294 plants were lost out of

6—2

88 Report on a Trial of Tarred Felt “Dises”

the same number of unprotected ones. A net gain of 57-9 per cent.
was thus effected by using the protective discs.

The results obtained with the cabbages were even more striking as
only one plant was lost out of 408 protected by the discs. The infesta-
tion, however, was not nearly so severe in the cabbages as in the cauli-
flowers. Only 13-2 per cent. of the unprotected cabbages were lost as
against 63-0 per cent. of the cauliflowers.

Market-gardeners, and gardeners generally, hold the opinion that
cauliflowers are much more lable to be destroyed by maggot attack than
cabbages, and the results obtained in these trials give strong support to
this view. Whether it is a fact that cabbage-root flies are more strongly
attracted to cauliflowers than they are to cabbages, or whether the
former plants succumb more readily to maggot attack than do cabbages,
remains a subject for further enquiry. The above results clearly show
that, on similar land, cauliflowers suffered much more severely from
root-maggot infestation than cabbages. The only factor that differed
in the two cases, was that the cabbages were planted out a month
earlier than the cauliflowers, and consequently were well established,
and growing vigorously, before the maggot attack became severe.

Gibson and Treherne, who discuss very fully this question of the
relative susceptibility of cauliflowers and cabbages to maggot attack,
make the following statement:

“We consider that the supposed susceptibility of cauliflowers over
cabbages is probably due to the lesser vitality of the former plants.”

During a period of seventeen days between June 25th and July 11th,
1915, at Agassiz, B.C., 1418 eggs were deposited on six cabbage plants,
while only 1038 were laid on six cauliflower plants; these results lead
them to believe that no special choice exists on the part of the fly to
deposit on cauliflowers over cabbages. On the other hand, in the table
given on p. 26 of their bulletin, they state that 6602 eggs were laid on
twelve cabbage plants during 133 days, an average of 550-1 per plant,
equal to four eggs per plant per day. On six cauliflowers 5515 eggs
were laid during 117 days, an average of 919-1 eggs per plant, equal to
seven eggs per plant per day; i.e. nearly twice as many eggs were laid
on cauliflowers as on cabbages. Although the periods of time were
not coincident—the countings of the eggs on the cauliflowers being made
later in the year than in the case of the cabbages—the results would
appear to indicate that the flies were more strongly attracted to the
cauliflowers than to the cabbages,

J. T. WApDSWoRTH 89

PRECAUTIONS TO BE OBSERVED IN APPLYING THE DISCS, AND
OTHER REMARKS,

In order to obtain the best results from the use of tarred felt discs
it is advisable that the soil should be in a fine friable condition when
the discs are placed in position, and further, that the plants should not
be placed in depressions of the soil but if possible on a slight ridge.
If the soil is lumpy the discs will not he flat on the ground and the
female flies will then be able to crawl underneath the discs to deposit
egos. If the discs are placed below the soil level they are liable to
become covered-with soil after rains, which renders them less efficient.
It is also of the utmost importance to place the discs round the plants
immediately after planting out, if this operation is performed later
than the first week of May. In the district around Manchester, eggs
of the cabbage-fly were found on May 14th in 1915, and on May 18th
in 1916; these were the earliest dates they were found in those seasons.
Probably in the South of England they would be found much earlier,
and in the North somewhat later. On seedlings planted out earlier in
the year, it is not so important to place the discs in position immediately
after planting, although there may be advantages to be gained by
doing so.

Several of the observers, who have tested the tarred felt discs, draw
attention to the fact that protected plants are usually larger, and mature
earlier, than unprotected ones. These differences are nightly attributed
to freedom from root injury owing to the absence of maggots on the
roots. Another possible factor, which contributes to this result of
increased size and early maturity, may be referred to here. If the soil
underneath the discs is examined on hot days, during the first two or
three weeks which follow the planting out of the seedlings, it will be
found in a moist condition, whereas the soil round the unprotected
plants will be dry. The disc, by preventing evaporation from the surface
of the soil, conserves the moisture in the soil directly beneath it. The
plants are thus enabled to recover more quickly from the effects of
transplanting, and the conserved moisture also helps them to withstand
better the effects of prolonged drought. More especially is this the
case with regard to cauliflowers, which suffer more from drought than
cabbages.

The question is frequently asked whether the discs, owing to the
smell of the creasote with which they are impregnated, deter flies from

90 Report on a Trial of Tarred Felt “ Discs”

laying eggs. Whilst the creasote may exercise a deterrent influence
on the flies for a short period of time after the discs have been placed
round the plants, this influence—supposing it to exist—soon disappears.
Frequently eggs may be found both on the surface of the discs, and also
underneath them: a fact which tends to disprove the idea that flies are
deterred from laying eggs owing to the emanation of odours from the
discs themselves. The chief function of the disc is to act as a mechanical
obstacle to the efforts of the flies to deposit eggs in the soil near the
plant.

Growers have also asked for information regarding the action of the
discs on slugs. No definite experiments have been made with the
object of determining the effects of the discs with regard to slugs. It 1s
worthy of note however that, in the row of 233 cauliflowers adjoining
the field of clover (see Fig. 1), only two plants were definitely proved
to have been injured by slugs. As numerous slugs were observed in wet
weather along the edge of this field, it would appear probable that they
do avoid the discs. Occasionally a slug was found sheltering beneath
a disc, but this was where the ground was lumpy and spaces were present
underneath the disc. If the discs are pressed closely to the soil, no
accommodation would remain for slugs, and they would thus be pre-
vented from crawling underneath.

Several market-gardens were visited during the summer months in
order to determine the relative proportion of plants destroyed by maggot
attacks on different types of soil. An interesting fact was observed in
a market-garden situated about two miles from the one where the above
described experiments were conducted. Although I examined hundreds
of cauliflowers there, it was difficult to find a single plant infected with
root-maggots. The soil is a heavy clay loam—the owner described it
as a good strong soil—and the gardener attributed to this fact his
immunity from root-maggot attacks. He also uses gas lime abundantly,
having a strong belief in the beneficial effects of this substance against
root-maggots and pests generally. The contrast between the two
cauliflower plots, at the end of July and early August, was very striking.
On the one plot, where the soil was light in character, from fifty to sixty
per cent. of the cauliflowers were destroyed by maggots: on the heavy
soil, hardly a single plant, in three or four hundred, was lost from this
cause. All the growers, with whom I discussed the matter, agreed that
cauliflowers and cabbages are more liable to become infested with root-
maggots when grown on light soils, than they are when grown on heavy
soils. Various reasons were given in support of this opinion, and

ee ee

J. T. Wapsworttt 9]

without attempting here to discuss this aspect of the question, | may
say that all the evidence obtained during the summer in this con-
nection supports their view.

Trials of the discs have also been made with winter-greens and
flowering broccoli. Very little advantage has been apparent with
regard to these plants up to the present time (end of September).
Probable reasons for this fact being that the intensity of attack of the
second brood of flies was declining about the time when planting out
took place, and also because at this period (end of July and early
August) the greatest numbers of brassica plants are growing in the
fields; consequently the numbers of eggs laid near any single plant are
not likely to be so numerous as when fewer plants are in the fields and
gardens, as is the case earlier in the season.

As stated above, tarred felt discs are regular articles of commerce in
America, where they are largely used by growers. The discs used in these
experiments were obtained from A. B. Cowles, 25,8. Water St., Rochester,
N.Y.; they cost two dollars per thousand (120 for a shilling) in addition
to carriage. In this country at the present time tarred felt discs are
not obtainable commercially. Various kinds of materials, however,
have been tested this summer and arrangements are in progress whereby,
it is hoped, suitable discs will be purchasable at a reasonable price in
England. Growers who suffer losses from root-maggot attacks will thus
be enabled to test for themselves the value of this method of protection.

LITERATURE REFERRED TO AND CONSULTED.

1) Buatr, W. 8. Experimental Farms Report, Canada, p. 362. 1904.

) Brirron, W. E. and Lowry, Q. 8. Experiments in controlling the Cabbage-
maggotin 1915. Report Agric. Exp. Stat., New Haven, Conn., pp. 114-118.
1916.

(5) Brrrron, W. E. and WatpeEn, B. H. Eighth Report, Agric. Exp. Stat. Conn.,

pp. 832-837. 1908.

(4) Caxsar, L. 37th Report of Ontario Agricultural College, Canada, p. 40. 1911.

(5) Carpenter, G. H. Injurious Insects observed in Ireland during the year
1901. Economic Proc., Royal Dublin Soc., Vol. 1, 1902, pp. 141-144.

(6) Grsson, A. and TREHERNE, R. C. The Cabbage Root Maggot and its Control
in Canada, etc. Bull. No. 12, Dominion of Canada, Dept. of Agric., Entom.
Branch. 1916.

(7) Gorr, E. 8S. A new preventive against the Cabbage Maggot. 8th Ann.
Report Agric. Exp. Stat. Univ. of Wisconsin, pp. 169-173. 1891.

(8) ScHornr, W. J. The Cabbage Maggot in Relation to the Growing of early

Cabbage. Bull. No, 382, New York Agric. Exp. Station, Geneva, N.Y.

1914,

(9) ScHornr, W.J. The Cabbage Maggot: Its Biology and Control. Bull. No. 419,

New York Agric. Exp. Station, Geneva, N.Y. 1916.

(10) ScuéyvEn, W. M. Beretning Skadeinsekter og Plantesygdomme, p. 23. 1896.
(Quoted from Schoene.)

(11) Strncertanp, M. V. The Cabbage Root Maggot with notes on the Onion
Maggot and Allied Insects. Bull. No. 78, Cornell Univ. Agric. Exp. Station,
Entom. Division. 1894.

(12) Smirn, J. B. Report Entom. Dept. New Jersey Exp. Stat., p. 439. 1907.

(13) THroBALD, F. V. Report on Economic Zoology, Wye, pp. 53-54. 1905¢

(14) ——— Report on Economic Zoology. South Eastern Agric. Coll., Wye, p. 93.
1913.

(15) Wasnpurn, F. L. Report on Injurious Insects of 1907 and 1908. Annual
Report Minnesota Agric. Exp. Stat. 1908-9.

92 Report on a Trial of Tarred Felt “ Dises”
:
;
i

DESCRIPTION OF PLATE XIV

Fig. 1. Photograph showing appearance of the cauliflower plot on July 4th. Rows I
and III (counting from the field on the left side of the photograph) were protected
with tarred discs, rows IT and IV were unprotected. The remaining rows of plants

were not utilised in these experiments.
Fig. 2. Photograph of a brussels-sprout plant showing a disc in position.

Fig. 3. Outline figure of a dise; actual size.

THE ANNALS OF APPLIED BIOLOGY. VOL. III, NOS. 2 and 38 PLATE XIV

Fig. 2 Fig. 3

A ll et Nit

93

ON THE RESISTANCE TO FUNGICIDES SHOWN
BY THE HOP-MILDEW (SPHAEROTHECA HU-
MULI (D.C.) BURR.) IN DIFFERENT STAGES
OF DEVELOPMENT.

By E. 8. SALMON,
Mycologist to the South-Eastern Agricultural College, Wye, Kent.

(With Plate XV.)

In the testing of the resistance of any fungus to a chemical substance
both the environmental conditions and the stage of growth of the
fungus are, obviously, important factors.

We know, e.g. from Clark’s! experiments, that the spores of a fungus
when supplied with nourishment in the form of a decoction of sugar-
beet show a greatly increased power of resistance to the toxic effects
of copper sulphate. With Qdocephalum albidum the lethal strength
for spores in water is -0013°%, whereas with spores supplied with
nourishment it is 0-1 °%,—about 80 times as strong.

In a recently published paper? dealing with the fungicidal action
of certain chemicals, principally sulphides, attention was drawn to the
fact that data as to the behaviour of spores placed in a fungicide have
little practical value as indicating the strength at which the same
chemical will be fungicidal when used against the growing fungus on
the plant. It was pointed out that while Foreman? found that a
0-16 % solution of caustic soda prevents the germination of spores of
Botrytis cinerea and Sphaerotheca mors-uvae, experiments have shown
that a 0-3 °% solution does not kill well-developed patches of the conidial
stage of S. Humuli.

In all our spraying experiments with “powdery mildews” (Ery-
siphaceae) carried out during 1914 and 1915 we were careful to select
for spraying well-developed patches of mildew showing dense clusters
of conidiophores bearing chains of conidia—a stage of development

1 Bot. Gazette, Vol. xxx (1902).

2 J. Vargas Eyre and E. 8. Salmon, “The Fungicidal Properties of Certain Spray-fluids,
Journ. Agric. Science, Vol. vit, 473 (1916).

3 Foreman, F. W., ‘“‘The Fungicidal Properties of Liver of Su.phur,”? Journ. Agric
Science, Vol. ut, 401 (1910).

94 The Resistance to Fungicides

which may be briefly described as a “powdery” patch. This stage was
chosen both in order to make the experiments as strictly comparable
as possible and on the assumption that in such a stage of development—
where the fungus has a well-developed mycelium furnished with very
numerous haustoria—the resistance to the fungicidal properties of the
chemical would be at its highest.

During experiments this season (1916), however, facts have been
observed which show that the mildew when in an earlier stage of develop-
ment on the host-plant is more difficult to kill. Our suspicions of this
fact were first aroused by the occurrence of minute tufts of conidiophores
at fresh places on the leaf two or three days after the spraying had
taken place. These fresh little patches of mildew could not have been
due to re-infections subsequent to the spraying, since a conidium takes
about five days from infection to develop mycelium and conidiophores.
The- pot was then studied more closely by means of hop-leaves
inoculated with conidia of the hop-mildew. It was found that very
soon after being inoculated the young hop-leaf, when rapidly growing,
shows small raised “blisters” or “humps” at the places where it has
been infected. At first the raised blister is green all over, and shows
no trace to the naked eye of a white mildew, although usually, under
a lens, a few delicate hyphae mav be seen radiating from the centre of
the “hump.” After a day or two, each “hump” shows a whitish
growth beginning to spread over its surface, due to the formation of
branched mycelial hyphae, and soon afterwards the first conidiophores
appear. An exactly similar appearance may be found on a young
Rose-leaf when attacked by the Rose-mildew (S. pannosa).

Spraying experiments on hop-leaves bearing numerous “humps”
showed that the mildew in this very early stage of development has
a greater power of resistance to sulphide solutions. Thus in experiments
it was found that a certain stock solution of ammonium sulphide when
diluted 1 to 100 with water was completely fungicidal to the hop-mildew
when in its “powdery” conidial stage; this solution did not, however,
prevent the youngest stage of the mildew, described above, from
developing and forming conidiophores. At double the concentration
however, the sulphide solution in question became fungicidal for this
earliest stage of development.

The details of two experiments may be given here. In Hxper. 1 the fungicide
used was a certain concentrated solution of ammonium sulphide diluted 1 to 100 with
water (and containing 1 per cent. soft soap). Three seedling hop-plants were used,
and the leaves on each which were sprayed bore the hop-mildew in the stages of
development noted below:

EK. S. SALMON 95

¢

plant a: leaf at 5th node (from apex) bearing numerous “‘ powdery” patches.

» 4th ,, bearing “humps,” showing afew very young sterile hyphae.

plant b: » Oth ,, bearing numerous “powdery” patches.
» 4th ,, bearing “humps,” showing a few young sterile hyphae.
plant ¢: , Oth ,, bearing numerous “powdery” patches.

» 4th ,, bearing green “humps” which showed under a lens no
signs of any hyphae.

All the plants bore at the respective nodes two leaves, each with the mildew in the
same stage of development, and one leaf at each node was left unsprayed as a control.
The spraying (using an atomiser) was carried out on June 6. By June 10 it was
apparent that on all the sprayed leaves where the mildew had been in the “‘ powdery”
conidial stage the ammonium sulphide solution had almost or quite killed the fungus ;
in every case, however, the leaves which bore the humps showed the mildew in a
further stage of development. On plant a there were numerous tiny tufts of conidio-
phores appearing from the ““humps”; on plant 6 there were two such tufts of conidio-
phores, and on plant ¢ nine such tufts. By June 17 the original *‘ powdery” patches
on the three sprayed leaves were all dead and in many cases semi-obliterated; the
control leaves all bore very numerous almost continuous densely powdery patches.
The sprayed leaves with the “humps” were as follows: a, ten very small though
densely powdery tufts had developed over the “humps”; 6, seven small densely
powdery tufts of conidiophores had developed where the youngest “humps” had
been,—where the ““humps” had shown, at the time of spraying, mycelial hyphae,
the fungus had been killed; c, twelve tiny powdery tufts of conidiophores had
*” In every case the control leaf showed much more
numerous and much larger densely powdery patches of conidiophores. It was clear

developed over the “humps.

from this experiment that the ammonium sulphide solution used was fungicidal for
conidial stage, but was not fungicidal for the mildew
in its earliest stages of development, although it appreciably checked its growth.

In Exper. 2 the fungicide used was the same concentrated solution of ammonium
sulphide diluted 1 to 50 with water (and containing 1 per cent. soft soap). The
seedling hop-plant used bore ‘at the 5th node (from the apex) two leaves, each
bearing very numerous “‘powdery” patches of mildew, and at the 4th node two
leaves, each bearing from thirty to forty “humps,” only a few of which showed
young mycelial hyphae on their surface. One leaf at each node was sprayed on
June 6; the remaining leaves served as controls. On June 10 the “‘powdery”
patches on the sprayed leaf were all dead and semi-obliterated; the “humps” on
the sprayed leaf showed no signs of any living mildew. On June 12 the control leaf
at the 5th node bore numerous densely powdery patches, while the sprayed leaf
showed no signs of any living mildew; at the 4th node the sprayed leaf bore about
thirty “humps” with no trace of mildew developing from them, while the control
leaf bore over thirty “humps” each of which now bore a small densely powdery
patch of conidiophores. On’June 20 the sprayed leaf at the 4th node was of a healthy
deep green colour and still showed the original “humps” but no trace of mildew
occurred anywhere on the leaf; a photograph of this leaf as it appeared at this date is
reproduced in Plate XV, fig. 1. The control leaf at the 4th node now bore densely
“powdery” patches covering over the original ““humps”; a photograph of this leaf
is shown in Plate XV, fig. 24.

1 T am indebted to Mr H. Wormald for taking these photographs.

>’

the mildew in the “‘ powdery’

96 The Resistance to Fungicides

Further, some evidence has been collected showing that the age of
the mildew even when in the “powdery” conidial stage is a factor of
importance. The older conidial patches have less power of resistance
to the soluble sulphide spray than the young patches.

The explanation of the difference in resistance to fungicides shown
by the earliest and the later stages of development cannot be given until
further data are available. It would appear that to some extent the
age and condition of the host-leaf are concerned; the mildew has less
resistance in its conidial stage when on an old hop-leaf than when on
a vigorous young leaf. But that this is not the whole explanation is
shown by the fact, mentioned above, that on one and the same leaf
a sulphide solution will kill the mildew where present in its well-developed
conidial stage and fail to kill it in its youngest stages of development.
It is possible that the earliest stage of development is more resistant
because the first-formed haustorium from the appressorium of the germ-
tube has more vigorous powers of resistance than the later-formed
haustoria, or because the appressorium offers powers of resistance; on
the other hand it may be that the vigorously growing conidial stage is
less resistant through the formation of thinner-walled hyphae, or that
the great number of haustoria present allows the fungicide to reach more
easily the epidermal cells of the host-plant and cut off the food-supply.

On the old hypothesis that the sulphide solution acts fungicidally by
virtue of the sulphur deposited, it could be held that the “powdery”
conidial patches are killed because a sufficient amount of the sulphide
solution is collected round them to give the requisite deposition of
sulphur, while this would not take place in the case of the earliest stage
of development when there are few mycelial byphae and no conidio-
phores. We are unable to accept this explanation, however, since our
work of last year has led us to the conclusion that with regard to this
class of fungicides the sulphides contained in solution act as such and
not by virtue of the sulphur deposited. ,

Whether it will be more economic to attempt to kill a “powdery
mildew” by spraying at a strength lethal to its earliest stage of develop-
ment, or to wait until the mildew has developed into the less resistant
conidial stage, must be decided by future experiments in the laboratory
and field. It is clear, however, that in the treatment of “powdery
mildews” generally this difference of susceptibility at different stages
of development must be kept in mind as a new factor of importance.

THE ANNALS

OF APPLIED BIOLOGY.

VOL. Ill, NOS. 2 and 3

PEATE

XV

OBSERVATIONS ON THE LARVAL AND PUPAL
STAGES OF AGRIOTES OBSCURUS, LINNAEUS.

By GEORGE H. FORD, M.Sc. (Vict.),

Technical Expert, International Institute of Agriculture, Rome.

(From the Department of Agricultural Entomology,
Manchester University.)

(With Plates XVI and XVII and 1 Text-figure.)

CONTENTS.
PAGE
1. Historical . : : : F : - : ; 97
2. Material and Methods , : : : : 98
3. Notes on the Biology of the ae Cee : - : 100
4. Description of the Larva. i . : : : 103
5. Description of the Pupa . : - : ; 106
6. Comparison with Larvae of close i allied Species . : 108
7. Natural Enemies : c : : : - : 110
8. Summary of General Conclusions : : : - 113
Bibliography : ; : : - : é . 114
Description of Plates . : - : : : elle

1. HISTORICAL.

ALTHOUGH the larval stage of Agriotes obscurus has undoubtedly
been of economic importance for a long period, yet there does not appear
to have been a good description of it until Westwood published a brief
account in 1839. Marsham in 1808(12) had figured a larva which he
considered to be a stage of Hlater segetis (= Agriotes lineatus), but the
figure is too indefinite for accurate determination. Furthermore, he
says he was unable to breed out the perfect insect, and was relying on
Bierkander for his information.

Westwood (18), however, describes the larva of Agriotes obscurus and
figures the antenna, the under side of the head and prothoracic segments,

98 The Larval and Pupal Stages of Agriotes obscurus

and the first maxillae and labium; his figures are too small to be of
much utility. He further notes that the larva had been figured pre-
viously by De Geer and Bierkander; but points out inaccuracies in
the figures of these authors, that seem to indicate faulty identification.

Chapuis and Candéze in their work published in 1855(3) do not
describe the larva, but merely quote references to the writings of Marsham
and Westwood. In 1860, Curtis published in his work on Farm Insects,
an account of Klateridae of importance to agriculture; he did not
describe the larva of Agriotes obscurus, and considered the larva of
A. lineatus to be the common and important species.

Schiodte (16) defined larval characters of the genus Agriotes in 1869,
and described those of A. lineatus. He figured the left mandible, as
well as the eighth and ninth abdominal segments. No mention is
made of A. obscurus. Perris in 1878(14) described many larvae of
Elateridae; but he merely quotes references to Marsham(12) and
Westwood(1s), and does not add any description of A. obscurus.
Beling(2), in a full account of the metamorphoses of Elateridae,
published in 1883—4, describes the larva of Agriotes lineatus very
completely, and states that he could only detect shght differences
between the larvae of A. lineatus and A. obscurus. Finally, the subject
is brought up to date by Xambeu(20) in 1912 and 1913 by his work
on the life histories of Elateridae. He adds little to our knowledge
of Agriotes obscurus, merely referring to Beling (op. cit.) and adding
a few descriptive remarks. It has been somewhat difficult to trace
literature definitely referring to Agriotes obscurus, owing to the long
established custom of terming Elaterid larvae under the collective
title of “wireworms,” without reference to any definite species. This,
together with a number of cases of inaccurate identification im the older
literature, has often rendered the information of doubtful specific value.

2. MATERIAL AND MetuHops.

The idea of studying the life history of certain Hlateridae, whose
larvae are known as “ wireworms,”’ was first suggested to me by Dr Imms,
to whom my thanks are also due for advice given from time to time.
The first problem requiring solution, was the determination of the exact
species of which the common wireworm forms the larval stage. With
this intention, the material was purposely obtained from as wide a
range of localities, and under as different conditions (of soil, crop, etc.),
as possible. The area covered was mainly in Cheshire and around

G. H. Forp QQ

Manchester, while a fair number of specimens were taken in various
districts of Mid-Lancashire and North Staffordshire. Wireworms
were taken in soils growing a variety of crops, but the most abundant
and sure source of specimens was found to be the potato crop; a visit
to a potato field rarely proving a disappointment. Pastures, especially
those that had been down for a long period, proved a good source for
material. They were, however, only utilised until sufficient observa-
tions had been made, owing to the lengthy examination required and
the uncertainty of results. The material when collected was put into
fairly deep tin boxes, loosely packed with damp soil. The larvae were
examined individually under a binocular microscope in the laboratory,
being measured and roughly classified into sizes. They were placed,
after examination, in plant pots previously prepared. As this type
of breeding cage proved to be useful, no trouble from disease having
been experienced, some further details may be of interest. The drain-
age hole of the plant pot was covered with a circular piece of fine wire
gauze strengthened by having a strip of zinc soldered round the edge.
The gauze was roughly two-thirds the width of the pot; so, no matter
how it slipped, the gauze would still cover the drainage aperture and
prevent the escape of any larvae. The top of the flower pot was covered
over by a perforated zinc plate, having a flange of about an inch in
depth. The flange and the weight of the zinc were sufficient to keep
the pot covered securely, even in the most boisterous weather. The
vessel was filled first with small pieces of broken pots, and then with
earth in the usual manner; a potato was placed about half way down
in the soil to provide food for the larvae. The pots were buried out
of doors in the ground, up to the level of the soil inside the pot. The
larvae lived quite healthily in these pots, and no sign of disease was
ever observed. For experiments requiring a greater depth of soil,
two ordinary three-foot drainpipes were used; the narrow end being
capped with cement, which was sloped inside to a central aperture
covered by wire gauze imbedded in the cement. These drainpipes
were very useful, but were too weighty to handle easily. They were
handicapped by being buried in stiff clay soil, which prevented good
drainage and rendered the drainpipes liable to become waterlogged.
Buried in soil of a more open texture and better drainage capacity,
they would be quite free from any disadvantage, save that of weight.

Throughout this work, the endeavour was to keep all the conditions
as closely approximating to nature as possible.

100 The Larval and Pupal Stages of Agriotes obscurus

The greater part of the laboratory work has been carried out in the
Department of Agricultural Entomology of Manchester and I am also
much indebted to Professor Hickson for affording me various facilities.
Acknowledgment has already been made to Dr Imms. I am also
indebted to Mr J. T. Wadsworth for the photograph of the pupal ce'l
with the pupa im situ. Mr J. C. F. Fryer of the Board of Agriculture
has kindly given me information as to distribution, etc., of the common
wireworms. The work has been carried out during my tenure of a
Board of Agriculture studentship and completed during the following
year.

9

3. NoTES ON THE BIOLOGY OF THE LARVA.

The larval stages of Elateridae are practically ubiquitous, occurring
in all types of soil. The larvae of Agriotes obscurus have been found
most commonly in the lighter types of soil, particularly loams rich in
organic matter. The range of soils that they inhabit is extraordinarily
varied, including the grades between fairly stiff clay to light loam.
They are practically always to be found in pastures, even when on clay
soils, though in such a case they are much less plentiful. In Cheshire
and Lancashire the common wireworm is the larva of A. obscurus, as
was demonstrated by breeding experiments. This seems to be corro-
borated by the fact that Newstead records adult A. obscurus (when ina
state permitting of identification) far oftener than that of A. lineatus,
as being the common species taken as food by wild birds over the same
area. Very probably, A. obscurus is the common species in, the north
of England; while A. lineatus seems to be the most common species
further south, though A. obscurus and A. sputator are not far behind
in point of numbers. An interesting feature with regard to A. obscurus
is, that the occurrence of its larvae seems to be, in some way, closely
connected with the presence of organic matter in the soil. The larvae
are most abundant in soils containing a large percentage of humus,
due either to the peaty nature of the soil, or to heavy manuring. The
same thing was found in cottage gardens, which are usually very rich
in organic matter. This point will be considered later in connection
with the feeding habits.

The larvae of A. obscurus were found in what may be roughly
described as three stages. These stages were, of course, not sharply
defined, but merged into one another. They may be given approxi-
mately :

Small stage (1) 7 mm. long by 0-75 mm. broad,

~~?

G. H. Forp 101

Medium stage (2) 10—15 mm. long by 1-0—1-25 mm. broad.

Large stage (3) 17—23 mm. long by 1-5—2-0 mm. broad.

It should be stated that no living specimen was ever found to be
more than 21 mm. long, and that 19 mm. is about the average length,
taking a number of specimens into consideration. The 23mm. specimen
was an extended spirit example. After about a year, the small stage
was no longer present, and only medium and large stages could be
found in the plant pots. According to most authorities the larval
period is supposed to last five years. From observations on the length
and rate of growth, I am inclined to place the period at four years
rather than five. The small stages, taken in one year from July to
October, varied much in size. It was found, after a couple of months
(November—December), that a number of these apparently small
specimens were really of medium size. It appears from this that the
breadth is a safer criterion than the length (which is difficult to ascertain
on the living larva). Judging from the size of specimens taken from
July to October, and also from the fact that the adult beetles are
common from April to June, it may be assumed that the newly hatched
larva commences its free existence in July. Again, there are three
larval moults, performed in the usual manner, by splitting the chitinous
skin along the thorax, and leaving the old exo-skeleton behind. From
these facts, and from the rate of growth, it may be assumed that the
larva is full grown in three years. Full size larvae are found actively
feeding up to October; but these same larvae apparently do not feed
again (at any rate on vegetation), and pupate the next summer. It is
thus probable that there are three stages limited by the three moults,
and occupying three years. Then there is a period of active feeding,
followed by a quiescent condition terminating in pupation, all this
taking place in a year. It was found that of the large larvae (17 to
23 mm.) a number fed through the winter, while some were not feeding
after October, and were not lethargic; these latter were broader than
were the former. This would appear to support the above theory.
If the life of the larva is one of five years, then it would seem that the
early (small) larval stages would continue over a longer period; this
is not confirmed by observations. The rate of growth was found to be
of such a uniform character as to suggest that the curve of growth is
fairly continuous rather than irregular. It seems to the author that
the period of growth is not five years, a period given since Bierkander,
but probably requires four years from the egg to pupation. The early
(small) stage larvae were never found feeding. It is stated that they

Ann. Biok ut i

102 The Larval and Pupal Stages of Agriotes obscurus

feed on organic matter in the soil. The medium and large stages were
almost invariably observed feeding (with the exception mentioned
above in the case of the larger stage), and it seems that they feed con-
tinuously, save in the months of December to February inclusive.
This observation was obtained from regular examination of potatoes
placed in the breeding pots. Feeding is usually active by March and
seems to decrease at the end of November. It is very difficult to starve
these wireworms. Of 10 large wireworms placed in sifted soil without
food November 3rd, 1914, nine were present January 22nd, 1915, and
seven on March 8th, 1915. The three missing ones had probably been
eaten by the others, as traces of the skins were found. These seven
larvae had thus existed in soil free from any vegetable matter visible
to the naked eye for 125 days; of the seven larvae only one remained
on May 4th, 1915. Ina duplicate experiment under similar conditions,
10 larvae lived from November 3rd, 1914, to February 11th, 1915, i.e.
100 days. The larvae will tolerate a large amount of moisture, and
will exist for at least a week in water-logged soil. On the other hand,
wireworms kept in soil not suppled with water and allowed to dry up
will die after a couple of days. It is hard to explain their independence
of vegetable growth (such as pasture grasses and various crops) for
food, in any other way than by assuming that they feed on the organic
matter in the soil. This would help to explain the usual occurrence
of these larvae in the lighter soils plentifully supplied with organic
matter. In several cases, small and medium sized larvae were found
buried in the farmyard manure under growing potatoes. It could not
be said definitely that the larvae were feeding, but the condition of
the mouth parts seemed to indicate that they were.

The usual depth at which these wireworms are found in the soil
varies from one inch in pasture, to eight or nine inches in potato fields.
In winter they bury themselves much deeper, and were found as much
as two feet deep in the large drainpipes previously described.

The larvae bury themselves in soil surprisingly quickly, though their
rate of progress on the surface is comparatively slow.

Kighty-three larvae had buried themselves at least two inches deep
in soil, of about the same consistency of that in a potato field, in less
than twenty minutes. On being exposed to light, they immediately
travel in the opposite direction to the source of the light. They travel
slowly, but fairly easily on an unpolished surface, and often progress
carrying the abdomen bent in a slight curve to one side. This is one
of the difficulties that hinder accurate longitudinal measurement.

G. H. Forp 105

4. DESCRIPTION OF THE LARVA.

The full-grown larvae of A. obscurus measure about 19 mm. « 2 mm.
(Plate XVI, figs. 1 and 2). They are semi-cylindrical and their colour
varies from pale yellowish white, to yellow brown. The young stages
are of a paler colour, which darkens as the larvae become older. The
freshly-moulted larvae are easily distinguishable by their pale yellow
white colour. The junctions of the body segments are of a slightly
darker shade than the rest of the body, and are in addition marked
with faint longitudinal striae (Plate XVII, fig. 6). The spiracles are
placed on darker chitinised areas. The muscular impressions of the
ninth abdominal segment are strongly chitinised dorsally and ventrally,
and slightly so laterally. The legs are of a darker brown colour.

The head is approximately as broad as long, with the anterior
margin of a dark brown, rounded at the angles. On each side of the
head, just behind the antennae, but slightly more towards the median
line, is a slightly raised, irregularly rounded, pigmented eyespot. Imme-
diately behind each eyespot is a long slender seta. The antenne are
short (Plate XVII, fig. 10), each consisting of three segments. The first
segment is stout, brownish round the base, with a clear area round the
summit. The second is dark brown, longer in proportion to its breadth,
narrower than the preceding segment, and somewhat enlarged on its
outer and upper margin where it bears two stiff upright hairs. The
third segment is dark brown, about one half the dimensions of the second
segment, and bears two terminal lobes which are slightly lighter in
colour than the second segment. The lobes of the third segment,
when seen from above, consist of a long narrow lobe (almost as long
as the second segment) flattened at the apex, and a smaller lobe placed
ventrally to the larger lobe and nearly half as long as, and a little
thicker than, the latter. The third segment bears, in addition, two
setae. The head region bears two pairs of setae on the lateral margin,
placed near the anterior and posterior edges. No attempt is made
to discuss the homologies of the head region.

The mandibles (Plate X VII, fig. 7) are of a dark brown colour, irregu-
larly flattened dorso-ventrally, and sickle shaped. The apex issomewhat
blunt, and just below this is a blunt tooth on the inner margin. This
blunt tooth is common to the genus Agriotes according to Schiodte (16).
A second tooth that is more pointed than the former, projects nearly
half way down. Just below, is a well-marked process (/a) projecting
inwards which is the “lacinia mobilis.”.. Mangan(11) in describing the

—

104 The Larval and Pupal Stages of Agriotes obscurus

mandibles of Periplaneta australasiae, quotes Hansen (Ann. Mag. Nat.
Hist., 6, vol. x11, 1893), as recording the “lacinia mobilis” as occurring
in certain Coleoptera. The mandible articulates with the head, by
means of the condyle (co) and the ginglymus (97).

The first maxillae (Plate XVII, fig. 8) are strongly chitinised around
their outer margin, and are of a uniformly brown colour. They possess
a distinct cardo (ea) mto which the chitinised margin continues, swelling
into a small dark area. There is no definite distinction between the
stipes (si) and the terminal lobe of the maxilla. The stipes bears,
on its outer margin, two long thick setae and three shorter ones, all
close together, and placed rather more than half way up. The maxil-
lary palp (mp) consists of four joints, of which the first is broad in
comparison with its length, the second nearly as broad but longer,
the third narrower and shorter; all three segments are brown, darker
at the edges, with clear areas at their summits. Of the three segments,
each bears setae; the first, two irregularly placed; the second, two
at the summit and two on the outer margin; the third, two on the
summit. The fourth segment is short, rounded terminally, and of a
light brown colour. <A galea (ga) and lacinia (la) are present in each
first maxilla. The galea is short, and composed of two joints; both
being about as long as the first two segments of the maxillary palp.
The terminal segment bears four short, slightly-twisted setae. The
lacinia is not plainly differentiated from the galea, and bears a large
number of closely-packed, slender, waving setae.

The labium (Plate XVII, fig. 9) consists of the basal mentum, which
is brown, and slightly chitinised at the lateral margins. It bears,
at the margin on each side towards the base, a long curved seta, and
another on each side towards the apex of the labium.

The mentum articulates with the palpiger, which bears a two-
jointed palp on each side. The palpiger is shield-shape, and of a dark
brown colour, darker at the sides, and lighter in the median area, and
with a clear space across the anterior margin. It bears two long setae
placed anteriorly. The first joint of the labial palp swells out towards
its extremity, and bears a whorl of small setae, commencing about
half way up from the inside, and progressing outwards and upwards.
The second joint of the labial palp is rounded, and appears to possess
a flattened, pale area at the tip.

Between the two labial palpi is the ligula (//) which bears a short
seta on each side. The further setae, much smaller than the previous
two, are placed at the base of the ligula.

LO 5 Wie

G. H. Forp 105

On the underside of the head, adjacent to the first maxillae, is a
dark, strongly-chitinised ridge, diverging anteriorly and converging
posteriorly. A dark line runs parallel, close to the preceding ridge.

The prothorax has a curved furrow running down each side, and is
nearly equal in length to the meso- and meta-thorax taken together.
The hairs on the body of the larva are arranged in a definite manner,
which seems to be fairly constant. The first eight abdominal sterna
each possess three pairs of hairs, arranged symmetrically along the
margins of these sterna. On examining the larva in its lateral aspect,
it is seen that the first eight abdominal terga, considered up to their
mid-dorsal line, possess two lateral pairs of hairs; of which the posterior
pair is placed near the margin, a little to the left of the plane of the
anterior pairs, which are placed nearer to their margin than the former.
Placed mid-laterally, and just over each spiracle (of the first eight
abdominal segments), is a pair of short hairs about half the length of
the other hairs. In a line with the spiracle and in the same plane as
the two posterior hairs, is a single hair of similar length to the paired
ones. The ninth abdominal segment bears, on the base of the pseudo-
podium (or anal papilla), two pairs of short hairs. Theré are nine
pairs of spiracles, the first being thoracic (Plate XVI, fig. 2, st), and
placed ventrally. The remaining eight pairs are laterally placed on
the first eight abdominal segments anteriorly in each segment. The
ninth abdominal segment affords a means of distinguishing the genus
Agriotes; Perris(14) describes the typical anal segment of the genus as
follows: “Dernier segment assez-longuement demi-ellipsoidal, terminé
par une pointe, ayant de chaque cété, prés de la base, une cavité de
Papparence d’un grand stigmate.”’

This applies truly to the larva of Agriotes obscurus, but requires to
be added to. The anal segment is terminated by a slight constriction,
which swells out into a blunt point, outlined in black (Plate XVII, figs.
5 and 6). The cavity, that Perris describes as being like a large stigma,
is not a respiratory opening, but is possibly a muscle attachment?.
The walls of the cavity are black, and thickened; only slightly at the
sides, but strongly dorsally and ventrally. In shape it is oblong, with
the anterior and posterior walls rounded. It is placed very high up,
and near the base of the segment. The stigmata are oblong, with
their dorsal and ventral walls closely apposed, and strongly chitinised ;
their median line points forward and upwards, save in the case of the
meso-thoracic stigmata, where the median line points forward and

1 Henriksen (8) also speaks of this structure as an ‘‘eye-shaped muscular impression.”

106 The Larval and Pupal Stages of Agriotes obscurus

outward. The ninth abdominal segment is equal to the length of the
seventh and eighth segments combined, and possesses a ventral papilla
or pseudopodium. This, according to Sharp(17), “represents a body
segment and is generally described as being the protruded termination
of the alimentary canal.” The pseudopodium is round, sloping gradu-
ally to meet the eighth segment, and with a sudden slope posteriorly
to meet the remainder of its segment. The apex is white and flattened
and bears a few short hairs. The pseudopodium is of some assistance
to the larva in walking. The three pairs of legs are provided with
spines and are dark brown, short and terminating in a curved claw.
The legs do not require any special description; the right leg of the
third pair is figured in Plate XVII, fig. 11.

5. DESCRIPTION OF THE PUPA.

Before proceeding to describe the pupa, some details will be given
with regard to pupation, etc. The first pupa was found on Aug. 14, 1915,
pupation had therefore-occurred within seven days, as the last previous
examination had taken place on Aug. 7, 1915. At the previous exami-
nation it was observed that several larvae were very sluggish, and sub-
sequent events showed that they were preparing for pupation. The
first adult Agriotes obscurus was observed in a freshly emerged condition
on Aug. 30, 1915. This would place the pupal period at from two to
three weeks. Observations on other pupae show that this period is
correct. Pupae were found up to the end of September, and all the
adults had emerged by the 9th of October. Pupation takes place in
an earthen cell at a depth in the soil varying from three to twelve inches.
The pupal cell is oval and allows ample space for pupation. The
average dimensions of three cells was 5 by 14 by 7 or 8mm. The
photograph (Fig. 1) shows a section through a typical pupal cell, with
the pupa lying in a characteristic position.

The internal walls of the pupal cell are smooth, and appear to have
been slightly cemented together with some glutinous secretion. The
cell will stand air-drying without crumbling away, to a greater extent
than an artificial cell made of similar soil but not cemented. The
method by which the data concerning the pupal cell were obtained
is as follows:—the plant pot, taken into the laboratory and the zine
cover removed, was filled level to the top with soil of a known depth,
the soil being slightly compressed to produce cohesion. A piece of stout
cardboard was placed firmly over the top and the pot then inverted so

Sor » ati. an teenie

——as

G. H. Forp , 107

as to stand upside down on the cardboard. The cardboard was gently
removed, and the soil in the pot loosened by pressing slightly through
the drainage aperture. The pot was then lifted up with a twisting
movement; a cast of the interior of the pot was left. This cast could
easily be examined with regard to the depth of any specimens in the
pot, as the depth of soil added just previous to examination was
known. With this method, pupal cells in perfect condition were
obtained without trouble. The pupa (Plate XVI, figs. 3 and 4)

Fig. 1. Pupa of Agriotes obscwrus in earthen cell. x2. (Phot. J. T. W.)

measures up to 13 mm. long by 3 mm. broad, gradually tapering in the
abdomen to a breadth of | mm. The general colour is white, speckled
on the apices of the abdominal segments with small patches of rusty
brown colour. The thorax is nearly as broad as long, and swollen.
Two spines project from the anterior angle of the thorax, which is
provided at the posterior margin, on each side of the middle line, with
a small protuberance. The long, spatulate, posterior corners are each
provided with a curved thorn-like structure, of a brown colour and
pointing outwards. The third to the sixth abdominal segments are

108 The Larval and Pupal Stages of Agriotes obscurus

slightly tooth-shaped at their lateral posterior margins. There are
nine abdominal segments, of which the ninth is provided with two
spines. Hach spine possesses a small spur at the base. According
to Curtis(6), these spines are movable. The ninth abdominal segment
is provided ventrally with a flap, broad at the base, suddenly narrow-
ing half way down the segment, and continuing in a smaller piece
with parallel sides, which terminates in two sharp points. Beling(2)
describes the pupa of Agriotes obscurus in a similar manner to the above,
and adds that the segments of the antennae are as long as broad, that
the wing cases reach to the middle of the fourth abdominal segment,
and that the third pair of legs reach to the middle of the fifth abdominal
segment. In the pupa described above, the third pair of legs only just
reach on the fifth abdominal segment, and do not stretch half way.

6. COMPARISON witH LARVAE OF CLOSELY ALLIED SPECIES.

The evidence showing that the common wireworms taken by the
author in Cheshire and other districts are larvae of A. obscwrus is as
follows :—

All the larvae were examined under a binocular microscope, and
were apparently all similar. At this time I was inclined to believe
that the common wireworm represented the larval stage of Agriotes
lineatus. A comparison of my material with Schiodte’s figure of the
anal segments of lineatus(16) did not solve the problem. The larvae
undoubtedly belong to the genus Agriotes, and agree in general with
the figure of lineatus given by Schiodte, but differ as to certain details.

Seven larvae at last pupated and the six adults that emerged were
Agriotes obscurus. This seems to be corroborated by the interesting
observations of Newstead (previously mentioned), where the area
covered was mainly Cheshire, and part of Lancashire, and the records
of adult Elaters devoured by birds give A. obscwrus far more commonly
than A. lineatus. As the occurrence of wireworms is admittedly
universal, at any rate in this country, and as their attacks are not
epidemic, these facts seem to point to A. obscurus as being the parent
beetle of the larva in question. The identification of the larva pro-
vides the next problem.

The author has been unable to procure an undoubted specimen of
the larva of A. lineatus for comparison, so reference was made to the
literature. It was found that there is a scarcity of good accounts of the
larvae belonging to the genus Agriotes. Some of the older accounts were

G. H. Forp 109

incomplete or even wrong, as Perris(14) quotes Bouché (Naturg. p. 186),
as describing a larva of an Athous or Corymbites as A. lineatus. The
best accounts of Elaterid larvae were found in Perris(14), Schiodte (16),
and Beling (2), from which sources the comparisons are procured. The
distinction between the genus Agriotes and that of Hlater seems fairly
clear. The possession of a blunt tooth, on the internal margin of the
mandible (Plate XVII, fig. 7), indicates the genus Agriotes according
to Schiodte, while Beling gives the presence of the chitinous cavity
on the base of the ninth abdominal segment as indicating the genus.
Beling describes A. lineatus and A. obscurus, as follows:

(1) Larve, sehr fein und seicht punktirt, fast glatt, blass braunlich
gelb. In erde, vorzugsweise auf Aeckern....Agriotes lineatus.

(2) Larve unregelmassig seicht gerunzelt, starker und dichter
punktirt, auch etwas dunkler als die vorhergehende gefarbt, mit der-
selben angleichenorten lebend....Agriotes obscurus.

Beling further says that the larva of Agriotes ustulatus Schall. is
similar, but is smaller. He adds that the larva of A. obscurus seems
more strongly and thickly hairy (punktirt) and wrinkled than lineatus ;
but characteristic differences remain to be found. His description of
the lineatus larva would apply equally well to that of A. obscurus.

The position of the muscle attachments on the anal segment seemed
to hold out a hope of identification, as the one illustrated by Schiodte
differed in position from those observed by the author. This is un-
fortunately of little account, as the method of preserving the larva
affects the position. If the larva is simply killed by being immersed
in alcohol, it usually contracts, and the wrinkling of the skin and the
position of the muscle attachment, shown in Schiodte, are obtained.
Boiling the larva, and then passing it through different strengths of
alcohol, preserve it naturally. Neither the muscle attachment placed
so closely to the base, nor the wrinkled skin as figured by Schiodte,
is ever seen in living or properly preserved specimens. There is a
difference, however, between the pseudopodia. Schiodte figures the
pseudopodium proper (as distinct from its base), as exhibiting three
divisions, the third probably pertainmg to the non-chitinous and
fleshy tip. In the larvae examined by the author, there were never
more than two (including the fleshy tip). As there is no reason to
suppose Schiodte’s figure is incorrect, this difference of the pseudo-
podium must be taken as specific. The mandible figured above
(Plate XVII, fig. 7) differs from that of Schiodte.

A number of mandibles (both left and right) were examined and

110 The Larval and Pupal Stages of Agriotes obscurus

were found to agree substantially with the figure given. Schiodte,
however, figures the blunt tooth near the apex of the mandible as very
slight, while the author’s figure shows it to be more pronounced.

Furthermore, the position of the spiracle and hairs on the eighth
abdominal segment, as represented by Schiodte, does not agree with the
condition observed by me in the larvae of A. obscurus. These differences
are constant in the specimens examined during this work?. The only
other description to which I have had access is that of the larva of -
Agriotes pallidulus Il. by Schiodte. The dimensions of this larva,
9mm. long by 1 mm. thick, and its intense yellow colour, together
with its anal segment, which is rounder than those of A. obscurus or
lineatus, prevent confusion with either the larvae of A. lineatus or
A. obscurus.

It would seem, that if the above differences between the larvae of
A. lineatus and A. obscurus do not apply generally, then one is possibly
a variety of the other, as it was formerly believed (Curtis, loc. cit.,
pee).

7. NATURAL ENEMIES.

It would be natural to assume that the wireworms, on account of
their underground existence, are fairly immune from enemies that
prey upon them. In spite of this, they are subject to the attack of
a number of persistent foes; particularly various species of wild birds.
Wild birds are accountable for the destruction of vast numbers of msect
larvae. This is especially so during the nesting season, as “ practically
all birds except doves and pigeons feed their young on an animal
diet, whatever may be the character of the food of the adult” (9).
Again, as Collinge remarks—“It should be remembered that the
nestling season is also that when the destruction of injurious insects
is most needed—i.e., at the period of greatest agricultural activity
and before the parasitic insects can be depended upon to reduce the
pests” (4 and 5). At present, definite information as to the food of
nestling birds is somewhat scanty; but according to Collinge, the

? Owing to Mr Ford’s departure to Rome he was unable to consult Henriksen’s
paper (8) before he left England. The Danish entomologist mentions that the larva of
A, lineatus is “brownish yellow, faintly rugose and punctulate, mostly rugose.”” With
regard to that of A. obscurus he adds “dark brownish yellow. On each side, between
“‘tergum and sternum a pale longitudinal stripe. Punctulate and with few rugae.” J may
add that the longitudinal stripe is not evident in any of Mr Ford’s specimens. Henriksen
does not appear to have reared any examples of A. obscurus from the larva to the adult.
A. D. Imms.

Gy. H. Forp 111

Starling, Thrush, and Blackbird seem to feed Klaterid larvae to their
nestlings as usual constituents of their diet. To quote one example—
that of the Starling (Collinge, loc. cit.)\—the food taken to one nest
throughout an averaged period of half a day, required 146 visits and
included 13 wireworms (species not stated); this was in the middle of
May. It is probable that a large number of Elaterid larvae (as well
as other pests) are thus disposed of during the nestling season. With
regard to the food of adult birds more information is available, and it
is possible to give a rough list of birds that are of importance as
destroyers of Elaterid larvae, for “the frequent occurrence of these
insects, and especially of the parent beetle, is very marked and goes
a long way to prove that they form part of the regular food supply
of various species of birds” (13).

The most useful in this respect are Plovers (or Lapwings), Gulls,
Rooks, Jackdaws and Starlings, all devouring both adult and larval
stages of Elateridae. In districts in proximity to the sea, (rulls are
very beneficial—they are usually present where land is being ploughed ;
and they follow the plough, in numbers, searching for insects that are
turned up. In one instance (Newstead, loc. cit.), forty-five Agriotes
larvae were found in the crop of a Black-headed Gull (Larus ridibundus).
Over the country generally, the Plover (or Lapwing) is one of the
most beneficial birds to the farmer—in spring, its habitation is
usually ploughed land. It undoubtedly destroys many injurious
insects.

The Rook is not quite so wholly depraved as it is thought; for though
it certainly takes a fair amount of seed corn, yet wireworms form about
9 per cent. of its food (Leigh(10)), and the young are fed on Elaterid
larvae (Newstead, loc. cit.). Jackdaws and Starlings, both as nestlings
and adults, feed on both larvae and imagines of Elateridae.

Curtis(6) states that the Sparrow preys on wireworms, but, up to
the present, there is apparently no evidence available that might
mitigate the uselessness of this ever-present pest. The author has
little definite information regarding the particular proportion of the
food of birds formed by the larva of Agriotes obscurus.

Wireworms appear to form a fairly constant constituent of the food
of the common Mole (Talpa europaea), in fact insect larvae and worms
form the greater portion of its diet. In an investigation on the food of
moles (19) it was found that out of 100 moles, 41 of the stomachs con-
tained wireworms. Though this investigation only extended over a
short period (from Dec. 5, 1913, to Feb. 5, 1914), it yet serves to indicate

Ww

112 The Larval and Pupal Stages of Agriotes obscurus

the value of moles from the economic point of view. Digestion in the
stomach of the mole is apparently very rapid, thus rendering identifi-
cation of the stomach contents difficult. It is, at any rate, certain
that the animal is a very voracious feeder as Scheffer(15) found that
Scalops aquaticus, a mole distributed over the eastern region of the
United States, will consume more than its own weight of suitable food
in a day. In the stomachs of nine moles which I examined it was
found that three stomachs contained remains of Elaterid larvae (the
remainder of the food was mainly various insect larvae, earthworms
and slugs). It is interesting to note that in the districts where these
nine moles were trapped, one farmer would not allow moles to be trapped,
save when they were becoming too plentiful; on all the surrounding
farms the moles were trapped regularly. The district has been examined
for wireworms, and it was found that on the farm where the moles
were protected, wireworms were very scarce, and then could only be
found in small numbers in a very old pasture. On the surrounding
farms, where moles were trapped, wireworms were taken in numbers.
The animal is often a nuisance in grassland and cornfields, but it should
be remembered that the mole is burrowing after food, and not for
amusement. Again, as its food consists principally of insect larvae,
and, as Adams(1) remarks, the mole does not hibernate, but continues
active all the year round, it can well be seen how useful the animal is.
These facts together with its voracity, and its need for meals at frequent
intervals (Adams, loc. cit.), show the utility of the animal. It certainly
destroys large numbers of wireworms, and should therefore be merely
prevented from becoming too abundant. According to Graf(7), an
important enemy of adult Elaterids is to be found in the Carabidae,
species of Calosoma being particularly important in the control of
Limonius californicus (sugar beet wireworm). Curtis (loc. cit.) mentions
that the Carabs Plerostichus madidus and Nebria brevicollis are enemies
of the wireworm. While collecting between 1914 and 1915, Carabid
larvae and small adults (Nebria, sp.) were commonly found in potato
drills, where wireworms were attacking the tubers. The adult Carabs
most commonly found were Pterostichus madidus and Nebria brevicollis.
It was found, as the result of carrying away two wireworms and a
Nebria brevicollis in a tin box containing earth, that the Carab attacked
the wireworms. On examination, after the day’s work, the Nebria
was alive, and the two wireworms dead and mutilated. Under natural
conditions, no Carabs were seen attacking wireworms, though injured
larvae were seen.
-

G. H. Forp. 113

Wireworms with their tough highly chitimised skins, and under-
ground habits, are almost free from internal parasites. Curtis(6),
however, says that wireworms have a Hymenopterous parasite and
quotes Bierkander as having found them. Forbes (18th Rept. State Ent.
p. 47, 1891—2) records a parasitic fly as having been reared from a wire-
worm. Out of over 300 larvae collected, not one attacked by an internal
parasite has been seen. Graf found the same result with 10,000 larvae of
Limonius californcus. In this connection, I am informed by Mr Fryer
that he has not infrequently found wireworms parasitised by what is
probably a Proctotripid, at all events a Hymenopteron, but has never
succeeded in breeding it out. No bacterial or fungoid disease was
seen in any of the larvae in the field, or in the laboratory. In this
connection, Graf reports as having occasionally observed fungoid
attack in the field, but says that one fungus and two bacterial diseases
were common under laboratory conditions. He also says that one
bacterial disease was very fatal to the young wireworms, while the
older stages seemed to be immune.

According to Leaflet number 10 of the Board of Agriculture (1916),
wireworms are parasitised by a species of the fungus /saria.

The pupae are difficult to rear; any disturbance or any slight
variation in humidity or temperature, was sufficient to cause them to
perish, and from the same causes some pupae passed into the perfect
state but failed to mature. No disease of any kind was noticed.
Graf notes that a few pupae of Limonius californicus were killed by a
fungus.

Graf records a fungoid disease as having caused the death of adult
Elaters under natural conditions.

Wireworms seem to be little affected by physical agencies under
experimental conditions. They were found to be little affected by
excess of moisture, but they could not resist dryness and were killed
by frost. Under natural conditions it is probable that these factors
are of no importance, as the larvae can always burrow down deeper
in the soil, out of the way of frost, or in the search of greater humidity.

8. SUMMARY OF GENERAL CONCLUSIONS.

1. This paper contains an account of the larval and pupal stages
of the Elater, Agriotes obscurus, whose larvae, together with certain
other related larvae, are known as “ wireworms.”

114 The Larval and Pupal Stages of Agriotes obscurus

2. The common “wireworm” in Cheshire, North Staffordshire,
and South Lancashire, is the larva of Agriotes obscurus.

3. The life of the larva is probably four, rather than five years.

4. The larva of Agriotes obscurus can be distinguished from that
of A. lineatus by means of the development of the blunt tooth, just
below the apex of the mandibles (more prominent in A. obscwrus);
by the orderly arrangement of the body hairs, and the more anterior
position of the stigmata; and by the pseudopodium or anal papilla
on the ninth abdominal segment of A. lineatus having three apical
divisions, while that of A. obscurus only has two.

5. Birds, particularly the Plover (or Lapwing), are of great value
in checking the larvae of Agriotes obscurus (as well as other related
species). The Plover, which is wholly beneficial, should be stringently
protected. The common Mole (Talpa europaea) is also of great value
in checking the pests, and should not be wantonly destroyed, unless
increasing in too large numbers. The amount of damage caused by
a mole is probably very small in comparison with the amount of good
it does.

6. The larva pupates in an earthen cell in the ground, down to
one foot deep; the pupal period is about three weeks; the imago
remains resting motionless in the pupal cell for roughly two months,
after which it comes to the surface, and hibernates under stones, clods,
ete., until the next season.

BIBLIOGRAPHY.

1. Apams, L. E. Talpa europaea, habits, etc. Mem. Manchester Soc., 1902,
Vol. xtvu. Art. 4. ;

2. Bertrne, TH. Beitrag zur Metamorphose der Kiferfamilie der Elateriden.
Deutsche ent. Zeit, 1883, xxv. p. 141; 1884, xxvin. p. 199.

3. CHapurs and Canpize. Larves des Coleoptéres-Elateridae. Mém. Soc. Roy.
des Sciences, Liege, 1855, p. 139.

4. CoLimincr, W. E. Food of nestling birds. Journ. Bd. Agric. Sept. 1912.

5. —— _ Food of some British wild birds, 1913. Dulau and Co.

6. Curtis, J. Farm insects, 1860, pp. 152-195.

7. Grar, J. E. Prelim. report on sugar-beet wireworm. U. S. Depart. Agric.,
Ent. Bull., No. 123, 1914.

8. Henriksen, K. L. Oversigt over de danske Hlateride larver. Entom. Meddel.
Iv. 1911, pp. 225-329, 76 figs.

9. Jupp, 8S. Food of some nestling birds. Yr. Bk., U. S. Depart. Agric., 1900.

Leteu, H. 8S. Interim report on feeding habits of Rook. Econ. ornith. Comm.,

1914.

>

THE ANNALS OF APPLIED BIOLOGY. VOL. III, NOS. 2 and 3 PLATE XVI

ry

THE ANNALS OF APPLIED BIOLOGY. VOL. Ill, NOS. 2 and 3 PLATE XVII

%

15.

Goal horn [15

Manean, J. On the mouth parts of some Blattidae. Proc. Roy. Irish. Acad.,
Vol. xxvi. B. No. 1, 1908.

MarsHam, T. Note on the wireworm. Trans. Linn. Soc., 1808, Vol. xx.
pp. 160-1, p. 118, f. 4.

NewsTEapD, R. Food of some British birds. Swpplt. Journ. Bd. Agric.,
Dec. 1908.

Perris, M. E. Larves des Coleoptéres, 1878, Paris, pp. 161-188.

ScHEFFER, J. H. The common Mole. Kansas Bull., No. 168, 1910.

Scuioptr, J. C. De Metamorphosi Eleutheratorum observationes, 1870,
Vol. u. pp. 470-536, Pl. VIII.

SHarP, D. Camb. Nat. Hist., Vol. vt. (Insects, Pt. 1m), p. 188.

Westwoop, J. O. Intro. to Mod. Class. Ins., 1839, Vol. 1. pp. 237-8, fig.

Wuitr, P. B. Food of the common Mole. Journ. Bd. Agric., Aug. 1914,
XXxI. pp. 401-7.

XaAMBEU, Cpt. V. Mceurs et Métamorphoses des Insectes. Ann. Soc. Linn.
de Lyon, 1912, p. 111; 1913, p. 28.

DESCRIPTION OF PLATES.

PLATE XVI

FIGURE

Ile

~I

9.

10.
ne

Fully grown larva of Agriotes obscurus, lateral aspect, drawn from an extended
spirit specimen. ant. antenna; mu. muscle attachment; sf. spiracle; ps.
pseudopodium. x8.

Mature larva, as above, ventralaspect. ps. pseudopodium; si. first spiracle. 8.

Pupa of Agriotes obscurus, ventral aspect. ant. antenna (left); le. elytron (left).
x 8.

Pupa of same, dorsal aspect. 8.

PLATE XVII
Lateral aspect of the ninth abdominal segment of the larva of Agriotes obscurus.
ps. pseudopodium; mu. muscle attachment. x 24.
Ventral aspect of the ninth abdominal segment of the same. ps. pseudopodium.
x 24.
Right mandible of the larva viewed from above. gi. ginglvmus; co. condyle;
la. lacinia mobilis. x 50.

Right first maxilla of the same viewed from below. ca. cardo; sf. stipes; mp.

maxillary palp; la. lacinia; ga. galea. x 50.

Labium of same, viewed from below. m. mentum: pg. palpiger; Up. labial palp;
li. ligula. x 50.

Right antenna, seen laterally. -d/. dorsal lobe; vl. ventral lobe. — x 100.

Right lee of third pair of the larva. co. coxa; fc. trochanter; f. femur; ¢b. tibia;
ta. tarsus; cl. claw. x49. (a

116

ON THE BIOLOGY AND ECONOMIC SIGNIFICANCE
OF TIPULA PALUDOSA.

By JOHN RENNIE, D.Sc., F.R.S.E.
(North of Scotland College of Agriculture.)

Part II. Hatcuinc, GRowTH AND HaBits oF THE Larval.
-)

(With Plates XVIII—XX and 3 Text-figures.)

THE most common species of Crane-fly larva occurring in grass
and corn land in the north-east of Scotland is Tipula paludosa. Tipula
oleracea occurs also, but is much less frequently met with. Along
with these, there has also been found in comparatively small numbers
in fields the larval stage of Pachyrhina histrio, but this species appears
to occur more frequently in garden ground. The following Tipulidae
in addition have been found in the winged stage in the district sur-
rounding Aberdeen:

Tipula varipennis, common and generally distributed.

T. gigantea, in small numbers.

T. lutescens, in small numbers.

Pedicia rivosa, L. widely distributed in the northern area, but not
common.

The Egg.

Hatching of the flies goes on during the months of June, July,
August and September, and as already recorded (Part I) the first
mating and oviposition may take place within a very short period.
In captivity, hatching mating and oviposition have all occurred within
a few hours. A newly hatched female contains considerably over 400
shelled ova. In two such taken at random the actual numbers were
found to be 448 and 490. A third female captured out of doors in coitu
contained 255 black shelled ova together with a quite small number—
about 12—of pale coloured immature shelled examples. The form of

! The work recorded in this series of papers bas been carried out with the aid of
Grants from the Board of Agriculture for Scotland,

eR tld

J. RENNIE 117

this female when found indicated that oviposition had previously
taken place. A female Pachyrhina histrio taken in the open contained
259 black shelled ova.

The egg measures 1-1 mm. x -4mm.; it is black in colour, with
a dark purplish metallic lustre. As development proceeds this lustre
diminishes, and finally before hatching the shell is of a dull black colour.
The covering of the egg is a strong tough membrane, which is completely
formed around the egg before copulation takes place. I have been
unable to detect a micropyle, but this may be present. It is possible
that the membrane before coming in contact with the air is permeable
to the spermatozoon. The somewhat remote possibility of partheno-
genesis taking place with the first brood of eggs has not been overlooked,
and females have been kept apart from males from the period of their
hatching until death, but oviposition never took place under these
conditions.

The Early Larva.

The emergence of the larvae takes place in about 14 days after the
eggs have been laid. They are then of a pale reddish sandy colour,
and about 2°7 mm. in length, Plate XIX (b). When fully extended,
thirteen body segments can be made out. On each of these from
the third to the twelfth there is a small tuft of laterally placed,
moderately strong bristles. The thirteenth segment which bears the
spiracles and terminal papillae, has a pair of tufts of relatively stronger
and longer curved bristles, borne alongside the large lateral conical
para-anal papillae. These tufts constitute the most striking difference
between the early larva and the later form.

Through the skin the two longitudinal tracheal trunks are visible,
and also the alimentary canal with its four anteriorly and one posteriorly
placed diverticula. The masticatory apparatus is well developed, both
mandibles and first maxillae being strongly toothed.

In a short time the segmental bristles tend to become very short or
worn away, and so also do the anal tufts. Traces of the lateral bristles
persist even in the fully formed ‘ grub,” but the anal tufts disappear
completely. These changes appear to come about by attrition. In
about 12 to 13 days from the time of hatching the larvae are 4—5 mm.
in length when fully extended, and already resemble the older and more
familiar “leather jacket.” By about three weeks they have attained
a size of 6mm. They feed from the first day onward.

Ann. Biol. 11 8

118 Biology and Economic Significance of 'Tipula paludosa

The fully grown Larva: External features.

Owing to the difficulty of keeping alive recently emerged larvae which
were prevented from burrowing into the soil, it has not been possible so
far to follow the changes in external form effected at the various moults.
The larva when fully grown attains at full extension a length of about
40 mm. It is now of a brownish-grey earthy colour interspersed with
irregularly placed blackish dots. Frequently the longitudinal tracheal
trunks, two in number, may be seen through the skin. The shape is
cylindrical, slightly narrowed anteriorly, and expanded posteriorly into
a peristigmatic papillae-bearimg area. The skin, which is generally
tense in healthy larvae, exhibits the following characteristics :—along
each side there is a moderately wide strip which on the animal
contracting folds outward, forming a pair of blunt keel-lke longi-
tudinal ridges. Besides numerous transverse wrinklings, there are
slight but definite transverse furrows marking off distinct segments.
Eleven of these can generally be counted. Each segment bears on
its ventral surface four very minute bristles, two lateral and two near
the middle line slightly in front of the lateral pair. On the first four
segments behind the head dorsally, there is a row of bristles, and a pair
of dorso-lateral bristles on succeeding segments. The head bears a
pair of short jointed antennae; there is a very strong chitinous and
highly complex mouth armature which includes massive toothed
mandibles with palps, a pair of serrated first maxillae, notched
united second maxillae and elaborately folded and bristled labrum
(Plate XIX (a)). The whole set of structures is based upon a large
strong bivalved chitinous support which surrounds the gullet.

The anus, which is sub-terminal, is surrounded by large fleshy
lobes and a pair of large retractile laterally placed conical papillae.
Beling regards these papillae as characteristic for this species.

On the somewhat truncated terminal region there is a pair of large
brown coloured circular stigmata, each with a lighter glistening dull golden
marginal ring. This stigmatic area is expanded on its border into six
conical papillae, of definite form and arrangement. There is a ventral
pair whose tips are black with a clear central area. This pair appears
to have a sensory function, and may be seen at times in the living
animal apposed to the stigmata above. Below each of these ventral
papillae there is a pair of small pigmented spots which are some-
times united to form a short streak on each side, The remaining

J. RENNIE 119

four papillae project dorsally in two pairs. These bear on their exposed
surface numerous fine hairs which follow the boundary of an elongated
shghtly raised conical area; the outer pair is tipped in black. Fig. 1.

Fig. 1. Stigmatic area in Tipula paludosa, (a) showing papillae with hairs, stigmata,
and pigment spots; (6) showing anus and fleshy lobes, together with para-anai

papille.
DP. Dorsal papille ; LP. Lateral papille ;
VP. Ventral papillae; AP. Para-anal papille, outside stigmatic area: A. Anus.

Duration of Larval Period.

In the N.E. of Scotland the adult flies may be seen frequenting
cultivated land from early in June to the beginning of October.
It appears to be the accepted opinion in England (Theobald, Agricultural
Zoology, 1913, p. 228) that there are two generations of these flies,
T. paludosa, and T’. oleracea, in the course of the year. Our observations
have shown that in this area, probably owing to the higher latitude
and more rigorous climate, there is only one.

The following observations made upon a small collection of flies
reared from eggs which hatched in September, 1913, are typical of the
results obtained in rearing during several years. The parent flies had
hatched out within small laboratory cages or had been caught upon
the college farm and placed in these. The cages had wooden roof and

$—2

120 Biology and Economic Significance of Vipula paludosa

base, wire gauze sides and glass front. They contained a bed of turf two
to three inches deep. Mating and oviposition readily took place, and on
the turf being broken up later a considerable number of recently hatched
larvae was found. A later search in the month of October, however,
showed the mortality to have been considerable. On Ist November, the
turf was again broken up and the surviving grubs collected and measured.
The lengths were taken by allowing the larvae to crawl upon a sheet
of paper and pricking this at the moment of their maximum extension.
About one half the number of larvae was found to be under 16 mm.
in length, and the remainder from 30 to 35mm. It is remarkable
that the larvae may attain to nearly their maximum length quite
early, but it must be noted that they are relatively slender at this
period. Subsequent growth takes place in the direction of thickness.
By the end of the month the disparity in size was not so great. They
all conformed to Beling’s description of Tipula paludosa larva. In
November clover was sown in the cages; this for a time afforded
opportunities for feeding but was allowed to die out during winter.
These larvae lived throughout the greater part of the following summer
in soil containing decaying vegetable matter only, and duly pupated
and emerged as adults in July and the early days of August.

As a control upon the above, portions of second year’s grass upon
the farm were dug up in October, the turf was disintegrated, shaken
up in sieves, and the soil searched. Larvae were found and measured
in a similar manner, and the commonest size at this date was found to
range from 20 to 25mm. These also were identified as 7. paludosa.

Larvae collected out of doors during November and the first half of
December showed a maximum size of 20mm. These were relatively
slender at this degree of extension, and contracted when handled to
much smaller bulk than the laboratory reared specimens.

The larvae collected out of doors on 1st November, and which were
separately caged, but kept under similar conditions, when fully extended
measured 36 mm. in December, but in this state were distinctly more
slender in the body. Examination of the contents of the alimentary canal
showed that they had fed upon the decaying rootlets in the soil. As
the season progressed it was found that there was eventually no signifi-
cant difference in size between those reared indoors and those living
a natural life outside. The records of soil temperature taken on the
farm showed that there had been little frost during this experiment.
The winter was a mild one, and the facts suggest that some feeding at
least had taken place amongst the larvae out of doors, In a subsequent

- - we Gomer ew

J. RENNIE 121

season caged larvae kept short of food were found in February to
measure from 20 to 30mm. Generally, there has been found out of
doors great variation in the size of larvae at the end of winter in the
same district and even upon the same field.

Tipula paludosa has been kept under direct observation throughout
its whole life cycle, and owing to variations in the length of the larval
stage pupation and consequently hatching of adults is spread over a
considerable period, viz., in this district, June to September. (Rarely,
I have found adults in the cages in May.) Under experimental con-
ditions of limited food supplies larvae have been kept alive and been
continuously under observation for fifteen months. The minimum
duration of the larval period has been found to be about nine months—
September to June. Before all the larvae of a season have pupated
the next season’s larvae may have appeared, so that there may be
larvae present in the soil all the year round. There is a possibility that
this fact may have led to the view that there are two generations of
flies in the year. I have had under observation in breeding cages in
the month of July, larvae, pupae, and adults of one generation, together
with developing eggs and emerged larvae of the next generation—all
alive at the same time; and in the variable climate of the region under
observation such occurrences are not improbable in the field.

Bionomics of the Larva.

The newly emerged larvae are very susceptible to drought, and
when kept in dry soil were found to die off quickly. Strong sunlight,
even when the soil is moist, was also fatal in a short time. Artificially
reared larvae require to be kept moist and sheltered from direct
sunlight, otherwise the mortality in the early days is very great.
Larvae which are hatched from eggs which have been placed upon the
surface of the ground immediately burrow into the soil, avoiding the
light. A large proportion of larvae reared from eggs in 1913 died in
the course of the first eight weeks, especially towards the end of this
period, notwithstanding all attempts to reproduce natural conditions,
and only a comparatively limited number of flies have been reared
from many thousands of eggs laid in the laboratory cages.

In view of the fact that very large numbers of eggs are laid, and of the
probability that the adults only rarely approximate to these numbers,
there must be a considerable mortality in the course of the life-history,
due, of course, to various factors. Our experience suggests that the first

122 Biology and Economic Significance of Tipula paludosa

few weeks of life constitute a period in which the insect is particularly
susceptible to the prevailing physical conditions. While difficulty was
experienced in rearing large numbers of larvae from the egg to the
adult stage, no such difficulty was met with in rearing flies under the
same laboratory conditions from larvae collected in the field in late
winter. It may be suggested for example that wet weather in the end
of summer, and early autumn months, will favour the survival—apart
from natural enemies—of greater numbers of larvae, and that con-
versely, prolonged drought will tend to kill off numbers of those hatched
about this time. In this connection it may be worth while to quote the
opinion held by some farmers in this area that a wet summer and
autumn foreshadows a plentiful supply of crane-fly in the following
year.

The Larva on Farm Lands.

Published references to the activities of the larva as far as I have
been able to trace them deal exclusively with instances of serious or
even excessive damage effected by these insects upon grass or corn
crops. But in the course of the present investigation it has become
clear that Tipula larvae are very commonly present upon farm lands,
sometimes in considerable numbers, without their presence becoming
apparent. Cases of excessive damage have also been experienced, but
the following instance may be taken as an average experience under the
conditions named. It is quoted in full because it illustrates a number
of features related to the larval habits.

These observations were made in the spring and summer of 1913,
upon the College farm at Craibstone. Search was made for the presence
of Tipula larvae in the end of March and beginning of April. The
weather was cold at the time, and the searches were not very fruitful.
The Woodlands field (Fig. 3) which was in grass at this time was selected
for enquiry, samples being dug up at a number of places, and the turf
thoroughly examined, but no Tipula larvae were obtained. Grey slugs
were particularly plentiful. This was on 3rd April. On the 19th,
ploughing was in progress and the plough was followed, samples of the
furrow were taken, disintegrated, and searched, but no larvae were
found in this way. Further search by two observers resulted im four
larvae being found. These were found under stones at the surface,
on the part not touched by the plough. After the field had been sown
and rolled it was again examined on the 29th, and larvae were now
found to be very numerous under the turf clods upon the surface. In

J. RENNIE 123

the interval between the times of examination there had been a good
deal of rain.

On the 3rd of May the field was visited at 6 a.m. The two previous
days had been dry and sunny, but in the end of April there had been
much wet. The morning was fine, and at 6 a.m. the sun’s rays had
reached the western end of the field only. The eastern end was still
in the shadow of the trees.

A search was commenced at the eastern end where there was some
frost in the ground. In about 45 minutes 94 larvae had been collected.
In the southern hollow where the sun had now reached, 42 were obtained
in about 10 minutes; on the crest of the field at the west end (in sun)
75 were found in 20 minutes, and on a low part (N.W. corner), in the
sun, 15 in 15 minutes. Two collectors were at work. In all in about
14 hours 226 larvae were obtained. In some cases from six to a dozen
were found in a single piece of turf. The smallest number appeared
to be present in the shade at the highest level of the field (EK. end).
The larvae were found mostly under the turf clods, and largely in the
“mids”; sometimes they were lying on the soil below, and sometimes
embedded in the turf with heads weil buried amongst the roots. They
were not seen distributed generally amongst the sown grain.

The field was again examined on the 10th of May. There had
been continuous and heavy rain for several days previously, and the
ground was very wet. The oats had “brairded” early in the week,
but the wet weather had rendered rolling impossible. At the time of
search rain was not falling, but there was a mist close down to the ground.
Larvae were frequently found beneath the loose turfs upon the surface,
and they were also to be seen crawling freely on the ground. Some
trouble was taken to find them in the act of attacking the young crop,
but with no success. The ground in a number of places was scraped
with a toothed digger and the plants turned over. A few larvae were
found in the ground in this situation, ie., free from turf clods and
amongst the soil in which the oats were growing, but none was seen
actually attacking any part of the crop. The oats were scarcely an
inch above ground. In many places no corn could be seen and here
the ground was turned with the digger. A few larvae were found in
this way, but it could not be said that they were more numerous than
in other situations. It was found that mostly the seed in these places
lay deeper and was growing all right.

During the following week the weather improved and there had
been several warm and dry days. On the 17th the field was again

124 Biology and Economic Significance of 'Tipula paludosa

examined. It had been rolled during the previous day. The crop
looked well, and there were no indications of Tipula attacks. Larvae
were again found in the most usual situation, viz., below the turf
clods; 40 were collected in a few minutes. Search amongst the
growing oat plants resulted in only one larva being found; none was
seen upon the surface, and none detected attacking the crop.

On the 21st another examination was made. The weather in the
interval had been showery, but not very cold. The day was warm
and there was some wind. The field was carefully searched, particu-
larly for traces of larvae moving freely in the soil or actually attacking
the crop. They were found in the usual places below or burrowing
into loose “foggage” upon the surface. In a few cases they were
found below flat stones at the surface. Only a very few were obtained
by searching the open soil around the oat plants. In this search the
soil was turned over with a digger and the oat plants uprooted. Some-
times the ground was scraped and stirred. The examination ought
to have discovered larvae if they were present in the soil in proximity
to the roots of the oat crop, and it is concluded they were absent in
this situation at the time of search, viz., between 10 a.m. and 12 noon.
The weather at the time was warm and showery, and larvae could
always be found amongst the decaying turf. The crop at this date,
notwithstanding the undoubted presence of Tipula larvae in large
numbers, showed no bad effects. Up to the beginning of July, when
the last search for larvae was made upon this field, there was no apparent
effect of the presence of Tipula upon the crop. On this occasion the
search was effected by cutting the crop over certain areas, and sifting
the soil by spade and sieve. Larvae were obtaimed, but no pupae
were seen.

It should be mentioned that in view of the presence of Tipula m
appreciable numbers and the possibility of an attack ensuing, a plot
experiment was early arranged upon the field with the object of testing
the effect of rolling and of some common manurial substances. The
experiment, though negative in its results as far as its original purpose is
concerned, is given here because it confirms the conclusion that Tipula
was not visibly damaging the crop. The field, which is surrounded by
trees, was rich in humic matter. Below is given a diagram of its
situation, and of the experimental plots together with the report of
Mr W. Findlay, N.D.A., Superintendent of Field Experiments.

J. RENNIE 125

S
Treated) with) Paperworks
Lime

E
t iy

i}
Rolled 3/5/13
J

16/4/13

3/5/13

Fig. 2. Diagram of Experimental Area of Field.

Report. The Woodlands field at Craibstone was ploughed out of Lea during the
second week of April, 1913, and immediately sown with Sandy oats at the rate of
six bushels per acre. A heavy rainfall prevented the whole of the field being rolled
at that time, and an interval of about two and a half weeks elapsed before the soil
was dry enough to finish the operation. One strip the whole length of the field
was rolled three times, and other two strips were rolled twice.

On the 21st May three plots were laid off and treated as follows:
1. 4 ewts. Paperworks’ Lime per acre.
2. 1k ,, Nitrate of Soda per acre.
3. 2 ,, Common Salt per acre.
The accompanying plan will show the scheme of the different treatments and
cultivations.

At no time was there any difference either in the thickness or strength of the
crop between the parts rolled at different times.

The plot to which Nitrate of Soda was applied showed an increased crop of
about 20 per cent., but those to which Lime and Salt were applied could not be
distinguished from the rest of the field.

Wo. M. Frnpiay.

126 Biology and Economic Significance of Tipula paludosa

The experience here recorded has been general for a series of
years; large numbers of Tipula larvae have been regularly obtained
from lea fields upon farms on Deeside and elsewhere in the neighbour-
hood, which during the periods tested had no crop losses due to their
attacks. In a good many such cases the numbers obtained from single
fields were considerable. A case where serious damage was effected
is given upon page 132.

“AN
"2 tete Cat ee
: gets i

‘

Fig. 3. Plan of Craibstone Home Farm.
Experimental Field, No. 5.

Summarising the outstanding features of this record we note:
(1) An apparent scarcity of larvae in spring before ploughing took place.
The failure to find Tipula was not as subsequent finds proved due to
their smallness of size, and it does not seem likely that they were deeply
situated in the soil at this time. They have rarely been found below
six or eight inches under the surface. Tests made with deep cages in
winter (Feb. and March) yielded only insignificant numbers (eight
per cent.) below six inches from the surface.

(2) When known to be present in the field they could not be found
in the act of attacking the crop. The suggestion that they feed at
night is plausible, but observations on larvae in cages have shown that
they feed readily at all times. The fact. that there was an abundance
of humic matter in the soil is probably not without significance in this
connection (see pp. 127—8).

J. RENNIE 127

(3) The occurrence of most of the larvae at the surface beneath
loosened decaying turf. Their presence here in the spring is general;
they occur both loosely below, and also very frequently deeply embedded
in the turf. The disturbance of the soil in ploughing and harrowing
is probably the cause of their gathering in these situations, and their
presence is probably primarily due to the need for shelter and moisture.

(4) The absence of harmful effect upon the crop notwithstanding
their presence in considerable numbers throughout the spring and
summer. This feature is considered in connection with the further
data given below.

Further consideration of feeding habits.

With a view to rendering clearer the feeding habits of the larvae
and to throw light upon the circumstances under which they attack
growing crops numerous experiments were made of which the following
are illustrations.

I. A small lot of larvae reared from eggs laid in September was
kept in ordinary field soil covered with loose turf in a small laboratory
cage. No crop was sown in the soil, but it was watered from time
to time. They wintered under these circumstances and continued
throughout the following summer. The larvae pupated in July and
the last of them emerged as flies on the 4th August and mated on the
same day. The life cycle in this case occupied about eleven months.

II. A collection of larvae was kept out of doors in small boxes contain-
ing ordinary garden soil without growing vegetation during the months of
May, June, and July. They survived this treatment, but were under-
sized. Some managed to pupate, but others died in the larval stage.
One imago was observed to fail in the act of emerging from the pupal
case. The larvae of this group did not on the whole do so well as those
of lot I, and the mortality towards the end of the experiment was
high. Dissections showed the presence of vegetable fibres in the
intestine, and a considerable amount of gritty material. It may be
mentioned that this latter is normally present in the intestine.

III. A collection of larvae was kept in small cages with no growing
plants, and a limited amount of decaying vegetable matter amongst
the soil. These conditions were maintained during the months of
May and June of the present year. At the end of June all were alive
and healthy looking, and some were well grown. The cages were
set in a large field rearing box containing washed sea sand, amongst

128 Biology and Economic Significance of Tipula paludosa

which the metal cages containing the larvae were sunk. A consider-
able number of larvae wandered from the cages and were subsequently
recovered amongst the sand, alive and to all appearances quite healthy.
Adult flies were later seen emerging from the cages in the laboratory
to which they were removed during July. The larvae of this and
Experiment II were collected in the fields and had wintered out of doors.

The foregoing, together with other experiments and observations of
a like nature have shown clearly that the Tipula larva may subsist in
the soil and complete its development independent of the presence or
absence of a growing crop upon the ground. The results here obtained
have led to the institution of further and more exact experiments
dealing with the nutrition of soil larvae including Tipula. These are
at present in progress.

IV. A number of larvae caged in the autumn of 1915 were kept
in soil without growing vegetation except for a short period when a
small quantity of corn sown in the cage germinated. In February
they were found to have reached a fair size. Several killed and ex-
amined on the 19th were found to contain soil particles and fragments
of vegetable fibre. At this date they were found mostly in compact
earthen cells formed against the sides at the bottom of the cage. This
habit has been frequently observed in winter and suggests a quiescent
period under the adverse conditions of cold, confinement and restricted
food supply.

In order to determine more clearly the circumstances inducing
destructive attacks upon preee the following type of experiment was
resorted to:

Groups of larvae were put up in large cylindrical glass cages of
about 10 inches diameter, in prepared soil, in which the visible amount
of vegetable matter was very slight. This soil was, further, mixed
with well washed sea sand. Around the cylinders between the glass
and the soil, oat seed was introduced. The cages were kept at room
temperature and were examined daily. After the corn had germinated,
the larvae were kept under close and continuous observation for pro-
longed periods at a time. The larvae appeared sluggish, and not much
movement was seen in the day time, although their burrows soon
became very numerous between the soil and the glass (Plate XX).
They could be seen lying in these, and after the corn had germinated,
or even before this, they could be seen attacking it, gnawing at the

J. RENNIE 129

husk, radicle and plumule. They were also seen eating the blades
which had come above the ground. The glass cylinders had removable
ends of perforated zinc of fine mesh. A few larvae passed through
the perforations at the bottom although these were small Within
the cage they tunnelled freely to a depth of six or seven inches. After
a week, when all the corn appeared to have germinated, and both
radicle and plumule were of some length, the cylinders were removed
and the state of the seedlings ascertained. These were separated
out carefully and placed in water. The soil was removed as far
as possible by gentle washing, and each seedling examined in turn.
Care had been taken that no other creatures were present in the soil
capable of damaging the oats. There were usually about 25 larvae
present. <A typical result is given:

34 seedlings apparently sound.

1] had radicle or rootlets more or less eaten away.

39 had plumule cut through or bitten into. This includes a number attacked

in the blade, above ground.

8 seeds were attacked. In some cases this had been effected before germi-
nation, in others, afterwards. In some of these the food store was com-
pletely eaten out, in others only partially so.

92

a Equal to 63 per cent. of the plants damaged. ~

The factors contributing to such a result appear to be:

(1) The absence of decaying or other vegetable matter, with con-
sequent restriction to the crop for food supplies.

(2) The large proportion of larvae to the cubic content of soil—
two dozen to about 14 cubic feet of soil.

(3) The confinement of the larvae to a limited amount of space;
it was not possible for them to leave the cages.

A further factor to be noted is the favourable conditions of warmth
and suitable moisture, inducing rapid germination. This would be
of some importance in the field.in favour of the crop “growing away”
in a proportion of cases before the damage done became irremediable.

In other experiments of this nature with germinating oats, clover,
or timothy grass, the numbers of larvae used were very much greater,
with correspondingly serious damage. It was found subsequently in
examples of very severe attacks upon crops in the field, that the numbers
used in these experiments were in excess, and that serious loss may
ensue where the numbers, in average samples of an affected area, are
about 10 to 15 per square foot of surface (see p. 133).

130 Biology and Beonomic Significance of Vipula paludosa

Migrations.

The use of large numbers of larvae upon experimental plots or
open shallow cages brought to light the fact that scarcity of food in
spring due to over stocking results in the larvae migrating. The
first proof of this was obtained from a lot of 120 larvae contained
in a small box in soil in which clover had germinated. As the food
supply became exhausted the larvae left the box in large numbers
and were found crawling about the laboratory. Two other instances
of this occurred in connection with field plot experiments at Craibstone.
In these experiments, small plots of 2} feet by 6 feet were walled off
with wooden boards sunk six inches in the ground and standing about
three inches above the surface of the soil. It had been established
in experiments in a previous winter that these larvae even in very cold
weather rarely went below this depth of six or seven inches, and it is
not considered likely that they burrowed beneath the limiting boards
of the plots. The plots were stocked with large numbers of larvae;
in one instance three adjacent contained 900 larvae each and in other
three plots, there were 700 each. When it was discovered that the
crops in these had failed, at the end of June the soil in each was sifted
for larvae. The hatching season had just commenced, but only a
very few flies had been seen; of the 2100 larvae in the second group
of three plots only 295 were recovered, together with a small number
of pupae and two or three empty pupal skins.

Another plot was not disturbed. It had stood exposed in the same
way as its neighbour, for some weeks, but later a large field cage was
fitted upon it. Here there had been placed originally 2700 larvae,
and now to these were added the 295 recovered from the other
plot. In due course flies hatched out and appeared in the cage, but
the numbers up to the end came very far short of those of the grubs
introduced. The soil was searched and no trace in the form of empty
pupal cases or dead larvae was found.

In a duplicate experiment on Holm Farm, in Lewis, carried out at
the same time, designed to test the relative resisting or recovering
power of selected oat varieties, it is probable that a similar migration
took place. This experiment consisted of plots of Hamilton, Sandy
and Potato oats, into each of which 900 larvae were placed. There
was also a control plot of Hamilton oats, which received no larvae.
The oats were sown on 18th May, and “brairded” on 3rd June.

J. RENNIE 131

The plots were under continuous observation till the end of June, during
which time no difference was observed in the growth of the different
varieties. At this date the plots were searched and only a few larvae
could be found. The local observer states that he is confident birds
did not interfere with the plots, and the only conclusion that can be
drawn is that the larvae migrated from the plots before the crop had
germinated. Had they remained and continued their development,
pupae and empty pupal cases would have been found in the plots.
As a matter of fact these were examined before adult flies had begun to
appear. Had the larvae died within the plots it is certain that the
remains of some at least would have been found, when the soil was
passed through the sieve. But nothing of the kind was found.

The question remains—do they migrate in appreciable numbers
normally, in search of food? On the fields under observation I have
found them wandering upon the surface, but so far only occasionally.
With regard to the plots, it is to be borne in mind that the larvae had
been placed in excessive numbers on soil which did not contain any
growing plants (except the ungerminated corn), while around there was
abundant green vegetation. Their power to climb up even a few
inches of a vertical board was not expected; that they did so was also
evidenced by their being found in the control plots in which none had
been placed.

A further interesting case, bearing on the point, is recorded in an
editorial note in the Scottish Naturalist, 1915, p. 1.

The reference is brief and may be quoted in full:

““A few weeks ago it was reported to us that the inhabitants of a
certain district in Perthshire had been seriously alarmed by an invasion
—not of Germans—but of an immense number of small worm-like
creatures which crawled over the road near the houses in such numbers
as to make walking decidedly unpleasant. Examples of the creatures
were brought to us, and were recognised as ‘leather jackets’—other-
wise the larvae of Tvpula oleracea, the commonest species of crane fly.”

As bearing further on this subject, the Editorial note proceeds to
quote from an article by P. Desol, in Compt. Rend. Soc. Biol., Paris,
uxxvil. No. 21, June, 1914pp. 126—7.

The summary given is quoted from the October 1914 issue of the
Review of Applied Entomology (Series A). As the conclusions here are at
variance with my own on 7’. paludosa and with the occurrences just
described they are reproduced. The observations were made in the
meadows of Avesnois. “In the spring of 1914, circular patches of from 15

152 Biology and Heonomic Significance of Tipula paludosa

to 60 feet in diameter, or the entire areas of meadows, were to be seen
covered with yellow and dead grass the roots of which were found to
be full of the earth-coloured larvae of Tipula oleracea. Grasses and
clovers are chiefly attacked while plants with hard or thick roots are
not affected. Where the infested zone borders on a furrow the larvae
fall into it during the night and being incapable of climbing out may be
collected in large numbers. They are not migratory, so that healthy
meadows may be found close to infected ones. Infection is due to
chance circumstances bringing fertilised Tipulids to the ground.” Iam
satisfied that the statement that the larvae are not migratory is erroneous
and that the existence of healthy meadows near them may easily be
otherwise explained.

The conclusion to the experiments just described—unsatisfactory
as regards the object for which they were planned—was due to the
excessive numbers of larvae used on the plots. The large numbers
were decided on with a view to avoiding an indifferent result, and this
was supported by field experience, where large numbers had been
seen and which had not proved harmful. In its result, however, it
is useful and the possibility of migrations from waysides, waste land,
or grass fields to germinated corn crops cannot be neglected. The
experience recorded on p. 122 may be recalled, in which a field ex-
amined in April, and found almost negative as regards the presence
of Tipula larvae, was later found to have an abundant supply. This
problem is under continued investigation.

Various field experiments and observations have been made with the
view of testing the effect of manurial agents and various insecticides
upon the larvae. One of these is quoted below, but in general the results
obtained were not consistent, and in cases of some well-known insecti-
cides were not corroborated by direct tests upon larvae in confinement.
The experiments were useful mainly in bringing into prominence some
factors which are contributing elements affecting the degree of attack
upon the crop, and which are considered in the succeeding section.

©
Field Experiment: Holm, Stornoway. Oat Crop.

In the previous season the field was grazed by cattle right on to the
end of November, and with sheep and cattle on to January. Ploughing
began early in January, and was finished by the end of February.

J. RENNIE 133

Weather.

From the beginning of October to the middle of March it was
exceptionally wet and boisterous. There was practically no frost
and the temperature was normally mild. The Farm is worked on
the five shift system; the present tenant came to the farm in 1910,
and previous to this the farm had been much neglected, especially as
regards drainage and weeds. Oats were sown in the last days of March.
After sowing, the field was immediately harrowed—one single and two
double. It was not rolled till 23rd May.

The oats “brairded” in about a fortnight from sowing time and
looked strong and healthy for a week. About 20th April the crop:
began to look sickly in parts and by the end of another week the
presence and activities of Tipula larvae became apparent. The
weather for some weeks after sowing was warm and sunny.

The state of the crop was reported to the College authorities in May,
and on 24th May arrangements were made to treat the field experi-
mentally. At this time, it showed parts fair, parts very thin, and
fairly large tracts were quite blank. All over the field leather jackets
could be readily found, and on the bare parts they were present in very
large numbers; scores could be found near the surface an a square
foot of soil. A part of the field including the worst portions was
plotted as under:

( |
peg Plot Plot Plot | Plot Plot
{ Rolled |
once | = F a
No. 5 No. 4 No. 3 | No. 2 No. 1
4 ewts. 1 cwt. | 1 cwt.
Unrolled No manure common Nitrate 2 ewts. salt Nitrate
salt of Soda | of Soda

The manures were applied on the 6th of June, and two-thirds of
the total area rolled early the following morning. On the 14th June
one half of that part was rolled a second time. The other third was
not rolled.

On June 26th the larvae were still to be seen, and active. Ten
trial counts were made on square foot samples taken at random. These
were dug to a depth of six inches and the larvae counted. The follow-
ing are the numbers found: 4, 5, 4, 4, 5, 4, 12, 14, 14, 13. No sample

Ann. Biol. mm 9

134 Biology and Beonomic Significance of Tipula paludosa

tested was found blank. The last four lots were from Plot 5, unrolled
part. Plots 1, 3, and 4, showed improvement on the crop at this date.
On July 14th the results showed as follows:

Plot 1. Originally the thinnest. Recovery good.

Plot 2. Originally 2nd thinnest. Now same as Plot 1.
Plot 3. Originally the best. Much ahead of all.

Plot 4. Originally only medium. Second best.

Plot 5. Originally 2nd best. Worst of all.

There was no very obvious difference between the once and twice
rolled parts of the plots, but the unrolled parts were decidedly poorer.
Larvae were still active in the field at this date.

On 15th September the field was visited and the crop as a whole
found to be in a very good condition, it having tillered remarkably
well. The variety of oat was Hamilton. Signs of the larvae having
been present were scarcely apparent, any patching being extremely
slight. The nitrate plots did extremely well, the salt ones also were
very good. From the appearance of the crop in May, it would have
been impossible to have foreseen such an excellent recovery. Local
circumstances did not allow of the estimation of the yield of the
separate plots by weight per plot.

Factors favouring the Larva.

It may be regarded as well established that Tipula larvae are
generally distributed and ordinarily present in the soil, though not
necessarily only upon farm lands. Even upon these they have been
found to be present in appreciable numbers, when no recognisable loss
in crops resulted. The question of importance is,—what are the cir-
cumstances determining a destructive attack upon a crop?

Naturally the first condition of importance is the presence of
excessive numbers. The various factors contributing to the periodical
appearance of large numbers of any species of insect have not hitherto
been appreciated in advance, and indeed cannot be said to be well
understood. In general with regard to Tipula they may be held to
include :

(1) favourable weather conditions for the survival of the large
numbers of young which are generally produced:

(2) unfavourable conditions for competitors:

(5) ready supply of food:

(4) absence or diminution of numbers of natural enemies,

J. RENNIE 135

As regards the influence of the weather in affecting numbers it has
been already noted that the newly hatched larvae are very susceptible
to drought and that the mortality in laboratory reared larvae which —
are not kept moist is high. There is a fairly general popular impression
that severe winter with plenty of frost kills the larvae, but our experi-
ence so far does not bear this out. Larvae were left in the open exposed
upon stones during a night of severe frost, and were found alive the
following morning. The apparent beneficial result following frosts
seems to be due to the improved tilth favouring a more rapid primary
growth of the crop in spring. The environmental factors are the subject
of continued investigation. Unfortunately, owing to the presence of the
larvae on cultivated land, the food supply is adequate. Food is not
ordinarily scarce, and as already indicated a high percentage of larvae
kept im captivity complete their development upon ordinary turf in
the soil.

The question of natural enemies is dealt with in a succeeding paper.

Tipula attacks on Oats.

Conditions unfavourable to the oat crop may render it suscep-
tible, and Tipula, even when the numbers are not excessive, may
work havoc. Apart from the question of manuring and _ general
farm practice which are not considered here, the weather in spring
has been found to be a significant factor. These observations have
been carried on during a period of seven years; along with these
the experience of over 130 farmers on a wide area has been collated
and there is a universal testimony to the fact that a cold late spring
in which the primary growth of the plant is delayed constitutes one
of the most certain conditions for crop failure due to Tipula attack.
It is during the early days of growth from the time of sowing until the
adventitious root system is established that Tipula attack results in
the destruction of the individual plant. After this period the plant
may be regarded as out of danger; it may be weakened but probably
it will not be killed outright by subsequent attack. Any cause there-
fore tending to extend the period of germination or immediately
subsequent growth increases the liability of the crop to loss from
Tipula attack. In this area the time between sowing and brairding
for oats in an average season is 10 to 14 days. In a series of seven
cases which came under observation in one season in which failure of
parts of the oat crop, attributed to Tipula attack, took place, this

136 Biology and Economic Significance of Tipula paludosa

period ranged from 16 to 21 days. There was one case in which,
owing to severe drought, it lasted six weeks. The actual proportions
of crop failure were stated thus:

Period between sowing

and germination . Proportion of crop lost

16 days one-fourth of whole crop
6 weeks one-third of whole crop

17 days one-tenth of whole crop

3 weeks loss incurred; proportion not estimated

16-18 days almost one acre destroyed

3 weeks one-third of crop

3 weeks one-third of crop; it recovered later

On this account early sowing, especially in our north-eastern
climate, is attended with a certain amount of risk. The same risk
would apply to all late districts in seasons when fine weather early in
spring tempts the farmer to sow early.

Tipula and Oat Varieties.

The question of the relative resisting or recovery power of different
oat varieties in relation to Tipula attack has been investigated both
in the field and in plot experiment. The experience of the farming
community within the College area has been obtained and discussed,
and the general view is given below. There exists some opinion that
the larva exercises a selective capacity and prefers some varieties to
others, as is evidenced by the following statements:

“They (i.e., the Tipula larvae) like Potato oats better than Sandy.”

“Potato oats are more liable to attack than Red Oats or Sandwich.”

“They are more fond of ‘ Potato’ varieties.”

“Waverley suffers from attack more than Potato or Sandy.”
*** Banner’ oats suffer more than Potato.”

All the evidence, however, goes to show that this is not the correct
interpretation. It is not the larvae which “ prefer’’ particular varieties,
but particular varieties which show better power of recovery from
their attack. Some of the writers expressed their experiences more
carefully :

“Of the ‘Waverley’ and ‘Hamilton’ oats which were sown side by side, the
‘Waverley’ had to be resown; ‘Hamilton’ oats were only slightly thinned.”

“The ‘Red Oats’ resisted poorly, ‘Providence’ and ‘ Potato’ did better.”

Taken collectively the views of over a hundred farmers who have
had experience of Tipula attack is contradictory with regard to the

THE ANNALS OF APPLIED BIOLOGY. VOL. III, NOS. 2 and 3 PLATE XVIII

THE ANNALS OF APPLIED BIOLOGY. VOL. III, NOS. 2 and 8 PLATE XIX

| ~
| S
2
hel =
a\\ =
\\\ yn
\\ [a]
~~
S
a
—
a)
—
ae —
LE
if X
i we
ee CS,
Sa
Wes
~ SS
te
SSeS
—
=
=

Del, E, Harvey

THE ANNALS OF APPLIED BIOLOGY. VOL. III, NOS. 2 and 8 PLATE XX

pa Sse

Del, B. Simpson

ne ne eri een ee ne

J. RENNIE 137

relative merits of particular varieties. This is likely to be the case
so long as other factors such as quality and source of seed, type of soil,
particular agricultural practice, manuring, and so on, are not con-
sidered. But there is general agreement that the newer varieties of oats
have distinctly less power of recovery, because they do not tiller well.
This is well borne out by wide experience.

Another practice which has been found of service in overcoming
Tipula is to effect a “change of seed.” Seed grown on the coast and
sown in an inland locality, and seed grown in the Lothians and sown
in Aberdeenshire have in both instances proved more resistant where
larvae were at work, than native grown oats. Further, seed from an
early district and sown in a late one, also from a light soil to a heavy
one, have in both instances been proved satisfactory measures against
losses from Tipula attack.

The subject of preventive agricultural practice and remedial
measures in relation to Tipula are dealt with in a further paper.

REFERENCES TO PLATES.

Plate XVIII. Tipula paludosa. Larvae. 3 times natural size.

Plate XIX. a. Head armature of Tipula paludosa. (A) antenna; (B) mandibles;
(C) first maxillae; (D) fused second maxillae; (E) labrum; (F) internal basal support
of appendages.

b. Newly hatched larva of Tipula paludosa.

Plate XX. View of larvae in soil within glass rearing cage, showing their burrows and
damaged oats, which were grown between the glass and the soil. Sketched from nature.

REFERENCE.

Bexine, TH. Beitrag zur Naturgeschichte verschiedener Arten aus der Familie
der Tipuliden. Zool. Bot. Gesellsch. in Wien, xxtu, Bd. 1873, pp. 575-592.

158

NOTE ON ATTACKS OF PHYLLOTRETA
VITTULA ON SPRING CORN.

Phyllotreta vittula has recently been recorded by several continental
writers as a serious pest of spring corn. It has also been recorded by
Lind Rostrup and Ravn! as a serious pest of winter barley in Denmark.
The following extract from Baranoy’s account of this pest in the Govern-
ment of Moscow is taken from the Review of Applied Entomology,
Series A, Vol. 1, page 214. “This beetle becomes yearly more numerous
and the damage done by it to the summer-sown crops is increasing.
The author never found it or noticed any damage done by it on the
winter-sown crops. It winters in the province of Moscow in the imago
stage among the roots of summer stubble and of Triticum repens.
It emerges from the earth with the return of warm weather (the middle
of May in 1912) and immediately attacks the sprouting grain. It does
the most injury at the time of pairing, and at the beginning of June
its injurious activity decreases. It feeds on the epidermis and paren-
chyma of the leaves, forming channels in them parallel with the veins.
If the number of channels is not great the leaves may recover. The
author did not succeed in finding eggs of this insect. The harm done
to the plants by the larvae was considerably greater than that done
by the adults, and the author noticed that the former sometimes pass
over to another plant. The plants attacked by the larvae generally
die or are greatly retarded in their development. After the 15th June
the insects decreased in numbers and by the beginning of July there
remained only single individuals, but they reappeared on the 15th August
in great numbers.”

On April 27th, 1914, the writer received from Mr E. 8. Beaven of
Warminster, Wilts, specimens of seedling barley plants in which long
strips were eaten between the veins of the leaves, very similar to the
damage described above. With these plants were sent specimens of
Phyllotreta vittula which were said to be present in large numbers and
eating holes in the young barley plants soon after they come above the
ground. The attack was not so severe as that described by Baranov.

1 Rev, App. Ent., Ser. A, Vol. m0, p. 698.

EF. R. PETHERBRIDGE 139

This year a similar attack was noticed at the Rothamsted Kxperi-
mental Station by Mr A. W. Rymer Roberts, who writes as follows:
“On one of the fields at the Rothamsted Experimental Station I noticed
during the course of last May considerable damage done by Phyllotreta
vittula to young barley of three or four inches in height. The field had
been ploughed out of ley the previous winter, and though not ordinarily
one of the experimental fields, had this year crops of various kinds in
strips. In addition to the barley, spring-sown oats on the next strip
were also afiected, though not so severely as the barley.

“The beetles usually seemed to attack the upper surface of the
leaf first, eating through the parenchyma and leaving only the epi-
dermis of the under surface. The holes so made, as you will see by
the photograph I am sending, are in general oval, running between
the ribs of the leaves, as Baranov describes; only in our case the pest
was not bad enough to render the holes confluent, nor did I notice that
plants were killed by it. ;

“The beetles disappeared about the end of the month, and I did not
notice a second brood, though I may have overlooked them owing to
pressure of other work.

“Though the damage done was not very material in permanent
injury to the crop, in view of the seriousness of the attacks on spring-
sown corn in Russia, it will be well to watch for any extension.”

It will be noticed that the attack at Warminster was earlier than
that at Rothamsted which occurred at about the same time as that of
Baranov’s. The attack in both cases occurred when the plants were
very young, and therefore at a time when the plant is most likely to
receive a severe check.

As this pest is so widely spread on the Continent, and has occurred
at such widely separated places as Rothamsted and Warminster (where
attacks would not easily be overlooked) it is possible that the damage
done by this pest in other parts of England may have escaped notice.

F. R. PETHERBRIDGE.

ScHOoL OF AGRICULTURE, CAMBRIDGE
December 1916.

Votume III APRIL, 1917 Nord

ACCESSORY WETTING SUBSTANCES WITH SPECIAL
REFERENCE TO PARAFFIN EMULSIONS.

; By A. H. LEES, M.A.,
Plant Pathologist, University of Bristol, Agricultural
and Horticultural Station.

WITHIN the last three years or so the opinion has been gradually
gaining ground that one of the most important points in a spray fluid
whether for fungicidal or insecticidal purposes is its wetting power.
The author’s attention was first called to this point when using certain
proprietory insecticides which, used according to the directions given,
were not very effective but which, on addition of soft soap, had their
wetting powers and therefore their killing powers greatly increased.
Reference was made to this subject in the Annual Report for 1913 of
the Research Station, Long Ashton!, and since that time other papers
have appeared dealing more fully with the physical problems involved
in the question’. It is not proposed here to touch this side of the
problem as the author is not a physicist but to give certain results
bearing more on the immediately practical aspects of spraying.

It has long been known that the addition of soft soap to a fluid
containing no substance capable of precipitating soap, causes an in-
crease of wetting power, No quantity of soap, however, has proved
sufficient to wet such resistant insects as the colonies of Woolly Aphis
which occur on apple trees. These insects excrete a covering of waxy
threads on their back which render them impervious to ordinary spray
fluids. The same difficulty was found in the experiments of Prof.
Barker and the author against American Gooseberry Mildew. There
was therefore a real need to find a satisfactory accessory wetting agent.

1 The Annual Report of the Agric. and Hort. Research Sta., Long Ashton, Bristol, p. 70.

2 Lefroy, ‘‘ Insecticides.” Ann. App. Biol. Jan. 1915.

3 W. F. Cooper and W. H. Nuttall, “The Theory of Wetting and the Determination of

the Wetting Power of Dipping and Spraying Fluids containing a Soap Basis.” Journ.
Agric. Sci. Vol. vu, pt. 2, Sept. 1915.

Ann. Biol. 11 10

‘

142 Accessory Wetting Substances: Paraffin Bmulsions

In this connection paraffin emulsions were tried. These emulsions
have been often recommended by many writers as insecticides and
there is no doubt that they are of value. They have, however, a
reputation of being dangerous to use owing to their liability to scorch
foliage. There are, the author believes, very definite reasons for this
hability but by suitable means it may be got rid of.

An emulsion may be stable or unstable, its state depending on its
constitution. If stable, every small paraffin globule is surrounded by
a film of soap and the film is sufficiently strong to keep one globule
from touching and coalescing with its neighbour. If very little soap
is originally present, or if where present in sufficient quantities it is
withdrawn by chemical means, there comes a point where the film is
no longer strong enough to prevent paraffin globules from touching
and fusing with their neighbours. When this happens throughout the
emulsion, paraffin appears in quantity on the surface and the emulsion
is said to be unstable.

Pickering?! has pointed out the great diversity of formulae for paraffin
emulsions that have been recommended. In a table given by him the
proportion of paraffin to soap in one formula is as great as 100: 1-2,
in others as low as 100 : 240, while there are all kinds of intermediates.
This enormous range in quantities suggests that some of these formulae
have been advocated in a somewhat haphazard spirit, since if formulae
at one end of the series were successful it does not seem necessary to
make use of others which contain such divergent quantities.

Of the 24 formulae quoted 15 contained paraffin in proportion to
soap in ratios varying from 100: 23 up to 100: 1-2. That is to say
in those mixtures of these fifteen where the paraffin bore the lowest
ratio to the soap, it was present to the extent of four times the amount
of soap and in those which bore the highest ratio, to as much as eighty-
three times. It was therefore at first sight probable that the liability
to scorching was caused by the high content of paraffin. Experiments
were therefore started to see how little paraffin need be used in order
to be lethal and at the same time to avoid scorching. It was at once
found that with the Bristol water which has a hardness of about 12°
it was impossible to obtain any kind of stable emulsion with even small
quantities of paraffin unless at least 4 °% of soft soap were used. The
paraffin simply floated on the surface of the liquid and when sprayed
on to foliage naturally produced scorching.

In this connection it is interesting to notice that out of 23 formulae

1 6th Report, Woburn Exper. Fruit Farm, 1906.

ee ee

Ae HH, Lass 145

quoted by Pickering in which the amount of soap was stated, in 7 it is
-5 % or lower. In these 7 therefore, supposing hard water had been
used, however little paraffin were added, scorching would be probable.
In the others where the paraffin content was relatively high and the
soap content relatively low there would also be a likelihood of scorching
since, as will be shown later, the amount of paraffin that can be
held in stable emulsion depends largely on the amount of soap
present.

Preliminary experiments had shown that it is possible to obtain
wetting from paraffin emulsions in two ways. If the ratio paraffin
to soap was low, wetting, if it occurred, was due to the aqueous solution,
while if the ratio was high, wetting was due to a paraffin film. It was
clearly necessary to avoid this wetting by paraffin as to it, moderate
scorching, when it occurred, was probably due.

In order to test the properties of paraffin emulsions of different
formulae with regard to stability and wetting power a series of experi-
ments was started in the laboratory. The percentages of paraffin
tried were, 0, 1, 2, 5, 10, 15, 20 and 25 and the percentages of soap },
1, 1$ and 2. Mixtures were made with both distilled and tap water.
Each emulsion contained 100 c¢.c. of water and was made by syring-
ing the mixture of soap and paraffin, without the use of heat, several
times through the “rose” of a garden syringe.

Two points were then investigated :
(a) The wetting power of the mixture,
(6) Its stability.

Wetting Power. As these experiments were done during the winter
it was necessary to select some surface other than insects or fungi on
which to test the wetting. The glossy American cloth cover of an
exercise book proved a fairly good substitute. When the surface was -
worn by the action of the liquids used it was renewed by applying a
thin film of boiled linseed oil which, when dry, gave a new brilliant
surface.

Wetting took place either by means of the aqueous solution or by
the paraffin or by a combination of both. A drop or two of the liquid
to be tested was placed on the American cloth surface and gently made
to roll. If the drop immediately spread and would form a thin stable
aqueous film it was given 3 marks for aqueous wetting. If it spread
but would not form an absolutely stable thin film it was given 2 marks,
while if it wetted slightly only it was given 1 mark. If the drop
failed to spread much and left the faintest film of paraffin behind it on

10—2

Amount of Wetting

144 Accessory Wetting Substances: Paraffin Emulsions

being made to roll it was given 1 mark for paraffin wetting, if the
paraffin mark was definite, 2 marks and if abundant, 3 marks.

In the accompanying table the continuous line represents wetting
by water and the dotted line wetting by paraffin. For the sake of
clearness, where the dotted line lies along the zero line or along the
thick line, it has been raised slightly above.

TABLE I

Tap Water Distilled Water

0 1 2 5 AOU 1S: R20 2Owo 1 2 5 10 15

at

Percentage of Paraffin Percentage of Paraffin

Ay EL Dimes 145

In considering these plotted results it is probable that as soon as
the paraffin wetting rises above zero the mixture would have a slight
tendency to scorch and a much greater tendency when wetting by
paraffin attained a higher value than. wetting by water. Considering
the results obtained with tap water first it will be noticed that with
only $ % of soap no wetting by water could be obtained and that with
increase of paraffin the wetting by it increased to a maximum. The
lower percentages therefore would be unsafe and the higher percentages
would certainly produce scorching effects.

With 1 % soap there would be a slight tendency to burn even with
only 1% of paraffin but in practice, owing to the small quantity of
paraffin present, this does not occur. With larger quantities such as
5% and over burning does actually take place. With 1} % of soap
the doubtful point is only reached with 10 % of paraffin and there would
be no serious danger until 20 and 25 % was reached. For 2 % of
soap probably all strengths would be safe, though for tender foliage
there might be some small risk above 5 % of paraffin. With distilled
water the doubtful points occur at 5 % for 4 % soap, at 15 % for1%
soap, at 20 % for 11 % soap and at 25 % for 2 % soap.

It will be noticed that the } % distilled water results are intermediate
between the 1 % and 14 % tap water results. -This indicates that in
water with a hardness of 12° (such as we have in Bristol) about 3? %
soap is used up in precipitating the metallic salts which cause hardness.
That amount of soap is therefore wasted. (This represents about
? lb. per 10 gallons and involves a loss of about one penny for that
amount of spray fluid.)

From a practical point of view the following recommendations may
be drawn from these results. Using tap water of a hardness of 12° or
using soft water the amounts of soap to be used for various percentages
of paraffin in order to obtain a safe mixture are shown in the following
table :—

TABLE IT,
Tap water Soft water
% paraffin % soap required % paraffin % soap required
10—25 2 20—25 2
2—10 13 10—20 14
1 1 5—10 1
— — 1—5 4

In cases where the hardness of water is more or less than 12° the
greater or less quantity of soap may be approximately calculated by
assuming that 12° hardness equals #°% soft soap Thus for water of

146 Accessory Wetting Substances: Paraffin Emulsions

24° hardness 24 °% soap will be required for a 10 % paraffin emulsion
while for only 6° hardness only 1) % will be required.

Stability of Emulsions. In considering stability of paraffin emulsions
it must be pointed out that none can be kept indefinitely without
de-emulsification setting in. For practical purposes it is sufficient to
study their behaviour after comparatively short periods of time follow-
ing their manufacture. If they will remain stable for one hour it is
sufficient for practical purposes and experience shows that if any decided
de-emulsification is going to set in it does so in the first hour. In the
laboratory, therefore, after the various emulsions had been tested for
wetting power they were poured, after thorough agitation, into burettes
and allowed to stand for one hour.

It was found that all those formulae containing only $°% of soap
in tap water were unstable and therefore dangerous to use. This agrees
with what might be expected from the plotted results of wetting power,
where the line indicating wetting by paraffin is always equal to or above
the line indicating wetting by aqueous solution. Similar emulsions
made with distilled water were stable even in cases (paraffin 10 %—
25 %) where, in the plotied results, the paraffin wetting line is above
the water wetting line. These emulsions therefore, though not certain
to scorch, cannot necessarily be classed as safe. With 1%, soap and
tap water the emulsions were still slhghtly unstable, as might be
expected from the plotted results where the paraffin wetting line is
nearly always above the water wetting line. These emulsions are
therefore dangerous except in those cases where the paraffin content
is low. Most plants will stand a very small quantity of free paraffin
if well distributed, since, under conditions of spraying, much of it
is evaporated before it can produce scorching effects. When however
much is present scorching is bound to occur. The other emulsions
tried, namely 1, 14 and 2 % with distilled water and 14 and 2 % with
tap water, gave stable emulsions. It will be seen that in the case of
13% soap and tap water at 20 and 25 % the paraffin wetting hne was
above the water wetting line, indicating that the emulsion was unstable.
Actually however when tested for an hour in a burette no appreciable
de-emulsified paraffin appeared. This should not be taken to indicate
that it is a safe emulsion to use. It contains a high content of paraffin
which is only just held in emulsion by the soap. In other words the soap
pellicule that separates the paraffin globules is nearly as thin as it can
be without actually parting asunder. So long as the liquid is at rest
it is stable, but when forcibly beaten against a solid object as in spraying,

A. H. Lxss 147

the soap pellicules break and liberate a certain amount of free paraffin.
Such an emulsion is therefore unsafe to use. It is thus possible to
divide emulsions into three classes:

1. Dangerous. 2. Unsafe. 3. Safe.

The first includes those which, like the $ % soap with tap water,
de-emulsify on standing. The second, those like 14% soap with tap
water and 20 or 25 % paraffin, which de-emulsify on being sprayed.
The third, those like the 2% soap emulsions, in which no de-
emulsification occurs at all.

Wetting Power of Paraffin Emulsion compared with other
Auxiliary Wetting Agents.

‘In the previous portion of this paper the wetting power of paraffin
emulsions on a certain artificial surface, American cloth, has been con-
sidered. Table III gives the results of the wetting power of some of these
emulsions and of other auxiliary wetting agents on certain natural
surfaces.

TaBLeE III. Wetting Power of Various Fluids.

American Gooseberry
Wetting Substance Gooseberry Leaf Sea-kale Leaf Mildew
Water Nearly complete None None
Gelatine 1 in 1000 Complete Very slight =
Gelatine 1 in 10,000 55 9 ”

Casein 1 in 1000
Casein | in 10,000
Soap 3 %
Soap 1%
Soap 2%
Emulsion
Soap 1%
Paraffin 1%
Emulsion
Soap 14%
Paraffin 15%
Emulsion
Soap 2%
Paraffin 2% |
Emulsion
Soap 1%
Paraffin 5 %

Emulsion
Soap 2%
Paraffin 1%

Slight
Nearly complete

9°

Complete

Wets by paraffin but
not by aqueous
solution

Complete

Very slight
Moderate
Nearly complete

Complete

Nearly complete

Moderate

148 Accessory Wetting Substances: Paraffin Eimulsions

These surfaces were chosen to represent types that could be wetted
with ease, with moderate difficulty and with great difficulty. They
are a gooseberry leaf, a sea-kale leaf, and the white felted stage of
American gooseberry mildew. The gooseberry leaf is smooth without
being waxy and is easily wetted, the leaf of sea-kale is slightly waxy
and is only wetted with moderate difficulty, while the surface of
American gooseberry mildew offers great difficulty.

Of the other wetting agents soap, gelatine and casein were tested.
Both the latter have recently been recommended, the former in an
acid or neutral medium, the latter in an alkaline. For both one part
in a thousand is relatively strong while one part in ten thousand is
relatively weak; they were therefore tested at those strengths.
Even at the weaker strength both increased the wetting power but
the increase was only slight. They caused complete wetting of the
gooseberry leaf but only raised that of the sea-kale leaf from none to
very slight while the effect on the mildew was mil. For waxy and
hairy surfaces, therefore, they are useless though they are to be re-
commended where extra spreading power is desired on an easily wetted
surface. With arsenate of lead and more especially with Bordeaux
mixture, where a cheap substance is required to obtain even spreading,
they are most efficient.

Soap and paraffin emulsions having a soap basis may be considered
together as they can only be used in mixtures where soap gives no
chemical reaction. Where soluble salts of bases other than potassium,
sodium and ammonium are present insoluble stearates are formed
which prevent the use of soap. Half a per cent. of soap proved itself
slightly superior to both gelatine and casein while the higher percentages
gave nearly complete wetting on sea-kale and a very slight wetting on
the mildew. It is clear, however, that the use of soap is limited as,
even at 2% strength it proved itself inefficient when tried on the
most difficult surface. The paraffin emulsions proved themselves much
more potent.

The strengths tried were as follows:

Paraffin Emulsion 1, 1
Tee i
” ” i 15
a. +3 2 2
3) 93 iB 5
” go” 2 1

In each case the figure first given is the percentage of soap and the

A. H. Lares 149

second that of paraffin. (Thus P.E. 2, 2 indicates Paraffin Emulsion,
2 % soap = 20 lbs. to 100 gallons of water and 2 % paraffin = 2 gallons
to 100 gallons of water.)

P.K. 1, 1 gave superior results to 2 °% of soap in that it gave better
wetting of the mildew. P.H. 14, 1} was still better while P.H. 2,2 gave
perfect wetting. P.E. 1, 5, though it nearly completely wetted the
mildew, was found to wet by means of paraffin instead of by aqueous
solution and so, though a cheaper mixture than P.H. 2, 2 would have
great danger of scorching foliage. P.H. 2, 1 stands intermediate in
wetting power between 2 % soap and P.H. 2,2 and shows the influence
of introducing emulsified paraffin on the wetting properties of the liquid.

There is no object in introducing greater quantities of either paraffin
or soap as P.K. 2, 2 gives perfect wetting. On the other hand the
behaviour of P.E. 1,5 and P.E. 2, 1 show that itis not possible to reduce
the quantities of either without destroying the desirable qualities of
the mixture. P.E. 2, 2 is, therefore, the cheapest mixture that can be
used which at the same time has.the highest wetting power. The
value of this 2 % emulsion hes not so much in its own killing power
as in the fact that it can act as a carrier, so to speak, for other fungicidal
or insecticidal bodies which, used alone, would prove themselves in-
sufficient to kill. Thus liver of sulphur, used alone, has no great
controlling effect on American gooseberry mildew but, combined with
paraffin emulsion, has given promising results in a commercial scale
experiment undertaken by Prof. Barker and myself. In the direction
of insect control it also shows promise. While dilute solutions of
nicotine are without decided action on adult caterpillars or difficultly
killed beetles such as Byturus tomentosus, the Raspberry Beetle, it has
been found possible, at any rate on the small scale, to kill these by
uniting the same nicotine solution with 2 °% paraffin emulsion.

EMPUSA MUSCAE versus MUSCA DOMESTICA L.

By H. T. GUSSOW,

Dominion Botanist, Department of Agriculture, Ottawa, Canada.

(With Plate XXI.)

THE biological control of insect pests has received considerable
attention for some years past, and success has without doubt accom-
panied research work along these lnes. The common house fly
(Musca domestica, L.) is more than ever now, because of additional
proof furnished by more recent investigations into their responsibility
as carriers of wound infections, recognised as an important source of
danger to public health.

The most destructive disease, to which the house fly is subject, 1s
that caused by the fungus Hmpusa Muscae. The question whether
this fungus may be used in the control of this pest is of decided interest,
and has received more or less careful attention from writers, as well
as from investigators.

It has already been pointed out in an earlier paper? that it is an
easy matter experimentally to spread the disease among flies in
captivity. That is to say material from flies recently dead from the
disease produced infection in others brought into contact with the
spores of Empusa. Yet, fatal as the disease is among flies, it was for
long a matter of conjecture whether it really constituted an important
factor in the death of flies. No doubt the very fact of its being fatal,
coupled with the number of dead flies to be seen every year towards
autumn, helped to attach an importance to this disease, which a close
study failed to confirm. Close observations made in places, where
flies are present by thousands throughout the season, indicate that
perhaps one in a thousand flies is killed naturally by the disease,—
in some instances even a smaller proportion. The determination of
the mortality among flies from this disease is comparatively simple,
since, as a rule, the fly killed by Empusa is and remains attached to

1 Giissow, H. T., “Lmpusa Muscae and the Extermination of the house fly.” Rep. Loc.
Govt. Board, on Publ. Health and Medical Subjects. New Series, No. 85, p. 10.

H. T. Gussow 151

the substratum where it died. In the localities referred to, many
thousands of flies frequented the places daily, but only from 1 to 4 at
the most of newly deceased flies’ bodies were observed on any one day
during a period of 22 days. Hence in nature this fungus does not
appear to account for the general disappearance of flies in winter.

Though I am quite satisfied that one cannot speak of this disease
as the cause of an epidemic among flies, and though, endemically, it
may be present every season, the fact that it is so fatal to individual
flies made it desirable to experiment with the fungus with a view to
spreading the disease more widely and rapidly. This might be done
were it possible to grow the fungus conveniently in artificial culture,
and use it in the manner of, for instance, D’Herelle’s Coccobacillus
acridiorum, pathogenic to grasshoppers.

Methods of securing artificial cultures have been tried by a number
of observers, and because of some recent publications on the subject,
which, perhaps, have not come to the attention of scientific workers,
it will not be out of place to discuss the opinion of these writers here.
In the last note which I prepared (loc. cit.), I briefly dealt with the
investigations of Mr Edgar Hesse, on the subject of culture of Empusa.

The seventh report of the same Board (No. 102, 1914) contains
further contributions to the subject under discussion. In the letter
of transmittal Dr Newsholme, the medical officer of the Board, briefly
reviews the original contributions in the report. We now take excep-
tion to the statement that the Hmpusa “is known to be the cause of
much of the mortality among the house flies during the autumn months,”
because of careful observations on the rate of mortality. Notwith-
standing my friendly criticism of Mr Hesse’s 1912 work in the sixth
report, his work receives much support from Dr Bernstein who
was appointed by the Board “to supervise and control Mr Hesse’s
investigations.” It must be pointed out here that Mr Hesse claimed
and “proved” (!) “the possibility of obtaining from spores of #. Muscae
derived from flies which had died of the disease, a culture which, under
suitable conditions as to moisture and food material, gave a luxuriant
growth in a few days, followed by the production of spores. These
spores when mixed with syrup and fed to normal flies, induced in them
a rapidly fatal disease, which, as judged by the naked eye and by
microscopical examination, was indistinguishable from the disease
found in the flies attacked by Hmpusa Muscae in the ordinary course.”
Mr Hesse, be it noted, refers to luxuriant growth and production of
spores, but does not here state what growth or spores.

152. Empusa Muscae versus Musca Domestica L.

When, through the courtesy of Dr. C. Gordon Hewitt, some original
material of Mr Hesse’s culture was placed at my disposal, I could,
on examination, find nothing else but Mucor racemosus, which certainly
produced after a few days a luxuriant growth of the same fungus: we
also observed “spores,” but not Hmpusa spores. Since that time
experiments have been made with spores of this universal mould fungus
on living flies, but, notwithstanding an ample supply quite sufficient
to fatten them, they died of starvation; but under no circumstances
was death due to Mucor racemosus, since sound flies, kept under iden-
tical conditions minus the fungus, died about the same time from the
effects of confinement. Mr Hesse solved the difficulty apparent in
this instance by concluding that this fungus was polymorphic, and
“that Empusa was merely a parasitic form of the Mucor.” Mr Hesse
supplements this statement by the following paragraph :—

“The idea that the Hmpusa Muscae might be polymorphic now
presented itself, and also that this might be the reason that its culti-
vation had not apparently been successful hitherto. I realised that
one should not expect to produce a purely parasitic growth on a dead
medium (szc/), but that, as in the case of other fungi when artificially
cultivated, a morphological change might be expected; and came to
the conclusion that Mucor racemosus, although a saprophyte, might,
when the spore is ingested by the fly, assume a parasitic existence
and produce the growth, in the tissues of the fly, known as Empusa
Muscae, whose spores, when cultivated on artificial media, revert to
the saprophytic stage.”

It is regrettable to notice the author’s attempt to interpret an error
of his by proposing a somewhat fantastic theory. Surely before such
explanation is necessary, he should be sure that what he recorded
as occurring did actually occur, viz.: the growth of Mucor from ger-
minating Hmpusa spores. It is so clear that it hardly needs stating
that he has confused an adulteration of his “cultures” with what he
thought he had produced. The use of an egg. yolk medium “in thin
layers in a Petri dish, sterilised by steam at 100° C. for one hour on three
days in succession,” clearly indicates the degree of importance attaching
to the author’s further results, whatever they may be.

He proceeds then by inserting bodies of flies which had died of
Empusa during the previous autumn. Control dishes were also pre-
pared. In five days sporangia had formed on the dishes containing
the egg medium, whilst the control dishes remained quite sterile. The
author, unfortunately, does not indicate whether he placed on his

>) See ee SB eS el

‘
1
:

H. T. Gussow 155

control dishes dead flies, not infected with Hmpusa, but at any rate
collected at a similar date (the previous autumn), which precaution
one would naturally expect in controlling the results of the first experi-
ment. It appears that he placed no flies whatsoever on his control
dishes, which naturally remained sterile. Had he done so, we are
prepared to predict the identical formation of sporangia observed in
his experiments with Hmpusa killed flies.

Next he made transfers from his culture, sporangia forming in the
secondary culture. “Meanwhile,” he states, “the fungus of his
secondary culture, which in every respect was identical with Mucor
racemosus, was transferred to a syrup made of sterilised sugar and
water, and put aside for experiments.” How long he does not say.
Nor whether the spores had time to germinate in the sugar solution in
the “‘meantime.”’

Now comes the most surprising effect! In his experiments Mr
Hesse uses “flies all bred from insects in confinement”; in June “the
chrysalides of the greater and lesser flies commenced hatching out”
(2? Musca domestica L. the common house fly, and Fannia canicularis, L.
the lesser house fly). In his first experiment all negative results were
secured, owing apparently to absence of plum jam. Notwithstanding
an earlier statement by the author that many insects became entangled
in the sticky mess of plum jam, in the second experiment this was used
together with, though not specifically so stated, a filter paper saturated,
as in the first experiment, with a syrup in which had been incorporated
spores (we take it Mucor spores). By the fourteenth day the first
deaths were noticed; by the twenty-first day all flies were dead. “All
presented typical outward (sic) appearances of infection with Empusa
Muscae.” No control was made in this case, he states.

In. experiment three,—a repetition of No. 2,—the first deaths
occurred in four days, and by the eleventh day all were dead. Cause
of death is not stated.

In the fourth experiment there was a control with flies not fed
with “spores,” and some fed with spores from the fourth generation
of Mucor. “The results were the same” (as which?). In the control
“no deaths occurred and the flies appeared to be normal four weeks
later.” This is most remarkable; and, if Mr Hesse had not reported
this as an actual observation, I should feel inclined to challenge the
accuracy of these observations.

Cultures were then made by this indefatigable worker from “flies
which had died in the cages with the naked eye appearance of infection

154. Empusa Muscae versus Musca Domestica ZL.

with Hmpusa Muscae.” These resulted in a prolific growth of Mucor
racemosus. Why, I wonder, the cautious expressions of “outward”
and “naked eye” appearances? One must instinctively ask:—Was
Empusa Muscae present, or did it only appear to be present? A state-
ment of this most important fact is carefully avoided. Why?

Dr Bernstein thus adds his opinion: “There can be no doubt than
that Mr Hesse has succeeded in producing a growth, ingestion of the
spores of which results in the death of all the flies from Empusa Muscae.”
One cannot but feel greatly surprised at such a statement from the nature
of the research work quoted.

After consultation with Dr Copeman and Mr Ramsbottom (of the
Nat. History Museum) the investigators, Dr Bernstein states, agreed
on this fact. Here I should wish to ask:—Did they agree on the fact
that the flies dead from Empusa Muscae were the flies used, or did they
merely agree that all the flies, presumably submitted for examination,
were dead from Hmpusa Muscae ?

Dr Bernstein now steps in the breach and tries a number of experi-
ments. Cage 1 was not interfered with; Cage 2 received a paper
saturated with sterilised syrup of cane sugar; Cage 3 received in addi-
tion a paper saturated with syrup containing the spores of Mucor
hiemalis supplied by Mr Ramsbottom. Cage 4 contained a paper
saturated with syrup containing spores of Mucor racemosus, “which
had been cultivated on egg yolk from EHmpusa Muscae (sic). The
cages were kept in a warm room (in whose charge, or under whose
observation, we would gladly have seen stated). Abnormal changes
only occurred in Cage 4. In fourteen days 75 % of the flies were dead.
Mr Ramsbottom, examining one of the flies from this cage, expresses
his opinion that the manifestations were identical with those of Empusa
Muscae. No fly died in the other cages from Hnpusa.

The general conclusions are stated as follows:—“It would seem
then that there could be no doubt that the deaths of the flies in Cage 4
were due to a fungus indistinguishable from Mucor racemosus, but
which can be readily cultivated in great quantities from the bodies of
flies dead of Empusa Muscae.” This is more correct; we note “from
the bodies of flies dead of Empusa”; what role did the Empusa play
therein ?

Empusa Muscae spores under no circumstances have produced in
the cultures made by myself and many other workers anything else
but what belonged exclusively to that fungus; certainly never anything
like Mucor racemosus. Mucor racemosus, or Mucor resembling what

— | seer

re ee a

H. T. Gussow 155

may now be somewhat loosely called Mucor racemosus, was readily
secured from the bodies of any dead fly. We cannot, of course, doubt
the correctness of the recorded observations; but we are of the opinion
that if a Mucor proved fatal to flies in the manner described in these
experiments, it must have been one of the pathogenic types or a different
species altogether. We expect close diagnostic studies of the patho-
genic organism would soon establish the identity. But we do absolutely
question that Mucor developes from an uncontaminated Hmpusa spore.
We are glad to note that Mr Ramsbottom has now become interested
in clearing up this somewhat involved research work; and we quite
agree with him that—if Mr Hesse’s assumption is correct that Hmpusa
is a polymorph—a fundamental biological principle would be absolutely
overthrown.

The critic, they say, but assumes the rdle when he has failed in
producing good or marketable merchandise in his own line. That I have
assumed the critic’s réle, I may not deny: in what follows, I would
play a constructive part, with what success I modestly leave in turn
to my critics to say.

I wish, therefore, in concluding to record some of my cultural ex-
periences and other observations on Hmpusa Muscae, which I have
not hitherto published. This may assist in simplifying the apparent
difficulties which Mr Hesse had in securing an uncontaminated growth
from Hmpusa spores.

In my last paper (loc. cit.) I described briefly a method of successful
auto-infection of flies coming into contact with flies recently dead from
Empusa. By this means one has under one’s own control the securing
of fresh material of the fungus for a considerable period. By placing
a fly showing the fresh belts of fungus growth on a pillar made of plas-
ticine, this latter can be bent and twisted under the microscope into
any position. When carefully adjusted one can observe the discharge
of spores very readily. These fly in every direction and for distances

’ of about seven centimetres.

Hence this fact was used for obtaining spore cultures quite easily,
and uncontaminated. A series of clear coverglasses was placed at the
bottom of a sterile jar, which was closed with an ordinary cork. A fly
which was observed to be surely “infected’’—experience having shown
the incipient symptoms of infection, viz. sluggishness, increase in volume
and lightening in colour of abdomen, and a peculiar reddish colour,
a dirty brick red, of the “eyes”’—was pinned on to the cork from below,
so that, when the stopper was replaced, the fly was securely held above

156 Empusa Muscae versus Musca Domestica L.

the coverglasses underneath. Spores were freely shed and collected
absolutely pure.

The usual methods used in spore cultures were then employed,
but the results were not satisfactory. I then used for a medium various
animal fats—a droplet of butter, of dripping, of lard, etc. These were
exposed to the spore bombardment, but no more appreciable results
were obtained. Moreover, these substances were found to be very
unsuitable for microscopic use owing to their annoying refractive
indices and their opacity. How much more unsatisfactory must the
egg yolk medium—stone hard, no doubt, from the method of treatment
given by Mr Hesse—have proved itself. I also tried sterile asparagus
tips—mainly because of the known lecithin contents of this plant—
but no satisfactory results were achieved on it either.

I then obtained interesting results from the use of a mineral fat,
i.e. common vaseline, as used for ringing slides in hanging drops. This
medium was sterile to begin with and made beautiful clear hanging
drops or surfaces. A small quantity was placed with a scalpel on a clean
coverglass and gently heated until evenly spread like a film over the
coverglass. When exposed to spore-shedding, the number of spores
desired thereon was easily adjusted by the removal of coverglasses
after one, two, three or more minutes’ exposure. The coverglasses
were then inverted on hollow ground slides and ready for examination.
» To obviate any misunderstanding, here let me remark parentheti-
eally that, in the description following, I am not detailing the result
of any one particular experiment, but am summarising the nett results
of some hundreds of observations.

Germination began very shortly in every instance. The plasmatic
mass which surrounds the spore when shot from the conidiophore
appeared to remain in a liquid or semi-liquid form; it never showed
the wrinkled appearance generally observed, and so often figured. The
first symptoms of active life appeared about one hour after the spores
had been shot on the medium. Never at any time did the original
spore produce a germinal tube; at first there appeared something
very much like a germinal tube in the shape of a small “exdentation”
of the cell wall of the mother spore. This gradually elongated, but
assumed, from ten to fifteen minutes later, a flask or clubshaped appear-
ance. About an hour later it was recognised with certainty as a second-
ary spore, the nucleus of the first spore having taken up its position
in the now forming secondary spore. As this spore grew, it, together
with the mother spore, presented a dumb-bell shape. Where joined

H. T. Gitissow 157

to the mother spore the “handle of the dumb-bell” elongated somewhat,
and the secondary spore also appeared surrounded by a_plasmatic
substance—just as did the mother spore—but in very much smaller
quantity.

Immediately after the secondary spore was formed, it actually
did germinate. It should have been said that in the mother cell the
“exdentation”’ occurred on almost any position of the spherical body.
But with the secondary spore, germination, as opposed to the “exden-
tation” of the mother cell, took place at the end towards the mother
cell. Indeed what appeared at first as a rather long “handle to the
dumb-bell” was later recognised as a mycelial thread, or, more correctly,
tube growing into the cavity of the old spore. Naturally in many cases
the secondary spore often separated from the mother spore.

The germ tube was seen to branch, and the contents of the spore
to be slowly used up, while the mycelium grew in size. This produced
sparse, short, stout branches at no special points. Later on a second
germ tube grew from the secondary spore body; on rare occasions
there were three in all. The growth made slow progress thereafter,
becoming detached from its spore shell, and assuming shapes and sizes
similar to the mycelial portion observed in the fat body of the fly. We
have not been able to see a true tertiary spore in Hmpusa Muscae after
germination took place, although peculiar club or flask shapes occasion-
ally appeared which resembled spores; but they could not, on examina-
tion, be determined as such.

The growth remained pure all the time and made progress for
28 days on this medium. Then signs of disintegration appeared, and
the mycelium became vacuoled more and more, less sharp in outline,
and later collapsed. This is, so far as I know, the longest time a
growth has been maintained outside a fly; but we cannot regard it
as a successful culture yet. Every effort to continue growth failed,
no doubt because it normally takes place in nature in the living fly
body.

These observations clearly show that Mr Hesse’s Mucor racemosus
was nothing else but an impurity, and polymorphism does not occur
in this fungus, as represented at one time by Mucor and another by

?

Empusa.

Every other experiment to continue the development of these
spores has failed so far. The absence of nutrients accounts for this;
for the various ingredients tried to furnish material for continued
growth did not suit the fungus. The main points, from a scientific

Ann. Biol. tt 11

158 KEmpusa Muscae versus Musca Domestica ZL,

point of view, viz. the hibernation, the question of resting spores,
so common in other species of this genus, and of artificial culture, —
still remain unsolved. But, while it is realised as possible that the
solution of these problems may have other than a scientific value, it
appears to me from my observation that we must look, for the control
of the fly problem, to other biological organisms—or remove systemati-
cally, by all necessary precautions, the insanitary conditions favourable
to the breeding of this annoying and dangerous pest.

EXPLANATION OF PLATE XxXl.

Fig. 1. Empusa spore shot on vaseline at 9.15 a.m.
Fig. 2. Same spore at 11.15 a.m.

Fig. 3. 2 BS 12.10 p.m.

Fig. 4. a5 PS 12.25 p.m.

Figs. 5, 6. Same spore, a little later.

Fig. 7. Same spore at 4.40 p.m.

Fig. 8. - 5 5.20 p.m.

Fig. 9. S A 5.30 p.m.

Fig. 10. a x3 6.5 p.m.

Fig. 11. Spore shot on glass slide, mass of protoplasm drying up.

Fig. 12. Same spore as before after 24 hours’ germination.

Figs. 13-17. “‘Involution forms.”

Figs. 18-20. Spore germinating into mother spore, and becoming separated from same.
Fig. 21. Proboscis of Musca domestica and spore of Empusa Muscae, relative sizes.

PLATE XxI

VOL. Ill, NO. 4

THE ANNALS OF APPLIED BIOLOGY.

j
!

H.1.G, del. ad nat.

a A

A BLOSSOM WILT AND CANKER OF
APPLE TREES.

By H. WORMALD, M.Sc. (Lonp.), A.R.C.Sc.

(Mycological Department, South-Eastern Agricultural College,
Wye, Kent.)

(With Plates XXII—XXIV.)

CONTENTS.

PAGE
I. Introduction c : : : : 3 : : : 159
II. Historical ¢ : : : : : 5 3 : é : 162
Ill. The Disease as observed on naturally infected trees . ; : 165
(a) Observations made in 1915 é : : 2 ; : 165
(6) Observations made in 1916 : , - : : : 168

(c) Comparison with other diseases producing a similar con-
dition . c : : : : : 3 : C 172

IV. The Blossom Wilt Fungus compared with other Monilias of Fruit
Trees : : - : , . 4 é 3 173
(a) Cultural Studies of Monilias found in this country : 173
(b) Dimensions of Conidia . : ¢ ; = : : 177
(c) American Strains of Monilia . - - : : : 180
(d) Continental Strains of Monilia ‘ : : : : 181
(e) Nomenclature of the various forms : : 2 : 183
V. Inoculation Experiments with Pure Cultures : : : - 185
(a) Inoculation of apple flowers in the greenhouse - ; 186
(6) Inoculation of apple flowers in the plantation : : 192
(c) Inoculation of twigs through wounds : : “ C 195
(d) Inoculation of the fruit . - : : : ‘ : 195
VI. Control Measures . : - : : ; - : : - 198

I. INTRODUCTION.

DuRING recent years many varieties of apples have been seriously
attacked by a disease which causes the wilting and death of the blossom,
frequently kills the twigs, and may produce cankers on the branches.
From information received at Wye College from various localities in

11—2

160) A Blossom Wilt and Canker of Apple Trees

Kent and Sussex it is evident that the disease is increasing in intensity
year by year; in some orchards and plantations it has now assumed
epidemic proportions and is causing considerable loss to the fruit
farmers.

The varieties which have suffered most are Lord Derby, Cox’s
Orange Pippin and James Grieve. A Sussex fruit grower! who supplied
the writer with specimens gave the following list of varieties affected
with the disease in one of his orchards in 1915: Duchess of Oldenburg,
Worcester Pearmain, Allington Pippin, Cox’s Orange Pippin (this
variety very badly attacked), Early Julyan, James Grieve, Lane’s
Prince Albert, Lord Grosvenor, Prince Bismarck, Chelmsford Wonder,
Newton Wonder, Domino and Lord Derby. In the same orchard the
varieties which the owner found to be free, or practically so, from the
disease were Charles Ross, Gladstone, Beauty of Bath, Lady Sudeley,
Blenheim Orange, Royal Jubilee, Bramley’s Seedling, Warner’s King
and Queen. Whether any of the varieties included in the latter list
are immune is at present an open question but some of them are known
to be susceptible, for Beauty of Bath, Warner’s King, and Bramley’s
Seedling have been found elsewhere with the disease, as have also
Duchess’ Favourite, Ecklinville Seedling, Fearn’s Pippin, Dartmouth
Crab and Rival. The same grower writing in May 1915 said, the
disease “is more extensive among apples than I have ever seen it before.
I reckon that in some varieties one-fourth of the trusses of fruit have
been ruined by it”; with reference to a similar outbreak in 1916 he
wrote, “The specimens enclosed are from Fearn’s Pippin which is very
badly damaged. One fairly large tree blossomed all over profusely
and I am sure that nineteen out of twenty of the trusses of fruit blossom
have gone off like those enclosed,” and in another letter he averred that
“Three-fourths of my anticipated crops of Cox and Domino, and half
that of some other varieties have been destroyed by the disease.”

Inspection of affected orchards and plantations in Kent during the
past season (1916) has shown that the experience of the Sussex grower
is by no means unique, for trees with from 50 % to 75 °% of the flowering
spurs killed by the disease were not uncommon. On one farm visited
by the writer there were thirty large standard trees, about twenty years
old, of the Lord Derby variety which had produced about 300 bushels
of fruit in 1914 when a little of the disease was noticed; in 1915 the
disease was worse, while last season (1916) the crop was practically nil
and hundreds of dead spurs recently killed were to be seen on each

1 T am indebted to Mr Wm E, Bear, Hailsham, for the information supplied,

Se

os

H. WorMALD 161

tree. The affected trees were in two rows; along one side of them was
a row of the variety Warner’s King bearing a few diseased trusses while
along the other side the trees (Newton Wonder) were apparently quite
free from the disease. This would appear to indicate that the Lord
Derby variety is particularly susceptible to the disease and this con-
clusion is supported by the fact that in one large orchard where there
were some hundreds of trees of this variety together with several other
varieties it was possible to detect the Lord Derby trees even in winter
by reason of the large number of dead spurs and twigs which they bore,
In the Weald of Kent where this variety is extensively cultivated so
much injury has been caused by the disease that it has been necessary
in some cases to “top-graft” the trees with a more resistant variety.

Not only are well-established trees attacked, but quite young trees
too are susceptible to the disease. In one case, observed in the fruit
plantation at Wye College, a voung cordon apple tree (of the variety
Rival) was attacked through a fruiting spur situated near the middle
of the stem during the first season after it was planted out; a canker
developed round the base of the affected spur and killed the upper half
of the tree.

The first symptoms of the disease are noticeable about a fortnight
after the tree comes into flower; it will then be observed that some
“fruiting spurs” of the trees affected not only fail to set fruit but the
flowers and leaves round the base of the inflorescence show evidence of
wilting, and, within a few days become dry and withered. Where such
a truss is borne on a short spur there will be found about the middle of
June a depressed, often cracked, canker-like area on the branch, around
the base of the spur. In some instances the canker within a few weeks
girdles the branch and so causes the death of that portion distal to the
canker.

Usually no external sign of any parasitic organism is to be found on
these newly killed spurs at this stage. During periods of wet weather
however the dead flowers and pedicels may produce pustules bearing
chains of elliptical to citriform conidia which are readily identified as
of the Monilia type. Again if sections are made through the base of
a dead truss, showing externally no trace of the fungus, and examined
microscopically, hyphae are invariably to be found. When particles
of these sections are placed on nutrient culture media the fungus con-
tinues to develop and may be induced to produce conidia when growing
in pure culture; under these conditions too the organism is recognised
as a Monilia.

162. A Blossom Wilt and Canker of Apple Trees

If the dead spurs are allowed to remain on the trees until the following
season the fungus appears at the surface of the spurs and over the
cankered areas during the winter and spring in the form of rounded
Monilia pustules which burst through the bark and produce numerous
chains of conidia (Fig. 1).

The constant association of the Monilia with the Blossom Wilt and
the appearance of that fungus on the diseased spurs and cankers
obviously suggest that the.organism is responsible for the damage
done. The Monilia that is generally assumed by plant pathologists,
in this country and abroad, to attack the apple is Monilia fructigena
Pers. (= Sclerotinia fructigena Schroéter). In its conidial form this is
the fungus that so readily attacks the ripening apple during the late
summer and often causes them to become “mummified.” On such
apples the fungus is usually to be seen in the form of yellow pustules
which appear more or less in concentric circles over the diseased fruit.
The form which I have hitherto always found to be associated with the
Blossom Wilt was found to be quite distinct from the typical Monilia
fructigena found on the fruit in summer and autumn, in the colour and
size of its pustules, in the dimensions of the conidia and in its habit
and growth on artificially prepared culture media. The pustules of
the Blossom Wilt fungus are in general smaller than in WM. fructigena,
are grey rather than yellow in colour and the conidia they bear are
smaller. This grey Monilia of the apple conforms more nearly to
descriptions of Monilia cinerea Bon. (= Sclerotinia cinerea Schroter)
which, according to those continental mycologists who recognise this
as a species distinct from M. fructigena, is the form responsible for the
majority of cases of Brown Rot in the “stone-fruit” trees (i.e. plum,
cherry, damson etc.).

II. Historica.

In 1888 Sorauer (21) pointed out that not only was Monzha destructive
to the apples themselves but that it could invade the woody tissue of
the twigs which were in consequence killed towards their tips, and since
that year frequent references to this “ Zweigdiirre” have appeared in
continental phytopathological literature. This condition on apple trees
is generally attributed to MW. fructigena, while a similar disease on cherry
trees is said to be caused by MW. cinerea by those who recognise the two
as distinct species.

Frank and Kriiger(i0) in 1899 make reference to an outbreak of
Brown Rot on apple trees in the neighbourhood of Berlin; on the

H. WorMALD 163

affected trees fruiting spurs were killed soon after the flowers opened,
the flowers and leaves became withered and the fungus penetrated into
the bark of the branches.

One of the most valuable contributions to our knowledge of the
Brown Rot Monilias is a paper by Woronin(23) published in 1899; in
it he not only supplied evidence in favour of his contention that Sclero-
tinia fructigena and S. cinerea were distinct species and could be dis-
tinguished even in the Monilia (or conidial) condition, but also used the
two forms in his inoculation experiments using, apparently, pure strains!
of these fungi. With regard to his results when the flowers of the apple
were inoculated, he found that the conidia of S. cinerea germinated on
the stigmas and attacked the styles shghtly hut the germ-tubes were
unable to penetrate further into the flower. The germ tubes of the
conidia of S. fructigena on the other hand grew out into all parts of the
flower and then into the pedicels and leaves, the latter gradually be-
coming withered. From these results he states emphatically, “Meine
Experimente beweisen ausserdem, dass die Laubdiirre bei den Aepfeln
nur durch Sclerotinia fructigena verursacht werden kann.”

In 1900 Aderhold (1) records having received specimens of apple twigs
which, from their appearance, he believed had been destroyed by “ fire-
blight,” the bacterial disease which ravages pear and apple trees in
America. The affected shoots contained a white mycelium which when
cultivated in hanging drops produced a Penicilliwm-like fructification?
which he considered was the conidial form of a Mollisia. Later how-
ever (2) having found that shoots killed during the previous year bore
Monilia pustules he attributed their death to the action of M. cinerea
and concluded that the form previously observed was of secondary origin.

Miiller-Thurgau(5) in the same year describes a similar disease of
apples and pear trees but refers the causal organism to M. fructigena.

Aderhold and Ruhland(4) in 1905 confirmed the work of Woronin
in that they found that inoculation of apple flowers with Sclerotinia
cinerea led to a weak infection while Sclerotinia fructigena readily pro-
duced the death of the blossoms.

Eriksson (8) in 1913 figured and described a canker-like disease of

1 Woronin is not quite clear on this point; on p. 18 he writes of “zahlreiche, reine,
streng controlirte und méglichst variirte Impfversuche mit diesen beiden Pilzen,” but
whether his “reine Impfversuche’? were made from cultures growing on previously
sterilised media is not stated.

* From the description and figures it would appear that these Penicillium-like fructifi-
cations consisted of “sporidia”’ or “microconidia”’ of the Monilia, as the production of
these bodies is readily induced in artificially prepared cultures of the Brown Rot fungi.

164 A Blossom Wilt and Canker of Apple Trees

apple trees which, from the description, appears to be identical with
that observed in this country. He found that the Monilia pustules
which developed on the affected parts were grey in colour but states
that he makes no distinction between Monilia cinerea and M. fructigena,
and assumes that the form which appears on the fruit in autumn is a
stage in the annual cycle of that form which kills the twigs and blossoms.
Broz(6) in 1913 describes the occurrence of the “Zweigdiirre” at
Vienna and states that late frosts favour the outbreak of the disease.

Brown Rot is also known in America where it is particularly de-

structive to peach trees, killing blossom, fruit and branches. The
American form, though it has been generally! named Sclerotinia fructi-
gena, is found by the more recent investigations of Matheny (14), Jehle (12),
Conel(7), and Bartram(5) to conform more nearly to the descriptions
given of S. cinerea, and is referred by them to that species.

In our own country it has been customary to refer all cases of Brown
Rot of fruit trees to Monilia fructigena, and although the Monilias are
probably responsible for greater losses to the English fruit growers
than any other genus of fungi (since in one form or another it attacks
the blossom, young fruit, ripening fruit, fruit in the store, twigs and
branches), no attempt appears to have been made until quite recently
to determine whether the conclusions arrived at by Continental workers
hold good for Britain.

In 1903 Mr G. Massee observes(13) that Monilia fructigena “is most
frequently seen on apples and although best known to the casual observer
on the fruit occurs also on the young shoots, leaves and even the flowers.”
That Monilia is capable not only of killing the flowering spurs but may
also produce large cankers on the main limbs of apple trees was recorded
by Mr Salmon (16) in 1910, Since that year many specimens of Blossom
Wilt and Brown Rot canker have been sent to Wye College, and it was
evident that the disease was becoming a serious menace to the cultiva-
tion of some varieties of apples in certain districts, particularly in Kent.

Mr Salmon has, on several occasions”, pointed out the serious nature
of this form of Brown Rot; since preliminary examination showed that
the results of investigations made abroad were not wholly in accordance
with observations made on the disease as affecting apple trees in this
country it was evident that it was a subject demanding further research,
and as Mr Salmon himself, from pressure of other duties, was unable to
continue the work, it was entrusted to the present writer.

1 Vide Duggar’s Fungous Diseases of Plants, footnote on page 187.
2 See Bibliography on p. 203.

H. WorMALD 165

The investigation is still in progress but it was thought desirable
that the facts already ascertained should be published in order that
steps might be taken to check the further spread of the disease.

III. Tuer DISEASE AS OBSERVED ON NATURALLY INFECTED TREES.

(a) Observations made in 1915.

An opportunity for studying the disease under conditions favourable
for close examination occurred in 1915 when a row of apple trees in the
fruit plantation at Wye College was found to be attacked by the Blossom
Wilt. The trees, forty-eight in number, were of the varieties Warner’s
King and Duchess’ Favourite planted alternately and were “closely
spur-pruned” bush trees about eight feet high. Detailed observation
commenced early in June 1915 and the following facts were noticed:

Not one of the trees was entirely free from Blossom Wilt though two
of them had but one dead truss each, while the tree which had suffered
most had 132 wilted trusses of blossom or about one-third of the total
number present. The affected trusses (inflorescences) were recognised
by their brown and withered drooping flowers and leaves; they were
often greyish since the edges of the withered leaves showed a tendency
to curl inward thus exposing the hairy under surface of the leaves.
That some disintegration of the tissues had occurred was evidenced by
taking a spur between the thumb and finger, immediately below the
insertion of the inflorescence, when it was found that, when pressure was
applied, the diseased spurs were more easily compressed than was the
case with healthy spurs, and wilted trusses were easily broken off at the
lower limit of the year’s growth. The short axis of the inflorescence
was at this time quite dead and when cut across was brown throughout ;
microscopic examination showed the presence of mycelium in the cortex,
xylem and pith, the last sometimes being almost replaced by interwoven
hyphae. The discoloration of the tissues extended to the older parts of
the spur particularly in the cortical region and frequently reached the
branch bearing the spur.

Some of these trusses, shortly after wilting, produced Monilia
pustules on the dead flowers during a period of wet weather but in the
majority the mycelium within the tissues apparently remained sterile
throughout the summer. When however some of the latter were broken
off from their spurs (i.e. at the lower limit of that year’s growth) and
placed on damp filter paper in a large petri dish, pustules of conidia
readily appeared on the exposed broken end and on the withered flowers ;

166 A Blossom Wilt and Canker of Apple Trees

in some cases well-developed chains of conidia were produced within
twenty-four hours after placing the dead trusses in the moist chamber.
Pustular outgrowths were observed on some of the recently produced
(from infection in 1915) cankers in August but they were barren and no
conidia were found on them before December.

Old dead spurs bearing conspicuous pulverulent pustules of conidia
during the summer months were present on some of the trees in the row,
and in those cases where cankers had been caused these too had the
conidial tufts scattered over the cankered surface, particularly towards
the periphery. These in all probability were the result of infection
during the previous flowering season of 1914. That these dead spurs
and cankers were the source of infection resulting in the blossom wilt
of 1915 and that the fungus on them was the cause of the injury was
suggested by the fact that the newly killed trusses were most numerous
on those trees bearing the greatest number of old spurs and cankers and
were most frequently met with in their vicinity. Trusses within a short
distance below such spurs and cankers were particularly liable to
infection; thus on one branch of a Duchess’ Favourite tree a portion
one foot in length bore nine dead trusses, three above and six below a
dead spur with its accompanying canker on which were numerous
conidial fructifications of the Monilia—three of the dead trusses were
on the spurs immediately below the canker. (This canker is shown in
Fig. 2.)

The relation between the number of dead spurs and the number of
wilted trusses per tree showed, although the disease may spread from
affected trees to others in the neighbourhood, it may be stated generally,
particularly if the trees are well separated from each other, that the
wilted trusses are infected from spurs or cankers on the same tree.

The forty-eight trees in the row had in all seventy dead spurs (some
with cankers) bearing the fungus; the number of dead trusses was 1319,
and since there was no other source of infection in the immediate
vicinity it must be assumed that the wilted trusses of these trees were
infected from the seventy dead spurs; that is to say that in that year
each old spur was responsible on the average for the death of nineteen
trusses of blossom. A search in the neighbourhood of a group of closely
ageregated dead trusses almost invariably revealed the presence of one
or more dead spurs bearing the fungus.

Infection occurred through the flowers and not through the leaves
as shown by the fact that primarily the flowering trusses only are
attacked. On one of the trees (variety Warner’s King) was a branch

H. WorMALD 167

with a single dead spur, near the upper end, bearing a number of
powdery pustules; all the flowering spurs, fifteen in number, on that
branch below the dead spur were killed, while all the barren spurs (i.e.
bearing leaves only), eleven in number, were unaffected. Barren spurs
and young vegetative shoots were not attacked directly but became
wilted when borne on that portion of a branch distal to a canker which
had encircled the branch, or when borne on a spur which also carried
one or more infected trusses.

By the middle of June it was found that where diseased trusses were
borne singly on short, simple (unbranched) spurs, 0-5 em. to 1-0 em. in
length, growing from the younger portions (5-7 mm. in diameter) of
the branches the fungus had traversed the tissues of the spurs and
penetrated to the branch itself producing a canker completely encircling
it and causing the wilting and death of those parts beyond the canker.
On older portions (to about 1-5 em. in diameter), bearing simple spurs
to 2 em. in length, the canker at the base of each affected spur had
reached about half way round the branch.

A similar condition obtained in the case of spurs killed during the
previous season, and it may be stated generally that when infection had
occurred through a spur to 3 em. in length a well-defined canker was
produced extending upwards and downwards along the branch for
several centimetres; the lateral extension of the canker was less than
the longitudinal but on those portions of the branches two or three years
old it was sufficient to completely girdle them. When the spurs were
longer they were often killed to the level of their insertion on the branch,
but on those 5 or 6 em. in length the fungus as a rule had failed to reach
the branch and the lower portions of such spurs were still alive.

The fungus may grow downwards from the infected truss along one
side of the spur while the other side remains alive and unattacked.
One spur had pustules over an area extending two-thirds round it but
at the apex there were two young growing apples. On another the
pustules had developed over the terminal portion and along the whole
of one side, the other side being still alive for about half its length as
shown by a living shoot near the middle; a section across this spur
showed the tissues on the one side to be brown as far as the pith, while
on the other they were still green.

A few old cankers were found from which the layers of dead tissue
were in process of being removed by the formation beneath them of
callus which was growing over and healing the cankered surface. These
cankers at this stage showed no trace of the fungus but they bore such

168 <A Blossom Wilt and Canker of Apple Trees

a general resemblance to the Monilia cankers that it seemed probable
that they had been caused by the fungus, whose further development
had been arrested. The condition of such cankers together with
observations made subsequently on younger cankers suggested that
the fungus did not survive after it had produced its crop of conidia.
In support of this it was found that the formation of callus had already
commenced beneath the bark of cankers still bearing the pustules of
the fungus (Fig. 8).

The observations made in June 1915 and recorded above suggested
the following hypotheses:

(1) Infection arises from conidia falling upon the open flowers.

(2) The conidia are produced on spurs killed during the previous
flowering season and on cankers arising round the base of the dead
spurs.

(3) When the fungus has produced its crop of conidia on a spur or
canker its further development ceases and in the case of cankers the
injured surface becomes healed over by a growth of callus.

That the fungus was a true parasite and able to produce both the
Blossom Wilt and Cankers was definitely proved in 1916 by means of
inoculation experiments with pure cultures as is shown later in this
paper. The other two points were established by attaching numbered
labels to affected spurs and cankers, and recording observations at
certain periods. After a few weeks this revealed another interesting
feature relative to the cankers. It was found that not only did the
older cankers (i.e. those produced in the previous year) show no further
extension of the injury produced, but that the newly formed cankers
also ceased to increase in size quite early in the season; many of them
had attained to their maximum size by the middle of June while in
others a slight increase in the cankered area could be detected towards
the end of June or early in July, but in no case did a canker show any
extension after July 12. It would appear therefore that the fungus
penetrates no further into the tissues of the host after some six or eight
weeks from the inception of the disease.

(b) Observations made in 1916.

The observations were continued in 1916, those spurs and cankers
labelled in 1915 being examined from time to time while other cankers
produced during May and June 1916 were also labelled; in all a record
was kept of over 100 spurs and cankers and in no case was there any
evidence that the fungus was able to make any further advance into

——

wee

H. WorMALD 169

the tissues of the trees after the end of June of the year in which infection
occurs. This was found to obtain not only for the two varieties Warner’s
King and Duchess’ Favourite, but also for Lord Derby and James
Grieve of which there were a few trees with the disease in the same
plantation.

On the cankers produced in 1915 conidia were first found early in
December of that year, and during January 1916 a considerable number
of spurs and cankers showed pustules of conidia which proved to be
viable and able to germinate at the temperature of the open air. On
January 27 two hanging drops of distilled water containing conidia
from separate spurs were set up and placed outside in the open air.
Within 24 hours many of the conidia in both drops had germinated
and had produced germ tubes up to 32 in length.

It was found necessary during the winter to remove the Duchess’
Favourite trees interplanted between the successive trees of Warner’s
King, so observations during the flowering season of 1916 were almost
confined to the latter variety; two Duchess’ Favourite trees were
however retained and as the disease followed the same general course
on these and on a few other trees of other varieties available for examina-
tion, the following account of the disease as it occurred on the Warner’s
King variety may be taken as describing a typical outbreak of blossom
wilt.

The withered trusses had been carefully cut off from ten trees at
one end of the row during the summer of 1915, while the rest of the
trees in the row were subjected to the usual pruning operations only
and many dead spurs and cankers, on which pustules of the fungus
subsequently appeared, were left on the trees. That there would be
another outbreak of the blossom wilt on these trees was expected there-
fore, and from the time the flowers began to open the trees were examined
frequently for the first signs of the wilt, in order that the earlier stages
of the disease could be recorded, as these had not been observed in
1915.

On May | it was noticed that a few of the flowers on these trees were
open and the rest continued to open normally during the fortnight
following. Nothing unusual was observed until May 17 when one truss
was found to be withered and several others were showing the first signs
of wilting as indicated by the flagging of the leaves at the base of each
affected truss. This wilting of the leaves is the first evident symptom
of the disease, for the flowers may fail to “set” and so wither from other
causes.

170 A Blossom Wilt and Canker of Apple Trees

From May 17 to the end of that month other trusses wilted and the
following sequence was recorded for one tree:

Date Affected trusses
May 17 2 trusses wilting (leaves beginning to flag)
ee 5 trusses dead: 2 others wilting
vee 18 trusses dead: 1 wilting
3. Gl! 19 trusses dead.

No trusses on this tree began to wilt after May 31.

In all cases where wilting of trusses occurred subsequently to May 3
they were found to be killed not by a direct infection through the blossom
but by the development of the disease in the spur or branch from
infection through flowers of other trusses. In several instances a spur
bore two trusses one of which became infected directly and the fungus
penetrated the tissues of the spur to the level of insertion of the un-
infected truss which, in consequence of the interruption to the upward
flow of sap to it, wilted from desiccation.

As an illustration of the rapid progress made by the fungus during
the month of Mav the condition of a small branch which was completely
girdled by the 31st of that month may be cited; on that day the shoots
and trusses borne at the nine nodes commencing at the distal extremity
were as follows:

Node 1. Vegetative shoot wilting (leaves yellow and drooping).

, 2. Barren spur wilting.
., 3. Dead truss with canker girdling the branch and extending
1 cm. upwards and 2-5 cm. downwards.
.. 4. Dead truss with canker nearly girdling branch.
cera: % ., halfway round.
6. - Re ,, nearly half round.
,, 7. Living healthy truss of flowers.
,, 8. Dead truss with canker just beginning.
, 9. Living truss.

The trusses at nodes 3, 4, 5, 6 and 8 had been killed by direct infection
while the shoots at nodes 1 and 2 were wilting by reason of the girdling
canker at node 3; all the spurs were unbranched and about 1-5 em. in
length; the branch, where girdled, was 0-8 cm. in diameter.

As infection probably did not occur before May 1 it will be seen
that in the course of about four weeks conidia had germinated and
produced mycelium which had traversed the tissues of the flowers and
the spurs and had invaded the branch itself to a distance of two and a
half centimetres.

H. WorMALD 171

On the older parts of the branches (1-5 to 2 em. in thickness) bearing
short infected spurs to 2-5 em. in length, there was evidence, towards
the end of May, that the fungus was advancing beyond the level of
insertion of the spurs, by the appearance of longitudinal fissures in
the bark immediately above and below the nodes. By June 7 the
affected areas on the branches were slightly sunken below the general
surface and extended round the nodes as shallow elliptical depressions.
Further extension of the fungus appeared to be inhibited about that
date and although the margins of the cankers later became more defined,
no further increase occurred superficially in the majority of those
examined. The bark over the affected areas gradually became sulcate,
often, on the smaller branches, with distinct “oyster-shell” markings
of raised lines.

Cankers which had produced the Monilia pustules during 1915 had
been labelled and all proved to be barren during 1916, no pustules being
present; no further increase of the cankered surface had occurred and in
each case callus was growing over it beneath the ruptured bark. Some
cankers, several inches long, were almost completely healed (Fig. 7).

Thus on trees where the disease has existed for two or more years
the following results are to be observed during the summer months:

A. Result of infection during that year:
(1) dead trusses of blossom ;
(2) cankers on the branches round the bases of the shorter spurs
bearing the wilted trusses.

B. Result of infection during the previous year:
(1) dead spurs bearing Monilia pustules ;
(2) cankers also bearing Monilia pustules.
C. Result of infection two years previously :
(1) dead spurs with rough cracked bark, not now bearing Monilia
pustules ;
(2) cankers more or less healed over by callus growing over the
cankered surface from the margin towards the middle line.
During winter affected trees are recognised by the dried remains of
the infected trusses (see Fig. 1). Flowers which fail to “set” into fruit
soon fall from the tree if the spur bearing them is not diseased. When
affected by the blossom-wilt Monilia, however, they are retained on
the tree through the winter. The leaves too, of these spurs, being killed
before the absciss-layer is formed, remain attached to the tree.
In December 1916 the dead spurs and cankers caused by infection in

172. A Blossom Wilt and Canker of Apple Trees

the spring of that year began to produce Monilia pustules which burst
through the dead bark. In the following month these became somewhat
pulverulent and particles of pustules placed in water set free mature
conidia. On January 24, 1917, conidia were obtained in this way in
a drop of water on a glass slide when the temperature of the air, as
registered by a thermometer placed on one of the trees, was 28-5° F.
(= — 2°C.). The shde was placed in a damp chamber at a temperature
of 6° to 8° C.; within 24 hours about 50% of the free conidia had
germinated with germ tubes to 52y in length.

This blossom-wilt fungus apparently occurs comparatively rarely on
the apples themselves, even on trees severely infected with blossom wilt.
On the apple trees at Wye it has been observed on the fruit on one
occasion only when it was found on a few very young apples soon after
they had set. In a plantation of Lord Derby trees in the Weald of
Kent there were found in January 1916 small “mummified” apples
nearly covered with a Monilia which in cultures was indistinguishable
from that found on the cankers; in the same plantation in mid-July
1915 (and again in 1916) a few young apples, about 1-5 cm. in length,
were obtained, apparently killed by the same fungus which appeared in
the form of numerous pustules bursting through the skin. I have not
yet seen this form on fully grown naturally infected apples.

(c) Comparison with other diseases producing a similar condition.

It is sometimes impossible to distinguish, from their general appear-
ance, between the Monilia cankers and those produced by Nectria
ditissima; the latter are usually recognised by the increased rate of
growth (set up in the tissues bordering the canker) which causes the
branch at that place to be swollen. Young Nectria cankers, also,
generally bear the sporodochia of the Fusarium stage in the summer,
or the pustules if not evident readily develop when placed in a damp
chamber; young Monilia cankers do not produce pustules until about
December. Again Nectria cankers originate at nodes bearing unopened
buds, or infection occurs through wounds; those produced by the
blossom-wilt fungus arise around the insertion of infected flowering
spurs.

The blossom-wilt condition bears a close resemblance to that caused
by the Apple Sucker (Psylla mali); when the damage is caused by this
insect, however, numerous cast skins are to be found attached to the
under side of the withered leaves. The larva of the Pith Moth also
produces a withering of the fruit buds but before the flowers themselves

H. WoRMALD Ive

expand, while in the blossom wilt caused by the fungus the effect is
not noticeable until about a fortnight after the flowers open.

The disease has by some growers been attributed to late frosts;
inoculation experiments (described later in this paper) carried out in
a greenhouse show conclusively that the fungus readily produces the
wilt in the absence of frost.

It would seem that the diseased twigs and branches bear a close
resemblance to those attacked by “ Fire-Blight” (Bacillus amylovorus)
in America. Prof. H. H. Whetzel, Professor of Plant Pathology in the
Cornell University, U.S.A., when he visited Wye College a few years
ago, saw trees badly cankered by Monilia, and informed us that the
general appearance of the diseased branches was indistinguishable from
the condition induced by the attacks of the Bacillus, and, as already
pointed out, Aderhold at first attributed the “Zweigdiirre” of apple
trees to the “Fire-Blight” organism. In this connection it may be
mentioned that dead apple twigs examined by Prof. Worthington KE.
Smith (20) during the winter of 1898-9 were pronounced by him to have
been killed by “Fire-Blight” since bacteria were found in the dead
tissues. No conclusive evidence that Bacillus amylovorus occurs in this
country has yet been published, and as old spurs killed by Monilia
often contain bacteria in abundance, it seems probable that Prof. W. E.
Smith was mistaken in his diagnosis and that the specimens seen by
him were twigs killed by Brown Rot.

lV. THe Biossom Witt FUNGUS COMPARED WITH OTHER
MOoONILIAS oF Fruit TREES.

a) Cultural studies of Monilias found in this country.
]

While circumstantial evidence obtained during observations made
in the open made it very probable that the Monzlia found on the spurs
and cankers produced infection of the flowers, for confirmation of this
it was necessary to reproduce the disease by the inoculation of flowers,
using pure cultures. The organism has therefore been cultivated on
sterilised media, and the cultures obtained have provided a means of
comparing this with other forms of Monilia with respect to certain
characters appreciable only in pure cultures.

Some thirty strains of the fungus have been cultivated in pure culture,
many of them in duplicate, and, without exception, no constant differ-
ences could be detected in the various strains. Two methods of obtaining
the cultures have been adopted. When the fungus is found in the

Ann. Biol. m1 12

174 A Blossom Wilt and Canker of Apple Trees

conidial stage, conidia are isolated on agar plates; the resulting spore-
lings which are seen to be uncontaminated are transferred to other
plates and when sufficient growth has taken place sub-cultures are
prepared from these. Cultures for the continued propagation of any
particular strain are made on agar slants in test tubes. Strains have
been obtained in this way from dead spurs, cankers and young apples;
one was prepared from conidia induced on a dead truss by placing it in
a damp chamber, and another from a “mummied” apple.

When the fungus is in the form of sterile mycelium only, as in the
majority of freshly killed spurs and young cankers, the surface of the
affected parts is gently rubbed over with cotton wool damped with
absolute alcohol!, and transverse sections are made through the diseased
tissues with a sterilised razor. These are dropped into a watch-glass
containing sterile distilled water and particles of the sections are then
removed with flamed needles (avoiding the outer layers of the cortex)
to another watch-glass of sterile water, from which they are finally
transferred to agar plates. As a rule cultures which were apparently
quite pure could be obtained directly in this way; sometimes, par-
ticularly in the case of spurs which had been dead for some weeks,
bacterial colonies appeared immediately round the section but sub-
cultures taken from the peripheral mycelium could be obtained pure.
Such cultures were indistinguishable from those obtained from conidia.

For comparison with these, cultures of the Brown Rot fungi from
other sources have also been prepared, e.g. forms growing on various
“stone-fruits” (plum, cherry, damson, peach), on apples and on pears.
Kach of the strains has been started from a single conidium. In those
instances where no viable conidia were present this could not be done
directly and the cultures were obtained initially from sterile hyphae;
the production of conidia was induced by growing on sterilised potato,
and “single-spore” strains were then prepared.

I have not yet obtained the ascigerous (Sclerotinia) stage of any of
these forms, so have been unable to compare them with the descriptions
of that stage as found on the Continent and in America. The fungus
which appears so generally on ripening apples is however undoubtedly
the M. fructigena as described in Woronin’s paper (23), from the large
size of its pustules (usually 1-5 to 2 mm. in diameter), the colour of the
pustules, which is a buff yellow, and the large size of the conidia; the
latter are usually about 20, in length but often exceed that. This
species I have obtained from the apple, pear, cherry, plum and peach,

* 94 % alechol has also been used with equally good results.

’

thee

A os Betteetetecetiabeed 2

on
—<—

a,

oR

H. WorMALD 17a

.

but, to the time of writing, only on the fruit and not on the flowers or
twigs!.

The blossom-wilt Monilia, as it occurs on the twigs, fruiting spurs
and cankers of the apple tree in winter and spring, has pustules generally
smaller (but they may reach a diameter of 1-5 mm.) and grey in colour
rather than yellow, though when very powdery they may become some-
what ochraceous and are then easily mistaken for M. fructigena, while
the conidia are much smaller, being usually 10-12, in length and
rarely exceeding 14. When, as occasionally happens, it occurs on
young apples in summer the dimensions of the conidia are greater and
this again may lead to some confusion.

The two species grow vigorously on suitable sterilised culture media
and the utilisation of this property will probably afford a ready and
more definite means of identifying them, particularly as cultural methods
can be employed when the fungi are in the condition of sterile mycelium
only. Differences in habit of growth and development of conidia dis-
tinguish them when growing on two easily prepared media, viz., prune-
juice agar and steamed potato.

The medium generally used by the writer to obtain mycelial growth
is a decoction of prunes (1 °%% stoned prunes) containing 1-5 or 2 % agar-
agar. M. fructigena produced, on plate cultures of this medium, a flat
dise of mycelium, circular in outline and growing out regularly to the
edge of the plate with a margin laciniate or almost entire. The growth
of the blossom-wilt Monzlia at first resembles that of M. fructigena but
soon shows a tendency to become lobed, the lobes being deltoid or
flabelliform with narrow sinuses between them; a characteristic feature
of these cultures is for growth to be arrested at about mid-way between
the centre and the edge of the plate and fresh growth commences as
flabelliform radiations from between the lobes. This condition has
not been observed in cultures of M. fructigena.

Neither species produces conidia on the prune-juice agar medium
although both give rise to numerous clusters of minute spore-like bodies
produced in radiating chains which in the earlier stages of development
resemble the fructifications of a Penicillium but later become dense
irregular clusters. These “sporidia” were observed on M. fructigena
by Humphrey (11) who states that they germinate and produce mycelium.
Woronin figures them for M. fructigena and M. cinerea but failed to

1 Since the present paper was sent to the press evidence has been obtained which
suggests that WM. fructigena may extend from the infected fruit into the branches and
form cankers, in the case of certain varieties of soft-wooded apple trees (e.g. James
Grieve); this point is being investigated.

12—2

176 A Blossom Wilt and Canker of Apple Trees

induce them to germinate; attempts by the present writer to obtain
germ tubes from these sporidia have also failed.

Of sterilised media the most suitable for the development of conidia
has hitherto proved to be steamed potato, and in diagnosing a fresh
strain this is at present always used. Roux’s tubes are found useful
for these cultures though ordinary wide test tubes are also employed.
Vigorously growing mycelium transferred from an agar plate will reach
the edge of the potato in four days and tufts of conidia are produced
within a week. Measurements of conidia from potato cultures were
taken when the cultures were 6-8 days old; if left for a longer period
the conidia, particularly those of the canker strains, became overgrown
by hyphae probably from their own germ tubes. The conidial tufts
of M. fructigena in these cultures are yellow in colour and usually form
distinct continuous raised zones towards the upper end of the potato,
while those of the canker strains were smaller, scattered, more diffuse,
and grey in colour. . .

An examination of the Monilias found on the stone-fruit trees
(Prunus spp.) in this country led to the discovery that there are (in
addition to M. fructigena which occurs frequently on the plum and
sweet cherry) two forms both of which may be referred to M. cinerea,
though they differ from one another culturally, and one of these appears
to be identical with that form which causes the destruction of the apple
blossom. The latter when growing as plate-cultures on prune juice
agar is at first hyaline, but, when growth has extended for about 2 em.
from the point of inoculation, a circular band appears, olive green to
dark brown in colour, approximately at 1 cm. from the centre, and
later other brown zones develop. This brown coloration is more
intense when the fungus is grown as agar slant cultures and appears
to be favoured by the depth of the medium, for the brown zone first
appears as a curved transverse band on the lower (deeper) side of the
point of inoculation. The colour becomes darker and more general
with the age of the culture which eventually is almost black throughout.
Strains which also become brown on this medium have been obtained
from the stone-fruit but whether any of these are able to infect the
apple flower has not yet been determined.

Other strains, obtained from plums and sweet cherries, remain
hyaline both in plate and tube cultures of prune-juice agar. Of two
strains obtained by isolating conidia from a damson, one retained the
hyaline appearance, the other produced a dark brown band when the
two were grown side by side on the same plate.

H. WorMALD Ure

Other differences have also been detected between the two types;
thus the blossom-wilt form produces conidia less freely than the hyaline
form from plums when grown on steamed potato. Again when inocula-
tions are made on mature apples the former soon causes the fruit to
become dark brown to black with conidial pustules absent or very few
in number, while with the latter the dark colour is less pronounced and
numerous pustules are readily developed.

Four strains have been obtained from flowers and twigs of acid
(Morello) cherry trees and these on prune-juice agar produce a brown
coloration which however is not so uniform as that of the strains from
the apple trees.

Whether the hyaline form is less virulent than the brown is at present
uncertain, and it is proposed to carry out inoculation experiments with
that object in view, but it is significant that all the strains obtained
from wilted trusses and cankers of the apple tree produce on the agar
cultures a coloration which eventually becomes almost black.

(b) Dimensions of Conidia.

Apart from the colour of the pustules, the size of the conidia is the
character most generally quoted as determining the species of Monilia
(in the absence of the ascigerous stage) found on cultivated fruit. Of the
strains of the blossom-wilt fungus at first studied only a few conidia
were measured; it was soon evident however that not only did the
average size of the conidia vary with the conditions under which they
were produced, but also that there was considerable variation in the
dimensions of conidia taken from the same pustule. The number taken
therefore in order to obtain a fair average was in each case 100, for those
strains obtained more recently.

It has already been stated that the Monilia of the apple blossom wilt
has in general smaller conidia than M. fructigena and this difference is
particularly well marked if a comparison is made of the dimensions in
the two forms when each is producing pustules most abundantly under
natural conditions. Thus if conidia of the blossom-wilt fungus are
taken from cankers or spurs in the spring when this form is most con-
spicuous, the typical conidia are found to measure 11-12 x 8-9u.
M. fructigena on the other hand is most vigorous in late summer and
early autumn when the majority of its conidia reach or exceed 20 in
length. If, however, the two are grown under the same conditions, as,
for example, on sterilised potato in the laboratory the difference is less

178 A Blossom Wilt and Canker of Apple Trees

striking, for the conidia of the blossom-wilt Monilia are then consider-
ably larger than when they are produced on spurs and cankers.

Four strains obtained from dead spurs and cankers from February
to April, 1916, were subsequently grown on steamed potato and in each
case 100 conidia were measured! with the result as shown in the table.
The conidia taken from the sterilised potato were measured when the
cultures were 6-8 days old.

Average of 100 conidia Average of 100 conidia

Strains of the Blossom Wilt Monilia taken from spurs taken from cultures

from spurs and cankers of apple trees or cankers on steamed potato
Strain XI from dead spur, Feb. 1916 12:0 x 8-5 u 19-0 x 15-0 4

» XII from dead spur, March 1916 11:5 x 85 19:0 x 15-5 4

» XIII from a canker, April 1916 11:5 x 8:5 yu 18:0 x 144

», XIV from another canker, April 1916 11-0 x 8-On 18:0 x 13-54

Strain XIV was subsequently used in inoculations on apple blossom
in the open, and conidia obtained from an infected flower stalk in July
had an average size (for 100 conidia) of 15-5 x 12-0, but conidia taken
during the following winter (February 1917) from a spur killed by
inoculation with this same strain gave an average of 11-0 « 8-5 which
is a close approximation to that obtained from conidia taken from the
original canker.

A strain obtained from young apples soon after setting (June 12, 1915)
was found to have conidia the majority of which were from 17 to 20u
in length but sufficient measurements were not made for an average to
be taken comparable with the above. Another strain obtained from
young apples in July 1916 gave an average of 18 x 13, for 100 conidia
taken from the fruit itself, and when grown on steamed potato the
average was about the same, i.e. 17-5 x 14-04. The other cultural
characters of these two strains, e.g. habit of growth and brown coloration
on prune-juice agar, and grey conidial tufts on potato, resemble those
of the typical blossom-wilt strains.

It has not yet been ascertained what factors determine the output
and the size of the conidia, but probably temperature and the degree
of moisture have some influence. The conidia are invariably smallest
if obtained in winter when the fungus is living on dead wood and bark ;
they are much larger when produced on young apples in the summer
or on potato at the temperature of the laboratory (15°-18° C.).

Therefore when a Monilia obtained from an apple tree is examined
with a view to its identification it is essential that the conditions under

’ The readings were taken correct to 0:5 and the averages were calculated also
correct to the nearest 0:5 u.

H. WorMALD 179

which the conidia are produced be taken into consideration. The range
of variation and the average size of the conidia of the blossom-wilt
fungus under certain conditions is indicated in the accompanying table :

Dimensions of the Conidia of the Blossom-Wilt Monilia.

Source of conidia Range of variation Average
Spurs and cankers in winter and spring 7x 5 to 20x 134 Ll-5 x 8-5)
(usually 10-14 x 7-9-5 1)
Young apples in the open in summer (June 9 x 7-5 to 25 x 174 18 x 134
and July) (usually 16-20 «x 11-14-5 u)
Potato cultures at temperature of laboratory 9 x 7 to 26 x 204 18-5 x 14-5
(15°-18° C.) (usually 16-20 x 11-14-5 p)

When this fungus is growing on young apples in the open or on
steamed potato in the laboratory the dimensions of the conidia approach
those of M. fructigena, but the smaller size and grey colour of the
pustules distinguish it from the latter.

Strains of erey Monilias on plums show a similar variation as here
shown:

Average of 100 conidia Average of 100 conidia

from original source taken from culture
Strains of grey Monilias from plums of strain on steamed potato
Strain VII, mummied plum, February 11-5 x 8-04 16:0 x 12:5
3 LEE as » April 11:5 x 8-Ou 16-0 x 12-04
yon SX *, » April 12:0 x 8-Ou 17:0 x 12:04
» - XX, dead twig, June 14:0 x 10-54 16-5 x 12-04
> XI, yofng green plum, June 16:0 x 10:04 16-0 x 12:04
» XIt, canker on plum tree, August no pustules present 15-5 x 11-5
> XIII, mature plums, August 14:5 x 11-04 15-5 x 11-54

Strain VIII on prune-juice agar produced a browning of the medium
comparable with that caused by the blossom-wilt form of apples; the
others remained hyaline except Strains X and XI which were inter-
mediate in this respect, the coloration being less pronounced (hardly
perceptible in plate cultures) than in Strain VIII.

These strains from plums generally produced conidia more readily
on steamed potato than did those from apple spurs and cankers; the
conidial tufts of the latter were always scanty whereas the former may
produce definite concentric circles of conidiophores.

Four strains have been isolated from Morello (acid) cherries and
though they have not yet been closely studied they behave in cultures
as the Strains X and XI from plums.

Monilia fructigena shows less variation in the size of its conidia but
I have not yet found this species producing fresh chains of conidia in

180 A Blossom Wilt and Canker of Apple Trees

the winter time; pustules examined in March and April have been
barren or the conidia present were not viable, a condition observed on
the Continent by Ewert(9). Three strains of M. fructigena of which a
sufficient number of conidia have been measured gave dimensions as
follows:

Average of 100 conidia Average of 100 conidia
from original source —_ taken from cultures

Strains of M. fructigena of strain on steamed potato
Strain VII, mummied apple, June... 19:0 x 12-54 19-5 x 11-54
» WWIII, young plums, July ao. 21:5 x 13:04 20:5 x 13-04
» IX, mature apple, October ee 20-5 x 13:5 20:0 x 12-5

(c) American Strains of Monilia.

As much attention has been given to the study of Brown Rot diseases
in America it was possible to obtain specimens and cultures from that
continent for comparison with those described in the preceding pages.
In all, ten strains have been cultivated and examined. Apples and
cultures were received from Dr G. B. Posey of the Oregon Agricultural
College, a culture (from a peach) and mummified plums from Prof.
L. R. Jones, Professor of Plant Pathology in the University of Wisconsin,
mummified plums and peaches, together with cultures prepared from
ascospores, from Mr W. A. McCubbin, Laboratory of Plant Pathology,
Ontario. Thus the ten strains were from three different hosts, viz.
apples, plums, and peaches, and represented three widely separated
recions. :

The strains are characterised by certain features common to all of
them but absent from the Monilias obtained from material collected in
this country. The most striking character peculiar to the American
form is the readiness with which it produces conidia on sterilised culture
media; even on prune-juice agar, as plate or as tube cultures, numerous
tufts of conidiophores develop, usually within three or four days from
inoculation.

When growing on plates of prune-juice agar the mycelium grows out
to form a regular circular dise with an entire or sub-entire margin,
resembling the growth of M. fructigena in its rate of development and
general habit rather than that of our grey Monilias; the production of
numerous tufts of conidiophores (often in concentric zones) in these
cultures serve to distinguish it from M. fructigena.

On steamed potato the conidial tufts are grey and are usually so
numerous that they form an almost continuous pulverulent layer over
the surface. In these cultures therefore the general appearance of the

H. WorMALD 181

American form is very unlike that of M. fructigena and the conidia too
are smaller than in the latter as will be seen by comparing the dimensions
given below with those of M. fructigena.

The dimensions of the conidia of those strains where 100 conidia

were measured were as follows:
Average of 100 conidia

OOO

From the Agar Potato

fruit culture culture
Strain C, mummied peach from Ontario 16-5 x 12:04 14:0 x 9-54 16:0 x 12:04
oD “A plum _,, ap 17:0 x 12-54 140x 954 16:0 x 12:54
» , culture (peach) from Wisconsin — 14:0 x 95u 15:5 x 10'5u
» F,mummied plum _,, Ps ‘No suitable 14:0 x 9-O0u 16:5 x 11:04

conidia present

» G,ascosporicculture(peach)from Ontario — 14:0 x 9:54 14:0 x 10-54
50) EE ay 5 (plum) ,, 55 = 140 x 80O0n 13:5 x 10:04

The most uniform results are obtained on prune-juice agar but as
all the British strains do not develop conidia! under those conditions
a comparison of the size of the conidia of the various forms from this
culture medium cannot be made.

Growth is vigorous on the media employed; in one case conidia
placed on agar germinated at room temperature (about 18° C.) producing
within three days sporelings 3 mm. in diameter, and already short chains
of conidia were present. On prune-juice agar plates growth takes place
at a more rapid rate than in the case of the grey Monzlias of this country,
and the periodicity in rate of growth observed in the latter is absent from
the American form which grows out uniformly from the point of inocula-
tion to the edge of the plate.

In a paper recently published Bartram(5), working with strains of
Monilia obtained in Vermont, concludes that the form in that state is
M. cinerea but finds that it produces conidia in agar cultures; it is
probable therefore that the Vermont strains resemble those received at
Wye from Oregon, Wisconsin, and Ontario and that that form is the
one commonly found in North America.

(d) Continental Strains of Monilia.

From our geographical situation and the frequency with which
nursery stock is imported from the Continent it is natural to expect
that the Monilias of this country would be closely related to, if not
identical with, those of the rest of Europe, and from descriptions

1 Very rarely have I seen an occasional short chain of conidia on plate cultures of
M. fructigena on this medium, but no definite pustules or tufts of conidiophores have
developed, nor sufficient conidia for comparative measurements.

182. A Blossom Wilt and Canker of Apple Trees

available it seemed probable that M. fructigena and M. cinerea of the
Continent were present on our own fruit trees. When it was found that
cultural methods were a great aid not only in distinguishing M. fructigena
from the grey Monilias, but could also be employed for the recognition
of certain forms among the latter, efforts were made to obtain strains
of the Brown Rot fungi from the Continent.

To the time of writing I have only succeeded in obtaining specimens
from Holland, Dr H. M. Quanjer of the Instituut voor Phytopathologie,
Wageningen, having on several occasions kindly forwarded mummified
fruit. The fungus usually present on such material was a yellow one, in
cultures indistinguishable from the yellow form of England, and there
can be no doubt that this is Monilia fructigena Pers. One strain from
these specimens was obtained from a mummified pear received in March
1916. Conidia were fairly numerous and on the average measured
19-5 x 9-5, dimensions which are greater than the average size of
conidia from grey Monilias in winter; these conidia failed to germinate
when placed on agar so evidently they had not survived the winter.
Particles of a pustule were teased out in sterilised water and placed on
a prune agar plate; these grew out but were impure. More successful
results were obtained by cleansing a portion of the skin with cotton-
wool soaked with 94 °% alcohol, then, raising a portion of this sterile
skin with a scalpel, particles of the pulp were removed and placed on
an agar plate. The resulting culture was apparently pure; sub-cultures
on potato yielded conidia and the isolation of some of these afforded
pure sporelings. Sub-cultures from the sporelings were then grown for
comparison with other strains of M. fructigena and no appreciable
difference has been detected either in the mode of growth or coloration ;
on potato the average size of the conidia was 21-0 « 12-5y.

One batch of fruit from Dr Quanjer received early in January 1917
included plums bearing grey Monilia pustules with viable conidia and
a strain was isolated from each of two of the plums. Conidia, taken
from the plums as received, and later obtained from potato cultures,
were measured and their dimensions found to be as follows:

Average of 100 conidia in each case

From mummied plums From potato culture

Grey Monilia from Holland, Strain 1 11-5 x 8u 16:5 x 12-5
11-0 x Tp 16-0 x 12-04

bo

9 99 bb °°

It will be seen that these figures approximate closely to those obtained,
under corresponding conditions. from strains found on plums grown in

H. WorMALD 183

this country. The mode of growth on culture media is also similar to
that of our grev Monilias. Neither of the strains from Holland has
produced definite brown zones in tube cultures of prune-juice agar but
irregular brown patches eventually appeared. They thus resemble
(and are probably identical with) a form frequently found on plums
and acid cherries in Kent.

I have not yet obtained diseased apple twigs from abroad.

(e) Nomenclature of the various forms.

In cultures it is possible therefore to distinguish four types of
Monilia as found on cultivated fruit trees of the genera Pyrus and
Prunus, as here tabulated:

Cultures on steamed
potato in Roux’ tubes

Prune-juice agar plate
cultures

(1) Monilia fructigena,
occurring commonly
on apples and plums,
and frequently on
sweet cherries.

Margin almost entire or laciniate; no

Conidial tufts yellow,
brown coloration; conidia absent.

well-developed at
upper end of potato,
forming raised zones.

(2) Blossom Wilt Margin with deltoid or flabelliform Conidial tufts grey,

Monilia of the apple,
also occurring occa-
sionally on plums.

(3) A grey Monilia
frequent on plums
and sweet cherries.

(4) American form of
Monilia.

lobes, growth usually arrested about
mid-way between centre and side of
plate, and new outgrowths as flabelli-
form lobes develop, usually from
the sinuses; olive-green, to brown,
zones appear, the first usually at 0-5
to lem. from the centre; conidia
absent.

As above but no brown zones appear.

Margin entire or crenate; conidial tufts

numerous, usually in concentric
circles; brown coloration of the agar
absent or appears as a_ peripheral
band near edge of plate; growth
generally more rapid than in (2) or
(3) and more uniform.

few and scattered.

Conidial tufts grey,
more numerous than
in (2), often appear-
ing in concentric
circles round point
of inoculation.

Conidial tufts grey,
almost covering the
whole surface in a
continuous Jayer.

As shown in the table the successive brown zones typical of the plate

cultures of the blossom-wilt Monilia do not appear in cultures of the
other forms; sometimes a brown coloration occurs in the plate cultures
of the American form but in that case it commences as a peripheral
band when the mycelium approaches, or has reached, the edge of the

184 A Blossom Wilt and Canker of Apple Trees

plate. When growing as “slant cultures” in test-tubes both M. fructi-
gena and the American Monilia usually produce a browning in old
cultures, but this commences at the extreme lower end of the slant and
gradually extends for some distance upwards, while in the case of the
blossom-wilt fungus the first brown zone appears usually from 0-5 to
1 em. from the point of inoculation when the culture is still quite young
(often within a week).

The typical hyaline form from plums and sweet cherries remains
colourless even in old slant cultures.

On the Continent, opinion is divided as to whether all instances of
Brown Rot, caused by Monilia, on fruit trees are to be attributed to
one species only (retaining for it the name Monilia fructigena), or to
more than one. Those who favour the latter view generally assume
that there are three species concerned, viz. M. fructigena, said to occur
chiefly on the pomaceous fruits, characterised by its yellow pustules and
comparatively large conidia which usually are not viable in winter:
M. cinerea, considered to be found almost exclusively on the “stone”
fruits, with grey pustules and smaller conidia which retain viability
throughout the winter: and M. laxa, a grey form, with small conidia,
occurring on apricots. Aderhold and Ruhland(4) claim to have proved
that these three species are to be distinguished by their respective
ascigerous (Sclerotinia) stages, which they describe. SS. fructigena they
obtained from mummified apples, and from the ascospores were able
to develop the yellow Monilia stage, M. fructigena. That the two forms
are stages in the life-history of one fungus appears therefore to be fully
established. With regard to Sclerotinia cinerea their conclusions are
less convincing. They were unable to obtain an ascigerous form from
peaches, plums or cherries, so described the form, which they refer to
S. cinerea, from apothecia (preserved in spirit) which had been found
on mummified peaches and had been sent to them from America, and
assumed that the Monilia cinerea of Europe is its conidial stage.
Mummified apricots yielded a Sclerotinia which they described and
named S. laxa, with Monilia lava as its conidial stage.

Since, as has been shown above, the Sclerotinia generally occurring
in America produces a conidial stage which can be readily distinguished,
in cultures, not only from the grey Monilias of this country but also from
those strains obtained by the writer from plums received from Holland,
it would seem probable that Aderhold and Ruhland included in their
diagnosis of the one species Sclerotinia cinerea two forms which are very
different when grown under certain cultural conditions.

H. WorRMALD 185

The description given of Monilia laxa of the apricot, as the conidial
stage of Sclerotinia laxa applies equally well to the grey Monilias of
other fruit. The most striking difference thev give is in the size of the
conidia, viz. M. cinerea 9-3-14-5 x 6-2-12-4y, M. laxa 12-4-23-8 x 9-3
-15-5 4, but this difference is no greater than that shown by the blossom-
wilt fungus or the Monilias of plums when growing under such diverse
natural conditions as (a) mummified fruit or dead spurs in winter,
(b) fresh fruit in summer.

The present writer is of the opinion from the evidence to hand that
the grey Monilias of this country are to be referred to the continental
form M. cinerea Bon., the Sclerotinia stage of which has not yet been
described, unless indeed it should prove to be identical with Sclerotinia
laxa (Khrenb.) Aderh. et Ruhl. The blossom-wilt fungus must pro-
visionally therefore be included under Monzlia cinerea Bon. This
species as at present delimited includes at least two forms, one which
remains colourless in prune-juice agar cultures, and another which
produces zones of an olive green to dark brown colour. The American
Brown Rot fungus appears to be a distinct form and it may be necessary
to distinguish it specifically from the European species.

V. INocULATION EXPERIMENTS WITH PURE CULTURES.

As observations in the open had indicated that in all probability
infection occurred through the open flower, it was necessary that
inoculation experiments should be carried out when the trees were in
bloom, and the time available under natural conditions was thus limited
to about a fortnight in the year. It was decided therefore that a first
series of inoculation should be performed under glass in order that any
information thus obtained, relative to the mode of infection, could be
utilised during the same spring in subsequent experiments in the planta-
tion. Another reason for conducting the experiment under these
conditions was to eliminate the possibilitv that the action of frost would
modify the course of the experiment.

Some difficulty was experienced in obtaining conidia in sufficient
quantity to ensure successful inoculation. With the object of inducing
the fungus to produce conidia readily on a sterilised substratum many
culture media were tried, e.g. agar-agar media: 1-5 °4 agar in an extract
of each of the following substances: celery, prunes, French beans,
apple twigs, malt culms (also malt culms + 10 °% starch), maize meal,
potato. Prune-juice agar containing strips of filter paper was also used.

186 A Blossom Wilt and Canker of Apple Trees

Woronin obtained conidia of Monilia cinerea on gelatine cultures pre-
pared with apple juice or bread broth: these were tried for the apple
blossom Monilia but without success. Nor were conidia produced on
bread steamed in test tubes or on pieces of apple branches sterilised in
the autoclave. The best results were obtained with semi-cylinders of
tubers and roots as potato, artichoke (Helianthus tuberosus), carrot,
mangel ang parsnip. The last two produced a trace of conidia, the
other three developed conidiophores in scattered tufts, and potato,
sterilised either by intermittent steaming (20 minutes at 100° C. for
each of three successive days) or in the autoclave for 20 minutes at
115° C., was generally used as the medium for cultures used in the
inoculation of apple flowers.

(a) Inoculation of apple flowers in the Greenhouse.

For this experiment young apple trees about three feet in height
were acquired; for convenience they were planted in pots early in the
year (1916) and left outside until the fruit buds showed the first signs
of expanding. The pots were then transferred to the greenhouse, with
the result that the earliest flowers were fully expanded during the first
week in April.

Conidia for the inoculations were taken from cultures, about seven
days old, on steamed potato (in one instance a culture on steamed
carrot was used). They were removed from the culture tube and placed
on the stigmas of the flower by means of a sterilised platinum needle,
which was re-sterilised in the flame of a spirit lamp after each inoculation.
One flower was successfully inoculated by inserting the needle between
the stamens and the styles so that the conidia were deposited at the
base of the styles, but this method was not generally adopted as usually
there was no space between the filaments for the insertion of conidia
within the hollow receptacle of the flower.”

One flower only of each inflorescence was inoculated, and of twenty-
four flowers thus treated thirteen not only withered but produced a
wilting of the trusses to which they belonged. The results obtained on
the four varieties tested were as here shown; in one column appears

No. of flowers No. of successful

Variety inoculated inoculations
Prince Bismarck 9 7
Lord Derby 4 2
Cox’s Orange Pippin 5 2
Worcester Pearmain 6 2

Totals 24 13

H. WorMALD 187

the number of trusses treated (i.e. one flower of each inoculated), and
in the other is the number of successful inoculations resulting in each
case in the death of the whole truss.

The blossoms of the Prince Bismarck trees were the first to expand
and the earliest of these were inoculated immediately after opening,
some of them on the first day on which the stigmas were exposed. The
flowers of the other varieties were as a rule rather older when inoculated
and this suggests (judging from results as indicated in the table) that
the flowers are most susceptible to infection immediately after they
open. None of the untreated trusses showed any signs of wilting
throughout the season except in those cases where the fungus had
extended from the inoculated flowers so far as to cut off supplies to
those parts not directly infected. ;

The results obtained from these experiments, in those instances
where inoculation of a flower was followed by the wilting of the remaining
flowers and also the leaves of the spur, are given below in detail to
illustrate the progress of the disease from the time conidia were placed
in the flowers.

All the trees were removed from the greenhouse about the middle of
May and were left in the open throughout the winter.

Variety Prince Bismarck.

(a) Flower of a spur 1 cm. in length inoculated April 4 by placing
conidia on the stigmas. On April 11 the styles were dark brown for
a distance of 4 mm.; the whole truss was dead before the end of April
and on May 5 a canker extended from the base of the spur to three-
quarters of the distance round the branch.

(b) Flower of a spur 2:5 cm. long inoculated April 8; another truss
of flowers borne on the same spur was untreated. The inoculated
truss was dead by April 26 and the disease had reached the base of the
other truss on that spur so that this also wilted.

(c) and (d) Flowers inoculated April 8. Both trusses were dead
on April 26.

The trusses thus treated were all borne on a branch bearing in all
six trusses of flowers, four of which were infected directly as shown;
the fifth became infected indirectly from truss (6), and the sixth also
wilted early in May as the result of a canker produced at the base of
the inoculated truss (d) at the next node below. The cankers developing
round the bases of the infected spurs later became confluent and by
June 8 that portion of the branch bearing the trusses was cankered for

iss A Blossom Wilt and Canker of Apple Trees

a distance of 19 cm. or from 3 cm. above truss (a) to 7 cm. below (d);
the terminal portion of the branch died as a result.

The fungus itself was not observed until the first week in December
when young pustules with chains of immature conidia (the conidia
remained cohering in chains when particles of the pustules were mounted
in water) were seen at the lower end of the canker. Later others appeared
and by Jan. 9 numerous conspicuous pustules were present along the
whole length of the canker and on the infected spurs. By this time the
conidia were more or less pulverulent as a considerable number became
free on mounting in water; such conidia were viable and many of them
germinated within forty-eight hours when the slide was kept in a damp
chamber at a temperature of 6° to 8° C.

(ec) On another branch a truss was infected by placing conidia at
the base of the styles of one flower on April 8. The truss wilted before
the end of the month and a canker developed which had nearly girdled
the branch when examined on June 8. Immature Monilia pustules
were present at the base of the spur on Dec. 6; these were well developed
by Jan. 9.

(f) On the same branch as (e) a similar truss was infected by
placing conidia on the stigmas of a flower. The spur was killed before
April 26, but no canker was produced on the branch. One Monilia
pustule was found on the spur on Jan. 9.

(g) A truss inoculated on April 11 was dead within fifteen days.
On April 27 the twig bearing it was removed from the tree and it was
found that the tissues of the spur (which was 1 cm. in length) were
brown to the level of its insertion on the twig and the disease was
already encroaching on the tissues of the twig itself. In sections made
transversely through the spur there were found numerous hyphae,
particularly in the pith, and particles of the sections placed on an agar
plate produced mycelial growth resembling that obtained in a similar
manner from naturally infected spurs.

The exact date when the trusses of this variety first showed signs of
wilting was not ascertained, but all those referred to above were dead
on April 26 and had probably begun to wilt some days previously. The
varieties which were treated at a later date were examined more fre-
quently and the earlier symptoms of the disease noted. _

H. WorMALD 189

Variety Cox's Orange Pippin.

(a) Truss of a spur 1 em. long on a branch (0-4 em. in thickness)
bearing a terminal vegetative shoot distal to the inoculated truss.

April 19. A single flower inoculated by placing conidia on the stigmas.

, 27. Stigmas brown to base, stamens drooping.

May 1. Truss wilting: flowers and leaves drooping, the latter curled

and brown.

Branch, bearing the truss, half girdled by a canker proceeding

from the base of the spur.

, 9. Branch three-fourths girdled: leaves of the vegetative shoot
on the same branch flagging.

June 8. Canker extending to 5 cm. below the infected spur. The
upper limit of the canker was not distinctly marked, but
the distal portion of the branch was quite dead by this time.

Jan. 9 (1917). Well-developed Monilia pustules present on the dead
spur.

(b) Truss of a spur 0-5 em. long on a twig (0-4 cm. in diameter)
bearing another flowering spur distally.

April 19. A single flower inoculated as in (a).

» 27. Stigmas of inoculated flower brown and withered, stamens
collapsed, pedicel discoloured to its base, sepals brown and
withered. Other flowers of the same and of the neighbouring
trusses showed a little browning of the stigmas on this date
but in every case the stamens were upright and the pedicels
and sepals were not discoloured.

May 1. Truss withered: flowers drooping, leaves curled and brown.

» 9 The other truss (not inoculated) on same twig wilting.

» 9. Twig completely girdled, both trusses dead.

June 8. ‘Twig dead to the level of its insertion on the stem, but no
canker had developed on the latter.

Jan. 9 (1917). One Monilia pustule present at the base of the spur.

‘(c) A flower inoculated on April 19 showed the same symptoms of

the disease as in (a) and (b) to April 27 when the stigmas were brown to
the base, the stamens collapsed and the pedicel brown throughout its
length; on April 29 however it was accidentally removed, evidently
before the fungus had invaded the tissues of the spur itself for the leaves
were still alive in June. None of the flowers of this truss set into fruit
but this was probably due to non-pollination.

Ot

39

Ann, Biol, ut 13

190) A Blossom Wilt and Canker of Apple Trees

Variety Worcester Pearmain.

(a) Truss of a spur 1 em. in length on a branch 0-4 em. in diameter.

April 19. One flower inoculated on the stigmas.

,, 27. Styles brown to the base, stamens collapsed. On a neigh-
bouring control truss of the same age the styles were not
brown and the stamens were upright.

,, 29. Flowers and leaves of the truss drooping.

May 1. Truss withered, leaves brown and curled.

June 8. Canker present on the branch, nearly girdling.

Dec. 6. Young pustules (bearing very short chains of immature
conidia) appearing.

Jan. 9 (1917). Conspicuous Monilia pustules present on the canker
and dead spur.

(6) Truss of a spur 1-5 em. in length on a branch 0-5 em. in diameter.

April 19. One flower inoculated on the stigmas.

, 27. Styles of inoculated flower brown, some of the stamens
collapsed; these features were not present on a neighbouring
control truss.

,, 29. Flowers and leaves of the truss drooping.

May 1. Truss withered, leaves brown and curled.

June 8. Canker present round the base of the spur, girdling the
branch and extending 2-5 cm. below the insertion of the spur;
that portion of the branch distal to the canker was dead.

Jan. 9 (1917). Several Monilia pustules present on the dead spur.

Variety Lord Derby.

On this variety the inoculations which resulted in wilting were made
on trusses borne on short spurs at nodes on the main stem of a young
tree. The flowering spurs were four in number and inserted at successive
nodes, the stem being 0-6 to 0-7 em. in thickness in this region. The
uppermost truss (a), borne on a spur 2 em. in length, was inoculated ;
at the next node below was an untreated control truss (b); the third (ce),
on a spur | em. long, was inoculated while the fourth (d) was another
control. Within a few centimetres of (a) the stem divided into three
branches all bearing leaves only.

April 18. The first flower to open (i.e. in the centre of the inflorescence)
of each of the two trusses (a) and (c) was inoculated by
placing conidia on the stigmas.

April 27.
Noe
May 1.
> 5.

4 ALO:
spon.
ie).

June 2
Dec. \6:

H. WorMALD 191

The styles of the inoculated flowers were withered and brown
to the base and the stamens had collapsed. In the corre-
sponding flowers (of the same age) on the control trusses
the styles were not discoloured and the stamens were upright.
All the flowers of the inoculated trusses were drooping and
the leaves were beginning to wilt; the stamens of the un-
treated flowers were however rigid (not collapsed with
distorted filaments as in the two inoculated flowers).
All the leaves of trusses (a) and (c) were by this time curled,
brown and withered; the stamens of the non-inoculated
flowers were still rigid. (This stage is shown in Figs. 3
and 5.)
From truss (c) the disease had traversed the tissues of the
spur and was invading the main stem as shown by a slight
sinking of the bark at the base of the spur and extending
about half the way round each side; truss (b) was just
beginning to wilt, indicating that the transpiration current
was being interrupted, and that the vessels of the xylem
were attacked.
The canker at truss (c) had almost girdled the stem and
extended 0-5 em. upwards and 1-5 em. downwards from the
base of the spur; the cankered area was indicated by a
distinct wrinkling of the bark which was also slightly de-
pressed below the general level.
The stem was now completely girdled; the lowest branch
distal to the affected spurs began to wilt.
The leaves of the other two branches above the spur were
drooping and turning brown.
A canker had also developed from truss (a) so that by this
date the two cankers had united and together they extended
along the stem for a distance of 10 em., i.e. from 4 em.
above (a) to 6 cm. below (c); truss (d) however was still
alive. Towards the upper end of the canker there was a
series of dark lines more or less parallel to one another giving
a zonate appearance. That portion of the tree above the
infected spurs was by this time quite dead. (This stage is
shown in Figs. 4 and 6.)
Young pustules were bursting through the bark at the lower
end of the canker; the immature conidia remained cohering
in chains when mounted in water.

13—2

192. A Blossom Wilt and Canker of Apple Trees

Jan. 9 (1917). Numerous conspicuous pustules were present along
the canker and on the infected spurs; the conidia were
more or less pulverulent and a considerable number became
free on mounting in water. Such conidia were viable for
many of them germinated within 48 hours in distilled water
at a temperature of 6° to 8° C.

(b) Inoculation of apple flowers in the Plantation.

The experiments carried out in the greenhouse furnished definite
proof that the fungus is a true parasite under favourable circumstances.
It was desirable however that similar experiments should be performed
in the plantation in order to obtain evidence that the disease could be
induced in a similar manner on trees growing in the open.

Early in May 1916 inoculations of apple flowers in the plantation
were made for comparison with the results obtained under glass. The
flowers of the variety Warner’s King began to open about May 1 and
as this variety was known to be susceptible to the disease it was selected
for the experiments.

As before, the conidia were obtained from cultures, about a week
old, growing on steamed potato, but the method adopted in the actual
inoculation of the blossoms was slightly modified, in order to avoid any
injury to the stigmas that might occur when using the platinum wire
in transferring the conidia from the cultures. The steamed potato,
with the fungus, was removed from the culture tubes (using sterilised
instruments) and placed in sterilised petri dishes; small portions
bearing tufts of conidiophores were cut off with flamed scalpels and
placed in another sterile petri dish which was taken direct to the
plantation. In each case only one flower of each truss was inoculated
and the inoculation was made by removing from the dish, on the point
of a sterile needle, a particle of potato bearing conidiophores and bring-
ing the conidia in direct contact with the stigmatic surfaces, so that the
needle itself in no case touched the stigmas. The selected flower of
the truss was marked, for identification and comparison with the rest,
by a few inches of white cotton tied leosely round the flower.

Variety Warner's King.
Results of inoculations:
(a) May 5. One flower of a truss inoculated on the stigmas.
» Ll. One style of the inoculated flower brown to the base.
18. The stamens of this flower had collapsed, while those of

be)

10

Or

H. WorMALD 193

other flowers of the inflorescence were upright. On
this date the treated flower was accidentally removed
during the examination and the disease made no further
progress, the spur remaining alive throughout the
summer,
One flower inoculated, borne on a spur 2 em. in length.
One style brown in the middle.
Stamens collapsed (stamens of the other flowers of that
truss upright).
Flowers and leaves of the truss wilting.
The bark round the base of the spur beginning to crack.
The disease had not extended further; the spur was
dead to the base and the bark on the branch was
fissured just below the insertion of the spur but there
was no definite depressed cankered area.

(1917). Conspicuous Monilia pustules present on the
dead spur.
One flower inoculated, on a branched spur bearing two
trusses.
Two styles brown to the base, two others brown for
about half their length.
The leaves round the truss wilting.
Monilia pustules present on some of the dead pedicels.
The infected half of the spur was dead to the base, a
distance of 3 cm.; the rest of the spur, including the
other truss, was alive.

10 (1917). Monilia pustules present on the dead spur.

5,

bo pS bo

—_
=

(Sy ey ey

=

One flower inoculated, on a branched spur bearing two
trusses.

Stamens of inoculated flower collapsed.

Truss wilting.

Infected half of the spur dead to the base with a canker
two-thirds round the main axis of the spur; the rest of
the spur was alive.

. 10 (1917). Monilia pustules present on the dead portion.

One flower inoculated, on a spur 5 em. in length.
Flower brown to the base of the pedicel.

Leaves round the base of the inflorescence drooping.
Truss quite withered, leaves brown.

Spur dead for a distance of 5 cm.

194. A Blossom Wilt and Canker of Apple Trees

Nov. 28. The spur bore several Monilia pustules with chains of
conidia; the latter were not pulverulent and evidently
immature.

Jan. 10 (1917). Numerous conspicuous pustules present along
the whole length of the spur; conidia more or less
pulverulent (a large number becoming free on mounting
particles of the pustules in water).

Although the conidia were obtained when the
temperature of the air was below 0° C. many of them
germinated within 48 hours, in a drop of water kept
at ‘6° to 8°'C.

(f) May 5. One flower inoculated of an unbranched spur 2 cm. in
length.
,, 16. Flower discoloured to the base of the pedicel.
, 18. Leaves at the base of the truss drooping.

June 2. Bark on the branch cracking immediately above and
below the insertion of the spur; a slightly depressed
area extended one-quarter round the branch.

.. 26.. Canker now extended half round the branch.

Oct. 19. The canker had not extended any further round the
branch; it was however quite well defined but at this
date bore no Monilia pustules.

Dec. 1. Immature pustules (conidia not pulverulent) present
along the whole length of the spur and on the canker.

Jan. 10 (1917). Pustules more conspicuous: conidia more or less
pulverulent; many germinated in distilled water within
48 hours at 6° to 8° C.

A flower on each of three other trusses was also inoculated; the
three flowers died but became detached about May 20 and the disease
did not extend into the axis of the inflorescence. The inoculation of
three flowers on May 10 and five on May 11 did not produce any wilting
of the trusses and it may be that in these cases the flowers were too old
for successful infection, although at the time of inoculation the stigmas
appeared receptive and showed no discoloration. These results corre-
spond to those obtained in the greenhouse where, as already shown, the
inoculations were most successful on those flowers which had recently
expanded.

It will be observed that in every case where inoculation with the
fungus was followed by the death of the truss, Wonilia pustules appeared
on the cankers and dead spurs during the succeeding winter. Usually

H: WorMALD 195

they began to burst through the bark during December and many well-
developed pustules with viable pulverulent conidia were to be seen by
the middle of January.

Although it has been noticed that in some cases spurs showing the
typical symptoms of the Blossom Wilt in the summer failed to develop
Momnilia pustules in the winter, this condition is exceptional, and where
cankers have been formed these invariably have produced the conidial
stage of the fungus. The results of the artificial inoculations too
demonstrate that a spur showing the wilt condition in summer is almost
certain to produce pustules of conidia before the following spring. The
danger of allowing such spurs and cankers to remain on the trees until
the next flowering season is obvious.

(c) Inoculation of Twigs through Wounds.

Twigs on trees of the Newton Wonder variety, which is known to be
susceptible, were inoculated with mycelium from a plate culture of
the same strain which produced wilting of the blossom in the inoculations
made on the Warner’s King variety. A A-shaped cut was made through
the bark, the triangular portion turned back, and agar bearing vigorously
growing mycelium was placed between the wood and the bark; the
latter was then gently pressed back and the wound covered with sterile
tinfoil which was secured in place by means of raffia. Three wounds
were treated in this way and three others, as controls, were not inocu-
lated. The inoculations were made on May 30, that is, at the time when
cankers were in process of development on trees infected through the
flowers. When examined some months later all the wounds were
covered with callus and no trace of canker was to be found on any of
them.

The result suggests that the fungus does not readily (if at all) produce
cankers by infection through wounds on the branches, and agrees with
observations in the open where, so far as my own experience goes, a
canker produced on an apple tree by this Monilia invariably originates
in a spur that has been infected through the flowers.

(d) Inoculation Experiments on the Fruit.

It has already been pointed out that the Monilia which causes the
Blossom Wilt may occur on the young apples. It will also grow readily
on apples approaching maturity and on ripe apples after picking, as
artificial inoculations have proved, although under natural conditions
instances of its occurrence on the mature fruit appear to be rare for

196 A Blossom Wilt and Canker of Apple Trees

the writer has not yet met with such cases eyen on trees seriously
affected with Blossom Wilt.

Apples of the varieties Warner’s King, Newton Wonder and Bramley’s
Seedling were inoculated on August 10 and 11 by placing mycelium from
a plate culture in wounds made with a sterile scalpel. Unfortunately
the apples became attacked by ants and all fell to the ground after a
few days; the experiments however continued long enough to show
that the fungus rapidly produced a brown rot appearing within four
days as a discoloured area 1-5 to 3 cm. in diameter round each point
of inoculation.

The experiments were then continued on apples of the Bramley’s
Seedling variety, which were brought into the laboratory and inoculated
with a strain isolated from a canker. The first inoculations of this
series were made on Aug. 17 and the experiments were repeated at
intervals throughout September and October. For comparison, other
apples were inoculated with Monilia fructigena, using a strain obtained
from a plum, and with a hyaline form of M. cinerea also isolated from
a plum. In some cases M. fructigena and the canker strain were placed
in wounds made on opposite sides of the same apple, while in others
the hyaline strain of M. cinerea and the canker form were grown on
the same apple. Under these conditions it was found that each of the
three forms produced a rot which extended approximately at the same
rate for all, i.e. 2-5 to 3-5 from the point of inoculation in seven days.
The canker form of Monilia however seldom produced pustules of
conidia; on some of the fruit a few scattered tufts of white barren
hyphae developed but on others no pustules were produced. On the
other hand the strain of M. fructigena used in these experiments freely
produced large yellow pustules more or less in concentric circles. The
hyaline form of M. cinerea also developed conidia readily but on smaller
erey pustules which were usually fairly numerous.

The skin of those apples inoculated with the canker Monilia rapidly
assumed a dark brown shade over the affected area, which gradually
became black. This nigrescence was to be detected towards the centre
of the discoloured area about a week after inoculation; it gradually
extended over the surface until the whole was black. The other two
strains also produced some blackening particularly in the later experi-
ments (i.e. on the more mature fruit) but not so readily. These differ-
ences were most striking in those cases where two forms were growing
together on the same apple by inoculations at opposite sides. For
example, two apples were inoculated on Aug. 17 and two others on

H. WorMALD 197

Aug. 21 by placing mycelium from an agar plate culture in wounds on
opposite sides using the canker Monilia on one side and M. fructigena
on the other. The result was the same for all four; that half of the
apple infected with M. fructigena produced numerous large yellow
pustules and remained brown, while the side infected with the canker
form became quite black in from three to four weeks and developed no
pustules at all or from one to three minute barren tufts of hyphae.
The side inoculated with M. fructigena became shrunken at a much
more rapid rate than the other.

Similar experiments were performed about the same time using the
canker-producing Monilia and the hyaline M. cinerea strain. Again
that side of the apple infected with the former soon became black and
bore few or no pustules, while the opposite side produced, as a rule,
numerous greyish pustules and remained brown for some weeks, becoming
however gradually darker until it was almost black; as before the
shrinking of the skin was most pronounced on the pustular side.

Later other apples of the same variety were inoculated using the
same three strains of Monilia but infecting each apple with but one of
the three. The results conformed with those obtained previously ;
those apples infected with WM. fructigena became almost covered with
large yellow pustules often becoming confluent, those with the hyaline
strain of M. cinerea produced smaller greyish pustules, while those with
the canker form remained sterile or produced a few scattered tufts of
aerial mycelium usually sterile.

The apples used in these experiments were such as showed no apparent
injury before inoculation; they were obtained from trees growing in the
College plantation and taken immediately to the laboratory and inocu-
lated. As the crop was picked during the first week of October sub-
sequent inoculations were made on apples (of the same variety) from
the fruit-storage shed. On such fruit the results were practically as
before except that the barren hyphal tufts produced by the canker-
strain were rather more numerous than in previous experiments. This
was probably due to small abrasions caused during the operations of
picking and storing or to minute cracks produced in the skin on drying,
thus allowing hyphae to grow out into the air. But even then the
difference between these results and those obtained with the other two
forms of Monilia was still conspicuous.

Whether similar results are to be obtained with other varieties of
apples has not yet been determined but it appears evident that the
Monilia causing the blossom wilt and canker of apple trees produces

198 A Blossom Wilt and Canker of Apple Trees

conidia on the ripening fruit much less readily than M. fructigena, and
this probably accounts for its inability to establish itself on the mature
apples, for the chief sources of infection from this fungus (i.e. the
pustules on the cankers and spurs killed in the previous season) have
shed most of their conidia and are becoming desiccated at the time when
the apples are reaching maturity. As indicated earlier in this paper
the newly formed cankers do not produce conidia until long after the
fruit is picked.

The blackening of the skin of apples produced by the blossom wilt
fungus may also be caused by M. fructigena and by the hyaline grey
Monilha, particularly on the mature fruit, but the result takes place
much more gradually. Black apples with few or no pustules are
frequently found among stored fruit; this condition appears to be
brought about by WM. fructigena which has invariably been isolated by
the writer from such apples obtained from fruit growers in Kent.

Spinks 22) finds that a similar “ Black Rot” of cider apples is also to
be attributed to M. fructigena.

VI. ContTrot MEASURES.

It is evident that cutting away the dead spurs and cankers removes
the chief source of infection, and an experiment, carried out on the row
of trees of the Warner's King variety referred to in preceding pages,
has demonstrated that where this can be done thoroughly the results
are highly satisfactory.

Ten trees at one end of the row were carefully pruned by the writer
on June 15 and 17, 1915, and so far as could be seen at the time every
withered truss was cut away until all the dead discoloured (brown)
tissues of the spur were removed. In those cases where cankers had
developed on the branches the operation involved the removal of dead
bark and wood in those places, until a clean cut, showing healthy tissues
only was made. The trees were not treated in any other way and the
wounds made by the pruning knife were left exposed. In all over 220
dead spurs were removed from these trees. When these ten trees were
examined in the summer of 1916 it was found that five were quite free
from the disease, one tree had but one dead truss, three had two each,
and one had six dead trusses; a search on the last mentioned tree how-
ever revealed the fact that two dead spurs had been overlooked during
the pruning operations and these now bore a number of Monilia
pustules.

The rest of the trees (fourteen in number) in the same row had been

H. WorMALD 199

pruned in autumn in the usual way and no special instructions with
regard to the disease has been issued to the men entrusted with the
work. The consequence was that although a certain amount of “dead
wood” had been cut away a considerable number of spurs and cankers,
which subsequently produced the conidial stage of the fungus, remained
on the trees. In some cases a dead spur had been cut off close to the
branch but the diseased tissues of the canker round the base had been
left and later the fungus appeared there. In the spring of 1916 it was
found that every tree bore the fungus in its infectious (conidial) con-
dition, the number of dead spurs and cankers with pustules varying
from one to twenty-five per tree with an average of eight. The number
of wilting trusses on these trees in May 1916 was 159 and varied from
two to twenty-nine per tree; these numbers would probably have been
considerably higher but for the fact that the trees had produced that
year exceptionally few trusses of bloom. On several trees however
more than one-fifth of the trusses wilted and in one case more than
half of those present were killed.

Thus on those trees from which all dead spurs had been removed
(omitting the one on which two had been overlooked) the number of
wilted trusses was reduced, on the average, to less than one per tree,
the infection in these cases being doubtless due to air-borne conidia
from the infected trees in the rest of the row, while those trees on which
infected spurs were left had an average of eleven wilted trusses per
tree.

The ideal mode of treatment would be the removal of diseased
trusses immediately after they first show signs of wilting; this would
involve examining the trees three or four times at intervals of about a
week between one examination and the next, all wilting trusses being
removed and the spurs cut back until all brown bark and wood is
removed, the first of these operations to be done about fourteen days
after the earliest flowers open. This method would prevent the develop-
ment of cankers which are not only much more troublesome to remove
than the spurs themselves but it would prevent the girdling and death
of the small branches.

As such measures would usually be impracticable except in small
plantations of bush trees where the disease has not yet become rampant,
an alternative would be to prune off all diseased spurs and cut out the
eankers as occasion permitted during the summer. This work should
be done as early as possible for the dead trusses with their withered
brown leaves are easily distinguished from the healthy ones and clearly

200 A Blossom Wilt and Canker of Apple Trees

indicate where pruning is necessary. If the operation is left over until
the winter or spring very careful search is necessary to avoid overlooking
some of the dead spurs and cankers, although even in winter the diseased
spurs are often easily recognised from the fact that at leaf-fall the leaves
and remains of flowers on such spurs still remain on the trees, and the
pedicels and petioles may still be found on them in the spring.

When, owing to difficulty in obtaining the necessary labour, the
removal of the spurs and cankers cannot be carried out before leaf-fall
it may be done any time during the winter, but it is imperative that all
diseased parts be cut off before any of the flowers begin to open.

It is important to emphasise the fact that the operation whenever
carried out must involve the removal of all brown and dead wood and
bark. Thus it is not sufficient merely to break off the withered trusses.
In one experiment six withered trusses were broken off from their
spurs early in June 1915, at the base of that season’s growth, the spurs
being then labelled for future reference; in April 1916 four of these
bore Monilia pustules.

As already shown the disease occasionally appears on the young
apples causing them to become dry and withered, and such “ mummified ”
apples may be retained on the trees until the following year. Since it
is impossible to distinguish, without close examination, between this
form of Brown Rot and Monilia fructigena when occurring in this way.
it is necessary that all “mummies” be picked and destroyed during the
winter, especially as Monilia fructigena itself is the cause of a serious
fruit rot.

The cutting out of all affected parts before the blossoms open is the
only treatment that can be recommended with confidence until further
investigation is carried out. Owners of large standard Lord Derby
apple trees have in some cases found, however, that to cut out all the
dead trusses (which in serious attacks often number some hundreds per
tree) is not practicable especially when skilled labour is difficult to
obtain'. In those cases where the disease is very severe, “ top-grafting ”
with a less susceptible varietv is to be recommended.

Whether spraying is of value in controlling the disease is at present
uncertain. As infection takes place through the open flower the use

1 Our Sussex correspondent writes: ‘To cut off and burn the millions of diseased
spurs and shoots on apples and plums is quite impracticable here.”

‘‘ Miilions” is probably no exaggeration in the case of Jarge orchards, for the writer has

counted over 130 wilted trusses (or about one-third of the whole) on a small bush tree, and
has seen large standard trees where the proportion of dead to living trusses was even

higher.

H. WorMALD 20

of a “cover-spray ” to protect the stigmatic surface is out of the question,
therefore a wash to be effective must be applied before the flowers open
and must be capable of destroying the powdery conidial stage, or at
least must prevent the conidia from falling during the period through-
out which the flowers are open and receptive.

Experiments have been tried in the open with the Lime-sulphur
Wash, which is frequently recommended for cases of Brown Rot, but
no favourable results have been obtained. ‘‘ Winter-washing” with
Lime-sulphur failed to check a serious outbreak of the disease in the
following spring. Facilities for testing the value of Lime-sulphur as
a summer spray were kindly offered by Mr P. Manwaring! of Hors-
monden, Kent, who permitted a plantation of Lord Derby trees (four
rows each with thirty-two trees), which had had a severe attack of the
Blossom Wilt in the previous year, to be used for experimental purposes.
Three rows were thoroughly sprayed with Lime-sulphur at ‘summer
strength” (s.c. 1-01) immediately before the flowers opened, the fourth
row remaining untreated and kept as control. In May all the rows
showed a serious attack of the Blossom Wilt and no difference in intensity
could be detected between the unsprayed row and the rest; many of
the latter had 50 °% or more of the trusses killed by the fungus, and the
numerous Monalia pustules on the dead spurs were apparently uninjured
by the spray fluid.

This result is doubtless due to the fact that such a liquid as the
Lime-sulphur solution is unable not only to penetrate but to adhere to
the powdery pustules. Experiments carried out in the laboratory
showed that Lime-sulphur solution even when applied in the form of a
very fine spray with an atomizer immediately ran off from the pustules
which appeared totally unaffected by the treatment.

Bordeaux Mixture applied similarly with an atomizer adhered a
little more readily but the majority of the pustules were not covered
by the spray.

An Ammonium Sulphide solution containing soft-soap, as recom-
mended by Dr Eyre and Mr Salmon? for use on the conidial stage of
the Hrysiphaceae, was also tried in the same way. This wash did wet
the pustules which in consequence became brown on the surface and
lost their pulverulent appearance. The pustules themselves were not

1 TI take this opportunity of thanking Mr Manwaring for the facilities offered for
investigating the disease in his plantations.

* Eyre, J. Vargas and Salmon, E. 8S. The Fungicidal Properties of Certain Spray
Fluids. Journ. Agric. Science, Vol. vit. pp. 473-507.

202. A Blossom Wilt and Canker of Apple Trees

killed and conidia placed in hanging drops germinated readily. It
would seem however that the surface layers were killed and the question
is whether this would be sufficient to prevent the fall of the conidia
during the critical period when the flowers are open if the spraying
were done as late as possible but before the flowers expanded. It is
proposed to carry out experiments in the open to test this point.

In conelusion I desire to thank Prof. L. R. Jones (Wisconsin),
Dr G. B. Posey (Oregon), Mr McCubbin, M.A. (Ontario), and Dr Quanjer
(Holland), who have kindly sent specimens of mummified fruit or
cultures from abroad, also Dr Pethybridge and Mr J. M. Hector, B.Sc.,
who sent mummified fruit and diseased branches. I am indebted also
to Mr Salmon (head of the Mycological Department at Wye College)
whose advice and criticisms throughout the investigation have been
invaluable.

SUMMARY.

1. A “Blossom Wilt and Canker” of apple trees, produced by a
species of Monilia, is causing great loss to fruit growers in the south-
east of England.

2. Infection takes place through the open flowers; the fungus
invades the tissues of the flowering spur, thus killing the inflorescence
and the leaves of the spur; the disease may reach the branch and
produce a canker.

3. Spurs killed during the summer, together with the accompanying
cankers, produce pustules of conidia during the following winter and
spring; these conidia, falling on the flowers as they open, give rise to a
new outbreak of the blossom-wilt disease.

4. When a canker has shed its crop of conidia it becomes covered
with callus which eventually heals the lesion.

5. Inoculation of apple flowers with conidia from pure cultures of
the fungus resulted in the death of the inflorescences and the spurs;
in some cases cankers were produced. Conidia-bearing pustules of the
organism appeared on these dead spurs and cankers during the following
winter.

6. The causal organism is a grey Monalia easily distinguished from
M. fructigena; at present it is to be referred to Monilia cinerea Bon.

7. On culture media the habit of the fungus is different from that
of the grey Monilia (also referred to M. cinerea by American workers)
which is commonly found in North America.

H. WorMALD 203

8. The disease may be kept in check by cutting out all dead spurs

and cankers before the flowers open; on the first appearance of the

disease all wilted trusses and dead spurs should be promptly removed.
Spraying operations can be efficacious only when they kill the conidial
pustules or prevent them from shedding their conidia during the
flowering period.

BIBLIOGRAPHY.

ApERHOLD, R. Zwei gefihrliche Erkrankungsfalle unseres Kernobstes.
Prosk. Obstbau-Ztg. Jahrg. v. No. 3, 1900, pp. 39-42.

—— Eine dem “‘fireblight” Amerikas fusserlich ahnliche Krankheit des
Apfelbaumes. Cent. f. Bakt. 2 Abt. Bd. vr. 1900, pp. 628-629.

—— Uber eine vermuthliche zu Monilia fructigena Pers. gehorige Sclerotinia.
Ber. Deutsch. Bot. Gesells. 22, 1904, pp. 262-266.

u. Runianp, W. Zur Kenntnis der Obstbaum-Sklerotinien. Arbeit.
biol. Abt. Land und Forst. Kaiserl. Gesundheitsamte, 4, 1905, pp. 427-442.

BarrraM, H. E. A Study of the Brown Rot fungus in Northern Vermont.
Phytopathology, Vol. v1. pp. 71-78, Feb. 1916.

Broz, O. Die Moniliagefahr. Separatabdruck aus “ Der Obstziichter,” Nr. 7,
1913.

ConEL, J. L. A study of the brown-rot fungus in the vicinity of Champaign
and Urbana Illinois. Phytopath. 1v. pp. 93-101, 1914.

Eriksson, J. Zur Kenntnis der durch Monilia-Pilze hervorgerufenen Bliiten-
und Zweigdiirre unserer Obstbiume. Mycol. Centralblatt, Bd. 1.
pp. 65-78, 1913.

Ewert, R. Verschiedene Uberwinterung der Monilien des Kern- und Stein-
obstes und ihre biologische Bedeutung. Zeitsch. f. Pflanzenkr. Bd. xx11.
pp. 65-86, 1912.

Frank, B. und Kriicer, F. Uber die gegenwirtig herrschende Monilia-
Epidemie der Obstbiitume. Landwirtschaftliche Jahrbiicher, Bd. xxvmt.
pp. 185-216, 1899.

Humpurey, J. E. On Monilia fructigena. Botanical Gazette, pp. 85-93,
1893.

JEHLE, R. A. The brown rot canker of the peach. Phytopath. Vol. m1.
pp. 105-110, 1913.

Massex, G. A Text-Book of Plant Diseases. London, 1903.

Martueny, W. A. A Comparison of the American Brown-Rot Fungus with
Sclerotinia fructigena and S. cinerea of Europe. Bot. Gaz. Vol. Lv1. pp. 418
—432, 1913.

Mtuier-THureav, H. Jie Monilienkrankheit oder Zweigdiirre der Kern-
obstbiume. Cent. f. Bakt. 2 Abt. Bd. vi. pp. 653-657, 1900.

Satmon, E. 8S. The “Brown Rot” Canker of Apple Trees. Journ. South-
Eastern Agric. Coll. Wye, No. 19, pp. 355-357, 1910.

—— A “Canker” of Apple Trees caused by the “Brown Rot” Fungus. Gard.
Chron. Vol. Xtvit *p. 327, May, 1910,

204 A Blossom Wilt and Canker of Apple Trees

Fi

_

g.

Fig.

Saumon, E.8. The “Brown Rot” Canker of the Apple. Journ. South-Eastern
Agric. Coll. Wye, No. 22, pp. 446-449, 1913.

The “Brown Rot” Canker of the Apple. Gard. Chron. Vol. Lyi. p. 85,
Aug. 1914.

SmirH, W. E. Diseased Apple-twigs. (Rept. of the Scientific Conmnittee of
the Roy. Hort. Soc.) Gard. Chron, Vol. xxv. (Third Series), p. 125, 1899.

Soraver, P. Die Schiiden der einheimischen Kulturpflanzen durch thierische
und pflanzliche Schmarotzers. Berlin, 1888, p. 235.

Spinks, G. T. A Black Rot of Apples. Ann. Rept. of the Agric. and Hort.
Research Sta. Long Ashton. Bristol, 1915, pp. 94-96.

Worontx, M. Uber Sclerotinia cinerea und Sclerotinia fructigena. Mém.
Acad. Imp. Sci. St Pétersbourg, vu. Series. Vol. x. No. 5. Phys. Math.
pp. 1-38. Oct. 1899.

DESCRIPTION OF PLATES XXIIl—XXIV.

1. Portion of cankered branch (var. Lord Derby) showing Monilia pustules; con-
dition as seen in winter. (Photographed April 8, 1916.)

2. Dead spur and canker bearing Monilia pustules, with neighbouring flowering
spurs; the wilted trusses below the canker were probably infected by conidia falling
from the spur and canker; condition as seen in summer. (June 4, 1915.)

. 3. Four trusses on the main stem of a young Lord Derby apple tree; the first and

the third (from above) were each inoculated, from a pure culture of the fungus, on

a single flower, Result fifteen days after inoculation—both inoculated trusses are
dead.

. 4. Asin Fig. 3 but four weeks later; the stem is cankered above and below the two

inoculated trusses.

ig. 5. The stem of the infected tree at the time that the photograph shown in Fig. 3

was taken; the four trusses there shown are to be seen immediately below the branch
on the right.

oe. 6. Asin Fig. 5 but taken four weeks later, i.e. on the same day as Fig. 4 was obtained.

The leaves on the branches above the canker are wilting.

. 7. A canker produced by a natural infection in 1914; it bore Monilia pustules in

1915, but when photographed in June 1916 was barren and was being covered over
by callus.

r. 8. The canker shown in Fig. 2 seen in transverse section (x 23). The development

of callus at the sides of the canker has already commenced at this stage.

(Figs. 1—4 and 7 are % natural size.)

CAMBRIDGE: PRINTED BY J. B. PEACE, M.A., AT THE UNIVERSITY PRESS.

THE ANNALS OF APPLIED BIOLOGY. VOL. III, NO. 4

PLATE XXII

THE ANNALS OF APPLIED BIOLOGY.

VOL. III,

NO.

PLATE XXIII

THE ANNALS OF APPLIED BIOLOGY. VOL. II

, NO. 4 PLATE XXIV

Fig. 8

be |

THE ANNALS OF
APPLIED © BIOLOGY

THE OFFICIAL ORGAN OF THE ASSOCIATION
OF ee Oe es
: 4 ae ive

r

" eahe
ie
/ /7/ @ Epvrep sy
’

E. E. GREEN, Way’s End, Camberley (late Government Entomologist, Ceylon)
| WITH THE ASSISTANCE OF
_ PROFESSOR B. T. P. BARKER, National Fruit and Cider Institute, Bristol
ie ee Dr S. E. CHANDLER, Imperial Institute, London
oy F. J. CHITTENDEN, Royal Fesieieal Society’s Gardens, Wisley
_ J. C. F. FRYER, Board of Agriculture and Fisheries, London

Proressor F. W. GAMBLE, The University, Birmingham

_PRorEsson PERCY GROOM, Imperial College of Science and Technology, London
Bie oe RS Dr A. D. IMMS, The University, Manchester
ee ae ~ __ Prorgsson R. NEWSTEAD, The University, Liverpool

"So eee _ Proressor J. H. PRIESTLEY, The University, Leeds

CAMBRIDGE UNIVERSITY PRESS
C. F. CLAY, Manacer
LONDON:. FETTER LANE, E.C.
EDINBURGH: 100, PRINCES STREET
5 EE SREY also
fee * H. K. LEWIS & CO., LTD., 136, GOWER STREET, LONDON, W.C. :
WILLIAM WESLEY & SON, 28, ESSEX STREET, LONDON, W.C
PARIS: LIBRAIRIE HACHETTE & CIE. :
CHICAGO: THE UNIVERSITY OF CHICAGO PRESS a
BOMBAY, CALCUTTA AND MADRAS: MACMILLAN & CO., LTD
TORONTO: J. M. DENT & SONS, LTD.
TOKYO : THE MARUZEN-KABUSHIKI-KAISHA

Price Seven Shillings and Sixpence net

The Association of Economic Biologists _
President
Proressor NEWSTEAD, F.RS.

Vice-Presidents

ProFEssor CARPENTER, M.R.I.A.

ProFressor HICKSON, F.R.S.

R. STEWART MacDOUGALL, D.Sc.

A. E. SHIPLEY, D.Sc., F.R.S. WR

Council
Pror. B. T. P. BARKER _ A. D. IMMS, D.Sc.
S. E. CHANDLER, D.Sc. Pror. J. H. PRIESTLEY
F. J. CHITTENDEN A. G. L. ROGERS
E. E. GREEN E. S. SALMON

Pror. P. GROOM, D.Sc.

Hon. Treasurer
J. C. F. FRYER, Esga.,
Craven House, 0
Northumberland Avenue ed

Hon. Secretary
Proressorn PERCY GROOM,

Imperial College of Science and Technology,
South Kensington

CONTENTS OF Vou. III, No. 1 | ne
“pace

1. The Fig ‘Canker,’ caused by Phoma Cinerescens Sacc. By E. §.
Satmon and H. Wormatp. (With Plates I and II, ie 1 Text-

figure.) . ; : ‘ ; : ; : : ¢ pat |
2. Shrinkage, Swelling and Warping of Cross-grained Woods: No. 1, Yang os oe
(Dipterocarpus sp.). By Percy Groom. (With a Diagram.) 13 ©
3. The Dalby Profile Recorder. By W.E. Darpy, MInst..B., PRS.
(With 7 Text-figures.) . : : : j TNE Pon ga a ae 39 “

4. The Action of sien Worms, By the Rev. Hinpertc FRIEND, — ee
ERM Be. Lie ed eR Oe ae

5. The Food of Slugs and the Pee of Anoeloeenianies, “By e 3
ProFessor A. RAILuieT. . saint Yee my bs

CAMBRIDGE UNIVERSITY PRESS

MIMICRY IN BUTTERFLIES. By R. C. Puynert,
F.R.S., Arthur Balfour Professor of Genetics in the University of
Cambridge. Buckram. With 16 plates. Royal 8vo. 15s net,

‘Though much has been written on the subject of late years, there has not heretofore
been any single comprehensive book dealing with it in the light of the latest knowledge.
This lack Professor Punnett has admirably supplied. Primarily he brings to bear on the
question the arguments of Mendelism....Professor Punnett collates the facts from various
sources, and presents them clearly and in due proportion. His explanations are helped by
excellent coloured plates illustrating the most important examples of mimicry from different
parts of the world....The whole subject is still beclouded. We have only the beginnings of
knowledge. But at least in Professor Punnett’s treatise we seem to see the foreshadowing
of something like a foundation of solid fact in place of what has heretofore been a rather
airy structure of fascinating guesswork.” —T'imes

TICKS. A MONOGRAPH OF THE IXODOI-
DEA. By George H. F. Nurratt, M.D., Ph.D., Sc.D., F.R.S.,
Ceci WarBurTon, M.A., F.Z.S., W. F. Cooprr, B.A., F.Z.S., F.L.S.,
and L. E. Roxsrnson, A.R.C.Sc. (Lond.). Royal 8vo. Part I.
Argasidae. 5s net. Part II. Ixodidae. 12s net. Part III. The
Genus Haemaphysalis. 12s net.

BIBLIOGRAPHY OF THE IXODOIDEA

Part I by G. H. F. Nurtatr, L. E. Ropinson and W. F. Coopzr. (2004 titles.)
6s net. (1911.) Not sold separately from Part II of Text as above.
Part II by G. H. F. Nurratt and L. E. Rogrnson. (462 titles.) 4s 6d net. (1915.)
Not sold separately from Part III of Text as above.
N.B. These Bibliographies are an essential part of the work as they contain complete

references to authors cited in the text and the references are frequently accompanied by
notes in parentheses. ;

FUNGOID AND INSECT PESTS OF THE
FARM. By F. R. Peruersrivgs, M.A., Biological Adviser, School
of Agriculture, Cambridge. Crown 8vo. 4s net.

This book has been written for those who wish to acquire some practical knowledge of
farm and garden pests. It does not aim at dealing with all the numerous diseases which
affect crops but rather at giving an accurate account of some of the commoner forms.
ConTENTS: Part I. Introduction to Fungi—Potato Disease and allied Diseases—Finger

and Toe, and Wart Disease—Mildews—Ergot and Clover Sickness—Rusts—Smuts.
Part II. Introduction to Insects—Butterflies and Moths—Beetles—Flies—Aphides
and Sawflies—Eelworms.

TYPICAL FLIES. A Photographic Atlas-of Diptera,
including Aphaniptera. By E. K. Prarce, With 155 photographs.
Royal 8vo. Paper boards. 5s net.

“<The photographs are as good as any we have seen of this class of insect—a peculiarly
difficult class to represent pictorially in any natural manner; the venation of the wings is
well brought out wherever the banding or colouring of the wings does not obscure it, and
there are excellent notes as to the habitat, larval habits, and so on under the pictures.
The diversity of appearance is well brought out; to the student the venation systems
depicted will be helpful, and to the beginner the variety of habitat and habit will be
distinctly stimulating.” —Nature

Cambridge University Press, Fetter Lane, London —
C. F. Cray, Manager

THE ANNALS OF APPLIED BIOLOGY —

In the Annals of Applied Biology papers will be published concerning the |
economic aspects of agriculture, botany, plant-breeding, plant-pathology, mycology,
horticulture, forestry, helminthology, entomology, veterinary zoology, medical 2
zoology, and marine zoology. :

Contributors to The Annals are asked to send type-written articles if possible, rs ;
and as far as possible to make illustrations in a form suitable for reproduction :
as text-figures. Contributors will receive free fifty copies of articles only. —

Editorial communications should be addressed to E. E. Green, at PY S End,”
Camberley, Surrey.

Terms of subscription.

Members of the Association will receive The Annals free. To others, the |
annual subscription price, including postage to any part of the world, for a single —
copy of each of the four parts making up the annual volume, is 25s. net; single —
copies 7s. 6d. net each. Subscriptions for Zhe Annals are payable in maya and
should be sent to Mr C. F. Cuay, Cambridge University Press, Fetter ae
London, E.C., either direct or through any bookseller. .

ae

Volumes I and II now ready. Price, in four Pere, paper covers, 25s. net each ; ot
bound in Buckram, 30s. net each.

The publishers have appointed the University of Chicago Press agents for ee ,
sale of The Annals of Applied Biology in the United States of America and have —

authorised them to charge the following prices: annual subscription, post Aree, $6. 00, , a
single copies, $2.00.

Claims for missing numbers should be made within the month flowing that a
_ of regular publication.

CAMBRIDGE: PRINTED BY J. B. PEACE, M.A., AT THE UNIVERSITY PRESS.

' THE ANNALS OF
APPLIED BIOLOGY

THE OFFICIAL ORGAN OF THE ASSOCIATION
OF ECONOMIC BIOLOGISTS

7 EDITED BY

_ E. E. GREEN, Way’s End, Camberley (late Government Entomologist, Ceylon)
WITH THE ASSISTANCE OF

Proressor B. T. P. BARKER, National Fruit and Cider Institute, Bristol
Dr S$. E. CHANDLER, Imperial Institute, London
F. J. CHITTENDEN, Royal Horticultural Society’s Gardens, Wisley
J. C. F. FRYER, Board of Agriculture and Fisheries, London

PRorFEssor F, Ww. GAMBLE, The University, Birmingham

Proresson PERCY GROOM, Imperial College of Science and Technology, Londor
: Dr A. D. IMMS, The University, Manchester

Proressor R. NEWSTEAD, The University, Liverpool
Proressor J. H. PRIESTLEY, The University, Leeds

CAMBRIDGE UNIVERSITY PRESS
C. F. CLAY, Manacer
LONDON: FETTER LANE, E.C.
EDINBURGH: I00, PRINCES STREET
also
“ H. K, LEWIS & CO., LTD., 136, GOWER STREET, LONDON, W.C.
WILLIAM WESLEY & SON, 28, ESSEX STREET, LONDON, W.C.
PARIS: LIBRAIRIE HACHETTE & CIE.

CHICAGO: THE UNIVERSITY OF CHICAGO PRESS
BOMBAY, CALCUTTA AND MADRAS: MACMILLAN & CO., LTD.
TORONTO: J. M. DENT & SONS, LTD.

TOKYO : THE MARUZEN-KABUSHIKI-KAISHA

Te a line Oh Atlan cee. wos

See

ree
ons

ae i i oe %

The Association of Economic Biologists

President
Proressor NEWSTEAD, F.R.S.

Vice-Presidents

Proressor CARPENTER, M.R.LA.
Proressorn HICKSON, F.B.S.

R. STEWART MacDOUGALL, D.Sc.
A. E. SHIPLEY, D.Sc., F.R.S.

Council
Pror. B. T. P. BARKER A. D. IMMS, D.Sc.
S. E. CHANDLER, D.Sc. Pror. J. H. PRIESTLEY
F, J. CHITTENDEN A. G. L. ROGERS.
E. E. GREEN EK. S. SALMON

Pror. P. GROOM, D.Sc.

. Hon. Treasurer
J. C.F. FRYER, Esq.,
Craven House, .
Northumberland Avenue i

Hon. Secretary

ProFEssoR PERCY GROOM,
Imperial College of Science and Technology, 2
South Kensington ie

CONTENTS OF Vou. III, Nos. 2 ann 3

A Bacterial Spot of Citrus. By Eruen M. tee D.Se., PLS. ca ae

(With Plates I1I—XIII.) Sage Bela 63
2. Report on a Trial of Tarred Felt “Discs” for Forni Cabbages a -
Cauliflowers from attacks of the es es By J. ns ii
Wapswortnh. (With Plate XIV.) . . oete g

3. On the Resistance to Fungicides shown by the Hop- hlden iSuhierlien :
humuli (D.C.) Burr.) in different stages of pea aas ed BB.)
Satmon. (With Plate XV.) . ba estate Rais ‘be

4. Observations on the Larval and Pupal Stages of doko ey oa
Linnaeus. By Georce H. Forp, M.Sc. (Vict.). eee Plates: x
XVI and XVII and 1 Text-figure.) Pie et aes OT ae

5. On the Biology and Economic significance of Tipula ners By | ey.
Joun Renniz, D.Sc., F.R.S.E. Part Il. Hatching, Growth and.
Habits of the Larva. whee Plates XVIJI—XX and 3 Text-
FOUIES ee ee ee Pat aoe ta Cla

6. Note on Attacks of Phyletreta clans on dong Conn EES TN eae

_ CAMBRIDGE UNIVERSITY PRESS

MIMICRY IN BUTTERFLIES. By R. C. Punyert,
F.R.S., Arthur Balfour Professor of Genetics in the University of
Cambridge. Buckram. With 16 plates. Royal 8vo. 15s net.

‘*Though much has been written on the subject of late years, there has not heretofore
been any single comprehensive book dealing with it in the light of the latest knowledge.
This lack Professor Punnett has admirably supplied. Primarily he brings to bear on the
question the arguments of Mendelism....Professor Punnett collates the facts from various
sources, and presents them clearly and in due proportion. His explanations are helped by
excellent coloured plates illustrating the most important examples of mimicry from different
parts of the world....The whole subject is still beclouded. We have only the beginnings of
knowledge. But at least in Professor Punnett’s treatise we seem to see the foreshadowing
of something like a foundation of solid fact in place of what has heretofore been a rather
airy structure of fascinating guesswork.’’—T'imes

TICKS. A MONOGRAPH OF THE IXODOI-
DEA. By Groren H. F. Nurrawy, M.D., Ph.D., Sc.D., F.BS.,
Ceci, WARBURTON, M.A., F.Z.S., W. F. Coopsr, B.A., F.Z.S., F.L.S.,
and L. E. Rosiyson, A.R.C.Sc. (Lond.). Royal 8vo. Part I.
Argasidae. 5snet. Part II. Ixodidae. 12s net. Part III. The
Genus Haemaphysalis. 12s net.

BIBLIOGRAPHY OF THE IXODOIDEA

Part I by G. H. F. Nurratr, L. E. Rozsrnson and W. F. Cooprr. (2004 titles.)
6s net. (1911.) Not sold separately from Part II of Text as above.

Part II by G. H. F. Nurratt and L. E. Ropinson. (462 titles.) 4s 6d net. (1915.)
Not sold separately from Part III of Text as above. ,

N.B. These Bibliographies are an essential part of the work as they contain complete
references to authors cited in the text and the references are frequently accompanied by
notes in parentheses.

FUNGOID AND INSECT PESTS OF THE
_ FARM. By F. BR. Pernersripes, M.A., Biological Adviser, School
of Agriculture, Cambridge. Crown 8vo. 4s net.

‘This book has been written for those who wish to acquire some practical knowledge of
farm and garden pests. It does not aim at dealing with all the numerous diseases which
affect crops but rather at giving an accurate account of some of the commoner forms.

Contents: Part I. Introduction to Fungi—Potato Disease and allied Diseases—Finger
and Toe, and Wart Disease—Mildews—Ergot and Clover Sickness—Rusts—Smuts.
Part II. Introduction to Insects—Butterflies and Moths—Beetles—Flies—Aphides
and Sawflies—Eelworms.

TYPICAL FLIES. A Photographic Atlas of Diptera,
including Aphaniptera. By E. K. Pearce. With 155 photographs.
Royal 8vo. Paper boards. 5s net.

“The photographs are as good as any we have seen of this class of insect—a peculiarly
difficult class to represent pictorially in any natural manner; the venation of the wings is
well brought out wherever the banding or colouring of the wings does not obscure it, and
there are excellent notes as to the habitat, larval habits, and so on under the pictures.
The diversity of appearance is well brought out; to the student the venation systems
depicted will be helpful, and to the beginner the variety of habitat and habit will be
distinctly stimulating.” —Nature

Cambridge University Press, Fetter Lane, London
C. F. Ciay, Manager

economic aspects of agriculture, botany, plant- -breeding, alee beni niycology, j
horticulture, forestry, helminthology, entomology, veterinary zoology, medion
zoology, and marine zoology.

Contributors to The Annals are asked to send type-written articles if pasteles
and as far as possible to make illustrations in a form suitable for epi.
as text- Sela Contributors will receive free fifty copies of res saege

Camberley, Surrey.

: Terms of subscription.

Members of the Association will receive The Annals free. To others, the
annual subscription price, including postage to any part of the world, for a ies
copy of each of the four parts making up the annual volume, is 25s. net; single ©
copies 7s. 6d. net each. Subscriptions for The Annals are payable in advan and
should be sent to Mr ©. F. Cray, Cambridge University vies Fetter Lane,
London, E.C., either direct or through pie bookseller, ees

Volumes I and II now ready. a in four Pes paper covers, 25s, net each ; :
bound in Buckram, 30s. net each. ote ae

The publishers have appointed the University of Chicago Press agents for the
sale of The Annals of Applied Biology in the United States of America and have
authorised them to charge the following prices: annual biroeghr post Wace. re 00, ‘
single copies, $2.00.

Claims for missing numbers should be made within the month following t that an
of regular publication.

CAMBRIDGE: PRINTED BY J. B. PEACE, M.A., AT THE UNIVERSITY PRESS,

Vol. III, No. eee April, 1917

E THE ANNALS OF
APPLIED BIOLOGY

Tice OFFICIAL ORGAN OF THE ASSOCIATION
OF ECONOMIC. BIOLOGISTS ’

EDITED BY

£, E. GREEN, Way’s End, Camberley (late Government Entomologist, Ceylon)
WITH THE ASSISTANCE OF

Proressor B. T. P. BARKER, National Fruit and Cider Institute, Bristol
Dr S. E. CHANDLER, Imperial Institute, London
F. J. CHITTENDEN, Royal Horticultural Society’s Gardens, Wisley —
J. C. F. FRYER, Board of Agriculture and Fisheries, London

Prorrssor F. W. GAMBLE, The University, Birmingham
e _ PROFESSOR PERCY GROOM, Imperial College of Science and Technology, London
Dr A. D. IMMS, The University, Manchester
Proresson R. NEWSTEAD, The University, Liverpool
Proressor J. H. PRIESTLEY, The University, Leeds

CAMBRIDGE UNIVERSITY PRESS
Cc, F. CLAY, MAanaGER
LONDON: FETTER LANE, E.C. 4
EDINBURGH: 100, PRINCES STREET
also
H. K. LEWIS & CO., LTD., 136, GOWER STREET, LONDON, W.C. 1
WILLIAM WESLEY & SON, 28, ESSEX STREET, LONDON, W.C. 2
PARIS: LIBRAIRIE HACHETTE & CIE.

CHICAGO: THE UNIVERSITY OF CHICAGO PRESS
‘BOMBAY, CALCUTTA AND MADRAS: MACMILLAN & CO., LTD.
TORONTO: J. M. DENT & SONS, LTD.

TOKYO : THE MARUZEN-KABUSHIKI-KAISHA

| Price Seven Shillings and Sixpence net

President Sr bah
ProFessoR NEWSTEAD, F.RS.

Vice-Presidents

ProFEssoR CARPENTER, M.R.LA.
Proressor HICKSON, F.R.S.

R. STEWART MacDOUGALL, D.Sc.
A. E. SHIPLEY, D.Sc., F.R.S.

| Council 2 Bat: Ea?
Pror. B. T. P. BARKER A. D. IMMS, DSo. ¢ Eas ie
S. E. CHANDLER, D.Sc. Pror. J. H. PRIESTLEY _
F. J. CHITTENDEN AL GL: ROGERS © .

E. E. GREEN E. S. SALMON ©
Pror. P. GROOM, D.Sc. : a.

Hon. Treasurer : “4

J. C. F. FRYER, Esq, «>
Craven House, ea
Northumberland Avenue a

Oy RE soe
Hon. Secretary

ProrEessor PERCY GROOM,
Imperial College of Science and Tec olc
South Kensington

CONTENTS OF VOL. IIL, ‘No. 4

1. Accessory Wetting Substances with Special Reference to Poratiin ne ; :
Emulsions. By A. H. Less, M.A.. IRN SME Oe ck,

2. Empusa Muscae versus Musca domestica Es ‘By H. T. Gissow. Bey
(With Plate XXTI.)

3. A Blossom Wilt and Canker of Apple Trees. “By H. Won,
M.Sc. (Lond.), A.R.O.Sc. (With Plates RH XXIV?

Pas

2 CAMBRIDGE UNIVERSITY PRESS

\

MIMICRY IN BUTTERFLIES. By R.C. Puynzrt, :

R.S., Arthur Balfour Professor of Genetics in the University of
ee cbdee Buckram. With 16 plates. Royal 8vo. 15s net.

_ Though much has been written on the subject of late years, there has not heretofore
been any single comprehensive book dealing with it in the light of the latest knowledge.
This lack Professor Punnett has admirably supplied. Primarily he brings to bear on the
question the arguments of Mendelism....Professor Punnett collates the facts from various
sources, and presents them clearly and in due proportion. His explanations are helped by
excellent coloured plates illustrating the most important examples of mimicry from different
parts of the world....The whole subject is still beclouded. We have only the beginnings of
knowledge. But at least in Professor Punnett’s treatise we seem to sec the foreshadowing
of something like a foundation of solid fact in place of what has heretofore been a rather
airy structure of fascinating guesswork.” —T'imes

TICKS. A MONOGRAPH OF THE IXODOI-
DEA. By Groree H. F. Nourrauz, M.D., Ph.D., Sc.D., F.RS.,
Crom, Warsurton, M.A., F.Z.S., W. F. Cooper, B.A., F.Z.S., F.L.S.,
and L. E. Roxpryson, A.R.C.Sc. (Lond.). Royal 8vo. Part I.
Argasidae. 5snet. Part II. Ixodidae. 12s net. Part III. The
Genus Haemaphysalis. 12s net.

BIBLIOGRAPHY OF THE IXODOIDEA
Part I by G. H. F. Nurrarz, L. E. Ropinson and W. F. Cooper. (2004 titles.)
6s net. (1911.) Not sold separately from Part II of Text as above.
Part II by G. H. F. Nurrart and L. E. Ropinson. (462 titles.) 48 6d net. (1915.)
- Not sold separately from Part II of Text as above.

N.B. These Bibliographies are an essential part of the work as they contain complete
references to authors cited in the text and the references are frequently accompanied by

_ notes in parentheses.

FUNGOID AND INSECT PESTS OF THE
FARM. By F.R. Petuersrinas, M.A., Biological Adviser, School
of Agriculture, Cambridge. Crown 8vo. 4s net.

This book has been written for those who wish to acquire some practical knowledge of
farm and garden pests. It does not aim at dealing with all the numerous diseases which
affect crops but rather at giving an accurate account of some of the commoner forms.

Contents: Part I. Introduction to Fungi—Potato Disease and allied Diseases—Finger
and Toe, and Wart Disease—Mildews—Ergot and Clover Sickness—Rusts—Smuts.
Part If. Introduction to Insects—Butterflies and Moths—Beetles—Flies—Aphides
and Sawflies—Eelworms.

TYPICAL FLIES. A Photographic Atlas of Diptera,
including Aphaniptera. By E. K. Pearce. With 155 photographs.
Royal 8vo. Paper boards. 5s net.

“The photographs are as good as any we have seen of this class of insect—a peculiarly
difficult class to represent pictorially in any natural manner; the venation of the wings is
well brought out wherever the banding or colouring of the wings does not obscure it, and
there are excellent notes as to the habitat, larval habits, and so on under the pictures.
The diversity of appearance is well brought out; to the student the venation systems

' depicted will be helpful, and to the beginner the variety of habitat and habit will be
ara stimulating. ”—Nature

amb idee University Press, Fetter Lane, London
C. F. Cray, Manager

> ee - of

THE ANNALS OF APPLIED BIOLOGY

In the Annals of Applied Biology papers will be published concerning the —
economic aspects of agriculture, botany, plant-breeding, plant-pathology, mycology, —
horticulture, forestry, helminthology, entomology, veterinary zoology, eae é
- zoology, and marine zoology. |

Contributors to The Annals are asked to send type-written articles if possible, %
and as far as possible to make illustrations in a form suitable for reproduction —
as text-figures. Contributors will receive free fifty copies of articles only. —

Editorial communications should be addressed to E. E. Green, at Way’s End, a am

_ Camberley, Surrey.

Terms of subscription.

Members of the Association will receive The Annals free. To others, the |

annual subscription price, including postage to any part of the world, for a single — ' aa
copy of each of the four parts making up the annual volume, is 25s. net; single
copies 7s. 6d. net each. Subscriptions for The Annals are payable in advance and

should be sent to Mr C. F. Ciay, Cambridge University Press, Fetter Lane,
London, E.C. 4, either direct or through any bookseller.

Volumes I, II and III now ready. Price, in four parts, paper covers, 25s. net
each ; bound in Buckram, 30s. net each. oh

The publishers have appointed the University of Chicago Press agents for the
sale of The Annals of Applied Biology in the United States of America and have ©
authorised them to charge the following prices: pee subscription, post free, $6.00, _
single copies, $2.00.

Claims for missing numbers should be made within the month following that ae

of regular publication,

CAMBRIDGE: PRINTED BY J. B. PEACE, M.A., AT THE UNIVERSITY PRESS.

~

ar VA \ Pees f

y

ay
‘if ’ ‘ \ ‘
f,

See reboot
~ tare wenn ee

apogee a ve
Pon LEO

a P
Onna dake Snes paeeeneananetaen
eo ee

no aan opewetonn von og
aasesiseepnie basa

rent <

. i~>
ey er ne a ae fe
a Fee oe eres
TN teen ~