Digitized by the Internet Archive

in 2011 with funding from

LYRASIS members and Sloan Foundation

http://www.archive.org/details/tobaccosubstatio1936ande

s

43

Bulletin 386 . July, 1936

TOBACCO SUBSTATION AT WINDSOR
REPORT FOR 1935

P. J. Anderson, T. R. Swanback and O. E. Street

(HamxtrtituA

JVgrttuiiural Experiment Jiiatttm

5tew Simian

CONNECTICUT AGRICULTURAL EXPERIMENT STATION

BOARD OF CONTROL

His Excellency, Governor Wilbur L. Cross, ex-officio, President

Elijah Rogers, Vice-President Southington

William L. Slate, Director and Treasurer New Haven

Edward C. Schneider, Secretary Middletown

Joseph W. Alsop Avon

Charles G . Morris Newtown

Albert B. Plant Branford

Olcott F. Kins: South Windsor

Administration.

STAFF

William L. Slate, B.Sc, Director and Treasurer.
Miss L. M. Brautlecht, Bookkeeper and Librarian.
Miss Katherine M. Palmer, B.Litt., Editor.
G. E. Graham, In Charge of Buildings and Grounds.

Analytical
Chemistry.

E. M. Bailey, Ph.D., Chemist in Charge.

C. E. Shepard \

Owen L. Nolan

Harry J. Fisher, Ph.D. (-Assistant Chemists.

W. T. Mathis

David C. Walden, B.S. J

Miss Janetha Shepapd, General Assistant.

Chas. W. Soderberg, Laboratory Assistant.

V. L. Churchill, Sampling Agent.

Mrs. A. B. Vosrurgh, Secretary.

Biochemistry.
Botany.

H. B. Vickery, Ph.D., Biochemist in Charge.
George W. Pucher, Ph.D., Assistant Biochemist.

G. P. Clinton, Sc.D., Botanist in Charge.

E. M. Stoddard, B.S., Pomologist.

Miss Florence A. McCormick, Ph.D., Pathologist.

A. A. Dunlap, Ph.D., Assistant Mycologist

A. D. McDonnell, General Assistant.

Mrs. W. W. Kelsey, Secretary.

Entomology.

W. E. Britton, Ph.D., D.Sc, Entomologist in Charge, Stale Entomologist.

B. H. Walden, B.Agr. ..

M. P. Zappe, B.S.

Philip Garman, Ph.D. \Assistant Entomologists.

Roger B. Friend, Ph.D. [

Neely Turner, M.A. '

John T. Ashworth, Deputy in Charge of Gypsy Moth Control.

R. C. Botsford, Deputy in Charge of Mosquito Elimination.

J. P. Johnson, B.S., Deputy in Charge of Japanese Beetle Control.

Miss Helen A. Hulse 1 _ , .

Miss Betty Scoville (-Secretaries.

Forestry .

Walter O. Filley, Forester in Charge.

H. W. Hicock, M.F., Assistant Forester.

J. E. Riley, Jr., M.F., In Charge of Blister Bust Control.

Miss Pauline A. Merchant, Secretary.

Plant Breeding.

Donald F. Jones, Sc.D., Geneticist in Charge.
W. Ralph Singleton, Sc.D., Assistant Geneticist.
Lawrence C. Curtis, B.S., Assistant.

Soils.

M. F. Morgan, Ph.D., Agronomist in Charge.
H. G. M. Jacobson, M.S., Assistant Agronomist.
Herbert A. Lunt, Ph.D., Assistant in Forest Soils.
Dwight B. Downs, General Assistant.
Miss Geraldine Everett, Secretary.

Tobacco Substation
at Windsor.

Paul J. Anderson, Ph.D., Pathologist in Charge.

T. R. Swanback, M.S., Agronomist.

O. E. Street, Ph.D., Plant Physiologist.

C. E. Swanson. Laboratory Technician.

Miss Dorothy Lenard, Secretary.

Printing by The Peiper Press, Inc., Wallingford, Conn.

CONTENTS

Fertilizer experiments with single sources of nitrogen 546

Nitrate nitrogen and soil acidity production by nitrogenous fertilizers

on

Measurements of the Production of Nitrate Nitrogen 553

Seasonal Characteristics of the Period 554

Residual Nitrogen and Early Availability 554

Maximum Levels 565

Relative Availability in Relation to Crop Needs 567

New Materials 568

Rainfall in Relation to Nitrates 569

Speed of Recovery After Rains 572

Soil Reaction as Influenced by Nitrogenous Fertilizers 573

Seasonal Fluctuation in Relation to Nitrates and Rainfall 573

Fluctuations Caused by the Fertilizer Material 574

Conclusions 574

Further fertilizer experiments with cottonhull ashes 574

Further investigations on the use of fertilizer magnesia 578

Color of Ash 578

Annual Applications of Magnesian Lime 580

Magnesia Application Based on Microchemical Soil Tests 581

Anhydrous Magnesium Sulfate as a Source of Magnesia 583

Experiments to determine the best time to harvest havana seed tobacco 585

Tobacco insects in 1935 587

Prevalence of Insect Species in 1935 587

Insect Control Experiments 589

Tobacco Budworm 589

Potato Flea Beetle 590

Flea Beetle Resistant Tobacco 591

Tobacco Thrips 591

Wireworm Investigations 592

Tobacco diseases in 1935 593

Dead-blossom Leaf Spot 595

Pythium Stalk-rot of Transplants 597

Studies on Pole Rot. 1 600

Effect of shade cloth on atmospheric conditions 607

TOBACCO SUBSTATION AT WINDSOR
REPORT FOR 1935

P. J. Anderson, T. R. Swanback and 0. E. Street

JL his, the fourteenth annual report of the Tobacco Substation,
describes the progress of experiments conducted at Windsor on fertilizers,
cultural practices, diseases and insects of tobacco. It has never been
our policy in these bulletins to review every line of work every year.
Rather, we have taken up certain projects that have been under way
for several years or that have yielded some reliable conclusions. The
same system holds in the following pages.

The character of the weather during the growing and curing season
has an important bearing on results for any one year. This is particularly
true in the case of fertilizer experiments, although it applies somewhat
to all field experiments. The point was well illustrated in the season
of 1935 at Windsor. Besides wind storms that blew the tobacco over,
there were heavy leaching rains at critical growth periods when an abundant
supply of nitrogen was most needed. Because most of our field experiments
were nitrogen tests and the results would have been vitiated by the addition
of mid-season nitrogen applications, no fertilizer was added after the
rains. As a result, most of the tobacco, when cured, was found to contain
a high percentage of starved, yellow, worthless leaves. Although this
ruined the crop from a commercial standpoint, it was of immense value
experimentally. It enabled us to compare the capacity of different
fertilizer materials and of different quantities of fertilizer to hold up
under such conditions.

In Table 1, below, the rainfall by months and by 10-day periods is
given. It will be noted that the June rainfall was about 2.5 inches above

Table 1. Distribution of Rainfall in Inches at the Tobacco Substation,

Windsob 1934-1935.

By 10-day periods

Year

1934
1935

May

. June

July

Aug

1-10

11-20

21-31

1-10

11-20

21-30

1-10

11-20

21-31

1-10 |

2 . 35
1.23

.86
.15

1.61

.02

i.57

3.18

1.43

.29

■2 :>:s

1.12
1.10

1 15

2.08
2.05

.51

.38 1

1.29
.16

By months

Year

May

June

July

August (total)

1934
1935
Mean

4.82
1.40
3.59

3.47
5.53
3.12

3.20
4.30
4.35

3.45
1.80
4.24

normal and that the excess was during the last few days of the month
(2.3 inches on June 29). This is near the beginning of the grand period

546 Connecticut Experiment Station Bulletin 386

of growth of tobacco plants, and chemical tests showed that the soil
was practically depleted of available nitrogen, a deficiency that was not
made up throughout the remainder of the growing season. Such a con-
dition called for immediate application of additional nitrogen as a side
dressing. The Station advised growers of the situation and many made
such applications with beneficial results.

With the installation this year of instruments for measurement of
maximum and minimum temperature and relative humidity, more com-
plete weather records are now kept at this Station. A shelter of approved
U. S. Weather Bureau type has been built to house these instruments
and also a hygro thermograph, which gives a continuous record of tem-
perature and relative humidity. The record of rainfall, which has been
kept for several years, will be continued. Wind velocity, wind direction,
and sunshine duration indicators, with their appropriate recorders, may
be added later.

The Station furnishes free service in testing soils as a guide to fertilizer
practice and selection of fields for planting. Approximately 2,500 soil
samples were submitted during 1935, indicating the increasing number
of growers taking advantage of the service. Formerly the soils were
tested only for acidity. Other tests were added as fast as their usefulness
and reliability were fully demonstrated, until now the routine tests include
potassium, calcium, magnesium, phosphorus, nitrate and ammonia nitro-
gen and, in special cases, aluminum, manganese and chlorine. These tests
are also proving valuable in diagnosing malnutrition troubles in the grow-
ing crops.

During the year Mr. Lacroix, Station Entomologist, who has been
studying the insects of tobacco for the last six years, has published a
bulletin on "'Insects of Growing Tobacco in Connecticut," which has
been distributed as Connecticut Agricultural Experiment Station Bulletin
379 to all tobacco growers. It should be of considerable service in identi-
fying and combating the imect enemies of tobacco.

FERTILIZER EXPERIMENTS WITH SINGLE SOURCES OF NITROGEN

A great many materials are available to furnish the necessary nitrogen
in the tobacco fertilizer mixture and new ones appear on the market
every year. It is a common belief among tobacco growers that certain
ones have specific effects on the crop. For example, some hold that
castor pomace makes the crop darker; linseed meal gives it a higher
"finish"; fish meal makes a red leaf, etc. For the most part, such ideas
are based on personal observations or hearsay, or on claims of salesmen,
and not on accurate experimental evidence. Also larger yields are claimed
by the champions of each material.

The scientific agronomist, on the contrary, believes little in specific
effects. He points out that these materials are first decomposed in the
soil and absorbed by the plant as nitrates, or, to a less extent, as ammoni-
ates. The nitrate is the same whether it originates from cottonseed
meal, dry ground fish, nitrate of soda, or other sources. Why then, should
the individual materials produce different results on the plants? One
can only surmise that if such specific effects do occur, they must be due
to different rates at which the materials decompose and become available,
or to subsidiary elements that enter the roots with the nitrogen.

Fertilizer Experiments 547

As early as 1892, the Connecticut Agricultural Experiment Station
began field experiments with the object of comparing the effects of
different nitrogenous materials on the yield and grading of tobacco.
Since that time there have been numerous similar experiments in this
and other states with various old and new nitrogenous materials, but
still there is no agreement as to the comparative merits of even the common
ones.

In 1926 this Station started a field experiment to compare four different
nitrogenous materials with respect to their rate of nitrification, leaching
and residual effect on the soil. From time to time since then, the project
has been enlarged to include tests of additional materials and more replica-
tions. During these years the yields and gradings have been carefully
determined and observations made to see whether any of the claimed
specific effects could be detected and whether they were constant.

Progress on these experiments has been described in some of the previous
publications, the latest of which was the report for 1931 (Conn. Sta. Bui.
335: 239-246), to which the reader is referred for more complete details.
Here we shall discuss only the results obtained since that time and present
some conclusions. The old plots have been continued in the same location :
a leachy, sandy loam on Field V, and a new series on a more retentive,
heavier soil on Field I. Havana Seed tobacco was used in these tests.
In a third set of plots, the organic nitrogenous materials were compared
on Broadleaf tobacco.

Table 2 gives a summary of the yields and grade indexes of the plots
of the old series on Field V for the last six years. Table 2A shows a similar
summary for the new series on Field I, where the organic nitrogenous
materials have been compared for four years. It will be noted that the
yields are considerably higher on Field I than on Field V. Table 4 sum-
marizes four years results on the Broadleaf field.

Looking over the results of the three series of tests on three different
soils over a series of years, the following general conclusions seem warranted :

Cottonseed Meal. The fact that, in the old series, the yield for cotton-
seed meal was on the average less than that of the other organics is due
probably to the unfavorable location of these two plots on more sandy
soil than the others. In 1935, another cottonseed meal plot was added
on more favorable soil on Field V and it was noted that for this year
the yield and grade index were in line with those for the other organics.
Also in the four-year experiment on Field I, it will be seen from Table 2A
that the average yield was about the same as for linseed meal and only
a little below that for the other organics. On the Broadleaf test, the
yield was next to the highest. The grade index also compared very favor-
ably. The common use of cottonseed meal as the standard base for
tobacco mixtures seems to be vindicated by these tests.

Castor Pomace. In the six-year experiment on Field V, this material
gave the highest yield of all, and in the four-year experiment on Field I,
it was almost as high as any. The differences, however, are not large.
Castor pomace has the reputation for producing a heavier yield of tobacco
than cottonseed meal and the results of these experiments possibly furnish
some support for such a claim. It is also a common belief that castor
pomace produces darker tobacco. Such a difference in color has never

548

Connecticut Experiment Station

Bulletin 386

CI LO

LO

(N

l-H

"*

CI

CI

SO

*

CO

m t-

eg

©

CI

lO

Os

©

SO

<!

CO CO

■HH

T(4

•«*

CO

CI

CO

—

CO

HHO\

t- CO

ON

CO

i-H

co

LO

-sf

©

t-

P5

LONVO

CO CI

CI

r-i

OS

CO

C4

CI

©

t-

O

MMCO

CO CO

CO

CO

CO

1—1

CO

CO

CO

co

i— 1 (N

CI LO

*

©

©

,_,

CI

ec

LO

~

O l-H

OS l-H

o

CN

Os

CI

so

©

—

X

o

CO T*

CO T*

"*

CO

CO

<N

CO

Tt<

CO

-

CO o

CO

r~

SO

ON

©

^4

Os

l-H

m

LO CI

"S?

CI

CO

SO

©

C)

SO

CI

•z
cs

Os

"? T}l

■**

LO

"*

r)4

rH

**

■^

■*

O

CO -H4

1—4

co

ON

CI

■<*

LO

©

1—4

SO l-H

r-^

co

ON

rH

CO

r>

1—4

Tf4

o>

CO r?

T*

CO

CO

-#

CO

CO

■"*

CO

■>* ■*

in

t-

,_,

©

CI

Os

so

©

M

CO CI

Tjl

CO

t>

©

LO

■»*

LO

1—4

OS

CO CO

T*

CO

CO

t—i

CO

-*

CO

■*

1 — 1 l-H

LO

r-

©

NO

SO

LO

CI

~

rH rH

so

CN

©

CO

SO

l-H

f-

ON

l-H

«* ■>*

■*

T?

LO

i-H

CO

■*

Tt

I— 1 CO

CI

CO

r>

SO

t>

©

CO

Os.

r-

Os Cs

CI

SO

■^f

i— 1

co

CN

SO

CO

■<

s£5 so

co

t>

t-

rf

CO

o

t-

b-

rH rH

1-1

r-^.

1— 1

rH

■"*

|-H

i-H

^

CJ LO LO

© ci

rH

■^<

CO

i-H

CO

CO

ON

ON

CO t- CO

so OS

CO

LO

T(<

t-

•**

i-H

CO

co

os

ON CO o

o co

O

co

co

©

©

©

ON

©

a>

rH rH C|

CI t-^

CI

r-i

I-H

M

c.

CJ

1—t

CI

CO CO

ciso

1— 1

CO

SO

CI

LO

©

ON

a

NO CI

CO SO

SO

!—l

t-

o

t>

©

OS

so b-

CO CO

t~

co

SO

co

co

co

co

1—1

1—4 r-i

i—i rH

1— 1

^

1 — 1

l-H

rH

r—l

l-H

GO

XI

1—4 ^4

SO

CO

co

CI

SO

CO

©

I-

0

^-< ©

o

CO

"*t4

SO

CO

l-H

rH

T?

s

OS

CO ON

o

OS

Os

t-

1^

co

©

1-

o

pH

rH i— 1

CN

1— 1

i-H

l-H

l-H

C|

r-l

CO ^

CO

©

^

fM

i-H

Tf

•*P

i-H

co t>

o

LO

CO

co

CO

CO

Os

ON

-

OS

CC t-

t>

1—

r-

SO

l>

r-

SO

r-

S*

OsCl

SC

Os

SO

co

eo

i-H

so

CO

CI CO

CO

CI

©

CI

t-

SO

"HH

LO

C

LO LO

b-

so

SO

I-

SO

t-

LO

SO

'— '

1-HrH

1— i

^

i—i

i-H

r-<

l-H

rH

■*^j4

ci

LO

CI

<N

SO

©

SO

l> t-

o

1—

CO

©

CO

rH

OS

CO CO

«*

■«*

LO

co

■<*

LO

LO

rt

l-H 1—4

1-4

rH

r-^

i-H

l-H

l-H

v^^^ s

^^

rH O CO

i-H d

.2

03

CO

3
g

2

15
CD

£

a

CO

1?

CO
O

d

o
£
£

13

CD

£

CD
2
"&E

13

CD

£

2

T3
03
CD
CO

c

£
o

<->

CD

£

CD

"CO
CD

o

CD

*->

CO

o

CD

O
i.

+s

'5
c

CO

+H

O

CO
CO

CD

CO

C

r3
CO

CO
u

CO
■—

3

cfl
CD
s-

£

13

C

tH
O

rQ

o

u

U

r-1

E

z

C/2

hJ

U

U

co

1 E

O <

Fertilizer Experiments

549

Table 2A. Single Sources of Nitrogen, New Series on Field 1.
Summary of Four Years' Results

Source of nitrogen

Acre yield •

Grade index

1932

1933

1934

1935

Av.

1932

1933

1934

1935

Av.

Cottonseed meal

(1)
(2)

1928
2053

1928
1891

2090
2171

2281
2416

2086

.353
.411

.466

.457

.470
.475

.366
.364

.420

Castor pomace

(1)

(2)

1776
2014

1884
1855

2111
2362

2448
2591

2130

.383
.448

.453

.470

.424
.430

.363
.356

.416

Linseed meal

(1)
(2)

2072
1789

1918
1880

2145

2298

2265
2308

2084

.407
.409

.472
.486

.435
.490

.360
.387

.431

Dry ground fish

(1)
(2)

2043
1966

1941
1880

2359
2390

2380
2470

2179

.387
.389

.444

.447

.447
.484

.363

.423

Corn gluten meal

(1)
(2)

2005
1745

1978
1829

2307
2400

2339
2466

2134

.391
.363

.444
.459

.456
.468

.338
.333

.407

been apparent during the six years of the Field V experiments, nor on
the Broadleaf. During 1934, however, the tobacco from the castor
pomace was judged to be darker on the heavier Field I plots, which in
this case seems to support the belief of some growers that too much castor
pomace should not be used on heavy soils. However, it seems to be a
very good source of nitrogen for use on more sandy soils.

Table 3. Limed and Unlimed Sulfate of Ammonia Plots on Field I.
Summary of Four Years

Acre yield

Grade index

Fertilizer treatment

1932

1933

1934

1935

Av.

2013

1932

1933

1934

1935

Av.

Sulfate of ammonia
With lime

1845

1996

1912
1994

2201
2204

2017
2016

.379
.404

.435
.409

-.442
.402

.347
.339

.395

Sulfate of ammonia
Without lime

1895
1861

1896
1659

1990
1746

1691
1505

1780

.436

.457

.413
.380

.374
.261

.335
.200

. 355

Neither sulfate of
ammonia nor lime

2035
2051
2031

1940
2050
2036

1974
1898
2001

2133
2045

2017

.404
.420
.429

.476
.506
.495

.409
.391
.405

.351
.355

.422

Linseed Meal has given favorable results in all tests. The yield is
no better than that from other sources, but it may be significant that
the average grade index has been invariably higher for this than for any
other material. This means that the tobacco from these plots has been
judged each year to be of a little better quality than that from the other
plots. At various times the leaves have been examined and compared

550

Connecticut Experiment Station

Bulletin 386

with those from adjacent plots to see whether the claimed "better finish"
could be observed. When the same grades were compared, this superiority
was not so apparent. It is true, however, that the linseed plots yielded
a somewhat higher percentage of leaves in the best grades and this accounts
for the higher grade index.

Linseed meal appears thus to be an excellent source of nitrogen for
tobacco. Unfortunately it is usually more expensive than some of the
other organic materials.

Dry Ground Fish gave very high yields in all tests and was like castor
pomace in that respect. The grade index has been about the same as
for cottonseed meal.

Corn Gluten Meal has also given very high yields in all tests. The
grade index, however, is on the average a little lower than that of the
other organics.

Soy Bean Oil Meal (a by-product of the soy bean oil industry) con-
tains approximately the same plant food elements and in the same pro-
portions as cottonseed meal. It has come on the market in ever increasing
quantities in recent years, due to increase in domestic production of soy
beans, and promises to be a permanent competitor of the cottonseed
product. Also during the last year the price has been lower. The soy bean
oil meal was tried on one plot on Field V adjacent to the others of the
old series for the first time in 1935. It produced very excellent tobacco
with a higher grade index than for any of the other organics, and did
not show the starved condition which was so evident on most of the
plots this year. A one-year test on a single plot, however, does not furnish
sufficient data from which to draw any sweeping conclusions.

Nitrate of Soda. This is the most available and most easily leached
of all the materials tested. During the first six years of the experiment,
all of the fertilizer was applied broadcast about ten days before the plants

Table 4.

Single Sources of Nitrogen. Broadleaf Plots.
Summary of Four Years.

Plot

Source of nitrogen

Acre yields

Grade index

No.

1932

1933

1934

2183
2123
2241
2206
2358

1935

1813
1866
1771
1871
1906

Av.

1932

1933

1934

1935

.231
.241
.303
.219
.248

Av.

N49
N50
N51
N52
N53

Cottonseed meal
Castor pomace
Linseed meal
Dry ground fish
Corn gluten meal

1721
1430
1600
1460
1562

1872
1938
1946
1892
1982

1897
1839
1889
1857
1952

.465
.346
.448
.396
.390

.426
.462
.453

.474
.462

.474
.455
.515
.467
.437

.399

.376
.430
.389
.384

were set (at the same time as the other fertilizers). Since the land was
quite sandy on this plot, most of the nitrogen was leached out each
year when there were heavy rains during the first part of the season.
As a result, the starved crop was almost worthless when cured. During
the last four years however, in order to overcome this, the nitrate was
applied fractionally in five applications at intervals of about ten days.

Fertilizer Experiments 551

By this practice, very good tobacco has been grown on sandy land with
nitrate of soda as the only source of nitrogen. The average yield on this
plot for the first six years was 849 pounds. During the last four years
(fractional application) it was 1,741 pounds, or more than double the
average of the previous six years but not quite so good as the adjacent
cottonseed meal plot at 1,811 pounds for these same four years. The
average grade index for the first years was .185 (with cottonseed meal
at .338) while for the last four years it was .415 (cottonseed .342). Particu-
larly interesting is the grading record of the tobacco from this plot in
1935. Most of the tobacco on the other plots of this series (except the
soy bean meal plot) was poor, yellow and starved, showing the results
of leaching. But the nitrate of soda plot showed very little starvation
and the grade index was highest of all on the plots of this series.

Thus it might be possible to grow good tobacco on a nitrate alone as
the source of nitrogen, if one distributed the applications properly through-
out the growing season. Such a program would involve additional labor
and would not be very practical for the average grower. The humus
content of the soil could be kept up by turning under winter cover crops.
It is probable that the principal use for nitrate of soda (or other nitrates)
in tobacco culture will continue to be as a side dressing, applied between
the rows of the growing crop for quick replenishment of the nitrogen
after leaching rains.

Sulfate of Ammonia. This material has been found to be the least
desirable of all the nitrogenous substances tested. Results of all our
experiments on sulfate of ammonia up to 1933 were published in full in
our report for that year (Conn. Sta. Bui. 359: 355-360) to which the
reader is referred for a full discussion. Continuous use of sulfate of
ammonia (without lime) makes the soil more acid each year until it finally
reaches a point where tobacco will hardly grow. Long before that, however,
the leaves become very poor in quality. Use of sulfate of ammonia
makes them darker, heavier and more prominently veined. The fire-
holding capacity is reduced, the ash darker, and the taste and aroma
inferior.

Since some of these defects are obviously related to the acidifying
effect of sulfate of ammonia in the soil, the possibility was suggested
that they might be overcome by adding sufficient lime to neutralize
this tendency. Accordingly, four plots were started on Field I in 1932
with sulfate of ammonia as the only source of nitrogen. To two of these
plots, 1,000 pounds of high calcic limestone to the acre was added the
first year, 1,200 pounds of magnesian lime the second year, and 2,000
pounds of magnesian lime the third year. In the spring of 1935, the
reaction of the soil on the limed plots was pH 5.55, and on the unlimed
plots was pH 4.25. Therefore no more lime was added.

The results of four years' test, as seen in Table 3, show that the use
of lime has kept the yield up to the level of other sources of nitrogen
and that the decreased yields that have been observed in our other experi-
ment may be prevented by liming. The grade index, however, has not
been kept up and the tendency to produce dark, heavy tobacco has not
been overcome.

Urea. As may be seen in Table 2, the yield on the urea plot has been

552 Connecticut Experiment Station Bulletin 386

maintained at a high level (second only to castor pomace) despite the
fact that the soil is light and sandy. The use of urea alone, at the rate
of 200 pounds of nitrogen to the acre, has had a tendency to produce
somewhat darker, heavier tobacco, and as a result the average grade index-
is not so high as for some of the other materials. It is probably best not
to use this material as the one source of nitrogen, but more extensive
tests recorded in our previous reports show very good results where urea
is used to furnish a percentage (up to 50 per cent) of the nitrogen of the
formula.

Calnitro, a mixture of ammonium nitrate and calcium carbonate, has
maintained both the yield and the grade index at a level almost as high
as other sources. It is hardly suitable to use as the only source of
nitrogen in the formula but might be substituted for a part of the more
expensive materials.

Discussion of Results. Averaging all the results of the experiments
through a series of years, one is impressed by two facts: So far as yield
is concerned, there is little choice among the different organic materials:
also among the grade indexes the differences are not impressive. Possibly
the better showing for linseed meal and somewhat lower grading for
corn gluten meal may be of sufficient magnitude to be significant. A
grower may reduce his fertilizer bill by watching the market and choosing
any of the other materials tested that happen to be cheaper in a given
season. A certain percentage of urea and calnitro may also be used to
advantage in such substitution. Sulfate of ammonia cannot be recom-
mended. If used at all, only a small percentage of nitrate of soda (and
other nitrates) should be included in the original mixture for broadcasting,
but the grower should keep a supply on hand for side dressing to be applied
immediately after leaching rains. Soy bean oil meal is a promising new
organic but has not been sufficiently tested to warrant recommendation
at present.

NITRATE NITROGEN AND SOIL ACIDITY PRODUCTION BY
NITROGENOUS FERTILIZERS

O. E. Street

Since the inception of the industry, nitrogenous fertilizers have occupied
the foremost position in the nutrition of Connecticut Valley tobacco.
The classic example of the Indian placing a fish in a hill of corn was soon
imitated by the early settlers and can reasonably be interpreted as the
earliest example of nitrogen fertilization. The use of farm manures for
tobacco growing was an early and wide-spread practice and still persists,
though on a much reduced scale. More than 50 years ago, cottonseed
meal was tried as a tobacco fertilizer and soon rose to the dominant position
it still occupies. Other vegetable organics such as castor pomace and
linseed meal; animal organics such as dry ground fish, Peruvian guano,
hoof and horn meal, tankage and blood; and nitrate of soda among the
inorganics, were added to the list and peculiar virtues attributed to each.
Sulfate of ammonia, a by-product of the gas and coke industry, also fomid

Effects of Nitrogenous Fertilizers 553

a limited use. Within the last 15 years, the development of nitrogen
fixation processes to a position of economic, production brought such
materials as urea, nitrate of potash, nitrate of lime, ammonium nitrate,
ammonium phosphate and cyanamid on the market and led to their
trial in tobacco fertilization. By-products of the manufacture of food
products recently added to the list include corn gluten meal and soy
bean oil meal, while the tobacco industry itself supplies stems and stalks
which contain from 1 to 1.5 per cent nitrogen as well as a higher potash
content.

A comparison of some of these sources as the sole carrier of nitrogen
has been in progress at this Station for 10 years and results have been
reported in Tobacco Station Bulletin 10, Report of 1927; Conn. Exp.
Sta. Bui. 299, Report of 1928; Conn. Exp. Sta. Bui. 335, Report of 1931.
Originally planned as a study of leaching and soil reaction, the records
now include yield and grading data for the entire period. Four plots
on which cottonseed meal, nitrate of soda, urea and sulfate of ammonia
were compared as single sources, comprised the entire experiment at
first. Gradually others were added until, in 1935, 12 sources were being
studied on 15 plots. The 4 materials first used have occupied their original
plots during the entire 10 years. Castor pomace, linseed meal and dry
ground fish have been undisturbed since 1929, cal-nitro since 1930, and
corn gluten meal since 1932. Fertilization has been on an acre basis
of 200 pounds of nitrogen since 1928, before which it was 164 pounds
per acre.

Measurements of the Production of Nitrate Nitrogen

The large differences in yield, quality and soil reaction produced by
the various classes of materials represented in the original experiment,
as well as the smaller differences produced by the materials later added,
led to the introduction in 1932 of measurements of the soil nitrate levels.
The tobacco plant takes up most of its nitrogen from the soil in the form
of nitrates. But, except for nitrate of soda and calnitro, the materials
mentioned contain little if any nitrogen in the form of nitrates. The
nitrogen in cottonseed meal and most of the other materials is
called unavailable and cannot be used immediately by the plant.
It is only by complicated decomposition processes carried on by
soil organisms that it is broken down to nitrates and becomes available
to the plant. The speed of the breaking-down process is different for
each material. By measuring the quantity of nitrate nitrogen present
in the soil at frequent intervals, it is possible to obtain a detailed picture
of growing conditions throughout the season. Each measurement repre-
sents the difference between the amount of nitrates produced from fertilizer
and soil organic matter and the amount taken out by the growing plant
and by heavy rains. As nitrate is also the form in which the nitrogen
is most easily washed out of the soil by heavy rains, these same tests
give an indirect measure of the amount of leaching. The need of these
measurements as an indication of the differences between these materials
in rate of availability, resistance to leaching, and recovery from heavy
rains was apparent. The most frequent question asked by the growers
of Connecticut Valley tobacco in reference to nitrogen fertilizers deals
with the speed at which these materials nitrify and the probable effect
of their behavior on the growth and quality of the tobacco.

554 Connecticut Experiment Station Bulletin 386

Measurements of the soil nitrates and soil reaction were made from
April through September, of the years 1932, 1933, 1934 and 1935, at
weekly intervals during June, July and August, and at wider intervals
in the spring and fall. These data for nitrate nitrogen are presented
in Tables 5, 6, 7, and 8, and for soil reaction in Tables 9, 10, 11 and 12.
while the four-year averages are to be found in Tables 13 and 14. The
average nitrate levels and soil reaction for all organic sources are graphically
shown in Figure 129, while the corresponding results for the inorganic
sources and the plot receiving no nitrogen are shown in Figure 130. The
correlation between nitrate levels and rainfall distribution and amount
for a year of low summer rainfall, 1932, and one of higher rainfall (as
well as irrigation), 1934, is shown for tvpical materials in Figures 131 and
132.

Seasonal characteristics of the period. The classification of
summer seasons as dry or wet in relation to nitrification, may not furnish
an accurate criterion of growing conditions, as such factors as distribution
of rainfall and average temperatures are not taken into account. Depart-
ures from the optimum in growing conditions may be due to low tempera-
tures or poor distribution of rainfall as well as to excessively wet or dry
conditions. Thus early availability may be low because of a cold, dry
spring, under which conditions the decomposition of protein materials
would be limited by both the shortage of water and unfavorable tempera-
ture. Again, the total rainfall for a month may equal the exact mean,
but if it all fell in one or two periods of two inches each, the result might
be serious damage to the crop rather than uniform growth. An abrupt
change in weather conditions from cool and moist to dry and hot will
often find the plants with poorly developed and shallow root systems
and the crop will almost cease growth. Wilting will be common, and the
value of the crop will be greatly impaired.

The four years during which the tests were made ranged from perhaps
an average growing season in 1932 to an abnormally wet season in 1935,
with 1933 rather dry throughout, and 1934 of adequate early rainfall but
very dry during the greater part of the growing season. Cool and dry
conditions were prevalent during the early part of the growing season of
1932, but a favorable rainfall distribution during most of the time the
tobacco was in the field maintained nitrates at a rather high level. During
1933 the general level of nitrate production was low for the entire season
because of inadequate rainfall. Irrigation failed to stimulate nitrification,
in contrast to a natural rainfall in early July, which produced a moderate
stimulation. The season of 1934 was one of moderate rainfall until June
20, followed by five weeks of drought, broken only by an inch of rain
during the third week. Nitrates were never at high levels. After ab-
normally dry conditions in April and May of 1935, June and July were
months of high rainfall. Precipitation of more than two inches fell on
June 29 and more than one inch on July 19, and nitrates were at a starva-
tion level during a large part of the growing season.

Residual nitrogen and early availability. Nitrate determinations
in April of three years show definitely that the quantity of nitrates left
after the winter and spring rains is very low, and that nitrification, either
of unused fertilizer of the previous year, or of the natural organic matter

Effects of Nitrogenous Fertilizers

ODD

-
-
O
-
-

H
H
<

H

«

as
-
O

a
o

3

-
3

K
>

a

a>

0 O

ZH

a

OS

o

£>

CS

£ "3

S3 og

3 fl

173 s

*o"g

z M

<D

O

03

S S-s

cj^S

t»

T3

a

3

O

ft

T3

Q 0«

c

60

<u

ac

o

+j

T>

a

O)

52 a)

.sa

03

j

+a

z

03

o
U

aj

+j

03

Q

t*O»»C0CMC0^N

H«C5HhO\OCOW
ON"OMHCO\OvO
i— I r— I CO <M ^H

VDO^OiOtJ'COOOnC©
(NCO^OOIOCOOi-ht*

C^l i — I ' r- I rH

t~-t-<UOOpHvOCOTjl

to m \o >— < t> o i- .

i— 'cneooomvooof-

ON CO i— i o

- t»3 m h

"# N i-l if)

0"\<Ncoco\ccococo^oc^csoTj<e\it^co

i-H O rji tJ( 00 |> (M

^O l-H

VD

CO

~

t~~

r— 1 CO

—

CI

(N

>> >=

<o

<B

QJ

V

CO 03

a

a

c

C

££

3

i-s

i-s

3

i-s

i-H i-H (M i-H (M _ rH

_£? i? i? _b 6D &E 6C Man,
'S'33'3^355 «f «

Connecticut Experiment Station

Bulletin 386

-
q
c
s
-

-

£

a

—
c
a

3

-

60

o = S

.3 £

3 S

si

a -a

^ r ci :i

t- eo o in en t- (M

H tl r W M K! ^ ^ O O C

t- ifi •*? O t-

iseoHo N^

i— I 5s! f- in \C vo

i/; « W 0\ <N i— t- i-H

O. M O "? CO M

h « « i: T* ^

S\ K M K K rt

h © h O M ^ W

C>, M K ■<*

ir h M rt M t- t» N

l« N K W t» N M H

- O « <* N M N M

CO © t~ ^ i— I

t— i ^H <N CO

tjc r-l in CO

^i ^ O (B O «

CO cB C

^2£

C C

S 3

C ^ — — ' ^ ^° ^ ^ ^

a o,

Effects of Nitrogenous Fertilizers

DO t

—

c
-
o
a

£

o to

= "2
IB

7D C

Q 2 =

I!

Z a
US.

|1

M « il N - fl M O

CO i— i CO CO 0\ Ss t* 1^ f-

1—1 CO lO O LO Tf l~ O I— I CO l-H

Cs CC CC O 3C O 50

* N H

10 O O l> <M

ci m a 55 « o

*m ci *c ci ci co io

| H IS H ^H} 1ft N « Tl
5 lO CO r-H CI CI CO — ' -H I-*

o v© \o >-o t— \o i— i i— i e*i o ci

<N LO C) t- Ov M 5S l~-

i— I <M ^< \C CI CO

C ic * ^ " * l« 51 a o M s '^ M O I- O ™

>> >-- >> >-■ bJO 6<5 6C 60

a* a a

™ — "- — — 'f-*^)(l>
h-S H-5 ^ <!-«<-< C/5 X

558

Connecticut Experiment Station

Bulletin 386

c

a>

i!

S%>

*e

u s

«3

u-g

d

05

p

<S

£ '3

-=<3E

1/3 I

$ a

*^"0

■fe o o

«J

2

o

CO

s

£r= as
7. °6

o

T3

in

fl

3

O

a.

i o

sec

c

S.S §

• i-i

i >• 60

c

CD

tuo

o

e c —
C.S 'S
O 3 S

3

U'Sa

<B

*J

CC

t-l

ts

z

h9a

G g«

tc

-a

is

.2 6

J

3§

05 C

O c.

-a

05

05 —

S Sf

3E

o

U

CO

+=

as

Q

N CO N O W O >* C I- <J N

n o h lo t* n <« h i— i i— i

COHMLONCSaH^^WVOM^Nt-M
N CO W O ft H f,] IN H i — 1 i — l CO i—i

MWMOHH^riiNhCOcO^WKCO

I— I CO t- I— I C3S H CO

COcONCOlOLChHlOt^ON^iiHij^
I— < r-< CO LO I— I CO CO r- 1 CM i— I i— IT?

•*hH*t>co«Hoee«\ccoHcoHb

NN^ONifiWN CO i— I CM W r- 1

lOt-HH^COLniO^^IOlOONCOCOWCC
CM CO t* CO CO \D lO CM H •* « Ol

WiMNWcOINcOoNHioceXCOClCivO
H N N i* C ci « * M l_l

>o LO CO O t—

Oh «

>s>v>^>^>^tlC53£ilCtlC

c s c ^ — ,i_

S 3 3 3 3 3 3

-5 t-j H; ►— H-5 r-; I— 5

3 "3 3 3 3 3 u

— 1-; -<■<;■<-< C/) CC

Effects of Nitrogenous Fertilizers

559

o an

1^1

a §73
S- *s 4)

^ Ml

■? 3 b

Q g«

v -

3E

2 «

si

US

§73
II

o r^ co t~ o t~

w w w io io w m in io w w w in w us io

iflwifl'j'^^'j^Tj^^ioTf^TfLn

'i'WttW^rfT'^TfrtTfTtTfTfTJIt*

■'i'^^'ij^^MrtTfcijnTfccwn^

w w lo lo w w ui io m w w i.o w in io *

iowwTf'*'*^^^.)(^in

CO eo ^o

■*T?T*T}<-<*Tt^J.^.^^*T*-*'*T*^*

lOlOr?T*-*^T*T?T}<Tj«T*

m c in c in
co ^ i* eo ■>*

■># ■* T? T?

minTfT«i*Tji^^^i^^i^,^^^Tji

Tj'mmmm^mmin'^ininin'^inLO

>0 ^H '■O CO © t—

iflHcowrtoomN^a

rH I— I <N I— I !M i— I

^j <U <V O <D

££

■< < < < rj) W

hh ^ h

560

PS

w

to

13
O

z

w

PS

-

Connecticut Experiment Station

Bulletin

— o
ffl i

ioE

55 c o

^•5

1

o g

1/5 LO LO LO

LI? LO LO LO LO LO LO LO LO LO LO LO

LOWLO^rfrtTf^^TfiJi^Tf^rt-t-"

CO <M ©

LO LO LO

T*r?«r!<-^>T?'-<#Tjf'#-^<T*-rr.^.-fTf<

^^'♦'•wMMcocommw'jtc^MM

O LO LO LO LO LO LO LO LO LO LO LO LO LO LO >0

N O CO © \0 <N \D '"*

(Ni-^OOiO-<*lO>0

lo w w w ^i •i1 ^i Tf <#

■* *# T* Tj< r* ^

LO<fiin'<si'*'*'*'t'*,J('*'^'*',?'*'f,Tf

O « H M ffl N O
H O O ^O t? r? CO

LO LO LO T? ■>* T* T*

Tf^'J^Tf^TfTiiri1'*

l.O"5'*^l"*'*T?'?'f*',#'*'*'?

U5WCiWNO\\0M

>i ^ 0) Si « 4!

£££

>5>^>^>^>^tict>cic.

S? * S c c c ^^^2?^ SF Si

5^333333333-;-;

1 — S ►— 3 I — S ' — S ^^. ^^> I — S 1 — S

3 3 3 ® ^
■< < •< as m

Effects of Nitrogenous Fertilizers

561

3-2 £

03 5

g-a

SB

>=a.c

in CO

ws w io m w 10

w in io ic Tf w w w io w io u)

muOr}'in'rJ,-!*T*T}<'<*^TfiTj<TfiTj<Tj<rJ«T*iO

o c t* *o o

O h ic h iO

m in t? ■** t?

^* ^* "^ ^* ^* "^ ^!? ^?* ^* ^* T? ^* ^^

'^T*T?LO'fT«-JlTJI'*^TJI^^TJI^T}''I*>#

lO cc C

i— i o> <

^W^tji^C(5MmnM(i5CncB(OMMnff5

m m m in w in m m m m in in m in m m ^o m

LOLOT*-«*T*-^>T*'^''#T*T*T*-<#-^r*T*T*<Tt

MaiTjcnONNa^^ci^oN^ccaa^
^OsCccmT*LO^,-^'fOT*T?T*inoT*int-

^*T?-<*r*^.T*-^-*^<Tf<Tj<T*^rJ<T*Tj"T*

WMO**0\CSIONcaCiNMNO\0\0\0

Kcccct-**nMW^L«Mionin^LnN

■«?•^•T*T*■^'■<*T*■r}<'<*,

«XCC01OTfCQC>|M«MtfL')COTfci5'*in

m m o
en o t>

wmT!'in'J,'J'f^,i,ii^'f't'jTiiT)iinin

^o*,,*!— 4eoioc-j©\v©co©\oco©r-©"*

i— i <N ^H f-h (N rt M M H fl M H N

< 5 S

562

Connecticut Experiment Station

Bulletin 386

H

b*

22

Oh > 5b

o 3 §

•-, s

°E

LO LO

LO

IO

LO

LO

LO

IO

LO

LO

LO

LO

LO LO

L0

LO

LO

Csl CI

eo

CO

CO

so

CO

o>

LO

so
o

CO

o
—

CO

t^

OS iM

t? CO

Os

-1-

o

CO

o

in lo

LO

LO

—

^

■>#

T*

"tf

"*

T*

T*

■«* -e

■>*

•*

r*

Cs CO
(M CO

CO

o

o
t>

LO

0>

CO

LO

CO

—

s

sC

2

OS Os

o

00

CO

in io

•&

Tf

■^

r?

■^1

-*

■*

"*

"*

T*

•^ ■*

■*

LO

LO

co so

"*

CI

o>

~4-

10

LO

CI

CO

Csl

T*

CO

O IM

co o\

p.

t>

Os

LO

T? T*

LO

LO

■<#

■«?

-*

t?

CO

■*

^<

-*

■^ T?

■<*

T*

•*r

i—i o

T-H CO

LO

—

CO

o

cc
o

CO

CO

0>

co

CO

o
[ -

LO

s

—

sO

0S

LO

T* T*

-t

CO

CO

CO

CO

eo

CO

CO

CO

CO

CO CO

eo

fY^

T?

IN SO

IO rH

CO
CO

o

CI
CI

so

CO

0"

CO
CI

—

—
eo

OS

eo

eo t>

CO SO

LO
IO

CO

o

**l
eo

sO SC

IO

LO

—

LO

IO

LO

LO

LO

LO

LO

LO LO

LO

LO

LO

CO CO
r-* CO

LO

CO

o>

CO
LO

o

O

CO

SO

—

CO

o

to

CO

LO

CO

c

eo co

O l-H

00

l>

so o

o

LO

IO

LO

so

so

LO

SO

SC

SO

SO so

LO

sO

LO

eo t-

i-h m

CO

CO

O

~

LO

10

so
CI

co

CI

o

OS.

—

CO

co co

O OS

5

CO
LO

s

SO o

so

LO

LO

so

so

SO

>o

SO

sC

so

O LO

so

IO

LO |

O LO

© o

—
o>

c:

LO

O
LO

CO

CO

0>

—

t>

o>

Os Os

IO

CI

0>

o

j; i

LO LO

T?

LO

^

—

■*

^

■*

"*

■^

•*?

"* "«*

-*■

■«*

T?

CO CO

Os Os

CO

CO

CO

CO

O

co

CO

CO

CO

0"

CO

LO

—

IO

eo

eo lo

T* Csj

CI
CI

SO

o

sC 1

-* r*

-J-

■«t

■*

'■*

■^

^*

■^

->*

^<

•*

r? t?

■*

■*

-

CO CO
Os Os

IO

CO

o1

CO

CO

CO

CI

CO

CO
CO

CI

CO

Csl

C-J CO

CO o

CO

LO

"* t*

■«*

Tf

■■*

T?

-"

T*

•«?

—

■*

T*

T* ■<?

■<?

■*

"T

co eo

CO Os

IO

OS

t-

Os

-*

<N

co

CO

12

0>

Os
CI

■::-

co

SO C-l
LO CO

CO

Os

LO

*# r*

->*

'*

■*?

•*

-#

■■*

■>*

-*

T?

Tj"

Tf T*

^"

■<*

"*

CO t-

CO r- <

co

N

CI

CO

CI

CO

LO

so

O

O

CO
so

^f

CO
CO

r-

r- t>

eo

(~-

t—

LO LO

LO

T*

>*

■»?

■*

■«?

T?

■<#

■<?

■*

■># ^

■*

lO

■^,

LO LO

CO

—

T— 1

—

<N

i-H

CO

IO

OS

LO

(M O

so
IN

Os

CO
CI

<i

c

c

3
— :

G

~3

>;

>

5

_^-

>~

-

tc si

be

a
r.

+i

c
c
IT.

Effects of Nitrogenous Fertilizers

56:5

1

a

85

I o

— u

B —

u'S

to

a)

Cm

p

C8

v a

-=°s

<K s

4)

SOS

Z

<0

■~

o

to

6^e

'fi

60

T3

C

O

a

c

c

be

Ca.iS
5£

O

b

*»

c

TS

45

ce

Z

o ft

-c

0>

a p

11

0

U

03

till

a v

t> t;

o *

>a

<

WftHOJOMflMOiW

ft ir> 10 o t> t- t- T?

T* CC IM O CO -H TT"

O f II-- N M H w o N ^

IM O CO ■— I TT1 TT O CI O r- CO i— i LO O (T-l

CCOO\HO\LOt-»NXHXLONN

eo ft o t- n B'vo i-

i-h (M LO LO W ^ U! (-

H CO "* <M O t-

M t- CO CO N

ON O CO t- CO CC

uo •> O ft

t* O » t- O <M

c-i in c-i

fo eo ro

i— lrH(Mr}<i^T*eOcV5

i— i <M

SO 1/! N ft

a*3

e c

3 3

,— r^i s-. s-. S~i s-

s "3 3 "3 3 "3

-5 ^5 H-5 H

3 3 3 3 57 57
-< «< <l <; cq c«

c

i64

Connecticut Experiment Station

Bulletin 386

<
a

o

2i

So 5

Z

a c-=
r £ <c

c a a

t- s m

Qoca

.2 E

cs

LO lO L.O LO LO LO LO 1/5 LO LO LO lo m LO LO

b> m e>. w cc ft « w lo wm(Cmo\h\oh ci
i— i t? lo cc e-i o c-i cs >; o> n \c « rf » ■*? cc ©

CaOOCCT*^niNOflNK«MT*'<*l>

"31 ■<* ■** -C1

LO LO

T? LO T* Tj< © -**

f- \o t-

co co co

T? \£ VO 0s. 0s.

VC 11 LO LO

LO LO tO LO LO LO

LO LO ^ T* T? ■<*

TT t? t* r?

""? "* T* T* T* -*

T?--^-T?Tr-5e-r'-<*-r*-r'

LO *^* LO

-# ■* T*

vc o cc

rt C t*
r? t* rj>

T* "<? ■* T? T?

CO CO ■<*

>* "*' T?

■# "* "*• ■<? ■"$• ■"# ■**

>e o vc © •<* >c

CO CO CO Tj> LO •*

-*•*?-*}■■'#■<-<-*■'*■**■'*■<*■<*

o ■* a o

^■r!'-r'T*-*Tj.T*T*-*T*T?"'*lr:

t* i— i cc lo m c>, vccoo^ccoor-r-^-i

■< i< ^ ^ *-; >^ — — -^-^^-5<-<<<c«C/2

CSCC — — ^

Effects of Nitrogenous Fertilizers

565

of the soil has proceeded at a very slow rate. Conditions by the middle
of May are not greatly different and it is quite apparent that no reliance
can be placed on residual nitrogen. Fertilizer applications are made dur-

6.00

O

O 4.50

Qc

^ 4.00-

O

3.50

com.

••:■-.. .-•• ,N-. .••■ /£&

CSM COTTONSEED MEAL
CP CASTOR POMACE
LM LINS EED MEAL
OOF DRy GROUND F I S H
COM CORN C LUTEN MEAL

APR 15 MAY 15 29 JU.4 II 18 25JY2 9 16 23 30AU.6 13 20 27 SEP. 7 21

DATES OF SAMPLING

Figure 129. Graph showing average nitrate nitrogen and soil reaction
levels of organic sources for the period 1932-1935.

ing the latter half of May and gradually make their presence felt by early
June.

Maximum levels. The time at which nitrates reach their peak varies
with the seasonal distribution of rainfall. A high point sometime during

J66

Connecticut Experiment Station

Bulletin 386

June occurred in each of the four years, ranging in date from June 13 in
1932 and June 19 in 1933 and 1934, to June 24 in 1935. This level was
not maintained, usually due to heavy rains, but in 1932 it was succeeded
by a much higher peak in late July, and still another after the crop was
off the field in August. | In 1933, a rather low June maximum was not

^ 500

o

^ 4.'.>G

Q:
O

>^^

^^\^*.r /

\ ._**/

_ \^ ,. ■'-^_ jS

"■'••-.. .No N ^ / ...

, u \,-""\

x-"\\ /<-" . — - -

\ v. ./■ V

\ N' ---^ / *■'

SA '"-•-. "~--"' '"—''"' Jl, '"

— "' '^-^^_ \ ^*' ""—.—"

^~*"*"~' "^ K-~''''^\ S'~~

X- N^^''

"~\ ' a y

-'-■< ; \ s

■~~.~. / \ *s

\ / V..--?.^-"

Z. EG E ND

NS NITRATE OF SODA

SA SULFATE OF AMMONIA

U UHEA

NITRO

rnoGEi,

APR/5 MAY 15 29 JU.4 II 18 25 JY2 9 16 23 30 AU6 13 20 27 SEP7 21

DATES OF SAMPLING

Figure 130. Graph showing average nitrate nitrogen and soil reaction
levels of inorganic sources and no nitrogen for the period 1932-1935.

exceeded with any consistency, as steadily dry conditions and the increas-
ing demands of the crop prevented any accumulation. A sharp drop
occurred soon after the June maximum in 1934, with nitrates maintained

Effects of Nitrogenous Fertilizers 567

at a lower level until July 16, when a still more severe reduction followed
an irrigation equivalent to nearly two inches of rainfall. In 1935 the heavy
rain of June 29 reduced the supply of soil nitrates sharply, while the suc-
cession of July rains allowed only occasional recovery. Thus in three
of the four years, the early peak was not surpassed, a condition which
can hardly be considered normal for our climate. It should be empha-
sized, however, that all weather is unusual under our climatic conditions.
The record for the month of July, with an average rainfall of 4.35 inches
for 64 years at the Hartford Station, shows extremes of 15.35 inches in
1897 and 0.54 inches in 1924. In only 13 years of this entire period was
the rainfall within one-half inch, plus or minus, of the mean. Thus it is
idle to expect ideal rainfall distribution in any given period more often
than once in five years.

Relative availability in relation to crop needs. The question of
supply and demand in the nutrition of the tobacco plant is the ultimate
measurement of the value of any material. Studies of the rate of growth
and nitrogen assimilation1 of tobacco indicate that the plants take in
an average of only 7 pounds of nitrogen to the acre during the first 30
days in the field. Subsequent assimilation is much more rapid with con-
sequently heavier demands on the materials furnishing nitrogen. At the
end of 50 days, 66 pounds have been used by the plant, and at 72 days,
the average date of harvest, 114 pounds of nitrogen are present in the
crop.

The first conclusion from these data is that early availability is not a
prime requisite. Materials such as the nitrates, if used in a single appli-
cation before setting, are in a precarious position due to danger of leach-
ing. The plant is not able to use any considerable quantity until about
July 1, and it is to be noted that heavy rains almost always occur some-
time during late June. Fractional application of these materials may be
practiced, and has been followed with the nitrate of soda treatment be-
ginning in 1933.

In the ordinary growth cycle of the tobacco plant, assimilation by the
plant will usually exceed the available supply in the latter stages of the
growing period. This is a desirable condition, as undiminished nitrogen
intake delays ripening and tends to produce dark tobacco. Some materials,
if used as sole sources, either produce nitrates too abundantly during the
latter part of the growing season or have accumulated so much that the
plant cannot use it all. Under unusual circumstances, nitrates are sub-
ject to this criticism. More frequently, however, this type of behavior is
characteristic of sulfate of ammonia and to a lesser degree of urea. Vege-
table organics are not apt to cause this trouble, but fish, as indicated by
its average nitrate level on August 6 (Table 13 and Figure 129) is on the
border line.

Among the organics, differences in maxima and general levels are not
great. Slightly more favorable location of the linseed meal plot and one
plot of castor pomace have been noted in previous reports. Fish and cal-
nitro are also located in this tier (see diagram page 241, Bui. 335). Corn
gluten meal has been moved since 1931 from this tier to a position on the
upper part of the field, adjoining the nitrate of soda plot on the south.

1 Rates of growth and Nitrogen Assimilation of Havana Seed Tobacco. M. F. Morgan and O.E. Street
Jour. Agr. Res. 51: 163-172, 1935.

568 Connecticut Experiment Station Bulletin 386

Replications of the cottonseed meal and castor pomace plot, in that order,
adjoin the sulfate of ammonia plot to the south. The cottonseed meal
plot on the lowest tier, was moved in 1932 but returned in 1935 to its
original position. Thus at the present time there is a cottonseed meal
plot on each tier and a castor pomace plot on each of the two lower tiers.
Over the four years under discussion in the present paper, a slight differ-
ence in soil type and topography operated to the disadvantage of cotton-
seed meal and perhaps corn gluten meal.

Taking into account these differences, the lowest level of nitrate nitro-
gen during the period the plant was in the field was maintained by cotton-
seed meal (Figure 129). Corn gluten meal started slowly, being the lowest
until June 25, but thereafter the decomposition was as rapid as any other
source. Post-harvest nitrification was rather high, but the drop at harvest
was normal. Contrary to popular belief that linseed meal nitrifies slowly,
it made a rather rapid start, maintained a good growing level, and dropped
to a satisfactory point before harvest. Castor pomace behaved in a
fashion closely paralleling that of linseed meal during the entire growing
season, but resembled cottonseed meal in remaining at a lower level after
the crop was removed. Dry ground fish had the highest average of the
organic materials but was not appreciably ahead of linseed meal and corn
gluten meal. As mentioned previously, it did not drop so sharply before
harvest, and maintained the highest late level. Among the inorganics
only urea is worthy of note. The nitrate production of this material was
greater than any of the organics, yet in the same order, and at a rate to
produce a curve of the same type, (Figure 130). Furthermore, the ready
decomposition of urea, as indicated by its rapid recovery after rains, per-
haps justifies the use of less total nitrogen in the formula where this
material is employed. The low post-harvest level of urea would indicate
that the greater part of this material becomes available during the grow-
ing season.

New Materials. Three materials have been added to the test in the
past two years, cyanamid in 1934, and soybean oil meal and Peruvian
guano in 1935. Cyanamid was used to replace one-fifth of the fertilizer
nitrogen on a cottonseed meal plot located on the southeast corner of
the field. Peruvian guano and soy bean oil meal were located in that order
in a tier north of the old plots. This latter area was occupied by an Adco
manure plot from 1926 through 1931. It was used as a forest nursery in
1932 and 1933 with no fertilizer, and returned to tobacco in 1934 receiving
a standard fertilizer furnishing 200 pounds nitrogen, three-fifths from
cottonseed meal, one-fifth from castor pomace and one-fifth from nitrate
of soda. No lime was added to the soil at any time since 1931.

A comparison of nitrate levels for the cyanamid-cottonseed mixture
compared with cottonseed meal during the past two years indicates
practically no difference.

The behavior of Peruvian guano and soy bean oil mealduring a single
high rainfall year is not a good measure of their ultimate value. Both
maintained nitrates at a distinctly higher level than cottonseed meal until
after the heavy rain at the end of June. The subsequent production of
nitrates by Peruvian guano was consistently lower than that of the cotton-
seed meal plot correspondingly located, while soy bean oil meal was con-

Effects of Nitrogenous Fertilizers 569

sistently higher. The soy bean oil meal produced superior tobacco and
seems to be a promising material.

Rainfall in Relation to Nitrates

The previous discussion of nitrate production has introduced most of
the information necessary to an understanding of the importance of
rainfall in nitrogen fertilization.

The presence of reasonable amounts of soil moisture is indispensable
for the decomposition of organic fertilizers as well as urea or sulfate of
ammonia. If the supply of soil water is so limited as to inhibit plant
growth, it will also limit the production of nitrates through interference
with biological activity in the soil. Such a condition prevailed during
parts of 1934.

A review of the processes which organic fertilizers undergo before they
become available to the plant is not amiss because of its fundamental
importance in an understanding of the problem of nitrogen fertilization
of tobacco. There are three major steps in the process of nitrification,
the nitrogen being changed from the organic forms, first into ammonia
salts, second into nitrites and third into nitrates. In each of these steps
a specific class of bacteria performs a distinct function. The final product
in the case of any material is a nitrate of potassium, calcium, magnesium,
sodium or even ammonium. Urea and sulfate of ammonia are already
intermediate compounds and hence should be more readily available.
It is only in the form of nitrates that the plant is able to assimilate any
appreciable amount of nitrogen. Therefore, unless the proper conditions
of temperature and moisture prevail, the plant will suffer from lack of
nitrogen.

Nitrate salts, either produced by nitrification or applied directly, are
at once the most available to the plant and the most readily removed
from the soil by heavy rains. An examination of Figure 131, which shows
the nitrate levels for nitrate of soda, urea, dry ground fish and cottonseed
meal in 1932, and the rainfall during the season, amply illustrates this
point. The fertilizer application was made between the determinations
of May 16 and May 31, and in that year the nitrate of soda was all applied
at one time. Nitrification occurred at a reduced rate until a rain of three-
cjuarters of an inch on June 12 stimulated activity. Up to this date,
there was insufficient moisture even to dissolve the nitrate of soda and
diffuse it in the soil evenly enough to secure a true sample. The other
three materials doubled their production of nitrates within a week. Heavy
rains, 1.61 inches, on June 16 and 17, promptly reduced the nitrate supply
in the surface soil and until the middle of July the accumulation was very
gradual. Rainfall was sub-normal in this period, only 1.30 inches until
July 17. A total of 1.13 inches in three rains within the next 8 days caused
renewed activity of all materials, and the peak for the season was reached
on July 25. During the following two weeks, rains of 1.78, .45, .70 and
.24 inches produced leaching on all treatments and the nitrates practi-
cally all moved out of the surface soil. Harvesting the crop about August
10 ended the drain on the nitrates from that source, but only dry ground
fish showed an immediate response, the others remaining low at this

.70

Connecticut Experiment Station

Bulletin 386

point. A temporary revival in early September found the fertilizers with
only a limited supply of readily nitrifiable material, and heavy rains soon
brought about the almost entire disappearance of soil nitrates.

300

L EGEND

CSM COTTONSEED MEAL
DBF Ot)V GROUND FISH

NS NITRATE OF SODA

U UREA

APR 16 MAY 16 3IJU6 13 20 27JY5 II 18 2SAU.I 8 15 22 SEP.6 19

DATES OF SAMPLING

Figure 131. Graph showing nitrate nitrogen production in relation to
rainfall in 1932. Nitrate of soda applied all at one time in late May.

A similar study of the same materials in 1934 is shown in Figure 132.
Nitrate of soda was applied in five equal lots on May 21, June 2. 12, 20

Effects of Nitrogenous Fertilizers

571

cc CO
<o O

to ^

wo

50

LEGEND

CSM COTTONSEED MEAL
Dar DRY GROUND FISH
NS NITRATE OF SODA
U UREA

AX

.dgf_/J .t \

MAVIS 29JU4 II 18 25JY.2 9 16 23 30AU.6 13 20 27
DATES OF SAMP LI NG

O O

It:

o ^

o

CSM COTTONSEED MEAL
DCF DRV GROUND FISH
NS NITRATE OF SODA

25

-

20

-

1.5

1.0

-

OS

-

LI i

0

- . 1.1.

i .Si.

■ s

I. 1

nil ■

;..

11 1 i.

In

j

,

I

A PR 1 5

J I I I L

J L

MAY 15 JU.3 10 17 24JY.I 8 15 22 29 AU.5 12 19
DATES OF SAMPLING

SEP 9 23

Figure 132. Graph showing nitrate nitrogen production in relation to
rainfall and irrigation in 1934 (above) and to rainfall alone in 1935
(below). Nitrate of soda applied in five equal amounts from
late May to early July.

572 Connecticut Experiment Station Bulletin 386

and July 2. Heavy rainfall before the fertilizer was applied kept nitrates
very low and the increase was not marked until .82 inch fell on June 12.
Determinations on the eighteenth indicated a rapid increase, but this was
quickly lost in the following week, when 2.35 inches of rain were recorded.
Aided by an inch of rain on July 7, urea gained quite rapidly up to July
16, while nitrate of soda, with the final 40 pounds of nitrogen applied on
July 2, rose to a fairly high level. Fish gained less rapidly than urea,
while cottonseed meal was only able to maintain its level of July 2. An
irrigation on July 17, equivalent to 2 inches of rain, brought a rapid drop
in all cases. Added to this was a natural rainfall of more than 2 inches
between July 26 and 28, and a partial irrigation which was interrupted
by the rain, and the soil nitrates were out of reach of the plant roots.
Urea and fish regained a low level in the middle of August, but excessive
fall rains soon dissipated these gains.

The nitrate levels in relation to rainfall for 1935 are also shown in
Figure 132. This year found the lowest levels of any year studied, a con-
dition almost equally true for all materials. Nitrate of soda was again
applied in five equal amounts but failed to make any proportionate re-
sponse. Early nitrate levels were low, but unlike 1934, this was not due
to too much rainfall. In fact, the low level of nitrates early in 1935 was
due to too little rainfall, which did not permit decomposition of the cotton-
seed meal, fish or urea, and even the nitrate of soda was not sufficiently
dissolved to get a true sample. Then almost as soon as nitrification had
made a good start as a result of moderate June rains, a very heavy rain.
2.30 inches on June 27, reduced the level to the danger point. Late July
rains brought the level below 10 pounds to the acre and the crop was
actually starved during the remainder of the growing period. Fish made
a fair post-harvest recovery, but the damage had been done.

Speed of recovery after rains. It is noticeable that the four materi-
als, each representing a different class of compounds, behave quite dif-
ferently following rains. Nitrate of soda fluctuates widely, in part due
to sampling error, but mainly due to the ease with which it is trans-
located by rain water. A moderate rain, up to an inch,will move the nitrates
down only a few inches, provided the soil is not already saturated. With
the onset of drier weather, the nitrates will return with the capillary
water and again become available to the plant. Heavy and persistent
rains will soon carry nitrate of soda into the drainage water and the loss
is then complete. At this point it should again be emphasized that other
nitrate salts, such as nitrate of potash or nitrate of lime, behave similarly,
as do all the nitrates formed by the complete decomposition of organic
or intermediate materials. Fortunately other materials decompose over
a considerable range of time, and unless they have been nitrified, they
do not leach readily.

It is apparent that cottonseed meal regains its nitrate level very slowly
after heavy rains. Fish decomposes more rapidly and urea still more so,
perhaps twice as fast as cottonseed meal. The practice of side dressing
after leaching rains is only valuable if nitrates are used, other materials
serving only to increase the ease of application.

Effects of Nitrogenous Fertilizers 573

Soil Reaction as Influenced by Nitrogenous Fertilizers

The acidic character of nearly all the materials used for nitrogenous
fertilization of Connecticut tobacco is not widely recognized. Among the
fertilizers used in this test, only nitrate of soda may be considered neutral.
Calnitro, a mixture of ammonium nitrate and calcium carbonate, might
seem to be in the same class, but it is moderately acidic. Organic materials
produce a greater excess of acid ions in their decomposition than is com-
monly supposed, and are classed as moderately acidic. Urea, by its
decomposition into nitrates and carbonates or unstable bicarbonates, is
rather strongly acidic, but not as much so as sulfate of ammonia, which
rapidly depletes the soil of all soluble bases.

No lime has been added to the original plots since the experiment was
started in 1926. The block of plots added in 1929, castor pomace, linseed
meal and dry ground fish, was on an area not previously in tobacco, as
was calnitro, added in 1930. Corn gluten meal, and the second repli-
cation of cottonseed meal added in 1932, were on unlimed potash plots.
The second replication of castor pomace and the present cyanamid plot
are on old potash plots which were limed inl931 at the rate of 1000 pounds
of hydrated lime per acre. However, the pH data for the limed castor
pomace plot are not included in the tables, even though the present dif-
ference is not great. The Peruvian guano and soy bean oil meal plots are
located on an old Adco manure plot, which had proved to be almost as
basic as stable manure; consequently the soil reaction data are not com-
parable to the older parts of the field. An application of 500 pounds of
land plaster per acre to the older parts in the fall of 1933 was the only
soil amendment made during the 10 years of the experiment.

Results of soil reaction studies on these plots have been reported in
Tobacco Station Bulletin 10, Report for 1927; Conn. Agr. Expt. Sta. Bui.
311, Report for 1929; and 306 published in 1929. In the last-named, pages
796-797, may be found a study of the effect of various fertilizers on the
soil reaction. Bulletin 311, pages 264-268, discusses the seasonal fluctu-
ation of soil reaction for these plots from May, 1926, to December, 1929. '

At the time nitrate measurements were started, the soil on most of the
plots was already rather strongly acid. Sulfate of ammonia has become
much too acid to produce good tobacco, as indicated by pH readings of
3.50 and lower during the summer. Urea has dropped quite markedly
from 1927 measurements, while all the organic materials are somewhat
lower. Nitrate of soda has not changed the soil reaction greatly during
the past 10 years. The data for no nitrogen are a good index of the normal
behavior of this soil. As may be seen from Figure 130 and Table 14, the
absence of nitrogenous fertilizer results in an equilibrium between pH
5.00 and pH 5.60 which does not vary from year to year.

Seasonal fluctuation in relation to nitrates and rainfall. All

measurements of soil reaction indicate a maximum of soil acidity some-
time during the growing season. This usually corresponds with the
maximum of soil nitrates, but does not necessarily coincide with it. The
presence of other acidic materials such as sulfates usually leads to a gradual
lowering of pH readings until late summer. The cumulative effect of the
acid ions does not entirely disappear until heavy and persistent rains have
carried out of the soil most of the other acidic materials as well as the

1 See also Conn. Agr. Exp. Sta. Bui. 334. Soil Changes Resulting from Nitrogenous Fertilization.

574 Connecticut Experiment Station Bulletin 386

nitrates. An indication of this fact is to be found in the lower pH read-
ings in late September as compared to April. Nitrates have practically
disappeared by late September, but subsequent removal of other acidic
ions causes a further rise in pH readings of from .10 to .30 pH unit.

Fluctuations caused by the fertilizer material. A detailed exam-
ination of Tables 9 to 12 shows wide fluctuations in soil reaction for any
one material during the period of measurement. The range was most
extreme with nitrate of soda and urea, differences as wide as 1.55 and 1.44
pH units being recorded in 1935 for the two materials. The range as
indicated by the four-year averages, was greater for inorganic than for
organic carriers, but the least for no nitrogen. Differences in range be-
tween the organic materials were negligible, with dry ground fish and corn
gluten meal showing the slightly wider ranges. No correlation with the
range of nitrate determinations was apparent. It is quite obvious that a
general positive relation between nitrification and soil acidification exists,
but is not subject to detailed agreement.

Conclusions

From the data presented, the following conclusions may be drawn:

Soil nitrate levels vary widely as a result of the interaction of factors
favoring the accumulation of nitrates — such as high temperatures and
adequate moisture — with the factors favoring their removal, such as
heavy rains and rapid plant growth.

Relative availability of organic materials in relation to plant needs
indicate that all of those tested are about equally valuable. Rapidity
of decomposition is in the following increasing order: Cottonseed meal,
corn gluten meal, castor pomace, linseed meal, dry ground fish.

Inorganic materials range from immediately to slowly available. Ni-
trates are subject to too ready leaching, while sulfate of ammonia produces
abundant nitrates but is objectionable because of its acid character.
Urea is a very good source of fertilizer nitrogen.

Among new materials, soybean oil meal seems very promising.

The relation of rainfall to nitrates is discussed. Moderate rains, of not
more than an inch per week, greatly stimulate nitrification. Heavy or
persistent rains, of two or more inches in a short period, cause a heavy
loss of soil nitrates.

Speed of recovery after rains varies with the class of material. Vege-
table organics are the slowest followed by animal organics and urea.

Soil reactions determined concurrently indicate the acidic character of
all materials used except nitrate of soda. Seasonal fluctuations averaged
.67 pH unit for organic materials, .85 pH unit for inorganic, and .43 pH
unit for no nitrogen.

FURTHER FERTILIZER EXPERIMENTS WITH COTTONHULL ASHES

In a previous publication (Conn. Sta. Bui. 334: 207-210. 1932) the
writers described preliminary experiments on the use of cottonhull ashes
as the source of potash in the fertilizer mixture. Results up to that time
did not indicate any improvement in yield or grading from the substitu-
tion of cottonhull ashes for other carriers of potash, but the need of further
tests before drawing conclusions was stressed.

Fertilizer Experiments with Cottonhull Ashes 575

Since many growers reported good success with a formula made up
only of cottonhull ashes and cottonseed meal, a new series of field tests
was started in 1932 in which such a formula was to be compared with a
standard formula supplying the same quantity of nutrient elements but
with the potash derived equally from sulfate and carbonate.

The formulas were as follows:

Cottonhull ash formula (K16)

3000 lbs. cottonseed meal
560 lbs.* cottonhull ash

Formula without cottonhull ash (KIT)

3000 lbs. cottonseed meal
109 lbs. carbonate of potash
146 lbs. sulfate of potash

93 lbs. magnesian lime

49 lbs. precipitated bone

Each formula supplies 200 pounds of nitrogen, 109 pounds of phosphoric
acid, 200 pounds of potash, 49 pounds of magnesia and about 75 pounds
of calcium oxide to the acre.

The test has been continued for four years on eight plots, each one-
twenty-fifth acre in size, with the four cottonhull ash plots alternating
side by side with the control plots (without cottonhull ash). These plots
are located on field I on a medium heavy soil that does not leach much or
suffer from drought and naturally produces a high yield of tobacco. The
soil is uniform except that in 1933 a "dead furrow", in which the water
settled and drowned the plants, upset the data on one plot (K17) and
the effect was still visible in the two following years. Each season the
fertilizer was applied about 10 days before setting. Havana Seed plants
were set during the first week in June and all cultural operations were
uniform. The harvested tobacco hung in the same tier in the curing
shed, was taken down and stripped at the same time and sorted in the
Station sorting shop.

During the entire four years, close observation in the field failed to
show any differences between the plots with respect to size, habit of
growth, appearance, or maturity of the plants.

Observations on the cured tobacco as it was sorted showed that all
plots produced a good quality and there was little choice between them.
During one year, however, it was our impression that the tobacco from
the cottonhull ash plots was a little smoother, with less prominent veins.
This was reflected in the sorting records by a somewhat larger percentage
of the grade, "light wrappers".

A summary of the yields and the grade indices for the four years is
given in Table 15. This shows that the yield per acre on all plots is almost
the same. In fact it is almost identical if plot K17 (damaged by dead
furrow) is omitted. It is apparent, then, that no increase in yield may
be expected from the use of cottonhull ash.

^Amount changed each year according to the percentage of potash found in the ashes.

176

Connecticut Experiment Station

Bulletin 386

~
o

t~

LO

CO

<M

■<

**

T?

©

CSC

cc

u

<

^3

COVON^

© i— i CO ON

0

<M "* co **

CO CO CM ©

-

^1 ^3" "^ *^

T* "■* t? ""#

m

CM © Tf CO

© t— t— ©

Ml

© © © LO

I— LO T? •"#

C*

CO CO CO CO

CO CO CO CO

(M LO © ©

lo Tf r- t-

u

co

i— i Tt (N O

<N CO CM i— 1

a

-

■>* T* T? ■*

T* T* •>* TT"

£

>>

X

V

CO

LO © CO <N

CSI © LO ©

d

1— 1

co

Cs © © ©

vCOKf-

5

■* i/M/JlO

t* rf t? ■<*

<N

■*# co to in

T? CSI CO i— 1

CO

TfVOMT*

© CO LO ©

C"

T? ^ T*4 ^

"^ "^ ^* "*T

CO

©

5

<1

©

©

CM

CS|

03

6D

ce

c

>

^)

r-l ©■* ■*

l-H t- t— T*

<

o

vO^HM

CO ■■* CM ©

-

© © © ,— !

© © © ©

Cs| <M <M CM

i— i CS) CS| CSI

in

© uo co o

■5? T? © LO

CO

NtOHin

co © © ©

o\

CM I— 1 I— 1 CO

© CM i— I CM

CSI CM CM CM

r— i (M (M CSI

■>*

l> © CO ©

CO CO t- LO

0)

CO

© CM CM CM

■** Tjl r— CO

>H

9

© I— 1 I— 1 F— 1

©© © ©

>>

1-1

CM CM CM (M

i— i CS| csi <M

XI

T3

0)

£

CO

HftlflO

© r~ ©

CO

-* CO •"* ©

© © co

03

OS

OS ON CO ON

CO CO CO

i— 1 t— 1 r— 1 i-H

^H ^H 1— 1

<:

CSI

© co -^ os

© t~- CO ©

CO

i— 1 LO © OS

.— co © co

Ov

© OS © ©

© © © i—i

(MHHN

CM ^H CSI CS|

j=

03

a

o

C8

Cis

Ch

3

CO CO
CB-C

o

C
O

*J

"3 CO

D

+J

3 O

o

o

CO

03

U

d

t— 1 CM CO

^H CM CO

© © © ©

r-- t- f- t-

0

c

H

i4tti4i4

XXtXi*

Fertilizer Experiments with Cottonhull Ashes 577

Considering the grade index figures, it appears that there has been a
small but rather constant difference in favor of the cottonhull ash tobacco.
Whether this difference is large enough to be significant is mathematically
questionable. The most we can conclude from this series of tests is that
the use of a formula composed of cottonhull ashes and cottonseed meal
has produced very excellent tobacco which is just as good, or possibly a
little better, than that grown with a similar formula where other sources
of potash are substituted.

Effect on the soil. Since this material contains considerable quan-
tities of alkaline salts (potash, lime, magnesia), there has been some appre-
hension lest the continuous use of cottonhull ashes over a series of years
on the same field might make the soil alkaline, or result in an unfavorable
accumulation of salts. In order to see whether such effects could be
detected after four years of treatment with 500 to 600 pounds of ashes
annually, soil samples from each of the eight plots, taken in November of
the fourth year, were analyzed. There was no significant difference in
the amount of potassium, calcium, magnesium, phosphorous or nitrogen,
as between the cottonhull ash plots and the control plots. The use of
ashes, therefore, has not resulted in any greater accumulation of these
salts in four years.

However, there has been some decrease in acidity. Each cottonhull
ash plot tested somewhat higher (pH) than its adjacent control. In
November, 1935, the average reaction of the ash plots was pH 5.07 and
of the controls was pH 4.91. Such an increase in reaction, however, is
not injurious on this field and would not be dangerous except on soils that
naturally have a high reaction.

Effect on the burn. In order to see whether cottonhull ashes have
any specific effect on the burn of tobacco, both strip tests and cigar tests
were made.

The object of the strip test is to determine the fire-holding capacity
of the leaf when a single leaf is stretched between the two hands. The
leaf is ignited from an electrically heated wire coil and the duration of
combustion (until last glow disappears) determined with a metronome.
Twenty tests of the "seconds" and twenty of the "darks" (fermented
crop of 1933) from each plot were made, making a total of 160 tests for
each treatment. The average duration of burn for all tests on the cotton-
hull ash tobacco was 41.5 seconds; for the control formula it was 43
seconds. The difference between them is too small to be significant and
thus there is no indication here that the use of cottonhull ashes has had
any influence on fire-holding capacity.

As a further comparison, 50 cigars were wrapped with leaves from the
cottonhull ash plots and another 50 from the check plots. These were
smoked in pairs by a considerable number of smokers. No consistent
differences were observed in color or coherence of the ash, width of coal
band or in the taste and aroma, of the cigars.

Effect on chemical composition. Cottonhull ash is used chiefly to
supply potash but it also supplies calcium and magnesium. In order to
see whether the tobacco plant absorbs more of any of these bases from
the cottonhull ash formula than from the check formula, samples of two

578 Connecticut Experiment Station Bulletin 386

grades, seconds and darks, were analyzed from each of the eight plots.
The average content of bases from all plots is shown in Table 16.

Table 16. Percentage of Bases in Tobacco from Cottonhull Ash Plots
and Adjacent Check Plots.

Fertilizer

Total ash

Potash
KsO

Lime
CaO

Magnesia
MgO

Cottonhull ashes
No ashes

17.84
17.76

4.23
4.43

6.03
5.63

1.95
2.06

It appears from these analyses that the basic content is not significantly
changed by the use of cottonhull ash.

FURTHER INVESTIGATIONS
ON THE USE OF FERTILIZER MAGNESIA

Work on magnesia in tobacco fertilization has been carried on for a
number of years at this Station. The latest extensive report of results is
found in our annual report for 1932 (Conn. Agr. Exp. Sta. Bui. 350). In
the following pages we include further studies on the effect of magnesia
on yield, grading and chemical composition; estimation of soil magnesia
and its relation to crop needs ; as well as a discussion of conditions govern-
ing the color of ash. In addition, results are given on tests with a neutral
magnesium salt, an anhydrous magnesium sulfate.

Color of Ash

Experienced cigar smokers frequently correlate the taste with the
appearance of the ash on the cigar. A dark or muddy ash in most cases is
associated with poor taste and aroma, while good cigars produce lighter
colored ashes.

An excessive amount of potash in tobacco makes a dark ash, and the
color is roughly proportionate to the excess of this element in comparison
to other bases. The dark color of the ash is due to the presence of un-
oxidized particles of carbon (charcoal) ; i.e. the color range — white, through
gray to black — indicates the degree of combustion. A high content of
potash causes incomplete combustion, since potassium salts have a tend-
ency to fuse and enclose the carbon particles, thus excluding the air and
preventing oxidation.

Since the gaseous products of incomplete combustion are less pleasant to
the taste and smell than those of complete combustion, it follows that
the darker the ash color, the less pleasant the taste and aroma.

Previous investigations at this Station have shown that the dark color
due to excess potash may be changed to a desirable light color by increas-
ing the magnesia, the compounds of which do not fuse when burned but
produce a porous ash that permits more complete combustion.

Then there is the familiar muddy or brick -colored ash which frequently
appears on cigars, irrespective of brands. The cause of this condition

The Use of Fertilizer Magnesia

579

has been revealed only recently. A study of the carbonates or oxides
(which constitute the ash) led us to believe that manganese compounds
may be a contributing factor in discoloring the ash, since there was a
resemblance in color between certain manganous compounds and the
brick-colored ash.

Samples of tobacco, known to produce brick-colored ash when wrapped
on a cigar, together with samples producing desirable grayish ash, were
analyzed with respect to their manganese content. The results are given
in Table 17.

Table 17. Content of Manganese (as Mn304) in Cigar Leaf Tobacco
and Color of Ash. Air Dry Basis

Sample No.

% M113O4

Ash color

1

.023

2

.034

3

.047

4

.057

5

.064

6

.067

7

.081

8

.277

9

.302

Light gray ash, no brick color

" very slight brick color
Gray ash with slight brick color

Very heavy brick color

From this table it may be concluded that brick-colored or muddy ash
may be prevented when tobacco contains less than .04 per cent Mn304.
How is it possible to avoid the brick-colored ash? A previous study* of
manganese in Connecticut soils showed that active manganese in the soil
increases as the soil becomes more acid, i.e. below pH 5.0 we may expect
the soil to contain sufficient active manganese to influence seriously the
color of ash.

With respect to the samples in Table 17, it should be stated that samples
1 to 6 were grown on soils with a reaction ranging from pH 5.40 to pH
6.15, while samples 7 and 8 were derived from soils with pH values of
4.90 and 4.45 respectively. The remedy then is to raise the reaction of
the soil enough to inactivate the manganese, which is accomplished by
liming.

Although calcium is the active element governing the absorption of
manganese**, magnesian lime may be used in liming an acid tobacco soil,
since such a soil is usually also low in magnesia.

Finally, there is the light gray to whitish ash which in extreme cases
will flake, thus lacking in coherence. This latter condition, as has been
fully discussed in our report for 1932 (Conn. Agr. Exp. Sta. Bui. 350) is
caused by the presence of too much magnesia as compared to other bases.
The work at this Station has been continued on the problem of determin-
ing the proper quantities of magnesia to apply to tobacco soils for obtain-
ing a desirable ash. The latest results are reported below.

* H. G. M. Jacobson and T. R. Swanback. Manganese content of certain Connecticut soils and its
relation to the growth of tobacco. Jour. Anier. Soc. Agron. Vol. 24: 237-245, 1932.

**Hans Burstrom. Uber antagonistische Erscheinungen bei der Kationenaufnahme des Hafers.
Upsala, 1934.

580 Connecticut Experiment Station Bulletin 386

Annual Applications of Magnesian Lime

Iii our previous experiments, reported in 1932 (see above) one-time
applications of magnesian lime were made and in succeeding years the
duration of effects of the treatments observed. It was concluded from
these results that long time residual effects could not be depended on, but
annual application of not more than 100 pounds magnesium oxide to the
acre should be applied. In order to determine more definitely the size
of the annual application, experiments were begun in 1933 with a set of
one-fortieth acre plots, formed by dividing the plots crosswise of the
original lime series, and treated as follows:

0 lbs. Magnesia (MgO) per acre (in hydrated magn. lime)

50 " " " " "

75 " " " " "

100 " " " " "

125 " " " " "

All plots received general fertilizers, composed of cottonseed meal,
castor pomace, nitrate of potash and sulfate of potash to equal about 200
pounds nitrogen and 200 pounds potash per acre.

Effect on yield and grading. There are no conspicuous differences
to be found in the results from the various amounts of magnesia applied
in regard to yield and grading, which may be learned from the three-year
summary in Table 18. It is obvious that from a practical standpoint,
moderate additions of magnesian lime to tobacco land will not alter the
yields or appearance of the tobacco grown on it. Smoke tests and the
chemical composition of the leaf, however, may reveal other qualities.

Table 18. Yield and Grading Records of Magnesia Plots. Summary of
Three Years Results, 1933-1935 (Quantitative Series)

Pounds MgO per
acre annually

1933

—Yields
1934

per acre
1935

Av.

1933

- Grade
1931

index —
1935

Av.

0

1655

1918

1984

1852

.535

.430

.383

.449

50

1710

1917

2032

1886

.550

.431

.358

.446

75

1703

1840

1988

1844

.535

.418

.343

.432

100

1823

2036

1983

1947

.568

.424

.335

.412

125

1862

1961

1947

1923

.543

.426

.337

.435

Smoke Tests. For the purpose of observing the effect on the ash
color, leaves from tobacco grown with the various quantities of magnesia
were wrapped on cigars, which were then smoked. It was noticed that
tobacco from the check plots, where no magnesia other than that inci-
dentally introduced through the organic materials had been applied to

The Use of Fertilizer Magnesia

581

the land, produced a dark, muddy ash, with bad, sharp taste. With the
50-pound application, a lighter ash was obtained but still a slight brick
color adhered to it. The 75-pound application produced as nearly a perfect
ash as could be desired — a very light gray to whitish, without flaking and
with no objectionable taste and aroma. Similar results followed from the
use of 100 pounds, although a tendency to flakiness occurred. The highest
application (125 pounds) produced an ash that was altogether too flaky.
From the results of these tests, it is evident that an annual application
of about 75 pounds of magnesia in the form of magnesian lime would be
a suitable amount.

Chemical Analyses. Samples of tobacco grown on these plots
during 1933, 1934 and 1935 were analyzed with respect to content of
magnesia. The results are shown in Table 19.

Table 19. Content of MgO in Tobacco Leaves of 1933, 1934 and 1935
Crops. Air Dry Basis

Grade

of
tobacco

i<ks:>,

1934

1935

Lbs. MgO
per acre

%MgO

Av. of
treat-
ments

%MgO

Av. of
treat-
ments

% MgO

Av. of
treat-
ments

None1

D1

1.05

1.09

1.10

S3

1.35

1.19

1.07

D

.99

1.19

1.08

1.12

1.11

1.06

S

1.38

1.14

.95

50

D

1.45

1.48

1.42

S

2.01

1.99

1.51

D

1.28

1.68

1.56

1.79

1.58

1.60

S

1.99

2.14

1.89

75

D

1.59

1.55

1.53

S

1.76

2.20

2.04

D

1.61

1.77

1.82

1.99

1.67

1.97

S

2.17

2.38

2.62

100

D

1.92

2.17

1.88

S

2.46

2.53

2.45

D

2.10

2.31

2.22

2.39

1.76

2.16

S

2.76

2.63

2.54

125

D

2.30

2.42

1.73

S

3.15

3.07

2.65

D

2.45

2.85

2.63

2.83

1.53

2. 12

S

3.55

3.22

2.56

Here it is evident that with an annual application of 75 to 100 pounds
of MgO per acre, producing a satisfactory ash, the leaf will contain about
2 per cent MgO, which is in agreement with our previous findings.

Magnesia Application Based on Microchemical Soil Tests

On 10 plots which in 1930 received varying quantities of magnesian
lime per acre (equivalent to 0, 100, 200, 400 and 600 pounds MgO), the

1 These plots were limed in 1930 at the following rates, respectively: 0, 100, 200, 400 and 600 pounds
MgO per acre.

3 "D" and "S" — "darks" and "seconds" from duplicate plots.

582

Connecticut Experiment Station

Bulletin 386

residual effects of these treatments were observed on the 1933 crop of
tobacco.

In the spring of 1934, it was decided to make microchemical tests* of
the soil on each individual plot of this series for estimation of available
(replaceable) magnesia. With these figures in mind, we planned to add
sufficient magnesian lime to make the total available MgO in all instances
100 pounds to the acre, including the amounts derived from the cotton-
seed meal and other fertilizer materials.

In 1935 the tests were repeated. The quantities of MgO in pounds per
acre applied in each of the two-year periods are given below:

A, Aj control plots

No magnesia

(other than

in organics)

B

100

45

100

Bx

100

65

300

C

200

5

100

Q

200

45

200

D

400

0

100
100

A

400

5

E

600

0

0

Ex

600

5

0

To all plots the same general fertilizer was applied as reported above for
the annual applications.

The three-year summary (Table 20) of results on the residual and equali-
zation tests give further evidence that moderate quantities of magnesian
lime affect neither yield nor grading.

Table 20. Yield and Grading Records of Magnesia Plots. Summary of

Three Years' Results, 1933-1935

(Residual Series. MgO Equalized in 1934-35)

Pounc
acr«

s MgO per
per year

1934 1935

Plot No.

Yields per acre

Grade Index

1933

1933

1934

1935

Av.

1933

1934 1935

Av.

0

0

0

A **

A,

1698

1895

1959

1851

.536

. 405 . 390

.444

0

45
65

100
300

B
Bx

1748

1825

1998

1857

.538

.419 .378

.445

0

5
45

100
200

C
Cx

1658

1748

1848

1751

.543

.422 .358

.441

0

0
5

100
100

D

1626

1798

1870

1765

.562

.424 .379

.455

0

0
0

0

E

Ex

1850

1789

2070

1903

.538

.451 .380

.456

Chemical Analyses. In the same manner as reported above, samples
of tobacco were analyzed for their content of magnesia. Results from
these analyses are given in Table 21.

* Morgan, M. F. Microchemical Soil Tests. Conn. Agr. Exp. Sta. Bui. 333. 1932.
** Listed in Station records as L9, L9-1, Liu, L10-1, Lll, Lll-1, L12, L12-1, L13, L13-1.

The Use of Fertilizer Magnesia 583

Since it is desirable to obtain about 2 per cent MgO in the ash, the
results indicate that the microchemical soil tests can safely be relied on
when there is already a relatively high content of soil magnesia. In
attempting to estimate the need of magnesia application through the use
of the microchemical soil test (more recently called Universal Soil Testing
System for Connecticut Valley), proportionately more magnesia should
be applied, the less magnesia the soil contains, in order to attain desirable
results. In other words, if it is decided to raise the magnesia content of
a soil containing about 25 pounds per acre, more than 75 pounds should
be added to form some 100 pounds available (replaceable) magnesia. A
lower content of magnesia in the soil is often accompanied by a higher
acidity (lower pH value) and it is possible that when magnesia is applied
in the form of magnesian lime, a part of it in the neutralizing action of
the lime becomes fixed and unavailable for plants. Further investigations
are needed to warrant recommendations on magnesia applications based
on soil tests.

Table 21. Content of MgO in Tobacco fbom "Residual" and "Equalization"
Plots. Crops of 1933, 1934 and 1935. Air Dry Basis

Grade

of
tobacco

1933

1934

l

193E

%MgO

Av. of
treat-
ments

% MgO

Av. of

treat-
ments

%MgO

Av. of

treat-
ments

Control

D

1.13

1.06

.94

AAj

S

1.37

1.18

1.17

D

1.14

1.26

1.03

1.14

.71

.93

S

1.41

1.29

.88

BBj

D

1.31

1.37

1.33

S

1.53

1.65

1.55

D

1.38

1.51

1.79

1.76

1.91

1.87

S

1.83

2.23

2.70

CQ

D

1.61

1.57

1.27

S

2.04

1.86

1.76

D

1.46

1.84

1.69

1.88

1.89

1.82

S

2.24

2.42

2.37

D Bl

D

1.66

1.82

1.96

S

2.04

2.27

2.03

D

1.64

1.96

1 64

2.01

1.77

1.96

S

2.50

2.30

2.08

EE,

D

2.15

2.19

1.30

S

3.29

2.68

2.08

D

1.77

2.41

1.72

2.19

1.58

1.73

S

2.44

2.19

1.94

Anhydrous Magnesium Sulfate (Kieserite, EMJEO)
as a Source of Magnesia

Anhydrous (without water) magnesium sulfate under the commercial
names of Kieserite and EMJEO recently have been put on the market
and are offered as substitutes for, or a supplement to, magnesian lime.
The materials contain about 30 per cent MgO and being soluble salts
might be expected to be more readily available.

584 Connecticut Experiment Station Bulletin 386

With the purpose of testing the availability of this magnesium salt in
comparison with magnesian lime, an experiment was made in 1934 with
one plot receiving the magnesium sulfate, another the magnesian lime
(hydrated) and a control plot with no special carrier of magnesia. The
magnesia was applied at a rate of 100 pounds MgO to the acre together
with a complete fertilizer mixture which was identical for all three plots.

The yield and grading is found in the summary below :

Treatment Yield, lbs. per acre Grade index
Control 1810 .367

Magnesian lime 1753 . 314

Kieserite 1863 .353

There seems to be a slight falling off in yield and grading for the
magnesian lime, an unexpected development in consideration of our
previous experience with this source of magnesia. The result with Kieserite
was about equal to that with the control; thus with respect to yield and
grading the use of Kieserite should prove satisfactory.

Chemical analysis. Samples of "darks" and "seconds" from the
three plots mentioned above were analyzed with respect to their content
of magnesia. The results computed on the basis of air-dried leaf are
given in Table 22.

Table 22. Content

OF

MgO in Toba
Aib Dry

ceo Le
Basis

aves from Kieserite Test.

Treatment

Per cent MgO

Darks

Seconds

Av.

Control

Magnesian lime
Kieserite

1.026
1.427
1.645

1.023
1.717
1.943

1.02
1.58
1.80

Results from these analyses indicate that magnesia supplied in the
form of sulfate is somewhat more readily absorbed by the plants than
magnesia in the form of magnesian lime. In this test the sulfate proved
to be about 10 to 14 per cent more efficient than the lime; hence in using
this material one should be guided by this apparent difference in availability
of the magnesia in these two materials.

This magnesium sulfate (Kieserite, EMJEO), being a nearly neutral
salt, should not affect the soil reaction either Avay, and may be recom-
mended for tobacco on soils having a desirable soil reaction (pH) and a
satisfactory calcium content, but in need of considerable magnesia.

Summary and Conclusion

The amount of magnesia in the fertilizer did not greatly affect yield
or grading of the crop but had a decided influence on the combustion
of the leaf. The color of the ash of a cigar is an index of its chemical
composition. Thus a relative excess of potassium may produce a dark
ash, while increasing amounts of magnesium will brighten the ash pro-
portionately. An excess of manganese in the ash has been found to give
it an undesirable color known to the tobacco trade as "muddy" or '"brick
color." At a minimum of less than .04 of 1 per cent manganese (Mn3OJ
in the leaf, the discoloration does not appear.

Harvest Time for Havana Seed Tobacco 585

In order to produce a proper combustion of the leaf, an annual applica-
tion of 75 to 100 pounds magnesia in the form of magnesian lime has
been found preferable to larger applications at intervals of several years.
Chemical analyses have revealed the presence of about 2 per cent MgO
in the leaf at the above rate of application. Smoke tests have corroborated
these findings.

An attempt was made to determine the need of magnesia applications
through the use of microchemical soil tests. The results are not conclusive,
but indicate the possibilities of the method, which, however, requires
further investigation.

An anhydrous magnesium sulfate (about 30 per cent MgO), sold under
the commercial name of Kieserite or EMJEO has been tested as a source
of magnesia. It was found that this material is 10 to 14 per cent more
efficient than magnesian lime. It may be used on land in greater need
of magnesia where the soil reaction might be adversely disturbed when
larger quantities of magnesia were to be applied.

EXPERIMENTS TO DETERMINE THE BEST TIME
TO HARVEST HAVANA SEED TOBACCO

When the first blossoms open, it is time to break the tops off the tobacco
plants. Most growers agree on this point but there is considerable diver-
sity of opinion and practice as to how many days should elapse between
topping and harvesting. Naturally the period varies with the character
of the .weather.

All growers recognize the signs of leaf ripening, such as yellow mottling,
down curving of the margins, and the thickening of the blades, and
they try to harvest when they think the tobacco is ripe. But the ripening
process extends over a considerable period of time. It begins with the
lower leaves and progresses upward for a week or longer before the top
leaves show the proper symptoms. Some growers cut the tobacco when
the center leaves have reached this stage; others prefer to have the top
leaves quite ripe. In the latter case, the lower leaves become overripe
and are spoiled, or at least of inferior quality. If the tobacco is harvested
too soon after topping, the leaves are green and thin and it is a common
belief that they do not burn so well and are inferior in taste and aroma.

There are no recorded experiments in which an attempt has been made
to determine the time at which Havana Seed or Broadleaf tobacco should
be harvested in order to secure the best quality and the maximum yield.
Therefore, such an investigation was planned, beginning in 1935, to be
carried out through a series of years in order to minimize the effects
of seasonal variation. Some interesting results of the first year's experiment
are presented at this time.

From a field that was quite uniform in appearance and growth, three
plots, each of three rows and containing 400 plants, were used. All plants
were topped on July 22. After removing the border rows, the first plot
was harvested one week after topping, the second plot two weeks after
topping, and the third plot three weeks after topping. All plants were
hung on the same tier in the curing shed. The excessive rains of June
and July caused a shortage of nitrates in the soil and many yellow, starved

586

Connecticut Experiment Station

Bulletin 386

leaves were found at sorting time. This starvation complicated the
data and must be kept in mind in interpreting the year's results. The
tobacco harvested the first week after topping was obviously too green,
according to our standards, while that harvested three weeks after topping
was completely ripe, even to the top leaves, possibly overripe. It was
observed in the curing shed that the plants harvested last cured in the
shortest time, while those harvested one week after topping cured very
slowly and there were still green leaves long after the others had cured.
Thus it was demonstrated strikingly that the maturity of the leaves at
harvest has considerable influence on the speed of the cure.

On the sorting bench, a distinct difference was also evident. Tobacco
harvested one week after topping was thinner, somewhat shorter, and
all grades had an olive green cast. It appeared silky and of excellent
quality otherwise. From the later harvestings the leaves were distinctly
heavier, coarser with more prominent veins, longer, and the green cast
was gone. There was a higher proportion of starved yellow leaves, classed
as "brokes", especially among those harvested three weeks after topping.
Thus the grade index was considerably reduced. Presumably this fault
would not be so damaging during a season of less rainfall.

S"

\ /

y

'~"~V-

v' \
\

4-

i /

/

\
\
\
\
\

1

of'

\

\

\

/■-

\ \

LENGTH OF LEAVES IN INCHES

Figure 133. Leaf length as affected by time of harvesting.

The yield and grading records taken at the time of sorting are presented
in Table 23. The figures show that there was a steady increase in weight
after topping up to at least three weeks. In order to see how much of
this might be due to increase in the actual size of leaves, they were sized
(intervals of 2 inches) and the percentage of the crop in each length was
computed. The results, presented graphically in Figure 133, show that
the actual leaf area increases continuously after topping.

Harvest Time for Havana Seed Tobacco

587

In order to see whether there was an increase in the absorption of
salts from the soil, and whether a part of the increase in weight was due
to accumulation of mineral salts, analyses were made of samples from
two grades of each of the lots harvested at different periods after topping.

Table 23.

Experiments on Time of Harvesting. Sorting and Yield Records
for 1935

Time of harvesting

Yield in pounds
per acre

Percentage of Grades

Grade

L

M

1
1

LS

24
30
19

ss

9
4
1

LD

39
43
45

DS

13
6
4

F

10
8
8

B

4

8

index

One week after
topping (July 29)

1518

.334

Two weeks after
topping (Aug. 5)

1856

.355

Three weeks after
topping (Aug. 12)

2003

23

.291

The results presented in Table 24 show that there has been no actual
increase or accumulation of salts as expressed in percentage of the dry
weight of the leaves. In other words, the salt intake has merely kept
pace with the increase in size and weight of the leaves so that the more
mature leaves do not contain proportionately any more mineral elements
than the green leaves.

Table 24. Analyses of Leaves Harvested at Different Intervals

After Topping

Interval after

Grade

Percentage of

topping

Total ash

Potash (K2O)

Lime (CaO)

Magnesia (MgO)

1 week

D

S

18.24
20.53

5.70
5.92

5.46
5.68

1.37

1.54

2 weeks

D

S

16.67
19.54

5.16
5.20

4.23

6.42

1.09
1.41

3 weeks

D

S

15.98
19.65

4.65
5.21

4.93
6.06

1.22
1.47

TOBACCO INSECTS IN 1935

Donald S. Lacroix
Prevalence of Insect Species 1935

The spring of 1935 was cold, and both plant and insect life were from
ten days to two weeks later in development than in seasons past. This
was quite noticeable in the case of the potato flea beetle1, the overwinter-
ing adults of which did not appear in tobacco seedbeds until after the

1 Epilrijc cueumeris Harris.

588

Connecticut Experiment Station

Bulletin 386

middle of May. There was a moderate to heavy infestation of this insect
throughout the Valley on young tobacco plants in the field in early June.

During the last week in July and the first week in August, a heavy
infestation of the potato flea beetle developed on Broadleaf tobacco
in East Windsor, South Windsor, and Ellington and caused serious injury.
Only by thorough and constant dusting was the pest held in check on
Shade-grown tobacco.

Wireworms1 were as injurious as usual, taking the tobacco-growing
area as a whole. In some isolated instances they worked later than they
did last season.

Cutworms of various species were very persistent in 1935. The black
cutworm2 injured Broadleaf plantations around Windsorville during the
last part of June, and on plantations not treated with poisoned bran bait,
infested about one-third of the plants.

Figure 134. Larva of dark-sided cut\Vbrm.
(One and one-half times enlarged.)

Another species, the dark-sided cutworm3, discovered in seedbeds in
Windsor the third week in May. was also taken in late June and early
July in the field.

The spotted cutworm4 was observed the third week in July on the
outside rows of Broadleaf tobacco at Warehouse Point. A field of alfalfa,
adjoining the tobacco, had just been cut, and the cutworms were migrating
into the tobacco in large numbers.

The well-marked cutworm5 was not seen on tobacco this year.

A few specimens of the common stalk borer6 were brought in from
East Hartford on Shade-grown tobacco in mid-July.

The tobacco budworim was far more injurious to tobacco on the Experi-
ment Station farm at Windsor than it had been in the past. Unlike some
of the other pests in 1935, there was no delay in its seasonal appearance
as the first brood of larvae was active from early to late July and the
moths emerged early in August. Injury from this pest was found in East
Granby, West Suffield, East Windsor, South Windsor and Manchester.

The tobacco thrips8 caused more injury to Shade-grown tobacco during
1935 than it did in 1934, but was not so serious as it was during the dry

1 Pheletes eetypus Say.

2 Agrolis ypsilon Roll.

3 Euxoa messoria Harris.

4 Agrolis c-nigrum Linn.

5 Agroils unicolor Walker.

6 Papaipema nitela Guen.

7 Heliolhis virescens Fabr.

8 FranklinielUi fusca Hinds.

Insect Pests in 1935

589

season of 1933. This species was taken in small numbers on Havana
Seed tobacco and Broadleaf tobacco throughout the Valley.

Grasshoppers were normally abundant. Several cases of injury by
the redlegged grasshopper1 to Shade-grown tobacco came to our attention.
In each case the land had been idle for a year or two before tobacco was
planted or had been in hay. In West Suffield the Carolina grasshopper2,
was more abundant on Havana Seed tobacco than any other species.

Hornworms of both species3 were normally abundant in and around
Windsor and about a week later than usual in first appearing. Through-
out the Broadleaf plantations of East Windsor, South Windsor, and
Ellington, they were not so numerous.

Figure 135. Dark-sided cutworm injury to plants in seedbed.

There was practically no injury to tobacco from the tarnished plant
bug4.

The spined stink bug5 caused little trouble this season.

Insect Control Experiments

Tobacco bud worm.6 The increase in population of the Tobacco
budworm on the Experiment Station farm at Windsor necessitated prompt
control measures. The recommended poisoned corn meal bait used in
the South was applied. This bait was prepared by mixing 25 pounds
of corn meal with one pound of arsenate of lead. A "pinch" (about half

1 Melanoplus femur-rubrum De G.

2 Dissosteira Carolina Linn.

3 Phlegelhontius quinquemaculata Haw., and P. sexia Johan.

4 Lygus praiensis Linn.

r> Enschislus variolarius Beauv.
6 Heliolhis virescens Fabr.

590

Connecticut Experiment Station

Bulletin 386

a teaspoonful) of the mixture was dropped in the bud of each of 25 infested
plants (July 18, 1935). Twenty-four hours later these were examined
and compared with 25 untreated plants. Dead worms were found on
23 poisoned plants and none on two. Of those untreated, 21 had live
worms and 4 had none. This insect is not gregarious; only one worm
is found on a plant in most cases.

Potato flea beetle.' Recent events have led agriculturalists away from
the use of insecticides poisonous to man. This is particularly true in the
case of plants grown for human consumption.

Barium fluosilicate has proven most satisfactory as a poison for the
potato flea beetle on tobacco but there appears to be some question as
to its effect upon human beings. Comparatively large quantities must be
ingested by a person to cause any trouble, and the amounts used in dusting
tobaco are very small.

Figure 136. Dark-sided cutworm injury to young plant in field.

However, tests have been made with pyrethrum and derris compounds
during the last few years, and in every case their toxicity to flea beetles
has been pronounced. Unfortunately their insecticidal quality is not lasting
and repeated dusting is necessary. In 1933 and 1934 "Cubor" (.5 per cent
rotenone) and pyrethrum dusts were used successfully. In 1935 "Cubor"
(.75 per cent rotenone) dust was used on several plantations with satisfac-
tory results.

Another dust containing rotenone and known as "Ku-Ba-Tox" was
tried on Shade and Havana Seed tobaccos during 1935. At full strength
this material killed flea beetles rapidly and retained its toxicity for three
or four days. When diluted with equal parts of tobacco dust, it gave
equally good control. When mixed with equal parts of "Dutox,"? immediate
control of the flea beetle resulted and the "Dutox" retained its killing
action, keeping the beetle population down for more than a week. Table
25 indicates in actual figures just how the materials worked out.

No injury to the tobacco foliage was apparent on any of the treated
plots.

The figures below indicate that the materials used were equally effective
in control of the beetle, but the tests ended prematurely as the insects

1 Epitrix cucumeris Harris.

2 Barium Jiuosilicate

Insect Pests in 1935 591

disappeared from the whole Station plantation (even from untreated
plots) while these experiments were in progress.

Flea beetle resistant tobacco. Several different types of Havana
Seed tobacco are grown at the Experiment Station in the strain studies.
It was observed late in July that certain strains were much more seriously

Table 25. Insecticide Tests on the Potato Flea Beetle
(Havana Seed and Broadleaf Tobacco)

Number

of Flea Beetles per

25 plants

Material used

At time of

24 hours after

48 hours after

treatment

treatment

treatment

"Ku-Ba-Tox" and

tobacco dust,

178

31

26

equal parts
"Ku-Ba-Tox" and

"Dutox", equal

257

30

18

parts
"Dutox" and tobacco

dust, 1:3

204

32

21

Check

193

236

283

infested with the potato flea beetle than others. A series of population
counts showed that the Brown strain (No. 8) was much more seriously
injured than the No. 211 strain. The following table (Table 26) contains
the figures resulting from these counts.

Table 26. Prefebence of the Potato Flea Beetle fob Cebtain Tobacco Strains

Strain Number of beetles per

25 plants

Adjacent rows „ 1 ^"l

.,. . 211 257

Adjacent rows „ i 491

(10 rows from nearest ^^ Ar.

No. 8 strain) iLL . 4U

Another observation made in connection with these studies indicated
that the whole plantation was not so seriously infested this year as in
the past. All the Havana Seed tobacco at the Experiment Station (with
the exception of the strain test plots) was No. 211 in 1935, and in previous
years was the Brown strain. Then, too, the Broadleaf and Shade tobacco
plots were well infested with flea beetles this year while the No. 211 strain
was not. All these facts naturally bring up the question, "Is No. 211
Havana Seed tobacco distasteful to flea beetles?"

Tobacco thrips.1 Many attempts have been made during the last few
years to find a dust which would control the tobacco thrips, and in every

1 Frankliniella fusca Hinds.

592 Connecticut Experiment Station Bulletin 386

case the material used has been most unsatisfactory. Various pyrethrum
and derris dusts have been compounded with carriers of all types and
given thorough trials. On the other hand, sprays containing pyrethrum
or derris have consistently given fair to good control.

During 1935 investigations were continued on control measures for
thrips and the same results were obtained. Dusts proved of no value
while sprays gave good results. "Ku-Ba-Tox" liquid spray and "Cubor"'
spray were promising in preliminary tests and were tried again on a larger
scale. The following table shows how these materials worked under field
conditions.

Table 27. Insecticide Tests on Tobacco Thbips Control (Shade Tobacco)

Number of thrips

on 25

leaves

Material applied

24 ho

irs later

48 hours later

July 29, 1935

Dead

Alive

Dead

Alive

"Ku-Ba-Tox" and

57

0

51

1

water, 1:400

"Cubor" and water.

2 lbs. to 50 gals.

64

2

47

1

"Loro" and water,

1:600 with soap

47

11

39

15

Check

0

34

0

41

"Ku-Ba-Tox" and "Cubor" sprays gave a very good kill but "Loro"'
was only fair. "Cubor," when used at the strength indicated, must be
very carefully mixed and used in an absolutely clean sprayer or it will
not all dissolve. This, of course, leaves an undesirable residue. None
of the materials injured the leaf.

In view of these encouraging results, it seems most advisable to carry
on further spray trials on a large scale to determine how often and at
what times to apply them. In any case, it would be advisable to start
spraying in late June when the plants are comparatively small and when
the thrips first appear.

Wireworm investigations. Studies on the life history of the Eastern
Field Wireworm1 have been continued in an effort to get all information
possible on this pest. Particular attention has been devoted to the earlier
stages in the life of this insect. Also a study of the development of infesta-
tions on a number of fields is being continued over a period of years to
accumulate desirable ecological data.

1 Pheletes ectypus Say.

Tobacco Diseases in 1935 593

TOBACCO DISEASES IN 1935
P. J. Anderson

Diseases in general were somewhat more prevalent in 1935 than in
the two previous years, probably due to the greater rainfall. Some of
them were of sufficient importance to warrant special investigations
described below. The occurrence and degree of severity of the rest are
also recorded here. Most of our tobacco diseases are of minor importance,
but some of the major ones should be more thoroughly investigated.

Black rootrot (Thielavia basicola) was found causing considerable
damage in some of the seedbeds that had not been steamed. All the
symptoms of yellow, stunted plants and rotted black roots were evident.
If our beds were not sterilized so regularly, this disease would cause a
great deal of trouble in Connecticut. Only once in 11 years has the
writer found a serious case of black rootrot in steamed beds. As a rule,
this disease does not often seriously affect Broadleaf in the field. In
1935, however, a two-acre area in a field in South Windsor produced
such small stunted plants that the crop was hardly worth harvesting,
while the rest of the field had tobacco of normal size. The roots of the
dwarfed' tobacco were seriously affected with rot. This part of the field
was nearest the barn and heavy applications of manure had been applied,
a treatment that is known to favor black rootrot.

Wildfire {Bacterium tabacum). After the serious epidemics of wild-
fire in the early "twenties", the disease became less and less prevalent
here with each succeeding year. By 1934 it had practically disappeared
from the Valley and there was hope that we would not be troubled with
it again. No cases were found in the seedbeds in 1935, but in the late
summer it appeared again in two or three localities and in some cases
the fields were seriously affected at the time of harvesting. Beports from
other sections of the United States suggest that after years of limited
occurrence, wildfire may again be spreading. This year was marked
by an unusually severe epidemic in Western Tennessee and Kentucky.
It warns of the necessity of continuing the precautionary control methods
that were developed and practiced by most of the growers when the
disease was more prevalent.

Blackfire (Bacterium angulatum), a bacterial disease that is destruc-
tive in the more southern tobacco growing areas, was observed in Con-
necticut both on Broadleaf and Havana Seed this year in small amount.
It is not a conspicuous or a serious disease here and does not seem to
be spreading, although it can always be found somewhere. On one large
field of Broadleaf this year, 10 per cent of the plants were affected. The
spots are usually larger, less numerous and more definite in outline on
Broadleaf than on Havana Seed. Usually only a part of the leaves on
a plant are affected; in the field under consideration most of these were
the bottom leaves. The dead spots vary in size from an eighth to a half
inch or more in diameter. Sometimes the spots run together and kill
large areas of the leaf. The most characteristic feature, as distinguished
from other types of spots on tobacco, is the irregular, angular or jagged
outlines. They also show wide variation in color, ranging from almost

594

Connecticut Experiment Station

Bulletin 386

white through various shades of gray and brown to almost black. Figure
137 is a photograph of light-colored spots. On heavier leaves the spots
may be more circular and less irregular in outline and the surface marked
by somewhat concentric ridges. Leaves severely affected when young
often become distorted. After curing, the spots remain white, conspicuous
and rigid and do not soften easily during "casing." A severely affected
leaf is worthless.

Microscopic examination of young spots showed no fungous mycelium
in the tissues but great numbers of rod-shaped motile bacteria. These
were isolated and the cultural characters found to agree with those recorded
for Bacterium angulatum. Inoculation experiments in the greenhouse
did not give conclusive results but probably the conditions here were
not favorable to infection.

1

Hk ■MSBE

£.• *' *"'ll^

Figure 137. Angular spots on Broadleaf caused by the
bacterial disease, Blackfire.

Mosaic (Calico) in the seedbeds. The method by which the virus
of calico passes the winter and starts new infection each year is still a
matter of considerable uncertainty. It is not improbable that there are
several sources of the spring infection. There is pretty good evidence,
however, that the seedbed is the principal avenue by which it arrives
in the field. Yet one rarely sees a calico plant in the beds. This year,
however, one Broadleaf grower brought to the Station some plants from
his beds that showed plainly the symptoms of the disease. Examination
of the beds revealed that plants were affected in a few small areas in
some, and it was hoped that in the field the infection would be only local
and could be removed. Although the plants were free from symptoms
when they were set out, more and more of them developed the disease
as they started to grow, until about 90 per cent were affected. All were
removed in early July; the field was harrowed and set again with plants
from a disease-free bed, and a good crop with only a small percentage

Tobacco Diseases in J 935

595

of mosaic was harvested. It is obvious from this experience that
it is not safe to set plants from a bed where there is any calico.

It has been demonstrated in Kentucky that a large percentage of the
infection may be traced to the use of infected chewing and smoking tobacco
by workmen. The virus gets on the hands and then is passed on to the
plants that they touch in weeding, pulling, setting, etc. We have found
many indications that at least a part of our bad infestations may start
from the same source.

Some other diseases have been observ ed this year but were not sufficiently
prevalent to warrant special investigation or note. Pythium Damping-
off and rootrot (P. debaryanum) and the later bedrot (Rhizoctonia)
were found in a few beds. Non-parasitic leafspots were present in the
usual number.

Figure 138. Dead-blossom leaf spot. Note the

adherent, partly decayed blossom and

the dead leaf area about it.

Dead- blossom Leaf Spot

During the latter part of August and in September, this spot became
prevalent on Shade tobacco and was brought to our attention by several
growers. It occurred particularly on the later primings and was undoubt-
edly favored by the wet weather that prevailed in most of the Valley
during that period. The trouble is not new but has been known to Shade
growers for many years. It is primarily a disease of this variety of

596 Connecticut Experiment Station Bulletin 386

tobacco — although occasionally seen on other kinds — because blossoms
in great numbers are allowed to develop on Shade, while the tops and
suckers are removed from the Sun-grown types. When the flowers are
mature, the corollas, or pink-colored parts, drop, and many of them,
especially in wet weather, adhere to the gummy leaves and decay. As
shown in Figure 3 38, the dead spot spreads in all directions from the
decaying blossom as a center. The higher humidity under the tent also
favors the development of the spots on this type of tobacco.

Symptoms. Spots are irregular and wavy — not angular — in outline,
up to an inch or more in diameter, with a quite definite margin. In color
they vary from light tan to reddish brown, while some are faded to ashen
gray. Spots are usually bordered more or less by an indefinite, narrow,
light yellow halo that fuses outward gradually into the normal green of
the leaf. The dead tissue is brittle and breaks out easily. The dead
blossom remains tightly adherent to the center of the dead spot and
usually becomes more or less blackened by masses of Alternaria spores.
There may be one to a half-dozen spots on a leaf, thus rendering it worth-
less for wrapper or binder.

In the curing shed, where the leaf is in close contact with others on
the same lath, the rot frequently spreads from one to the next until a
dozen may be rotted. In the shed, under these conditions, the trouble
has all the symptoms of pole rot.

Fungi associated Math the spots. Examination of the dead blossoms
shows the presence of spores of a considerable number of fungi. Alternaria
tenuis is universally present and is the predominating fungus. Botrytis
cinerea and Cladosporium herbarum are also very common. Sometimes
there are also spores of Fusarium, Macrosporium, Monilia, Phoma, and
other genera. Examination of the tissue beneath and adjacent to the
dead blossoms shows spores and sporophores of Alternaria tenuis on the
surface and Alternaria mycelium inside the tissues of the leaf. Other
species are found more rarely on the adjacent leaf tissues. Frequently
bacteria also are present in abundance inside the dead leaf. Isolations
from the margins of the spots where no spores were found on the surface
yielded almost invariably pure cultures of Alternaria tenuis.

In view of the considerable number of species of fungi and bacteria
that are found in the decaying blossoms, it is difficult to decide which
one, if any, is the primary parasite. Alternaria is the most universally
present and its mycelium is constantly found inside the leaf tissue. As
indicated by previous experiments of the writer, however, (Conn. Sta.
Bui 367, p. 134), the spores of this fungus sprayed on the leaves did not
produce spots on the tobacco except very slightly on overripe, injured
leaves. This indicates at the best a very weak degree of parasitism. It
is possible, however, that in moist, warm weather, beneath a wet rotting
corolla, the conditions might be such as to permit infection and a limited
spread of Alternaria in the tissues. Or perhaps the virulence of Alternaria
is "stepped up" by previous growth on the tobacco blossom to such a
degree that it becomes truly parasitic on the green leaf. Possibly the
combined influence of all the rot-producing organisms, in direct contact
with a somewhat etiolated part of the leaf, might produce infection where

Tobacco Diseases in 1935 597

any single one of them would fail. The presence of the dead blossom
on the leaf is absolutely essential to the parasitism of whatever organism
or organisms produce this type of spot. In the absence of blossoms the
spots do not occur.

The spread of the rot in the curing shed from one leaf to adjacent leaves
is readily understood in view of the known capacity of Alternaria to
infect curing leaves under these conditions. This is no different from
the usual pole rot except that the presence of the rotten corolla and
adjacent dead tissue provides a very favorable nursery to serve as a
starting point for spreading to the adjacent leaves under the moist con-
ditions of a packed shed.

Previous observations. Leafspots caused by decaying blossoms have
been noted aud described b> other observers both on other types of tobacco
and on different species of plants. Thus in Germany, Gleisberg, (Gleisberg,
W. Botrytis-Erkrankungen. Gartenflora 70: 13-18. 1921) found that
similar dead brown spots were produced on Primula leaves and on about
25 other species of plants when blossoms of the locust tree (Robinia)
fell on the leaves and rotted there during wet weather. He found that
the blossoms become covered with a growth of the fungus, Botrytis
cineria, which also spread to the leaf tissue. Although pure culture
inoculations were not made, he considered the evidence sufficient to
warrant the conclusion that this fungus was the direct agent in the disease.
During the same year, 1921, Pape (Pape, H. Beobachtungen bei Erkran-
kungen durch Botrytis. Gartenflora 70: 48-51. 1921) observed the
same type of spot about blossoms which had fallen on tobacco leaves.
His description and photographs which accompany his article show that
he had under observation spots very similar to those which we find in
Connecticut. He also found the spores of Botrytis cinerea on the surface
and cultured this same fungus out of the affected leaf tissue. Therefore
he concludes that it is the causal agent.

A more complete description of the trouble as it occurs on tobacco in
Germany was later published by Boening1, who alludes to a popular belief
of growers that the tobacco blossom contains a poison that has a corroding
effect when in contact with the leaf. He finds the spots mostly on the
middle and lower leaves of the plant, explaining this on the theory of
better moisture relations for fungus growth on these leaves than on the
upper leaves. Like the two writers just mentioned, he considers Botrytis
cinerea the causal agent which is able to pass from the dead blossoms into
the leaf tissues under the moist conditions which prevail under the
adhering blossoms. He also noted that when hung up to cure, the infected
leaves serve as foci from which the rot passes into the other curing leaves
which are in contact, and thus many leaves may be ruined.

Control. No practical method of preventing the spots in the field
under the usual cultural practices for Shade tobacco has been found.
In the shed, the same methods must be used as for the control of pole
rot. The surfaces of the leaves must be dried as soon as possible by proper

'Boening, Karl. Kranliheiten des Tabaks. Eine durch herabgefallenen Blueten hervorgemfene
Botrytisblattfaule. Arbeiten aus der Bayer. Landesanstalt fur Pflanzenbau u. Pflanzenschutz, Heft
4: 18. 1928.

598

Connecticut Experiment Station

Bulletin 386

firing and ventilation. (See Conn. Sta. Bui. 364 for more detailed
description of firing) .

Pythium Stalk-rot of Transplants

This disease was first brought to our attention about the first of June,
when a Shade grower brought to the Station a basket full of plants which
had died in the field a few days after transplanting. There had been
frequent soaking rains during the transplanting season, and the constantly
wet soil undoubtedly influenced the prevalence of the trouble. The

Figure 139. Pythium stalk-rot. Note the dead
black stem and bases of leaves.

disease is probably not new here but no published description of it in
America has come to our notice. In Sumatra, however, it has been known
since about 19171 under the name of "Stengelverbranding" (Stemburn).

1 Meurs, A. Parasitic Stemburn of Deli Tobacco. Phytopath. Zeitschr. 7:169-185. 1931.

Tobacco Diseases in 1935 599

Meurs found that it occurs both in the seedbeds (rare however) and in
the fields shortly after transplanting. Although the trouble here has
been observed only in the field, it is quite possible that it may be present
also in the beds and has been confused with ordinary bed rot which it
resembles. Examination of a number of fields showed that the disease
was general throughout the Valley during 1935 and occasioned much
extra labor in restocking.

Symptoms. This is a soft, watery, black (or at least very dark),
mushy rot that completely disintegrates all the tissues of the stalk. It
starts just at, or below, the ground level, but under favorable conditions
quickly works upward even into the bases of the leaves (Figure 139)
and causes the plant to fall over and completely collapse. The stalk
is so rotted that it no longer holds together when the plant is pulled up.
If the plants are kept over night in a moist chamber, they become covered
with a thick white felt of mold.

Causal organism. Microscopic examination shows the tissues com-
pletely permeated with the mycelium and spores of a fungus, Pythium
debaryanum. This fungus was easily isolated and grown in pure culture.
In every respect it was identical with the fungus which was described
and illustrated in our report for 1933 (Conn. Sta. Bui. 359: 336-354)
as the cause of the early damping-off of tobacco seedlings in the beds.
This fungus is present in most soils and it was not possible to determine
whether it was carried on affected plants from the beds to the field or
whether primary infection came from the field soil. Possibly both methods
of infection play a part.

In Sumatra, four different species of Pythium have been found in the
diseased plants and each, on re-inoculation, proved to be capable of
producing the rot. One of these species was doubtfully called P. debary-
anum* The other three were studied by Meur (citation above) and
found to be P. aphanidermatum (Eds) Fitz., P. myriotylum Dreschsler
and P. deliense Meur. None of these last three species has been found
in Connecticut but as far as symptoms are concerned, our disease appears
to be identical with the Sumatra "stemburn."

Control. Prevalence of the disease is undoubtedly influenced by
environmental conditions, being worse where moisture is greatest. We
may therefore expect to find more or less of it any year when there are
frequent rains during the setting season. Since the fungus is not new,
there is no reason to fear that the disease will increase and become epidemic
any more than it has in the past. Since there is a possibility that at
least some of the infection starts in the beds, one obvious control measure
would be to keep the bed as free as possible from disease by sterilizing
the soil, by aeration and keeping the stand thin, and by avoiding too
moist conditions. Succulent, tender, fast growing plants appear to be
more subject to attack than those which are hardened. Meurs finds
that the plants of the first pulling from the beds are more susceptible
to attack than the later pullings. The use of well hardened and disease-
free plants should do much to keep this stalk-rot under control. It might

*Jochems, S. G. J. Stengelverbranding bij Deli-Tabak. Med. Deli Proefstation te Medan-
Sumatra. Ser. 2, No. 49. pp. 35, 1927.

600 Connecticut Experiment Station Bulletin 386

prove profitable to discard the first pulling and use only plants from the
later pullings.

Studies on Pole Rot. I

Of all the diseases that plague Connecticut Valley tobacco, pole rot
probably presents the most baffling problem for the plant pathologist
and for the grower and packer. Our knowledge of the causal agencies,
influence of environmental and other factors,and of control is inadequate
and confusing. Half of the farmers make no effort to use the recommended
measures of control and many who use them are only partially successful.
Some even believe that heating with charcoal — the commonly recom-
mended measure— frequently aggravates rather than remedies the trouble.
In view of the meagerness and uncertainties of our knowledge of pole
rot, there is need of a thorough re-investigation of every phase of the
trouble with the ultimate object of perfecting a method of control that
will be easier to apply and more certain of success than any that we now
have. This is not an easy task ; perhaps we shall never be entirely success-
ful. The chances of finding a better control method, however, will be
more favorable after we have more fundamental information about the
causal organisms, their mode of attack, response to environmental factors,
etc. It was with the object of acquiring such fundamental information
that the investigation here reported was started. The present discussion
is a preliminary statement and will be followed by others as the work
progresses.

Name. The rotting of tobacco during the time that it is hanging on
the poles in the curing shed has been designated by a variety of names
in different sections of this and foreign countries. Sweat, pole sweat,
shed burn, house burn, pole burn, stem rot and pole rot are the most
common names. ("Stem rot" is the name applied only to the decay
of the midribs.) The use of the term "burn" probably comes from the
German name "Dachbrand" (literally "roof-burn"). Although the dark-
ened dead appearance of the leaf produced by one of the types of this
disease might by a stretch of the imagination be compared to the effect
of burning, the term "burn" is certainly not descriptive of the various
other symptoms. There is also an objection to the use of the term "sweat"
because, in the trade, this term is used to designate the fermentation
process, which has no relation to pole "sweat". The fact that sometimes
drops of water collect on the leaf surfaces during damp weather, and
may favor rot, probably explains the use of this term. The disease in
all its various manifestations is nothing more than a "rot" of vegetable
matter as this word is used and understood by English speaking people.
This term is, therefore, used here to avoid hazy misconceptions on the
part of the reader which the other names frequently connote. The term
"pole rot", then, designates all forms of rot that affect the curing tobacco
while hanging on the poles, and distinguishes it from other rots that may
occur in the bundle, the case, or in the field.

Symptoms. Pole rot as a rule does not occur during the first stages
of curing — that is, while the leaves are still green. Only occasionally
and under very moist conditions, where the leaves have remained green
for a long time in the shed, has the writer ever seen pole rot on a green

Tobacco Diseases in 1935

601

leaf. In general the symptoms may be expected to appear on leaves
which are in the yellow stage and just turning brown. The rot may appear
on any of the leaves of the plant which are just passing through this
critical stage, if continued moist weather conditions prevail at just that
time. When the leaves wilt down around the stalk, the middle and lower
ones form an overlapping sheath around the top of the stalk. As a result,
the inner leaves remain under more moist conditions because of lack of
air circulation and the greater quantity of water that is transpired from
the green stalk and the crowded leaves. Such conditions are very favor-
able nests for pole rot to start and it is here, rather than on the outside
leaves, that one usually finds the trouble first. Even where charcoal
firing is resorted to, it is difficult to dry out the leaves in such places,
and an insufficient firing is apt to make conditions more favorable to
rot because the temperature is raised but the surfaces are not dried.
Rarely, if ever, is a shed of stalk tobacco cured without an occasional
such pocket where rot has started. The primed leaves of Shade tobacco
are better distributed and there are no green stalks to furnish a continuous
supply of moisture. Therefore pole rot is not so common in Shade tobacco
as in stalk-cut tobacco. As a rule also, Shade tobacco is more carefully
fired during the cure.

■ %

<J]f\ly ;

^^p?

, ' -nts ..»' ,,"^4*^t. »,-

■ *

. » f AT if -.'*--■ ' ,

Figure 140. Pole rot. This spotting is the
characteristic freckle rot symptom.

There is considerable and often confusing variation in the symptoms
exhibited, depending on the part of the leaf attacked, severity of infection,
time of infection and location of the leaf with respect to the other leaves.
For convenience in discussion we may distinguish three types of pole
rot: (1) freckle rot, (2) web rot and (3) vein rot. One, two or all of these
may be present on the same leaf and they may intergrade into each other.

Freckle rot has usually been considered the initial stage of the disease,
but this is not universally true since either of the other symptoms may

602 Connecticut Experiment Station Bulletin 386

appear first. Small dark specks, from pinhead size up to a quarter of
an inch in diameter, appear on the blade of the leaf. They resemble
in size, shape, distribution, and often in color, the freckles on a red-
headed urchin's face. On light leaves they are deep, reddish brown,
while on dark, heavy leaves they are almost black. When the leaf is
held up to a strong light, the spots appear translucent. There is no fungous
growth on the surface. Freckle rot is shown in Figure 140.

In web rot, large areas of the leaf blade are rotted and discolored and
there is no indication of small local spots. The affected areas are usually
several inches in diameter or may include the whole side of the leaf. In
severe cases in the shed, the tissue is so disintegrated that the leaf tears
apart between the hands when one tries to open it; that is, it no longer
has any tenacity. The color, although varied, is always darker than
that of the unaffected part of the leaf. Such areas, after curing, do not
become soft and pliable when the rest of the tobacco comes "into case",
but remain stiff and brittle and are easily broken during the sorting and
packing. There is no elasticity and affected leaves are worthless for
cigar wrappers and binders. Sometimes there seems to be a connection
between the web-rot type and the freckle-rot type when a large lesion
of web rot is bordered by freckles which gradually become less noticeable
and disappear in the healthy tissue. One gets the impression that the
web-rot type of lesion is produced by the enlargement and coalescence
of the freckle spots. This theory, however, does not explain the many
cases in which there are no freckles at all in connection with many pole-
rotted leaves. It is apparently possible to have web rot without an initial
stage of freckle rot.

Vein rot (including stem rot) is the name applied where the rot involves
principally the mid-vein or the large lateral veins. The veins, especially
the midribs, retain their moisture long after the leaf web has dried and
therefore furnish very favorable substrata for growth of rot-producing
organisms. They turn brown, become soft, and then are covered with
a growth of white, gray or variously colored molds. The point at which
the midrib is attached to the stalk is an especially favorable point of
attack. Frequently the rot progresses to a point here where the slightest
pull will make the leaf fall from the stalk, or sometimes it will even fall
to the ground of its own weight. The rot also usually passes out somewhat
into the adjacent web. When the leaf is stretched between the hands,
the blade pulls away from the large midrib and the latter, being all rotted
except the stringy fibers, frays out as the leaf is stretched. The same
may be true for the large lateral veins just as for the midrib. Leaves
thus affected are graded as "brokes" in the sorting, even though the
blade may be perfectly sound.

Cause of pole rot. From the beginning of tobacco culture it was
recognized that moisture bears an important relation to prevalence of
pole rot, because the rot is always worse during wet curing seasons.
Moisture, however, is not the direct cause, but is of importance only
because it provides conditions in the shed which favor the growth of
decay organisms that are the direct agents. Although it is now agreed
by all investigators that organisms produce pole rot, there is considerable
difference of opinion as to which of the various fungi and bacteria found
in rotted tobacco are the primary agents in decay. There is the possibility

Tobacco Diseases in 1935 603

that the organisms that cause the trouble in one section are not the same
as those that operate in another tobacco section; or that the organisms
that produce one type of rot are not the same as those that produce another
type. This uncertainty may be ascribed for the most part to the almost
insurmountable difficulty of making pure culture inoculations on leaves,
under natural, but at the same time, sterile conditions — the only method
of proving definitely the causal relation of an organism to a disease.

Mueller-Thurgau1, in 1885, stated that pole rot is caused by fungi and
warned against too high moisture in the air of the curing chamber since
"thus the curing process is too long drawn out and fungi of different
sorts settle on the leaves and have an unfavorable influence on the quality
of the product. The dreaded "Dachbrand" is of this nature and I have
regularly found on these leaves Pleospora-like fungi". Thus the first
investigator to ascribe pole rot to fungi, believed it was associated with
Pleospora (ascospore stage of Alternaria) .

Sturgis2, of the Connecticut Agricultural Experiment Station, in 1891
was apparently the first to make a serious attempt to find what organisms
are responsible. He decided that the rot was due to the growth of bacteria
but that a fungus (he mentions Cladosporium) first made it possible
for the bacteria to gain access. To quote from him: "Pole burn is due
primarily to the growth of a fungus upon the leaf which by disintegrating
and partially destroying the tissue of the leaf gives access to a bacterial
process of decay".

Two years later Behrens3 in Germany came to the conclusion as a
result of his investigations that pole rot is always caused by one of two
fungi, Selerotinia Libertiana Fuck, or Botrytis cinerea which he considered
the conidial stage of Selerotinia Fuckeliana de B. Although he found
species of Aspergillus, Penicilleum, Cladosporium and Alternaria present,
he considered them as secondary saprophytes which had no relation to
the rot. He found that the Sclerotinias usually occurred first on the
midribs and only later passed to the leaf, an observation that indicates
he was investigating a trouble that corresponds more nearly to the type
we have called vein rot.

In 1893 Sturgis4 states that stem rot is due to Botrytis longibrachiata
(- Botryosporium pulchrum Cda., according to Clinton in Conn. Sta.
Rpt. for 1903, p. 362). After referring to the work of Behrens, he reiter-
ates his belief in the bacterial nature of the other types of pole rot: "We
are, therefore, convinced that although the initiative in pole-burn may
be taken by a Botrytis or some similar fungus, the preliminary growth
of which may even be essential to the succeeding condition of decay,
the latter is largely bacterial in its nature."

Six years later Sturgis5 made a further study of pole rot in the initial
stages, and after making cultures and microscopic examination, stated:

1 Mueller-Thurgau, Herman. Ueber das Verhalten von Staerke und Zucker in reifenden und troch-
ennden Tabaksblattern. Landw. Jahrb. 14: 485-515. 1885.

2 Preliminary report, on the so-called pole burn of tobacco. Conn. Agr. Exp. Sta. Bpt. 15, 168-
186. 1891.

3 Behrens, J. Trockene und nasse Faule des Tabaks, "der Dachbrand." Zeitschr. f. Pflansenkr.
3: 82-90. 1893.

4 Sturgis, W. C. Further notes on the cause of "pole sweat," and "stem rot" of tobacco. Conn. Agr,
Exp. Sta. Rpt. 17 (1893): 84-85. 1894.

5 Sturgis, W. C. Further notes on the pole-burn of tobacco. Conn. Agr. Exp. Sta. Bpt. 23 (1899):
255-269. 1900.

604 Connecticut Experiment Station Bulletin 386

"I can only conclude from these cultures that as far as regards the organisms
associated with the earliest stages of pole burn — this species of Alternaria
is the only one which occurs in any abundance and that it occurs only
on the surface of the leaf, not in the internal tissues". "The fungus
which was associated with the trouble [referring to his previous report]
was unquestionably a species of Cladosporium ; in the present instance
unquestionably a species of Alternaria. These statements are not contra-
dictory. They merely indicate that under certain atmospheric conditions,
any saprophytic fungus which may be present in the curing barn may
attack the dead tissues of the leaves and start in them a process of disin-
tegration which will almost surely be followed by true bacterial decay."

In Wisconsin, Johnson1 isolated fourteen different forms of fungi
from decaying tobacco, only three or four of which were of general occur-
rence under field conditions. The most common was Fusarium, but
frequently he found Trichothecium roseum and Penicillium brevicaule
and occasionally, Alternaria and Botrytis. He states: "No particular
organism can be universally ascribed as the cause of this disease. More-
over, it is to be expected that dead or dying tissues, such as occur in
curing tobacco, will become substrata for various saprophytic organisms
when conditions favorable for their development occur. Furthermore,
it seems most reasonable to suppose that such organisms as sometimes
exhibit considerable parasitic action should be the first, and the ones
most likely to produce the decay of curing tobacco. For this reason we
may expect that the purely saprophytic fungi should occur more rarely,
if at all. The occurrence of species of Fusarium, Botrytis, Sclerotinia
and Alternaria seems to indicate that these organisms exhibit some para-
sitism upon the slowly dying leaves.

"These observations indicate that the causal organisms of shed burn
and stem rot vary in different sections of the country, and that it is in
all probability a matter of the organisms with which the leaves happen
to become infected, either from the field or from infectious material in
the curing shed. The organism actually producing the decay may also
vary with the temperature and humidity as well as with the stage of
curing, and consequently we may find a considerable variance in the
different sheds and in different years in the same locality."

Although admitting that bacterial decay may occur under extreme
conditions of humidity, he concludes that: "The ordinary shed burn,
as far as my observations go in Wisconsin, seems to be due entirely to
fungi ... In Wisconsin, Fusarium particularly seems to cause the
decay known as stem rot and shed burn equally well, and for this
reason it appears that the two diseases may be caused by the same organ-
ism and that they differ mainly in the point and time of decay".

Recently a more thorough investigation of the disease has been con-
ducted in Wisconsin by Johnson and Ogden2. After several years of
investigating the small spots, (presumably of the freckle rot type described
above) they conclude: "Practically without exception, only one fungus
has been found to predominate in these spots". This fungus was identified
as Alternaria tenuis Nees. With regard to the stem rot, they state that

1 Johnson, J. Black rot, shed burn and stem rot of tobacco. Wis. Agr. Exp. Sta. Research Bui. 32:
81-85. 1914.

2 Johnson, James and W. B. Ogden. The relation of air conditions to tobacco curing. Wis. Agr.
Exp. Sta. Res. Bui. 110: 40-45. 1931.

Tobacco Diseases in 1935 605

"the evidence seems to indicate that a number of different fungi may be
concerned" and the reader is referred to Johnson's previous work quoted
above for the names of these species.

A review of all that has been published on the subject shows that there
is thus considerable diversity of opinion as to the organisms responsible
for pole rot, or that there are different causal organisms depending on the
locality in which the investigation was conducted. In view of this rather
uncertain state of our knowledge of causal agencies, the writer began an
investigation of the organisms associated with rotting tobacco leaves
with the hope of determining whether any single one is the initiator of the
trouble in Connecticut, and the relationship of the others to the various
forms of pole rot. Since it is possible that the organisms concerned in
one form of the trouble may not be the same as those associated with
another form, it was planned to take up each type of the disease by itsel
without generalizing on all its manifestations. In the present discussion,
therefore, there are recorded only those investigations and observations
that concern the "freckle rot" type of pole rot, reserving for a later paper
the "web rot" and "vein rot" types.

The small, isolated, dark spots characteristic of freckle rot have been
described above under "symptoms". Examination of a large number of
the smallest (presumably youngest) of these spots in the sheds failed to
show spores of fungi on the surface. When the leaves are kept in a moist
chamber for a sufficiently long time, fungi of several species develop on
the surface. However, to assume that these were the same fungi that
produced the spots might easily lead to a false conclusion. Very small
spots — not larger than a pinhead — taken directly from the shed and show-
ing no surface growth, were cut out of the leaves, the tissues thoroughly
teased apart under a dissecting microscope and examined under high
magnification. Examination of hundreds of these spots showed that the
tissues were filled with fungous mycelia of only one type. Bacteria in
any quantity were not found in these youngest spots. The fungous
mycelium is hyaline (dark only when outside the matrix), stout (3.7
microns in diameter) thick-walled, distinctly septate and usually con-
stricted at the septa, much branched, winding and irregular in direction
and in size of cells. The cells are usually two or three times as long as
they are broad and inclined to be slightly dumb-bell shaped. Frequently,
however, there are chains of globular, shorter, subglobose cells of larger
diameter (Figure 141), a condition that seems to be distinctive for this
species. Another distinctive feature of this mycelium is the manner of
branching, which is commonly dichotomous from the enlarged end of a
pear-shaped basal cell, although true monopodial, and other irregular
types of branching, may also be found. After a little experience, this
mycelium could readily be differentiated from all other species of mycelium
found on tobacco leaves. It was obviously the only kind of organism
that occurred commonly in these spots.

The fungus was next isolated by cutting out tiny pieces from the smallest
spots with a flamed scalpel, washing through several changes of sterile
water and transferring to potato dextrose agar plates. From about 95
per cent of such transfers, only one fungus grew on the agar. This same
fungus was present on the other 5 per cent but was accompanied by other
fungi. It quickly sporulated, was transferred, measured and studied and
found to be Alternaria tenuis Nees. Since this species was fully described

606

Connecticut Experiment Station

Bulletin 386

and pictured in last year's report, (Bui. 367: 132) in connection with
"white speck" on tobacco leaves, no description of its morphological
characters is repeated here.

No satisfactory method of inoculating half-cured leaves under aseptic
conditions with this fungus has yet been found by the writer and to that
extent the conclusions drawn are subject to criticism. Nevertheless, after
studying and isolating the organism from hundreds of these spots from
many different sheds during the last two seasons, he is convinced that as
far as the Connecticut Valley is concerned, the freckle rot type of pole
rot is produced by Alternaria alone. This conclusion is in entire agree-
ment with that expressed by Johnson and Ogden for the Wisconsin area.
It may also be stated here that Alternaria has been constantly found in
connection with the other types of pole rot, but the presence of several

Figure 141. Alternaria tenuis, the causal organism of freckle type of
pole rot. Spores (left) magnified 550 diameters. Internal myselium
(right) magnified 1000 diameters.

other species of fungi (Botrytis cinera, Cladosporium, Fusaria, etc.)
as well as bacteria in many cases, makes it difficult to determine
which have a causal relation. There is need here of further investigation
to determine the role played by these various species. Certainly one is
not warranted in concluding, since freckle rot is produced by only one
organism, that web rot and vein rot are also produced by the same fungus.
Since spores of these different species of fungi are probably present in
great numbers on the leaves, the question naturally arises: Why is Alter-
naria the only one that produces the freckle rot type of spot? Previous
investigations by the writer on white speck (Bui. 367: 134) and on dead
blossom leaf spot (p 596), of Tisdale and Wadkins1 on brown spot and
of Ghimpu2 in Europe, indicate that this species of Alternaria, although

1 Tisdale, W. B., and R. F. Wadkins. Brown spot of tobacco caused by Alternaria longipes. Phyto-
path. 21:641. 1931.

2 Ghimpu, V. La maladie des tabac causee par Pleospora Alternariae. Bui. Cult, si ferment.
Tutunului 20. (No. 1): 1-27. 1931.

Tobacco Diseases of 1935 607

not a virulent parasite, is nevertheless in the group of fungi known as
facultative parasites or semi-parasites. Since, at the critical stage of
curing mentioned above, the leaf is still alive but at the same time is
getting weaker daily and losing its resistance, we may suppose that the
fungus which comes nearest to being a parasite will be the first one to
gain entrance to the tissues, and that it may be followed later by others
of weaker parasitism until, when the cells are quite dead, the pure sapro-
phytes will cause rot. The most nearly parasitic fungus in this case is
Alternaria tenuis, hence it is the one that produces the first stages- of rot.
Another question to be answered : How do the spores of Alternaria come
to be present in such numbers on curing tobacco leaves? The writer has
found this species sporulating on white speck lesions, blossom rot lesions,
and angular leaf spot and mosaic rust, in fact almost any injured part of
the leaf becomes infected and produces many spores which are carried into
the curing shed. Even more important is the great abundance of these
spores on the dead sand leaves. Since the species sporulates on many
other species of plants and vegetable debris from which the spores may
be blown or ortherwise carried to tobacco leaves, it is easy to see how
they would be present on the surface of the leaves in great number when-
ever tobacco is taken into the shed.

EFFECT OF SHADE CLOTH ON ATMOSPHERIC CONDITIONS

Observations on the effect of shade cloth on light intensity reported last
year (Conn. Agr. Exp. Sta. Bui. 367: 143) were repeated at intervals
during the growing season of 1935. In addition, measurements of tem-
perature and relative humidity were made with a sling psychrometer.
The instrument and technique employed for light measurements were the
same as in 1934. Table 28 summarizes the results.

Reductions of light intensity under shade tents ranged from 30 to 63
per cent, almost identical with last year. Measurements under trees
were added to show the relative reduction under heavy shade, and ranged
from 83 to 95 per cent of intensities in the open.

Temperature and relative humidity measurements disclose interesting
facts. Temperatures were lower under the shade tent in three out of five
cases and equal in the other two cases. Relative humidities were from
11 to 23 per cent higher in all but one case, a foggy morning. The com-
mon belief that air temperatures are several degrees higher under shade
tents is thus an error. The observer gets this impression because of the
lower evaporating power of the air, and the consequent higher body
temperature.

Relative humidity measurements are of much value in an interpreta-
tion of the effect of shading. An average increase of 15 per cent in rela-
tive humidity, together with a decrease of 50 per cent in light intensity,
is the equivalent of a southern climate many degrees of latitude removed
from Connecticut conditions. The effect on the physical characteristics
and chemical composition of the plant is very pronounced, as was noted
in last year's report.

608

Connecticut Experiment Station

Bulletin 336

Table 28. The Effect of Shading on Light, Temperature and
Relative Humidity

Light in

%

Date

Hour

Weather

Location

candles

Dry

Wet

relative

per sq.
ft.

bulb

bulb

humidity

July 19

11:30 A.M.

Calm

Under tree

150

90

77

55

In open

900

92

79

56

_

Under tent

400

91

83

71

July 23

1:15 P.M.

Light

Under tree

90

89

77

57

breeze

In open

1000

91

79

58

Under tent

700

90

85

81

July 26

1 P.M.

Intermittent

Under tree

50

77

64

49

strong breeze

In open

1000

81

67

48

Under tent

600

81

72

64

Aug. 2

7 A.M.

Light

Under tree

35

75

68

70

breeze

In open

350

77

70

71

Under tent

130

75

71

82

Aug. 17

9:30 A.M.

Calm, fog

Under tree

40

75

70

78

dissipating

In open

700

79

73

75

Under tent

400

79

73

75

v) 0

62

University of
Connecticut

Libraries

3Ui53UZBBb4/52