B-1144
July 1974

     

Bollworm Attack
On Experimental
Semidwarf Cottons

Texas A&M University System, The Texas Agricultural Experiment Station, J. E. Miller, Director, College Station, Texas

Contents

SUM MARY . . . . . . . . . . . . . . . . . . . . . . . . . .
INTRODUCTION . . . . . . . . . . . . . . . . . . . . .
METHODS . . . . . . . . . . . . . . . . . . . . . . . . . .

1971 Study - Bollworm Attack on Isolat
of Three Genotypes . . . . . . . . . . . . .  5%‘-

Deltapine 16 . . . . . . . . . . . . . . . .
DSR 1X6-56-66 . . . . . . . . . . . . . . 
DSR 6-19-66-6M-1O . . . . . . . . . . l"

1972 Study - Bollworm Attack on Cot
Planted in 6-Foot-Row Plots ., . . . . 

RESULTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . l’

1971 - Bollworm Attack 0n Isolated Pl
1972 - Threshold Studies . . . . . . . . . ..

DISCUSSION . . . . . . . . . . . . . . . . . . . . . . . . .
ACKNOWLEDGMENT . . . . . . . . . . . . . . . .
REFERENCES CITED . . . . . . . . . . . . . . . ..

  
   
  
    
   
  
  
  
  
  
  
 
  
  
 
    
   

Summary

Investigations of bollworm, H eliothis ze‘
attack on cottons differing in growth and fruiti
teristics were conducted in Texas, 1971-72.' ‘y:
fruits damaged on experimental semidwarfcq
no greater than those 0n Deltapine 16; squ
tended to be less on the compact cotton i
worms fed, they moved much less on the;
cottons in comparison to Deltapine 16. stu?
that 9, O00 worms per acre in three infestatio
not produce severe effects on lint yields of t
investigated. Delayed maturity, as a result g
tack, occurred in Deltapine 16; maturity w 
in an experimental semidwarf cotton. 

Bollworm Attack
On Experimental
Semidwarf Cottons

J. L. Baldwin, J. K. Walker,
J. R. Gannaway and G. A. Nile_s*

  
   
   
 

i.

l , research assistafit and associate professor, The Texas
Experiment Station (Department of Entomology), and

("te and associate professor, The Texas Agricultural Ex-
‘n (Department of Soil and Crop Sciences).

trademark or a proprietary product does not constitute a
lvlarrantyof the product by The Texas Agricultural Exper-
 and does not imply its approval to the exclusion of other
_ also may be suitable.

Insect problems associated with cotton production
in much of the United States are many times a conse-
quence of the extended production period required to
produce an acceptable crop. Pest populations tend to
increase as the growing season progresses, and insectici-
da] protection is often necessary. The development of
experimental semidwarf determinate cottons by breed-
ers of The Texas Agricultural Experiment Station offers
a potential for reducing the production period. These
strains produce their crop in a shorter time than many
indeterminate varieties and, consequently, may escape a
great deal of the late-season insect damage (Walker and
Niles, 1971). However, it has been suggested that the
problem of the bollworm, H eliothis zea (Boddie), may be
greater on semidwarf plant types (Bradshaw, 1972).
More damage was observed on a semidwarf strain than
on Deltapine 16, possibly as a result of increased feeding
efficiency of the pest on the diminutive cotton; that is,
these genotypes possess short internodes on the fruiting
branches with resulting close fruit placement. The dis-
tance of bollworm travel among fruit forms would seem
to be reduced; consequently, more squares and bolls
could be attacked in a given period of time.

Investigations were designed to study bollworm at-
tack on semldwarf cottons and to develop information on
economic damage thresholds for the bollworm on these
genotypes.

Methods

1971 Study - Bollworm Attack on Isolated Plants of
Three Genotypes

Three cottons were examined in the present study.

Deltapine 16: This widely grown, commercial vari-
ety possesses an indeterminate fruiting habit in that it
sets fruit throughout much of the season. Deltapine
plants are large and produce extensive branching with
long internodes.

DSR 1X6-56-66: The experimental cotton strain
DSR 1X6-56-66 (hereafter identified as 1X6-56) displays
an accelerated fruiting character which enables it to set
and mature a large portion of fruit in a relatively short
time. 1X6-56 is often referred to as a semidwarf genotype
since its plants are small with short internodes and com-
pact fruit placement.

4

DSR 6-19-66-6M-10: Another experimental cotton
strain, DSR 6-19-66-6M-10 (hereafter identified as
6M-10) combines a determinate fruiting habit with a
semidwarf appearance. 6M-10 is very similar to 1X6-56;
however, internode distance on the fruiting branches,
although relatively short, is somewhat longer and fruit
placement is not as compact as on 1X6-56. 1X6-56 and
6M-10, when grown in high-density plantings, have pro-
duced more lint and matured earlier than when grown in
standard, single-drill culture on 38- to 40-inch beds
(Niles, 1969).

The genotypes to be examined were planted April
27 in the conventional system, one drill per 40-inch bed.
The plantings were arranged in a randomized block with
six replications. Plants were thinned so that certain
plants could be isolated from adjacent ones by approxi-
mately 8 feet. The isolated plants were checked periodi-
cally throughout June for bollworms and boll weevils,
Anthonomus grandis Boheman. Very few pests were
found, and their damage was insignificant. Each isolated
cotton plant was confined in a 6'x6'x6' plastic screen cage.

Third instar bollworms reared on an artificial diet
were used for the experiment. One larva was released on
the main-stem terminal of each test plant in the early
evening of July 4. Younger larvae were not used because
of their high mortality rate and limited damage, perhaps
two or three small squares (Quaintance and Brues, 1905).
Undamaged fruit counts were recorded on the day before
release of larva. Beginning July 5, the plants were in-
spected twice daily, and new fruit damage was recorded.
This procedure was followed with each plant until its
respective bollworm had ceased feeding. The numbers of
undamaged fruits remaining at the termination of each
planfs bollworm infestation were recorded. Damaged
fruits were tagged when initially discovered on the
plants. Date of injury, including morning or afternoon
inspection times, was recorded on each tag. The feeding
trail of each bollworm was approximated using this pro-
cedure.

A second bollworm was released on each of the test
plants on July 19. The methods and procedures used to
determine the feeding trail of the bollworm were identi-
cal to those used for the July 4 infestation. These proce-
dures subjected each test plant to two infestations by
bollworms.

After the second bollworm infestation, the test
plants were isolated until August 16 at which time they
were defoliated, removed from the field and pictorially
diagrammed to illustrate their general appearance and
locations of damaged fruit. The bollworm feeding pat-
terns and movements over the plants were reproduced
from the data recorded on the plant identification tags.
Measurements were taken of the internode lengths on
the main stem and on each fruiting branch. From these
data, the bollworm traveling distance (total distance

   
  
 
   
  
 
  
    
   
    
   
   
   
   
    
  
   
   
   
   
  
 
  
   
  

crawled between damaged fruit locations) was i,
for each plant, providing a relative means of ¢_
bollworm movement on the three genotypes. 

The relationship of bollworm numbers to I =30
Deltapine 16 and 1X6-56 was examined. Del 
was planted May 1 ir1 the conventional systemf
per row on 40-inch beds. The 1X6-56 geno
planted May 1 in two drills per 40-inch bed wi 
inches apart. a

1972 Study -- Bollworm Attack on Cottons ‘if
G-Foot-Row Plots ‘l

The experiment was arranged in a split-p
with three replications — main plots were Q
and subplots were worm-infestation releases. 
plot contained a 6-foot section of row. Each 
Deltapine 16 contained approximately 24 l 
proximately 48 plants were in each 1X6-56 su n!‘
plants in the subplots were caged with pla 
cages approximately 2 weeks before worm =5,
time of enclosure, the subplots were spraye
insecticide to destroy arthropod enemies of h‘
There was no evidence that the released y,‘
attacked by predators or parasites during the 

Infestations of worms were released in th 
at different times. Four, early, third-instar 
6-foot subplot comprised an infestation (eq 
approximately 9,000 worms per acre). The first f
release was made on nine subplots of each ge
July 4. Six of the subplots of each genotype 
first infestation had been released were rei 
second time on July 13. A third infestation __
on three subplots of each genotype on July j
subplots had experienced attack from the first x
infestations. Three subplots of each genotype;
controls. All of the plants in each subplot were“,
daily; damage to fruits and the date and 
damage on each plant were recorded. This l
new damage to be recognized each day apart c‘
ously damaged forms not yet shed from the p 
these data the number of damaged fruits (n 
damaged fruits encountered at each daily e ,1
was associated with a population of approxi 
worms per acre. Squares that were one-thi F
larger were ‘recorded. Fruits were classified 
proximately 5 days after blooming. Damaged y‘
and shriveled, were not included in the co __

After boll opening began, semiweekly 
made of the numbers of total bolls and open 
experiment was harvested on September 15.

Results 1
1971 -- Bollworm Attack on Isolated Plants }

Damage from worm attack on the cotton ; 
shown in Table 1. Bollworm damage to 

 
  
 
  
  
  
 
 
  
  
  
 
  
 
  
  
 
  
    
  
 
  
 
 
 
  
  
 
  

was similar during both infestations. Damage
f- er on the two modified semidwarf genotypes
 ltapine 16.

 't counts taken after each of the infestations
l the bollworm releases did not have a great
.1 the fruit loads of the plants. There was an
 uction in the number of squares on the two
 cottons after the second infestation; how-
 due to the determinate fruiting habit of
g — the decline in number of squares from
 e count was greater than the number of
raged by bollworms. Both 1X6-56 and 6M-10
‘iderable number of squares and small bolls
it second infestation.

damaged-square counts recorded at the ter-
i the second infestation indicate the advanced
 the two semidwarf cottons. Deltapine 16
 uld have been much more susceptible to
.5 orm attack because of its greater number of

 traveled significantly greater distances
f“ e 16 than on 1X6-56 and 6M-10 (Table 2). The
‘was related to the contrasting plant mor-
 fbollworms did not have to travel as far to
 on 1X6-56 and 6M-10 as on Deltapine 16. The
‘cottons fruited on short internodes and, dur-
i 3 weeks of fruiting, had more large squares
 16. However, the reduced movement of
ion the semidwarf genotypes did not enable
‘V age greater numbers of fruit (Table 1). Fig-
_ 2 are diagrams of one of the experimental
 -10 with locations and kinds of fruit damaged

by two bollworm infestations. Travelling distances of the
worm were figured from these diagrams.

This experiment was repeated with similar results.
Bollworms destroyed similar amounts of fruits on the
three cottons, and despite attack from two infestations,
the isolated plants set and matured 15-30 bolls per plant.

1972 - Threshold Studies

Bollworm attack on the genotypes is illustrated by
the data from the subplots of Deltapine 16 and 1X6-56
where the first and second infestations were released
(Table 3). Counts varied from day to day, but upper and
lower limits of square damage are apparent. The number
of old and newly damaged squares encountered daily in
the 6-foot subplots of Deltapine 16 never exceeded 23,
which, for the date, represented 17-percent square dam-
age. Most daily counts ranged from 9 to 17 old and newly
damaged squares per 6-foot subplot. The percentage of
total squares damaged during this time ranged from 5 to
15 percent in most cases. Boll damage was never high
since the worms attacked considerably more squares
than bolls. Despite the loss of substantial numbers of
squares to the released worms, boll counts progressively
increased during the duration of the experiment. Num-
bers of damaged squares and percentage of square dam-
age were less in the semidwarf cotton 1X6-56. There
were more large squares on this compact cotton with
each furnishing more food; consequently, fewer squares
were eaten than on Deltapine 16.

Table 4 provides the results of the statistical analysis

i? ERAGE FRUIT DAMAGE PER PLANT FROM TWO RELEASES OF BOLLWOFlMS ON THREE COTTON GENOTYPESl

Fruit count before
bollworm release

   
 

 9e Test.

Fruit count after

Bollworm damage infestation

Genotype Squares Blooms Bolls Squares Blooms Bolls Total Squares Blooms Bolls
Deltapine 16 32.0 1.1 3.3 9.3a 0.2a 1.6a 11. 1a 27.1 1.6 6.8
1X6-56 32.5 1.3 7.1 7.5a 0.8a 0.8a 9.1a 24.0 1.8 9.5
6M-10 27.6 1.0 4.3 8.3a 1.0a 0.6a 10.0a 21.1 2.1 7.6
Deltapine 16 30.6 1.5 13.1 6.0a 1.1a 1.6a 8.8a 32.8 2.8 19.0

1X6-56 11.3 2.1 17.6 2.0a 1.1a 1.6a 4.8a 3.8 1.3 15.0
6M-10 19.1 1.5 16.8 3.3a 1.0a 1.1a 5.3a 3.6 0.6 14.8
‘l; Total damage for two releases

Deltapine 16 15.3a 1.3a 3.3a 20.0a

1x553 9. 5a 2. 0a 2. 5a 14.0a
6M-10 11.6a 2.0a 1.8a 15.5a

mpared-vertically: those followed by the same letter are not significantly different at the 5% level of probability using Duncan's

5

   
 
  
    
   
 
   
 
   
  

TABLE 2. AVERAGE DISTANCES TRAVELED PER PLANT BY first and gecgnd infegtatigng and the first, 3 s.

slNsl-E BOLl-WORMS ON THREE COTTON GENOTvpEsl third infestations were released. Total fruit 0- if
pears t0 be excessive, but lint losses were su ii

Distance traveled, centimeters SmalL i‘

Release Deltapine 1s 1x6-5s 6M-10 ._ f
The statistical analysis of the fruiting and '

Julv 4 193a 72b 94b formance of the two genotypes, 1X6-56 and l’
MY 19 172a 40b 55b 16, are summarized in Table 5. 1X6-56 accumu 

much faster than Deltapine 16. At the last boll ~'
lMeans are compared horizontally: those followed by the same September 15, Significantly fewer bolls werep
letter are not significantly different at the 5% level of probability both genotypes Where three infestations had -
usmg Dunn's Mumple Range Test‘ leased. The other infestation releases did not i’

nificantly from the control subplots in numbers __

of the damaged fruit records for all worm releases. The rapid maturity of 1X6-56 is shown in T’
Square damage was significantly less in 1X6-56 where the its larger percentage of open bolls earlier in t 

A SQUAR 

A SQUARES .
PREVI 

i BLOOMS

‘ 10 cm. I

20am
20am r
22 am
(T)
l7)
Figure 1. Attack pattern of first worm infestation release on DSFi a‘
6-19-66-6M-10. Numbers are July dates; am and pm indicate morn-_ Figure 2. Attack pattern of second worm infestation r a 4
lng and evening records; symbols designate location and kinds of 6-19-66-6M~10 (same plant shown in Figure 1). Sym p
fruit lost to bollworms. location and kinds of fruit lost to bollworms. 

6

  

‘F

in GE NUMBER OF BOLLWORMS AND DAMAGED AND UNDAMAG ED FRUITS RECORDED IN 6-FOOT SUBPLOTS OF

 
 

 Number Total squares Total bolls
 released (damaged and Old and newly Damaged squares, (damaged and Old and newly Damaged bolls,
it - ms found undamaged) damaged squares % of total undamaged) damaged bolls % of total
Four larvae released per subplot per infestation
First infestation released July 4; second infestation released July 13
Deltapine 16
3.6 141 14.0 9.9 10 0.0 0.0
3.3 140 8.0 5.7 12 0.0 0.0
3.3 136 11.0 8.0 13 0.3 2.3
3.6 130 20.0 15.3 16 2.0 12.5
3.0 139 21.0 15.1 19 3.0 15.7
2.6 135 23.0 17.0 20 3.0 15.0
1.3 141 17.0 12.0 27 3.0 11.1
0.3 136 12.0 8.8 27 1.0 3.7
3.6 147 11.0 7.4 33 0.0 0.0
3.6 140 14.0 10.0 33 1.0 3.0
3.0 120 16.0 13.3 37 1.0 2.7
3.0 128 14.0 10.9 42 1.0 2.3
2.3 106 16.0 15.0 44 1.0 2.2
1.3 115 14.0 12.1 53 1.0 1.8
1.3 107 11.0 10.2 56 1.0 1.7
1.0 94 13.0 13.8 59 1.0 1.6
0.6 96 3.0 3.1 80 1.0 1.2
1X6-56
4.0 210 6.0 2.8 62 1.0 1.6
3.3 196 7.0 3.5 72 1.0 1.3
as 205 8.0 3.9 s9 1.0 1.1
2.6 185 9.0 4.8 92 3.0 3.2
Z3 176 9.0 5.2 126 4.0 3.1
1.6 172 10.0 5.8 112 3.0 2.6
0.6 150 6.0 4.0 112 2.0 1.7
0.6 132 7.0 5.3 142 2.0 1.4
3.0 143 3.0 2.9 123 4.0 3.2
3.6 129 5.0 3.8 138 2.0 1.4
3.0 113 7.0 6.1 146 3.0 2.0
4.0 100 2.0 2.0 144 4.0 2. 7
3.3 84 2.0 2.5 144 2.0 1.3
1.6 69 1.0 1.4 157 3.0 1.9
0.3 57 1.0 1.7 154 3.0 1.9
0.3 50 0.3 0.6 158 3.0 1.8
0.0 40 1.0 2.5 163 2.0 1.2

 
 
 
  
 
 
  
 
 
  

Q Deltapine 16. The released bollworms
layed maturity of Deltapine 16, but not
may have been the result of the greater
 of 1X6-56 early in the squaring period.

d ‘ diagrams of plants in control subplots and
‘ere three infestations of bollworms were
delay in maturity (percent open bolls)
 ltapine 16 can be explained by Figure 5;
l» be set at distal locations on the fruiting
A consequently, later). The feeding attack
 ever, did not result in a delay (Figure 6).

The release of three infestations did not result in
statistically significant reductions of lint yield; about
1,000 pounds of lint per acre were produced on both
cottons (Table 5).

Discussion

The damage potential of bollworms is not increased
by enhanced feeding efliciency on semidwarf cottons;
however, reduced traveling distance for worms on these
strains could lessen the efficiency of insecticidal treat-
ments. Reduced exposure to the chemicals may result as

7

TABLE 4. TOTAL NUMBER OF coTToN FRUITS PER 6-FOOT
SUBPLOTS DAMAGED BY DIFFERENT BOLLWORM RELEASES
ON TWO COTTON GENOTYPESI

First and First, second
First second and third
infestation infestation infestation
Genotype Control release release release
Squares
Deltapine 16 0a 46.3b 80.3c 79.3c
IX6-56 0a 22.6ab 28.0b* 26.3b*
Blooms
Deltapine 16 0a 3.3b 7.3bc 11.6c
IX6-56 0a 5.0ab 10.0bc 12.3c
Bolls
Deltapine 16 0a 5.6b 10.9b 23.6c
IX6-56 0a 8.6b 16.3c 31 .6d
Total fruit damage
(not analyzed)

Deltapine 16 0 55 98 114
IX6-56 0 36 54 70

lMeans within genotypes are compared horizontally: those followed
by the same letter are not significantly different at the 5% level
of probability using Duncan's Multiple Range Test.

*Differences between genotypes are compared vertically: an asterisk
indicates that means differ significantly at the 5% level of
probability using Duncan's Multiple Range Test.

a consequence of the decreased movement on the di-
minutive cottons. In areas where dependence on insec-
ticide is heavy, semidwarf genotypes might aggravate an
already difficult situation.

These threshold data demonstrate the ability of the
cotton plant to withstand considerable damage to squares
and yet produce high lint yields. For example, square
damage in Deltapine 16 where two infestations of worms
were released attained a level of 15 percent or greater on
several dates. Cotton with 15 percent of the squares
damaged by bollworms appeared to be suflering consid-
erable injury; yet these studies showed that yield losses
were not severe. The release of an additional infestation
increased fruit loss, but lint production despite the worm
attack should be viewed in perspective. Insecticide pro-
grams for bollworm control are expensive, and the or-
ganophosphate materials used today do not perform as
well as did DDT in the 1950s Ironically, insecticidal
applications, through destruction of the bollworm-
regulating populations of arthropods in cotton fields,
increase by many times the number of bollworms to be
killed. Further, late-season populations of the insectici-
dally resistant tobacco bollworm H eliothis virescens (F.)
may develop in treated cotton.

8

  
   
  
  
 
  
  
   
  
   
  
   
  
  
   
   
  
   
  
   
  
  
  
  
 
  
   
  
  
   
  

TABLE 5. NUMBER OF BOLLS, PERCENT OF 
AND LINT PRODUCTION m GRAMS PER 6-FOOT 
m Two COTTON GENOTYPES SUBJECTED T0 i»
BOLLWORM BELEAsEs1

First and
First second
infestation infestation
Genotype Control release release
Number of bolls, July 14 
Deltapine 16 42 23 32 A
IX6-56 155a* 121 b* 119b*

Number of bolls, July 21

Deltapine 16 110a 63b 54b
1X6-56 199a* 16Gb‘ 151 b‘

Number of bolls, August 14

Deltapine 16 153 159 150
1X6-56 169a 160a 156a

Number of bolls, August 24

Deltapine 16 153 151 153
1X6-56 159 160 152

Number of bolls, September 15

Deltapi ne 16 146a 139ab 1303b
1 X6-56 148a 1 38ab 139ab

Percent open bolls, August 31

Deltapine 16 54a 33b 23b
1X6-56 86* 79* 72*

Percent open bolls, September 4

Deltapine 16 70a 52b 47b
1 X6-56 94 88* 87*

Grams of lint harvested

Deltapine 16 265 249 233
1X6-56 251 246 238

200 grams per 6-foot cotton row = 960 pounds of li 

‘Means within genotypes are compared horizontallyi;
by the same letter are not significantly different» _,
of probability, using Duncan's Multiple Range Test:

*Differences between genotypes are compared verti - _
indicates that means differ significantly at *
probability using Duncan's Multiple Range Test.

The threshold research presented w '_
pected to simulate every natural bollwo w;
For example, the age structure of pop 
worms attacking cotton is complex; these ~» “_
only the effects of a population of fixed 1Q
less, this research provides understandingf

m
came F-QQ

    
   
  

 

 

 

d - A I
; I- n: I i ‘i
_ i _.. 2 __
fe :;- *——_;_ _&_ -L
ams of all "“' _,_ __ -l— _,_ i
f» t-row con- -— a *4“; _L L_ —'-
3, Deltapine , , ,
"-1- '33.
L__::f LL-4.
n_ __'“J
iii-u L‘ h’
iioutlook toward the bollworm problem that is insecticides cautiously, thereby preserving the benefi-
;~je=| ong many growers in Texas (Sterling and cial arthropods. Growers have found that many early Iuly
 infestations of bollworms can be ignored, even though
_nsider the bellworm to be largely a seeom square damage may seem severe. Such tolerance of ap-

preciable worm damage is feasible, however, only when
insecticidal programs are not underway for control of the
boll weevil.

 induced, in many instances, by applications
 s which disrupt the normal complex of pre-
he». asites. Consequently, it is a policy to use

|20cMl

ll llililialiii  
a i? 11%.

prams of all plants in a 6-foot-row control subplot (two drills per bed) of DSR 1X6-56.

we. a m. a a a‘   r  mama“  a '1 ra"-"W-=I"-“-Y ~i

O Bolls

w Dam. Squares
0 Dam. Blooms
O Dam.BOIls

'
F.

 s  v.11 v w. .;

   

Figure 5. Diagrams of all plants in a 6-foot-row subplot of Deltapine — third infestation release. Symbols indicate undamaged bolls and z
fruit. Numbers are dates in July when fruits were attacked.

O Bolls

w Dam. Squares 

o Dam. Blooms
O Dam. Bolls

Figure 6. Diagrams of all plants in a 6-foot-row subplot (two drills per bed) of DSR 1X6-56— third infestation release. Symbols indicate A.
bolls and damaged fruit. Numbers are dates in July when fruits were attacked. ”

1O

  
  
 
  
 
   
   
  
  
 
 
 
 
 
  
  

ence Foundation (NSF GB 34718) and a
‘iagreement with the Agricultural Research
. Department of Agriculture.

Literature Cited

. 1972. Eflects of boll weevils in diflerent generations
on genotypes planted in two densities. Unpublished
Texas AGIM University.

. Growth and fruiting modifications for mechanized
n. Beltwide Cotton Production Research Confer-
p. 114-117.

1..., and C. T. Brues. 1905. The cotton bollworm. USDA
tomol. Bull. 50. 155 p.

A veld, and Robert L. Haney. 1973. Cotton yields climb,
A 7 through pest management systems. Texas Agricultural
.Tex. Agr. Exp. Sta. and Tex. Agr. Ext. Serv. 19(3):4-7.

and G. A. Niles. 1971. Population dynamics of the boll
and modified cotton types. Tex. Agr. Exp. Sta. Bull.

The Texas Agricultural Experiment Station, I. E. Miller, Director, College Station, Texas
3M——7-74

1e Texas Agricultural Experiment Station
Texas A8cM University POSTAGE AND FEES m
College Station, Texas 77843 u.s. DEPARTMENT OF


_ _ , AGRICULTURE ‘RT?
J. E. Miller, lhrector AGR m‘ n»
Publication

Penalty For Private Use, $300