Front Cover
 A procedure to asses temporal risk...
 A procedure to asses temporal risk...
 Blast development on the new rice...
 Sweet potato feathery mottle virus,...
 Some chracteristics of rice dwarf...
 Management of papaya ringspot by...
 Symptomatology, transmissibility...
 Information for contributors
 Back Cover

Group Title: Journal of Tropical Plant Pathology
Title: Journal of tropical plant pathology
Full Citation
Permanent Link: http://ufdc.ufl.edu/UF00090520/00038
 Material Information
Title: Journal of tropical plant pathology
Series Title: Journal of tropical plant pathology.
Alternate Title: Journal of Philippine phytopathology
Philippine phytopathology
Physical Description: v. : ill. (some col.) ; 26 cm.
Language: English
Creator: Philippine Phytopathological Society
Publisher: Philippine Phytopathological Society
Place of Publication: Philippines
College Laguna
Publication Date: January-June 1996
Frequency: semiannual
Subject: Plant diseases -- Periodicals -- Philippines   ( lcsh )
Plants, Protection of -- Periodicals -- Philippines   ( lcsh )
Genre: periodical   ( marcgt )
Dates or Sequential Designation: v. 1, no. 1 (January 1965)-
General Note: Title from cover.
General Note: "Official publication of the Tropical Plant Pathology."
 Record Information
Bibliographic ID: UF00090520
Volume ID: VID00038
Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
Resource Identifier: oclc - 54382605
issn - 0115-0804

Table of Contents
    Front Cover
        Front Cover 1
        Front Cover 2
    A procedure to asses temporal risk of tropical rice blast
        Page 1
        Page 2
        Page 3
        Page 4
        Page 5
        Page 6
        Page 7
        Page 8
        Page 9
        Page 10
        Page 11
        Page 12
        Page 13
        Page 14
        Page 15
    A procedure to asses temporal risk of tropical rice blast
        Page 16
        Page 17
    Blast development on the new rice plant type in relation to canopy structure, microclimate, and crop management practices
        Page 18
        Page 19
        Page 20
        Page 21
        Page 22
        Page 23
        Page 24
        Page 25
        Page 26
        Page 27
        Page 28
        Page 29
        Page 30
        Page 31
        Page 32
        Page 33
        Page 34
    Sweet potato feathery mottle virus, its association with "Kamote Kulot" and its effect on sweetpotato yield
        Page 35
        Page 36
        Page 37
        Page 38
        Page 39
        Page 40
    Some chracteristics of rice dwarf phytoreovirus (reoviridae) isolates from Nepal
        Page 41
        Page 42
        Page 43
        Page 44
        Page 45
        Page 46
        Page 47
        Page 48
        Page 49
        Page 50
    Management of papaya ringspot by isolation and intercropping of papaya
        Page 51
        Page 52
        Page 53
        Page 54
        Page 55
        Page 56
    Symptomatology, transmissibility and serological identification of mussaenda virus
        Page 57
        Page 58
        Page 59
        Page 60
    Information for contributors
        Page 61
    Back Cover
        Page 62
Full Text



Business Manager
Board Member
Board Member
Board Member
Board Member
Board Member
Immediate Past President





Subcrtis Communicaions should be odc
Pathology, UPIB, College. Loguna 4031. Phdipine
pubhcalian of the Phtippie Phylopahoogicl Socae
usMaining Associles For others, t i P100.00 per copl
and payable in advance. Memberhp in Me Pipp
membenrhip wi be suppled by the Secreary upon req
charge some authors a modest amount of each pubAe
their research projects or supporting institutions. Advert
No endorsement of any stateent of claims made in a
Phytopathtological Society, Inc.

10 rBlyLupalLilUluu,.al. Qui.iLy, 111i..

rORS 1996 1997


RD 1996 1997

Associate Editor
Associate Editor

S., INC.

sed to Me TP ASUR. P.P.S. c/ Deparmnet of Plart
rtopdhology, publshed emi-cnnuay. is Me aofcial
n 6 i sert bee to memibs n good ImnPig and
medic; and $ 2500 per copy eoeeese. pMae bee
PhMyp dXagiucdSod Sociy ic: hwnarmalon mrganng
t. Page Cfhwre: The Edb arl Board rw rvme e rig
1 page coimenurwje ton he p ymer c apaby
nernt: Rotes may be secured fom the Busines Manager.
rtisremd is aounred by this Journal or by the Philppine

1996 Phil. Phytopath. 32(1):1-171



A portion of MS thesis of the senior author submitted to the Graduate School,
Oregon State University, Corvallis, Oregon, USA.

1Senior Research Assistant and 2Plant Pathologist, Entomology and Plant
Pathology Division, International Rice Research Institute, Los Bafios, Laguna;
3Professor, Department of Botany and Plant Pathology, Oregon State University,
Corvallis, Oregon, USA

Key words: temporal risk, rice blast, Pyricularia grisea, multivariate analysis,
cluster analysis, principal component analysis, discriminant analysis

One approach in disease management is to avoid planting at the time when
conditions favor infection. To assess temporal risk of rice blast disease (caused by
Pyricularia grisea), we developed a procedure where patterns in the relationship
between blast proneness and time of sowing at three tropical sites in Asia were
analyzed using multivariate statistical procedures. A matrix data set comprising
predicted diseased leaf area and panicle blast severity as column variables and 24
hypothetical sowing dates as row variables was constructed at each site to determine
such patterns. Sowing dates were grouped according to proneness of rice to blast
using cluster analysis (CA). Three groups of sowing dates at each site were identified
by CA. The proneness characteristics of the groups were further described using
principal component analysis (PCA) by relating leaf and panicle blast with weather
factors. At Cavinti ir, the Philippines, it was shown that IR50 is prone to both leaf
and panicle blast if sown from July to December, and to panicle blast alone if sown
from January to May. With cv. C22 at the same site, leaf blast infection is high
during June to December, while panicle blast infection is high during January to

2 1996 Phil. Phytopath. 32(1):35-40

conditions favor infection. An important
consideration here is the influence of weather on
blast in determining temporal risk of rice to the
disease. Although several statistical approaches
are available to describe such an influence, the
use of multivariate procedures has been very
limited. Classification and ordination are
multivariate techniques that can be used to
identify temporal risk. Classification is concerned
with separating distinct sets of observations and
with allocating new observations to previously
defined groups (Johnson & Wichern, 1992).
Ordination attempts to find major axes of variation
among observations in order to reduce the many
dimensions of a data set to a very few, with
minimum loss of information (Anderson, 1971;
Beals, 1984).

In this paper, multivariate techniques were
used to identify temporal risk of rice to blast at
Cavinti and the IRRI blast nursery in the
Philippines and Sitiung in Indonesia. These
statistical procedures will help predict the
potential of blast epidemics if cultivars are sown
at different times of the year. Such predictions
allow early, cost-saving management decisions.
These methods may also help identify the
appropriate time and place for establishing
blast-related research sites to maximize
exposure of plants to the disease.


General Procedure

The flow diagram of the procedure used is
shown in Figure 1. This procedure was followed
at every site and for every cultivar at each site.
Initially, blast and weather databases were
obtained for Cavinti, the IRRI blast nursery, and
Sitiung. The blast databases were gathered from
the field experiments using cultivars IR50 (at
Cavinti and IRRI) and C22 (at Cavinti and
Sitiung). The weather databases had daily
values of weather variables recorded for several
years. From the disease and weather data,
empirical models predicting rice blast were
developed using multiple linear regression
(Calvero. 1994). At the same time, a 1-year
simulated weather database was constructed
for disease predictions and for use in
multivariate analysis (Fig. 1) to summarize the

overall weather trend of a particular site. Using
weather factors from a simulated weather
database as predictors in the models, blast
variables such as maximum and final diseased
leaf area (or final leaf blast index at Sitiung) and
panicle blast severity (or panicle blast index at
Sitiung) were also predicted for each cultivar for
24 hypothetical sowing dates. A matrix data set
(termed main matrix) comprising the predicted
disease values as column variables and the
sowing dates as row variables was assembled
for each cultivar at each site-for multivariate
analysis. Similarly, another matrix data set
(termed secondary matrix) was constructed with
different weather factors occurring from sowing
to the time initial blast symptom was observed
and from sowing to the time of flowering of the
cultivar as column variables and sowing dates
as row variables. This matrix served to relate
weather factors with the proneness of cultivars
to blast infection. Specifically, cluster analysis
was applied to the main matrix data set to identify
groups of sowing dates (or blast proneness
groups). Principal component analysis was
applied to both the main and secondary matrix
data sets to characterize the groups generated.
Discriminant analysis was then used to develop
discriminant functions to identify which group a
sowing date belongs. These methods are
described in detail below.

Models of Blast Variables

Empirical regression models of maximum
and final diseased leaf area and panicle blast
severity'were obtained from the equations
generated by Calvero (1994). These models
were developed using the results of the
WINDOW PANE program (Calvero etal., 1994),
procedure of which will not be discussed here.
The models are presented in Table 1.

Construction of a Weather Database

Historical meteorological database
containing daily values of temperature
(maximum, minimum, and mean in C), rainfall
(in mm d-1), relative humidity (in %), wind speed
(in m s-1), and solar radiation (in MJ m-2) were
obtained for each site from the IRRI Climate
unit. Wind speed values were not available at
Sitiung. Two (1985-1986), seven (1987-1993), and

1996 Phil. Phytopath. 32(1):1-17 3



disease data

Main matrix




Identify BPG
of sowing


Match new
sowing dates tc

Figure 1. Flowchart of the procedure for determining the temporal risk of tropical rice to blast dis-
ease. (BPG=Blast Proneness Group).







weather data


1996 Phil. Phytopath. 32(1):35-40

Table 1. Empirical equations used in predicting blast variables at Cavinti, the IRRI blast nursery, and

Models: Y = a + b,X, + b2X, +...+ bX

MDLA = [9.85 0.02 MMAX 10B16 0.46 DWS35 20B56 + 0.31 N1']2
FDLA = -35.11- 8.80 DWS35 10B46 + 1.33 MRH 30A6 + 2.04 log,(N)
PBS = [2.41 + 0.02 CDWP 40A39 3.00E-4 PFREQ 10A542]5

MDLA = [-8.48 + 3.47 loge(MMIN 10A5) 0.02 PFREQ 30A34 + 3.61E-9 N3]5
FDLA = [5.05 + 0.23 CDWP 20B37 + 2.86E-4 MSR 10A83- 2.66E-11 N4]4
PBS = [1.82 2.49E-10 TPREC 50A244 + 0.01 N1'2]6

IRRI blast nursery
FDLA = -11210.00 + 5757.22 DG25C 20B571' 0.05 CDWOP 10B352
PBS = EXP[-95.18 + 50.78 DG25C 20B541'6 + 1.96E-8 MRH 20B944]

FLBIn = 8.50 0.27 TPREC 10A1012
PBIn = 78.95 21.92 log,(DG25C OB30)

SMDLA and FDLA are maximum and final diseased leaf area (in %); PBS = panicle blast severity (in
%); FLBIn and PBIn are final leaf blast and panicle blast indices (in 0-9 values); MMAX = mean
maximum temperature (in C); DWS35 = number of days with wind speed 3.5 m s-1; N = nitrogen
amount (in kg N ha-1); MRH = mean relative humidity; CDWP = consecutive days with precipitation;
PFREQ = precipitation frequency (or number of days that are wet); MMIN = mean minimum temperature
(in C); MSR = mean solar radiation (in MJ m-2); TPREC = total precipitation (in mm d-1); DG25C =
number of days with maximum temperature > 25C; CDWOP = consecutive days without precipitation;
log, = natural logarithmic function; EXP = exponential function. MMAX 10B16 means that this weather
factor has duration starting at 10 days before sowing to 6 days after sowing. Similarly, PFREQ 10A54
means that this weather factor has duration starting at 10 days after sowing to 64 days after sowing.
The convention 3.00E-4 is also equivalent to 0.0003.

1996 Phil. Phytopath. 32(1):1-17 5

eight (1985-1992) years of data were available at
Sitiung, Cavinti, and the IRRI blast nursery,

In predicting blast variables using the
empirical models, a 1-year weather database
representative of the historical weather pattern of
a site was used. Using a simulated weather
instead of actual weather avoids the bias in
selecting what year to use in predicting the blast
variables at a particular site since not all years
are blast conducive years (Teng & Yuen, 1990).
The simulated weather offers an unbiased
approach in the selection process since it
summarizes to a single year, the statistical
properties inherent to weather variables running
for several years. Thus, simulated weather
considers both the conduciveness or non-
conduciveness of certain years in a weather
database to blast epidemic occurrence. The
simulated weather database was constructed
using a weather generator computer program
named SIMMETEO (Geng et al., 1988) with
monthly mean values of fraction of wet day
periods, maximum and minimum temperatures,
solar radiation, vapor pressure (in kPa), and
wind speed as input to the program.

Matrix Data Set

Main and secondary matrix data sets were
constructed for each site. The main matrix data
set had blast variables (i.e. maximum and final
diseased leaf area or disease index, and panicle
blast severity) as the column variables and the
24 hypothetical sowing dates (at half-month
intervals from January 1 to December 15) as
the row variables. The secondary matrix data
set had row variables similar with the main
matrix data set but the column variables are
weather factors occurring from sowing to the
time initial disease symptom become visible
(termed here as disease onset) and from sowing
to the time flowering occurs. In general, sowing
dates in each main matrix data set were grouped
based on leaf and panicle blast as measures of
blast proneness. The secondary matrix data set,
on the other hand, was used to investigate
weather influences on such proneness to further
characterize the groupings of the sowing dates.

The main matrix data set was transformed
to equally weigh the column variables in
multivariate analysis. Transformation also
reduced the occurrence of outlying sowing dates
and improved normality of the values in the data
set. Following conversion of percentage values
of leaf and panicle blast to proportion of disease,
attributewise (by column variables) relativization
by norm (Greig-Smith, 1983) was done as:

(E X2)1/2

where b. is the transformed value at ith sowing
date and jth column variable; and X, is the
untransformed value.

Statistical Analysis

Sowing dates were classified into distinct
blast proneness groups (EPGs) using the PC-
ORD system (McCune, 1993). The hierarchical
grouping through Ward's criterion with relative
Euclidean as the distance measurement was
used in clustering. This criterion gave the lowest
chaining (chaining is the sequential addition of
small groups to one or a few large groups) of
cluster dendrograms. Separation of BPGs was
done by slicing cluster dendrograms at a specific
distance measure. Group membership was
further assessed using discriminant analysis
procedure (DISCRIM) in SAS (SAS Institute
Inc., 1988).

Principal component analysis (PCA) with
variance-covariance as the resemblance
measurement was applied to ordinate sowing
dates based on the magnitude of predicted leaf
and panicle blast values. A parametric
correlation analysis of blast variables and
weather factors with principal component scores
was done to further explore the relationship of
the weather factors to the ordination of sowing
dates for characterizing the BPGs. Discriminant
analysis was then employed to generate
empirical equations which can be used to predict
the possible blast proneness group of a
particular sowing date, serving as explanatory
variables, the weather factors that had high

i iuO rnl. rnytopain. aJ I);O-QU

correlation with principal component scores. proneness of IR50 to blast when sown during
certain dates, the number of days with wind speed
RESULTS above 3.4 m s-1 occurring during sowing to
disease onset (or To) and the consecutive days
:lusterAnalysis with precipitation occurring during sowing until
flowering stage of the crop (or T,) had the highest
Three blast pronenesstgroups (BPGs) were correlation with the first (r = 0.85) and second (r
determined for each site. Fewer or more groups = 0.88) principal components, respectively. For
...,n4i h,- h~an rlicfinnnicha hult the thra r.99 tntal nrnPinitation occurrina during T had

iroup level in cluster analysis provided
distinction of dry and wet season and the
transition from dry to wet. Misclassified sowing
I-,- ------- S -- 4-k- &.s A- n--' '

between cvs. II
)last pronene:
group II of IR5
C22 were fou
R50 (Table 2).
at Cavinti an(
R50 showed
composition .
:he IRRI blast
sowing dates
Cavinti. At Siti
fell under grot

Principal Con

with the blast
the first print
- ri mtirn amn l

months at Cavinti differed
d C22. Sowing dates under
ip III of C22 mostly fell in
ing dates under group II of
er under group II or III of
rly, blast proneness groups
e IRRI blast nursery with
ly different membership
majority of sowing dates at
/ fell under group I. These
lassified under group II at
e majority of sowing dates
*able 2).

it Analysis for the Cavinti

icipal component scores
les on IR50 showed that
amponent describes the
-_ J-A.-- _- 4--k

with m
(r = -0.

the occ
this an
dates I
to Ma)
blast s,
would I
(Fig. 3
(Fig. 3
and mi<
group I
3D) bi
-6 ---

it r = -u.oo), wnie, consecutive uays
i temperature of 20-27C occurring
and number of days with wind speed
n s-1 occurring during To had the highest
I with the second principal component

ibove suggests that only wind speed,
re, and precipitation factors influence
ence of leaf and panicle blast at Cavinte
rtain times of the year. Based frorr
sis, planting IR50 and C22 during the
er blast proneness group I (January
r IR50 and January to mid-May anc
At for C22) would result to high panicle
rity (Fig. 3C and 3F) but less leaf blasi
ig. 3A, 3B, 3D, and 3E). Planting IR5C
i-May to June and mid-September tc
r (blast proneness group II for IR50;
I to moderate to high leaf blast severity
nd 3B) but low panicle blast severity
Planting C22 during June to Octobei
ovemberto December (blast proneness
r C22), would lead to high leaf infectior
i early phase of an epidemic (Figure
w to no leaf infection during the late
n 'F= In ariditinn r.99 wnilld likely

- - - -- .1 __ __ ..-- 'r_ 1 : ------ I ------ -- :_I_~ L_1__& ... L _ A4 -:~ ~ Cr~


1 v W I nI. Il LJ J LII. VAl I. II I

Table 2. Group membership of 24 hypothetical sowing dates at Cavinti and the IRRI blast nursery,
Philippines, and Sitiung, West Sumatra, Indonesia using rice blast parameters as attributes for
classification by cluster analysis.a

Cavinti IRRI Sitiung
Month Day SOW-ID IR502 C22 IR50 C22

January 1 1 I I
15 2 I I I
February 1 3 I II II
15 4 II III
March 1 5 III II
15 6 II Ill
April 1 7 I II Il
15 8 I
May 1 9 I I II
15 10 II II
June 1 11 II II I
15 12 III III I 11
July 1 13 III II I
15 14 III II II
August 1 15 III II I II
15 16 III I II
September 1 17 III II II
15 18 II II II
October 1 19 II II II
15 20 II II II I
November 1 21 II III II III
15 22 II II III Ill
December 1 23 II II III III

al= Group 1; II= group 2; and III= group 3. Groupings were generated by slicing cluster dendrograms at
specific distance measures.
bCultivars used at the sites.

8 1996 Phil. Phytopath. 32(1):35-4(

Table 3. Parametric correlation ot predicted bla
axes of 24 hypothetical sowing times at the thrn


Variable IR50
Axis 1 Axis 2 Axis 1

Disease variables"
Maximum leaf blast -0.89 -0.28 -0.08
Final leaf blast -0.99 -0.45 -0.98
Panicle blast 0.43 -0.90 0.61

Weather factor variables"
TPREC, -0.44 0.70 -0.68
TPREC, -0.68 0.49 -0.47
PFREQo -0.11 0.78 -0.46
PFREQf -0.44 0.86 -0.58
CDWPo -0.10 0.83 -0.44
CDWPf -0.37 0.88 -0.58
CDWOPo 0.13 -0.64 0.44
CDWOP, 0.53 -0.77 0.54
DR840 -0.25 0.53 -0.57
DR84, -0.62 0.41 -0.30

MMAXo -0.76 -0.18 -0.31
MMAX, -0.66 -0.58 0.11
MMINo -0.73 -0.24 -0.26
MMIN, -0.61 -0.62 0.15
DG25Co -0.76 -0.07 -0.29
DG25C, -0.65 -0.57 0.11
MAVEo -0.75 -0.20 -0.29
MAVE, -0.64 -0.59 0.12
DOPTo 0.36 -0.21 0.36
DOPT, 0.55 -0.64 0.58

variables and weather factors with the ordinatior

IRRI Sitiung

2 IR50 C22
Axis 2 Axis 1 Axis 2 Axis 1 Axis 2

0.75 -
-0.20 0.99 -0.12 0.32 0.95
-0.72 0.88 0.48 0.99 -0.06

0.44 0.07 -0.28 -0.54 -0.66
0.54 0.44 -0.49 -0.55 -0.24
0.33 0.08 -0.39 -0.51 -0.59
0.40 0.32 -0.54 -0.45 -0.26
0.34 0.14 -0.22 -0.47 -0.53
0.35 0.42 -0.54 -0.43 -0.25
-0.32 -0.01 0.42 0.53 0.60
-0.49 -0.20 0.53 0.44 0.26
0.16 -0.30 -0.22
0.53 -0.50 -0.26

0.49 0.15 -0.21 -0.79 -0.19
0.41 0.50 -0.60 -0.68 -0.37
0.49 0.67 -0.38 -0.79 -0.24
0.38 0.56 -0.57 -0.64 -0.33
0.53 0.59 0.13 -0.90 0.03
0.42 0.71 -0.21 -0.75 -0.21
0.49 0.67 -0.37 -0.81 -0.21
0.40 0.50 -0.60 -0.67 -0.36
-0.21 -0.57 0.54 -0.57 0.15
-0.50 -0.46 0.66 -0.60 -0.22

1996 Phil. Phytopath. 32(1):1-17 9

-0.28 0.16 -0.19 0.10 0.57 0.3(

temperature; DG25C= number of days with maximum temperature above 25C; MAVE= average mean
temperature; DOPT= number of days with MAVE of 20-27C; CDOPT= consecutive days with MAVE
of 20-27C; MRH=average RH (%); DRH80= number of days with RH > 80%; CDRH80= consecutive
days with RH 3 80%; MWS= average wind speed (m s-1); DWS35= number of days with wind speed
3.5 m s1; MSR= average solar radiation (MJ m-2). Weather factors with subscript "o" had durations from
sowing to disease onset, while "f' had durations from sowing to flowering stage of the crop.

10 iwo nll. mnytopain. z(1):j3-4u

blast at Cavinti and the IRRI blast nursery,
In characterizing the blast proneness respectively, showed no misclassified
groups using IR50 as the cultivar, sowing dates observations at E(AER) of 0%. Similarly,
under groups I (January and mid-April to equations predicting blast proneness group III at
October) and' II (February to April and mid- Sitiung had E(AER) of 0%.
October to November) may lead to high leaf (Fig.
4A) and panicle blast (Fig. 4B) infections DISCUSSION
However, sowing dates under group III (mid-
November to December) may lead to a very low The results show that the occurrence of blast
panicle blast infection (Figure 4B). outbreaks is related to weather conditions that
differ between sites and between cultivars at the
Principal Component for the Sitiung Site same site. At Cavinti for example, IR50 is less
likely to escape blast infection at Cavinti than C22
At Sitiung final leaf blast index had high because upland (dry) cultivation of IR50 may have
correlation with the second principal component disposed it to infection by P grisea. Although IR50
(r = 0.95), while panicle blast index had high and C22 are both susceptible to blast, IR50 is
correlation (r = 0.99) with the first principal more adapted to lowland conditions than C22.
component (Table 3). The number of days with
maximum temperature above 250C occurring Temperature appears to be directly linked to
during To had the highest correlation with the increased predisposition of IR50 at Cavinti. High
first principal component (r = -0.90). Total correlation of temperature factors to principal
precipitation occurring during T, had the same component scores with IR50 as the cultivar
degree of relationship with the second principal suggest this (Table 3). Similarly, a temperature
component. Therefore, only temperature and increase also trigger the decline of host resistance
precipitation factors could influence the to pathogen attack (Rotem, 1978).
occurrence of blast disease on C22.
At Cavinti, a wind speed factor was positively
If C22 is sown from January to December, correlated with the first principal component for
the cultivar may possibly get moderate to high IR50 and C22 (Table 3). On the other hand,
level of leaf blast infection (Fig. 4C). The level of maximum and final diseased leaf area on both
panicle infection is also high throughout the year cultivars had negative correlations with the first
except during mid-February to April, mid-June, principal component. Such a relationship
and November to December (Fig. 4D). suggests that high wind speed reduces the
possibility of leaf blast infection. The effect of wind
Discriminant Analysis speed above 3.4 m s-1 is related to the violent

1996 Phil. Phytopath. 32(1):1-17 11

01 2 3 4 5 6 7 8 0 1 2 3 4 5 6 7 8
Axis 1 (90.1%) Axis 1 (66.8%)
0.29 C f D
0.27 Group II Group I 12.5 Group III
0.25 ?11.5 E24' Group I1

.. %Gro u p II -

0.19 / 85 3 58 181

0.15 6.5

Axis 1 (94.0%) Axis 1 (84.8%)

Figure 2. Groupings of 24 hypothetical sowing dates relative to principal component axes. A = cv.
IR50 at Cavinti; B = cv. C22 at Cavinti; C = IR50 at the IRRI blast nursery; D = C22 at
Sitiung. Cumulative variances explained by the principal component axes are enclosed in
parentheses. Numbers beside the markers are sowing dates identification numbers (see
Table 2).

o o Group II

< Group II
o 0

Group 11

Group III Grou I Group

Figure 3. Ordination of 24 hypothetical sowi
Cavinti. Size of diamonds repres
Large-sized diamond figure means
(D), high final diseased leaf area c
on IR50 (C) and C22 (F).

cGroup I I Group 11

> G/op0

Sroup I 0o o (0 / Group I n
Group I Group II


Group I Group Il Group I
Wxis 1

Jates when overlaid with different blast variables z
s the degree of proneness of rice to the disease
h maximum diseased leaf area on IR50 (A) and C2
550 (B) and C22 (E), and high panicle blast severity

I A~~nn& na.:l 6.A -- &L. -2114%.IJ

IN Phil Phvtonnth. 32(11.1-17 1

Group III Group I

r Group II


< o

SGroup I


gure 4. Ordination of 24 hypothetical sowing tir
the IRRI blast nursery (A, B) and Sitiune
of proneness of rice to the disease. Larc
infection (A = IRRI blast nursery; C = i
blast nursery; D = Sitiung).

Group III group I
Group I

Group II

/ DGroup I


When overlaid with different blast variables at
3, D). Size of diamonds represents the degree
sized diamond figure means high final leaf blast
ing) and high panicle blast severity (B = IRRI

14 1996 Phil. Phytopath. 32(1):35-4C

Table 4. Discriminant functions generated for
3itiung, West Sumatra, Indonesia that classic
characterized by their proneness to blast outbre

Discriminant function

Cavinti, Philippines
LBc: D(Group 1)d= -30.550 + 5.172 I
D(Group 2) = -23.899 + 1.978 I
D(Group 3) = -8.955 + 0.468 D
IR50-PB: D(Group 1) = -324.182 0.077
+ 22.719 DWS3i
D(Group 2) = -180.529 0.043
+ 8.890 MMAXo
D(Group 3) = -186.327 0.051
+ 9.643 MMAXo
C22-PB: D(Group 1) = -3.406 + 1.296x1
D(Group 2) = -6.920 + 0.038 T
D(Group 3) = -10.453 + 0.047

IRRI blast nursery, Philippines
LB: D(Group 1).= -1518.000 + 178.
D(Group 2) = -1517.000 + 178.
D(Group 3) = -13A5.000 + 168.
PB: D(Group 1) = -29531.000 5.1'
223.995 PFRE
D(Group 2) = -28454.000 1.3'
219.292 "FRE(
D(Group 3) = -28167.000 11.A
219.461 PFREI

Sitiung, West Sumatra, Indonesia
LB: D(Group 1) = -98.849 + 9.090 I

intii and the IRRI blast nursery, Philippines, and
!4 hypothetical sowing dates into three groups

E(AER)b (%)

S350 + 0.510 CDWPo + 1'380 CDOPTo 11.10
S35. + 1.455 CDWPo + 1.128 CDOPTo 22.20
;35 +01.018 CDWPo + 0.670 CDOPTo 16.70
RECf + 4.088 CDOPT,
11.413 MMAXo 0.00
REC, + 2.666 CDOPT, + 16.334 DWS35o
REC, + 2.581 CDOPT, + 15.107 DWS35o
TPRECo + 1.336 DWS350 0.00
ECo + 8.116x10- DWS350 33.33
(ECo 0.106 DWS350o 50.00

DG25Co 0.463 CDWOPo 21.40
I DG25Co + 0.119 CDWOPo 14.30
I DG25C, 0.458 CDWOP, 33.30
)G25C, + 835.639 DG25C,
)G25Co + 819.276 DG25C,
DG25Co + 817.796 DG25C,

Kumagawa etal., 1957). However, wind speed that he likelihood of getting blast infection during
above 4 to 5 m s-1 also tend to injure plants, thus, sowing dates falling within blast proneness groups
nay facilitate pathogen penetration of hosttissues I and II was related to the number of days with
Sakamoto, 1940). maximum temperature above 25C and relative
humidity (RH) occurring from sowing to flowering.
At Cavinti, total precipitation, number of The positive relationship of these weather factors
lays that are wet (a day is wet when precipitation with the first principal component showed that
s greater than 0 mm d-1), and consecutive days increased leaf and panicle blast infections are
with precipitation occurring during sowing to crop due to more days with temperature above 250C
lowering stage had high positive correlations and high RH. However, since the temperature at
with the first two principal components where the nursery is commonly over 25C, it is likely
,anicle blast severity had negative correlation that RH effect on blast is greater than temperature
with. This suggests that precipitation factors are effect. Studies have shown that optimum leaf
directly involved in limiting the amount of blast development requires humidity of more than
noculum during the crop flowering stage for 80% (Suzuki, 1975). El Refaei (1977)
)anicle infection to be initiated mainly because demonstrated that stages in the P grisea
)f P grisea spores being washed off from monocycle are either directly or indirectly driven
potential inoculum sources (such as blast by RH. In panicle blast development, high RH
infected leaves). Ishiguro (1986) demonstrated prolongs wetness period in spikelets, which
hat large amount of spores is needed for triggers spore germination and aids in the infection
infection on panicles to be successful because process (Ishiguro & Hashimoto, 1991).
)f the erect orientation of the panicles and the
mpaction mode of spore deposition. At the IRRI blast nursery, sowing dates
within blast proneness groups I and II may lead
Precipitation factors at Cavinti occurring cv. IR50 to the same level of proneness to both
luring sowing to disease onset had favorable leaf and panicle blast mainly because of high
effect on leaf blast (Table 3). This relationship RH. Sowing dates in group III may lead to low
vas shown with leaf blast variables (maximum leaf and panicle blast outbreaks on the same
and final diseased leaf area) and precipitation cultivar because of low RH.
actors both having negative correlations with the
irst principal component in both the C22 and IR50 At Sitiurng, all sowing dates within blast
:ultivars. That is, low levels of leaf blast correspond proneness group II and some within group I may
o low values of the precipitation factors. Although lead to a high degree of proneness of C22 to
precipitation may wash off newly produced spores both leaf and panicle blast. Sowing dates in group
rom leaves, it may provide moisture for other III, on the other hand, may lead to high leaf blast
spores remained attached on conidiophores or infection than to panicle blast infection. From the
)n leaf surfaces. This free water is required by analysis, it appears that proneness to leaf blast
nost air-borne pathogens for the infection at this site is mainly due to the low amount of
processs to continue (Gunther, 1986). precipitation occurring from sowing to disease
onset. Several reports supported this analysis
In characterizing the blast proneness (Gunther, 1986; Kato, 1974; Suzuki, 1975)
groups generated for Cavinti, sowing dates showing that low rainfall prevents washing off of
within group I are prone to panicle blast due to spores from leaves allowing more pathogen
nore days with wind speed above 3.4 m s1 and propagules to be retained in the canopy. Teng et
ess precipitation. Sowing dates within group II al. (1991) further probed that reduced wash off of
nay produce high leaf blast severity due to low P grisea spores in plant canopies increases
vind speeds or frequent rainfall. Sowing dates autoinfection.
within group III may produce high leaf blast
infection due to low wind speed with Proneness to panicle blast at Sitiung is
considerablyy high temperatures. affected by temperature factors, although the
effect of which is rather indirect. An increased
At the IRRI blast nursery, it was observed number of days with maximum temperature

16 1996 Phil. Phytopath. 32(1):35-40

beyond 250C from sowing to disease onset (Table
3) actually reduces sporulation potential of
lesions on leaves (El Refaei, 1977; Kato &
Kozaka. 1974). This, in turn, leads to a reduction
in the amount of inoculum that would initiate
panicle infection later on (El Refaei, 1977; Kato,
1974). Similarly, several days of high temperature
produce non-viable air-borne spores due to
desiccation (Rotem, 1978).

Differences in the sowing dates falling into
blast proneness groups between sites are
extremely difficult to identify by just investigating
the long-term weather patterns at those sites.
Based on discriminant analysis, various weather
factors may influence the classification of
sowing dates into these blast proneness groups.
it is important to note that discriminant empirical
equations generated in this paper are cultivar-
and site-specific. However, verifying the ability
of these mathematical equations to predict blast
outbreaks at a given site using various blast-
susceptible cultivars can be the next priority for
research. A generalized method is potentially
useful in categorizing tropical blast hot spot
areas into groups according to similarities of the
physical (weather factors, soil factors)
environment. Likewise, it is possible to
categorize different rice cultivars into
susceptibility groups based on their reactions
to blast. The generalized method can result in
the use of mathematical models to predict blast
severity on cultivars with different genotypic
backgrounds and at different locations.


ANDERSON, A.J.B. 1971. Ordination methods
in ecology. Journal of Ecology 59:713-726.

BEALS, E.W. 1984. Bray-Curtis ordination: an
effective strategy for analysis of
multivariate ecological data. Advances in
Ecological Research 14:1-55.

R.J. 1992. Breeding rice for resistance to
pests. Annual Review of Phytopathology

CALVERO, S.B. Jr. 1994. Developing models
to predict favorable environments for rice

blast. MS thesis. Corvallis, Oregon State
University. 279 p.

McDANIEL, L.R., TENG, P.S. 1994. A
weather factor searching program for plant
pathological studies: Window Pane version
W1B00003. IRRI Discussion Paper Series
No. 5. Manila, International Rice Research
Institute. 43 p.

EL REFAEI, M.J. 1977. Epidemiology of rice
blast disease in the tropics with special
reference to leaf wetness in relation to
disease development. PhD Thesis. India,
Faculty of the Postgraduate School of
Indian Agricultural Research Institute. 195

LI, B. 1988. A program to simulate
meteorological variables: documentation for
SIMMETEO. Agronomy Report No. 240.
Davis: University of California

Biological control of rice diseases (blast
and sheath blight) with bacterial antagonists:
an alternate strategy for disease
management. p.87-110. In Pest
Management in Rice (ed. B. T. Grayson et
al.). New York, Elsevier Applied Science.

GREIG-SMITH, P. 1983. Quantitative plant
ecology. Oxford, Blackwell Scientific
Publications 359 p.

GUNTHER, H. 1986. Simulation of the
epidemiology of Pyricularia oryzae in rice.
Report of a 3-month M.Sc project.
Wageningen, Wageningen Agricultural

ISHIGURO, K.1986. Composing a simulation
model for epidemics of rice panicle blast.
Nogyo Gijutju 41:491-495.

Computer-based forecasting of rice blast
epidemics in Japan. p. 53-68. In Rice Blast
Modeling and Forecasting (ed. P.S. Teng).
Manila, International Rice Research

1996 Phil. Phytopath. 32(1):18-34 17


JOHNSON, R.A., WICHERN, D.W. 1992. Applied
Multivariate Statistical Analysis. New
Jersey, Prentice Hall. 642 p.

KATO, H. 1974. Epidemiology of rice blast
disease. Review of Plant Protection
Research 7:1-20.

KATO, H., KOZAKA, T. 1974. Effect of
temperature on lesion enlargement and
sporulation of Pyricularia oryzae on rice
leaves. Phytopathology 64:828-840.

Annual change of silicate absorption and
effect of calcium silicate in the rice plant.
Report of the Tokushima Agricultural
Experiment Station 2:13-14.

MCCUNE, B. 1993. Multivariate analysis on the
PC-ORD system. Corvallis, Oregon State
University. 139 p.

ROTEM, J. 1978. Climate and weather influence
on epidemics. p. 317-436. In Plant Disease,
Volume 2: How Disease Develops in
Populations (eds. J. Kranz, J. Rotem). New
York: Academic Press Inc.

SAKAMOTO, M. 1940. On the facilitate infection

of the rice blast fungus, Pyricularia oryzae
Cav. due to wind. Annals of the
Phytopathological Society of Japan 10:119-

SUZUKI, H. 1975. Meteorological factors in the
epidemiology of rice blast. Annual Review
of Phytopathology 13: 239-256.

guide for Release 6.03 edition. Cary, SAS
Institute, Inc.

TENG, P.S. 1994. The epidemiological basis
for blast management. p.409-434. In Rice
Blast Disease (eds. R.S. Zeigler, S.A.
Leong, P.S. Teng). Oxon, UK, CAB

analysis of the blast pathosystem to guide
modeling and forecasting. p. 1-30. In Rice
Blast Modeling and Forecasting (ed. P.S.
Teng). Manila, Philippines.

TENG, P.S., YUEN, J.E. 1990. Epidemic
models: Lessons from plant pathology. p.
211-220. In Risk Assessment in Genetic
Engineering: Environmental Release of
Organisms (eds. M. Levin, H. Strauss).
New York, McGraw Hill Publishing Co.

18 1996 Phil. Phytopath. 32(1): 18.



A portion of the MS thesis of the first aut
Baros, College, Laguna.

Respectively, Research Assistant, Plant
Research, Entomology and Plant Pathology
Baflos, Laguna; Associate Professor, Departi
Los Bafos, College, Laguna

Key words: epidemiology, Pyricularia g
spacing, simulation

A new rice plant type (NPT) is b.
This study was conducted to analyze
grisea on a NPT line, IR64454-81-1-3-;
experiments using a leaf blast model,
(N) level and plant spacing on blast
canopy structure and microclimate. In
on IR72 than on NPT at all N levels. I-
blast incidence were higher on NPT 1
blast severity on IR72 may be explain
canopy relative humidity, and lower le
the NPT
N level and plant spacing influen
glasshouse and simulated leaf blast
Leaf blast severity and collar blast inc
10 x 10 cm plant spacings than at 20 x;
as plant spacing increased.
The effects of N level and plant sl
the modification of crop structure an
experiment. Increasing plant spacing
canopy air temperature, and decrease
amount, leaf wetness duration, and
level increased, leaf area, light interce
and' leaf wetness duration increased,
mean tip angle.decreased.


The yield potential of high tillering, semi
dwarf varieties must be further increased frorr
10-11 to 15 t/ha by developing a new plant type
(NPT) to meet the growing demand of ar
increasing global population. This NPT should



* submitted to the University'of the Philippines Lc

thologist and Program Leader, Cross Ecosystem
vision, International Rice Research Institute, Lo
nt of Plant Pathology, University of the Philippine

sa, rice blast, new rice plant type, nitrogen, plar

g developed to further increase rice yield.
)last development caused by Pyricularia
nd IR72 in field, glasshouse, and simulation
ASTSIM.2. The effects of applied nitrogen
/elopment were determined in relation to
experiments, leaf blast severity was higher
ever, levels of field and glasshouse collar
n on IR72 at all N levels. The higher leaf
by a longer leaf wetness duration, higher
area and light interception in IR72 than in

j blast development in both cultivars. The
!verities increased as N level increased.
nce were significantly higher at 5 x 5 and
;m. Simulated leaf blast severity decreased

:ing on blast development were caused by
nicroclimate in both cultivars in the field
jsed an increase in leaf area and midday
n mean tip angle, light interception, dew
jday canopy relative humidity. As the N
on, dew amount, canopy relative humidity
while midday canopy air temperature, and

have a yield advantage of 20-30% over current
varieties. The physiological characteristics o
the NPT are: increased photosynthetic activity
increased light interception, improved assimilatE
partitioning, and improved nutrient uptake (IRRI
1989a). Some of the agronomic characteristic!
of the NPT are: low tillering, heavy and medium

)6 Phil. Phytopath. 32(1):18-34 19

zed grains, thick culm, erect and thick leaves,
nd slow senescence (Vergara et al., 1991).
management practices such as direct seeding,
maintaining a high plant population, and timing
nd methods of fertilizer application would have
be modified in growing the NPT (IRRI, 1989b).

Blast, caused by Pyricularia grisea,
mains a serious problem in irrigated rice in
mperate and subtropical lowland areas, and
I tropical upland areas (Bonman, 1992). High
population densities and rates of fertilizer
application are known to favor the development
f foliar diseases such as rice blast (Ou, 1985).
hus, it is important that the epidemiological
consequence of modified management
practices on rice blast development on the NPT
e understood before deployment over large
reas is undertaken.

The cultivation of NPT with modified
gronomic practices is expected to affect the
rop environment via changes in canopy
structure and microclimate. Therefore, the
environmental factors influencing blast
development should be studied in relation to
ie modified canopy architecture and
microclimate of NPT to provide insights into
last epidemiology and management.
furthermore, simulation using BLASTSIM.2, a
topical leaf blast simulation model, can be used
) explore the consequences of planting NPT
t different N levels and plant spacings as
affected by driving variables such as
environmental factors.

Calvero and Teng (1991) developed the
ILASTSIM.2 model which simulated the leaf
,last monocycle by calculating state variables
n a daily time step according to values of
riving variables. The state variables included
pore production and deposition, lesion
Drmation and expansion, and others. The
Iriving variables consisted of temperature,
elative humidity, lesion age, plant age, varietal
action, nitrogen, plant spacing, dew period,
Aind speed, and rainfall (Calvero et al, 1993).

This study was conducted to determine the
Effect of applied nitrogen (N) level and plant
;pacing on leaf and collar blast development
.- IDDIl DT 1 n-r I;- DA AAr-R1_1_-II_'_0 anrl

ami-dwarf indica variety, IR72, serving as the
antrol. Both cultivars were susceptible to leaf
nd collar blast. Disease development was
tudied in relation to canopy structure and
microclimate in field and simulation
experiments. A glasshouse experiment was
conducted to determine the effect of N level
nly on leaf and collar blast development on
oth cultivars under optimum microclimatic


experimentall Design and Treatment

A dry season irrigated lowland field
experiment was conducted in a split-split plot
esign in randomized complete block (RCB)
6ith three replications. Mainplot treatments
included the inoculated and control treatments.
he subplot factors were the combinations of N
wvels of 0, 90, and 180 kg N/ha and cvs. IR72
nd IR64454-81-1-3-2. Sub-subplot factors
included the plant spacings of 5 x 5, 10 x 10,
nd 20 x 20 cm. Each microplot corresponded
o a sub-subplot treatment.

A split-plot design in RCB was used in the
lasshouse experiment. The control and
loculated treatments served as the mainplot
actors. Subplot treatments included
combinations of N levels equivalent to 0, 90,
ind 180 kg N/ha and cvs. IR72 and IR64454-

.rop Establishment

The field and glasshouse experiments were
conducted at IRRI from Jan-April 1994. In the
eld experiment, levees werg constructed to
uild microplots measuring 3 x 1.5 m.
'regerminated seeds were dibbled hill-wise
allowing the different plant spacings with three
seeds per hill. Plants were thinned after a week
o one plant per hill. In the glasshouse
experiment, pregerminated seeds of the two
:ultivars were directly sown in plastic pots (500
;m3), with one seed per pot. Nitrogen in the form
)f urea (46-0-0) was applied in equal splits,
Before planting and early tillering stage (25 days
ifter sowina (DAS)) in both experiments.

ZU lss Phnil. Phytopath. 32(1): 18-

Intensive pesticiae application we

Inoculum Preparation and Application

P grisea isolate PO-6-6 was grown i
prune agar media for 7 days. The surfaces <
the plate cultures were gently scraped. Th
scraped plate cultures were exposed t
continuous fluorescent light for 3 days to indue

The spore suspension was prepared t
pouring sterile distilled water (10 ml) in eac
plate culture, scraping each plate culture gentle
and filtering the suspension in a beaker wii
layers of nylon mesh. The spore suspension we
then standardized at 150,000 spores/ml usin
a hemacytometer.

Field inoculation was done by sprayin
uniformly 1 L spore suspension onto the 18-dab
old plants in each sub-subplot at growth stag
21 (Zadoks et al., 1974). Spraying was don
using a hydraulic knapsack sprayer at 1700 i
Four inoculations were done at 3-day interval
starting from 18 DAS to induce the leaf bla,

In the glasshouse experiment, 15-day ol
plants at growth stage 21 (Zadoks et al, 1974
in the mainplot factor for inoculation wer
sprayed with the spore suspension inside
wooden inoculation chamber at 1700 h. Th
chamber was lined with wet jute sacks an
polyethylene sheets. Inoculation was perform
by spraying 100 ml of the inoculum as a mis
using an atomizer attached to a vacuum pump
model 5KH33DN'16X (General Electric Inc
U.S.A.) at 10 psi on each subplot for inoculatior
The inoculated mainplots were covered wit
polyethylene sheets after inoculation which wer
removed the following morning at 0700 h
Polyethylene sheet covering was done at nigl
consecutively up to 5 weeks after inoculation
to provide optimum conditions for infection.

Monitoring of Meteorological Factors

Meteorological factors were recorded usin!
automated Easylogger Weather Stationm
(Omnidata International, Inc., Utah, U.S.A.) ii

tne tieia experiment starting at the fir
inoculation date. Relative humidity and a
temperature within the crop canopy wei
monitored every 15 min using Physchel
relative humidity and temperature sensors
all treatments for inoculation in one replicate
Solar radiation, rainfall, and wind speed wei
measured for BLASTSIM.2 simulation. Le;
wetness duration was monitored-at a 15-m
interval by placing plate-type leaf wetnes
sensors attached to wooden stands at about tw
thirds height of the plant canopy. Leaf wetness
duration was measured in all treatments f(
inoculation in one replicate.

Dew amount in the crop canopy wa
measured using the method by Friedrich et a
(1991). "Dew papers" or filter papers (8.4 ci
diameter) supported by bamboo sticks with 0.
g crystal violet were installed at different height
of the canopy (30 cm interval) in each microplo
The "dew papers" were exposed overnigi
before each weekly sampling for bla,
assessment. Dew deposition was classifie
using a visual rating of the intensity of th
colored water on the filter paper surface, which
was calibrated into mm of water.

In the glasshouse experiment, relativ
humidity, air temperature, and leaf wetnes
duration were monitored at a 15-min interv,
using a hygrothermograph (Lufft, Inc., German)
installed per replicate starting at the firs
inoculation date. Daily air temperature range
from 24.0-32.20C, while daily relative humidil
ranged from 61.0-96.7%. Leaf wetness wa
maintained continuously for 24 h per day b
spraying distilled water on the leaves hourly
daytime and covering the set-up with wet jut
sacks and polyethylene sheets at nighttime.

In both experiments, canopy ai
temperature and relative humidity wer
measured per plot at midday (1100-1300 h
during the sampling dates for disease
assessment. Midday canopy air temperature
was measured using an AG42 Infrare,
Thermometer (Telatemp Corp., California
U.S.A.). Midday canopy relative humidity wa
measured using a temperature and relative
humidity meter (hygrometer) (ELE Internationa
Inc., England).

196 Phil. Phytopath. 32(1):18-34 21

measurement of Disease and Canopy
tructural Parameters

Three hills were destructively sampled at
indom from each plot at a 7-day interval
arting at 12 days after the first inoculation in
oth experiments. Leaf area, lesion number, and
visual estimate of leaf blast lesion area based
i the method of Pinnschmidt et al (1992) were
measured per hill. Measurement of leaf area
as done using an LI-3000 Portable Leaf Area
eter(Licor, Inc., Nebraska, U.S.A.). Leaf blast
severity (%) per hill was computed as the
estimated lesion area divided by total leaf area
ultiplied by 100. Collar blast incidence (%) was
assessed as the number of infected leaf collars
er total number of leaf collars per hill multiplied
r 100. Leaf nitrogen content (%) was analyzed
er hill on the leaf area samples using the micro-
eldahl procedure.

Mean tip angle and photosynthetic active
idiation (PAR) were measured at midday
100-1300 h) at the same sampling dates for
af area and disease assessments. Leaf
clination angle was measured as the mean
) angle using an LAI-2000 Plant Canopy
ialyzer (Licor, Inc., Nebraska, U.S.A.). Mean
Single was calculated as the mean inclination
igle from five angles of view (zenith angle) of
e LAI-2000 sensor that viewed the whole
inopy (Welles and Norman, 1991). Light
perception (%) was calculated from the PAR
easurements above and below the canopy
sing a Ceptometer (Decagon Devices Inc.,
'ashington, U.S.A.). Light interception was
ilculated as follows: Light interception (%) =
- (I/10)] x 100 where I = PAR at ground level
id o1 = PAR above the canopy.

ata Analyses

Disease, canopy structural, and canopy
icroclimatic parameters (midday
measurement) were analyzed using the ANOVA
ocedure (SAS Institute, Cary, NC, U.S.A).
anopy structural and microclimatic parameters
nong the treatments were analyzed using the
,sign of the experiments and Fischer's
otected least significant difference (LSD)
latment comparison. Leaf blast severity and

,UDPC). AUDPC was computed as:

UDPC (% day) = [(X,, + X,)/2] [t, t]

where X, = leaf blast severity or collar blast
cidence at ith i=1 assessment, t, = time
lay) between assessment dates, and n = total
amber of assessments. An appropriate data
ansformation was used for the particular
variable to be analyzed. Canopy relative
jmidity and air temperature, and leaf wetness
ration measured continuously were analyzed
sing a two-sample test of means and variance
Statgraphics, U.S.A.).

imulation Experiments

Leaf blast development on both cultivars
based on an inoculation at 18 DAS was
mulated using the BLASTSIM.2 leaf blast
odel. This was done to analyze the effect cf
eteorological factors on the leaf blast epidemic
5 influenced by the contrasting crop canopy
ructure. The first set of simulation used a 1994
weather data set from the IRRI wetland site,
tRI Climate Unit, which served as the control.

f simulated state variable
, U.S.A.). AUC was computed a

J. -70 ay) = L L~^1+1* A^,lLJ -l ,+1-+ I)J

where Xi = spore or lesion parameter at ith\
)servation, t, = time (day) between

22 1996 Phil. Phytopath. 32(1): 18-

variables such as spore production, release, ar
deposition; lesion number and size; an
predicted leaf blast severity were compare
among cultivars, N levels, and plant spacing

The receptivity factors (RF) of the hos
pathogen combination of P grisea isolate POI
6 with IR72 and IR64454-81-1-3-2 wei
determined in three glasshouse experiment
The receptivity factor or infection ratio is an inp
variable in BLASTSIM.2 which is determine
for each combination of P grisea isolate ar
cultivar. Plants were grown in 500 cm3 plast
pots with one plant per pot and fertilized wi
120 kg N/ha. Inoculated plants were of the sar
physiological age. The spore suspension we
prepared following the inoculum preparatic
procedure in the field and glasshous

Inoculation was done by spraying the spo
suspension (150,000 spores/ml) on 2% wat
agar slides placed equidistantly on a revolvir
wooden turntable at the bottom of a settlir
tower (1.12 m in length x 0.28 m in diamete
made of mylar film. Spraying was done in c
upward direction for 10 sec at 10 psi using
deVilbiss atomizer located at the bottom cent
of the settling tower which was attached to
vacuum pump, model 5KH33DN16X (Gener
Electric, Inc., U.S.A.). Spores were allowed
settle on the water agar slides for 2 min. Ti
rotation of the wooden turntable was control
by 3 volts output of an AC/DC adapt
connected to a Sankyo DFR 2R dynamo.

One inoculated water agar slide was place
horizontally on each leaf of 10 sample plan
per cultivar. Slides were removed after 24 h fro
inoculation. Inoculated plants were incubate
inside a mist room. Lesion number was count(
per leaf after 5 days from inoculatio
Receptivity factor (RF) was computed as tl
number of lesions produced on the plant divide
by the number of spores deposited on the plar
RF was determined at 10, 25, 40, 55, 70, 8
and 100 DAS.

The BLASTSIM.2 leaf blast model w.
supplied with the following daily data f'
simulation. Weather data included sol;

radiation, maximum air temperature, minimum
air temperature, rainfall, relative humidity, ar
wind speed. Crop data were collected from tt
lowland field experiment at 7-day interva
starting from 12 days after the first inoculatic
which included crop age, crop height, leaf are
and leaf width per hill.


Relationship Between Rice -Bla
Development and Crop Managemei

Plant spacings of 5 x 5 and 10 x 10 c
showed significantly higher AUDPC values
leaf blast severity and collar blast incidence the
the 20 x 20 cm plant spacing in both cultiva
(Fig. 1). This result confirmed those previous
reported which showed a higher blast severe
in closer plant spacing (El Refaei, 1977; Riberi
1982; Sah and Bonman, 1992). However, the
were no significant interactions between pla
spacing and cultivar-N combinations.

AUDPC (% day)
400 Ileaf blast severity DCollar blast incidenci

300- ____ ---



5x5 10x10 20x20
Plant spacing (cm)

Fig. 1. Area-under-the-disease-progress-cur
(AUDPC) of leaf blast severity and collar bla
incidence at different plant spacings averaged acro
applied nitrogen levels and cultivars (Fie

196 Phil. Phytopath. 32(1):18-34 23

There was no significant effect of N level at 0 kg N/ha, while IR72 at 90 and 180 kg N/ha
n the AUDPC of leaf blast severity in both had the significantly highest leaf areas.
iltivars (Fig. 2). This may be due to low blast Mean tip angle was not significantly different
verities in the field because microclimate was between inoculated and control treatments in the
at manipulated by polyethylenesheet covering field experiment (Table 1). Highest mean tip
nd hence, not optimum. However, IR72 at 90 angles were observed for IR72 and NPT at 0 kg
nd 180 kg N/ha showed significantly higher N/ha, while lowest mean tip angles were found
UDPCs of leaf blast severity than NPT at all in both cultivars at 180 kg N/ha. As N is applied
levels. High leaf nitrogen contents of both to rice cultivars with large leaves, the leaves
jltivars in all N levels may have also caused become longer and more drooped causing low
o significant differences in leaf blast severity mean tip angles (Tanaka etal., 1966; Chandler,
among N levels. 1969). Significantly higher mean tip angles were
found at plant spacings of 5 x 5 and 10 x 10 cm
The AUDPC of leaf blast severity showed than at 20 x 20 cm. Increased plant density lead

). IK/z at iu ana iou K9g Iina proaucea
significantly higher AUDPCs of leaf blast
everity than the rest of the treatments. In both
ultivars, the AU DPCs of leaf blast severity were
higher at 90 and 180 kg N/ha than at 0 kg N/ha.
*hese results follow those of previous studies
thich showed that a high N supply induced a
igh blast severity (Kawai, 1952; Beier et al.,
959; Kozaka, 1965; Sridhar, 1970; El Refaei,
977; Amin and Venkatarao, 1979).

The AUDPC of collar blast incidence in the
eld and glasshouse experiments indicated no
differences among N levels in each cultivar (Fig.
and 3). However, NPT had a significantly
higher AUDPC of collar blast incidence than
R72 across N levels in the field experiment.
i the glasshouse trial, the AUDPC of collar blast
icidence was higher on NPT than on IR72 but
nean comparison yielded no clear trend.

relationship Between Rice Blast
developmentt and Canopy Structure

The field experiment showed no significant
differences in leaf area per hill between
loculated and control treatments (Table 1).
significantly higher leaf areas were observed
in NPT than on IR72 at 90 and 180 kg N/ha.
.eaf area significantly increased as plant
pacing and N level increased in both cultivars.

In the glasshouse experiment, the control
treatment produced a significantly higher leaf
irea than the inoculated treatment (Table 1). This
howss the defoliating effect of leaf blast. The
,...,,r 1--i X i- :...--^ 11-441 -, :** l^ l .1:.-1-


No significant differences were found in
nean tip angles among inoculated and control
treatments, and cultivar-N level combinations
i the glasshouse experiment (Table 1). This
"an be due to reduced solar radiation.
furthermore, leaf inclination can be affected by
either factors such as cultivar difference, stage
if growth, leaf silica, water balance, crop
lensity, nutrients, etc. (Tsunoda, 1965; Trenbath
Ind Angus, 1975).

In the field experiment, the inoculated
treatment showed a significantly higher light
interception than the control treatment which
:an be influenced by low leaf blast severities
Table 1). Light interception was significantly
highest on NPT at 180 kg N/ha and lowest on
R72 at 0 kg N/ha. Light interception decreased
is plant spacing increased.

In the glasshouse experiment, the control
treatment had a significantly higher light
iterception than the inoculated treatment which
;howed the reducing effect of leaf blast infection
Table 1). Both cultivars at 0 kg N/ha showed
he lowest light interception. There were no
Significant differences in light interception
between the two cultivars at 90 and 180 kg N/

No significant differences in leaf nitrogen
;ontent were observed among all treatments in
he field trial (Table 1). This can be caused by a
linh nitmnrpn i intnk in ll N trPqtmPntfq h .,ai misp


Table 1. Summary of the comparisons of experiments and treatments on canopy structural parameters.'

Canopy Structure

Experiment Treatment Mean Tip Leaf Light Leaf
Angle Area Interception Nitrogen
(deg) (cm2/hill) (%) (%)

Field Inoculated ns ns higher ns
Control lower

Glasshouse Inoculated ns lower lower higher
Control higher higher lower

the 0 kg N/ha treatment showed a mean le;
nitrogen content of 4.4% at 30 DAS, which we
above the critical leaf nitrogen content of 2.51
(Tanaka and Yoshida, 1970). On the other han(
inoculated treatments showed a significant
higher leaf nitrogen content than contr(
treatments in the glasshouse trial. Howeve
there were no differences among cultivar-N levi

Two possible relationships can b
postulated on blast and N interaction. One i
the effect of leaf blast on leaf nitrogen, and th
other is the effect of leaf nitrogen on leaf bias
However, leaf blast severity in the field was nc
sufficient to cause differences in leaf nitroge
content or vice versa. Novero et al. (1992
showed that high amounts of N was nc
associated with high nitrogen concentration c
grains and straw in rice because the nitroge
use efficiency of rice in kg grains per kg nitroge
applied was low at 30%. Thus, increasing N ma
not affect the nitrogen content of any plant par

On the other hand, high leaf blast severit
in the glasshouse trial was adequate to caus
changes in leaf nitrogen composition. A high
leaf nitrogen content was observed in inoculate,
treatments than in control treatments. Thi
implies a direct effect of leaf blast on the
physiology of the rice plant. The increase in lee
nitrogen content due to leaf blast infection i
supported by previous studies reporting positive
correlations between leaf blast severity and lea
nitrogen content (Tokunaga, 1959; Kurschne
et al., 1992). A study showed that the progress

of leaf nitrogen content across time had a high
curve in the inoculated treatment than in th
control treatment from about 35 days aft(
transplanting up to harvest (Bastiaans, 1993).

The predisposition effect of increase
amount of N on an increased blast severity wa
reported in various studies. Amin and Katy;
(1979) found more total nitrogen in susceptible
than in resistant varieties. Increased nitroge
uptake by seedlings was associated with hig
susceptibility to blast (Otani, 1959; Kozak-
1965). Ammonium in rice leaves increased ce
water permeability which reduced its punctur
resistance (Suzuki, 1975). However, Roume
(1993) stated that the increase in susceptibilil
was more likely to result from a facilitated growth
after penetration due to an increase in solubl
nitrogen in the cytoplasm than from a reduction
of resistance to penetration of the cell wall.

Relationship Between Rice Bias
Development and Canopy Microclimate

Significant differences in dew amount wer
observed among N levels, plant spacings, an
vertical layers. Dew amounts in 5 x 5 and 10
10 cm plant spacings were significantly higher
than the dew amount in 20 x 20 cm plant spacin
at 37 DAS. Dew amount decreased from th
bottom up to the top of the vertical canop
layers. In both cultivars, the dew amount at 18
kg N/ha was significantly higher than the de\
amount at 0 kg N/ha. This can be supported b
a study which showed that cv. C22 with a dens,
canopy and erect leaves had the highest lee

)96 Phil. Phytopath. 32(1):18-34 25

AUDPC of leaf blast severity (% day) AUDPC of collar blast incidence (% day)
500 a eNew plant type U1R72 b iNew plant type I1R72
a D

400 LSD 400- LSDE.

300- 300-

200- 200-

0 90 180 0 90 180
Nitrogen levels (kg N/ha) Nitrogen levels (kg N/ha)

<* A ^ -_ l _1.-- -J: -_ *_ -_ .* jk.M / li kn ^\ ^ 1-\ -;r~ kl^|-* Mf ril 4r M-* ^ lrrll p

AUDPC of collar blast incidence (% day)
)0- IR72 iNew plant typ

b LSO,, D


0 90 180
Nitrogen level (kg N/ha)

JDPC) of a) leaf blast severity and b) collar blast
t different applied nitrogen levels (Glasshouse


AUDPC of leaf blast severity (% day)
100- MlIR72 UNew plant type

a LSD,.. I



0 90 180
Nitrogen level (kg N/ha)

Fig. 3. Area-under-the-disease-progress-curve
incidence in IR72 and the new plant typ

26 1996 Phil. Phytopath. 32(1): 18-34

blast severity and dew amount among cultivars
with contrasting canopy structures (Friedrich et
al, 1991).

IR72 had a longer leaf wetness duration
than NPT across N levels arid plant spacings
(Table 2). Leaf wetness duration increased as
the level of N increased. Leaf wetness duration
decreased as plant spacing increased. This
result follows the conclusions of past studies
which indicated longer dew period in dense
canopies than in sparse canopies (El Refaei,
1977; Blad etal., 1978; Scott eta., 1985; Royle
and Butler, 1986; Tompkins et al., 1993;
Deshpande et a., 1995). A low vapor pressure
in a less dense canopy can accelerate
evaporation from the canopy surface causing
a shorter dew period (Scott et a., 1985).

The effect of inoculated and control
treatments on midday canopy air temperature
was not significant in both experiments.
However, differences were observed among
plant spacings and cultivar-N level combinations
in the field trial. Midday canopy air temperature
increased as plant spacing increased. IR72 and
NPT at 0 kg N/ha had the highest midday
canopy air temperatures which were
significantly higher than the canopy air
temperatures of the rest of the treatments.
Midday canopy air temperature decreased as
N level increased in both cultivars.

Canopy air temperature (continuous)
showed no significant difference between
cultivars (Table 2). However, significant
differences were observed among N levels and
plant spacings, but the trend was different for
both cultivars. The result on canopy air
temperature (continuous) on NPT conforms with
the result on midday canopy air temperature.
Canopy air temperature (continuous) on NPT
decreased as N level increased. Canopy air
temperature (continuous) was lowest on the
closest plant spacing of 5 x 5 cm. Previous
studies showed that dense canopies have
reduced temperatures within the canopy (Blad
et al., 1978; Cappaert and Powelson, 1990;
Tompkins eta., 1993). Plants grown in a dense
canopy extract a high amount of soil moisture,
and produce high amounts of water vapor by
transpiration that cause a cooling effect

(Burrage, 1976).

A different result for canopy air temperature
(continuous) was observed in IR72, in which
canopy air temperature (continuous) increased
as N level increased. This can be explained by
the effect of leaf blast severity, as IR72 showed
a higher leaf blast severity than NPT in all N
levels. Measurement of canopy air temperature
detected increased temperature levels as the
area of senesced leaves increased. A previous
study on Septoria blotch on wheat showed that
canopy air temperature increased with an
increased disease coverage, a decreased leaf
area, and an increased number of dried leaves
(Eyal and Blum, 1989).

Inoculated and control treatments had no
significant effects on midday canopy relative
humidity in both experiments. However, the
effect of plant spacing was significant in the field
experiment. Significantly higher midday canopy
relative humidities were observed in both
cultivars at plant spacings of 5 x 5 and 10 x 10
cm than at 20 x 20 cm in the field experiment.
In support, many studies showed that relative
humidity increased with increasing plant
population densities (Colville, 1967; Rotem,
1982; Smith etal., 1988). High relative humidity
associated with a dense canopy can be related
to reduced air movement, increased
transpiration per unit volume of canopy, and
increased shading (Burrage, 1976).

The cultivar- N level combination effect on
midday canopy relative humidity showed no
clear trend in the field trial and it was not
significant in the glasshouse trial. However, a
higher canopy relative humidity (continuous)
was observed in IR72 than in NPT (Table 2).
High N levels also resulted in high canopy
relative humidity (continuous).

Interrelationships of Rice Blast with Crop
Management Practices, Canopy Structure,
and Microclimate

The AUDPC of leaf blast severity was
significantly higher on IR72 than on NPT at all
N levels in both field and glasshouse trials (Fig.
2 and 3). Under natural conditions in the field
experiment, there were no differences in leaf

1996 Phil. Phytopath. 32(1):18-34

Table 2. Summary of the field comparisons of cultivars on canopy microclimatic parameters (continuous

Canopy Microclimate
Air Temperature Relative Humidity Leaf Wetness
(C) (%) Duration (h)

IR72 nsd higher longer

NPT2 nsd lower shorter

'nsd no significant differences
2NPT new plant type

blast severity among N levels for both cultivars.
Under optimum conditions in the glasshouse,
differences in leaf blast severity were observed
among N levels. The higher leaf blast severity
on IR72 than on NPT can be partly explained
by the infection ratio or receptivity factor. The
receptivity factor determines the number of leaf
blast lesions that can be formed (Calvero et al.,
1993). Infection ratio is a critical epidemiological
parameter which is a component of resistance
that depends on a particular combination of host
and pathogen genotypes (Teng et al., 1977).

Collar blast incidence was significantly
higher on NPT than on IR72 at all N levels in
the field trial (Fig. 2). Differences in collar blast
incidence among cultivar-N level treatments in
the glasshouse experiment did not show any
clear relationship which can be due to high collar
blast pressure (Fig. 3). Torres (1986) found that
the same variety may respond differently to leaf,
collar, node, and panicle blast. In addition,
Pinnschmidt (pers. commun.) stated that the
effects of environmental factors on collar blast
can be different from leaf blast. The
susceptibility of NPT to collar blast can be due
to the cultivar's genetic susceptibility, and/or the
canopy structure that promotes a favorable
microclimate for collar blast development. The
steep angle of the leaf collar with the stem and
the high leaf area that shades the inside part of
canopy may induce longer dew retention in the
leaf collars.

The direct effect of N on leaf blast
development was observed in the glasshouse
study under optimum conditions without the

effect of plant spacing. Significant differences
in leaf blast severity were observed among N
levels. In addition, leaf nitrogen content was
higher in inoculated treatments than in control
treatments. Without the effect of the differences
in canopy microclimate among the treatments,
N influenced the development of leaf blast.

The levels of leaf and collar blast infection
did not cause differences in leaf area and light
interception between inoculated and control
treatments in the field because microclimate
was not optimum for infection. Higher leaf blast
severity and collar blast incidence in the
glasshouse caused the reduction in leaf area
and light interception in inoculated treatments.

Crop structural parameters varied between
the two cultivars in the field and glasshouse
experiments (Table 3). The NPT had a higher
leaf area and light interception than IR72 in the
field experiment. In the glasshouse trial,
differences in light interception and leaf area
between the two cultivars were not evident.
There was no significant difference in mean tip
angle between the two cultivars in both
experiments. Trenbath and Angus (1975) stated
that little emphasis has been focused on studies
on the comparison of leaf inclination between
canopies because very similar mean leaf
inclination angles were found between wheat
cultivars with contrasting canopy structures.

Significant differences in microclimatic
parameters were found between the two
cultivars in the field experiment (Table 2). IR72
had a longer leaf wetness duration and higher

laDle j. .ummary or me comparisons or experiments ana culuvars on canopy siruciur

Canopy Structure
Experiment Cultivar
Mean Tip Leaf Light
Angle Area Interception
(deg) (m2/hill) (%)

Field IR72 nsd lower lower

NPT2 higher higher

Glasshouse IR72 nsd nsd nsd


1 nsd- o1 significant differences
ns no significant trend
NPT -new plant type

canopy relative humidity (continuous) than NPT. progress of leaf and collar blast through
However, the cultivars were observed to have modification of the canopy structure ar
the same air temperature, dew amount and microclimate because leaf and collar bla
midday canopy relative humidity. decreased as plant spacing increased in bo
cultivars (Table 4). Leaf area and midday canola
The amount of N indirectly affected leaf air temperature increased as plant spacir
and collar blast development by modifying crop increased. Mean tip angle, light interception
structure and microclimate of both cultivars dew amount, leaf wetness duration, and midd;
(Table 4). As the N level increased in the field canopy relative humidity decreased as pla
experiment, corresponding increases were spacing increased. In support, other studio
observed for leaf area, light interception, dew have shown that an increase in host pla
amount, canopy relative humidity (continuous), density modifies the microclimate through
canopy air temperature (continuous) for IR72, reduction in air movement, increase in relatih
and leaf wetness duration, while corresponding humidity and leaf wetness duration, ar
decreases were observed in mean tip angle, reduction in temperature and amount
midday canopy temperature, and canopy air intercepted radiation (Rotem, 1982; Cappae
. ...... -.... /^, .;....... \ X, i-. IT" U .......;.. onrl rnd/M mlrtn 1QQf- Tnrnleine af n 10QQQ\


assnouse (increasing)

ncri Haroaci



py relative increase
lity (continuous)

Dy air variable
irature (continuous)

vetness duration increase

Dy relative nsd
lity (midday)

py air decrease
erature (midday)

)last severity nsd

Blast incidence nsd

- no significant differences

'or both IR72 a
,aqsp nr dcrerpa;

I new




30 1996 Phil. Phytopath. 32(1): 18-

LProduction IRelease EDeposition

20 LSDo.. LSD. I LSDo.,



5 -

!R72 NPT
LJProduction MRelease *Deposition

25- b LSD, 1 LSD., I LSDo,,,





0 -
0 90 180
Nitrogen level (kg N/ha)

3 UProduction Release EDeposition

25- c LSD.. LSD.. LSDo..,





5x5 10x10 20x20
Plant spacing (cm)

Fig. 4. Area-under-the-curve (AUC) of spor
production, release, and deposition
between cultivars (a), and among
applied nitrogen levels (b), and plar
spacings (c) (BLASTSIM.2 simulatio
using control weather data).

Area-under-the-curve (AUC)
sLesion number EOLesion size MSeverity

a8 LSDo. LSDo I LSDo.s I



b MLesion number DLesion size ESeverity

800 LSDoos LSDo, I LSD,,,r




0 90 180
Nitrogen level (kg N/ha)

sLesion number ELesion size ESeverity

800 LSDoI LSDo,,I LSDoo,I




5x5 10x10 20x20
Plant spacing (cm)

Fig. 5. Area-under-the-curve (AUC) of lesio
number and size, and leaf blast several
(%) between cultivars (a), and among
applied nitrogen levels (b), and plar
spacings (c) (BLASTSIM.2 simulatio
using control weather data).

o rnll. nytopatn. jz(1):18-34 31

Significant differences in leaf blast state
bles were caused by N levels and plan
;ings. The AUCs of spore production
ise, and deposition; and lesion number anc
were lower at 0 kg N/ha than at 90 anc
kg N/ha. The AUC of leaf blast severity
significantly higher at the plant spacing o
* cm than at 10 x 10 and 20 x 20 cm. Nc
ficant differences were observed on leal
severity among N levels (Fig. 4 and 5 ).

BLASTSIM.2 simulation using actua
her data generally produced similar results
the simulation using control weather data
pt for some minor differences. The AUCs
iore production, release, and deposition;
eaf blast severity were significantly higher
.72 than in NPT. The AUCs of spore
auction, release, and deposition were
:icantly lower at 0 kg N/ha than at 90 and
kg N/ha. There were no significant
ences in AUCs of lesion parameters among
'els but there were significant differences
AUCs of spore production, lesion number,
eaf blast severity among plant spacings.
AUCs of spore production and lesion
)er were significantly higher at the plant
ng of 10 x 10 cm than at 20 x 20 cm. The
of leaf blast severity significantly
ased as plant spacing increased (Fig. 6

Ir 1, iii LII .r 1 \, Iillll,, 11.,II L.

the simulation experiments showed no
icant differences in leaf blast severity
g N levels. This result conforms with the
:s of the field experiment, in which
*nces in leaf blast development were not
cant among N levels. This may be due to
accurate simulation of the leaf blast
mic in the field experiment by

nationss for the Cultivation of the New
'lant Type

he yield potential of NPT can only be
red by a close plant spacing as in direct
d rice because a dense plant population
ded to offset the low number of tillers. A
ve consequence of the direct seeding of
Likely to be the increase in leaf and collar
infection in close plant spacing. Thus,
dance to leaf and collar blast must be
orated in NPT before deployment over

J.C. KATYAL. 1979. Incidence
+^ ---.41;-- A 4., ,lnri(.

cing, and plant age. The receptivity due to
ogen and plant spacing is estimated by a
-ession equation: R2 = (-0.269 + 9.199*SP-1
0295*SQRT(N))2 where: R2 = receptivity due
nitrogen and plant spacing, SP = plant
cing, and N = nitrogen amount applied. In

Z. 96:140-145.

iTIAANS, L. 1993. Understanding yield reduction
in rice due to leaf blast. Ph.D. Thesis,
Wageningen, The Netherlands. 127 p.

D F n In DAIM7CD nnr Cr TI I ICIIIC in

32 1996 Phil. Phytopath. 32(1): 18

I20- llProduction MRelease =Deposltion
15- LSD, s, LSD.. I LSDo o
15 1



25 L-JProduction FReLease EDeposition
b LSDO., LSD .. I LSDo ,




0 90 180
Nitrogen level (kg N/ha)
25 LJProducion flRelease EDepositlon

C LSD..o SD, I LSD,,,




5x5 10x10 20x20
Plant spacing (cm)

Fig. 6. Area-under-the-curve (AUG) of spo
production, release, and depositic
between cultivars (a), and amor
applied nitrogen levels (b), and pla
spacings (c) (BLASTSIM.2 simulatic
using actual weather data).

Area-under-the-curve (AUC)
a ELesion number ELesion size MSeverity

600 LSDo.os LSD., I LSDo., I



0 b mLesion number 0jLesion size MSeverity

b8 LSDo.oi LSD... LSDo.,5




0 90 180
Nitrogen level (kg N/ha)
,Lesion number [jLesion size MSeverit

800 C LSDo. LSDo. [ LSDo.oi




5x5 10x10 20x20
Plant spacing (cm)

Fig. 7. Area-under-the-curve (AUC) of lesic
number and size, and leaf bla
severity (%) between cultivars (a), ar
among applied nitrogen levels (b), ar
plant spacings (c) (BLASTSIM
simulation using actual weather datE

96 Phil. Phytopath. 32(1):18-34 33

The interrelationship of nitrogen and other
factors affecting the blast disease of rice caused
by Piricularia oryzae Cav. Plant Dis. Rep.

3LAD, B.L., J.R. STEADMAN, and A. WEISS. 1978.
Canopy structure and irrigation influence on
white mold disease and microclimate of dry
edible beans. Phytopathology 68:1431-1437.

30NMAN, J.M. 1992. Durable resistance to rice blast
disease-environmental influences. Euphytica

3URRAGE, S.W. 1976. Aerial microclimate around
plant surfaces. In C.H. Dickinson and T.F.
Preece. (eds.) Microbiology of Aerial Plant
Surfaces. Academic Press, London. pp. 173-

Leaf Blast Model User's Manual. International
Rice Research Institute. P.O. Box 933, 1099,
Manila, Philippines. 85 p.

1993. Simulation of the tropical rice-leaf blast
pathosystem using BLASTSIM.2. (in

Canopy density and microclimate effects on
the development of aerial stem rot of potatoes.
Phytopathology 80:350-356.

CHANDLER, R.F. 1969. Plant morphology and stand
geometry in relation to nitrogen. In J.D. Eastin,
F.A. Haskins, C.Y. Sullivan, and C.H.M. van
Bavel. (eds.) Physiological Aspects of Crop
Yield. ASA-CSSA. Madison. pp. 265-289.

COLVILLE, W.L. 1967. Influence of plant spacing
and population on aspects, of the microclimate
within corn ecosystems. Agron. J. 60: 65-66.

1995. Estimating leaf wetness in dry bean
canopies as a prerequisite to evaluating white
mold disease. Agron. J. 87:613-619.

EL REFAEI, M.I. 1977. Epidemiology of rice blast
disease in the tropics with special reference to
leaf wetness in relation to disease development.
Ph.D. Thesis, Faculty of the Postgraduate
School, Indian Agricultural Research Institute,
Muw nfalhi 1QK n

:YAL, Z. and A. BLUM. 1989. Canopy temperature
as a correlative measure for assessing host
response to Septoria tritici blotch of wheat.
Plant Dis. 73(6):468-471.

KRANZ. 1991. A simple method for measuring
dew in upland rice canopies. J. Plant Prot.
Tropics 8(1):69-79.

1989a. IRRI Toward 2000 and Beyond. P.O.
Box 933, Manila, Philippines. 65 p.

1989b. Program Report for 1989. P.O. Box 933,
Manila, Philippines. pp. 5-7.

KAWAI, I. 1952. Ecological and therapeutic studies
on rice blast. Noji Kairyo Gijutsu Shiyo 28:1-

KOZAKA, R. 1965. Control of rice blast by cultivation
practices in Japan. In International Rice
Research Institute. The Rice Blast Disease.
John Hopkins Press, Baltimore, Maryland. pp.

ESTRADA 1992. Effects of nitrogen timing and
split application on blast disease in upland rice.
Plant Dis. 76(4): 384-389.

LUO, W. and J. GOUDRIAAN. 1990. Leaf wetness
in rice crops caused by dew formation: a
simulation study. Paper presented at the WMO-
Symposium on "Practical applications of
agrometeorology in plant protection", Dec. 4-
7, 1990, France.

1992. Grain yields and nutrient uptake of
irrigated maize, sorghum, and rice fertilized
with different levels of nitrogen. Philipp. J. Crop
Sci. 17(1):37-43.

OTANI, Y. 1959. Studies on the relation between the
principal components of rice plant and the
susceptibility to the blast disease and on the
physiological characters of the blast fungus.
J. Fac. Agric. Hokkaido Univ. 51(1):1-179.

OU. S.H. 1985. Ride Diseases. (2nd ed.)

,1 isuao mil. rnytopain. jz(l: 1i-.

BONMAN. 1992. A new assessment, key for A simulation analysis of crop yield loss due
leaf blast. Int. Rice Res. Newsl. 18(1):45-46. rust disease. Agric. Syst. 2:189-198.

RIBIERO, A.S. 1982. Effects of row spacing in the TOKUNAGA, Y. 1959. Studies on the relationship
rice blast disease attack. Pesq. Agropec. Bras. between metabolism of the rice plant and i
17(11):1691-1694. resistance to blast disease. I. Correlation
nitrogen and sugar contents of rice plant
ROTEM, J. 1982. Modification of plant canopy and blast disease. Bull. Tohoku Natl. Agric. Ex
its impact on plant disease. In J.L. Hatfield, Soc. Japan 16:1-5.
and I.J. Thomason. Biometeorology in
Integrated Pest Management. Academic Press, TOMPKINS, D.K., D.B. FOWLER, and A.
New York. pp. 327-342. WRIGHT. 1993. Influence of agronom
practices on canopy microclimate and Septor
ROUMEN, E.C. 1993. Partial resistance in rice to development in no-till winter wheat produce
blast and how to select for it. Ph.D. Thesis., in the Parkland region of Saskatchewan. Ca
Wageningen, The Netherlands. 108 p. J. Plant Sci. 73:331-334.

ROYLE, D.J., AND D.R. BUTLER. 1986. TORRES, C.Q. 1986. Effect of plant age on th
Epidemiological significance of liquid water in expression of resistance to Pyricularia oryzg
crop canopies and its role in disease Cav. in upland rice varieties. Ph.D. Thesi
forecasting. In P.G. Ayres and L. Boddy (eds.) University of the Philippines at Los Bahio
Water, Fungi, and Plants. Cambridge Univ. Laguna, Philippines, 82 p.
Press, Cambridge. pp. 139-156.
TRENBATH, B.R. and J.F. ANGUS. 1975. Le,
SAH, D.N., and J.M. BONMAN. 1992. Effects of inclination and crop production. Field Cro
seedbed management on blast development Abstracts 28(5):231-244.
in susceptible and partially resistant rice
cultivars. J. Phytopathol. 136:73-81. TSUNODA, S. 1965. Leaf characters and nitroge
response. In International Rice Researc
SCOTT, P.R., P.W. BENEDIKZ, H.G. JONES and Institute. Mineral Nutrition of the Rice Plan
M.A. FORD. 1985. Some effects of canopy John Hopkins Press, Baltimore, Maryland. pl
structure and microclimate on infection of tall 401-417.
and short wheats by Septoria nodorum. Plant
Pathol. 34:578-593. UDAGAWA, T. and Z. UCHIJIMA. 1969. Studies (
energy and gas exchange within canopies
SMITH, V.L., C.L. CAMPBELL, S.F. JENKINS, and Geometrical structure of barley canopies an

Phytopathology 8:595-600.

SRIDHAR, R. 1970. Nitrogen fertilization and the
rice blast disease. Plant Dis. Rep. 54(7):632.

SUZUKI, H. 1975. Meteorological factors in the
epidemiology of rice blast. Annu. Rev.
Phytopathol. 13;239-256.

TANAKA, A. and S. YOSHIDA. 1970. Nutritional
disorders of the rice plant in Asia. Int. Rice Res.
Inst. Tech. Bull. 10:1-51.

1966. Photosynthesis, respiration, and plant
type of the tropical rice plant. Int. Rice Res. Inst.
Bull. 7, 46 p.

VISPERAS. 1991. Rationale for a low-tillering
rice plant type with high-density grains. In
International Rice Research Institute. Direct
Seeded Flooded Rice in the Tropics. P.O. Box
933, Manila, Philippines. pp. 39-53.

WELLES, J.M. and J.M. NORMAN. 1991. Instrument
for indirect measurement of canopy
architecture. Agron. J. 83:818-825.

1974. A decimal code for the growth stages of
cereals. Weed Res. 14:415-421.

196 Phil. Phytopath. 32(1):35-40 35



Portions of undergraduate theses of the fir
problemm of the third author.

'Department of Plant Pathology, College of
031 College, Laguna.

Key words: "kamote kulot", sweetpotato fee

Enzyme-linked immunosorbent ass
used to detect sweetpotato feathery mott
Lam. on both symptomatic and sympto
from the field. Using ELISA, SPFMV v
33%, 53%, 67%, 71% and 82% of the tc
Tarlac, Laguna, Pampanga, and Zar
symptomatic and ELISA-positive sample
the development of vein clearing on lea,
Sweetpotato feathery mottle virus sig
Root yield losses of 98.10%, 97.28%, 95.J
UPL-SP., UPL-SP2, UPL-SP,, and Campl
of storage root was 50% in cultivars UPL
SP2 and 80% in UPL-SP2. The virus re(
85.4%, 86.3%, and 83,5% in UPL-Sf


Sweetpotato is one of the world's most
important crops. Historically, it had played a
significant role in providing relief food against
amine in the event of calamity in certain parts
)f the world. In the Philippines, it is the crop
recommended to be planted in areas affected
by the eruption of Mt. Pinatubo. It has been
utilized primarily for human consumption
because of its high nutritive value.

Sweetpotato is commonly propagated
sexually usually through cuttings to preserve
varietal purity aid to produce tubers earlier.
-owever, the vine cuttings are usually
-nntaminatpH hv in


id N.B. BAJET1

and second authors and of Special Research

riculture, 2BIOTECH, UP Los Bafios,

,ry mottle virus, yield loss assessment

(ELISA) and graft transmission were
/irus (SPFMV) in /pomoea batatas (L.)
iss samples of sweetpotato collected
detected in the leaf extracts of about
samples from the provinces of Albay,
ales, respectively. Grafting of the
>nto stocks of /. setosa Ker. resulted in
22 days after grafting.
cantly affected the yield of sweetpotato.
o, and 84.55% were incurred on cultivar
,respectively. Reduction in the number
1, and Campbell while 66.67% in UPL-
;ed sweet potato top weight by 92.0%,
UPL-SP2, UPL-SPs, and Campbell,

including plant viruses. The vegetative
propagation provides an excellent vehicle for
perpetuating the pathogens.

Among the various diseases affecting the
crop, the virus diseases constitute a serious
constraint to sustained and efficient sweetpotato
productivity. Some of the identified viruses
infecting sweetpotato include sweet potato
feathery mottle virus (SPFMV), sweetpotato
latent virus (SPLV), sweetpotato vein mosaic
virus (SPVMV), and sweetpotato mild mottle
virus (SPMMV). Of the identified viruses,
SPFMV is one of the most common and widely
distributed. It induces vein clearing, mosaic,
and mottling and can result to very poor tuber
production (Clark and Mover, 1988). SPFMV

36 1996 Phil. Phytopath. 32(1):35-,

has not been reported in the Philippine
(Tangonan and Quebral, 1992; Benigno ar
Quebral, 1976) until Nazarea (1990) an
Nazarea et al., (1992) reported its occurrence
"Kamote kulot", as popularly called by th
farmers, was noted in Paniqui, Tarlac. As tlh
word "kamote kulot" implies, the plants sho
curling and mottling of leaves, veinal chloros
and stunting. These symptoms still appe<
despite the application of fertilizer suggesting
that these symptoms were not due to nutriei
deficiency. In this paper, we report th
association of SPFMV with "kamote kulot", i
distribution, and effects on yield. Preliminai
results have been reported (Nazarea, 1991
Nazarea et al., 1992).


Sources of Plants and Preparation of
Samples for Assay

Plant samples were obtained from th
provinces of Albay, Laguna, Pampanga, Tarlai
and Zambales. Samples were collected L
taking two to three leaves starting from th
second youngest leaf down to the lower leave
About one gram of leaf disks from each sample
was obtained using sterile cork borer. Eac
sample was separately homogenized usin
mortar and pestle in carbonate coating buffe
pH 9.6, containing 0.01 M sodium diethy
dithiocarbamate (Na-DIECA) at a ratio of on
part sample and nine parts buffer. The extra
was either passed through three layers of gauz
cloth or briefly spun in centrifuge to remove plal
debris and the supernatant was immediate
tested using indirect ELISA.

Indirect ELISA

The standard indirect ELISA metho
described by Clark et al. (1986) was followed
with slight modification (Nazarea, 199(
Villegas, 1992). Microtiter plates (Dynatec
Laboratory, Alexandria, Virginia, USA) wer
coated with the extracts at 150 ul per wel
incubated overnight at 40C, and thoroughly
washed with phosphate buffered saline (PBc
containing Tween 20 (PBST). About 300 ul (

incubated for one hr at room temperature
emptied and washed as above.

Antiserum against SPFMV (obtained fror
Dr. J.W. Moyer, North Carolina State Universit
USA) was prepared using antibody buffer (PB
+ 0.2% skim milk) at a dilution of 1:1000 an
was added at 150 ul per well. The plate we
incubated for three hr and washed as above
Goat anti-rabbit enzyme conjugate (Sigm
Chemicals Co., St. Louis, MO; Catalog # ,
8025) diluted with the antibody buffer was adde
at 150 ul per well in the microplate. The plat
was incubated then washed as above. Th
alkaline phosphatase substrate, pare
nitrophenyl phosphate (Sigma, Catalog #10L
105), dissolved in 10% diethanolamine buffe
pH 9.6 was added to each well at 150 ul. Th
reaction was allowed to take place for
maximum of one hr and then stopped by addin
50 ul per well of 3N NaOH. Results wer
evaluated by reading the absorbance of eac
well at 410 nm using ELISA minireader

Graft Transmission

Seeds of indicator plant, /. setosa, wer
planted onto sterile soil. When two full
expanded leaves appeared, grafting was don
by partially cutting the portion between the ster
and petiole of the indicator plant. A branch fror
sweetpotato plant suspected to be infected wit
the virus was inserted into the cut portion. Th
graft was wrapped with a strip of Parafilm M t
prevent desiccation. The grafted plants wer
maintained in a shaded area and were observe
for symptom development.

Effects of SPFMV on the Yield of Sweetpotat

Sources of Planting Materials. Cutting
of commercially grown swveetpotato cultivars
UPL-SP,, UPL-SP2, UPL-SP5, and Campbe
were provided by Prof. R. Payson and Mr. F
Manguiat, Institute of Plant Breeding (IPB) an,
Dr. A.C. Sumalde, Department of Entomology
.UPLB. Sweetpotato plants of the same cultivar
exhibiting symptoms of SPFMV were als
collected in the UPLB experimental site of IPF

96 Phil. Phytopath. 32(1):35-40 37

Experimental Design. A split plot design
ith three replications was used. The cultivars
ere assigned in the main plots and the type of
planting materials (diseased or symptomless)
1 the subplots. The distance between rows Was
.75 m while the length of each rows was 10 m.
ine tip cuttings. 30 cm long, were planted on
dges at the rate of one cutting per hill and 50
n between hills.

Care of Test Plants. Standard
Neetpotato crop husbandry was followed.
andweeding was done by hilling-up. Care was
iken to prevent diseased vines from
itermingling with healthy vines in other rows.
missing hills in all treatments were replaced
nmediately only within two weeks after

Data Gathered. The following data were
gathered: growth and development, types of
symptoms, weight of tubers, top weight, number
tubers and percent yield reduction associated
ith different cultivars.

Percent yield loss, was determined using
ie formula of Hahn (1979):
L = (H-D)/H x 100, where: YL = yield loss
pressed in percent, H = plot yield of healthy
plants and D = plot yield of diseased plants.


urvey of SPFMV

SPFMV was found in all provinces
irveyed. It was detected by ELISA test in
2% of samples from Zambales, 71% from
ampanga, 66% from Laguna, 53% from Tarlac,
nd 33% from Albay (Table 1). ELISA results
ere confirmed by graft transmission. SPFMV
,as present in the infected plants as the
idicator host /. setosa exhibited symptoms 22
ays after grafting. Symptom appeared first as
ein clearing (2-4 leaves above the graft union).
eaves that developed above the first 2-4 leaves
,ere crinckled with vein banding and mild
istortion. These observations typical for
PFMV were similarto those reported by Moyer
nd Kennedy (1978).

Some sweet potato samples showing
different symptoms that were grafted to /. setosa
id not result to the production of symptom.
additionally of the grafted I. setosa plants that
Pere tested against anti-SPFMV serum, only
ne reacted to the serum. This observation
lay suggest that the symptomatic sweetpotato
lants were infected by viruses not related to
,PFMV or by completely different pathogens
rhich were.graft transmissible. Nome (1973)
as shown that viruses other than SPFMV, were
Iso graft transmissible to L setosa. Likewise,
eetham and Mason (1992) mentioned that
lentification of SPFMV in all major sweetpotato
rowing areas in the world is difficult because
f the many isolates and strains of SPFMV. It
'as also noted thatwhile infected sweet potato
lants normally gave positive results infected
amples with low level of virus did not give
positive result in ELISA.

SPFMV was readily detected in the
symptomatic sweetpotato samples except for
few samples with vein clearing or leaf curling.
ikewise, some symptomless plant samples
,ere also found positive for the virus (Table 1).
his indicates the presence of SPFMV and is
responsible for those symptoms (i.e. "kulot") on
Neetpotato. It also confirms previous report
iat SPFMV was detected in samples obtained
om the plots of the IPB, UPLB, and a number
f samples from Paniqui, Tarlac (Nazarea, 1990;
lazarea et al., 1992).

Our detection of SPFMV in symptomless
,af samples supports the findings of Green et
I., (1988) who reported that SPFMV was
detected in leaves with and without symptoms.
hey also found that SPFMV was not regularly
distributed along the length of the infected
weetpotato vine. SPFMV can be detected in
ome leaves along a vine but not in other leaves.
however, Cadena-Hinojosa and Campbell
1981) detected SPFMV only in sweetpotato
maves with symptoms.

Based on the above observation, SPFMV
s the possible cause of the "kamote kulot"
symptoms in sweetpotato and it appears
Widespread in the Philippines.

38 1996 Phil. Phytopath. 32(1):35-

Varietal Reaction to SPFMV Infection reduction was observed in cultivar Campbi
with 84.55% (Table 2).
Generally, the SPFMV-infected plants

SCampbell ar

I I I IO IUOlJ3 VVlI U 111 11 ai LU L IUO I UJUI L
e by Ngeve and Bouwkamp (1991) who observe

vigorous comp
UPL-SP, and I
with survival c

observed in e,
exhibited initi
which later be
leaf lamina e
noticed as th
cupping of the

JI- -Il-'.- were llilrnlv susGUutllU

5' --1 -to-
e infected I
f. was attrib
e losses d
S the differ
g and the e

JI ,JU LU ou-O 111 r
0 reduction in roc
f infection (Olive
ences in the ext
could be attribi
sed, strain of the
-lahn (1979) attr


by UPL-SP, we;e vein clearing, leaf distortion, SPFMV is likely the virus responsible for tl
chlorotic spots, puckering and yellowing. Vein "kulot" symptoms on sweetpotato. The extend
clearing and yellowing were the common severity of the "kulot" is affected by the cultivw
symptoms observed in Campbell. Leaf of sweetpotato. The presence of the virus
pigmentation which is inherent to the cultivar vines induces the leaves to curl which severe
may have contributed to the different types of impaired or hampered vegetative growth
symptoms associated with virus infection, shown by the poor ground cover. Po(
Alconero (1972) found a high correlation vegetative growth resulted to very poor tubi
between the color of the leaf petiole and development. The use of virus-free cuttings
symptom type. He observed chlorotic spots in planting materials is highly recommended
cultivars with no apparent purple pigmentation, minimize yield loss due to SPFMV.

Yield Loss Assessment ACKNOWLEDGMENT

Infection of the cuttings used as planting This research was supported by grant fro
materials by the SPFMV reduced significantly the Department of Science and Technolo(
the top weights of all four sweetpotato cultivars (DOST), Research Grant B2904Ag to N. I
tested as compared to the healthy plants (Table Bajet and in part by the Institute of Pla
2). The poor vegetative development of the Breeding (IPB), UPLB. We would like to thar
infected vines resulted to very minimal or poor Dr. J. W. Moyer for the gift of antisera and al.
ground cover up to the time of harvest or 3 5 to Drs. R. Bader (IPB), E. T. Rasco, Jr. and I
months after planting. The highest top weight Chujoy, CIP/SAPPRAD. The assistance of M
reduction was observed in cultivar UPL-SP, with V.Elbo and Mr. J. E. Padua are als
92.0% and the lowest reduction was observed acknowledged.
in cultivar Campbell with 83.5%. Top weight
reduction of 85.4%-and 86.3% was observed in LITERATURE CITED
UPL-SP2 and UPL-SP3, respectively. These
differences in top weight reduction among ALCONERO, R. 1972. Effect of plant age, lig
infected cultivars were not statistically intensity, and leaf pigments c
significant. symptomatology o7 virus-infectE
sweetpotato. PI. Dis, Reptr. 56:501-504
The reduction in top weights of the infected
cultivars resulted to the significant reduction in BENIGNO, D. R. A. and F. C. QUEBRAL. 197
both the size and number of tubers produced. Bibliography of plant diseases in tt

,I vie

uI ItLs. A/ L*O.0-/

Phil. Phytnpath. 32(1):35-40 39

>le 1. Incidence of sweetpotato teatnery mottle

No. samples
Province tested

ay 15

juna 15

mpanga 7

lac 15

bales 22

le 2. Effect of sweetpotato feathery mottle vir
sweetpotato cultivars.

%atment UPL-SP, UPL-S

Top Weigh

salthy 351.18a 300.

fected 31.27b 44.

%Yield Reduction 92.00 85.

Average 191.22a 172.

Root Weigt

s (S~-'MV) oasea on symptoms ana inairect

Incidence (%)

ymptom ELISA-Positive

40 33.3

00 66.7

71 71.4

66 53.3

72 81.8

(SPFMV) on top and root weights of some

UPL-SP, Campbell

ant (grams)

308.83, 426.06a

42.42, 70.34,

86.30 83.50

175.62, 248.20,

4U I0 u rmlll. rilyLUpVd u. JLt II.J

BEETHAM, P. and A. MASON. 1992. in sweetpotato (/pomoea batatas) usi
Production of pathogen-tested indirect enzyme linked immunosorbe
sweetpotato. Austr. Centr. Int. Agric. Res. assay (ELISA). Special Problem. UPL
Tech. Rep. No. 21, 47pp. CAS 17p.

CAMPBELL. 1981. Serological detection MALABANAN, L. D. VALENCIA and N.
of feathery mottle virus strains in BAJET. 1992. Detection of feathery mot
sweetpotato and /pomoea incarnata. PI. virus in sweetpotato. Philipp. Phytopath
Dis. Reptr. 65:412-414. 27.:50 (Abstract).

CLARK, M. F., R. M. LISTER and M. B. NGEVE, J. M. and J. C. BOUWKAMP. 19(
JOSEPH. 1986. ELISA Techniques. Meth. Effects of sweetpotato virus disease
Enzymol. 118:742-779. (SPVD) on the yield of sweetpote
genotype in Cameroon. Expt'l. Agr
CLARK, C. A. and J. M. MOYER. 1988. 27:221-225.
Compendium of sweetpotato diseases.
The Am. Phytopathol. Soc., St. Paul, NOME, S.F. 1973. Sweetpotato vein mos;
Minnesota, USA. 73p. virus in Argentina. Phytopathol. Z. 77:Z
GREEN, S. K., Y. J. KUO and D. R. LEE.
1988. Uneven distribution of two OLIVERO, C. A. and T. OROPESA. 19f
potyviruses (feathery mottle virus and Effects of sweetpotato feathery mottle vir
sweet potato latent virus) in sweetpotato on yield and other agronomic parametE
plants and its implication on virus indexing of sweetpotato. Agr. Trop. 35:167-172.
of meristem derived plants. Trop. Pest
Management 34(3):298-302. TANGONAN, N. G. and F. C. QUEBRAL. 195

Agric. 15:253-256.

MOYER,J.W. and G.G. KENNEDY. 197
Purification and properties uf sweetpotz
feathery mottle virus. Phytopatholo(

NAZAREA, A. P. 1990. Detection
sweetpotato feathery mottle virus (SPFM

273 p.

VILLEGAS, L. C. 1992. Detection
sweetpotato feathery mottle virus
/pomoeabatatas(L.) Lam. Undergradu
Thesis. UPLB-CAS. 45p.

)6 Phil. Phytopath. 32(1):41-50 41



'Institute of Agriculture and Animal Sciences
agriculture, Hokkaido University, Sapporo 060, J1
nes Centre, Norwich Research Park, Colney Lar

Key words: rice dwarf phytoreovirus, leafhop

Monitoring of rice fields from 1992 to
indicated that rice dwarf phytoreovirus ( R[
valley in Nepal. The incidence of RDV rar
15% during 1994, in Lalitpur district ( Lubh
appeared 30-40 days after transplanting ( I
the rate of about 0.45 hill/day, attained te
and caused more than 84% gain yield loss
The virus was transmitted by Nephott
and Recilia dorsalis. Partially purified prepare
polyhedral particles of about 72 nm diamet
rice and Echinocloa crusgalli leaves n
differentiated from Japanese isolates he


Rice dwarf is a destructive viral disease of
:e reported to occur in the temperate regions
f central and southern Japan, central and
authern China, Korea and.Nepal ( Hibino,
992). It is also reported to occur in the southern
hilippines, (Cabauatan et al., 1993). The
sease is caused by rice dwarf virus (RDV)
lenus: Phytoreovirus, family: Reoviridae),
which has icosahedral double-shelled particles
About 70 nm in diameter ( Uyeda and Shikata,
982), containing 12 segments of double-
randed (ds) RNA (Fuji-Kawata etal., 1970) and
even proteins (Nakata et al., 1978). The
iaracferistic symptoms of the infected plants
clude stunted plant growth and fine chlorotic
pecks on leaves which often coalesced to
irm longitudinal streaks along the midrib
-ibino, 1992). Rice dwarf is transmitted by rice
,afhoppers, Nephotettix cincticeps, N.
rescens, N. nigropictus and Recilia dorsalis


UYEDA2, and R. HULL3

IAAS ), Rampur, Chitwan,- Nepal; 2Faculty of
in, and 3 Virus Research Department, John
Norwich NR4 7UH, UK

rs, ds RNA, incidence, yield loss.

14 covering 81 locations of 21 districts
) was mainly restricted to Kathmandu
d from 5 40% during 1993 and 13 -
Ind Imadol). In this district, rice dwarf
r), spread very slowly until 90 DAT at
nal average incidence of about 15%,

< nigropictus, but not by N. virescens
)ns from infected rice leaves contained
The ds RNA genome of RDV-infected
rated as 12 bands and could be
g segment 8 ( S8s) of lower mobility.

y N. cincticeps which is the major vector in
apan, Korea and central China (Hibino, 1992).

RDV was first observed at Khumaltar
:ation in the Kathmandu valley, Nepal in 1977on
-e cultivars KT 32-2 and Taichung 176 with
characteristic symptoms such as chlorotic flecks
r streaks on leaves, reduced root or shoot
stem and either no or reduced panicles (John
Sal., 1979). Only three species of leafhoppers,
virescens, N. nigropictus, and R. dorsalis have
sen reported in Nepal (Sharma and Mathema,
977). N. nigropictus was considered as the
principal vector of RDV (John et al., 1979; Omura
tal., 1982 ). Using N. nigropictus, preliminary
rus-vector interactions were determined
'radhan and Khatri, 1980; Upadhyay et al.,
982) and rice germplasms were screened for
assistance (Upadhyay and Lapis, 1982). The
entity of the causal virus was confirmed by
ectron microscopic examination of leaf thin
Bctibns or crude extracts, and double gel

42 1996 Phil. Phytopath. 32(1):41-50

The occurrence of RDV was surveyed in bags and anesthesized for examination. Field
the major rice growing areas of the Terai, mid identification of leafhopper species was mainly
and high hills covering 81 locations of 21 based on the principal body parts (Wilson and
districts from 1992 to 1994. The incidence of Claridge, 1991). Five such samples were taken

rice hills witn dwarf symptoms was determined
from five randomly selected spots with about 100
to 200 rice hills for each cultivar. In the
Kathmandu valley, where rice dwarf was
previously reported (John et al., 1979), more
regular visits were made during the main rice
growing seasons (June/July to November /
December). During these visits, about 10-15 cm
portions of leaf samples were collected from
30-40 randomly -selected rice plants for each
cultivar from a field, wrapped in a moist paper
towel in a plastic bag, brought to laboratory of
the Plant Pathology Department, Institute of
Agriculture and Animal Sciences (IAAS),
Rampur and stored at -200C or preserved in a
desiccant until further examination. During these
visits, some diseased plants with typical rice
dwarf symptoms were also collected,
transplanted in plastic buckets with fertile
lowland soils and placed in a screenhouse. The
diseased plant samples were used for virus
transmission experiments.

Monitoring of Rice Dwarf and Leafhopper

Rice nursery beds and transplanted fields
at Imadol and Lubhu were regularly visited in
1993 and 1994 to determine the rice dwarf

from each cultivar and location, and the average
value was used to represent population
abundance. For species confirmation, a small
sample of the leafhopper population was
collected, preserved in glass vials with a
desiccant, brought to the laboratory and stored
at 40C until examination. The laboratory
identification was based on morphological and
genital characteristics (Wilson and Claridge,

Effect of Rice Dwarf on Growth and Yield

Plants from forty hills with and without rice
dwarf symptoms were collected randomly from
rice fields at Lubhu and Imadol in 1994. The
samples were grouped into five batches each
from plants with and without symptoms. For
confirmation, a composite leaf samples (taken
from leaf portions of each hills) from each batch
was collected, and preserved with desiccants
until examination by immunosorbent electron
microscopy (ISEM). Parameters responsible for
yield and yield components such as total and
effective number of tillers, plant height, panicle
length, filled or unfilled grains, and grain weight
were determined from 40 randomly selected
(20 from each group) panicles.

1996 Phil. Phytopath. 32(1):41-50 43

Particle Purification and Electron Microscopy These samples included some of the
symptomatic composite leaf samples from the
The particle purification and electron field study and infected plants used in yield
microscopy (EM) were performed at the John component experiment. The ds RNAwas directly
Innes Centre, Norwich, UK. A crude preparation extracted from these samples as they were small
of RDV was prepared basically as described by and not suitable for particle purification. The
Uyeda and Shikata (1982). About 50 g of frozen leaves were cut into about 2 mm pieces using
RDV-infected rice leaves.was cut into about 2 scissors and ground with an extraction buffer
mm pieces with scissors, ground in liquid [ (10 mM Tris-HCI, pH 7.0; 0.1 M NaCI; 1 m M
nitrogen into fine powder and thawed with 100 EDTA; 0.1% SDS; 0.1% mercaptoethanol -
ml of prechilled 0.1 M Na-K phospate buffer (pH bentonite (1 mg/ml )] at 1:2 dilution. Two hundred
6. ) in a beaker. The mixture was homogenised ul of the extract was transferred to an 1.5 ml
in a polytron for about 5 min and followed by microtube, a half volume of phenol: chloroform
another cycle after addition of 30 ml of chloro- added and the mixture vortexed. The ds RNA
form. The mixture was centrifuged at 6,000 rpm was precipitated with 600 ul of 99% ethanol at -
for 15 min and Triton X-100 was added to 1% 200C overnight and recovered by centrifugation
and stirred for 1 hr at 6C. The mixture was at 15,000 rpm for 10 min. The precipitate was
centrifuged at 27,000 rpm for 60 min in a RP 30 vacuum dried and sent by courier mail to the
rotor and the pellets were suspended in 2 ml of University of Hokkaido for further analysis.
phosphate buffer and kept at 4C overnight. The
suspension was layered on 5% sucrose cushion Electrophoresis of ds RNA
and centrifuged at 25,000 rpm for 90 min in a
Beckman 30 rotor at 4C. The pellets from 50g The genomic ds RNAs were characterized
of infected leaves were re-suspended in 400 ul by electrophoresis in 7.5% polyacrylamide gels
of phosphate buffer and are referred to as crude (20x40x0.08 cm) as described by Uyeda et al.
virus preparation. (1995). The genomic ds RNA extract from an
infected leaf of Lubhu and Imadol isolates of
The virus preparation was observed under RDV from rice and one ds RNA extract from an
the EM using the immunosorbent electron -infected leaf of Echinocloa crusgalliwere located
microscopy (ISEM) as described by Roberts and together with ds RNA extract of an RDV isolate
Harrisson (1979) with antiserum of the Japanese maintained at Hokkaido University as control.
RDV isolate.The grids were observed under a The electrophoresis was performed for 20 hr. at
JOEL transmission electron microscope at 30 100V using TAE buffer and bands in the gel
K magnification and measurements of virus visualized by silver staining (Sambrook et al.,
particles were taken from more than 50 viewing 1989).
fields from 4-5 grids.

ISEM was also used to confirm the .RESULTS
presence or absence RDV from crude extracts
of randomly collected composite leaf samples Occurrence and Symptomatology
with or without dwarf symptoms collected from
field survey (Kathamandu and nearby areas, and During 1992 to 1994, more than 116 rice
Parwanipur) and the yield component growing locations from different geographic
experiment. Some of these samples with typical regions (200 m to 2,200 m) covering 24 rice
rice dwarf symptoms were also used for growing districts of the country were visited and
extraction of ds RNA. rice hills with symptoms suggestive of rice dwarf
were observed only in the Kathmandu valley.
Extraction of Genomic ds RNA The symptoms of infected rice hills were similar
to those described previously (John et al; 1979;
The rice dwarf infected leaves preserved Omura etal., 1982) and included stunted growth
over a desiccant were used to extract genomic and whitish to cholorotic flecks on leaf which
A,4 DMA A, rncrihnri hw Mhniirr f nt (I Q194 r.nal-rsced to form lonaitudinal streaks alona the

44 1996 Phil. Phytopath. 32(1):41-50

Figure 1. Symptoms of rice dwarf disease in Ne
Lubhu, Lalitpur, Kathmandu valley, I
showing chlorotic flecks or streaks oi
veins (Figure 1 a,b). Similar symptoms were also
observed on a weed, E crusgalli but not on other
weeds species or on wheat.

During 1992, a few symptomatic rice hills"
were observed during flowering to grain-filling
stage of the crop from Lubhu and Imadol in
Lalitpur districts but a detailed survey was not
conducted. In more intensive surveys of rice
fields in Kathmandu valley and nearby areas
(Kathmandu, Lalitpur, Bhaktapur and Kabre
districts) during 1993-1994, rice hills with the
dwarf symptoms were observed in five locations
in the Kathmancu and Lalitpur districts (Table
1). ISEM examination of crude leaf extracts from
randomly collected composite leaf samples with
dwarf symptoms confirmed the presence of
RDV but no RDV particles were observed in
samples of leaves without dwarf symptoms
collected from adjoining areas of Kathmandu
valley (Banepa, Bhaktapur, Kabre) and
Parwanipur. During 1993, the incidence was
higher in Lubhu (40.2% ) than in Imadol ( 5.0%
) but negligible ( only one hill in one ha ) in three
other locations ( Sitapaila, Dhapakhel and

t .t 4

a section of rice field infected with rice dwarf at
al, (A), and a close up of rice dwarf symptoms
af lamina (B).
rice dwarf symptoms were observed in Sitapila
and Dhapakel (Table 1). The relative abundance
of leafhopper populations in these locations was
comparable during 1993 and 1994 (Table 1).

Leafhopper Abundance and Rice Dwarf

The relative abundance and species
composition of rice leafhoppers from all the
survey locations are reported elsewhere and only
the population abundance from rice fields where
rice dwarf was observed are presented here. In
these locations, N. nigropictus was predominant
(more than 98%) while N. virescens was
observed only in occasional samplings. Besides
rice plants, N. nigropictus was also collected
from grasses and weeds along the irrigation
canals and levees surrounding the rice fields.
Occasionally, a few N. nigropictus adults were
also collected in wheat fields planted after rice.
In rice fields above 500m, N. nigropictus was
also more dominant (above 73.3%) throughout
the country.

J I I I. I, ytq \ I 1. 1+0t U 4

2. Rel





I.U I / t U.
4.0 94.4 + 5
) Q QAi Q -L ,c


ease incidence ana learnopper aounaances were taKen as aescriDea in taoie 1.
I = nursery beds.

46 1996 Phil. Phytopath. 32(1):41-50

During early July, the leafhoppers were collected
from rice nursery beds .10 days after sowing,
thereafter the population increased until the crop
harvest (Table 2). In the nursery beds, no rice
seedlings with rice dwarf symptoms was
observed. Rice dwarf symptoms were first
observed 30-40 days after transplanting ( DAT)
only in rice fields. The disease incidence was
very low (1-4%) until about 90 DAT, thereafter it
increased at 0.41 hills/day per 100 hills and
attained average incidence from 12.5% to 14%
during the flowering and grain-filling stage of the
crop. During 1993, the incidence in some
isolated patches in the same location reached
up to 60% (Table 1). The presence of RDV in
composite leaf samples was confirmed by ISEM.
The leafhopper abundance during this period
was high.

Effect of Rice Dwarf on Growth and Yield

Under natural conditions, rice dwarf
significantly affected growth and yield
components (Figure 2). Rice hills infected at the
younger stages had more height reduction
(about 40%), produced 10% more tiller/hill but
most of them did not produce any panicle at all.
Rice hills infected at more advanced crop growth
stages did not produce symptoms on the main
tillers, but produced many infected side tillers.
The rice dwarf-affected hills produced 70% less
effective tillers and smaller panicles which
accounted for the 88% less filled grains.
Grain weights were also reduced drastically
(Figure 2A D )

Particle Purification and EM

Electron micrographs of a negatively
stained crude virus preparation (Fig. 3A) showed
polyhedral particles similar to those described
earlier (Omura et al., 1982). There were 20-60
times more virus particles on the RDV-
antiserum-treated grids than on untreated grids.
The particles were uniform in size with a
diameter ranging from 60 nm to 86 nm with two
peaks (67 nm and 76 nm) but with an overall
average diameter of 72 nm (Fig. 3B).

Analysis of ds RNA by PAGE

The electrophoretic mobility pattern of ds
RNAs from rice dwarf-infected rice leaves of the
Imadol and Lubhu isolates, from infected
leaves of E. crusgalli and of the Japanese
(Hokkaido) isolate are presented in Fig. 4. In
general, most of the segments of the Imadol and
Lubhu isolates migrated slower than those of
the Japanese isolates. However, mobility
patterns of the segment 8 (S8), S9 and S10 of
Nepal and Hokkaido isolates were much
different. The S8 of both Imadol and Lubhu
isolates migrated slower than the Hokkaido
isolate. The mobility pattern of S9 and S10 were
also distinct except for one sample from Imadol
isolate (NEI-1), which requires further
clarification. The migration pattern of ds RNA
from rice weed, E. crusgalli, was similar to the
migration pattern of rice isolate.


In a previous study (John et al., 1979), the
rice dwarf was reported from Lalitpur and nearby
areas of Kathmandu valley, and Parwanipur
Agriculture Station of Bara district from the
southern terai plain. Based on electron
microscopic observation of leaf thin sections and
gel diffusion serology of crude leaf extracts, the
causal virus from Lalitpur district was confirmed
as RDV ( Omura et al ., 1982 ). The results
presented in this paper further confirmed the
occurrence of RDV in Kathmandu valley 15 years
later and further characterize the virus. Based
on our survey results, the virus appears to be
endemic in some isolated areas of Kathmandu
valley with low incidence. These results differ
from the earlier reports (Johns et al., 1979;
Amatya and Manandhar, 1986) that the rice
dwarf was present in both the hills (Kathmandu
valley and nearby areas) and terai plains
(Parwanipur and Hardinath). However,
observations from these locations were based
on limited number of leaf samples, and those
from other regions were based on the absence
of typical rice dwarf symptoms. In these
experiments, no RDV particles were observed

1996 Phil. Phytopath. 32(1):41-50 47

Length (cm)
100 1

60 -

40 1-

Plant height Panicle length

Grains (No.)/panicle

80 -

SorrI grams

Filled grains

Tiller (No.)/hill


2 -


Grains weight (g)/panicle
3 1---

Total grains Filled grains

Figure 2. Growth and yield components of healthy and rice dwarf-infected rice plants. Plant height
and panicle length (A) total and effective tillers/hill (B) total and filled grains/panicle (C),
and weight of total and filled grains/panicle (D).

48 1996 Phil. Phytopath. 32(1):41-!




C 100


60 62 64 67 69 71 74 76 79 81 83 86
Width (nm)

Figure 3. Electron Micrograph of the virus particles present in crude virus preparation staine
with 2% uranvl acetate (A). Bar represents 200nm. Histoaram of width distribution of th

1996 Phil. Phytopath. 32(1):41-50 49

in the leaf samples without rice dwarf symptoms.
However, in the light of a recent report that RDV
occurs in a tropical area like the Philippines
(Cabauatan et al., 1993), and previous reports
on its occurrence in the tropical terai plains
(John etal., 1979), examination of large number
of leaf samples'are essential to confirm its
existence in the terai plains of Nepal.
Nevertheless, our results, together with earlier
reports, indicated that in spite of changing
agricultural practices, cultivation of virus-
susceptible cultivars, and the presence of high
vector population the incidence of rice dwarf has
remained low in Kathmandu valley for about two
decades. From the epidemiological point of view,
this finding is important and warrants further
studies identifying factors) responsible for its
low spread.

During the surveys, E. crusgalli plants with
typical rice dwarf symptoms were also frequently
observed in most of these infected rice plots.
Incidence of typical rice dwarf symptoms on the
ratoons of rice and E. crusgalli weeds, in a
previously rice dwarf-infected rice fields, was
also commonly observed during the off-rice
season. These observations suggest that the
virus possibly perpetuates from one season to
another on rice ratoons and some weeds, and/
or on leafhoppers ( Omura etal., 1982; Hibino,
1992). Further detailed epidemiological studies
similar to those in Japan and China ( Hibino,
1992 ) are essential to identify the factors)
responsible for slow virus spread and designing
appropriate management strategies in
anticipation of sudden virus outbreak.

The occurrence of RDV in the Kathmandu
valley appears to be a unique situation. The
nearest sites where the virus has been reported
are in Central and South China ( Hibino,1992 ),
which are separated from the Kathmandu valley
by the Himalayan mountain range. As it is
unlikely that infective leafhoppers could cross
this barrier it would seem that the Kathmandu
valley forms a closed ecosystem in which the
disease exists. How it got there originally is open
to speculation. Comparative analysis of both the
biological and genomic characteristics is
essential to clarify its origin. Some comparative
analyses of the RDV genome, initiated recently
in Japan, indicated heterogeneity in the genomic

-1 4 -7 4

.. -
S2t 111


84 pow
85 r-001

Figure 4. Comparison of the genomic ds RNA
migration pattern of Imadol ( NEI ) and Lubhu (NEL)
isolates of rice dwarf virus from rice and Echinocloa
crusgalli. The ds RNA directly extracted from rice and
Echinocloa leaves were electrophoresed in 7.5%
PAGE gel (20x40x0.08 cm ) and silver stained. The
isolates used are indicated above by abbreviations:
H= RDV Hokkaido University isolate, NEI= Imadol
isolate, NEL= Lubhu isolate, and NEL-E.c.=
Echinocloa crusgalli from Lubhu. The number of the
isolate represents the composite rice sample where
the RNAs were extracted. RDV-H was used as
standard control.


u iuo rnil. rnyiopain. 3L1):41-.

profile of field isolates of RDV from one location Rice Diseases. ( R. K. Webster and P. ;
(Murao et al., 1994) or from different Gunnell, eds. ), American Phytopathologic;
geographical areas. Society Press, St. Paul, Minnesota, USA.
JOHN, V. T, M. H. HEU. W. H. FREEMAN, and I
N. MANADHAR, 1979. A note on dwarf disease
Some biological characteristics of the N.MANADHAR,1979.Anoteondwarfdiseas
of rice in Nepal. Plant Dis. Reptr. 63: 784-78!
disease were studied earlier using N. nigropictus MURAO, K., N. SUDA, I. UYEDA, M. ISOGAI, I
and results showed that both the males and SUGA, N. YAMADA, I. KIMURA and E
females transmitted the virus efficiently SHIKATA, 1994. Genomic heterogeneity of ric
(Pradhan and Khatri, 1980). Screening of ricEr dwarf phytoreovirus field isolates and nucleotid
cultivars and advanced breeding lines indicated sequence variants of genome segment 12.,
that most local cultivars and many breeding lines Gen. Virol. 75: 1843-1848.
were susceptible to rice dwarf (Upadhyay and NAKATA, M., K. FUKUNAGA and N. SUZUKI, 197(
Lapis, 1982). We tested transmission using other Polypeptide components of rice dwarf virus
leafhopper species and showed that only N. Ann. Phytopath. Sc. Jpn 44: 288-296.
nigropictus transmitted the virus. The Philippine THAPA, 1982. Identification of rice dwarf viru
isolate of rice dwarf was also transmitted only in Nepal. Jpn Agri. Res. Qrt. 15: 218-220.
by N. nigropictus and not by N. virescens and OU, S. H. 1985. Rice Diseases. 2nd Editior
R. dorsalis (Cabauatan and Koganezawa, 1994). Commonwealth Mycological Institute, Ken
Recent comparative genome analysis indicated Surrey, UK. 380 pp.
that isolates from Nepal and Philippines had very PRADHAN, R. B. and N. K. KHATRI, 1980. Repoi
distinct electrophoretic profiles from Japanese on the occurrence of rice dwarf virus i
isolates, with S8s of slower mobility. Kathmandu valley. The Seventh Summer Crop
Workshop, Feb 24-27, 1980, Parwanipu
Agriculture Station, Parwanipur, Bara, Nepal.
of potato leafroll and potato mop top viruses b
We thank Mrs. R. B. Shrestha and N. K. immunosorbent electron microscopy Ann. App
Khatri for their assistance in insect and disease Biol. 93: 289-297.
monitoring, and Mr. Gary Lee for his assistance SAMBROOK, J., E. F. FRITCH and T. MANIATIS
in electron microscopy. This research was 1989. Molecular Cloning: a Laboratory Manua
supported by the Rockefeller Foundation's 2nd ed. Cold Spring Harbour Laboratory, Nev
International Program on Rice Biotechnology York.
SHARMA, K. C. and S. R. MATHEMA, 1977. Studie:
(Grant No: R 91003 # 119). on the field biology and control of leafhoppe
associated with paddy crop. pp. 33-34. Thi
Fourth Rice Improvement Workshop, March 15
LITERATURE CITED 18, 1977, National Rice Improvement Program
Parwanipur, Bara, Nepal. (mimeo).
AMATYA, P. and H. K. MANADHAR, 1986. Virus UPADHYAY, B. P. and D. B. LAPIS, 1982. Tetep:
diseases of rice and legume crops in Nepal; potential source of resistance to rice dwarf ii
status and future strategies. Trop. Agr. Res. Nepal. Int. Rice Res. Newsl. 7(5): 7-9.
Series 19: 3-13. UPADHYAY, B.P.H.E. KAUFFMAN and D.B. LAPIS
rARAlI IATAKI D n' '- V VCrVrAK1CI7AA1A bOA I -- - ... .. . - ..

16 Phil. Phytopath. 32(1):51-56 51

.J Il 1 II 1 L I II Il Ul- vV1 J I 1 I IL.. -I I .- r I Ii jl I I CA' J C'..L-*l I LS ..1^ i L I I I -'jI I- -III -.LII J I I I ... u
with canopy index. Higher crop canopy significantly delayed and reduced the rate of
PRS progression, but further reduction of infection rate was achieved by increasing
,i .. .. :,-. .

J. It Inousanu Iine alrdIes w r.it: Jdani~u lu ppaya I Ile uIbasc was 1L It~puI Ltu 111 Lrnl
arlywith annual production of 80.28 to 104.52 Philippines by Opina (1986) and he established
T. (BAS 1981 1993). the disease symptomatology, host range,

rphology of the causal virus. Based on the epidemic development of PRS.
)rmation generated, he concluded that the
ease was caused by Papaya Ringspot Virus
RSV). Ramos (1987) demonstrated that the MATERIALS AND METHO
is can also be readily transmitted through cleft
lifting, while Jovellana (1989) and Bayot etal. Preparation of Planting Materials
190) claimed that PRSV could be transmitted
ough seeds based on growing-on tests. Seeds of papaya cv. Solo were
igdalita etal. (1989) established the temporal allowed to germinate in seed bo
j spatial spread of PRS. They showed that emergence, the seedlings were tran
Disease progress curve was consistent with plastic bags with garden soil at the r
igmoid curve, typical of polycyclic diseases. seedlings per bag. The seedlings \
ing the logistic model, they estimated an in an isolated area, fertilized, frequer
rage infection rate of 0.74 per unit per week and sprayed with insecticide to ensu
ey also demonstrated a flattened disease and PRSV-free seedlings One rr
adient suggesting that the virus can be seedlings were transplanted in the fal
ciently dispersed over long distances. The few weeks ( May ) before the onset
3ve information indicate that PRS is capable season.
causing an explosive epidemic which could
ider any control measure designed to reduce Experimental Treatments
ial inoculum less effective.
The papaya seedlings were plc
Several attempts to develop effective three farmer's fields that were pre'
itrol measures for PRSV in the Philippines separately planted to three different(
ve proved either ineffective or of marginal field was regarded as treatment wi
iefit. Magdalita et al. (1988) demonstrated attributes as follows:
citation or systematic removal of infected
ints in papaya orchards as a worthwhile Treatment 1 -It was represent(
ictice against PRS, but under high disease ha field planted to 1-year old gu
assure, sanitation alone cannot sufficiently (cv Guaple) with planting distance c
ntrol the disease. They also evaluated The potential PRSV inoculum soL
ative crop isolation in conjunction with located within 100 m distance.Th
citation and they found significant disease canopy index (ICI) or the ratio of thec
ppression compared to the control. In a relative to the ground area was apl
ated. experiment, they also demonstrated 50%. Papaya seedlings were grow
integration of vector control by chemical guava rows following the same plantir
h sanitation and crop isolation, but failed Three months after transplanting,
provide additional benefit for the control of intercrop was shown to be lower than
:S. plants.

Intercropping of papaya with taller crops Treatment 2 It was represent
ve gained acceptance among papaya growers ha field planted to 6 year old .mE
PRSV endemic areas. This practice had (cv Carabao ) at a distance of 10 x
alved naturally because of the common degree of crop isolation was mode
servations that papaya plants growing along potential inoculum sources wei
er plants are less vulnerable to PRSV and beyond 199 m, but less than 500
i infected later When the plants developed was approximately 25%. Papaya
irketable fruits. were planted in double rows betwt
rows with planting distance of 3 x

1996 Phil. Phytopath. 32(1):51-56 53

Treatment 3 -It was composed of 0.7- ha development started 5 weeks after transplanting
field planted to 5 year old citrus ( cv Szinkom) and progressed slowly at the rate of 0.041 unit
with 3 x 3m planting distance. The. potential per week. At harvest time or about 25 weeks
PRSV inoculum point sources were located after transplanting, the incidence of PRS was
beyond 500 m The intercrop was rigidly pruned 3.4% or about 32 infected plants were rogued
out to maintain an ICI of 75%. Papaya seedlings out. When the potential inoculum sources were
were planted between citrus with a distance of extended over 100 m, but < 500 m and 5-year
1.5 m between hills. Three months after old mango trees were used as intercrop with
transplanting, the citrus crop was still higherthan canopy index of 25%, and with canopy height
the papaya plants, higher than the papaya plants (Treatment 2), the
onset of PRS epidemic development was 15
Crop Management Practices weeks after transplanting. This would mean a
10-week delay of PRS epidemic compared with
Three papaya seedlings were planted per treatment 1. The infection rate was 0.023 unit
hill and later thinned out to one hermaphrodite per week or a 2-fold reduction in the infection
plant per hill. Plants showing apparent PRSV rate compared with treatment 1. When the
infection and only female flowers were promptly potential inoculum sources were located beyond
rogued out. Fertilization was done at the rate of 500 m and citrus served as intercrop with 75%
120 Kg NPK (14-14-14) per hectare and was canopy index and higher canopy than the
applied at three splits, at planting time, at papaya crop (Treatment 3), the onset of PRS
flowering stage and 3 months after the second epidemic was further delayed by 14 weeks and
application. Weed control was alternately done 4 weeks compared it with treatment 1 and 2,
using powergrass cutter and glyphosate (Round- respectively. PRS epidemic however, failed to
Up) Whenever needed, papaya plants were progress up to harvest time or up to 25 weeks
sprayed with insecticide/acaricide to get rid of after transplanting. Isolation of papaya crops
insect pest or mite problems. Old or non- from potential sources of inoculum apparently
functional leaves and deformed fruits were tended to delay the onset of disease and reduce
promptly removed, the rate of PRS epidemic progression. The
further epidemic delay and reduction of infection
Data Gathered and Analysis rate appeared to be associated with the
increasing distance between the papaya crop
The incidence of PRS in each treatment and the potential sources of inoculum. Since
was monitored at weekly intervals until harvest aphid vectors usually find their potential hosts
time. A plant was considered infected if it showed through random probing, the probability of a
early disease symptoms as described by Opina viruliferous vector feeding on a potential papaya
(1986). Cumulative incidence of PRS was plotted host is reduced with increasing distance from
with time and infection rate was estimated using the inoculum sources. The above observations
logistic model. Area under the disease progress were consistent with the findings of Magdalita
curve (AUDPC) was also determined, et al. (1988) when they evaluated the effect of
relative crop isolation and sanitation on the
epidemic development of PRS. They. reported
RESULTS AND DISCUSSION a three-fold reduction of infection rate and a
delayed epidemic development by as much as
The effects of crop isolation, intercropping 15 weeks under relative isolation and rigid
and canopy structures of intercrop on the sanitation. However, because of relatively
epidemic development of PRS are presented in high infection rate after the onset of PRS
tables 1 and 2 and illustrated in figure 1. Results epidemic development, they concluded that
showed that when the potential inoculum while relative isolation and strict sanitation
sources were located within 100 m and when are worthwhile measures, these measures
guava served as intercrop with canopy index of would not satisfactory control PRS. Satisfactory
50%, and canopy height lower than the papaya solution of PRS, therefore, requires the
crop (treatment 1), the PRS epidemic integration of other control tactics.

54 1996 Phil. Phytopath. 32(1):51-56


0.03- Trt 1
o Trt 2
0. "-a- r
Trt 3

4 0.015-

E 0.01-
) ;--

0 - I-a---i- r-r, x-+-;,-K--i,,-x-)K----"

1 2 3 4 5 6 7 8 9 10111213141516171019202122232425
Time (Weeks)

Figure 1. Disease progress curves of papaya ringspot at various papaya croping systems.

Results suggest that the intercropping secondary virus transmission within the field. It
significantly suppressed the progression of PRS. also appeared however, that the effect of
The mechanism where intercropping reduce intercropping was not related .to the kind of
PRS epidemic progression may be related with intercrop, but rather associated with the intercrop
non-persistent relationship between PRSV and canopy structures such as canopy index and
aphid vectors. In this type of relationship, the canopy height. Despite the observed abundance
virus particles are stylet-borne and they are easily of aphid vector colonies due to the presence of
removed from the stylet during brief random preffered hosts ( eg. Corn, beans, etc. ) within
probing of the viruliferous vectors. Intercropping the vicinity of the citrus field, the occurrence and
therefore, will provide barrier or sieve to filter the progression of PRS appeared to be significantly
virus particles before they could be introduced lower compared with guava and mango
to papaya host. In effect, intercropping tends to intercrops. Results suggest that the height of
reduce the efficiency of incoming viruliferous intercrop exerted more effects compared to the
vectors thereby delaying the epidemic canopy index or the ratio of canopy coverage
progression. In addition, intercropping will further relative to the ground area. When the intercrop
reduce the PRS progression by reducing the canopy height was lower than the papaya crop,
probability of non-viruliferous vectors to acquire PRS occurred as early as 5 weeks after
PRSV or by further reducing the efficiency of tansplanting or even earlier than what Magdalita

Cumulative number of papaya plants infected with papaya ringspotvirus monitored
various experimental treatments


0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0

26 2
9R 9


Treatment 1 = crop area 0.6 ha, intercrop-guava, potential.inoculum point sources within 100 m, intercrop canopy index
ICI) 50%, intercrop canopy lower than papaya; treatment 2 = crop area 1.0 ha, intercrop mango, potential inoculum
sources > 100 m, but < 500 m, ICI = 25%, intercrop canopy height greater than papaya; treatment 3 = crop area 0.7 ha,
ntercrop citrus, potential inoculum sources > 500, ICI 75%, canopy height greater than papaya.

able 2. Proportion of papaya ringspot (PRS) at harvest time, rate of PRS progression, area under the
disease progress curve (AUPC) associated at the various experimental treatments.

Treatments' Disease Infection rate AUPC
(per week)

1 0.034a 0.041a 0.032a

2 0.007b 0.023b 0.007b

3 0.002b 0.000c 0.002b

Treatment 1 = crop area 0.6 ha, intercrop-guava, potential inoculum point sources within 100 m, intercrop canopy index
ICI) 50% intercrop canopy lower than papaya; treatment 2 = crop area 1.0 ha, intercrop mango. potential inoculum
sources > 100 m, but < 500 m, ICI = 25%, intercrop canopy height greater than papaya; treatment 3 = crop area 0.7 ha,
ntercrop citrus, potential inoculum sources > 500, ICI 75%, canopy height greater than papaya.

56 luu9o nii. rnytopatn. iz(i):o

( 1988 ) and Canlas (1991) observed in their LITERATURE CITED
field experiments without intercropping.
However, when the intercrop canopy height was BAYOT, R.G., V.N. VILLEGAS, P.M. MAGDALI
higher than the papaya crop, significant delay M.D. JOVELLANA, T.M. ESPINO and S
EXCONDE. 1990. Seed transmissibility
and suppression of the epidemic development EXCO trmissibili
papaya ringspot virus. Philipp. J. Crop I
were observed despite the reduction of canopy 15(12): 107-111.
index by 25%. Increasing the canopy index in CANLAS, M.L.J. 1991. Anthropod transmission
combination with higher intercrop canopy appear papaya ringspot virus. B.S. Thesis, UPL
to further delay the epidemic development and College, Laguna. 87 p.
reduce the progression of PRS. The further JOVELLANA, M.D. 1989. Transmission of papE
increase of the intercrop canopy index was ringspot virus. BSA Thesis, UPLB, 30 p.
observed to be counter productive since the MAGDALITA, P.M., O.S. OPINA, R.R. ESPINO E
intercrop will compete with the necessary solar V.N. VILLEGAS. 1988. Epidemiology
r n fr p a dt papaya ringspot in the Philippines. Phili
radiation for papaya development. Phytopathol. 25:1-11.
Phytopathol. 25:1-11.
OPINA, O.S. 1986. Studies of new virus disease,
This study clearly demonstrated the papaya in the Philippines. In Plant Vii
feasibility of satisfactory PRS management by Diseases of Horticultural Crops in the Trop
integrating strict sanitation practices with crop and Sub-tropics. FFTC. Book Series No.
isolation and intercropping papaya with crops PABLO, S.J. 1988. Commodity industry situati
having higher canopy height and canopy index In State of the Art and the Abstract Bibliograp
not greater than 75%. Such practices tend to Papaya research. Crop State of the
reduce the initial inoculum and the subsequent Bibliography Series No. 14: p. 1-6
PCARRD. 1977. The Philippine Recommends
rate of PRSV progression. Papaya. Los Baios, Laguna. 54 p.
RAMOS, C.S. 1987. Characteristization of
unidentified virus attacking papaya (Car
papaya L.) in the Philippines. M.S. The!

6 Phil. Phytopath. 32(1):35-40 57

'hytopathological Notes:



Respectively, University Reseacher, li
rofessor,Department of Plant Pathology, College
anos, College, Laguna

Key words: mussaend# PTA-ELISA, TMV

Mussaenda plants showing ch
streaks and color deviation in the flower al
putative virus can be transmitted by mec
react to an anti-tobacco mosaic tobamov
cucumovirus, tobacco etch potyvirus, chili
wilt tospovirus. Based on host range reach
virus belongs to the tobamovirus group. S
material both here and abroad, measures
and transfer of virus-infected mussaenda
potential source of inoculum for other cro


Mussaendas, collectively called Doias,
re perennial ornamental shrubs. They are
characterized by the presence of one or more
f the calyx lobes developing unusually into a
irge, colorful petaloid structure. They have
een widely utilized as accent landscape
materials in the Philippines and abroad. They
re considered to be one of the most
outstanding contributions of the Philippines to
ie world of ornamentals.

Napi (1959) on his study on the
ansmission of different plant viruses to abaca,
sported the occurrence of mosaic in
lussaendaphilippicavar. Aurorae. Similarly,
quino (1981) reported a disease symptom
similar to the mosaic pattern on mussaenda
plants collected at Los Baios, Laguna. He
demonstrated that mussaenda with typical virus-
ke symptoms have intracellular inclusion
odies which included needles and crystals, X-



itute of Plant Breeding, and Associate
Agriculture, University of the Philippines Los

)tic mosaic pattern in the leaves and
,arto be associated with a virus. The
nical means and by grafting and can
s serum but not to cucumber mosaic
n mottle potyvirus, and tomato spotted
n and serological analysis, the putative
e mussaendas are popular landscape
iuld be done to prevent the movement
ants. Infected mussaenda could be a

odies, and granular bodies.

Recently, we have observed a number of
ur mussaenda plants in the germplasm
collection showing typical virus symptoms.
;ymptomatic plants were not only confined to
4. philippica var. Aurorae where the putative
irus was first reported, but to other mussaenda
varieties as well. This study aims to
haracterize and describe the symptoms
associated with the disease, elucidate how the
disease is transmitted and identify the putative


Symptomatic mussaenda plants were
characterized and leaves with typical virus
symptoms were collected.

Sap was extracted and a plate-trapped
ntigen-enzyme-linked immunosorbent assay
PTA-ELISA) test was performed using different

58 1996 Phil. Phytopath. 32(1):35-,

irus (CMV), tobacco r
rus (TMV), tobacco etch p
li vein mottle potyvirus (CVM
dotted wilt tospovirus (TSWV,

remaining sap was inoc
ally to other mussaenda pla
hosts, such as Nicotiana gl/
im and Chenopodium amai
re maintained in the greenhc
development. When syn
id another PTA-ELISA te
i ii inn the nFrevious antiserur

necrotic local lesions that de,
tinosa were cut out, homogel
in mortar and pestle and s
ally inoculated to mussaende
elated plants were maintaine
se for symptom development

another experiment, ir
ja shoots were grafted onto
Sand vice versa. Plantswere
ed in the greenhouse for s)

few virus-like particles of
mussaendas were rigid rods typi
results confirm and extend the
(1959) and the findings of Aqui
reported that the symptomal
plants he examined for inclusi
infected with TMV as the I
characteristics of those inclusior
described for TMV.

iv. giulinosa mecnanica
the virus showed typical lo(
while that of N. tabacum

symptoms on tneioung lear
mussaenda plants mechanic;
the virus showed typical vir
month after inoculation. Th
virus in this inoculated mus.
confirmed through PTA-ELI,

Similarly, grafting of p
healthy shoots grafted to
developed chlorotic mosaic I
after grafting and older leave
grafted to infected shoots be
manner. PTA-ELISA test also


saenda plants showed
ern, vein clearing and Aqui.,o, V.M.
/es and color deviation virus-inf
. Symptoms appeared Philippin
its and flowers and Laguna.
ves and flowers.
Napi, G.N.
racted and run on PTA- transmis
-ent virus antisera, the abaca.
positively to the anti-TMV Laguna.
-a of CMV, TEV, CVMV,

1981. Inclusion bodies of
cted economic plants i
s. MS Thesis. UPLB, Co
8 p.

959. Further studies o
ion of different plant viru,
3S Thesis. UPCA, Col

-- - -- -u V -" - --

.. . . . . . . . . . .

'^"'"' """'

ol nins aner.

able 1. Reaction of infected mussaenda to different virus antisera in the PTA-ELISA.

Reaction (+/-)
ntisera Healthy Infected
mussaenda mussaenda

Dbacco mosaic virus +
ucumber mosaic virus
tobacco etch potyvirus
hili vein mottle potyvirus
tomato spotted wilt tospovirus

able 2. Reaction of different mussaenda plants and other diagnostic host to TMV antiserum
and their corresponding absorbances at 405 nm in PTA-ELISA test.

Entries Absorbance Reactionb
(405 nm)a

ifected Mussaenda hybrid 1 0.097 +
ifected Mussaerlda "Dofa Luz" 0.076 +
controll healthy Mussaenda Doia Aurora 0.030
dussaenda flava 0.012
;rafted Mussaenda 0.0665 +
V. tabacum inoculated with Mussaenda virus 0.069 +
. glutinosa inoculated with mussaenda virus 0.066 +
/4. philippica inoculated with mussaenda virus 0.066 +
flussaenda inoculated with local
lesion of N. glutinosa 0.082 +
'MV infected N. tabacum 0.102 +


60 1996 Phil. Phytopath. 32(1):35-A


Figure 1. Mussaenda leaves and flowers infe
slight vein clearing/curling; (B) flow
leaves, and (D) healthy flowers.

Figure 2.Test plants mechanically inoculate

d with a tobamovirus;(A) leaves showing mosai(
with streaking and color deviations; (C) health

vith extracts of symDtomatic mussaendin sec


1. Membership in the Philippine Phytopath
ing in Philippine Phytopathology or at I
society. The Editorial Board, however, r
tions of exceptional merit. It may also i
articles of interest to the Society.

2. Manuscripts must be reports of original
should have not been published elsewhel
accept or reject the manuscript is final.

3. The manuscript should be typed on one

4. The author's name should follow the tit

S. Papers other than Notes may be organize
tion, Materials and Methods, Results, Di
Literature Cited.

6. In the text, citations should be by name-a
With 3 or more authors, use et al. (e.g. O
et al. (1); the number in parenthesis shou
cited (or referred to in the text) under litel

7. Literature citation should be in alphabetic
published work; it should appear as fo<
Serials with Title Abbreviations must b
journals. Examples of abbreviation: Phi
Mol. Biol., Plant Dis. Reptr., J. Ag. Res.,

8. Acknowledgements should be placed at

9. Tables should be numbered consecutively
must have descriptive headings and shou
the text. Lower case superscript letters ai
containing tables should follow Literatui

10. Figures should add clearly to an underst
ments of figures (graphs, line, drawings
Journal page. Combine illustrations in co
each unit to correspond with the text
numerals. Label each illustration in pencil
and author's name. Legends for figures st
bered page following the tables.

11. See latest journal of Philippine Phytopat
papers to oe submitted to the journal.

12. Articles published are not paid but aut


gical Society is prerequisite to publish-
t one author must be a member of this
relax this rule in the case to contribu-
te distinguished scientists to contribute

search, except meritorious reviews, and
The decision of the Editorial Board to

of 8% x Il inch paper, double spaced

Author's position and institutional ad-

:onveniently under: Abstract, Introduc-
ssion, (or Results and Discussion) and

-number system, e.g. Ou and Nuque (1).
Nuque and Silva (1) should appear Ou,
correspond to the number of the article
are cited.

>rder and with numbers. Do not cite un-
)te. Biological Abstracts' 1968 List of
onsulted in abbreviating the names of
,. Entomol., Philipp. Phytopathol., J.
er. J. Bot.

end of the article i.e. after Literature

nd each typed on a separate page. They
be understandable without reference to
o be used for footnotes to tables. Pages
'ited and should be numbered accord-

ling of the paper. The size and arrange-
nd photographs) should correspond to
>osite cuts when possible, and number
ire reference, using consecutive Arabic
the reverse side with the figure number
Id be typed together on a separate num-

ogy for more details on the format of

s foot the bill for reprints

University of Florida Home Page
© 2004 - 2010 University of Florida George A. Smathers Libraries.
All rights reserved.

Acceptable Use, Copyright, and Disclaimer Statement
Last updated October 10, 2010 - - mvs