Dispersal of Conidia of Zygophiala jamaicensis in Apple Orchards

T. B. SUTTON, Department of Plant Pathology, North Carolina State University, Raleigh 27695-7616

ABSTRACTSutton, T. B. 1990. Dispersal of conidia of Zygophiala jamaicensis in apple orchards. PlantDis. 74: 643-646.

Conidia of Zygophiala jamaicensis, the causal agent of flyspeck, were trapped in apple (Malusdomestica) orchards or near reservoir inoculum sources from late May or early June throughharvest (mid-September). Hourly spore concentrations were positively correlated with temper-ature and windspeed and negatively correlated with relative humidity and leaf wetness. Conidiaconcentration in the air was characterized by a distinct diurnal periodicity; most conidia weretrapped between 0700 and 1300 hours. Fruit infections were usually observed about 1 moafter the first conidia were trapped. Management practices for flyspeck in apple orchards needto account for the presence and abundance of reservoir hosts, environmental conditions, orchardmanagement strategies, and fungicide choice and timing.

Flyspeck (caused by Zygophialajamaicensis Mason; teleomorph,Schizothyrium pomi [Mont. & Fr.] Arx),is one of the most important diseases ofapples (Malus domestica Borkh.) in the

southeastern United States. Z. jamai-censis overwinters on apple twigs and onnumerous reservoir hosts (2,9). Asco-spores mature in late spring and initiateprimary infections on apples and reser-voir hosts. Conidia, produced from pri-mary infections, initiate secondary cyclesthroughout the growing season. Numer-ous reservoir hosts are sources of

inoculum. Rubus spp. are the most abun-dant reservoir hosts in North Carolina,however, at least 37 additional hosts havebeen identified (9).

Control of flyspeck is based on theapplication of protectant fungicidesapplied at 10- to 14-day intervals. Notall fungicides registered for flyspeckcontrol are equally effective against Z.

jamaicensis. For instance, the ethylene-bis[dithiocarbamate] (EBDC) fungicidesare much more effective than captan (1);however, captan is often preferredbecause it provides better fruit finishearly in the season and is more activeagainst white rot (caused by Botryo-sphaeria dothidea [Moug. ex Fr.] Ces.& De Not.) and black rot (caused by

Botryosphaeria obtusa [Schwein.] Shoe-maker). Consequently, growers are oftenfaced with the problem of knowing in

Paper 12478 of the Journal Series of the NorthCarolina Agricultural Research Service, Raleigh27695-7643

The use of trade names in this publication does notimply endorsement by the North Carolina Agricul-tural Research Service of the products named orcriticism of similar ones not mentioned.

Accepted for publication 26 February 1990(submitted for electronic processing).

what periods the likelihood of infectionare greatest and deciding when to initiatea program for control of flyspeck. Know-ledge of periods of inoculum availabilitywould help growers decide when to bemost concerned about the possibility ofinfection. Thus, the objective of thisstudy was to determine periods of sporeproduction and release and to investigatethe environmental variables favoringspore release.

MATERIALS AND METHODSSpore trapping studies. Airborne

conidia of Z. jamaicensis were monitoredwith Burkard volumetric spore traps(Burkard Scientific Sales, Ltd., Rick-mansworth, Hertfordshire, England) atfive locations in North Carolina: at theedge of a patch of blackberry (Rubusargutus Link) (Pope82) from 1 May to3 September 1982; at the edge of a patchof blackberry adjacent to a commercialorchard (Laughter8l) from 2 June to 29September 1981 and MHCRS82 from 15May to 15 September 1982; and in theedge of a research orchard (CCRS81)from 27 May to 5 October 1981 andNesbitt84 from 16 May to 31 August1984. Each site was located approxi-mately 10 m from blackberries and other

(9) reservoir hosts. CCRS and MHCRSrefer to the Central Crops ResearchStation in Clayton and the MountainHorticultural Crops Research Station inFletcher, respectively. Pope82 andCCRS81 are located in Wake County,MHCRS82 and Laughter8l in Hender-son County, and Nesbitt84 in HaywoodCounty.

Traps were placed on the ground withtheir orifices approximately 40 cm abovethe ground except for the trap atCCRS8 1, which was placed in the canopyof a tree about 2 m above the ground.Traps were adjusted to sample 10 L ofair per minute. Tapes from the trap werecut into 48-mm sections, mounted on a

glass slide, and spores were stained withcotton blue in lactophenol and countedat X250 by making one traverse (0.70mm field diameter) through the centerof each hourly exposure.

Temperature and relative humidity(RH) were monitored at each location

with a hygrothermograph located in a

standard weather shelter. Rainfall wasmeasured with a top-weighing, recordingrain gauge. Leaf wetness at CCRS81,MHCRS82, and Pope82 was measuredwith a DeWit wetness meter (ValleyStreams Farm, Orano, Ontario). Wind-speed was measured at MHCRS with anautomatic weather system (10) locatedapproximately 0.5 km from the orchardsite.

Measurements of disease incidenceand severity. The percentage of fruitinfected with Z. jamaicensis and thenumber of colonies on infected fruit weredetermined in unsprayed portions of theorchards at CCRS and MHCRS. At

1 -u - _ ,ulUU

v 80Q)

0 60

_ 40._

ZL 20

30

0

co

E0)

0)aC)

20F

101

0)co

aE(-

CCRS 1981

% affected

,' No.colonles -

I

4-

ElE _,- 30 -

c 20 -.0

IX 10 _I I L. .

I;

16

12

8

4

251 5 15 251 5 15 251 5 15 251 5 15May June July Aug. Sept.

Fig. 1. Percent of fruit infected with Zygo-phialajamaicensis, the number of colonies perfruit, concentration of airborne conidia, andtemperature and rainfall for CCRS81 from27 May to 15 September 1981. Spore trapdata for 30-31 August are missing.

Plant Disease/September 1990 643

0

I-

(A

._

0

0

z

© 1990 The American Phytopathological Society

u0I r-

. . .. .... . .. ... 1 .. .1, . ,,,A

I �1"Illllldllll�

CCRS in 1981, 25 fruit on each of fourtrees were chosen arbitrarily to representthe various locations within the tree,tagged, and observed weekly for disease.At MHCRS in 1982,25 fruit were chosenarbitrarily, tagged on each of four trees,and examined every 2-3 wk. The or-chards Nesbitt84 and Laughter8lreceived standard fungicide sprays.There was no orchard near Pope82.

Data analysis. The relationship amongthe hourly environmental variables andconidia catch was examined by linearcorrelation analysis with the StatisticalAnalysis System (8). Analyses wereperformed on days with complete datafor the period of 30 June to 17 Augustat Laughter81, 16 June to 15 Septemberat MHCRS82, and for the 10 days withthe greatest spore catch at Laughter8l,MHCRS82, and Pope82.

RESULTSThe first conidia of Z. jamaicensis were

trapped in late May to early June at alllocations (Figs. 1-5). There was a generalincrease in density of airborne conidiathrough the growing season at CCRS81and Nesbitt84; at Pope82 there was aslight decline through the season. Therewas no seasonal trend at Laughter8l orMHCRS82. Greatest conidial concentra-tions were found at Laughter8l wherehourly concentrations averaged almost300 conidia per m3 of air sampled on

5 15 251 5 15 25J 5 15 25June July Aug.

Fig. 2. Concentration of airborne conidia ofZygophiala jamaicensis and temperature andrainfall for Laughter8l from 2 June to 30August 1981. Spore trap data for I June and15-22 June are missing.

26 July. Conidia were generally trappedthroughout the season although numberswere often lower during periods withoutrainfall. For instance, conidial concen-trations were low in June 1981 at CCRS(Fig. 1), in late August 1981 at Laughter(Fig. 2), and in June 1984 at Nesbitt (Fig.5). However, conidial concentrationswere not greatly diminished at MHCRSin 1982 during dry weather in late Augustand early September (Fig. 3).

Peak conidial catch occurred between0700 and 1300 hours; few spore weretrapped between 1500 and 0400 hours(Fig. 6).

Correlations between meteorologicalvariables and conidial concentrationswere greatest when the natural log trans-formation of conidial concentration wasused. The hourly conidia concentrationwas positively correlated (P = 0.05) withhourly temperature and windspeed andnegatively correlated (P = 0.05) withrelative humidity and leaf wetness (Table1). There was no significant correlationwith rainfall. The correlation coefficientsincreased when data from the 10 daysof highest conidial concentration wereused in the analysis; however, the general

V

C)

a)co

.3

a)L-

aEa)

11ll1,1ii.,1111 ii II I _ II IIII

Minimum'

EE30 -

mlm0

cc 10 , 1 I I II1

- . I

24

18

12

6

relationships between the spore concen-tration and the meteorological variablesmeasured were similar to those obtainedfor the complete data set (Table 1).

Colonies of Z. jamaicensis wereusually observed on fruit approximately1 mo after the first conidia were trapped.Fruit infections at CCRS81 wereobserved first in mid-July, and incidenceincreased to about 90% by mid-August

70

, 60

., 50

E 40

0 300

CO 20

10

401

a 30

a 20

E 10

78EE 30

= 20. 1M 10

xc

0.c

0

.z

Pope 1982

- II I

Maximum

Minimum,

' ~ ~ ~ ~ ~ ~ - - - _ m - - m m r fr15 25 5 15 25 15 15 25 5 15 25 5May June July Aug. Sept.

Fig. 4. Concentration of airborne conidia ofZygophiala jamaicensis and temperature andrainfall for Pope82 from 10 May to 3 Sep-tember 1982. Spore trap data for 23-29 Mayare missing.

~c._

CO

E

0a)CL

cn

O

a)

m

C.,

a)

Ea)

E

E

cx251 5 15 251 5 15 25 1 5 15 25 1 5 15

May June July Aug. Sept.Fig. 3. Percentage of fruit infected withZygophiala jamaicensis, number of coloniesper fruit, concentration of airborne conidia,and temperature and rainfall for MHCRS82from 20 May to 15 September 1982. Sporetrap data for 28 June-13 July and 21-22 Julyare missing.

25 1 5 15 251 5 15 25 5 15 25May June July Aug.

Fig. 5. Concentration of airborne conidia ofZygophialajamaicensis and temperature andrainfall for Nesbitt84 from 20 May to 31August 1984. Spore trap data for 24-29 Mayare missing.

644 Plant Disease/Vol. 74 No. 9

.- 220L-

. 180(C)

E 140a)

O 100Q

CI arc

.-.

C)L-

C(a

a)

EC)

I-

0%._

C:

Rol _

I . .

t1 I

... , ,.. .I I , I.. Il 1, I I, I ., I., I al I I I.

1111--

I 1. I

1�

CD I -

I I

1 AL_

.1 IL I .11 11 111.11 I 11-111.

I

I'llL IH I I -

III , II.-i

II. 111111,

(Fig. 1). Number of infections increasedfrom a mean of approximately one perfruit in early August to almost six perfruit by the end of August. At MHCRS82(Fig. 3), fruit infection was observed firston 7 July, and by mid-August 100% ofthe fruit were infected. The number ofcolonies per fruit increased from a meanof approximately three in early Augustto almost 30 by mid-September.

DISCUSSIONEnvironmental variables affecting

dispersal of conidia of Z. jamaicensis inapple orchards were similar to thosereported in a banana plantation (5). Air-borne concentrations of conidia were re-duced by free water (leaf wetness) and

1.0

ca)a

L.CS%-

cv

a)a.

*

.8

.6

.4

.2

4 8 12

high relative humidity and were charac-terized by a distinct diurnal periodicity.Meredith (5) found that airborne conidiaconcentrations of Z. jamaicensis in-creased following rains; however, asimilar trend was not observed in thisstudy. Spore concentrations were lowestduring prolonged periods without rain.The exception to this occurred atMHCRS82 where spores continued tobe trapped during mid-August and earlySeptember when rainfall was light.Heavy fog is common at this locationin the mornings in the late summer, andit is probable that sufficient moisture waspresent for production of conidia.Ocamb-Basu and Sutton (6) found thatconidia were produced at relative humid-

16 20 24Hour

Fig. 6. Percentage peak frequency of airborne conidia trapped at Laughter81, CCRS81,MHCRS82, and Pope82. See text for dates conidia were trapped at each location.

Table 1. Correlation coefficients between weather variables and natural logarithm (numberof airborne on conidia + 1) of Zygophiala jamaicensis per hour for the seasona and the 10days of peak spore trap catch at Laughter8l and MHCRS82, and for the 10 days of peakcatch at Pope82

Variablesb

Orchard Temp. RH Rainfall LW WS

Laughter8lSeason 0.16 -0.22 0.02 .. ..

(N = 804) **** NS10 days 0.35 -0.32 0.05 .. ..

(N = 238) ** ** NSMHCRS82Season 0.23 -0.12 -0.01 -0.27 0.27

(N = 1,212) ** ** NS ** **10 days 0.46 -0.23 -0.05 -0.50 0.35

(N = 240) ** NS ** **

Pope8210 days -0.02 -0.16 -0.08 -0.20 ..

(N - 236) NS ** NS **

aSee text for dates.bTemp. = temperature (C), RH = relative humidity,

wetness, WS = windspeed.CNS = nonsignificant, ** = significant at P = 0.01, * =

Rainfall rainfall in mm, LW = leaf

significant at P= 0.05.

ities greater than 96%.Conidia were not trapped until late

May to early June, which is 4-6 wk afterfull bloom in North Carolina. Althoughthe exact time of primary infection hasnever been determined, pseudothecia ofS. pomi mature in North Carolina fromearly to mid-April to mid-May (T. B.Sutton and C. M. Ocamb-Basu, unpub-lished). Because colonies become visiblewithin 10-12 days under optimumconditions, infections from ascospores inlate April or early May could result incolonies capable of producing conidia inlate May. Infections of Z. jamaicensis onapple fruit have been observed as earlyas mid-May (E. M. Brown and T. B.Sutton, unpublished). Secondary infec-tions by conidia are probably much moreimportant than primary infection byascospores in disease developmentthrough the season. Ascospores are pro-duced only in a distinct period duringthe spring, but abundant conidia areproduced on lesions on reservoir hostsand apple fruit (C. M. Ocamb-Basu, un-published, [3]) through the growingseason. Consequently, control of second-ary infection should be the focus of dis-ease management programs for flyspeck.

Management strategies for flyspeckneed to take into account the presenceand abundance of reservoir hosts, en-vironmental conditions, orchard man-agement, and fungicide choice andtiming. The potential for disease occur-rence is greater in orchards surroundedby reservoir hosts. The large number ofspores trapped in proximity to reservoirhosts in this study is evidence of theirimportance. It may not always be pos-sible to remove many tree species, butRubus spp., which are the most wide-spread hosts, can be mowed in ditch-banks and orchard borders. Optimumconditions for disease development aretemperatures ranging from 16 to 24 Cand wet conditions or relative humiditiesof 96% or greater (6). Thus, the potentialfor disease development is greatest inorchards in warmer and wetter growingregions and in orchards on slow-dryingsites or in valleys in which fog and dewpersist in the morning (3). Culturalpractices such as pruning can signifi-cantly reduce flyspeck incidence andseverity in some seasons (4,7) and arean important component of a manage-ment program for flyspeck.

In this study, secondary inoculum wastrapped at most sites by 4-6 wk afterfull bloom. This is additional evidenceof the need to initiate a fungicide controlprogram for flyspeck by second or thirdcover in North Carolina. A similarsituation is likely to occur in other south-eastern apple-growing regions of theUnited States. In eastern and midwesternapple-growing regions of the UnitedStates where temperatures and moistureconditions are less favorable for diseasedevelopment, controls may not be neces-sary until mid- or late-season.

Plant Disease/September 1990 645