Fungi Genus and Concentration in the Air of Onion Fields
and Their Opportunistic Action Related to Mycotic Keratitis
CHING-WEN CHANG
Institute of Occupational Safety and Health Council of Labor Affairs
Taipei, Taiwan, Republic of China CHI-KUNG H O
Department of Occupational Medicine Kaohsiung Medical University Hospital and
Graduate Institute of Occupational Safety and Health Kaohsiung Medical University
Kaohsiung, Taiwan, Republic of China ZUEI CHING CHEN
Department of Botany Nat^ional Taiwan University Taipei, Taiwan, Republic of China YAW-HUEI HWANG
Institute of Occupational Medicine and Industrial Hygiene
National Taiwan University Taipei, Taiwan, Republic of China
CHEN-YANG CHANG SHAO-TAUN LIU
Department of Occupational Medicine Kaohsiung Medical University Hospital Kaohsiung, Taiwan, Republic of China MEI-IU CHEN
Graduate Institute of Occupational Safety and Health Kaohsiung Medical University
Kaohsiung, Taiwan, Republic of China MAO YEN CHEN
Department of Botany National Taiwan University Taipei, Taiwan, Republic of China
ABSTRACT. The authors determined fungi concentrations and genera from onion fields lo-cated in monsoon and nonmonsoon areas of Taiwan, as well as concentrations from a coun-ty housing location, which served as the comparison site. The authors explored the rela-tionship(s) among onion harvesting, monsoon characteristics, airborne fungal levels, and mycotic eye infection. Airborne fungal levels in the onion field in the nonmonsoon area were the highest, followed by decreasing levels in the monsoon area and control sites, re-gardless of the sampling periods. The geometric mean of airborne fungal concentrations was the highest during onion harvesting. Cladosporium was the most dominant fungus during pre- and postharvest seasons, whereas Aspergillus, Fusraium, Acremonium, Alternaria, Peni-cillium, Monilia, Mycelia, and Rhizopus were increased significantly during harvesting. Al-though fungal concentrations in monsoon onion fields were lower than concentrations mea-sured in the nonmonsoon area, the invasion of dangerous fungal pathogens following eye trauma by airborne sharp exotics (e.g., onion flakes) dispersed in high wind may account for the increase in corneal ulcer cases that occur in the monsoon area.
<Key words: fungi, mycotic keratitis, onion>
MYCOTIC KERATITIS, or kerafomycosis, can lead to a significant reduction in vision. Cases of mycotic ker-atitis have been reported in children,' but most cases are associated with oufdoor agriculture activities.^"'' Trauma, typically resulting from vegetation, is the com-mon predisposing factor.^-'^ Other factors, such as
ster-oid therapy for intraocular surgery and chronic ocular Inflammation, might also contribute to corneal mycotic infections.''
Most studies on mycotic keratitis report the fungi gen-era isolated from the eyes of patients.''^""^ The fungi genera include Aspergillus, Acretnoniutri, Fusarium,
At-temaria, Curvularia, Penicilliutv, Tetraploa, and Cat^di-da. However, knowledge of the ecological distribution
of fungal genera in agriculture fields associated with mycotic keratitis is very limited.''
Recently, 5 onion harvesters in the Heng-Chun Penin-sula of southern Taiwan simultaneously reported suffer-ing from fungal corneal ulcers.'^ Four of the 5 cases com-plained of ocular trauma from pieces of onion skin or plant leaves that drifted into their eyes while they were harvesting onions. The researchers used corneal scraping to clinically identify the fungi in the injured eyes of all 5 cases. Three patients received penetrating keratoplasty, whereas the remaining 2 recovered following treatment with Amphotericin B and Fluconazole. The authors indi-cated that their study was the first report of a group of fungal corneal ulcer cases associated with onion-harvest-ing activities, and they considered the unique atmos-pheric characteristic of strong monsoons, which blew during the onion harvest season, and to be a contributing factor of eye trauma. As is the case in most reports of corneal mycotic infection, fungal genera identified from the eyes of 4 of the patients included Candida,
Cephalosporiurn, and Cryptococcus. However,
informa-tion about the ecological distribuinforma-tion of fungal genera in the onion fields was not provided. To evaluate the possi-ble link between fungal corneal ulcer cases and occupa-tional environments and atmospheric characteristics, we conducted the present study. We sought to determine the variations and concentrations of airborne fungal genera that existed during a period of onion growth in 2 separate geographic areas characterized by differing atmospheric characteristics of monsoons. The association between fungi occurrence and potential ocular infection of onion harvesters is also discussed.
Materials and Method
Environments. The major cultivated onion farmlands
in Taiwan are located in 4 southern-end townships: (1) Lin-Yuan, (2) Heng-Chun, (3) Shen-Chen, and (4) Fong-Kung. Lin-Yuan is a suburban township located approx-imately 70 km south of metropolitan Kaohsiung city. There are no strong monsoons in the Lin-Yuan area. Heng-Chun, Shen-Chen, and Fong-Kung are typical rural townships, surrounded by mountains and sea, in the Heng-Chun Peninsula. Downslope wind storms occur from October to April southeasterly across the central mountain chain of Taiwan. According to the local meteorological records of the Heng-Chun area, the monthly average wind velocity during the periods January-April 1988 and October-December 1998 was, for example, 2.8-5.9 m/sec (6.3-1 3.2 mph). The maxi-mum wind velocity (10-min average) is 10.1-16.2 m/sec (22.6-36.2 mph}, and the instant maximum wind velocity can reach 20.6-32.4 m/sec (46.1-72.5 mph).'^ The period for onion growth and harvest is usually from November to April (i.e., within the monsoon season).
Air sampling strategy and procedures. The authors randomly selected 1 onion field from each of the 4 townships to be the exposed sites, which represented a nonmonsoon field (Lin-Yuan) and 3 monsoon fields
(Heng-Chun, Shen-Chen, and Fong-Kung). All selected fields were distant from traffic roads and the residential community, and the fields were surrounded by other onion fields. An outdoor parking lot at a residence lo-cated in the rural township of Fung-Leu served as the comparison site.
Fungal samples were collected from the air of the 4 selected fields and from the comparison site—before, during, and after onion harvesting. Air samples were also taken simultaneously at harvest from 3 onion-gath-ering sites, respectively, located (a) in the middle of an onion field; (b) in an outdoor, barren area that was not adjacent to the aforementioned onion field; and (c) in an indoor gathering-distribution house.
For every sampling period, 30 airborne fungi samples were collected at 80 cm above the ground at the center and 4 corners of the selected onion fields. Measurement was achieved with a single-stage Andersen Microbia! Sampler (AMS, Graseby ISmyrna, Georgial), which op-erated at 28.3 l/min. Six samples were collected simul-taneously at each sampling site (i.e., 6 AMSs at 3 differ-ent sampling times). The sampling times for the preharvest survey were set at 5 min, 10 min, and 15 min, whereas sampling times for harvest and posthar-vest measurements were set at 1 min, 2 min, and 5 min. The AMSs were preautoclaved in the laboratory and disinfected with 70% ethanol-immersed cotton balls between each sampling. Fungi were collected directly onto plates containing 20 ml of Sabouraud dextrose agar with 40 |j.g/ml chloramphenicol (SDAC, Difco Lab-oratory [Detroit, Michigan|). Calibration of airflow was performed before and after sampling occurred. Two field blank samples were also taken from the field. Sam-pling at the parking lot was conducted in a manner that was similar to the procedures used at the other sites.
The sampling plates were sealed, sent back to the laboratory by express mail, and incubated at 25 °C for 4-7 days. Colonies that formed on the plates were then counted. The positive-hole method was applied to the AMS samples for correction of microbial coincidence.'"* Fungal colonies on plates were examined under a dis-secting microscope (modei SZ-PT, Olympus [Tokyo, japan]). The spores of each colony were then picked and examined under a high-resolution microscope (model BH-2, Olympus [Tokyo, japanj) for identifica-tion of genus in accordance with available refer-ences'^-'^ A total of 279 air samples were used for ad-ditional data analysis.
Environmental factors. Temperature, wind velocity,
and relative humidity were recorded with direct-reading instruments (Testo, GmbH&Co. [Lenzkirch, Germany]) on each sampling day. At least 3 readings were record-ed at each sampling site.
Statistical analysis. Airborne concentrations of
cul-turable fungi were determined by dividing the correct-ed colony number by the volume of samplcorrect-ed air. The values were then calculated and expressed as geomet-ric means with standard deviations. The percentage of an individual fungal genus was calculated as 100 times a ratio of concentration of an individual genus to the culturable fungi in all of the samples.
Results
Table 1 illustrates the concentrations of atmospheric fungi from the onion farmlands and parking lot before, during, and after onion harvesting. Regardless of the sampling periods, the geometric mean concentrations of airborne fungi at the nonmonsoon onion farmland were always the highest, followed by concentrations determined from the farmlands in the monsoon areas and the parking lot. We also observed that, in all of the farmlands measured, the mean fungi concentration was increased significantly when onions were harvested. The increased levels, shown as the harvest/before har-vest (H/BH) values in Table 1, were 2.7 for nonmon-soon farmland and 3.7-6.8 for monnonmon-soon sites. Once the onions were harvested, all airborne fungi levels were decreased—but to different degrees, compared with
levels present before harvesting.
Significantly high fungal levels were identified at onion-gathering sites (Table 1). A mean of 9,063.9 colony-forming units {CFU)/m^ existed at an outdoor onion-gathering ground in Heng-Chun. This value is 6.6 times the average of airborne fungi (1,363.6 CFU/m^) at the onion farm in the same township, mea-sured during the same sampling day. In Heng-Chun, an elevated fungal level of 6,999.1 CFU/m^ was also found in the air of an indoor house designated for onion gath-ering and distribution. Given the unexpected abun-dance, most of samples collected from both sites could not be enumerated and were, therefore, discarded prior to data analysis. On the other hand, a temporary gath-ering site at the center of an onion farm in Fong-Kung did not have an elevated fungal level; rather, its fungal concentration of 427.9 CFU/m^ was lower than the mean of that onion farmland (922.6 CFU/m^^).
The genus-specified concentrations of airborne fungi are presented in Table 2. With respect to the
nonmon-soon onion farmland, Cladosporium was the most pre-dominant genus before, during, and after harvesting, accounting for 92%, 59%, and 85%, respectively. In the period of onion harvesting, mean concentrations of
As-pergillus, Mycelia, and Fusarium were increased 30.9-,
15.2-, and 22-fold, respectively, compared with prehar-vesting concentrations, and they accounted for 24%, 8%, and 6%, respectively, of airborne fungi flora. Air-borne Acremonium concentration in the harvesting pe-riod was also 10.2 times the level determined at the time of preharvesting, but it was nonetheless present in a relatively low concentration of 3.9 CFU/m\
In monsoon area onion farmlands, Cladosporium was dominant in pre- and postharvesting periods, ac-counting for 55-56% of the airborne fungi; 22-29% of airborne fungi were identified as Aspergillus. During onion harvesting, Aspergillus was greatly aerosolized, resulting in an increase of 7.6 times the concentration at preharvesting, and it represented 38% ofthe airborne fungi, which was even higher than the percentage of
Cladosporium (30%). Other frequently identified fungi
included Fusarium (11%), Mycelia (8%), and
Penicilli-um (8%), with concentrations increased by 5-, 7-, and
5.7-fold, respectively. Airborne Acrernotiium and
Rhi-zopus concentrations were also increased by 14.4-fold,
respectively, during onion harvesting, but both levels were comparatively low. Almost all of the observed fungi were reduced or unidentified in the air of farm-lands after onions were cropped, except for
Trichoder-tria, Curvalaria, and Candida. The latter 2 were found
oniy in postharvesting samples—at rather low concen-trations.
One of the significant findings in the parking lot dur-ing the harvest period was the identification of abun-dant airborne Candida (35%) and other yeasts (24%). In addition, in contrast to the results of onion fields,
Fusar-ium, Mycelia, and Acremonium were all slightly
dimin-Table 1.—Concentration of Airborne Fungi Measurec During, and After Onion Harvesting
Measurement site/town Onion farmland:
Nonmonsoon area: Lin-Yuan Monsoon areat
Heng-Chun Shen-Chen Fong-Kung Onion-gathering site:
Onion farmland (Fong-Kung) Outdoor ground (Heng-Chun) Indoor house (Heng-Chun) Parking lot (Fung-Leu)
Notes: CM = geometric mean, and 'Colony-forming units/m^ (CFU/m^
at 4 Before harvesting (BH) CM* CSD 913.5 2.3 201.4 2.7 366.8 3.1 188.2 2.1 135.1 2.4 — — _ — — — CSD = geometric •fThe results presented here are a combination ofthe areas. n 19 55 16 18 21 — — — 6
Onion Farmlands, 3 Onion Accumulation Sites, and
Harvesting CM* 2,485.2 932.3 1,363.6 767.3 922.6 427.9 9,063.9 6,999.1 899.5 standard deviation. data collected from
(H) CSD 1.7 2.4 2.5 2.4 2.0 2.0 1.4 1.3 1.7 the onion n 13 80 21 40 19 12 2 2 6 After harvesting CM* 887.1 557.1 594.0 602.6 483.0 — — — 408.0 (AH) CSD 2.0 1.9 2.1 1.6 2.0 — — — 1.3 a Parking Lot—Before, n 20 60 20 20 20 — — — 4 Ratio of mean concentration H/BH 2.7 4.6 3.7 4.1 6.8 — — — 10.2
farmlands in Heng-Chun, Shen-Chen, and
A H/BH 1.0 2.8 1.6 3.2 3.6 — — — 4.6 -ong-Kung
"5 ing [ s an d a 1 lan d E u . c ' c 0 o lin g a. n <UJO . • • -c (rati o c fi e Con e u JS-S p c • o o n a n "J3 fic a BC C Tabl e 2 . Jo 0 0 ki n ro a. ro 0) ro c monso o E ro c .9 *E 0 are a c o c o E c o c and . E -2 f— Onio i % on . ltr a h _ c 0 u a) <D "E ion , CF L ltra t a; c O u c
s
"E on , CF U ro icent t i_ o ro QJ I H/ l , , II X II X 3 II -S - rs
X ca X (0 9 II X (0 9 II c X LO I O II X CQ X H/ B o II X II X II X Fungu s CTi d o o d d r^ d Acremo j LO d rv - o LO " ^ rN Alternar " ^ r o d LOg
r o -—• CO LO ro r v 9% ) CN t v 190 . CO CO f N CO gS ^^ r o UO T f <M CN r v CO CM ro I N LO •O CM 355 . (% 0 ro ro r o f N (N O^ 30 . ) 63 . • - V CN CO CO LO lus Aspergil CO in ro t v 85% ) ) 893 . LO CO CM o CO r o CO on1
a-: r o d CO d .12 Curvalai CO d LO ^—. 7%) 1 — "^—• CO q Lri CM 10 . c r o ro o 22 . r o 37 . LO r o 1 ^ Fusariun CO d q q r o CO r o q r v r o ?^ 2-CM r v a i • ^ r v CC • * r v LO . D 0 o> 1— r.1 Monilia Mycelia t M rv q f N r v _ ! LO 8% ) 1— — • LO o i r-r v LJ-i Tt" CO CTi rN 0 ,9) 6 r^ t v d 6% ) 0 60 . 2! f N ro d !N E Penicilli r o d rN d -5 r^ Rhizocti CN ( N d Tf-• ~ •.0 d U l Rhizopu 1 1 1 1 ro S .—- ^^^ IN ^ IN -rf ro fN ro CM f N ro tM ro ro -Ti-ro d CM ro LO r o Tricbodt Yeas t Candi Othe r cC JZ g 'c 0 CJ = af t It X ''-a; £ ( 0 io n 1 c 0 CO c '5 II X CC .E > ro c CO cu har v or e oni o n befor e be f II X cn cib 0 0 c 0 U-T5c _ • • QJ .r: ihe n ,-Chu i c ati o a; u 0 QJ - C T; OJ -n IVI i T J C £ c: tn QJ ro JZ c oni o CO c area : H e ine d dur i c 0 lOSU O 5 Jan . .ing-Y i —' rea : ro 0 0 lOUJ U 0 7 Notes: 1 LU J det e c 0 trat i once n u CC c .9 ra t 11 X COi
i(pe r 111. , un i CO E 0 lony -0 u II J QJ 1 parent h L: IE Lr wi t ^u Q) CL CL ro t fung i c ro c E 0 • a CL es o f CC S c u D.ished during the harvest season. Furthermore, the levei of airborne Aspergillus (50.3 CFU/m^) was only approx-imately 1/10 of levels determined at onion farmlands (473.1 CFU/m^ and 598.8 CFU/m'), and it was not in-creased appreciably during the harvest season (H/BH value = 1.4) as shown in the onion farmlands (H/BH values = 30.9 and 7.6). Aspergillus, however, account-ed for 33% of airborne fungi for both pre- and posthar-vesting seasons. There was a high prevalence of
Cla-dosporiurn (27-45%) present in the air, regardless of the
sampling periods.
The respective averages of atmospheric wind veloci-ty measured before, during, and after onion harvesting were 1.15 m/sec, 0.95 m/sec, and 0.74 m/sec, respec-tively (2.57 mph, 2.13 mph, and 1.66 mph, respective-ly) for the nonmonsoon field; 7.98 m/sec, 2.88 m/sec, and 1.85 m/sec, respectively (17.85 mph, 6.44 mph, and 4.14 mph, respectively) for the monsoon farm-lands; and 0.31 m/sec, 0.68 m/sec, and 0.53 m/sec, spectively (0.69 mph, 1.52 mph, and 1.19 mph, re-spectively) for the parking lot. Mean temperatures of these 3 locations during 3 concomitant sampling occa-sions were 21.6 -C, 24.9 -C, and 32.2 X , respectively (nonmonsoon field); 26.9 ^C, 32.2 ^C, and 31.3 °C, re-spectively (monsoon farms); and 29.9 -C, 32.9 -C, and 28.8 °C, respectively (parking lot). The corresponding relative humidity was averaged at 62.6%, 72.9%, and 71.4%, respectively; 54.3%, 66%, and 68%, respec-tively; and 62.2%, 46.3%, and 71.5%, respectively. Monsoon onion fields tended to experience higher wind speed and atmospheric temperatures—but lower relative humidity—than nonmonsoon farmland. Discussion
More than 70 species of fungi belonging to 40 genera are reportedly associated with the occurrence of human keratitis, among which Fusarium, Aspergillus,
Curvular-ia, and yeast-like fungi have been identified frequently.^
Although many fungi are possible pathogens, the fre-quent Isolates of fungi from eyes of individuals with ker-atitis vary geographically. For example, Fusarium is the most commonly isolated genus in south Florida,^'^ whereas Aspergillus is prevalent in India and Thai-lgp(j_i,18,19 Candida has been isolated frequently in Britain,•'^ but it has not been commonly reported as being the cause of keratitis in tropical countries.^ Per-haps this difference results from the climate characteris-tics of various regions. The climate of the present study areas was similar to that in south Florida, India, and Thailand. We observed that, during onion harvesting, the concentrations of Apsergillus and Fusarium were no-tably increased and prevalent in the air. Inasmuch as
Fursarium and Aspergillus may cause moderate to
se-vere outcomes following corneal infection,^ the preva-lence of relatively high levels of these fungi in the air im-poses a potential threat to onion harvesters. Increases in airborne concentrations of other keratitis-related agents, such as Acremonium, Alternaria, and Penicillium, should also be considered as potential hazards.
In addition to the presence of dangerous fungi in
onion farmlands during harvest, trauma of eyes caused by exotic bodies, such as onion flakes or dusts, is more likely to occur in monsoon areas with gusty winds. In an animal study, Hwang et al.^^ demonstrated that onion outer scales may significantly damage the cornea of rats in environments in which the wind velocity ex-ceeds 10 m/sec; such conditions occur often in mon-soon regions during onion planting and harvesting. It is speculated that notably increased levels of dangerous fungi with a likely occurrence of eye trauma during onion-harvest operations would contribute to the inci-dence of keratitis among onion harvesters in a monsoon area. This implication is supported, in part, by our clin-ical findings of fungi in the injured corneas of the clus-ter cases, and by the cases' self-reported statement: "Eye trauma by onion skins or plant leaves during onion har-vesting operation."^-' \sic] In addition to the environ-mental characteristics, host factors, including wearing of contact lens, use of mascara, use of certain medica-tions, use of antibiotics and/or corticosteroids, insuffi-ciency of tear secretion, presence of ocular diseases and/or systematic disorders such as diabetes, might also contribute to the development of mycotic keratitis.^
To protect onion harvesters from suffering ocular trauma and mycotic keratitis, workers should be well educated about the contributing factors of such disor-ders, as well as the protective measures they should take. It is strongly suggested that onion harvesters wear protective spectacles that afford full enclosure of their eyes. If exotic materials enter the eyes, workers should go at once to an ophthalmologist for proper diagnosis and treatment, instead of seeking medication from local pharmacies. The workers should learn that delay of proper medical care might result in blindness. Also, lo-cal ophthalmologists should be informed about this en-demic ocular disorder and prepare for suitable treat-ment. With appropriate education of both workers and ophthalmologists, use of eye protection in fields, per-formance of personal hygiene, and provision of medical care, it is expected that the occurrence of mycotic ker-atitis in onion harvesters will be significantly reduced.
The authors gratefully acknowledge the farmers and the farmer co-operatives in Lin-Yuan, Heng-Chun, Shen-Chen, and Fun-Shen areas for their participation and assistance.
This study was funded by the Institute of Occupational Safety and Health, Council of Labor Affairs, Executive Yuan of Taiwan, Republic of China, IOSH88-H307.
Submitted for publication October 20, 2000; revised; accepted for publication June 29, 2001.
Requests for reprints should be sent to Dr. Ching-Wen Chang, In-stitute of Occupational Safety and Health, Council of Labor Affairs, Executive Yuan. #99, Lane 407, Hengke Road, Shijr City, Taipei, 221, Taiwan, Republic of China.
References
1. Panda A, Sharma N, Das F, et al. Mycotic keratitis in children: epidemiologic and microbiologic evaluation. Cornea 1997; 16 (3):295-99.
2. Thomas PA. Mycotic keratitis—an underestimated mycosi!^. J 12. Med Vet Mycol 1994; 32:235-56.
3. Clinch TE, Robinson M|, Brron BA, et al. Fungal keratitis from
nylon line lawn trimmers. Am I Ophthalmol 1992: 114:437-40. 13. 4. Ishibashi Y, Hommura S, Matsumoto Y. Direct examination vs.
culture of biopsy specimens for the diagnosis of keratomycosis. 14. Am I Ophthaimoi 1987; 103:636-40.
5. Liesegang T|, Forster RK. Spectrum of microbial keratitis in south
Florida. Am I Ophthalmol 1980; 9 0 : 3 8 ^ 7 . 15. 6. Polack FM, Kaufman HE, Newmark E. Keratomytosis: medical
and surgical treatment. Arch Ophthal 1971; 85:410-16. 16. 7. Winchester K, Mathers WD, Sutphin IE. Diagnosis of aspergillus
keratitis in vivio with confocal microscopy. Cornea 1997; 16: 17. 27-31.
8. Fincher RE, Fisher IF, Lovell RD, et al. Infection due to the fungus 18. Acremoniutii {cephalosporium). Medicine 1991; 70:398-409.
9. Kinnas IS. Ophthalmic disease caused by a mycete of the giant 19. cane. BrJ Ophthal 1965; 49:327-29.
10. Byers |L, FHoiland MC, Allen IFH. Cephalofiporium keratitis. Am J 20. Ophthalmol 1960; 49:267-69.
11. Cuero RC. Ecological distribution of Fusarium so/an/and its op- 21. portunistic action related to mycotic keratitis in Cali, Colombia.
J Clin Microbiol 1980; 12:455-61.
Ein SH, Lin CP, Wang FHZ, et al. Fungal corneal ulcers of onion harvesters in southern Taiwan. Occup Environ Med 1999; 56: 423-25.
National Meteorological Station. Daily Meteorological Records in Heng-Chun Area in 1998, A-Table. 1999; (in Chinese). Andersen AA. New sampler for the collection, sizing, and enu-meration of viable airborne particles. | Bacterioi 1958; 76: 471-84.
Barnett FHE, Hunter BB. Illustrated Cenera of Imperfect Fungi. New York: Macmillan, 1987.
Larone DH. Medically Important Fungi: A Guide to Identifica-tion. New York: Elsevier Science, 1987.
Rosa RH, Miller D, Alfonso EC. The changing spectrum of fungal keratitis In south Florida. Ophthai 1994; 101:1005-13. Chander ), Sharma A. Prevalence of fungal cornea! ulcers in northern India. Infect 1994; 22:207-09.
Imwidthaya P. Mycotic keratitis in Thailand. J Med Vet Mycoi 1995; 33:81-82.
Ainley R, Smith B. Fungal flora of the conjuncttval sac in healthy and diseased eyes. Brit J Ophthal 1965; 49:505-15.
Hwang YH, Chou Ej, Chang CW, et al. Suspended matters in onion farms and their potential effects on corneal injury of the harvesters. Arch Environ Health 2002; 57(1):78-84.
