Two new field techniques for detection and counting of Schistosoma haematobium eggs in urine samples, with an evaluation of both methods*

Bull. World Health Organ. 1974, 51, 399-408Bull. Organ. mond. Sant5 Two new field techniques for detection and counting of Schistosoma haematobium eggs in urine samples, with an evaluation of both methods* B. C. DAZO1 & J. E. BILES 2 Two new techniques for the quantitative determination of Schistosoma haematobium egg densities in urine samples, the first using sedimentation and the second filtration, have been developed and evaluated. In the first method the concentrated sediment is examined in a counting chamber under a low-power microscope and in the second method the final concentration achieved by filtration is also examined microscopically. Both procedures are easily carried out under primitive field conditions, while retaining their reproducibility and reliability. The techniques can be useful in mass surveys of populations suffering from schistosomiasis. Recent efforts to control schistosomiasis have led to a renewed interest in quantifying egg counts. The need for a precise, standardized method to determine schistosome egg output was felt by Barlow, who in 1931 introduced a method for the detection of eggs in urine samples (1). This was followed by a number of quantitative studies (2, 10-16). More recent studies by Bradley (3-7) and by Davis (9), in which miracidial counting is described, have emphasized the need for a satisfactory method of determining egg density. The techniques used in most of these studies require improvement and the equipment and mate- rials needed are difficult to handle and cumbersome to carry in the field. Alternative techniques were therefore investigated. Two quantitative methods for detecting and counting Schistosoma haematobium eggs in urine specimens have been developed and tested in Ghana, the first depending on sedimenta- tion and the second on filtration. * The studies described in this article were carried out as part of WHO interregional project IR-0052: Field investiga- tions on schistosomiasis. 1WHO Scientist, National Institute of Medical Re- search, Jakarta, Indonesia. 2 Technical Officer (Laboratory Operations), Field Inves- tigations on Schistosomiasis, Division of Malaria and Other Parasitic Diseases, World Health Organization, Geneva, Switzerland. COLLECTION AND SUBSAMPLING OF URINE SPECIMENS Urine was collected in Akokoma village, Lake Volta, Ghana, between 10 h 00 and 12 h 00 in wide- mouthed snap-top plastic containers (300 cm3), which were then labelled with adhesive plaster strips. On arrival at the field laboratory, each urine sample was agitated by one of 2 techniques to distribute the schistosome eggs uniformly through- out the specimen. Paired 10-cm3 subsamples were then withdrawn from each gross urine specimen in order to compare both the results of 2 counting techniques and the efficacy of the 2 mixing pro- cedures. The specimens were agitated for 15 seconds either by manual shaking or by stirring with a battery operated stirrer. In order to randomize sampling, on the first day of collection the samples were shaken. After the paired 10-cm3 subsamples were taken, the specimen of urine was set aside for 1 hour, then stirred mechanically before another pair of 10-cm3 subsamples were withdrawn. The proce- dure was reversed the following day; the samples were first stirred then, after an interval of 1 hour, shaken. In this way the results of the 2 methods of egg dispersal were comparable between paired sub- samples. THE SEDIMENTATION METHOD The 10-cm3 subsample is withdrawn from each individual urine specimen and allowed to sediment 3282 - 399- 400 B. C. DAZO & J. E. BILES l * > t.; ..o Si'. _' S .S;.... ~~~~~~~~~~~~~~~~~~~~~~~~~~. .. ...... .~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~. .t... .. _.. tj Fig. 1. Sedimentation method. Tdhe plasteicsyringesinareplacedin for ivrtedoa psto n inuper liquid into a discard vessel. for 1 hour in an inverted syringe to give sufficient time for the eggs to settle. The supernatant liquid is discarded and the sediment is transferred for egg density determination into a counting chamber with a capacity of exactly 1 cm3. Materials Plastic containers, 300 cm3 capacity, with wide snap- top cover, for urine specimens Plastic syringes, 10 cm3 capacity Sedgwick-Rafter chamber;a this chamber is 50 mm long x 20 mm wide x 1 mm deep and holds 1 cm3 of liquid Transparent plastic tubing, 15 cm long, to fit syringe outlet Test tube racks Plastic graduated cylinder, 100 cm3 capacity Procedure Freshly voided urine samples are collected in plastic containers. The sample is shaken manually for 15 seconds and a 10-cm3 subsample is taken at a Manufactured by A. H. Thomas, Philadelphia, PA, USA. the centre of the container and at mid-depth of the urine specimen using a plastic syringe. As shown in Fig. 1, the syringe is placed in a test tube rack in an inverted position and the subsample is allowed to sediment for 1 hour. The excess volume of the specimen collected is measured in a graduated cylin- der. The supernatant liquid is removed from each subsample into a discard vessel by attaching trans- parent plastic tubing to the outlet of the syringe and pressing gently upwards on the plunger until only the last 1 cm3 of the urine, with the sediment containing the eggs, remains. The tubing is then taken off and the syringe is removed from the rack, the plunger being pulled back a little to let in air so that a bubble forms inside. The syringe is shaken well in order to mix the sediment properly with the remaining urine, and the mixture is transferred to a Sedgwick-Rafter chamber. The cover glass is placed diagonally on the chamber and slowly set straight to avoid inclusion of air bubbles in the sample as the chamber gradually fills. The counting chamber must be transferred to the stage of the microscope with care to avoid spilling. DETECTION AND COUNTING OF S. HAEMATOBIUM EGGS The total number of eggs present in the chamber is counted under a low-power microscope objective or under a dissecting microscope. The plastic syringe is rinsed twice using as little distilled water as possible (not more than 0.25 cm3 each time), and the wash- ings are also examined for the presence of eggs, which are included in the total. Advantages 1. This sample method requires no fixing, stain- ing, or filtration. 2. The eggs in the 10-cm3 subsample can be counted directly within an hour after the urine sample is collected. Disadvantages 1. The sample has to be processed without undue delay; otherwise the eggs will start to hatch. 2. Standardization of 10-cm3 subsamples using a plastic syringe relies on the accuracy of the syringe. 3. The counting chamber sometimes overflows during transfer to the microscope stage. 4. If gross haematuria is present the red cells mask the eggs. 5. Not all the eggs are expelled and washing out the syringe is necessary. THE FILTRATION TECHNIQUE A standardized volume of urine specimen (in this case a 10-cm3 subsample) is fixed and stained. Later, filtration is carried out with a Millipore filtration system, using a low-pressure vacuum pump. The eggs trapped on the filter paper are counted under a low-power microscope objective with the aid of a graticule. Reagents needed Stock solution of 1 0 Nile blue (aqueous) Formol-saline 10%. This should be made up with acidified distilled water, pH 4-5. As the staining reaction of the eggs is essentially acid, neither buffer salts nor lysing agents can be added without detri- ment to the stain. For use, take 100 cm3 of formol-saline (10%.) and add 1.0 cm3 of 1% Nile blue sulfate. This stain/fixa- tive solution is then dispensed in 5-cm3 lots into either plastic bags or bottles using a Cornwall B.D. automatic syringe with side-arm assembly. The 10- cm3 urine sample is later added, giving a dilution factor of 1 part stain/fixative to 2 parts urine sample. The stock solution should be frequently inspected for sediment, which can be filtered off quite easily when necessary. Filtration equipment needed Steriffil aseptic filtration system-Millipore plastic apparatus, 47 mm diameter 47 mm diameter punch for cutting filter papers Whatman No. 1 filter papers cut to 47 mm diameter Vacuum pump, either hand or electric Scissors Forceps Wash bottle with distilled water Millipore petri-slides Processing of samples On arrival at the field laboratory or even while still in the village, each sample is shaken well (15 seconds) and a 10-cm3 subsample is withdrawn using a Cornwall B.D. automatic pipette with can- nula. The calibration of the pipette should be checked using a 10-cm3 volumetric pipette, after which the lock-nut must be tightly secured. The subsample is pipetted into the plastic bag or bottle containing the 5 cm3 of stain/fixative solution and mixed gently by either squeezing or inverting the container. The bag or bottle should be labelled with the sample number. The excess volume of the urine sample is then measured and the total volume collected is computed. To achieve optimum staining the urine, stain and fixative should be left for 12-24 hours, but not longer, because of oxidation of the stain. Procedure The base of the Millipore filter carrier is attached to a vacuum source through a water trap. A second inlet pipe to this base is used for regulation and release of pressure. An assistant holds the bag, or bottle, while the operator labels the back of a filter paper with a soft- lead pencil. The sample is mixed by prior shaking if in a bottle or by finger pressure if in a plastic bag. The paper is then placed on the filter carrier and wetted thoroughly with distilled water from a wash bottle. This allows the paper to stretch before the carrier top is screwed down (otherwise creasing results). Next, the bag is held corner down over the filter apparatus and a small cut with the scissors is made in an upward direction in the corner of the bag. The 401 402 B. C. DAZO & J. E. BILES blades of the scissors are then washed with the wash bottle and a further cut is made some 5-7 cm higher up the bag and the point of the wash bottle inserted to wash the inside of the bag, including the bottom corners (Fig. 2). If bottles are used the contents are poured into the filter apparatus and the bottle is washed out twice into the filter. Suction is then applied. When the liquid is nearly through the filter the sides and shoulders are washed down with distilled water. Suction is maintained until filtration is complete. The vacuum is then released completely and the top of the filter un- screwed. If the pressure is not released the edge of the paper tears. The forceps are used to pick the filter paper up and place it in a Millipore petri-slide for drying. When the paper is completely dry it is lightly sprayed with hair lacquer. Experiments have shown that this has no effect on the staining, even when heavily used, and that eggs are fixed to the filter paper and cannot move or roll, which would make counting inaccurate. Comparison of Millipore membrane and filter paper for egg filtration Two kinds of filter media were tried for the filtration technique: a Millipore filter membrane with 14 ,um pore size, and Whatman No. 1 filter paper. We found that the former filter clogs very rapidly and is not easily wetted by water. Once it does become wet the membrane tends to stretch badly and is difficult to handle. It is also expensive. On the other hand, the filter paper proved to be efficient and satisfactory for our work. The stained eggs give sufficient colour contrast to the paper to make egg counts relatively easy. Reading and counting Reading is accomplished under a low-power scan- ning objective and x 6 eye pieces of the microscope. In order to count the eggs, the filter paper is dampened with a little distilled water and a graticule is gently pressed down so as to exclude bubbles but not enough to cause any great overflow of water and possible detachment of eggs. The graticule is made by marking out the top of a mini-petri dish into 2-mm squares using a razor blade. A 47-mm diameter circle is then drawn with a pair of dividers and the whole filed down to this circle. The two diameter lines should be more heav- ily scored to divide the circle into quadrants. Advantages 1. The sample size is accurately standardized. 2. Statistically acceptable results are obtained. 3. Filters can be stored for re-checking. 4. It is possible to read filters when the sample has been unreadable on the Sedgwick-Rafter chamber because gross haematuria is present. 5. Reliable counts are obtained in comparison with the sedimentation technique using the Sedgwick-Rafter counting chamber. Disadvantages 1. The technique is time-consuming, for both filtration and reading. 2. The logistics of collection, filtration and read- ing of processed samples have to be carefully worked out and enough staff should be made available for such work in a large survey. 3. Crystallization occurs occasionally in some spe- cimens for unknown reasons, rendering some filtered samples unreadable.a ADDITIONAL FINDINGS Staining and fixation To aid detection and counting of eggs trapped on the filter paper, 7 different stains in 2 separate concentrations were tested on eggs in 10-cm3 sub- samples. As Table 1 shows, Nile blue in a 0.001 % aqueous solution gave the best visual differentiation, followed by safranin and brilliant green at the same concentration. Therefore, Nile blue stain was used throughout our study. Eggs in certain samples did not stain adequately when left overnight in the stain/fixative solution. This was detected only after filtration. Such eggs were restained by flooding the filter paper on the plastic petri-slide carriers with the stain/fixative solu- tion and allowing the paper to dry. Rate of egg sedimentation after shaking of urine sample This was studied to find how long the specimens could be left after shaking before taking subsamples. A total of 11 trials was made using urine samples of a Studies in Ghana since this report was prepared have shown that the occasional crystallization occurring with the filtration method can be eliminated by using 1 % carbol fuchsin (Ziehl-Neelsen) as the stain; the results obtained with this stain are very satisfactory (E.C. England, tinpublished data, 1974). DETECTION AND COUNTING OF S. HAEMATOBIUM EGGS A B Fig. 2. Filtration technique. A: Cutting corner of bag so that fixed, stained sample runs into filter apparatus. B: Washing out inside of bag to rinse out any eggs that might remain. 403 B. C. DAZO & J. E. BILES Table 1. Results of staining S. haematobium eggs in 1 0-cm3 urine subsamples using different stains in aqueous solutions Stain Concentration 0.001 % 0.0003% 1. eosin 2. carbol fuchsin 3. methylene blue 4. Nile blue 5. safranin 6. cotton blue 7. brilliant green eggs partially stained pink; red blood cells stained bright red filter paper and eggs overstained and showed no contrast at all filter paper background stained green; eggs stained blue filter paper background stained blue; eggs stained dark blue background orange-red; eggs stained red paper and eggs unstained filter paper background stained medium green; eggs dark green eggs very light pink; red blood cells bright red everything stained but not as deeply; eggs stained light red greyish blue background; eggs showed up blue greyish blue background; very dark blue eggs greyish-pink background; eggs stained red paper and eggs unstained, possibly due to age of stain light green background; eggs stained dark green Table 2. Results of study on rate of egg sedimentation in urine sample: number of eggs found in Sedgwick-Rafter counts on paired 1 0-cm 3 subsamples at intervals after shaking Volume of Interval (minutes) Trial original No. urine sam- ple (cm3) 0 2 4 8 16 32 45 64 75 90 1 240 ~15 15 14 8 6 3 0 01 240 5 13 12 10 7 7 5 1 0 37 24 11 4 1 1 2 02 215 29 22 9 3 2 0 0 0 08 7 4 3 2 1 0 1 3 300 ~~~10 7 6 4 4 2 1 0 19 10 8 9 6 3 1 0 0 04 265 16 16 13 7 3 2 0 0 0 0 5 295 8 9 5 3 2 1 0 0 0 012 7 3 3 4 0 0 0 0 0 6 231 27 20 17 17 10 3 1 0 0 030 17 16 14 11 1 0 0 0 0 7 277 22 10 13 7 2 0 0 0 041 30 14 15 10 5 2 0 0 0 240 ~15 17 8 9 3 5 2 0 0 08 240 18 10 12 10 7 2 2 1 0 0 9 265 6 4 3 0 1 1 1 0 0 07 6 4 2 3 0 0 0 0 0 25 21 7 10 6 3 1 010 223 19 19 9 10 6 2 2 0 1 1 261 7 4 4 2 2 2 1 011 ~~~5 3 1 2 1 1 0 0 404 DETECTION AND COUNTING OF S. HAEMATOBIUM EGGS 405 over 200 cm3 taken from different persons positive for S. haematobium. After shaking the specimen vigorously for 15 seconds, paired 10-cm3 subsamples were taken just below the meniscus at 0, 2, 4, 8, 16, 32, 45, 64, 75, and 90 minutes. The results of this study are given in Table 2. It is evident that sub- samples can be taken up to 2 minutes after agitation of the urine specimen without appreciable change in the corresponding egg counts. Likewise, the data suggest that most eggs have precipitated within 1 hour. Estimating the egg count During the initial phase of our work, all the eggs trapped on the processed filter paper were counted with the aid of a graticule. When the egg counts are high (22 and above), reading becomes time-consum- ing. The accuracy of counting eggs on half of the filter paper and multiplying the count by 2 was therefore investigated. The results, shown in Table 3, indicate that there is no significant difference be- tween the half counts and the total egg counts. Thus the distribution of the eggs on the filter paper appeared to be random. As the number of eggs on the paper increases there are, of course, larger numerical differences in the counts, but in the lower counts of 21-150 the numerical difference is negli- gible. Table 3. Results of counting eggs on half of the filter paper No. of Actual half- Total eggs Half the total count counts counted egg count 1 10/12 22 11 2 31/34 65 32.5 3 36/39 75 37.5 4 34/38 72 36 5 41/44 85 42.5 6 51/54 105 52.5 7 56/64 120 60 8 62/66 128 64 9 81/90 171 85.5 10 165/176 341 170.5 11 274/282 556 278 12 403/437 840 420 13 628/651 1279 639.5 14 986/1090 2076 1038 Table 4. Results of sensitivity tests to determine limitation of the sedimentation egg counting method using the Sedgwick-Rafter counting chamber TilVolume (CM3) No. of eggs Actual eggTrial of blank urine added to Estimated countsin10-cm3N.sample used bak egdniy subsamplessample 1 95 205 21.5/10cm3 12 13 2 87 76 8.7/10cm3 2 4 2 1 3 76 70 9.2/10cm3 2 2 4 191 77 4.0/10cm3 2 1 5 140 11 0.7/10cm3 0 1 0 1 6 49 8 1.6/10cm3 1 7 59 8 1.3/10cm3 I 1 8 69 6 0.8/10cm3 0 1 9 150 15 1/10cm3 1 1 0 1 10 98 10 1/10cm3 2 0 1 1 11 160 8 1/20cm3 0 0 0 0 12 120 8 1/15cm3 I 0 0 0 13 80 4 1/20cm3 0 0 In view of the findings shown in Table 3, the following criteria are suggested for use in counting: If 20 or fewer eggs are counted on one half of the filter paper, all the eggs on the filter paper should be counted. If 21-150 eggs are counted on one half of the filter paper, read only that half and multiply by 2. If 151 eggs or more are counted in a quadrant, read only one quarter and multiply by 4. At least 10% of all negative filter papers should be retained for checking. Sensitivity of the sedimentation andfiltration methods The 2 methods used for detection and counting of S. haematobium eggs in urine samples were com- pared and analysed. In order to determine the limitations of the 2 egg- counting methods several sensitivity tests were per- formed. In the first series of trials, known volumes of urine samples taken from individuals a who were free a Ghanaian Government staff seconded to project IR- 0658 who had been repeatedly examined and found aegatfve for S. haematobium infection. B. C. DAZO & J. E. BILES TuuuI 1600. a 0 *r- 1400. C E 'DC*12.00. .0 nU) a E 10ooo cn .0 = 900- C) m E 0 0 700 C U)0)CD 6 500- D .0 E Z 300. 100. 100 300 500 700 900 1000 1200 1400 1600 Number of eggs in 1 O-cm3 subsamples by filtration Fig. 3. Scattergram of S. haematobium egg counts in paired 1 O-cm3 subsamples of the same urine specimens examined by the sedimentation and the filtration methods. Of the 65 speci- mens sampled, the egg count was higher by the filtration method in 57 and by sedimenta- tion in 8. of S. haematobium infection were used as blanks. Known numbers of eggs were introduced into each of these blank urine samples. The samples were then shaken and two 10-cm3 subsamples were withdrawn. The eggs they contained were first counted by the sedimentation method. The results, shown in Table 4, seem to indicate that at an egg density of 1 per 10 cm3 and above, there is a good chance of recovering some of the eggs in the 10-cm3 sub- samples. At a much lower egg concentration, the chance gradually diminishes until at a density of 1 egg per 20 cm3 the chance of recovery becomes virtually nil. In a separate experiment, using an 80-cm3 blank urine sample, 4 eggs were introduced and the speci- men was agitated by shaking for 15 seconds. Two 10-cm3 subsamples were withdrawn and later pro- cessed by filtration. Our reading showed 0 in one filter paper and 1 in the other, implying that by filtration a density of 1 egg per 20 cm3 could be detected only half of the time. In a third study, the efficiency of the filter paper used to trap schistosome eggs was investigated. This was done by refiltering all the filtrate of 20 10-cm3 subsamples with counts ranging from 50 to 1000 eggs per 10-cm3 subsample. Microscopic examina- 406 0~~~~~~~~~~~~~~~~~~~ 0~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 0 * :*0 * * *g,*0 .0 0 - .00 0 1E00 DETECTION AND COUNTING OF S. HAEMATOBIUM EGGS Table 5. Comparison of the results of two methods for estimating the number of eggs of S. haematobium in urine specimens 1. Sedimentation vs. filtration method (using shaker): n = 65 pairs of 10-cm3 urine subsamples; t-test for paired samples= 2.48; P= < 0.01 Sedimentation Filtration mean count 196.7 eggs/10 cm3 358.2 eggs/10 cm3 standard deviation ± 222.4 ± 476.3 2. Filtration (shaker) vs. filtration (stirring): n = 58 pairs of 10-cm3 urine subsamples; t-test for paired samples = 0.39; P = > 0.5 Filtration (shaker) Filtration (stirring) mean count 316.6 293.9 standard deviation ± 337.4 ± 274.0 tion of the processed filter paper yielded only 4 eggs present in the combined filtrate, indicating that the filter paper was very efficient in trapping the schisto- some eggs. DISCUSSION The need has long been felt among workers on schistosomiasis for a technique to estimate egg den- sities in urine that is easily adaptable to field condi- tions but will retain the statistical reliability of more complicated procedures. In a situation where known biological variations in egg output complicate the final egg count, tech- nical variation must be minimal in order that the collection of base-line data and subsequent evalua- tion in control programmes is meaningful and sta- tistically acceptable. It was with these considerations in view that the 2 new techniques were devised. The filtration method must be considered the more reliable and repro- ducible since the opportunities for technical error are less than with the sedimentation method. In the latter, small variations in the volumes of the plastic syringe will produce variations in egg counts in addition to those resulting from the intrinsic bio- logical variations; furthermore, remnants of fluid containing eggs may be retained within the syringe after this is washed. Both techniques have been evaluated under primi- tive field conditions and the filtration technique is Table 6. Standard deviation of determination and coefficient of variation in schistosome egg counts in duplicate 10-cm 3 samples by the filtration method Egg count in first sample Standard Coefficient ofEgg coun ampledeviation of variation Range of Number differencea (100 x stan-counOgt True mean examined between two dard deviation subsamples + mean) 1-49 36.0 12 i 3.97 11.0 50-99 73.3 16 i 9.65 13.2 100-499 228.1 52 ± 39.36 17.3 500-999 718.7 20 ± 84.12 11.7 1000-1999 1412.8 9 + 126.93 9.1 2000-3999 3066.0 3 t 394.10 12.9 1-3999 444.2 112 ± 86.6 19.5 a Standard deviation of difference = 2d , where d = dif- ference between duplicate samples of the same specimen, and K - number of paired samples. currently being utilized for the collection of data in a large population in Ghana. Here it has proved to be an ideal method for the mass screening of the study population of some 4000 persons in multiple surveys, demanding a minimum of the complicated logistical support that is invariably needed when other tech- niques are used. In order to compare the 2 methods more thoroughly, both were used to examine urine speci- mens collected daily from 33 children for 1 week. Four 10-cm3 subsamples were taken from each speci- men that was large enough (a total of 260 sub- samples from 65 specimens). Two subsamples from each specimen were processed by the sedimentation method and 2 by filtration. Table 5 and Fig. 3 are based on the means of these paired subsamples; the results clearly indicate that the filtration method is more sensitive and reliable than the sedimentation method. Similar findings were obtained by Cheever & Powers (8) in comparing a filtration and a dilution technique for counting S. mansoni eggs in faeces. The consistency of the filtration method and its reliability in obtaining reproducible counts were tested by calculating the standard deviation of de- termination and coefficients of variation. The results as shown in Table 6 indicate that the filtration technique permits efficient detection and counting of large numbers of S. haematobium eggs. The method also provides a permanent filter paper preparation that can be referred to at any time. 407 408 B. C. DAZO & J. E. BILES ACKNOVVLEDGEMENTS We express our sincere thanks to Dr D. Scott, Project Manager, UNDP/WHO Project on Research on the Epidemiology and Methodology of Control of Schistosomiasis in Man-Made Lakes (IR-0658) and his staff in Ghana for their cooperation and assistance. Thanks are also due to the staff of the Health Statistical Methodology unit, World Health Organization, Geneva, Switzerland, for processing the data. We are very grateful to Dr A. Buck, Epidemiologist, Division of Malaria and Other Parasitic Diseases, World Health Organization, Geneva, for analysis and final interpretation of the information collected. RESUMt DEUX NOUVELLES TECHNIQUES UTILISABLES SUR LE TERRAIN POUR LA DETECTION ET LA NUMERATION DES CEUFS DE SCHISTOSOMA HAEMATOBIUM DANS DES ECHANTILLONS D'URINE; EVALUATION DES DEUX METHODES Deux nouvelles methodes quantitatives pour la numeration des ceufs de S. haematobium ont ete mises au point et essayees au Ghana. La premiere, tres simple et rapide, fait appel a la sedi- mentation. On preleve a l'aide d'une seringue en plastique un 6chantillon d'urine de 10 cm3 qui est laisse au repos pendant une heure, ce qui permet aux ceufs qu'il contient de se deposer. Le surnageant est ensuite elimin6 et 1 cm3 du culot est transfer6 dans une cellule de Sedgwick-Rafter pour la numeration. La seconde est basee sur la filtration. L'urine est recueillie dans un recipient en plastique. On l'agite pendant 15 secondes puis on en preleve 10 cm8. Cet 6chantillon est ensuite transfere dans un autre recipient contenant 5 cm" de solution colorante et fixatrice, agite doucement et laisse au repos pendant une nuit. Apres filtration, les ceufs deposes sur le filtre sont comptes a faible grossissement du microscope. Ces deux methodes se sont affirmees d'un emploi facile sur le terrain dans des conditions rudimentaires, tout en temoignant d'une bonne reproductibilite et fiabilite. Elles peuvent etre utilisees lors des enquetes de masse au sein de populations atteintes par la schisto- somiase. REFERENCES 1. BARLOW, C. H. A new method for examining urine for helminth eggs. Am. J. Hyg., 14: 212-217 (1931). 2. BENNIE, I. Urinary schistosomiasis. The best time to obtain specimens. The effect of specific therapy on egg output. S. Afr. med. J., 23: 97-100 (1949). 3. BRADLEY, D. J. A quantitative approach to bilharzia. E. Afr. med. J., 40: 240-249 (1963). 4. BRADLEY, D. J. A simple and rapid method for counting schistosome eggs in urine. Trans. R. Soc. Trop. Med. Hyg., 58: 291 (1964). 5. BRADLEY, D. J. The measurement of bilharziasis. Prevalence and schistosome egg output. Aims and techniques, with an account of a field method. Bull. World Health Organ., 33: 503-508 (1965). 6. BRADLEY, D. J. The measurement of schistosome populations. In: Mostofi, F. K., ed. Symposium on bilharziasis. Berlin, Springer, 1967, pp. 301-327. 7. BRADLEY, D. J. Modified apparatus for parasite filtration. Bull. World Health Organ., 38: 828-832 (1968). 8. CHEEVER, A. W. & POWERS, K. G. Counting of Schistosoma mansoni eggs in feces. Comparison of a filtration technique and a dilution technique. J. Para- sitol., 54: 632-633 (1968). 9. DAVIS, A. Comparative trials of antimonial drugs in urinary schistosomiasis. Bull. World Health Organ., 38: 197-227 (1968). 10. GERRITSEN, T. ET AL. Long-term investigation of blood loss and egg load in urinary schistosomiasis in the African Bantu. Trans. R. Soc. Trop. Med. Hyg., 47: 134-140 (1953). 11. JORDAN, P. Periodicity of ova output and intensity of S. haematobium infection. In: East African Institute for Medical Research. Annual Report, 1959-60. Nairobi, Government Printer, 1960, pp. 25-26. 12. JORDAN, P. Egg output in bilharziasis in relation to epidemiology, pathology, treatment and control. In: Mostofi, F. K., ed. Symposium on bilharziasis. Berlin, Springer, 1967, pp. 93-103. 13. ONORI, E. Observations on variations in Schistosoma haematobium egg output, and on the relationship between the average egg output of infected persons and the prevalence of infection in a community. Ann. trop. Med. Parasitol., 56: 292-296 (1962). 14. Scorr, J. A. Dilution egg counting in comparison with other methods for determining the incidence of Schistosoma mansoni. Am. J. Hyg., 25: 546-565 (1937). 15. SCOTT, J. A. Egg counts as estimates of intensity of infection with Schistosoma haematobium. Texas Rep. Biol. Med., 15: 425-430 (1957). 16. STIMMEL, C. M. & SCOTT, J. A. The regularity of egg output of Schistosoma haematobium. Texas Rep. Biol. Med., 14: 440-458 (1956).