MORPHOLOGY OF THE VIRUS OF AVIAN ERYTHRO- MYELOBLASTIC LEUCOSIS AND A COMPARISON WITH THE AGENT OF NEWCASTLE DISEASE' D. G. SHARP, EDWARD A. ECKERT,2 DOROTHY BEARD, AND J. W. BEARD Department of Surgery, Duke University School of Medicine, Durham, North Carolina Received for publication July 26, 1951 In a recent report (Beard et al., 1950) there were described briefly the isolation and some of the characters of a particulate component from the plasma of chickens affected with erythro-myeloblastic leucosis. Continued study of the material has resulted in slowly accumulating evidence suggesting the identity of the component with the viral etiologic agent of the disease. Electron micrographs of the component isolated by ultracentrifugal procedures revealed a population of particles apparently of considerable homogeneity with respect to kind but of variable size and shape. In the preliminary studies, it was evident that the morphologic character of the particles was greatly influenced by various factors, certain of which were encountered in the technical procedures for preparing the material for electron micrography. Consequently, an investigation has been made of the component under various conditions with the object of learning, ultimately, the character of the virus as it exists naturally in the circulating blood plasma. The results of this work and a comparison of the findings with leucosis virus with those of the virus of Newcastle disease, which it resembles in morphology, are described in the present paper. MATERIALS AND METHODS The BAI strain A of erythro-myeloblastic leucosis, originally isolated by Hall, Bean, and Pollard (1941) and obtained from Dr. E. P. Johnson, was used as in the work previously reported (Beard et al., 1950). The character of the disease has been described by Hall et al. (1941) and by Johnson (1941). Plasma containing virus was obtained from chickens injected at 3 days of age with filtered plasma from previous passage birds. Blood was drawn by heart puncture, usually about 40 days after inoculation and 10 to 15 days after onset of the disease, into heparin ("liquaemin," Roche-Organon, 1 ml = 1 mg) in the ratio of about 9.0 ml blood in 1.0 ml heparin. Cells were separated from the plasma by centrifugation at 2,000 X g for 15 minutes; in some instances the plasma was used immediately, and in others it was frozen and stored for a time. Filtered plasma was obtained by passing the centrifuged plasma under 4 lb 1 This work was supported by research grants to Duke University from the National Cancer Institute of the National Institutes of Health, U. S. Public Health Service; from the American Cancer Society on recommendation of the Committee on Growth; by a gift to Duke University from Lederle Laboratories Division, American Cyanamid Company; and by the Dorothy Beard Research Fund. 2 Public Health Service Research Fellow of the National Cancer Institute, Federal Security Agency. 151
152 SHARP, ECKERT, BEARD, AND BEARD [VOL. 63 positive pressure through 2-inch 02 Selas filter candles (no. FP52, Selas Corporation of America, Scientific Equipment Division, Philadelphia, Pennsylvania). Purified virus was obtained from filtered plasma by alternate high- and lowspeed spinning in the ultracentrifuge as already described (Beard et al., 1950). The concentrates were resuspended in various saline solutions including Ringer's fluid of ph 7.2 to 7.5; Simms' solution (Simms and Sanders, 1942) of ph 7.2 to 7.3; or 0.07 M NH40H-NH4Cl buffer of ph 7.4. Specimens for electron micrographic examination were prepared in several ways: (a) Purified virus in one of the preceding saline solutions was placed on the collodion screen; the excess fluid was pipetted off and the remainder allowed to dry. Some of the preparations were washed quickly with water, and others were not. Such preparations were examined shadowed or unshadowed. (b) Purified virus in saline solution was placed in a drop on the surface of an agar3 plate (1.5 to 4.0 per cent agar in distilled water) and allowed to dry at room temperature. In some instances, the virus was "fixed" with osmic acid on the agar surface by inverting over the dry area a flat-bottomed dish, 1.5 cm I.D. by 1.5 cm deep, carrying several drops of 2 per cent osmic acid on its inner surface. The process required about 20 minutes for the best results. A block of the agar about 1.5 cm square, with the virus area in the center, was cut out of the plate and placed on a microscope slide. The surface of the block was covered with a 0.5 per cent solution of collodion in amyl acetate. With the slide tilted on edge, the excess collodion was drained away. When the film was dry, it was floated off onto the surface of distilled water or Ringer's solution, mounted on the wire screen, and examined in the microscope directly or after being shadowed with chromium. The surface of the collodion that was in contact with the agar was the one shadow-cast. (c) Purified virus suspended in Ringer's solution was deposited on the agar surface, dried, and fixed with osmic acid. The block of agar was cut out, and the area was heavily layered over with collodion in 2.5 per cent concentration. The collodion was allowed to dry for 15 to 20 minutes without draining, and the resulting thick, stiff membrane was pulled off the agar with forceps. The side of the membrane which had been applied to the agar surface was shadowed with chromium at an angle of 100 to 15 in the usual way. The preparation was then rotated in the chamber to receive at 90, from a second helix, a coating of silicon monoxide (Hall, 1950). In order to eliminate the collodion, which was too thick for electron micrography, the shadowed membrane was now floated, collodion side down, on the surface of amyl acetate and left for 90 minutes. After this interval, the remains of the preparation, held together by the silicon monoxide film, were still visible but very fragile. A specimen screen held with forceps was lifted up through the amyl acetate beneath the preparation, a portion of which was brought up on the screen. This was allowed to dry and examined in the microscope. It will be noted 3 The technique employing agar in this way was developed after conversation with Dr. C. E. van Rooyen in whose laboratory it had been used previously.
19521 AVIAN ERYTHRO-MYELOBLASTIC LEUCOSIS 153 that no water came in contact with the preparation after it had dried on the agar surface. (d) In another procedure, virus, in purified preparations or in plasma, was sedimented in the ultracentrifuge onto an agar surface for treatment as described in (b). For this, there was employed the ultracentrifuge rotor cell constructed for counting virus particles as previously described (Sharp, 1949). A block was cut out from a layer of agar, about 2 mm thick, to fit in the bottom of the cell. The cell was then filled above with the virus preparation and spun at 16,400 X g for 30 minutes. The supernatant fluid was pipetted off and replaced 3 times with calcium-free Ringer's solution containing 1 per cent heparin. The agar block was then removed from the cell and placed on a glass slide. From this point on, the procedure was the same as that described in (b). (e) Purified virus in Ringer's solution was dried on the collodion membrane in the conventional procedure as described in (a). Some of the preparations were then examined without shadow casting or washing; the particles of others were fixed while dry with osmic acid vapor under a small dish. In some cases, then, the screen was washed with water, and electron micrographs were made either with or without shadow casting. RESULTS Electron micrographs of purified virus obtained by procedure (a) are shown in figures 1 and 2; the former is an enlargement of the micrograph previously reported (figure 1, Beard et al., 1950), and figure 2 is the micrograph of an entirely different preparation. The micrograph of the same preparation as that of figure 2, but of a different screen, is shown in greater magnification in figure 3. The virus of figure 1 was dried from NH40H NH4CI buffer solution, while that of figures 2 and 3 was dried from Ringer's solution. The characters of the particles, as indicated by the images in the micrographs, are clearly shown in figure 1 as previously described. Drying of the residual salt of the NH40H NH4Cl solution produces relatively small and dispersed masses of salt deposit in which small groups of the particles are caught. Ringer's and Simms' solutions containing principally NaCl dry to form large crystal masses containing apparently considerable amounts of the particles with relatively few of the particles dispersed between the masses as seen in figure 2. The characters of the particles are well illustrated at high magnification in figure 3. Despite the variation in form, there are clearly evident fundamental similarities in character indicated by the occurrence of intermediate forms from the tailed to the spheroidal shapes. In the tailed particles, the head pieces, about 75 to 100 m,u diameter, seem well-formed and vary from spheroidal to arrow-head shapes. The tails, of about 200 m,u length, are narrow at the departure from the head, and at the distal end, they seem widely spread and flattened in some instances. It would appear that, in contrast with the tailed bacteriophages, the structure is continuous with that of the head. The surface of the head is not smooth, but folded or corrugated. The spheroidal particles without tails closely resemble in contour the heads of the tailed particles. Betwveen the spheroidal and
154 SHARP, ECKERT, BEARD, AND BEARD [VOL. 63 the tailed forms there are particles of many shapes and sizes, some with one end swollen and others of nearly filamentous character, with relatively small heads and a thick tail. As noted before, these forms of the leucosis virus are very much like those of the virus of Newcastle disease (Cunha et al., 1947; Bang, 1948; Elford et al., 1948; Reagan et al., 1951). The question of the significance of the tail structures of the Newcastle disease virus has remained obscure (Bang, 1948; Elford et al., 1948), though Cunha et al. (1947) and Reagan et al. (1951) have regarded the tailed Downloaded from http://jb.asm.org/ Figure 1. A concentrate of purified leucosis virus obtained by 4 ultiacentrifugal cycles in NH40H-NH4CI buffer solution at a concentration of about 100 times with respect to the volume of the plasma employed. X 12,250. Figure 2. The concentrate of another preparation of leucosis virus (second ultracenitrifugal cycle) suspended in Ringer's solution. X 12,250. Figuire S. The same material as that of figure 2 at higher maginification. X 22,520. particles as the typical form of this agent. Bang (1948) was unable to demonstrate tailed forms of Newcastle disease virus in chorio-allantoic fluid or in media of low salt content. The latter finding was reported also by Cunha et al. (1947) and by Elford et al. (1948). Because of this recognized influence of the conditions of examination, attempts were made to devise procedures for the elimination of factors which might be expected to change the form of the leucosis virus existing under natural conditions to that viewed on the screen of the electron microscope. A possibly potent factor in this eategory is the drying of virus from saline suspension on a collodion screen. As drying progresses, the salt concentration in- on March 11, 2019 by guest
1952] AVIAN ERYTHRO-MYELOBLASTIC LEUCOSIS 155 creases to saturation, and amorphous or crystalline deposits develop. Thus, in the last stages of drying, the particles must become impregnated with salt. This sequence might well be expected to result in extensive alteration in the shape of the virus particle. Such complications are not a problem with many viruses which can be studied in distilled water, particularly some responsible for plant disease, or with those animal viruses which seem not to be affected by the drying salt and which can be washed with water for removal of the salt from the screen. The particles from leucosis plasma, however, seem to swell and to become amorphous and of very low electron absorbing power when dried from water or after treatment with water on the screen. In order to avoid these difficulties, resort was had to the use of the agar surface as described in (b). Under these conditions, fluid and salt diffused together into the agar, leaving the partially dried virus behind on the surface, without having been exposed to high concentrations of salt. The characters of the virus prepared by this technique are shown in figure 4. Here it is seen that the particles, which adhered to the collodion floated from the agar surface, are essentially spheroidal and relatively uniform in shape and size, about 120 m,u in diameter. It will be noted at once, however, that the particles of figure 4 are definitely larger than the head structure of the tailed particles of figure 2 (the small difference between the magnification of figure 2 and that of figure 4 does not account for this, and both preparations are shadowed). Now and then, suggestions of tails were seen, but the symmetrical tailed particles observed in figure 3 were never in evidence. Micrographs of the same material without treatment with osmic acid showed essentially masses which appeared to be remains of these spheroidal particles. When the collodion film was floated off on the surface of Ringer's solution, salt crystals appeared on the dry film. In the preceding technique, water or saline solution was necessary for floating the collodion from the agar, and procedure (c) was employed to circumvent these complications. In order to obtain electron micrographs of the preparation made in this way, it was essential to dissolve the thick collodion film, and the silicon monoxide was employed to form the supporting structure in the absence of the film. A micrograph of such a preparation of virus fixed with osmic acid is shown, in figure 5, in which there are seen only spheroidal particles essentially identical in shape and size with those seen in figure 4 and with no evidence of tails. Elimination of the possibly adverse influence of repeated sedimentation of the virus in the ultracentrifuge and the consequent prolonged exposure of virus to saline solution were accomplished with the procedure of (d). Here the virus could be deposited directly from the plasma and examined with only brief contact with the saline solution employed to wash the virus on the agar surface after sedimentation. In figure 6 there are seen the particles derived in this way from plasma diluted 100 times with Ringer's solution. While there is no evidence of the characteristic tailed forms, some of the particles do show the suggestion of a tail. Calculation indicated a count of about 10106 particles per ml of this specimen of plasma. (Particles in approximately this order of concentration have been observed in the plasma of many of the individual birds as will be described in another report.) An occasional plasma showed, in addition to the spheroidal
156 SHARP, ECKERT, BEARD, AND BEARD [VOL. 63 particles, a wealth of "peanut" forms, illustrated in figure 7, which give the appearance of short rods in the process of segmentation and division. Forms somewhat like these, "dumbbell" shapes, have been seen in preparations of influenza virus (Sharp et al., 1944, figure 4) for which there is very suggestive evidence of segmentation of filaments (Dawson and Elford, 1949; Murphy, Karzon, and Bang, 1950) to form spheroidal particles of influenza virus. Downloaded from http://jb.asm.org/ Figure 4. Purified leucosis virus prepared on an agar surface as described in procedure (b). X 13,250. Figure 6. Purified leucosis virus prepared on an agar surface as described in procedure (c). X 11,000. Figure 6. Leucosis virus sedimented from plasma diluted 100 times on an agar surface as described in procedure (d). X 13,250. Figure 7. Particles from another specimen of leucosis plasma, diluted 50 times, sedimented on an agar surface as described in procedure (d). X 13,500. The preceding results, though revealing, were, nevertheless, unsatisfactory in the clarification of the fundamental interrelations of the various forms of the agent. It seemed clearthat osmic acid was acting to fix the virus particle in a relatively rigid state in the shape existing on the agar at the moment the reaction with osmic acid occurred. This was in contrast with the effect of osmic acid on the particles still suspended in saline solution which produced only atypically spherical bodies of very high electron absorbing power. Efforts were now made to see if particles dried from salt could be fixed so that they could then be washed with water without disruption. Such proved to be the case on on March 11, 2019 by guest
1952] AVIAN ERYTHRO-MYELOBLASTIC LEUCOSIS 157 application of procedure (e). In figure 8 there are shown, unshadowed, the tailed forms dried from Ringer's solution. The material of figure 9 was the same as that of figure 8 after fixation with osmic acid and washing with water. The particles of figure 10 were like those of figure 9, shadowed with chromium. In figure 9 there are seen essentially spheroidal particles of very high contrast due probably to impregnation with salt, surrounded by or imbedded in a clearly Downloaded from http://jb.asm.org/ Figure 8. Leucosis virus dried on collodion from Ringer's solution as seen without shadow-casting. X 12,500. Figure 9. Leucosis virus dried on collodion from Ringer's solution, fixed with osmic acid, and washed with water. The preparation was not shadow-cast. X 12,500. Figure 10. Leucosis virus treated as that of figure 9 and then shadow-cast. X 12,500. Figure 11. Newcastle disease virus purified and suspended in Ringer's solution and prepared for electron micrography essentially as described in procedure (c). X 12,500. Figure 12. Purified Newcastle virus in Ringer's solution treated with 0.5 per cent formaldehyde. The material was dried on the screen from Ringer's solution and lightly washed with water before it was shadowed. X 13,500. limited and outlined substance of low contrast. These two associated materials, of low and high contrast, in most cases appear, taken together, to constitute single entities. Actually, the low contrast material is frequently so related to the other as to simulate broad, blunt tails. The particles of figure 10 obviously are of the same size and shape as the combinations of high and associated low contrast materials of figure 9. It should be noted that the shapes of these coated particles are intermediate between the tailed forms of figure 3 and the nearly spherical particles also seen in figure 3. on March 11, 2019 by guest
158 SHARP, ECKERT, BEARD, AND BEARD [VOL. 63 For comparison of the characters of the leucosis virus with those of the agent of Newcastle disease, there are shown figures 11 and 12, which were obtained after the previous work was reported (Cunha et al., 1947). In figure 11 there is seen the similarity of the forms of Newcastle disease virus to the particles of leucosis virus in figure 3. Figure 12 illustrates the forms of purified Newcastle disease virus fixed in solution with formaldehyde of about 0.1 per cent concentration, dried from Ringer's solution, washed with water, and coated with chromium. Here there. are seen spheroidal forms, about 170 m,u in diameter, closely resembling, except in point of size, the particles of leucosis virus illustrated in figures 4 and 5. In addition, there is seen, frequently, material of low contrast surrounding or extending out eccentrically from the higher contrast, more elevated and nearly spheroidal material. DISCUSSION The present work has shown that the virus of avian leucosis may be demonstrated consistently in two morphological states, each directly related to the conditions of electron micrography. When the virus preparation is dried on the collodion screen under conditions resulting in progressive increase in concentration and crystallization of salt, there occur images of bizarre character varying from spheroidal to filamentous or well-defined sperm-like entities. When great increase in salt concentration and salt crystallization is avoided, only spheroidal particles of relatively uniform shape and size are seen. Evidence for the existence of the spheroidal particles in the plasma and identical forms in the purified preparations now seems unequivocal. Such particles can be recovered regularly as the only component, except for the "peanut" forms, of related characters from either plasma or purified preparations by means of the agar and osmic acid procedure. The factors common to the various procedures employing the agar technique are fixation with osmic acid and absence of saturation with salt. Possible effects of treatment with water are ruled out by the use of the thick collodion membrane and silicon monoxide. Though it would appear that the spheroidal particles constitute the natural form of the leucosis virus, it is, nevertheless, very difficult to dismiss the tailed and varied pleomorphic forms wholly as artifacts. This is particularly true for the symmetrical, sperm-shaped particles such as those seen in figures 1-3. Furthermore, regard for the possible reality of the existence of such particles is not lessened by experience with the Newcastle disease virus revealed as the beautifully shaped sperm forms such as those seen in figure 1 of Cunha et al. (1947) and figure 1 of Reagan et al. (1951). The principal problem with the morphology of the leucosis virus is thus reduced to the necessity for finding the relation between the two types of particles. It is possible to imagine that the virus exists in both forms together or reversibly transformable the one into the other; or it might be thought that the bizarre forms might constitute reproductive or maturation forms as recently suggested, prematurely, by Beard (1951). The present work, however, provides direct evidence of a more reasonable explanation of the phenomenon and one capable
1952] AVIAN ERYTHRO-MYELOBLASTIC LEUCOSIS 159 of explaining not only the relation of the sperm shapes to the spheroidal ones but also the occurrence of apparent pleomorphism. The train of evidence relating the two forms may be seen through examination, in sequence, of figures 2, 8, 9, 10, and 4. Typical pleomorphism attending drying from salt solution is evident in figure 2 where the particles range from the spheroidal to the sperm shape. The micrograph of figure 8 was an unusual one obtained from a screen and a region of it showing small salt deposit. Here, as expected, are principally the tailed forms showing, without shadow-casting, heads of essentially spheroidal shape. The most striking feature of the micrograph is the obvious difference in the size of these spheroidal heads and the spheroidal particles seen in figure 4. While it might be said that this difference is related to the layer of metal used to shadow-cast the particles of figure 4, comparison of figure 8, not shadowed, with figure 2 shows that the heads of the shadowed sperm shapes are comparable in size with those of the unshadowed forms of figure 8. Some of the particles of figure 2 are similar in size to those of figure 4, but most of such particles are spheroidal, or nearly so, in shape. Comparison now of figure 8 with figure 9, which also was not shadow-cast, reveals in the latter nearly spheroidal bodies of very high contrast and of the same size as the high-contrast heads of the sperm shapes of figure 8. In addition, it is evident that with these high-contrast bodies of figure 9, there is associated a well-defined material of low contrast which may be seen surrounding the highcontrast material or extending out from it. If both the high-contrast and the associated low-contrast materials are regarded together as an entity, shadowcasting would be expected to yield the forms of figure 10. In this case, the chromium coats both materials, rendering both equally impervious to electrons and thus producing shadowed forms nearly of the size of the spheroids of figure 4 and transitional in shape between the sperm forms of figure 2 and the spheroids of figure 4. The rounded or blunt shape of the low-contrast substance, which, in the present interpretation, represents the tail material of the forms of figures 3 and 8, is reasonably explained as due to removal of salt and imbibition of water and swelling of the structure. These findings suggest that the leucosis virus consists of a small inner structure of relatively high absorbing power for electrons lying within another substance of low absorbing power analogous, respectively, to the nucleus and the cytoplasm of cells. If it is conceived that the outer substance is high in water content without supporting rigid or semirigid structures, drying of the material with the attendant shrinkage would be expected to cause the formation of many different shapes. Further, more extensive variation in the shapes would occur should the inner material usually lie eccentrically placed within the watery surrounding substance for which there is evidence in figure 9. Under these conditions, drying and shrinkage of the watery substance eccentric to the inner material would be expected to result inevitably, some of the time, in sperm shapes, and, some of the time, in spheroids, but most frequently in particles of highly varied shape such as those of figures 2 and 3. The particles of figures 2 and 3 may then be regarded as the resultant of the effects of saturated and
160 SHARP, ECKERT, BEARD, AND BEARD [VOL. 63 crystallized salt on this particular physical structure of the virus. The findings with particles suspended in or washed with water are likewise clarified, for such particles might well swell and even rupture under these conditions. The evidence and views presented here, with respect to the leucosis virus, are somewhat similar to those of Elford et al. (1948) who regarded the Newcastle disease virus as essentially spheroidal and "jacketed with a readily deformable, osmotically sensitive mucinous substance." If the views of Elford et al. are interpreted correctly from their publication (1948), it would appear that these authors regarded the mucinous substance as extraneous to the virus particle. In the present work, the low-contrast, deformable substance is regarded as a part of the virus particle and bounded by a limiting membrane. Actually, the latter view would appear to be as well, if not better, illustrated in this respect for the virus of Newcastle disease by figure 4 of Elford et al. (1948) than by these micrographs of the leucosis virus. It seems likely that the particles of the viruses of Newcastle disease and erythro-myeloblastic leucosis are closely similar in physical characters though greatly different in size. In these respects, these two viruses differ greatly from other viruses causing plant, animal, or bacterial disease. SUMMARY Studies of the morphology of the purified virus of avian erythro-myeloblastic leucosis have been made under a variety of conditions of electron micrography. When preparations of the virus were allowed to dry in the presence of salt, the particles exhibited extreme pleomorphism, varying from spheroidal to sperm shapes. In contrast, the particles of the virus dried without associated salt crystallization were uniformly spheroidal, though variable in diameter, the average being about 120 m,u. Virus obtained by sedimentation for electron micrography directly from the plasma was likewise spheroidal. The spheroidal particles are regarded as the form of the virus existing under natural conditions, and evidence was advanced of the derivation of the sperm-shaped particles from the spheroidal forms. A comparison was made between the virus of leucosis and that of Newcastle disease, and the findings revealed striking similarities in the morphological aspects of the two agents. REFERENCES BANG, F. B. 1948 Studies on Newcastle disease virus. III. Characters of the virus itself with particular reference to electron microscopy. J. Exptl. Med., 88, 233-251. BEARD, J. W. 1951 Physical and chemical characteristics of viruses. Ann. Rev. Microbiol. In press. BEARD, D., ECKERT, E. A., CsLKY, T. Z., SHARP, D. G., AND BEARD, J. W. 1950 Particulate component of plasma from fowls with avian erythro-myeloblastic leucosis. Proc. Soc. Exptl. Biol. Med., 75, 533-536. CUNHA, R., WEIL, M. L., BEARD, D., TAYLOR, A. R., SHARP, D. G., AND BEARD, J. W. 1947 Purification and characters of the Newcastle disease virus (California strain). J. Immunol., 55, 69-89. DAWSON, I. M., AND ELFORD, W. J. 1949 The investigation of influenza and related viruses in the electron microscope, by a new technique. J. Gen. Microbiol., 3, 298-311.
1952] AVIAN ERYTHRO-MYELOBLASTIC LEUCOSIS 161 ELFORD, W. J., CHU, C. M., DAWSON, I. M., DUDGEON, J. A., FULTON, F., AND SMILES, J. 1948 Physical properties of the viruses of Newcastle disease, fowl plague and mumps. Brit. J. Exptl. Path., 29, 590-599. HALL, C. E. 1950 Electron microscopy of crystalline edestin. J. Biol. Chem., 185, 45-61. HALL, W. J., BEAN, C. W., AND POLLARD, M. 1941 Transmission of fowl leucosis through chick embryos and young chicks. Am. J. Vet. Research, 2, 272-279. JOHNSON, E. P. 1941 Fowl leukosis-manifestations, transmission, and etiological relationship of various forms. Va. Agr. Exptl. Sta. Tech. Bull. 76. MURPHY, J. S., KARZON, D. T., AND BANG, F. B. 1950 Studies of influenza A (PR8) infected tissue cultures by electron microscopy. Proc. Soc. Exptl. Biol. Med., 73, 596-599. REAGAN, R. L., SMITH, E. J., AND BRUECKNER, A. L. 1951 Electron micrographs of Newcastle disease virus propagated in the cave bat (Myotus lucifugus). J. Bact., 61, 37-40. SHARP, D. G. 1949 Enumeration of virus particles by electron micrography. Proc. Soc. Exptl. Biol. Med., 70, 54-59. SHARP, D. G., TAYLOR, A. R., MCLEAN, I. W., JR., BEARD, D., BEARD, J. W., FELLER, A. E., AND DINGLE, J. H. 1944 Isolation and characterization of influenza virus B (Lee strain). J. Immunol., 48, 129-153. SIMMs, H. S., AND SANDERS, M. 1942 Use of serum ultrafiltrate in tissue cultures for studying deposition of fat and for propagation of viruses. Arch. Path., 33, 619-635. Downloaded from http://jb.asm.org/ on March 11, 2019 by guest