t Features Bacterial Variation since Pasteur* Rummaging in the attic: antiquarian ideas of transmissible heredity, 1880-1940 JOSHUA LEDERBERG I began my scientific work on neurosporas in 1942 (at the age of 17 at Columbia University in New York). Avery, Macleod, and McCarty’s 1944 report on the chemistry of the pneumococcal transforming factor—as DNA!—irrevocably oriented my scientific career to questions of the genetic structure of bacte- ria. Part of my preparation for experimental study in that field was a review of the literature: I was guided particularly by René Dubos’ The Bacterial Cell, which truly is an encyclopedic bellwether of the modern era, as well as a guide to much of the older literature. In 1946, I moved to the laboratory of E. L. Tatum at Yale University and successfully consum- mated the first experiments on crossing Escherichia coli K-12, begun at Columbia, in time for the Cold Spring Harbor Symposium in July 1946. That occa- sioned my first introduction to André Lwoff and Jacques Monod and my unbroken connection with the Pasteur Institute. So much of the early groping towards a bacterial genetics was flawed by both intellectual confusion and messy experimental reporting. Work of this kind is worse than useless for several reasons: it creates an aura of disreputability that repels younger investiga- tors, and it establishes spurious theoretical paradigms. In addition, what incentive is there to clean up messy work? If it is disconfirmed, that is a social service that tends to glean little credit for the labor invested. If by some chance it should be confirmed, it will merely exalt the reputation of the first claimant, and for Joshua Lederberg is university professor at Rock- efeller University, New York, N.Y. *Adapted from a presentation at the centennial of the Pasteur Institute, Paris, France. VOL. 58, NO. 5, 1992 limited deserts. No wonder that so many early obser- vations were simply allowed to molder. On the other hand, it is exciting to see some of the more clear-sighted early explorations, like those of Pasteur, Theobald Smith, and MacFarlane Burnet, which have given us an enduring legacy. Not that they have been, or could have been, always on the right side of every argument: we can even glide over Pasteur’s attach- ment to vitalism, when he gave us so much of the evidence needed to overthrow that philosophy! The lessons to be learned from excavations of the scientific literature, besides giving due historical trib- ute, include a series of clinical case studies in the pathology of scientific development. In addition, there have been and doubtless will continue to be more “sleepers,” forgotten ideas and experimental observa- tions that deserve reexamination today for fresh interpretation, sometimes with quite exciting conse- quences. My main themes will be: © Attenuation, exemplified by Pasteur’s attenuation of anthrax with high-temperature cultivation in 1881, and its recent corroboration as plasmid loss. © Paragglutination: exchange of “receptors” between bacteria in mixed culture and infections. © Lysogenic conversion—and some possible anticipa- tions: “filtrate effects.” @ Griffith’s discovery of transformation in pneumo- cocci, 1928. @ Unfinished business (the Weil-Felix reaction): anti- genic cross-reactions of rickettsiae with Proteus X strains. Attenuation In 1881, Pasteur developed a protective vaccine for ASM News 58: 26(-265° 261 anthrax by growing his cultures at marginal growth temperatures of ~43°C. This initiated the prevalent empirical practice of “attenuating” virulent patho- gens, be they bacteria or viruses, by cultivation in adverse environments or unconventional hosts. Be- fore the 1940s, little sophisticated analysis was de- voted to the genetic mechanisms that might be in- volved; indeed, few would have troubled to think about a genotype as distinct from the expressed phenotype. Theobald Smith apart, few of the earlier workers even thought of a culture as a population of possibly disparate cells. Pasteur devoted little tech- nique or trouble even to achieving pure culture. Koch, who was its zealous proponent, blamed most observa- tions of bacterial variation on contamination, and with F. Cohn upheld the “monomorphism” (genetic invariance) of bacteria. As genetic approaches began to emerge in bacte- riology, the only systematic study of variation of virulence factors in aging bacterial cultures by Werner Braun supported a rather banal interpreta- tion of selective advantage of “rough variants” in such cultures. Nevertheless, the idea of symbiotic particles, later called plasmids, playing a role in microbial heredity was so intriguing that in 1949 I gambled by using Pasteur’s expression “attenuation” for plasmid loss. This has lately been wonderfully vindicated: avirulent cultures of anthrax lack a crit- ical virulence plasmid. Ezzell and colleagues point out, ironically, that Pasteur’s nonchalance about pure clone cultures, so much criticized by Koch, was the key to the success of his anthrax vaccine, which was a mixture of plasmid-deprived and still toxic (and immunogenic) plasmid-positive cells. The removal of plasmids by “pasteurization” of bacterial cultures has played an equally incisive role in the analysis of plasmid function for the production of toxin crystals in Bacillus thuringiensis. Many other virulence func- tions are also attributed to plasmids in contemporary microbiology. The loss of “killer plasmids” by heating is paral- leled by the removal of kappa particles of paramecia by heating and has a close analog in the removal of chloroplasts with streptomycin. Before long, I encoun- tered the F factor as a plasmid in my own work with E. coli K-12 and was delighted eventually to find that it could be attenuated by cultivation in the presence of acriflavine but not heat. Boris Ephrussi’s success in inducing cytoplasmic “petites colonies” mutations in yeast was the immediate inspiration for that effort. Many other substances, mainly DNA-binding mole- cules, will remove plasmids from bacteria. In my 1949 review, I mentioned two other reports of attenuation that might deserve further scrutiny— they still do. Agrobacter tumefaciens cultivated on glycine loses its plant tumor-inducing competence, a trait now well known to be dependent on a large plasmid. Mucoid group C hemolytic streptococci lose their mucoid capsules when grown on acetate-free media. I am not aware of any follow-up to those reports in light of current knowledge. 262 Features Paragglutination The possibility of the transfer of traits, particularly antigenic markers, from one bacterial species to an- other is a recurrent theme in the bacteriological liter- ature between 1890 and 1930. The theoretical basis for these speculations was often the “absorption of heter- ologous receptors,” exemplified in the passive sensiti- zation of erythrocytes, bacteria, or collodion particles. For example, in 1909, Kuhn and Woithe noted that E. coli from a dysentery case reacted with Shigella flexneri Y antiserum, cross-reacting with dysentery organisms later isolated from the same patient. Spec- ulations of biological transfer led to experiments in which E. coli cultures were grown together with the shigellae and acquired a cross-reaction with the shigellae. Aware of these reports, Eugene and Elisa- beth Wollman in 1925 recorded a similar observation with E. coli and Salmonella paratyphi B. They devoted their attention thereafter to studies of lysogenicity at the Pasteur Institute. They coordinated this hypothet- ical transfer of receptors with the phenomenon of bacteriophagy and suggested the term “paraheredity.” This brilliant theoretical anticipation of lysogenic con- version was, unfortunately, not matched by further experimental inquiry. Retrospective efforts to explain paragglutination should distinguish in vitro experiments from the natural occurrence of exceptional cross-reactions, e.g., of E. coli strains with more or less complete Salmonella or Shi- gella antigens. The latter reactions certainly occur, and today we would not resist interpreting them as instances of genetic recombination. In addition, many organisms share a cosmopolitan rough antigen, giving rise to sec- ondary cross-reactions of bacteria tested as rough vari- ants. Such variants may also be selected during labora- tory culture and by encounters with antigen-specific bactericides such as phages, colicins, and complement. There has been great skepticism about the invocation of a “transfer of receptors”, viz., genetic exchange, espe- cially for want of crisp experiments with well-defined - markers. Many of the earlier experiments suffer from an uncritical faith in the specificity of immunological re- agents, as well as from a confusion of phenotype and genotype. Both errors of dogma litter the history of immunology as well as bacteriology. Nevertheless, in hindsight, some of these antique experiments plausibly were a window onto real phe- nomena, perhaps of conjugal transfer of bacterial chro- mosomes or plasmids, perhaps of lysogenic conversion. that is, the dual role of a gene in the economy of a virus and of its host cell. Lysogenic Conversion An early experiment that may have anticipated lysogenic conversion was reported by Sanfelice-—a name we associate with the etiology of cryptococco- sis—in 1893. In experiments conducted barely a year after the demonstration of the toxin, nontoxic “pseudotetanus” clostridia were inoculated in filtrates of the anaerobic sporeforming Clostridium tetani and ASM News Features Joshua Lederberg pioneered micro- bial genetics. For example, by showing that genetic information could be passed from one bacterium to another, his work was among the first manipulations of any organism’s genetic material and in large part established the importance of bacteria in genetic research. For his work on genetic structure and function in microorganisms, he shared the 1958 Nobel Prize with George W. Beadle and Edward L. Tatum. “‘Rummaging in the attic’ of the biomedical literature played an impor- tant part in the development of my . ideas on bacterial genetics,” Lederberg notes. “While working on my doctoral thesis (in microbiology), I spent a sum- mer at Woods Hole [Marine Biological Laboratory} and found the library to be a remarkable resource. A Fruitful Search in the Attic of Ideas As a university professor, he continues his research at Rockefeller in the field of transcriptional specificities in mutagen- esis in bacteria. “Even with all of its problems, sci- ence today is better off than a half. century ago,” Lederberg notes. “As a newly minted Ph.D., I was facing the issue of ‘could you find a job to earn a living at doing research?’ “Even though we’ve seen some dimming of unblinking support for sci- entific research and molecular biology is of course much more crowded, any of my students still has a crack at revo- lutionary discovery if they will but seize the day,” he said. Although public scrutiny of scien- tific research and standards of account- ability are more stringent, perhaps more hostile than in the recent past, “The historical literature was of great interest to me when I was first wondering ‘Is there a genetics of bac- teria?’ I examined the existing obser- vations and bacterial systems and their credibility and found many things that informed my later work. “After I read about schemes for sero- typing in salmonella and saw how each new antigen combination was given a new species name, I had added incentive for investigating genetic recombination going on in the microbial world. Not many years later this led to the finding (with Norton Zinder) of transduction of markers by viruses as a method of ge- netic exchange,” Lederberg explained. After a period of study at Columbia University School of Medicine, Leder- Lederberg berg received his Ph.D. in microbiology at Yale. Although he has worked with many outstanding scientists, he speaks most fondly of his mentor, Francis Ryan. “He was an extraordinary, bright, gener- ous human being, a wonderful sparri partner who offered both discipline and respect. Everybody adored him, and some of his work in mutagenesis is being ‘exhumed’ today.” Following the receipt of his doctor- ate, Lederberg served as professor of genetics at the University of Wisconsin and then at Stanford School of Medi- cine. From 1978 to 1990, he served as president of The Rockefeller University. Lederberg doesn’t see a recrudescence of scientific McCarthyism. “Yes, the screws are a little bit tighter, and people are going to look more closely at mar- ginal research, including plagiarism as well as imputed fraud. However, any- one who exercises a modicum of com- mon sense and integrity has no rational basis for being deterred. “Unfortunately, social vigilance about the integrity of scientific research may create the impression that the dis- cipline is loaded with crooks and preda- tors,” Lederberg said. “We urgently need to dispel the idea that the primary moti- vation of researchers is to beat their competitors. I firmly believe that ideal- ism and the excitement of discovery are necessary parts of science.” thereby became toxicogenic, the acquired toxicity per- sisting for several transfers. He attributed the effect to the toxin itself: one can excuse him for not anticipating a post-Mendelian distinction between genotype and phenotype! It is easy to suggest that Sanfelice’s fil- trates still had some spores. I am not aware of any effort to confirm Sanfelice’s findings during the cen- tury after his report; of course a precise repetition is hardly feasible without Sanfelice’s identical strains. In recent work, a conversion of toxigenicity in C. botuli- num has been authenticated. C. tetani may have its toxin encoded on a large plasmid. In their pioneering study of lysogenicity in staphylo- cocci, Burnet and Lush (1936) reported a pigment change of a Staphylococcus albus to S. aureus after exposure to a phage B—this, to me, has the “smell” of lysogenic induc- tion. However, their investigations focused on a phage C, work that led them to the discovery of a mutation that VOL. 58, NO. 5, 1992 leads a bacteriophage to lose its lysogenizing power. Re- ferring to the insusceptibility of the lysogenic bacteria, they concluded, “According to [the] Wollmanns’ hypothe- sis, the distinction between the two alternatives faltered genetic constitution of the bacteria or induction by the phage] would disappear, the phage being regarded as a gene re-introduced into the genetic make-up of the orga- nism.” These thoughts had a profound effect on my own thinking about virus-gene relationships and the concept of the plasmid. Empirical corroboration of this paradigm came from Victor Freeman in his discovery of lysogenic conversion of Corynebacterium diphtheriae in 1951. According to his personal account: While carrying on investigations with phage B it was discovered that this phage had a marked activity on several avian cultures of C. diphtheriae. In view of this result it was decided to test for the possibility of 263 the presence of a dermal necrotic endotoxin after the method of Lazarus and Gunnarson, J. Bact. 1947, 53:705-714. The phage lysates of the avian cultures did produce a dermal factor toxic for guinea pigs. Further investigations soon revealed that the toxic substance being produced was true diphtheria toxin. In view of these findings the principal emphasis of the research project was modified to take advantage of these very pertinent observations etc. There was no practical devel- opment of a phage typing system. Several interesting patterns of specific lysis were developed through phage adaptation but owing to this instability these tests could not be utilized for typing purposes. In 1927, Frobisher and Brown had the idea that scarlatina was caused by a virus only secondarily related to the streptococcus. They evidently did con- vert a “cheese strep” to an erythrogenic one with filtrates. Their work was repeated by Bingel in 1947 and taken up again by John Zabriskie 20 years after that. Finally, Ferretti’s group in Oklahoma recently put the finishing touches on this as another canonical lysogenic conversion, the toxin genes having been located on the phage (plasmid). In 1953, I was collaborating with P. R. Edwards on the serotype of the H antigens of salmonellae. He told me of the conversion of some of the O somatic types with antisera. I tried to persuade him that antibody was a very unlikely reagent for transformation and that he was just seeing another example of selection. Zinder and I had just demonstrated phage-mediated transduction. So I asked, was there possibly some phage in his antisera? He countered that there was probably dissolved antigen being taken up by the altered bacteria. Indeed there was a phage! Iseki and Sakai, in the first of many such examples, demon- strated in 1953 the modulation of somatic antigen specificity in Salmonella spp. by lysogenic conversion. That is, the phage genome includes determinants altering its host’s development. DNA Transformation Fred Griffith discovered type transformation in pneumococci in 1928. This was followed by the work of Oswald Avery. The epochal paper by Avery, Macleod and McCarty followed 16 years later in 1944, in effect the discovery of DNA as a genetic agent. Griffith was not very revealing about the inside his- tory of his experiment; he was generally a private person. Hardly any of his papers survive. A civil servant in the British Health Service, Griffith initiated the serotypic classification of streptococci and pneumococci that we use today. That classification is invaluable in the tracing of epidemics and in planning for serotherapy. A “smooth rough” (S < R) variation was often observed, usually reversible on further cultivation or inoculation into mice. Griffith was obviously much concerned about the condi- tions of this variation and its bounds, but he never saw mutation of one type to another. Griffith believed that the revertible R strains con- tained some residue of S “antigen”; large inocula might provide enough “pabulum or stimulus” to enable the 264 Features reversion from R to S. However, he was fairly careless about phenotypic versus genotypic effects—and he cer- tainly lacked that vocabulary. His colleague Stewart Elliott suggested to me that Griffith may also have had in mind effects like those of mucins in protecting inocula during their incubation in test animals. Griffith conducted a series of transformation exper- iments using homologous strains. A stable R, derived from S-1 (which we designate R-1 although it exhibits no S-1 antigen), would engender smooth virulent re- vertants in the mouse when inoculated with heat- killed cultures of type 1. This would be in accord with his “pabulum” hypothesis. However, he later got S-1 as well when he used live R-2 cells! This is the paradigm of the transformation experiment; in a variety of other combinations, the killed S (smooth) cells transmitted their antigenic specificity to the rough recipients, regardless of their original type. Griffith knew he was onto something. And from this point, he was much more discerning about his terminol- ogy. Since the antigen itself is stable to heat whereas the transforming factor is destroyed by boiling, he clearly recognized the distinction between antigen and antigen- forming system (what we today call the gene): “By S substance I mean that specific protein structure of the virulent pneumococcus which enables it to manufacture a specific soluble carbohydrate.” We are at a loss trying to trace the intellectual influence behind his experiment, and such a statement, so modern in its outlook! He must have been aware of . paragglutination, if only the Weil-Felix reaction. He cites other articles from the Zentralblatt fiir Bakteriologie (Seligmann, 1917) regarding the diversification of dys- entery strains. But he left no tangible clues. What happened after that is, as they say, history. Unfinished Business: Proteus Cross-Reaction with Rickettsiae In 1916, Weil and Felix isolated Proteus strain X-19 from a patient; it agglutinated in cross-reaction with convalescent typhus sera. Other Proteus isolates, the XK and X2 strains, also cross-react with other rickettsiae. However, proteus is discounted as an eti- ological factor in rickettsial disease. Although it is generally assumed that these are merely accidental cross-reactions of polysaccharides, this coincidence seems remarkable. Rickettsiae are not believed to have a close affinity to proteus in terms of their global DNA homologies, but that is no argument against the borrowing of genes across taxonomic boundaries. The Weil-Felix test is still occasionally used for diagnosis but has been supplanted with preparations from cul- tured rickettsiae. It is not clear whether purified Proteus preparations would fare less well. There are many cross-reactions known today of hu- man blood group antigens with bacterial polysaccharides, e.g., with type XIV pneumococci. We can even refer to a tradition of parasites borrowing their hosts’ immunospec- ificity. It may be unlikely, but it is too soon to exclude that the transaction can occur at the DNA level. ASM News Untapped Lessons from History What is the use of rummaging through the attic of forgotten experiments? Most science is done in response to immediate contemporary stimuli such as methods, pro- vocative data, or trendy notions about what is exciting and important. Most students have an absolute aversion to historic reminders, an allergy to the dust in the library stacks. All they need is in this month’s journals. Paradoxically, it may be in the opening of new territories, more than in the cultivation of the neigh- boring fields, that historical retrieval may have the greatest use. There are still old treasure maps with hints of buried troves. In my own experience, historical material moti- vated new experiments on several occasions. Above all, the Salmonella serotypes worked out in the 1920s and 1930s were a distinct inspiration to me in 1945, encouraging the search for bacterial recombination in E. coli and what turned out to be transduction in 1952. Salmonella flagellar phase variation, which dates to Andrewes (1922) and was taken up by lino and me in 1956, proved a gold mine of new perspectives on gene regulation. Massini’s 1907 E. coli “mutabile,” Lac” > Lac*, spawned even more illustrious progeny, nur- tured at the Pasteur Institute. A century of argument on the biology of symbiosis is the patrimony of the plasmid concept. Somatic cell genetics was given a fresh start by antiquarian pathological histology on cell fusion in viral lesions. More problematic are historic reports in which only by luck can the experimental claims be reconstructed with enough rigor to instill confidence in them. Bacte- rial variation, unfortunately, has more than its share of such publications. This field has also been hindered by the lack of communication between the new biology of Mendel and Darwin on the one hand and the empirical successes of “classical” medical bacteriology on the other. For the latter, population analysis was very slow in taking hold, although it should have been implicit in the Koch-Cohn paradigms of pure culture technique. Per- haps we should also be more sympathetic with the old-liners as we realize that bacteriology did have some authentic phenomena of environmentally induced change, such as plasmid attenuation, on top of the selective phenomena and phenotypic enzyme induction that occasioned so much confusion. As we trace the history of bacterial variation, we see a fluctuating development that is by no means linear progress. The turn-of-century insights of Theobald Smith, Beijerinck, and a handful of others were supplanted by the morass of “bacterial dissocia- VOL. 58, NO. 5, 1992 Features tion” and cyclogeny. One may well say that knowledge can be lost as well as gained from generation to generation; stalled theories are better forgotten than taxed with the aura of disrepute and futility. In his address to the British Association in 1884, Lord Rayleigh remarked, “By a fiction as remarkable as any to be found in law, what has once been published, even though it be in the Russian language, is usually spoken of as ‘known’, and it is often forgotten that the rediscovery in the library may be a more difficult and uncertain process than the first discovery in the laboratory.” I am not sure that I would match my present labors with those in the laboratory, but I have inestimable help from a technology of bibliographic retrieval unknown in Rayleigh’s day. That technology may assist in the union of past, present, and future. O Suggested Reading Amsterdamska, O. 1991. Stabilizing instability—the controversy over cyclogenic theories of bacterial variation during the interwar period. J. Hist. Biol. 24:191-222. Burian, R. M., J. Gayon, and D. Zallen. 1988. The singular fate of genetics in the history of French biology. J. Hist. Biol. 21:357—402. Burnet, F. M., and D. Lush. 1936. Induced lysogenicity and mutation of bacteriophage within lysogenic bacteria. Aust. J. Exp. Biol. Med. Sci. 14:27-38. Ezzell, J. W., P. Mikesell, B. E. Ivins, and S. H. Leppla. 1985. The genetic basis of Pasteur’s attenuation of Bacillus anthracis cul- tures, p. 107-116. in H. Koprowski and S.A. Plotkin (eds.), The Wistar Symposium series, vol. 3. World’s debt to Pasteur. Alan R. Liss, Inc., New York. Freeman, V. J. 1951. Studies on the virulence of bacteriophage- infected strains of Corynebacterium diphtheriae. J. Bacteriol. 61:675-688. Galperin, C. 1987. Le bacteriophage, la lysogenie et son determin- isme genetique. Hist. Phil. Life Sci. 9:175-224. Lederberg J. 1949. Bacterial variation. Annu. Rev. Microbiol. 3:1-22. Lederberg J. 1987. Genetic recombination in bacteria: a discovery account. Annu. Rev. Genet. 21:23-46. McIlroy, A. P., A. E. Axelrod, and R. R. Mellon. 1948. The influence of sodium acetate upon the dissociation of a stain of hemolytic streptococcus. J. Bacteriol. 56:547-553. Sanfelice, F. 1893. Untersuchungen uber anaerobe Mikroorganis- - men. Z. Hyg. Infekionskr. 14:339-393. Smith, T. 1894. Modification, temporary and permanent, of the physiological characters of bacteria in mixed cultures. Trans. Assoc. Am. Physicians 9:85-109. Summers, W. C..1991. From culture as organism to organism as cell— historical origins of bacterial genetics. J. Hist. Biol. 24:171-190. Tatum, E. L., and J. Lederberg. 1947. Gene recombination in the bacterium Escherichia coli. J. Bacteriol. 53:673-684. (See this article for extensive bibliography on paragglutination.) Van Lanen, J. M., I. L. Baldwin, and A. J. Riker. 1952. Attenuation of crown gall bacteria by cultivation in media containing glycine. J. Bacteriol. 63:715-721. Wollman, E., and E. Wollman. 1925. Sur la transmission “para- hereditaire” de caracteres chez les bacteries. C.R. Soc. Biol. 93:1568-1569. * This article was originally presented at a Centennial Celebration of the Pasteur Institute, Paris, June 1988. Several important works on related themes have been published in the interval [54-57]. BIBLIOGRAPHY [On account of limitations of space, the bibliography published in the ASM News was substantially truncated.] 1. Almquist, E. 1924. Investigations on bacterial hybrids. Jour. Inf. Dis. 35: 341-346. 2. Andrewes, F.W. 1922. Studies on group agglutination. Jour. Path. Bact. 25:515-521. 3. Avery, O.T., MacLeod, C. M., & McCarty. M. 1944 Studies on the chemical nature of the substance inducing transformation of pneumococcal types. J. Exptl. Med. 79: 137-518. 4. Braun, W. 1953. Bacterial Genetics. Philadelphia: W.B. Saunders. 5. Bruner, P. and Edwards, P.R. 1948 Changes induced in the O antigens of Salmonella. Jour. Bacteriol. 55: 449. 6. Burian, R.M., Gayon, Jean; Zallen, Doris, 1988. The singular fate of genetics in the history of French biology." J. Hist. Biol. 21(3):357-402. 7. Burnet F.M. & White, D.O. 1972 Natural history of infectious disease. (Cambridge: University Press,). 8. Burnet, F.M., and Lush, D. 1936. Induced lysogenicity and mutation of bacteriophage within lysogenic bacteria. Austral. Jour. Exptl. Biology Med. Sci. 14:27-38. 9. Carlton, B.C. and Gonzalez, J.M. 1985. The genetics and molecular biology of Bacillus thuringiensis. 211- 249 in: Dubnau, David A, ed. The molecular biology of the bacilli. Orlando FL: Academic Press. 10. Cohen, S. and Warren, K.S., (eds). 1982 Immunology of parasitic infections. 2d. ed. Oxford: Blackwells. 11. Dubos, R.J. 1945. The Bacterial Cell. Cambridge, Mass.: Harvard University Press. 12. Eklund, M.W., Poysky, F.T., Reed, S.M. and Smith, C.A. 1971 Bacteriophage and the toxigenicity of Clostridium botulinum type C. Science 172: 480 - 482. 13. Elwell, L.P. and Shipley, P.L. 1980 Plasmid-mediated factors associated with virulence of bacteria to animals. Ann. Rev. Microbiol. 34: 465- 496. 14, Englund, P.T. & Sher, A. (eds). 1988 The biology of parasitism. New York: Alan R. Liss, Inc. 15. Ephrussi, B., Hottinguer,H., and Chimenes, A.M. 1949. Action de l’acriflavine sur les lévures. I. La mutation "petite colonic”. Ann, Inst. Pasteur 76:351-364. 16. Ezzell, J. W., Mikesell, P., Ivins, B. E. and Leppla, S. H. 1985. The genetic basis of Pasteur’s attenuation of Bacillus anthracis cultures. pp. 107-116 in Koprowski, H. and S. A. Plotkin (eds.). The Wistar Symposium Series, Vol. 3. World’s debt to Pasteur. Alan R. Liss, Inc.: New York, NY 17. Fairweather, N.F., Lyness, V.A., Pickard, D.J., Allen, G. and Thomson, R.O. 1986. Cloning, nucleotide sequencing, and expression of tetanus toxin fragment C. in Escherichia coli. Jour. Bacteriol. 165: 21-27. 18. Ferretti, J. J, Yu, C E, Hynes, WL, Weeks, C R 1987 Molecular characterization of the group A streptococcal exotoxin type A (erythrogenic toxin) gene and product. pp. 130-135 in Ferretti, J J, and Curtiss, R, eds. Streptococcal genetics. Washington DC: American Society for Microbiology. 19, Finn, C.W., Silver, R.P., Habig, W.H., Hardegree, M.C., Zon, G., and Garon, C.F. 1984. The structural gene for tetanus toxin is on a plasmid. Science 224: 881- 884. 20. Freeman, V.J. 1951 Studies on the virulence of bacteriophage - infected strains of Corynebacterium diphtheriae. Jour. Bacteriol. 61: 675- 688. 21. Frobisher, M. and Brown, J.H. 1927. Transmissible toxicogenicity of streptococci. Bull. Johns Hopkins Hosp. 41: 167-173. 21a. Galperin, C. 1987 Le bactériophage, la lysogénie et son determinisme génétique. Hist Phil Life Sci 9: 175-224. 22. Garfield, E. 1988 Announcing the SCI Compact Disc Edition: CD-ROM gigabyte storage technology, novel software, and bibliographic coupling make desktop research and discovery a reality. Current Contents (22):3-13. 23, Green, B.D., Battisti, L., Koehler, T.M., Thome, C.B. and Ivins, B.E. 1985. Demonstration of a capsule plasmid in Bacillus anthracis. Infec. and Immun. 49:291-7. 24. Griffith, F. 1928. The significance of pneumococcal types. Jour. Hyg. 27:113-59. 25. Hadley, P. 1939. Bearing of dissociative variation on the species concept among the schizomycetes. Jour. Inf. Dis. 65: 265-272. 26. Harris, H. and Watkins, J.F. 1965 Hybrid cells derived from mouse and man: artificial heterokaryons of mammalian cells from different species. Nature 207: 606-608. 27. Hirota Y. 1960. The effect of acridine dyes on mating type factors in Escherichia coli. Proc. Natl. Acad. Sci. U.S. 46: 57-64. 28. Inoue, K. and lida, H. 1970. Conversion of toxigenicity in Clostridium botulinum type C. Japan. Jour. Microbiol. 14: 87-89. 29. Iseki, S. and Sakai, T. 1953 Artificial transformation of O antigens in Salmonella E group. Proc. Japan Acad. Sci. 29: 121-126; 127-131. 30. Kauffmann, F. 1941 Die Bakteriologie der Salmonella-Gruppe. Copenhagen: Einar Munksgaard. 31, Lederberg J. 1949 Bacterial variation. Ann. Rev. Microbiol. 3: 1-22. 32. Lederberg J. 1952. Cell genetics and hereditary symbiosis. Physiol. Rev. 32: 403-430. 33. Lederberg J. and Edwards P.R. 1953. Serotypic recombination in Salmonella. J. Immunol. 71: 232-240. 34. Lederberg J. and lino T. 1956 Phase variation in Salmonella. Genetics 41: 743-757. 35. Lederberg J. and Tatum E.L. 1946 Novel genotypes in mixed cultures of biochemical mutants of bacteria Cold Spring Harbor Symp. Quant. Biol. 11: 113-114. 36. Lederberg, J. 1987. Genetic Recombination in Bacteria: A Discovery Account. Ann. Rev. Genet. 21:23-46. 37. Lederberg, J. 1988. The ontogeny of the clonal selection theory of antibody formation: Reflections on Darwin and Ehrlich. Ann. N. Y. Acad. Sci. 546:175-187. 38. Massini, R. 1907. Uber einen in biologischer Bezeichnung interessanten kolistamm (Bacterium coli mutabile). Archiv. Hyg. 61: 250-292. 39. McIlroy, A-P., Axelorod, A.E., and Mellon, R.R. 1948 The influence of sodium acetate upon the dissociation of a strain of hemolytic streptococcus. Jour. Bacteriol. 56: 547- 553. 40. Pasteur, L. 1881. De l’attenuation des virus et leur retour a la virulence. C.R. Acad. Sci. Paris 92:429-435, 40a. Provasoli, L. 1951 Destruction of chloroplasts by streptomycin. Cold Spring Harbor Symp. Quant. Biol. 16: 113- 120. 41. Rayleigh, Lord 1885 Report of the British Association for the Advancement of Science for 1884. pp. 1-23. (at p. 20.) 42. Sanfelice F., 1893 Untersuchungen uber anaerobe Mikroorganismen. Z. Hyg. Infektionsk. 14: 339 - 393. 43. Smith, Th. 1894 Modification, temporary and permanent, of the physiological characters of bacteria in mixed cultures. Trans. Assoc. Amer. Physicians, 9: 85- 109. 44. Smith, Th. 1934. Parasitism and Disease. Princeton: Princeton Univ. Press. 45. Sonneborn, T.M. 1946 Experimental control of the concentration of cytoplasmic genetic factors in Paramecium. Cold Spring Harbor Symp. Quant. Biol. 11: 236- 255. 46, Tatum E.L., Lederberg J. 1947. Gene recombination in the bacterium Escherichia coli. J. Bacteriol. 53: 673-684. {See this article, my doctoral dissertation, for extensive bibliography on paragglutination. } 47. Trager, W. 1986. Living together: the biology of animal parasitism. New York: Plenum Press. 48. Trevors, J. T., 1986 Plasmid curing in bacteria. FEMS Microbiology Reviews 32: 149-157. 49. Van Lanen, J.M., Baldwin, I.L., and Riker, AJ. 1952. Attenuation of crown gall bacteria by cultivation in media containing glycine. Jour. Bacteriol. 63:715-721. 50. Weil,E. and Felix,A. 1916. Zur serologischen Diagnose des Fleckfiebers. Wien. Klin. Wochenschr. 29: 33. 51. Weisburg, WG et al., 1989 Phylogenetic Diversity of the Rickettsiae Jour. Bacteriol. 171: 4202-4206 52. Wollman, E. & Wollman, E. 1925. Sur la transmission “parahéréditaire" de caractéres chez les bactéries. CR. Soc. Biol. 93: 1568-1569. 53. Zuckerman H.A. & Lederberg, J. 1986 Forty Years of Genetic Recombination in Bacteria. Postmature Scientific Discovery? Nature, 327 :629-631. 54. Brock, T. 1990 The emergence of bacterial genetics. Cold Spring Harbor Laboratory Press. 55. Burian, R., Gayon, J., and Zallen, D., 1988 The Singular Fate of Genetics in the History of French Biology, 1900-1940, J. Hist. Biol. 21: 357-402. - 56. Summers, W.C. 1991 From culture as organism to organism as cell: historical origins of bacterial genetics. J. Hist. Biol., 24: 171-190 57. Amsterdamska, O. 1991 Stabilizing instability: the controversy over cyclogenic theories of bacterialk variation during the interwar period. J. Hist. Biol. 24:191-222. 58. Sapp, J. 1987 Beyond the gene - Cytoplasmic inheritance and the struggle for authority in genetics. Oxford Univ. Press.