UNIVERSITY OF CAMBRIDGE PROFESSOR R. A. FISHER, Sc.D., F.R.S. DEPARTMENT OF GENETICS WHITTINGEHAME LODGE MISS M.F.E SPEYER, B.Sc. > Research Assistant and Secretary co. : . 44 STOREY’S WAY , CAMBRIDGE | Tal. 55822 Dear Lederberg: Gavi di Te 19/8/49 Thank you for your letter of the 27th July. Your findings were of extreme interest to me, and I am very glad to hear that, you have been able to give a demonstration that Hfr is actually a sexually more potent strain,rather than 2 less intersting strain owing its vroperties in recombination experiments to sone vhysiologicsl activities. I had not been able to show that Hfr gives recombination @lso0 when recombination is not necessary jenaelt had enplpyédz faifferent tehbhnique resistant to virus T I tried twice to grow together in broth Hfr/end W 583,eafter 24 h incubation I platec yen, EMB lactose plates coated with phages Tj end tT; 3; results were always obscure¢ by a high mutation rate to resistance in the controls,and I never thought of testin- colonies for rec-mbinetion ,2s you did. Your results are,l think,also of great inte- rest in shoving thatrecombination occurs also if it is not neededsand they are very encouraging for my plans of following with micromanipulations what happens during and after fusion. If you go on with your experiments of copulation on comolete medium,you may ‘be. interested to knov that Hfr is rather slow in "copulation". I have followed the growth of colonies of prototrophs obtained with Hfr mmt or with 58-161 ,in compa- rison of controls :; 58-161 hes an aversge lag of about 6 hours ,;vhile with Hfr ihe deley is more than double. There is a great variation ,in both cases, between the -lag of single prototrophs. As to the differences betyeen your data and mine, I think two points,which may not be idertical in your and mine technique, may be of importance. Wren I mean "moist" plates,I actually mean pletes which have been prepared shortly before use. No more than 0.05 ml. can be plated on these plates,to obtain clear cut colonies. XNorewveryixtiink Results are,however,usually good, even if the plates are used as “goon as ager is solidified. Moreover, I think salts have great importance.I have “not yet Ry systematic research on the subject,hut have many hints for it,and I think it may be wobth while planning a research about it. I prepare a solution of salts in 10 times concentration,which is then added in the plate before pouring a test tube containaing a measured quantity of gbucose and agar. Preparing a 10 x concentrated salt solution, some precipitate is formed even when salts are added one after the other,dissolving them carefully each time. This precipitete is removed uy filtration, but precipitation continues slowly in the concentrated solution®. Once -that I used a salt solution in which a larger precipitate then usualikx had occurred, all recombinations were extremely low,and returned normal only when I changed the -galt solution. I am using the same fumuta salts,in the same proportions you susgest in your 1947 paper. I het am quoting alj théser2ther poor data at length, to suggest possibilities, ESSE SE think Ys Guite worth while testingx possible origins of the discrep2ncies you find. As to the efficiency of the exveriment with "asymmetrical" numer input of “ gells of either strain,with 58-161, I have done it only in the absence cf vitamin Be and I did not notice any high efficiency. I am very interested to know If you heve been able to récover the Mfr behaviour “in recbnhinants. + tested only 8 recombinants for this behaviour,and none of them we Hfr; two were,hovever,ratner higher than 58-161 ,giving nearly 10x more prototrorhs. As to the oppositional character of the Hfr effect , I think to remember correctly saying that Ar- x B gave about 1074 prototrovhs. However,I could not localize, 17 jntercre ing due to the difficulties of BSUVSRALLGNXASNLLONSAXIN crossover deta with Hfr men“ions in the Cambridge paper,the two genes, and they may be far par apart or close toge- ther; it may well be worth repekiing the mxperiment with other substrains of Hfr. Concerning strain 123, exneriments of outcrossing K 12 hed been sterted,after finding Ufr, with a view to test the possibility of nating types: however, Hfr is no more effective then 5€-161 whtn crossed to 123 . The result itself seemed to me hewever,intersting en-ugh ,though I have been long time thining about the persiste - of parental characters,and itsmmeaning. There remains the possibility,which you | ‘mention, shat syntrophism, Lectose instability of 123 (I have no data about it), lysogenicity of K12 concur to give an illusion of recombination. However, crossing 133 Vas with 58-14. one gets approximately the reverse proportions of sensitives and resistants. The ak "fermentation characters which do not recombine at all are th. more closelyxlinked with M- ; I could never use other biochemical markers on the Kl2 side»because W583 x 125 gives extremely tiny colonies,tinier than 123 its@af, * COL wen of very slow growth,ond therefore mut difficult to test. I was very interested to hear about your Auktaxeuxthe exneriments with Salmonel- lae. My experimenis withfoott /n ad been started,curiously enough, with the idea of using ad drug resistances,as you did. However, I had only streptomycin resistance in mind, after Monod's suggestion; Ny" did not work,for this purpose,there being no concentration of x Ny which kills ail sensitives and ajiows survival of alla resistants. So I started using streptomycin-sensitiveness on the side of the new coli strains(w@ used ,in the vs first instance,all those giving growth on minimal medium) and biochemical deficiéncies _ in K 12 streptomycin res’ stant strains, on a minimal medium with streptomycin . The trouble was that, to avoid killing alla bacteria at once,wé had to add streptomycin in a fourth layer,some hours later,and we never had any satisfactory results . Heve you good experience of sinhilar techniques? T shall certainly send you 123,as soon as possible; however,nobody remains in) Sambridge, during my absence, who can send you the strains, I shall be there only at the end of Septemmber,so if you hre interested to have it before then, please write to the - National Collection of Type Cultures, The Lister institute, Elstree,(England); they sin should send you the strain in 2 few dsys,charging 2s 6d. The N@.123 is the number of the ~Hh. chemin strain in the collection,and #”is labelled Esch.coli var.ecidi Lectici. The Collection may have changed its address recently, but they would forward your let:er in this case. I have had t-oubles to assess the biochemical requirements of the strain. It needs pro- bably cystine or methionine, but results were not reproddmibits with different samples. a, T hope to hear soon from you of new results. My best italian address,in vier” of the fact thet I shal] be moving about a while, is the one given below. Wishing you a@ good work \ Yours Ax 2 Cay L.L.kCavalli i Viale Vittorio Veneto 24 | Milano (Italy) ii { Jk ne > P.S. IT am sorry about the insufficient quotation wit: reference to Prof;Tatum's work. I have not yet had the proofs of the letter to Nature,so I am in time to worrect it