420 



Cytological observations on Bact. coli 



On the other hand, Miss E. Klieneberger (unpublished 

 observations) has recently demonstrated unmistakable 

 dumbbell bodies and their V-shaped division stages in 

 large streptococci from a human case of broncho- 

 pneumonia, and I have seen the same appearances in a 

 large, colourless unidentified coccus which was encoun- 

 tered as a contaminant on an old agar plate. One or two 

 Feulgen-positive bodies, similar to those which I have 

 previously described in two sarcinae are also present in 

 the cells of Staphylococcus aureus and of Streptococcus 

 pyogenes (unpublished observations). I cannot, there- 

 fore, agree with the conclusions drawn by Knaysi & 

 Mudd (1943) from electron micrographs, that these 

 organisms lack discrete nuclear structures and that the 

 organization of nuclear material varies widely among 

 different species of bacteria. 



Their small size renders cocci rather difficult objects 

 for the study of the chromatinic bodies and even in the 

 more readily investigated rod-shaped bacteria, resolution 

 of the chromatinic structures into chromosome-like 

 bodies is not always possible (cf. B. mesentericus, PI. 6, 

 figs. 17, 19). 



It seems plausible that the chromatinic bodies of 

 bacteria are homologous with the chromosomes of 

 more highly differentiated organisms. The positive 

 Feulgen reaction of these bodies, their special affinities 

 for nuclear dyes, their regular cycle of multiplication 

 co-ordinated with cytoplasmic division and their simple 

 and constant numerical relationships strongly suggest 

 that they are true nuclear structures. Moreover, their 

 mode of division appears well suited for the transmission 

 of hereditary factors in linear array such as occurs in 

 chromosome division, though whether they actually 

 perform this function is of course not yet known. 



Also, direct proof that the chromatinic bodies contain 

 nucleic acid is still lacking; to obtain this it will be 

 necessary to determine whether they maximally absorb 

 the same wave-lengths of ultra-violet light as the nucleic 

 acids, an undertaking rendered difficult by the small size 

 of the bacteria. Piekarski's claims (1938) to have 

 obtained a selective absorption of ultra-violet light 

 require confirmation. 



Further evidence for the nuclear character of the 

 dumbbell bodies in bacteria is afforded, however, by a 

 comparison of these bodies with the nuclear apparatus 

 of yeasts. Badian (1937) has studied several species of 

 yeasts and has shown that they possess two chromo- 

 somes which in form and size closely resemble the dumb- 

 bell bodies of bacteria. I have been able to confirm these 

 observations for the vegetative stages of Saccharomyces 

 cerevisiae and Schizosaccharomyces Pombe. My results 

 will be described fully elsewhere but in the present 

 connexion it is interesting to note that in yeasts delicate 

 strands extend between the two pairs of chromosomes 

 at anaphase and between daughter nuclei, which exactly 

 resemble the strands so often seen between separating 

 chromatinic structures in bacteria. 



Statistical evidence that the bacterial cell contains an 

 organ of higher radiosensitivity than the rest of the 

 cytoplasm has been provided by D. E. Lea and his 

 collaborators in a series of radiological investigati is= 

 (1937, 1940, 1941). This organ cannot be regenerated by 

 the cell once it has been destroyed by irradiation and is 

 probably identical with the nuclear material. It is hoped 



to supplement these observations with morphological 

 data on the effect of ionizing radiations on the chroma- 

 tinic structures. 



The chambered appearance that Giemsa's stain lends 

 to bacteria which have been fixed through the agar with 

 Bouin's fluid, can also be produced by iron alum haema- 

 toxylin in the case of the aerobic spore-formers. Using 

 this stain, Guilliermond (1908) has described such boun- 

 daries ('cloisons transversales') in B. mycoides. The 

 pattern of alternating dark and fine transverse markings 

 is also reminiscent of the chambered structure of large 

 spirochaetes as described by Gross (1911) and Dobell 

 (1912) (cp. PI. 8, figs. 28, 29 with figs. 143 and 144 of 

 Dobell's memoir). 



The existence of a bacterial cell wall as a structure 

 differing physico-chemically from the cytoplasm has 

 been recognized since the work of A. Fischer (1895) 

 and his contemporaries, but the clear conception of the 

 cell wall, derived from plasmolysing experiments, which 

 prevailed forty years ago has become somewhat obscured 

 by the difficulty of demonstrating it satisfactorily in 

 most of the bacteria which interest the medical bacterio- 

 logist. 



Knyasi (1930) has described the morphological rela- 

 tionship of the plasma membrane and the outer sup- 

 porting cell wall in plasmolysed bacteria. My own 

 results with the Bouin-Giemsa and NaOH-crystal violet 

 method fully confirm his findings With regard to this 

 relationship, but differ from them in showing that the 

 aerobic spore-formers have a multiple coll structure 

 from the earliest growth stages onwards. 



NOTE 



In the present account observations on tb cell wall 

 in B. megatherium have been used to supplement the 

 description of the chambered cytoplasm in rod forms 

 and filaments of Bact. coli and B. mesentericus, i.e. in 

 two species in which the demonstration of the outer 

 supporting cell wall which surrounds the cytoplasm 

 had not proved successful. Since this article first went 

 to press I have learnt distinctly to stain the cell wall 

 in all three organisms by mordanting Bouin-fixed 

 preparations with , % tannic acid and staining for 

 10 sec. ( ! ) with 0-02 % crystal violet in water. The 

 new preparations further emphasise the distinction 

 between cytoplasmic cell boundaries and transverse 

 cod walls (septa) drawn in the present article and have 

 allowed the formation of the transverse septa to be 

 studied in detail. These observations will be published 

 elsewhere. 



SUMMARY 



1. The basic observations of Piekarski (1937-40) and 

 F. Neumann (1941) on Feulgen-positive, chromatinic 

 structures, going through a regular cycle of division, in 

 the cells of Bact. coli and Proteus vulgaris are confirmed 

 and the view that these structures are nuclear in nature 

 is accepted. 



2. The chromatinic structures in bacteria from old 

 cultures, although usually distinguishable from the 

 cytoplasm, are too small to be resolved accurately. 

 After transfer to a fresh nutrient medium the chroma- 

 tinic structures increase in size and give rise to short, 

 often dumbbell shaped, rods (chromosomes) which 



179 



