137 



in 1909. His investigations, made in collaboration with his assistant 

 MINKMAN, were published in the next year ). 



In the first place a detailed description is given of various en- 

 richment cultures for denitrifying bacteria. This part of the investiga- 

 tion was more or less based on work which VAN ITERSON had performed 

 several years earlier in BEIJERINCK'S laboratory 2 ). An analysis of the 

 gas developed in these crude fermentations led to the unexpected 

 result that in all cases nitrous oxide was present, although in greatly 

 varying quantities. Especially in experiments with high concentra- 

 tions of nitrate (8 to 12 per cent) a large percentage of the gas appeared 

 to be nitrous oxide, and BEIJERINCK rigthly emphasized the remark- 

 ableness of a biological process leading to the production of a con- 

 tinuous stream of gas containing about 90 per cent of nitrous oxide. 



A closer study of denitrification showed that in media of high ni- 

 trate concentration two special types of spore-forming bacteria, were 

 active. These unknown denitrifiers could be isolated, and were de- 

 scribed under the names of Bacillus sphaerosporus and Bacillus ni- 

 troxus. 



The main interest of the paper is, however, to be found in the 

 definite experimental proof that nitrous oxide is not only formed by 

 bacterial activity, but that there are also numerous bacteria which 

 are able to consume this gas. This holds in the first place for many of 

 the denitrifying bacteria themselves, which of course means that 

 nitrous oxide - - or the hyponitrous acid from \vhich it is an anhydride 



- is for these bacteria just an intermediate product in the reduction 

 of nitrate to free nitrogen. But also some bacteria which do not attack 

 nitrates themselves were able to decompose nitrous oxide. 



Most striking is finally the demonstration of a new case of "chemo- 

 synthesis", namely, the biological production of organic matter from 

 carbon dioxide with the aid of the energy derived from an inorganic 

 reaction. BEIJERINCK showed that a mixture of hydrogen and nitrous 

 oxide makes possible the development of a luxuriant microflora in an 

 inorganic medium containing carbon dioxide. In this case the energy 

 necessary for the carbon dioxide reduction is derived from a reaction 

 between the hydrogen and the nitrous oxide leading to the formation 

 of nitrogen and water. It is clear that this process is quite analogous 

 to the long-known bacterial utilisation of a mixture of hydrogen and 

 oxygen by the so-called hydrogen bacteria. 



Another more or less bewildering aspect of denitrification had 

 already been reported by BEIJERINCK in 1903 3 ). A study of the bac- 

 teria oxidizing hydrogen sulphide, thiosulphate, etc., as first described 

 by NATANSSOHN, had given BEIJERINCK the conviction that they 



1) Centralbl. f. Bakt. u. Parasitenk. II, 25, 30, 1910. 



2) G. VAN ITERSON Jr., Centralbl. f. Bakt. u. Parasitenk. II, 11, 689, 1904; Ibid. 12, 

 106, 1904. 



3 ) Handelingen van het 9e Nederl. Natuur- en Geneeskundig Congres p. 195, 1903; 

 cf. also: Arch, neerl. d. sciences exactes et naturelles Ser. II, 9, 131, 1904. 



