SUMMARYThe response of aerosolized viruses to relative humidity depended greatly on the composition of the fluid from which the viruses were sprayed. For example, the removal of salts from the spray fluid diminished the loss of infectivity of Langat virus (a group B arbovirus) at intermediate relative humidities. Salts were less toxic towards aerosolized Semliki Forest virus (a group A arbovirus) but did cause some loss of infectivity at higher relative humidities after prolonged storage of the aerosol. Polyhydroxy-compounds reversed the virucidal effects of salts on arboviruses. The removal of protein from the spray suspensions of arboviruses caused rapid loss of infectivity in the aerosol at very high relative humidities but had no detrimental effect at lower relative humidities. Poliovirus and T coliphage, which possess no structural lipid, retained high levels of infectivity following aerosolization at relative humidites of 7o ~ or above; at lower relative humidities they were inactivated rapidly. This rapid inactivation was increased further at lower solute concentrations of the spray fluid.The infectivities of aerosolized polioviruses and coliphages depended on the mode of rehydration during collection of the aerosols, but the infectivities of arboviruses in aerosols were unaffected by this. Atmospheric oxygen was not toxic to viruses in the aerosol state.

The bacterial and viral filtration performance of 12 breathing system filters was determined using test methods specified in the draft European standard for breathing system filters, BS EN 13328‐1. All the filters were of two types, either pleated hydrophobic or electrostatic, and these two types differed in their filtration performance. The filtration performance is expressed in terms of the microbial penetration value, defined as the number of microbes passing through the filter per 10 million microbes in the challenge. The geometric mean (95% confidence limits) microbial penetration value was 1.0 (0.5, 3.5) and 2390 (617, 10 000) for the pleated hydrophobic and electrostatic filters, respectively, for the bacterial challenge, and 87 (48, 212) and 32 600 (10 900, 84 900), respectively, for the viral challenge. In general, there was little change in the microbial penetration values following 24 h simulated use. It is concluded that results from the tests specified in the draft standard will allow comparisons to be made between different manufacturers’ products enabling an informed choice to be made.

SUMMARYDifferences in survival of Escherichia coli (strain B) sprayed from distilled water into air and into nitrogen as a function of relative humidity (RH) are reported. Two mechanisms which may contribute to death of airborne bacteria are described. In air one death mechanism occurring at low RH is attributed to the action of oxygen causing damage to flavin-linked enzymes as a result of free radical activity. Free radical suppressors are therefore expected to protect airborne E. coli B. Also, electron transport inhibitors like sodium azide, 2,4-dinitrophenol and potassium cyanide are shown to protect E. coli B against lethal effects of oxygen. An analogy is drawn with effects of oxygen on freeze-dried E. coli B. A second death mechanism of E. coli B in air occurs at higher RH's and is considered to result from the effect of aerosolization on RNA synthesis. The activation of RNAse as a possible protection to bacteria in the post-aerosolization medium is discussed. I N T R O D U C T I O NBecause the effect of aerosolization upon bacteria survival is related in a complex manner to relative humidity (RH) of the atmosphere, earlier reports (Dunklin & Puck, 1948;Davis & Bateman, 1960;Bateman, McCaffrey, O'Connor & Monk, 1961) have suggested that bacterial sensitivity increases at certain critical degrees of dehydration. Later reports (Hess, 1965;Cox, 1966) have shown that other factors, e.g. spray fluid, oxygen content of the carrier gas and the collecting fluid, influence the disruption of vital structures and the unbalancing of metabolism. Kinetic studies indicate that : (i) the loss of viability of bacteria is not a simple process; (ii) first-order reactions do not apply; (iii) typically, death rate is initially high and subsequently diminishes progressively.The possibility that overlapping lethal mechanisms exist, makes it extremely difficult to identify bacterial death by a given mechanism in a given set of conditions. One approach to the problem is the use of inert atmospheres to limit the effect of gaseous factors. Another is comparison between results of analogous studies, such as freeze-drying experiments, to derive support for evidence produced in aerosol tests. This paper reports results of such investigations. METHODS Growth of organismsCultures of Escherichia coli (strain B) were grown in 100 ml. of a chemically defined medium for 16 hr at 37", using a shake flask technique. The medium consists of 21-2

SUMMARYSemliki Forest virus preparations of different degrees of purification were sprayed into atmospheres of different relative humidity. Inactivation of airborne virus was greatest at high relative humidities and it decreased gradually as the relative humidity was decreased. Removal of salts from the sprayed suspension resulted in improved survival over the whole range of relative humidity tested. Extraneous protein was essential for survival at high relative humidities. Polyhydroxy-compounds protected airborne virus very well in conditions of low relative humidity.

Airborne Semliki Forest virus and T coliphages were inactivated at a considerably enhanced rate in open air compared with enclosed air. Open air exerts its maximum sterilizing activity on viruses contained in the smallest sized particles.

Tests described in this paper show that gelatine membrane filters used in the MD8 microbiological air sampling system collected monodispersed aerosols between 0.7 and 1.0 microns containing viable Bacillus subtilis var. niger spores, with an efficiency of 99.9995%. Gelatine membrane filters linked to the MD8 control pump system were as effective as the well established Casella slit-to-agar device for collecting some viable bacteria, nebulized under controlled experimental conditions and naturally occurring airborne micro-organisms in a pharmaceutical plant. By using a long flexible hose connection to the control pump, the head could be positioned where sampling was required in locations remote from the pump exhaust, making it suitable for microbiological monitoring in critical locations such as laminar flow stations and isolators.

S U M M A R YThe maximum bactericidal effect of oxygen on airborne coliform bacferia occurred at the lowest relative humidity tested. Escherichia coli COWUNE and Klebsiella aerogenes were very much less sensitive than E. coli B and E. coli JEPP. Only a small partial pressure of oxygen in a nitrogen-oxygen mixture at atmospheric pressure was required to produce the maximum effect. The oxygen toxicity for each bacterial strain was altered by changing the growth conditions or by including certain chemicals in the spray fluid. These chemicals are protective to freeze-dried bacteria and therefore inactivation of oxidative enzymes in freeze dried and airborne bacteria probably occur by similar mechanisms. The same oxidative enzymes in bacterial colonies were slightly inactivated by hyperbaric oxygen pressures.

No abstract