View report page

11
after some experience as the most suitable for the following reasons—Moulds were often found
sporulating in two days, and the spores disseminated from such gave rise, if the plate were kept
too long, to secondary colonies which vitiated the results. The large mould colonies tended also
to overgrow the bacterial colonies in their neighbourhood and prevent satisfactory sub-cultures.
I satisfied myself that the count on the fourth day was a reasonably fair one by repeating the
count, in many cases on the fifth and sixth days, and although I found that a few fresh colonies
became apparent on these later days, yet they seldom formed five per cent, of the total. Inasmuch
as the object was to compare different samples of air, it is plain that this comparison could be
effectually carried out by counting all the plates after a uniform time had elapsed, provided that
this time were such as to allow the great majority of the colonies to have developed. On only
two occasions was I obliged to relax this rule of counting finally on the fourth day; these were
in June, when the unseasonably cold weather had caused the temperature of the cool incubator,
from which the gas was turned off in summer, to fall to 16° C., much retarding the growth of the
colonies. On these occasions the count was taken on the fifth and sixth days respectively, which,
as I estimated from the rate of appearance of the colonies, corresponded fairly with the fourth day
under normal conditions. In counting the colonies, a map was made of each plate, every colony
being numbered for future reference. (3) Having determined the total number of colonies, subcultures
were made to determine their nature, and from these sub-cultures a number of further
cultures were usually requisite upon different media. Where the number of colonies in a plate
did not exceed 20 or 30, every colony was sub-cultured. Where the numbers exceeded this, the
quantity of sub-cultures would have become unmanageable, and I therefore marked out a half, or
sometimes even a quarter, of the plate and sub-cultivated every colony in this, making from the
rest of the plate, cultures only from such colonies as appeared manifestly diverse from those
which I had already taken. The total number of colonies microscopically and culturally studied
in the series of twelve complete observations was nearly 600.
Each complete observation involved the making of four plate cultures. I was anxious to
determine, if possible, whether bacteria could be found, in the air of the railway carriages or
elsewhere, to which a probable origin from human sources could be ascribed. It is well known
that a very large proportion of the bacteria of the air and of water are species incapable of growing
at the temperature of the human body. Pathogenic organisms, however, grow more rapidly and
better at the body temperature than at the ordinary air temperature. It was therefore plain
that, whilst I must rely upon the colonies arising in gelatin plate cultures incubated at 20° C., for
the total number of organisms present in a given volume of air, I should have a better chance
of discovering the presence of pathogenic organisms by incubating agar-agar plates at 37° C. I
hoped, too, that some interesting data for the comparison of the railway air with the fresh air
above might be obtained by finding the ratios between the numbers of organisms capable of
growing at blood heat, with the total numbers present—a line of observation not so far attempted,
to my knowledge, in air investigations. I therefore, in each experiment, took two samples of
railway air, each of 5 litres, and two similar samples from the fresh air. In each case one sample
was incubated in gelatin at 20° C., and one in agar-agar at 37° C. In only one experiment, and
that by inadvertence, did I fail to carry out this plan; this was in experiment IX., where I
accidentally made two gelatin plates from the two samples of tunnel air taken. The two plates
served, however, for the comparison of two different sections of the tunnel. I selected 5 litres,
as the volume of each sample, in preference to 10 litres, because the smaller number of organisms
obtained proved more manageable, and because, with the larger amount, the plates would have
been often so overgrown with moulds as to render sub-cultures very difficult. The number of
organisms in 5 litres was, moreover, as will be seen, quite sufficient for all practical purposes of
comparison.
I conducted two additional experiments, in the course of my investigations, with a view
to determining the presence of anaerobic organisms in the air of the railway. In these I aspirated,
in each case, 10 litres of air through glass wool and sugar plugs, as above described, and transferred
the plugs to culture tubes containing glucose-broth (2 per cent, glucose). These were
enclosed in "Buchner tubes," deprived of free oxygen by pyrogallic acid and caustic potash, and
were then incubated at 37° C. After the lapse of three or four days the cultures were examined,
and sub-cultures, both aerobic and anaerobic made from them. In neither of these experiments
did I find any evidence of the presence of anaerobic organisms, and I therefore made no more.
I varied the place of collection of the samples as much as possible. The twelve experiments
were distributed as follows—Three samples were taken from different station platforms, two from
the passages leading from the platforms to the lifts and stairway, two from the lifts themselves
as they were going up and down with passengers, three from the carriages in actual use, and two
from the air of the tunnels. The last-named were taken from the front platform of the carriage
immediately behind the engine, shut off by doors from the carriage and exposed to the tunnel air
only. The fresh-air samples were taken at various places near the line of route of the railway,
and at various heights above the ground. Quiet spots were selected for choice, as in crowded
thoroughfares throngs of people were apt to collect to witness the proceedings, and this might have
vitiated the results. In all cases great care was taken to collect fair samples of the air under
every-day working conditions, and to avoid abnormal and disturbing influences.
The details of the twelve observations are as follows—I have added to each the proportions
of carbon dioxide and organic matter found in the same air at the same time, the data being kindly
furnishe3 to me by Dr. Clowes. The CO2 is stated as a percentage—the organic matter as
grammes of oxygen absorbed by 100,000 c.c. of air.
I.—On March 14th, 1902, two samples, each of 5 litres of air, were taken on the down platform
of the Post Office station at 11.30 a.m. Trains were passing regularly, but there were not many