342 PROCEEDINGS OF THE SECTIONS. 
volume of this gas with nine-tenths of air is as explosive as gun- 
powder when fired. 
An accident of this kind might affect miles of sewer, and prove 
very disastrous by throwing up streets or buildings. 
Several accidents of this kind have happened in London and 
Paris, and even here one such case occurred in 1883, tearing up 
many chains of a newly constructed line along the Darling Har- 
bour railway. 
It is not necessary to adduce any more proof against the adop- 
tion of this kind of ventilator as there is little fear of its being 
selected by the authorities of this city, excepting in very special 
cases. 
One such case would be an instance, where a large trunk sewer 
received a number of small branches, inasmuch as a large sewer, 
when not running, rapidly generates much gas ; and it would be 
judicious, if space is obtainable, to erect one such shaft here to 
obviate the charging of the branches with gas formed in the 
larger sewer. 
MipstreeT GRATING VENTILATORS. 
entilation threw it 
out of competition, two things became manifest, viz. : that m 
order that the sewer air might not at any time assume sufficient 
expansive power to force itself through intercepting contrivances 
into the dwelling, appropriate openings should be —- and 
that those openings must be frequent along the course of the 
sewer. 
To meet these requirements in a practical way, that is to say, 
that should at once be effectual without being inordinately costly, 
the plan was adopted of making openings in the crown of the 
sewer, and building a pit or shaft from the crown of the sewer 
up to the street surface. 
By this method a series of pits covered with gratings were con- 
observable. “ 
To remove another objection, viz., that dilute as it might be, it 
_ night contain poisonous germs, the sewer air was made to filter 
itself through charcoal filters, for which it is claimed by some of 
