598 ANKUAIi EEPOBT SMITHSONIAIT INSTITUTION, 19 3 



feet. Both of these tunnels are ventilated by the natural movement of 

 air through the shafts and portals. The Holland Tunnel, with a total 

 length of 9,250 feet, an under-river length of 5,480 feet, and a capacity 

 of 1,900 vehicles per hour in each direction, or 46,000 per day, obvi- 

 ously required something more than natural ventilation. To this 

 end the ventilation of the tunnel was studied under three heads : 



1. The amount and composition of exliaust gases from motor vehicles. 



2. The dilution necessary to render the exhaust gases harmless. 



3. The method and equipment necessary for adequate ventilation. 



The impurities in the atmosphere of a tunnel used by motor vehicles 

 are the product of the combustion of gasoline. If complete combus- 

 tion occurred, the carbon content in the gasoline would be in the 

 form of carbon dioxide, which can be tolerated in considerable quan- 

 tity without injurious effects. In a gasoline engine, however, com- 

 plete combustion seldom, if ever, takes place. The exhaust gases 

 contain varying amounts of carbon monoxide, depending on such 

 variable factors as the quality of the gasoline, conditions of car- 

 buretion, etc. 



Carbon monoxide is a highly poisonous gas, injurious to health in 

 minute quantities if breathed for a long time, and if present in large 

 quantities is injurious even when breathed for a short time. Venti- 

 lation requirements are determined by the quantity of this gas in 

 exhaust gases. If sufficient fresh air is supplied to reduce this gas to 

 a safe percentage, other gases and impurities, such as carbon dioxide, 

 methane, and smoke, will also be diluted sufficiently. The first con- 

 sideration, therefore, was to determine the amount of carbon mon- 

 oxide that would be liberated in the tunnel. 



Investigations were carried out at the Bureau of Mines experiment 

 station at Pittsburgh. The schedule called for the testing of pas- 

 senger cars and trucks of various makes and capacities. The tests 

 were made with cars loaded and light, standing with engine racing 

 and idling, accelerating from rest on level grade and on maximum 

 grade, running at 3, 6, 10, and 15 miles per hour on level and up and 

 down a grade of 3i/2 P^r cent, corresponding to the maximum tunnel 

 grade. A total of 101 cars were tested. Gas samples were taken 

 directly from the exhaust pipe throughout the entire duration of 

 the test. 



In general, the results showed that the exhaust gases contained 

 about 6.8 per cent carbon monoxide and 8.4 per cent carbon dioxide, 

 developing only 67 per cent of the heat value of the gasoline. About 

 one-third of the gasoline fuel was wasted through incomplete 

 combustion. 



Experiments to determine the proper dilution to render the 

 exhaust gases harmless were conducted at the Bureau of Mines 

 experiment station at Yale. They were performed in a gas-tight 



