1914] Sumner, et al.: Physical Conditions in San Francisco Bay 79 
river water which, on the average, must leave the bay with each ebb 
35,600,000 
706 
22 and 24) that the total amount of water which passes out during 
each ebb tide is, on the average, about 1,077,600 acre-feet. From 
this it follows that the river water which enters the head of the bay 
tide is thus or 50,425 aere-feet. It has been shown (pp. 
in the course of a single tidal eycle amounts to less than five per cent 
of the total quantity of water discharged through the Golden Gate 
during the same period. Accordingly, more than ninety-five per cent 
of this quantity must have previously entered from the ocean. 
The year during which our observations were conducted was a 
particularly dry one. The rainfall for the three representative points 
above referred to (pp. 74, 75) was only seventy-six per cent of the 
average during the year 1912, while the discharge of the rivers for 
the first nine months of that year (beyond which the figures are not 
yet available) was only about forty per cent of the normal. It prob- 
ably would be safe to assume, therefore, that, within this period, not 
more than 2.5 per cent of the total volume of water discharged from 
the bay during an average ebb tide, came from the rivers. The re- 
maining 97.5 per cent must have entered from the ocean during the 
preceding flood. 
It might seem, on first thought, that we should have found a mean 
salinity in the bay such as would result from combining 97.5 per cent 
ocean water and 2.5 per cent fresh water. Assuming for the former 
a salinity of 34 per mille, the bay water should have had, according 
to this mode of reasoning, a mean salinity for the year of 33.15. In 
reality, it was found to be 27.48. 
Even if we substitute for pure ocean water, in these computations, 
water of the mean salinity actually met with during the flood stream in 
the Golden Gate (30.77) ,?° we have, as the result of such a mixture, a 
mean salinity of 30 per mille. In order to bring about the condition 
actually found, we should have to combine 89 per cent of water from 
the Golden Gate with 11 per cent of the river water. In other words, 
it would be necessary to assume a river discharge more than four times 
as great as that calculated from the published records. 
This contradiction rests upon a false conception of the actual 
movements of water in the bay. If sea-water of the assumed salinity 
and fresh water from the rivers really mixed throughout the bay, in 
20 Based on samples taken during the flood series of observations at station 
H-4967, which was the nearest of all to the Golden Gate. 
