1398 - 99 .] Admiral Makaroff on Oceanographic Problems. 395 
The evaporation of water from the Mediterranean is greater 
than the quantity supplied by rivers and rains. For this reason, 
the water becomes more dense, settles down, and goes back to the 
Atlantic by the under current. 
I wish to point out here that the temperature of the lower 
strata of the Mediterranean coincides with the mean winter tem- 
perature of the air in the eastern part of the sea. This is quite 
evident, because in winter the temperature of the water to a great 
depth corresponds to the temperature of the air. In summer, the 
surface water is much warmer, but this high temperature cannot 
penetrate to a great depth. I am sorry that I have not time to 
discuss more fully this question, but in the Straits of Bab-el- 
Mandeb we have the same phenomena as in the Gibraltar Strait 
and Mediterranean. Here again — by my observations — the tem- 
perature of the lower water strata coincides with the winter tem- 
perature of the air at the place where the water settles down. 
In the three straits already mentioned, we have a double current : 
superficial and bottom current. In the Straits of Formosa and La 
Perouse there are also two currents, but both are superficial. 
I ought to mention that the influence of the rotation of the 
earth on the direction and velocity of the currents cannot be over- 
estimated. I shall not discuss this question fully, but the fact 
that in every salt inland sea there is a circular rotation of the 
water in a direction opposite to the apparent movement of the sun, 
shows thalj the rotation of the earth has very much to do with 
the direction of the currents (see paragraph 222 of Le Vitiaz et 
V Ocean pacijique). In the vicinity of islands, for the same 
reason, the water follows a direction coinciding with the apparent 
movement of the sun. It is for this reason also that the water 
alongside the Chinese coast flows to the south during the 
north-easterly monsoon as well as during the south-westerly 
monsoon. The Kuro-Siwo current going to the north and north- 
east cannot touch the Chinese coast because there is brackish water 
flowing to the south-west. 
Plates IV., V., VI., and VII. show that in the Strait of Formosa 
the specific gravity and temperature of the water at the Chinese 
coast are quite different to what is observed off the coast of Formosa. 
This difference in the temperature and specific gravity may give to 
