548 Dr. W. B. Carpenter— 



upon one or the other of two conditions ; first, the mean "Winter temperature or 

 isocheimal of the surface; and «eco«rf, the temperature of the coldest water that is 

 admitted into it from the Ocean outside. If its communication with that ocean be 

 so shallow that the temperature of the water admitted through it is never lower 

 than the isocheimal, then the latter will be the uniform temperature of the entire 

 mass beneath the variable surface-stratum ; but if the communication be deep enough 

 to admit the water of a colder stratum of the Ocean, the bottom-temperature of the 

 Inland Sea will be the temperature of this stratum. 



Let us see how this view applies to two other cases. The Red Sea, like the 

 Mediterranean, is almost entirely cut off from communication with the deeper and 

 colder stratum of the Arabian Gulf, with which its surface-layer communicates 

 through the shallow Strait of Babelmandeb ; for whilst the lowest temperature 

 observed in that surface-layer, even in the northernmost extension of the Red Sea 

 known as the Gulf of Suez, is 71° (as I learn from Captain Nares, who has been 

 recently engaged in its survey), that temperature is there uniformly carried down to 

 the bottom at 450 fathoms ; and it may hence be pretty certainly affirmed that no 

 lower temperature than this will be found in the southern portion of the Red Sea, 

 even on a bottom exceeding 1000 fathoms in depth, since the loioest surface-tempera- 

 ture of that portion is probably never- less than 75°, the highest being nearly 90°. 

 Yet, in the Arabian Gulf, the temperature at a depth of 2,000 fathoms is certainly 

 not above, and is very probably helow, 361°. Here, therefore, the uniform temperature 

 ■which will probably be found to prevail throughout, beneath the surface-stratum, will 

 be the isocheimal. 



Now it seems to be the universal opinion of those who have most carefully studied 

 the existing Coral Formations, that the reef-building corals do not live and grow 

 at a greater depth than 20 fathoms; and since it is affirmed by Mr. Dana as a 

 deduction from the distribution of coral formations, that the existence of the reef- 

 builders is geographically limited by the isocheimal line of 68°,^ the Author cannot but 

 suspect that the bathymetric-al limit may be essentially a thermal one.* For all we know 

 of the relation of Temperature to Depth would indicate that even within the Inter- 

 tropical area of the open Ocean, the temperature at 20 fathoms may not be above 

 68°, and that in the next 10 fathoms it suffers considerable reduction. Now if the 

 temperature of the Red Sea nowhere falls below 71°, it is obviously a most interesting 

 question to determine whether the reef-building Corals are or are not to be found in 

 that sea at a greater depth than in the Oceanic area ; and if so, what is the greatest 

 depth at which they exist there. For since the Author's previous inquiries have shown 

 that Stony Corals, similar in all essential particulars to the reef-building types, can 

 live and grow at the depth of many hundred fathoms, there seems no a priori reason 

 why the latter should not thrive at like depths, if the Temperature be congenial to 

 them. 



A similar contrast is shown by the Temperature-soundings of Commander Chimmo 

 (which have been kindly communicated to the Author by the Hydrographer) between 

 the deep temperature of the Sulu Sea, a small area between the north-eastern portion 

 of Borneo and Mindinao, and that of the China Sea. The former, though not 

 ostensibly an Inland Sea (being but very partially surrounded by land), is so shut in 

 by reefs and shoals, as to have only a very superficial and limited communication 

 either with the China Sea or with the Celebes Sea. Notwithstanding this inclosure, 

 its depth is very great, ranginf; to 1,603 fathoms; and its Temperature-phenomena 

 present exactly the same contrast with those of the China Sea,^ that the Temperature- 



1 " On Corals and Coral Formations," p. 108. 



2 It is a very significant fact that the cold current ■vrhieh comes up from the south on the 

 eastern coast of South Anjerica, and which the Author regirds as the indraught of the Pacific 

 Equatorial current (as the similar current on the eastern coast of South Africa is of the Atlantic 

 Equatorial current), pushes the southern isocheimal of 68'^, the Coral Sea boundary, to the 

 notlh of the Equator, between the South American coast and the Galapagos, which, though 

 under the Equator, lie outside of that boundary. 



s The Temperatures here given are ihose of the boUnm at different depths in the line of the 

 teleiiraph cable between Singapore and Hong-Kong. Why the fall of temperature from 51° to 

 37° should seem to take place in the China Sea at so much smaller a depth than it does in the 

 Atlantic, cannot be positive. y affirmed until we have serial sotmdings which shall give the 

 Temperatures of successive strata in the deepest part of that sea. But as the temperatures 

 given above are those of the bultoin at various depths, on the sides of a iHtUeiu and as the care- 

 ful researches of the United States Coast Survey have placed it beyond all doubt that the colder 



