GC 231 
 
 .W68 
 Copy 1 
 
Digitized by the Internet Archive 
 in 2011 with funding from 
 The Library of Congress 
 
 http://www.archive.org/details/oncirculationofoOOwilk 
 
i 
 
 ,-y\ l v 
 
 GC 231 
 
 .W68 
 Copy 1 
 
 \"UJ48' 
 
 
 
 ON THE 
 
 CIRCULATION 
 
 OF 
 
 
 T 
 
 H 
 
 E OCEAN 
 
 BY 
 
 s. 
 
 
 CAPTAIN CHARLES WILKES, U.S.N. 
 
 
 
 PHILADELPHIA: 
 
 
 
 
 1859. 
 
 
ON THE 
 
 CIRCULATION 
 
 THE OCEANS 
 
 CAPTAIN CHARLES WILKES, U. S. N. 
 
 J 
 
 PHILADELPHIA: 
 
 1859. 
 

 ^ 
 
 ' ^ 
 
 The following Paper on the Circulation of the Oceans, is part of a chapter 
 of Vol. XXIV. of the Exploring Expedition, which is now being published 
 by the Government. I have availed myself of the meeting of the American 
 Association, at Springfield, to present my views on this interesting subject. 
 
 Charles Wilkes. 
 
 Washington Citt, August, 1859. 
 
 (H) 
 
THE 
 
 CIRCULATION OF THE OCEANS, 
 
 Heretofore, in the Narrative of the Exploring Expedition, (vol. 
 v. chap. 12), I have treated of the flow of the surface currents of the 
 ocean in connection with whaling, not only for the guidance of the 
 practical seaman, hut especially for those engaged in the whale fisheries 
 — that great interest of our country for which the expedition was fitted 
 out, and to which its hydrographical labors were more particularly to 
 be devoted. 
 
 The subject of the Circulation of the Ocean, and the physical causes 
 which originate and keep it up, have claimed my attention for many 
 years. The facts elicited from the observations made during our 
 explorations, and which other navigators have since confirmed, have 
 led to the adoption of the theory which I shall explain in the sequel. 
 
 The circulation of the ocean is apparently one of great complexity, 
 and seems to offer a wide field for research and investigation. I shall 
 endeavor to limit myself to the causes which are known to exist, and 
 adduce such facts and analogies as have a bearing on the many phe- 
 nomena which have been observed, but remained unexplained, in the 
 movement of the waters. 
 
 To effect the circulation of the ocean, it is evident that the forces 
 must be of great power and universal action, to be adequate to produce 
 the vast results we witness. The combined effect of the temperature 
 of the ocean depths and surface, evaporation, the hydrostatical pres- 
 sure, gravity, and the action of the centrifugal and hydrodynamical 
 forces, we believe are commensurate to this end. While the centri- 
 fugal force and evaporation through heat tend to change the form of 
 the fluid mass from its normal state, gravity and the hydrodynamical 
 movements restore it : forces constant and continuous, acting and 
 reacting ; forces engendered by the rotation of our planet ; forces 
 derived from and governed by the well-known laws of the pressure 
 and movement of fluids, and resulting from the subtile agency of heat 
 
 ( 3 ) 
 
4 THEORY OF THE 
 
 emanating from the earth. These are ever present and always acting 
 on the aqueous parts of our globe, -which embrace three-fourths of its 
 surface. All experiments and observations go to show the effects of 
 such action, and we think they afford sufficient data to prove that the 
 circulation of the ocean must result from their combined power. 
 
 Before entering upon the circulation of the ocean, we will refer to 
 some interesting facts which experiments have brought to light, rela- 
 tive to the temperature and density of the sea at the surface, and at 
 great depths, which I deem essential to be well understood. 
 
 It has been found, by experiment, that a zone of the maximum 
 density of sea-water, as shown by the thermometer, lies between the 
 55th and 60th parallels of N. and S. latitudes. On this zone the 
 temperature of the sea-water remains constant, and it is found to be 
 invariable to whatever depth the sounding-line penetrates. 
 
 At the south,. this zone is nearly entire; but from the interference 
 of the land, and the comparatively circumscribed space the waters 
 occupy in the Northern Hemisphere, it varies, within those parallels, 
 somewhat from a direct line. 
 
 Assuming, as our guide, the temperature of maximum density of 
 sea-water at a few degrees below that of distilled water,* we are ena- 
 bled to trace this line in the depths of the ocean, both towards the 
 poles and the equator. This has not yet been done quite as far as 
 practicable, but sufficiently so to establish that it descends rapidly 
 towards the poles, the colder water rising above, forming a convex 
 curve ; while towards the equator it sinks to the depth of 2000 
 fathoms, or 12,000 feet, forming a concave curve between the two 
 zones, where the warmer water is found above. Beneath this curve 
 an invariable temperature exists, viz., that of the maximum density 
 of sea-water as shown by our thermometers. The diagram on page 
 7 shows these curves on a section passing through the polar diameter. 
 
 By this it will be perceived that the zones of maximum density 
 only touch the surface of our globe on two parallels of latitude, north 
 and south. 
 
 It has been long held that the heat of the ocean is derived from 
 the sun ; but such an assumption must be erroneous, as it has no facts 
 or analogy to support it. Experiments, of which we have a large 
 number, go far to disprove it, and to show that the heating powers 
 of that luminary can have but little effect upon the ocean. 
 
 * We are aware that the temperature of the maximum density of sea-water 
 depends upon its specific gravity, and that it varies according to the quantity of 
 salt it contains ; but with a specific gravity of 1.02800, the temperature of its 
 maximum density may be put down at 36.5. 
 
CIRCULATION OF THE OCEANS. 5 
 
 It is well known that water cannot be heated from the surface 
 downwards, and that it is a very slow conductor of heat in that direc- 
 tion ; therefore there is, in effect, a barrier at once raised against the 
 assumption. 
 
 The ordinary decrease of the temperature of the ocean, in descend- 
 ing within the tropics, is, for the first 100 fathoms, 6° to 7°, equal to 
 one degree of the thermometer for every 90 feet. In descending on 
 the land, the increase of the temperature is one degree for every 56 
 or 60 feet of descent. 
 
 It must be evident that the heat of the sun's rays cannot penetrate 
 to the depth of the rays of light. In my experiments to ascertain 
 the depth of the ocean to which light reaches, it was found to be in the 
 ratio of the heat of the water. When the sea had the temperature 
 of its maximum density, an object could be seen from 32 to 40 feet; 
 but when the temperature rose above 75° Fahr., it increased to 170 
 and 200 feet. Between these extremes, the progression of depth is 
 believed to be from 4 to 5 feet for each degree, at which distance the 
 rays of reflected light are visible under the same angle. The experi- 
 ments indicate that light penetrates the ocean to the depth of 85 
 fathoms, or 500 feet. If it be true that only one-half of the vertical 
 rays of light reach through the first 20 feet, one-fourth through double 
 this depth, and the Too'oooth P ar t penetrate to the extreme depths, it 
 must be evident, when the progression of heat from the surface of a 
 fluid downwards is considered, that the heat of the sun cannot pene- 
 trate as far as its light, and can only effect the evaporation, and does 
 not tend to raise the temperature, of the fluid, unless in confined 
 spaces or vessels. 
 
 When changes occur in the temperature beneath the surface, they 
 can be traced to other causes. The changes of the ocean incident to 
 the seasons are very small, as is shown by the isothermal lines ; and 
 proceed more from the effects of radiation and evaporation, than from 
 any heat acquired by the direct rays of the sun. The temperature 
 of the ocean does not vary, in the same locality, night or day, three 
 feet beneath the surface. The highest temperature of the ocean has 
 been estimated to be 6° Fahr. above that of the land. 
 
 We should not look, therefore, to the sun as the source from whence 
 the ocean derives its heat : it must come from the earth. Experiments 
 have shown us the ratio of the increase of heat in descending towards 
 the centre of the earth, and they also prove that the sea decreases in 
 heat until the curve of maximum density is reached. The average 
 depth of the ocean is computed to be 25,000 feet ; and in whatever 
 direction the centre of the earth is approached, the ratio of the increase 
 
t> THEORY OF THE 
 
 of temperature would be the same ; and at the above depth it would 
 be found to be, on the solid parts of the earth, about 400° Fahr., or 
 equal to that of red-hot iron. 
 
 It is scarcely to be supposed, from our experiences at the surface, 
 that the ocean could come in contact with such a high heat; yet, when 
 the enormous pressure is taken into consideration, it might be so, 
 without any remarkable effects ensuing. We know that the boiling- 
 point of water falls on removing pressure, and rises when it is increased, 
 and that water may be heated to very high temperatures. If such is 
 the case, the heat at the bottom of the ocean might be received as 
 latent heat, be disseminated throughout its mass, and be insensible to 
 our instruments as long as sufficient pressure continued. We believe 
 this point to be the curve of maximum density pointed out on the 
 various parallels before referred to. From this curve upwards the 
 pressure rapidly decreases, and becomes inadequate to confine the latent 
 heat which would be evolved to the surrounding waters, and finally to 
 the atmosphere at the surface. 
 
 Although the average depth of the ocean may be as before stated, 
 we are well aware that, in many places where the phenomenon of 
 greatest heat exists, its depth is much less than the curve of maximum 
 density of the latitude, and that the temperature at the bottom might 
 not be over 200° Fahr., and would be more expeditiously disseminated 
 to the waters, and to the atmosphere above. It is well known that 
 there are large submarine streams of a low temperature flowing into 
 immense basins, which, combined with the outflow of the warm water 
 on the surface, would soon reduce them to an equilibrium of tempe- 
 rature with the surrounding seas, if some source of heat did not exist, 
 other than that derived from the sun. As instances, we will cite the 
 Caribbean Sea and the Gulf of Mexico. The assumed depth of these 
 inland seas would assign to the bottom a temperature of from 150° to 
 200°. Furthermore, if it is a fact that the earth has a higher tempe- 
 rature at the bottom of the ocean than at the surface (which all analogy 
 would go to show), it must follow that this heat is conveyed, after pass- 
 ing into the waters, upwards through the intestinal currents (if the laws 
 of the propagation of heat through fluids remain as they are on the 
 surface), to whatever amount of pressure or compression it may be 
 subjected. As the warmer atoms rise towards the surface, the pressure 
 being gradually removed, its latent heat (if such it is) will be given 
 out, and finally reach the atmosphere above. In no other way could 
 the heat of these seas, including an area of 1,500,000 square miles, be 
 maintained at the temperature of 84° to 88° of Fahr. It is believed 
 that the ocean heat is thus derived and maintained. 
 
CIRCULATION OF THE OCEANS. 7 
 
 We will now recur to the forces which maintain the equilibrium, as 
 well as produce the motion or flow, of both surface and submarine 
 streams. The well-known action of the centrifugal force must tend 
 to draw the lighter particles of fluid of the earth's surface towards 
 the equatorial regions, cbanging the normal statical level to the form 
 of a meniscus, extending north and south of the equator ; while the 
 effect of the rotation is also to heap up the fluid on the east side of 
 the continents. This accumulation in height of the ocean lies upon 
 the waters of maximum density as a base, which unites the mass or 
 columns. Experiments have proved that this line is a curve, extending 
 from one zone to the other, the middle of which is 12,000 feet in depth 
 
 
 
 
 
 
 
 
 
 
 1 
 
 
 jr; 
 
 r — 2 
 
 
 
 
 
 A 
 
 
 
 
 
 
 
 
 
 
 
 
 
 Surfaces^ 
 
 
 o \ 
 
 
 di 
 
 
 
 r 
 
 j^S&r/aK 
 
 
 
 
 / 1 a a. Meniscus. * 
 
 /*" b b. Normal level. 
 
 c c. Arc of max. temperature. 
 
 d d. Convex curve 
 
 at the equator. It must result, therefore, that any variation or change 
 of height in any part of the column, would be felt throughout, and the 
 hydrostatical pressure would be changed into a hydrodynamical force, 
 and cause a flow towards the equator, which would be in proportion to 
 the weight and pressure. 
 
 The compression of water being as 1 in 20 of its bulk, will operate, 
 as well as its expansion by heat, to increase its volume as it rises from 
 the deep sea, and thus augment the elevation. It will thus be seen 
 that the action and reaction of these forces on each other must be 
 continuous — evaporation, and the centrifugal forces tending to carry 
 off, and destroy the equilibrium ; whilst it is restored by the force of 
 gravity, and the hydrodynamical forces engendered. The intestinal 
 currents disseminate the heat derived from the earth, and preserve 
 the temperature of the ocean. 
 
 I have made no allusion to the winds, which are often alleged to be 
 the prime cause of the circulation ; but I cannot consider that they 
 have any agency, nor can I attribute to them the great and constant 
 action that is necessary to keep up this circulation in the oceans, 
 which we everywhere witness. Doubtless, the winds have their effect 
 
8 THEOKT OF THE 
 
 in interrupting these currents, and producing the many anomalies we 
 constantly observe in the surface currents, which tend to retard, 
 change, or increase their flow — raising to a great elevation bodies of 
 water which are confined in bays or rivers ; but these are only tem- 
 porary, and the reaction -which gravity brings about soon restores 
 them to a level, or equilibrium. 
 
 Neither can I impute to the rotation of the earth the effects we 
 witness in the set and velocity of a stream or current, except so far 
 as the centrifugal force is concerned in elevating the mass. In all 
 the great oceans, we find the streams and. currents of great velocities 
 flowing in opposite directions on parallel lines, in the same latitudes, 
 as well as under the equator. The causes must be continuous, and 
 act and react upon each other ; and the rotation could not produce 
 like effects in opposite directions. 
 
 It will be observed, by an inspection of the globe, that there are 
 portions of the oceans where this pressure would act on the land, and, 
 therefore, cause less variation or change than those parts where the 
 ocean has the greatest range in latitude, when the fluctuations would 
 be smaller and less frequent, and the flow or hydrodynamical move- 
 ment would be less active or rapid; consequently, the submarine 
 streams would have less power and constancy, and not be so apparent. 
 The ocean, from these zones in the Northern and Southern Hemi- 
 spheres, is found to set towards the equatorial regions ; but the sub- 
 marine streams, it will be noticed, are found to flow towards and on 
 the eastern side of the great oceans. 
 
 Having referred to what I deem essential points, I shall endeavor 
 to explain the action of these forces, which are well known to exist, 
 and enter fully into the consideration of the circulation which they 
 bring about and keep up — noticing all the phenomena they offer. 
 
 As heretofore, the streams and currents will be designated as polar 
 and equatorial, from whence they flow, and be classed under four 
 heads, viz. — 1. Submarine streams ; 2. Surface streams ; 3. Eddy 
 currents ; 4. Intestinal currents. 
 
 Under the first of these divisions are comprised those great move 
 ments of the waters, from the polar towards the equatorial regions, 
 produced by the hydrodynamical forces. Where tho ocean is conti- 
 guous, there can be little doubt that the hydrostatical pressure exists 
 throughout the whole extent of our globe. As a large part of the 
 Northern Hemisphere is occupied by land, it will be seen that this 
 pressure must be very unequal; and no hydrodynamical action of 
 any extent can take place, except where there is no land interposed, 
 whereby its action is prevented or deflected. The tendency of the 
 
CIRCULATION OF THE OCEANS. 9 
 
 polar waters towards the equatorial regions, is a well-ascertained fact, 
 proved by the great masses of floating ice and ice-islands, or bergs, 
 which are annually borne to lower latitudes ; and by the low tempe- 
 rature of the ocean, traced in a continuous route. 
 
 Another fact will serve to prove the existence of these submarine 
 streams. The medusa, which is the appropriate food of the sperm- 
 whale, is generated in the cold waters of the Antarctic seas, and is 
 carried to the low latitudes ; and the places where they are deposited 
 are the feeding-grounds of the latter animals. The locomotive powers 
 of the medusa precludes all supposition that they could reach the lower 
 latitudes except through the agency of the polar streams. 
 
 There are but four submarine polar streams of which we have any 
 knowledge, viz., one in the Northern Hemisphere, and three in the 
 Southern. The first is found in the North Atlantic Ocean, under the 
 name of the Labrador and Great Northern Polar Stream. It is first 
 perceived off the coast of Labrador, in the latitude of 55° N., longi- 
 tude 60° W., and appears both as a surface and a submarine stream, 
 occupying a space of the ocean between Europe and America. On 
 reaching the banks of Newfoundland, its surface stream is interrupted 
 and deflected by the waters of the Gulf Stream, which causes it to 
 pass along the coast of Newfoundland, and that of the United States, 
 as far as the capes of Virginia. The great submarine polar stream 
 continues its onward course towards the Azores, Madeira, Canary 
 Islands, and Cape de Verdes, and thence to the equator. Its velocity 
 has been ascertained, by the movement of the icebergs, to be equal to 
 thirty miles in twenty-four hours. Its width continues until it reaches 
 the latitude of 35° N., where it is contracted to nearly one-third 
 its former width, and flows on to the west coast of Africa, surround- 
 ing the Cape de Verde Islands, and thence to the equator, where it is 
 found, by its temperature, at the depth of 100 fathoms, to be 200 miles 
 wide. Its whole route is through 50° of latitude, a distance of some 
 8000 miles. It first inclines to the southeast, and then to the south. 
 Encountering the Cape de Verde Islands, its waters accumulate, or are 
 elevated, and give rise to a surface current, which flows to the south- 
 ward and eastward, along the shores of Africa, into the Gulf of Biafra. 
 
 Of the submarine streams of the Southern Hemisphere, we name : 
 1st. The Agulhas Stream, which bifurcates on the Cape of Good 
 Hope — the greater part flowing along the west coast of Africa 
 towards the equatorial region, whence it sweeps towards the Bight of 
 Biafra, including the island of Annabon, and thence west towards the 
 equator, where, as before stated, it encounters the Northern Polar 
 Stream. 
 
10 THEORY OF THE 
 
 The Agulhas Stream begins its flow in the 55th parallel of S. lati- 
 tude, and between 20° and 40° W. longitude. Its onward course is 
 also marked by the ice-islands, or bergs. In its progress to the north, 
 its low temperature is perceived at the surface on the bank from which 
 it derives its name. As in the case of the Gulf Stream, which passes 
 over the Labrador, we here find the Mozambique current overlying 
 and deflecting the Agulhas. There is a difference of some 10° in 
 the temperature of their water. They do not commingle : the warm 
 current from the north sweeps the surface current away to the south- 
 west ; but the Agulhas, or Great Submarine South Atlantic Polar 
 Stream, continues its course onwards to the north. Its width at the 
 Cape of Good Hope is not more than 200 miles ; but as it sweeps 
 onwards to the equator, it attains, off* Cape Palmas, a width of 360 
 miles. Before it reaches this point, the surface current is again per- 
 ceived, and has acquired a velocity of from one to three miles an 
 hour, cutting the equator at an acute angle, about the longitude of 
 25° W., and flows towards the coast of Brazil, merging itself in the 
 Guyana current. 
 
 Under the equator, this great southern sub-stream encounters and 
 commingles with the great northern Atlantic submarine stream, which 
 gives rise to numerous rips, boilings, overfalls, bubblings, breakers, 
 and smooth spaces, alternating with rough waters — ■ evincing unmis- 
 takeable signs of a conflict by the waters of the submarine streams 
 of the great deep. 
 
 A part of the Agulhas Stream flows into the Indian Ocean, and 
 has been traced on the east side of the Mozambique Channel, deve- 
 loping itself along the island of Madagascar. We have no knowledge 
 of its having been discovered in any other locality ; and our deep-sea 
 soundings for temperature, as we crossed that ocean on the usual 
 homeward-bound route, failed to show any (unless it was below our 
 soundings) between the Straits of Sunda and the Cape. 
 
 The second southern submarine stream sets upon Cape Leeuwin, 
 the southwest cape of Australia, on which it divides — a small part 
 flowing off to the eastward, along the south coast ; but the largest 
 part passes on to the north, on the west side of Australia. This sub- 
 marine stream flows with less velocity, and the surface current is often 
 diverted by the strong winds which prevail at certain seasons of the 
 year. The velocity of the under-current is from fifteen to twenty 
 miles in twenty-four hours, and its extent in latitude is 2000 miles. 
 Owing to the configuration of the Indian Ocean, and its extending 
 but a short distance to the northward of the equator, its hydrostatical 
 pressure remains almost constantly the same, and without those flue- 
 
CIRCULATION OF THE OCEANS. 11 
 
 tuations which prevail in the ocean that lies in the great valley of the 
 North and South Atlantic, extending from pole to pole. It is asserted 
 that the Red Sea, and probably the Persian Gulf, have a higher level 
 than the Mediterranean. The elevation of the former above the lat- 
 ter is reported at from twenty-four to thirty-six feet , and this is said 
 to be accounted for by the peculiar form of the canal or straits of 
 Babelmandel, at the entrance of the Red Sea ! It is much more rea- 
 sonable to suppose that these waters are maintained at their higher 
 level (if it be true) by the hydrostatical pressure which exists at the 
 equator, acting upon the columns or masses to the north of it. 
 
 The outflow from this ocean, by the surface stream on the African 
 coast, is very limited. This, together with the evaporation, is the 
 only discharge we know to occur ; consequently, the hydrodynamical 
 movement must be small. The northern part of this sea maintains a 
 higher and more equable temperature, throughout the year, than the 
 Atlantic or Pacific Ocean. Surface currents result from the monsoons, 
 which alternate with those winds. 
 
 The width of the Leeuwin submarine stream, abreast the west coast 
 of Australia, does not exceed 150 miles. The first movement of the 
 waters is noticed on the 50th parallel of S. latitude, and 100° E. 
 longitude ; and the tendency of the waters of the Antarctic Ocean 
 is from the west towards this locality.* 
 
 The third and last great southern submarine polar stream is that 
 which sets on Cape Horn, which divides it into two branches, both of 
 great velocity and extent. The one passing to the east into the South 
 Atlantic is well known — its route is around Staten Land to the east 
 of the Falkland Islands, and from thence it flows northward. This 
 has been recognized by its low temperature, as well as by the con- 
 veyance of icebergs to the latitude of 39° S. and longitude 54° W. 
 The velocity of the surface stream, which corresponds with the sub- 
 marine in direction, has been found to be from 60 to 100 miles in 24 
 hours ; but the submarine stream, as shown by the drift of the ice- 
 bergs, does not flow beyond 20 miles a day. These, after passing the 
 Falkland Islands, take a more northerly direction. The floating fields 
 of ice spread over a large extent of the South Atlantic, to the east- 
 ward of the Falkland Islands, but do not attain so low a latitude as 
 the bergs. The influence of the low temperatures of this stream is 
 perceived as far north as the latitude of 20° S. ■ — showing that it 
 
 * As a part of this stream flows on the east side of Australia, towards New 
 Zealand, the low temperature has been found beneath, and the surface current 
 exists to the north of New Zealand — setting towards the northeast, at times, with 
 some velocity. 
 
12 THEORY OF THE 
 
 underlies the current on the coast of Brazil, which sets to the south. 
 This -would give it a length of some 2600 miles, and its width off 
 Cape Horn cannot exceed 120 miles; nor is it believed that it spreads 
 as it flows towards the low latitudes in the Southern Atlantic. This 
 polar stream originates, or is first observed, in 55° S. latitude, and 
 100° W. longitude. 
 
 That part of this stream which has been called the Chilian, seems 
 to contract and be deflected as it reaches the coast of South America. 
 About 48° S. latitude, it impinges strongly upon the coast, in the 
 neighborhood of the Chiloean Archipelago. Thence it flows along 
 the coast to the north, and in no part exceeds 100 miles in width. 
 Its velocity is from -J to l-£ mile per hour, its flow appears to be 
 regular, and the surface and submarine streams coincide in direction 
 and velocity. Icebergs have sometimes been seen drifting in it, but 
 vast fields of ice are more commonly encountered. The submarine 
 stream is found off the island of San Lorenzo, at about 12° S. lati- 
 tude, and is here 80 miles wide. It thence flows to the Gallipagos 
 Islands, and invests the whole group. The temperature of the water 
 on the south side is from 10° to 15° below that of the surrounding 
 ocean. The obstructions offered by these islands cause the water to 
 accumulate and be elevated, and to flow off to the west with increased 
 velocity, for a short distance. This great stream pervades the ocean 
 to the distance of 8500 miles, and is continuous; and there can be no 
 doubt of its being urged forward by a constant force. At times the 
 surface current ceases, or is turned aside, and then commences again, 
 probably by the action of the submarine stream, which continues to 
 flow onward, and gradually communicates its motion again to the waters 
 of the surface. 
 
 I have thus drawn attention to the four great submarine streams, 
 and their branches, of which we have any knowledge ; and have 
 traced them to their furthest limits. At their termini, they all show 
 the violent conflict of the submarine streams before spoken of, except 
 that of the Indian Ocean, which is much less apparent. 
 
 There is a large part of the Pacific Ocean where no submarine 
 streams have yet been discovered. I have reference to that portion 
 of the North Pacific lying between the Hawaiian Group and Kurile 
 Islands, and extending to the north of the Aleutian Islands, and 
 beyond 50° N. latitude. There can be no doubt, however, from the 
 configuration of the continents, that the line or curve of temperature 
 of maximum density must exist near this parallel ; that the extent 
 of the zone is here reduced to less than 1500 miles in width ; and 
 that, though we have no knowledge of the flow of any large submarine 
 stream, yet some of lesser magnitude must exist ; for the isothermal 
 
CIRCULATION OF THE OCEANS. 13 
 
 lines of less temperature, in the North Pacific, attain a lower latitude 
 than those of any other ocean. Few experiments to ascertain the 
 deep-sea temperature or submarine flow have been made within tbis 
 space; but those we have are reliable. They assure us that there is 
 a constant tendency of the water towards the tropical regions, either 
 by the route of the coast of California, or through the straits on 
 the west side of the North Pacific* 
 
 The warm surface stream which flows along the coast of Japan from 
 the equatorial regions to the northeast, is called by the Japanese the 
 Kuvo Sivo. It probably originates on the east side of the Philippine 
 Islands, thence crosses the entrance to the China seas, and flows 
 onwards with a velocity of from twenty to thirty miles a day. The 
 surface current is often interrupted by those caused by the winds, 
 setting at right angles to its course into the China seas. Opposite 
 the straits of Sangar, it makes a curve along the Aleutian Islands to 
 the eastward, and finally disappears on the coasts of Oregon and 
 California. 
 
 It has been before remarked, that the normal hydrostatical equili- 
 brium is changed by the centrifugal force raising the fluid mass into 
 the form of a meniscus. This meniscus, united with the volume 
 beneath the normal level, and bounded by the curve of density, having 
 the figure of a crossed lens, may be considered as constituting a series 
 of columns, as represented in the diagram, whose height diminishes 
 from the equator, north and south, to the zones of maximum density, 
 resting upon and united at the base, and obeying the laws of fluids. 
 Now if, from the effects of evaporation or the outflow of the warm 
 surface currents, the altitude of any of these columns should be 
 changed, a change must necessarily ensue successively in all the 
 columns, and give rise to a transmission of oscillations or waves, fol- 
 lowed by a hydrodynamical movement of the ocean waters towards 
 the equatorial regions by the submarine streams beyond the tropics. 
 From the amount of these actions, a correct estimate may be formed 
 of the extent of the others ; and where one is perceived, it necessa- 
 rily leads to the conclusion of the existence of the other. We have 
 strong grounds, therefore, for believing that a submarine polar stream, 
 equal to the evaporation and outflow, must exist in the North Pacific, 
 although none has yet been detected. 
 
 The breakers, boilings, rips, bubblings, and counterflows, with 
 
 * A cold current is reported to set with some velocity through the straits of 
 Sangar, and is again met with in the passage between the island of Formosa and 
 the coast of China. This latter, probably, passes beneath the Kuvo Sivo, or 
 Japan current. 
 
14 THEORY OF THE 
 
 overfalls, are also found in the Pacific, showing the conflict of sub- 
 marine streams, although they are less violent than those we have 
 witnessed in the Atlantic Ocean. The sea lying between the Galli- 
 pagos Islands and the coast of South America, is represented by all 
 navigators as the seat of these commotions. They are also found on 
 the west of the Pacific, beyond the Marian Islands, and are met 
 with in the Indian Ocean, near the south point of Madagascar. In 
 all cases, fluctuations of several degrees of temperature are observed. 
 
 There are some other features which should claim our attention, as 
 tending to prove the existence of these submarine streams. One is 
 the peculiar form of the continents exposed to them, and the abrasion 
 they have undergone. It will be perceived that all the capes and 
 headlands of the continents exposed to the submarine streams have a 
 peculiar form — a steep and iron-bound coast ; while those along 
 which the surface streams flow, are lined by extensive banks and 
 shoals of sand. Where the two are found to meet, the prevalence 
 of the one or the other may be indicated by the description of the 
 bottom. In the one case, they are of mud, or soft; in the other, of 
 sand, or hard. Where they meet the deposits are greater, and the 
 banks or shoals more extensive. 
 
 I have before (in the Narrative, vol. ii.) mentioned the oscillation 
 of the ocean, and will now endeavor to account for it. The changes 
 in the normal level which take place through the action of the op- 
 posing forces, evaporation and outflow, produce, as before remarked, 
 an oscillation as often as any sudden effect takes place. It will then 
 communicate a wave to each succeeding column, without any progres- 
 sive motion of the fluid ; but if obstacles should intervene, the wave 
 being interrupted, rollers or breakers would occur. These oscillations 
 are found to be more violent at one season than another. The waves 
 come from different directions, passing, according as the disturbance 
 may have originated, to and from the equatorial regions. The oscil- 
 lations only become visible where obstructions occur. 
 
 The rollers take place periodically in some places, and are much 
 greater at those islands which are situated in mid-ocean, Ascension, 
 etc. etc. These appear to take place when the earth is in her aphe- 
 lion, whilst the former occur when she is in her perihelion ; but we 
 require further and more extensive observations to ascertain if this 
 be the fact. 
 
 The remarkable oscillations mentioned, in the Narrative, Nov. 7th, 
 1837, at the Samoan and Hawaiian Groups, were quite different from 
 the rollers. They were found to be coincident with a severe earth- 
 quake which happened in Chili at the same time, and to which they 
 
CIRCULATION OF THE OCEANS. 15 
 
 have been ascribed. The velocity of the motion, and the action and 
 reaction which took place, prove that such oscillations of the ocean 
 do occur ; and when they are seen, we believe that close observation 
 will detect a difference, and assist in discovering the cause of the 
 phenomena. These oscillations precede the hydrodynamical move- 
 ment, and point to a change of pressure at the polar or equatorial 
 regions. 
 
 There is another phenomenon connected with these waves, which 
 is the elevation of the waters of the Pacific Ocean above their normal 
 level, and that it must have remained so for a period of time. All 
 my observations upon the coral islands have satisfied me that they are 
 undergoing alteration by abrasion from the ocean waves,* and that, 
 at times, many islands have been submerged, so as to have appeared 
 as reefs, and been so described. The wrecks of vessels have been 
 found lying on the tops of islands some twelve feet above high water- 
 mark, and many large trees have been noticed lying from twelve to 
 fifteen feet above the highest tides. The coral rocks show everywhere 
 the effects of abrasion, and large quantities of heavy coral slabs are 
 piled up eight and ten feet above the usual water-marks, on the high- 
 est part of the low islands ; and distinct marks can be traced on the 
 high islands, some twenty feet above the present high sea level, and 
 these are of comparatively recent occurrence. 
 
 On the whole extent of the west coast of North and South Ame- 
 rica, the phenomenon of rollers takes place periodically. Observa- 
 tions on them have been but little attended to, though the inhabit- 
 ants of the coast are familiar with them. They are imputed to the 
 tidal wave ; but this cannot be the case, for they continue several 
 days, whilst the tidal action endures but for a few hours, and only 
 during the flood. These phenomena do not strictly belong to the cir- 
 culation ; yet they are a part of the system, and show 7 the action 
 and reaction of the fluid mass, and cause some change in the parti- 
 cles of the water. It is, therefore, considered proper to include them. 
 
 The well-known Bore is conceived to be another effect of the pres- 
 sure wave in its onward progress. The oscillations which cause it 
 travel with great velocity, and, coming in contact with the inclined 
 surface of the bottom, the wave, on its deepest side, is no longer 
 balanced by the pressure on the shallowest. Then it breaks, topples 
 
 * This is a fact, although not generally admitted ; nor Is it surprising, when 
 the force of waves has been p^certained by distinguished engineers to be, in the 
 German Ocean, 1} ton to a square foot; and it is supposed to be double this quan- 
 tity in the Atlantic. In the vast Pacific and Indian Oceans, it may be considered 
 Btill greater. 
 
16 THEORY OP THE 
 
 over in a vast breaker, and rushes on -with great and overwhelming 
 force (in the Hoogly at the rate of twenty miles an hour), and elevates 
 the water to a full tide. 
 
 Now, according to authorities,* the moving force of each of the 
 particles of water is due to the slope of its surface, and also to the 
 hydrodynamical action through the pressure. The former is the cause 
 of the surface, the latter of the submarine streams. It is evident that 
 the action of the first begins at the lowest point where the level or 
 equilibrium is first to be established ; and when the lowest particles 
 are removed, they give place to the succeeding ones, until they reach 
 the highest point or level, when the whole acquires a uniform velocity, 
 if its slope is the same. The hydrostatical pressure, undergoing a 
 change by the evaporation and outflow, is restored by the water from 
 the next adjoining column, and this in turn affects its neighbor, and so 
 its action passes on to the zones. By this a submarine stream is 
 established, which becomes constant, although its strength and velo- 
 city may vary with the amount of evaporation and outflow. The flow 
 of water through water is retarded by friction against its particles, and 
 by its own viscidity. 
 
 We again recur to the system of circulation. It has been shown 
 that a great part of the ocean is elevated, by the centrifugal force, 
 above the normal level ; that it assumes the form of a meniscus ; that 
 the fluid becomes elevated on the western side of all the oceans, in 
 consequence of its accumulation there by the rotatory movement of the 
 earth from east to west ; and that it preserves this elevation, from which 
 the outflow passes off towards less elevated parts lying without the 
 tropics, to the north and south, on the eastern side of the continents. 
 In the North and South Atlantic, the Pacific, and Indian Oceans, 
 it is analogous; but the South Atlantic and Indian Oceans have the 
 simplest form. We find in all a large central space, lying north and 
 south of the equator, which remains comparatively undisturbed, and 
 around which the cold and warm streams circulate. This space main- 
 tains almost an equal temperature throughout the year; and as far as 
 our experiments have gone, no submarine stream is known to exist 
 within them. Between the surface and 200 feet, the difference of 
 temperature continues uninterrupted. The boundaries which those 
 spaces occupy, and the area they embrace, are as follows : 
 
 * Du Buat and others. 
 
CIRCULATION 1' THE OCEANS. 17 
 
 Latitude. Longitude. Area in sq. milea 
 
 North Atlantic 15° to 35° N 35° to 65° W. 
 
 South " 20 "30 S 10 E. " 55 W. 
 
 North Pacific, 20 " 33 N 95 W. " 145 B. 
 
 South " 20 " 35 S 95 W. " 170 W. 
 
 Indian Ocean 18 "35 S 55 E. " 100 E. 
 
 1,800,000 
 1,950,000 
 4,680,000 
 3,370,000 
 2,295,000 
 
 The North Indian Ocean has no regular circulation : its surface 
 currents depend upon the alternations of the monsoons. 
 
 We must not fail to observe that, in the Northern Hemisphere, the 
 ocean circulates from left to right ; while in the Southern its move- 
 ment is from right to left. There are connecting currents between 
 each of these circulations, but they form a very small part of the 
 gyration. The drift currents, produced by the winds, are at times 
 felt over these spaces, but are not to be considered as forming a part 
 of the great and constant circulation. Their flow is but temporary, 
 and in the direction of the wind. 
 
 The expanse of water in the North Atlantic, which is strewn over 
 by the fucus nutans, has been most generally conceived to be a sort 
 of eddy, where other matters, besides the Sargasso weed, are being 
 deposited. This is unquestionably a mistake ; for there is little or 
 no driftwood, or other terrestrial products, found within it, and the 
 weed is young and fresh — a fact which has given rise to the opinion 
 that it must be near its place of growth. The tendency of the sur- 
 face is from this area, which would indicate that it had a higher level. 
 The same remarks will apply with equal truth to the areas which 
 have been pointed out in the other oceans. No doubt, these great 
 revolving streams attract and draw off a portion of the waters of 
 these spaces, to join in the flow as it circulates. At times, driftwood 
 may be carried by an eddy and left within these spaces, and retained 
 there for a long period of time. 
 
 The cause which governs the warm surface streams to flow from 
 the equatorial regions towards the poles, is that they occupy a higher 
 level than the colder waters lying without the tropics, and, in obedience 
 to gravity, they flow off to restore the equilibrium. 
 
 When a surface stream is perceived, there can be no doubt that its 
 particles are descending an inclined plane to a lower level ; and this 
 must continue until the level is restored, either by filling up, or by 
 reduction in height; and should the elevation of the water change to 
 an opposite side, the flow would in like manner return. This we see 
 exemplified in the tides, which we daily witness. 
 
 Another cause of the flow of water (to which I have adverted 
 before) is lying above and being in contact with the flow of a sub- 
 
18 THEORY OF THE 
 
 marine stream. So situated, it would necessarily acquire the motion 
 of the fluid on which it rested, and would attain some of its velocity. 
 This effect sometimes causes a flow to begin ; and, if unimpeded by 
 any surface current, it would acquire a part of its velocity, and the 
 direction of the submarine stream. 
 
 The phenomenon of submarine streams causing an accumulation 
 when obstructed, or encountering a shallow coast, is well known. 
 This elevates the ocean in those places, and a current flows off on 
 either side, with greater velocity than the parent stream previously 
 had. This effect is witnessed in any running brook. The low tem- 
 perature of the water, when compared with that of the surrounding 
 ocean, satisfies us that it must be caused by a submarine stream. 
 As the distance from these obstructions increases, the velocity de- 
 creases, and is finally lost. There are many localities where this 
 effect is perceived ; viz. at the Cape de Verde Islands, on the coast 
 of Brazil, near Cape St. Roque, in the North Atlantic, around the 
 Gallipagos Islands in the Pacific, and at all islands and shoals situated 
 or lying in the route of the submarine streams, the temperature 
 of the ocean is found to descend several degrees. This effect of a 
 change of temperature on shoals or banks, has been attributed to 
 radiation of heat, in consequence of the decrease of depth ; but it 
 is inadequate to this end. This phenomenon does not appear where 
 submarine streams do not exist. 
 
 In consequence of there being a greater flow from the Southern 
 Hemisphere by the three great submarine streams, the line of the 
 greatest heat of the ocean lies to the north of the equator, except on 
 the western side of the Pacific, where the southern submarine streams 
 do not reach, from the interruption of the lands lying to the south ; 
 and although we have not found, as yet, any northern submarine 
 stream in the North Pacific, yet it gives another reason to suspect 
 that one exists. 
 
 The counter flows I have called eddy currents: they appear to have 
 well-defined limits in all the oceans. The one of greatest extent is 
 that of the Pacific Ocean, which runs between 2° S. and 9° N. 
 latitude, and from 130° E. to 120° W. longitude — a distance of 5500 
 miles. The eddy in the Atlantic runs between the equator and 5° 
 N> latitude, and from 40° W. to 15° E. longitude. Part of this is 
 known as the Guinea Current. Its western portion is intermittent, 
 but the Guinea Current is constant and has a great velocity, termi- 
 nating in the Gulf of Biafra, preserving a width of from 180 to 200 
 miles, and flowing on a parallel, though in an opposite direction, with 
 the Agulhas, or Great South Atlantic Polar Stream, exceeding the 
 
CIRCULATION OF THE OCEANS. 19 
 
 latter in temperature 8° to 10°. Its greatest velocity is from three 
 to three and a half miles an hour. Its temperature in the middle 
 is 84° ; on its outer edge, 82° ; hut next the land from 79° to 80° 
 Fahr. 
 
 These eddy currents are not, however, to be considered as origina- 
 ting from the action of the flow of the adjacent waters on each side 
 of them, caused by the trade-winds, but rather from the accumulation 
 of water on the western side of the ocean, from whence they flow, from 
 a higher to a lower level, in obedience to the laws of gravity. The 
 Pacific eddy pursues a course to the east, cutting the meridians and 
 equator at a very small angle. Its width is about 300 miles, and the 
 drift currents caused by the trade-winds, with a variable velocity, run 
 nearly parallel to it. The eastern part of this eddy has been de- 
 signated, by some hydrographers, as a counter equatorial current. 
 
 Of all the surface movements of the ocean, the Gulf Stream has 
 excited most interest ; and many have devoted their attention to the 
 investigation of its phenomena. Various causes have been assigned 
 for it, and many hypotheses have been advanced. Some think that 
 the flow of the great rivers which discharge into the Gulf of Mexico 
 and the Caribbean Sea, give rise to it; but a consideration of the 
 supply derived from these sources, soon dissipates the idea of the in- 
 adequacy both in volume and velocity. By others the fountain-head 
 has been considered as lying in the Southern Hemisphere, coursing 
 through currents which discharge themselves into the Caribbean Sea 
 and Gulf of Mexico. It may readily be shown that such is not the 
 case. 
 
 The elevation of the equatorial seas above the normal level, by the 
 forces already referred to, is adequate to preserve a constant higher 
 level of the waters within, above those without the tropics. The Gulf 
 of Mexico is connected with the Caribbean Sea, and the latter with 
 the Atlantic Ocean, by many deep inlets. These waters are supposed 
 to rest on the same base, and are maintained at the same level by 
 the hydrostatical pressure, and must preserve the same level as the 
 meniscus. There is no constant stream flowing into the Gulf of 
 Mexico from the Caribbean Sea, between the peninsula of Yucatan 
 and island of Cuba ; nor is there any current, but the tidal current, 
 setting round its shores. The whole Gulf of Mexico is to be regarded 
 as an immense reservoir, and its waters are maintained at their high 
 level by the hydrostatical pressure of the equatorial seas. The Carib- 
 bean Sea may be regarded also as an elongated basin. Both are of 
 great depth ; and its waters, like those of the equatorial seas, stand 
 above the normal level. The waters of the Gulf of Mexico are dis- 
 
20 THEORY OF THE 
 
 charged on the east through the channel of the Gulf Stream, at the 
 straits of Bimini. They also escape by a smaller current flowing to 
 the east on the south side of Cuba. The pass over the ledge of 
 Bimini, then, is the spout or outflow from this great reservoir — which 
 has been called the Gulf Stream, and its flow is great — under the 
 law that the pressure outward is equal to the weight of a column 
 whose base is the whole surface, and whose altitude is equal to the 
 depth of its centre of gravity below the surface. 
 
 From recent investigations by the officers engaged on the Coast 
 Survey duty, it has been ascertained that in the outlet of the Gulf 
 of Mexico, the depth on a section at the Bimini Straits is but 370 
 fathoms. There it is thirty-six miles wide. The outflow passes off 
 through this section. The difference in the level of the waters lying 
 to the north, and those of the Gulf of Mexico, permits it to do so, 
 and the difference in both will represent the velocity of the stream. 
 At times, this is much more rapid ; and, again, the current has been 
 known to cease or to set into the Gulf. The peculiar and sharp turn 
 to the north taken by this stream, may be occasioned by its meeting 
 with resistance from the banks and shallow water, which, by the facts 
 before alluded to, may be elevated as high as the waters of the Gulf. 
 Its pressure would then act to the northward, and cause it to be turned 
 towards the coast of the United States, which it follows with a de- 
 creasing velocity, spreading its waters over a wide area of the North 
 Atlantic, to the north of the space where no current exists ; and only 
 ceases to flow when the level is restored. 
 
 The Caribbean Sea maintaining the same level as the Gulf ot 
 Mexico, from the same causes, this sea has also a very extensive cur- 
 rent on its south side, from the peninsula of Yucatan along the coast 
 of Honduras and Columbia, through a distance of over 700 miles. 
 It has likewise frequent outflows through its outlets to the north. 
 Observations prove that its waters flow, at times, with velocity to the 
 eastward, even against the prevailing winds. It will, therefore, be 
 seen that the equatorial stream does not flow through this sea, but 
 is lost (as I have before remarked) in its conflict with the great north- 
 ern stream. There is, however, a very strong current known to set 
 from Cape St. Roque to the west, under the name of the Guyana cur- 
 rent ; but this is imputed to the elevation, from the accumulation of 
 the waters by the South Atlantic submarine stream, on the shoal 
 coast of Brazil. The velocity of this current is very variable, though, 
 when greatest, it equals the velocity of the Gulf Stream ; but its course 
 is a short one, and it rapidly decreases, and ceases to flow. 
 
 The Gulf Stream offers the best illustration of a warm equatorial 
 
Bottom. 
 
 Depth Fatb. 
 
 34° .. 
 
 .... 370 
 
 49 .. 
 
 .... 500 
 
 60 .. 
 
 .... 510 
 
 59 .. 
 
 .... 540 
 
 54 .. 
 
 .... 540 
 
 CIRCULATION OF THE OCEANS. 21 
 
 surface stream. From what has been said, it must be governed by the 
 Bame laws which control the flow of liquids. 
 
 Some anomalies, however, connected with the temperature of this 
 stream, and which are recorded in the reports of the Coast Survey, 
 must be referred to. By the observations on the section of this stream 
 across the Straits of Bimini, where it was found that the greatest 
 depth was 370 fathoms, the temperature at that depth was but 34° 
 Fahr., while that of the surface was 84° Fahr. Further to the north 
 we have other sections crossing this stream. The observations are 
 presented in the following table : * 
 
 Surface. 
 
 In the Straits of Bimini 84° 
 
 Off Cape Florida 83 
 
 Off St. Augustine 82£ 
 
 60 miles to the north of St. Augustine 82 
 
 123 miles further to the North 81 1 
 
 It will be observed, by the above record, that the temperature has 
 increased, above that at Bimini, but below the depth of that section, 
 to the distance of 420 miles to the north, 21J° of Fahr., and this 
 extends over an area of more than 8000 square miles, and exceeding 
 it 900 feet in depth; while the surface has expanded itself fully forty 
 miles in width. Now, it cannot be supposed that this vast volume of 
 heated water, existing below the depth of the Straits of Bimini, could 
 have been derived from the Gulf of Mexico, or that it has passed over 
 this ledge. If these results of the Coast Survey are to be depended 
 upon, it does prove that the heat which raised this vast volume of 
 water so many degrees, could not have originated from the equatorial 
 regions, and must have acquired its heat after it had passed the bar- 
 rier where the temperature of the water was found to be only 34° 
 Fahr. at a less depth. 
 
 The temperature of the waters of the Gulf of Mexico have been 
 ascertained to be about 44° Fahr. at the depth of 1000 fathoms. From 
 this depth the water may rise and flow out over colder strata; but we 
 are constrained to doubt that, after passing that strait, where so low 
 a temperature is said to be found, it should again, contrary to the 
 laws of the specific gravity of fluids, descend to occupy a lower level, 
 with a temperature so much higher. The same records, as before 
 observed, give evidence that the warm water of the Gulf Stream, in 
 its progress to the north, spreads to a great width. They also assert 
 that the cold temperature of the depth on the Bimini section, is a 
 
 * See Coast Survey papers for 1855, p. 55. 
 
22 THEORY OF THE 
 
 proof that a colder stratum of water underlies the Gulf Stream. Such 
 does not appear to he the case, from the facts of the record, which 
 offer the most satisfactory proof that can be adduced, that the heat 
 must have emanated from the earth by the intestinal currents. All 
 analogy would forbid us to assign its origin to the Gulf Stream, or 
 to suppose that its waters, which are specifically lighter, could flow 
 under those that are denser and colder. 
 
 The separation of the cold and warm-water streams are always well 
 defined, and in many localities more so than the Gulf Stream. I 
 refer to the equatorial stream and the Guinea current. The latter 
 I have classed under the eddy currents. These run side by side, in 
 opposite directions, for upwards of 1000 miles, and preserve through- 
 out a well-defined margin, without commingling. This is probably 
 owing to the intestinal currents, which, ascending and descending, serve 
 to prevent it. Experiments, on a small scale, exhibit this effect, and 
 show distinctly that the movements of the ascending and descending 
 currents are well defined, and have the same well-marked line of 
 separation. 
 
 The excessive heat to which the bottom of the ocean is subjected, 
 must act upon the earth's strata, and comminute its particles. The 
 specimens of the bottom brought up by the deep-sea soundings 
 made by Lieut. 0. H. Berryman, U. S.N., were examined by the late 
 Prof. Bailey, of West Point. His investigations showed that the 
 bottom of the ocean was composed of microscopic Crustacea, which he 
 believed to be generated in the localities from which they were raised. 
 They present organic forms of great delicacy and fragility of struc- 
 ture, and it is believed that they could not be produced without either 
 light or heat. We know that none of the former can reach them, and 
 therefore conclude that their generation depends upon the latter. The 
 gases for these organic forms, if required, would, from the enormous 
 pressure, be found in their liquid state. 
 
 Although I am convinced that there is no better mode of ascer- 
 taining the existence and course of the Oceanic Circulation, than by 
 thermometrical observations on the surface and in the depths of the 
 sea, I feel satisfied that we have, as yet, no reliable instruments for 
 making those observations with precision. 
 
 The instruments for recording the temperature of the great depths 
 possess much ingenuity. Saxton's metallic thermometer, in use by 
 the Coast Survey, is, when carefully used, one of the most perfect ; 
 but very many of its results are discordant and evidently erroneous. 
 Those obtained by Lieut. 0. H. Berryman, when engaged in making 
 the soundings across the Atlantic Ocean for the telegraph, are ex- 
 
CIRCULATION OF THE OCEANS. 23 
 
 tremely discordant, varying, at the same depths and nearly in the 
 same localities, from 30° to 24°, 20°, 15°, 10°, and 5° helow the 
 freezing-point. In some instances, the record shows twice the tempe- 
 rature at the same depth. Lieut. Berryman reports having great 
 confidence in Saxton's instruments, and believes that every reliance 
 can he placed in them. He cites instances where two have been used, 
 in taking the deep-sea temperature, giving the same result, within half 
 a degree of each other. From his experiments and those of others, 
 after a careful examination, I think it will be evident, that these in- 
 struments are not to be relied on when subjected to great pressure, 
 and are liable to become deranged by any concussion, even that caused 
 by the lead to which they are attached in striking the bottom. On 
 comparing this with Six's glass self-registering thermometer, they 
 have been found to agree, when the depths are less than 300 fathoms. 
 It is to be desired that some reliable test should be made, and the 
 defects obviated by its ingenious inventor, so as to avoid, if possible, 
 disturbance by pressure or concussion. 
 
 It is hoped that investigations will be directed to the deep-sea tem- 
 peratures, in which little progress has as yet been made — attention 
 having rather been given to obtain a knowledge of the actual depth 
 of the ocean — a result of comparatively little moment. Sufficient 
 has been done to show the direction we should persistently follow, 
 by making a regular series of observations on the temperature at the 
 surface, and at depths of from 100 to 500 fathoms, on parallels 5° 
 apart, and on meridians from 50 to 100 miles asunder. This could be 
 effected at little expense, as our national vessels pass to and from 
 the foreign stations. In the year 1846, the then Secretary of the 
 Navy, at my request, ordered the steamer Missouri, bound for the 
 Mediterranean, to be provided with suitable instruments, to make a 
 series of observations on the surface and deep-sea temperatures in 
 crossing the Atlantic ; but they failed to be carried out or to receive 
 any attention, in consequence of the haste to convey our Minister 
 to China, who took passage in that vessel. We trust that other attempts 
 may be more successful, and results obtained. 
 
 In reviewing the foregoing remarks on the circulation of the oceans, 
 we cannot fail to be struck with the order, uniformity, and system 
 which everywhere prevails. We perceive, on the one hand, a great 
 body of water flowing onward from right to left, and another, in an 
 opposite hemisphere, equally large and rapid, making its gyration from 
 left to right — deep seas giving out their waters almost continuously, 
 while an indraught is constantly found to set into shallow seas. Not- 
 withstanding these apparent anomalies, which, at first view, appear so 
 
24 CIRCULATION OF THE OCEANS. 
 
 contradictory, they all go to prove the truth of the laws which govern 
 and harmonize the forces we have cited in our explanation of the 
 general circulation. In thus accounting for the many phenomena by 
 these well-known forces, the pressure and movements of fluids accord- 
 ing to their established laws, joined with the influence of heat, we 
 are under the belief that they must be considered fully adequate to 
 the end of producing a constant and uninterrupted flow of the waters 
 by which the circulation is establish-ed and maintained. 
 

 b 
 
LIBRPRY OF CONGRESS 
 
 029 714 196 P *