592 



NA TURE 



[April 19, 1894 



stick of ebonile near it, without producing any effect on the 

 pointer. 



Again, these two clear glass shades, belonging to gold-leaf 

 electroscopes, are one of them coated with our varnish, and the 

 otht;r not. Which is the uncoated one is at once apparent from 

 the alteration produced in the deflection of the gold leaves when 

 I approach a stick of rubbed ebonite near the lower part of the 

 glass shade of the uncoated one, for no such change in the 

 deflection is produced, as you see, when the ebonite rod is 

 broughi near the glass shade, which is protected by a layer of 

 this varnish applied on the inside. 



THE CURRENTS IN THE GREAT LAKES OF 

 NORTH AMERICA. 



A 



PAPER, entitled "The Currents of the Great Lakes," 

 prepared by Prof. Mark W. Harrington, from data collected 

 by means of bottle-papers during the navigation seasons of 1892 

 and 1S93, has just been published by the U.S. Department of 

 Agriculture as a Weather Bureau Bidldin, We reprint some 

 of the more interesting parts of the paper, and reproduce a map 

 showing the results of the inquiry. 



Early in 1892 the Weather Bureau published a wreck chart of 

 the Great Lakes, prepared in the winter of 1891-92. The wrecks 

 noted on this chart were only those due to meteorological agents, 

 and a striking feature of the chart was the clustering of wrecks 

 in certain parts of the surface of the lakes. This suggested that 

 unknown currents might play a considerable part in wreckage, 

 and steps were at once taken to get some idea of what these 

 currents are. 



The method pursued was that of bottle-papers, which have 

 frequently been used to study ocean currents, but had not been 

 employed in the lakes, and is as follows : — A bottle, containing 

 a paper on which is written the time and place of floating, is 

 thrown overboard at some definite point, left to float freely, and 

 when picked up the enclosed paper is marked with the time and 

 place of finding. In this way two points in the line of current 

 are obtained, and by considering a very large number of these, 

 satisfactory conclusions can be drawn as to the currents which 

 convey the bottles. 



A large number of bottles were specially made for this pur- 

 pose, with the name of the Bureau blown into the glass. They 

 were of an unusual colour, but the contents could be easily seen 

 by anybody who picked them up. The weight of a bottle was 

 about 420 grams ; total external displacement, 460 cubic centi- 

 metres ; volume displaced when floating, 430 c.c. ; volume 

 exposed above water, 30 c.c. 



From the position of flotation it appears that enough of the 

 bottle was above the surface of the water to give the wind some 

 power in drifting it. This would probably make little difference 

 with the direction of the drift, for the wind that drifted the 

 bottles would drift the surface water in the same direction. It 

 may have made some difference in the speed with which the 

 bottles travelled, making them move, perhaps, faster than the 

 water, but this effect would be slight. Within each bottle a 

 franked envelope was placed, addressed to the chief of the 

 Weather Bureau, at Washington. Before a bottle was thrown 

 overboard a blank form in the envelope had to be filled up, 

 giving the name of the vessel and its captain, the date of float- 

 ing, and the place where floated. Another space was left lor 

 the finder to insert similar data. 



The bottles actually picked up were, for the most part, on 

 shore, very few of them having been found in the water. It is 

 impossible to say what proportion of the bottles was recovered, 

 but it was not great. Though probably more than five per cent., 

 it did not exceed ten per cent. The figures cannot be given 

 exactly, because it is not known how many of the bottles are 

 still in the hands of masters of vessels. A considerable portion 

 of the papers recovered was found on the Canadian shore, and, 

 curiously, a large number of them came from shores which are 

 for the most part uninhabited. 



The investigation covers the years 1892 and 1893, but it 

 must be remembered that the observations could be taken only 

 during the season of navigation. ' This practically limits the 

 conclusions to the summer months, as when the bottles are 

 floated in the spring they will probably be found in the autumn, 

 and those that are floated in the autumn will be lodged in ice, 

 and their routes be variously changed, so that trustworthy con- 

 clusions cannot be drawn from them. In general, the finds of 



NO. 1277, VOL. 49] 



the latter sort— that is to say, the finds of bottles floated in the 

 autumn and picked up in the spring — have been left out of con- 

 sideration. The currents shown in the accompanying map are 

 herefore those of the season of navigation, and practically the 

 currents of summer. 



Classification of the Currents. 



The currents in the Great Lakes can be grouped under the 

 four following heads : — 



(i) The Body Currents.- — These lakes all have an outflow, and 

 there must be a general motion of the water towards the outlet. 

 This is visible upon the map of each lake, and the currents 

 which result from it must be continuous throughout the year, 

 and must affect most of the water. 



(2) A Surface Current due to the Prevailing Winds. — That 

 the winds have great effect on the currents in large bodies of 

 water is widely recognised, and the more constant they are the 

 more marked is the effect. The westerly winds, in case of lakes 

 lying nearly east and west, cause a surface current from the 

 west which is in the same direction as the body current. In 

 the case of the lakes which lie across the direction of the wind, 

 the surface drift is from the west across the lake. The details 

 of the direction, however, depends on where the outlet is, on 

 the form of the lake, and on the position of the inlet. 



(3) The Return Currents. — It will be seen from the illus- 

 tration that, in the case of three of the lakes, the main currents 

 hug one shore. In the case of Lake Superior, it is on the 

 southern shore ; in the case of Lake Michigan, it is on the 

 eastern shore ; and in that of Lake Huron, it is on the western 

 shore. Lakes Erie and Ontario do not show this phenomenon 

 so plainly. This feature can be explained by the two sorts of 



I currents already mentioned, combined with the lay of the lakes 

 I as to the prevailing direction of the wind and the position of 

 the outlet. In any case, however, the drive of the v/ater from 

 one end of the lake to the other necessitates more or less a 

 return current, providing the outlet is not sufficiently large to 

 allow this water to pass through. In the Great Lakes, the out- 

 lets are comparatively small, hence in all these cases there are 

 return currents. 



(4) Surf Motion. — Owing to this motion, the bottles have 

 been found to show a decided tendency shoreward whenever 

 they came within its vicinity, and especially so when the water 

 was shallow. 



Velocities of the Currents. 

 The directions of currents can be ascertained with much more 

 precision by means of bottle-papers than can the velocities. It 

 has therefore been very difficult to arrive at any satisfactory 

 conclusions concerning the speed of the currents in the Great 

 Lakes. In a general way, the speed appears to vary from four 

 to twelve miles a day. In a few special cases, very much 

 higher velocities have been found, but these are probably due 

 to surf motion rather than the motion of the surface water as a 

 whole. It is not at all improbable that the general surface 

 motion of the lakes has a higher velocity than from four to 

 twelve miles a day, but the only conclusion which it seems safe 

 to draw from the data is that the velocities are at least as high 

 as the figures mentioned. 



Lake Superior. 



In this lake thirty- five bottfes were recovered in 1892, and the 

 same number in 1893. From the courses of the bottles it 

 appears that there is a general surface current along the south 

 shore of the lake, from the Apostle Islands eastward, and that 

 to the east of Keweenaw Point this eastern current has very 

 great breadth. Still further eastward, toward the eastern end 

 of the Lake, it spreads out in a fan-shaped way, and a branch 

 of it seems to pass to the northward and westward, reaching the 

 extreme northern coast of the lake. A branch of this current 

 also turns southerly round Keweenaw Point, and at the bottom 

 of the bay, on the south coast of the lake, an eastern current is 

 taken up, which joins the main current to the eastward of . 

 Marquette, Michigan. The minimum velocity of the main 

 eastward current of the lake appears to be fiom four to six 

 miles a day. 



Lake Michigan. 



In this lake 163 bottles were recovered in 1892, and 

 thirty-five in 1893. The currents indicated by the floating 

 of the bottles are of unusual interest. There is first a 



