Superior. No attenpt has been made to study them in Lake Huron and Lake 

 Ontario, and the data on those of Lake Erie are too meager to prove their 

 existence, but there is every reason to believe that careful study would 

 reveal the tides of these lakes. Those who may be interested in the data 

 on this subject should refer to Whittlesey (1859), Graham (I86I), Comstock 

 (1372 and 1873), Ferrel (I87U), Harris (1907) Endr'is (I908), and Krecker 

 (1928). Inaddition, many of the early references given in the section on 

 seiches give valuable historical infrrmation. 



Currents 



Currents in the Great Lakes other than Lake Erie have been discussed 

 by Clark (1892-1893), Goodwin (I892), Harrington (l895) , Nasmith and Adams 

 (I91M, Kindle (19l5a and 1925), Judson (1909), Cooley (1913), Mclaughlin 

 (1912), Warmund (1927-1928), and Deason (1932). Papers concerning Lake 

 "rie particularly are: Harrington (189$), Fell (I910), McLaughlin (I9II), 

 Krecker (1928 and 1931), and Parmenter (1929). 



The most important paper for our purpose is that of Harrington (l895), 

 based on drift-bottle experiments in I892, 1893, and l89U. Of the many 

 bottles released in Lake Erie, 97 were reported found, and of these, nearly 

 one-half had been released in Western Lake Erie. With regard to the cur- 

 rent? in this part of the lake, he says: 



"At the western end of the lake the presence of Point 

 Pelee, Pelee Island, and the archipelago to the south, 

 cause certain variations, the principal one of which is 

 the tendency of a whirl about the islands, noted in each 

 of the lakes so far discussed. The numerous passages 

 between the islands existing here, and the fact that the 

 western end of the lake is nearly cut off by the point 

 and islands, together prevent the development of a clear 

 symmetrical whirl of the character found before. It is 

 very much broken up into parts, and is possibly variable. "- 



In order to obtain more data on the surface currents, a series of 

 experiments with drift bottles was carried out in May and June of 1928. 

 The bottles were fitted vath drags which tend to minimize the effect of 

 winds, and were hence more effective than the simple bottles used by 

 Harrington. Ninety-eight bottles were released and $[1 were recovered. 



For various reasons it seems unnecessary to present the deta. led 

 data on these experiments. The courses of many of the bottles are shown 

 in Fig. 6. It will be noted that, in some cases, bottles set near each 

 other were recovered at widely separated points. With few exceptions, the 

 courses taken by the bottles in the experiments could be explained by ref- 

 erence to the data on wind direction. In most cases where the explanation 

 was not evident, it was found that the bottles had been adrift a long time 



39 



