AMERICAN AND WEST INDIAN FAUNA AND FLORA. 121 
admirably traced by Alph. Milne-Edwards, in an interesting 
memoir on the fauna of the antarctic regions. He calls at- 
tention to the effect upon the distribution of penguins of 
icebergs carried by currents. These birds are excellent swim- 
mers, and fond of resting on floating ice; they are prepared 
for long voyages, and thus may have been transplanted from a 
single centre to the varied localities where they are now found. 
In their search for food they must naturally migrate to great 
distances during the long antarctic winters, when they cannot 
carry on their usual fishing avocations. The path of the ice- 
bergs, as dependent on currents, may even enable us to trace 
the original penguin home. Milne-Edwards concludes that the 
inhospitable country of Victoria was the origin of the most char- 
acteristic animals of the antarctic region. 
It will be quite possible, when sufficient material has been col- 
lected, to map out the course of the pelagic fauna ; and from its 
bottom distribution much light will be thrown on the course of 
the currents. Thus the existence at Newport of Agalma, Clio, 
and Pneumodermon, — inhabitants of arctic seas, Labrador, 
Maine, and northern New England, —is direct evidence that 
the cold arctic current finds its way round Cape Cod to the 
opening of Narragansett Bay. In the same way we may trace 
the northern course of the Gulf Stream by the presence of Sar- 
gassum, Porpita, Leptocephali, and southern pteropods, etc., 
which are carried each year to the coast of southern New Eng- 
land. The range of southern globigerine and pteropods will 
fix the eastern limits of the Gulf Stream, as northern types 
determine those of the cold current along the east coast of the 
United States. 
It is most hazardous, especially on insufficient data, to spec- 
ulate upon former land connections; and yet, in many cases, 
the presence_of identical genera of mammals seems to leave us 
no alternative. When, however, there are powerful currents 
running through the narrow channels that separate adjoining 
йай, ог ibd these currents skirt the shores of island ranges 
and of neighboring continents, it is not always necessary to 
assume such an extinet continental connection. The careful 
study of the distribution of mammals on. the islands of the 
