414 THE AMERICAN NATURALIST. | [Vor. XXXIX. 
chiefly discusses those cases, which are not in favor of the former 
land connection between these continents, that is to say, which .do 
not directly bear upon this question, while other cases, which might 
possibly furnish additional evidence are dismissed shortly. 
The general trend of Pfeffer's argumentation is as follows. There 
are a number of instances, where discontinuous distribution in all or 
some of the southern. continents is evidently a remnant of a former 
universal or subuniversal distribution, which is clearly shown by the 
presence of fossil remains of the several groups of animals in other 
parts of the world. On the other hand, there are similar cases, in 
which fossil remains are not known at all, or are not known from the 
northern continents. Here, Pfeffer claims, we have the right to 
assume, that these groups nevertheless once existed on the northern 
hemisphere, and they also once were subuniversal in their distribu- 
tion. 
Here we are again confronted with a fallacious generalization of 
correct observations, a way of drawing conclusions that has so often 
given origin to incorrect general theories. Of course, if some groups 
of animals, that are now more or less restricted, were once universal 
in their distribution, it is evident, that very likely some others were, 
of which no direct evidence of universality has been found. But the 
conciusion, that all such cases are to be explained by this assump- 
tion, is, to say the least, a little rash. Moreover, we do not intend to 
find a theory that might under certain assumptions explain the 
present facts, but we ought to try to find the correct explanation, and 
thus no way of ascertaining the latter should be neglected, and 
doubtful cases should not be dismissed shortly, but subjected to a 
careful and thorough study. 
In the present paper, Pfeffer only talks of paleontological evidence, 
and almost entirely neglects the morphological (systematic) relations 
of the different forms. It is sufficient. for him, if certain forms are 
found in Africa and South America, to point out that they might 
have been once present also in North America and Eurasia, and that 
thus a connection may be established (over Bering Sea). If similar 
forms are also found in India, this assumption seems to him beyond 
question. He does not pay the slightest attention to the mutual 
relations of these forms. 
Now it is a fact, that the former land connections between tropical 
South America and Africa, and between southern South Africa and 
Australia have been supported chiefly by studies of the degree of the 
relation of their faunas. We: know cases where members of the 
