136 BULLETIN: MUSEUM OF COMPARATIVE ZOOLOGY. 
of observation and deduction should go on together in the field, each 
aiding the other, if the investigator would avoid as far as possible the 
disappointment of finding afterwards that the field records are deficient 
in some particular point where fuller record would have been of critical 
value in testing a deduction. Memory may sometime supplement written 
record, but it is notoriously dangerous to trust to unwritten notes. In 
my own experience, however, careful deduction is more difficult than ob- 
servation in the field, but it is greatly aided by deliberate thinking and 
writing while the facts are before the eyes. 
Second, it must not be assumed that a theory gains support because 
its consequences can be definitely deduced. However accurately one 
may argue out the details of form appropriate to a certain stage in the 
dissection of a faulted mountain block, the theory of block faulting be- 
comes a demonstrated occurrence only when the sharply deduced conse- 
quences of the theory are shown to accord with closely determined facts 
of observation. 
Not only is it important that an investigation should give equal at- 
tention to deduction and induction ; it is essential to clear presentation 
that both phases of inquiry should be sufficiently published. It is 
otherwise almost impossible for the reader to discriminate between sound 
and unsound conclusions. It is conceivable that an able observer should 
patiently collect and record a multitude of facts, and that he should very 
imperfectly set forth the reasons that lead him to the announced expla- 
nation of the facts. The hurried reader may in such a case quote the 
announced explanation and accept it, if he wishes, on the authority of 
the writer; but the more critical reader will wish to make his own 
measure of the validity of the announced conclusion, and this he will find 
difficult in the absence of explicit announcement of the method of reach- 
ing it. It is particularly important to consider the deductive side of 
any problem in which there is substantial agreement among different 
observers as to the facts directly observable, but in which there is differ- 
ence of opinion as to the explanation of the facts; for in such a case 
the correct solution of the problem turns essentially on the validity of 
the deductions by which the unobservable facts of the past are brought 
into mental vision. 
It may not be amiss to point out that the investigator’s effort in all 
such problems as the one here in hand is simply to supplement the di- 
rectly observable present facts by the discovery of the unobservable past 
facts, so that the entire phenomenon shall become known. If observers 
of sufficient penetration had been present in the Great Basin during all 
