214 University of California Publications in Zoology (Vou. 8 
arises, Why the absence of such movement when other related 
species perform daily excursions? Noticing the unusual trans- 
parency of the animals, Dr. Esterly makes the suggestion that 
their relative invisibility affords them a protection not enjoyed 
by less transparent species and so does away with a need for 
migration which the others have. How is such a hypothesis to 
be tested? Obviously in no other way than by laboratory 
experiments. Again, Miss Johnson and myself have learned a 
number of interesting things about the asexual propagation of 
certain species of salpa by studies for the most part on preserved 
specimens. But these studies have brought to light a number of 
questions which, so far as we can see, cannot be answered with- 
out keeping living, growing animals under observations for con- 
siderable pe1iods of time. But such observation is impossible 
without the best of aquarium facilities—and the aquarium is 
only one part of a laboratory. 
The mutually supplementary relation between field work and 
laboratory work in such an enterprise is so obvious as to make 
dwelling upon it seem superfluous. Actual conditions and prac- 
tices, and to some extent views, do nevertheless Justify insistence 
upon the point. For one thing the great cost and difficulty of 
bringing together, duly balanced, the two kinds of work are 
serious obstacles in many eases to the realization of the ideal. 
An obstacle still more unfortunate in some ways is the well- 
defined notion rather widely held, that field studies are of quite 
subordinate importance. I believe, however, that an open- 
minded review of past and contemporaneous biology will con- 
vince anyone of the danger that lurks in overconfidence in any 
single method of research. Possibly there are nooks of science 
somewhere in which one method is enough; but if so they have 
escaped my notice. The admonitions of experience as to exactly 
where monomethodie research tends, are useful. Field observa- 
tion alone unquestionably encourages illy supported and more 
or less sentimentally colored generalization. Unsupplemented 
description, whether of organisms as wholes or of parts of 
organisms, produces results that savor more of the collector and 
cataloguer than of the whole-hearted student of animate nature. 
The laboratory too singly confided in has still greater danger 
