LOW TEMPERATURE DEVELOPMENT OF FUNDULUS 477 



are tJie more likely to exhibit abnormality seem much less rec- 

 ondite. In the first place, from the mode of its formation by 

 confluence (concresence) the head end of the Teleost embryo is 

 the first part to be formed and differentiated, and any disturb- 

 ances of the normal processes of tissue and organ differentiation 

 are much more likely to be exhibited first in the region earliest 

 differentiating. And in the second place, the organs of the 

 head region, especially the sense organs^ brain and heart are in 

 general more highly and therefore more extensively differen- 

 tiated, their normal development involves the interaction of a 

 larger number of factors, than in most other parts of the embryo 

 where muscles and the connective tissues form the chief con- 

 stituents. Hence a slight initial disturbance would produce a 

 more frequent as well as more marked result in the anterior 

 part of the embryo, where a more precise arrangement of the 

 underlying conditions of differentiation is necessary, than in the 

 more posterior parts where the tissues and organs are simpler 

 and possess greater regulatory properties. 



And it should not be forgotten that as a matter of fact every 

 part of the embryo is subject to abnormality following treat- 

 ment with chemicals or low temperature; no part has been found 

 to be wholly free from abnormality in every case. This is an 

 observation, by the way, which has a decided bearing upon the 

 use of experimentally treated material as an aid to the solution 

 of problems in the normal development of Fundulus. If any 

 part, organ or tissue, whatever, may be affected abnormally by 

 such treatment, an extreme degree of caution should be exer- 

 cised in applying to the interpretation of the events of normal 

 development, the evidence drawn from the histogenesis of 

 embryos developing from treated eggs. 



In conclusion I should Hke to refer briefly to the bearing of 

 the hypothesis stated here, upon a group of observations of a 

 wholly different kind. I refer to certain results of Teleost 

 hybridization described by Moenkhaus ('10), Loeb ('12), and 

 Newman ('14). Moenkhaus found that in such hybrids ''de- 

 velopment in its early stages proceeds normally, i.e., when 

 superficially viewed the deleterious effects of the two strange 



