264 HABIT AND INTELLIGENCE. [chap. 



Characters alimentary canal being doubled back so as to bring its two 

 bryordo extremities near each other. But this character is not 

 are subject constant among the MoUusca : it is not found, for instance, 

 caption, among the chitons ; and it is also not an embryonic cha- 

 racter — on the contrary, the molluscan embryo is sym- 

 Unsym- metrical, and the unsymmetrical form is subsequently 

 molluscan produced by the right or left side, according to the species, 

 develop- growing more than the other.^ 



meut. ° ° 



The parts that appear the earliest are the most constant 

 Import- throughout wdde groups. It is a result of this fact that the 

 embryonic ™ost important characters for classification are frequently 

 characters those of the embryo or of the larva. It is usually said, de- 

 fication. velopment is the criterion of morphology ; but it would, I 

 think, be more intelligible to say that development is the 

 criterion of classification : in other words, the history of the 

 development of any organism is the test of its true affinities. 

 The earliest developed characters are the fundamental 

 ones, and true classification is classification by funda- 

 niental characters. The most remarkable instance of this 

 Cirrhi- is that of the Cirrhipedes, or barnacles ; the position of 

 which among animals was totally misunderstood so long 

 as they were known in the mature state only. They were 

 classed by Cuvier as Mollusca, which they resemble only in 

 the possession of shells, and in other superficial characters ; 

 their but now that their larval forms are found to be unmistake- 



llrvx!^'^^^ ably crustacean, they are classed among, or near, the Crus- 

 tacea. Thus, of two crustaceans, which are at first almost 

 alike,^ one may end its life as a crustacean, while the other 

 undergoes metamorphosis into the very different form of a 

 cirrhipede. But this, though I believe it is by far the 

 most remarkable instance of the kind, is not by any means 

 the only one. A similar case exists among the marine 

 Dorsibran- worms. The clorsihrcmchiata and the tuhicolce are very 

 and tuhi- ^i^iilike in external appearance : the former are free animals, 

 coice. -v^ith branchiae in rows along the back ; the latter are fixed 



animals, inhabiting tubular shells, with branchiae round 



1 Huxley on the Morphology of the Cephalous Mollusca, Philosophical 

 Transactions, 1853. 



2 Dr. Knox's translation of Milne-Edwards's Manual of Zoology, p. 448. 



