i897- MORPHOLOGY IN ZOOLOGICAL SCIENCE. 27 



g;ood instance is found in the limbs of perennibranchiate Urodela. 

 These are, it is true, constructed on the pentadactyle plan, but they 

 are exceedingly feeble, and the number of digits is in almost every 

 case reduced. They are thus unable to support the weight of the 

 body, the purpose for which the pentadactyle limb was originally 

 evolved, nor do they, on the other hand, show any of the characters of 

 fins. Hence, we conclude that perennibranchiates have been derived 

 from caducibranchiate forms — forms, at any rate, which walked more 

 and swam less, — and the life-histories of Siredon and Menobranchus have 

 now proved that we are right. Then, again, in the case of the head of 

 Amphioxus we notice that it is asymmetrical, and that the brain and 

 sense-organs are almost entirely obsolete. But we know that the 

 whole organisation of Vertebrata is permeated by bilateral sym- 

 metry, and that they all have well-developed brains and sense-organs, 

 and lead typically a free, roving life. To this life only can we 

 attribute the fact that Amphioxus, like other vertebrates, is flattened in 

 a vertical plane, since only in swimming could a body of such a shape 

 be easily maintained in equilibrium ; asymmetry and feeble sense- 

 organs are inconsistent with such a life, and hence we interpret them 

 as secondary features, due to the secondarily-acquired burrowing 

 habits. 



One of the most vexed questions in zoology has been the value 

 of the evidence afforded by embryology. The older anatomists 

 roundly asserted that the ontogeny of the individual was a recapitu- 

 lation of the phylogeny of the race. They endeavoured to find 

 ancestral meanings for all the embryonic structures which they had 

 observed. Lately it has become fashionable to look coldly on such 

 theorising, and some have even gone so far as to deny that there is 

 any evidence that ontogeny is in any sense a repetition at all. I 

 cannot but think that the latter class of zoologists are in much the 

 same position as the theological opponents of the doctrine of evolution 

 — they are most imperfectly acquainted with the facts. It must be 

 remembered that comparative embryology is only an extension of 

 comparative anatomy ; that it is most arbitrary to say that only sexually 

 adult forms are to be compared with one another; and that the conclu- 

 sion that resemblance between the immature stage of one anmial and 

 the mature stage of another is indicative of affinity, is precisely on all 

 fours with a similar conclusion drawn from a comparison with one 

 another of two adult stages. But there are many cases where no 

 one really doubts that the affinity indicated by ontogeny is correct ; 

 in other words, that in these particular cases ontogeny is a repetition 

 of past history. Such cases are the pentacrinoid young of Antedon, 

 with the conspicuous basals and orals, the cyclops-like larva of the 

 parasitic copepods, the tadpole-like larva of the ascidians, and, in 

 general features, the tadpole of the frog. Now, if there be this 

 undeniable hereditary basis for ontogeny in some cases, it is 

 exceedingly unlikely that it is a factor which is only sporadically 



