J. E. DUERDEN 183 



back towards the egg of the ancestral phylogenetic stages, as seems 

 sometimes to be understood, but an inhibition or failure to continue 

 somatic expression beyond a certain stage. What that later expression 

 was can only be surmised by comparison with what are now regarded as 

 related types ; but we have no certain, knowledge that our interpretation 

 is correct ; it may have been something beyond our experience of to-day. 

 Ontogeny does not repeat phylogeny; any early structural similarity 

 among phyla is only evidence of community of germ plasm as far as the 

 two continue to express themselves alike, but not beyond. Ontogeny 

 can only repeat phylogeny as far back as no factorial losses have occurred. 

 In the case of a retrogressive character it does this to a less and less 

 degree with succeeding generations. Once factorial losses have taken 

 place throughout the race we have no certain means of knowing what 

 this expression was; we can only surmise from closely related types. 

 The germ plasm of all chordates carries the genetic factors for visceral 

 arches and gill-clefts ; they persist in their expression throughout life in 

 the case of fishes and some amphibians, but lose it at an early develop- 

 mental stage in sauropsida and mammals. We presume that ancestrally 

 they continued longer, but have no certain proof thereof The relation- 

 ship involved is almost certainly that of community of factors so far as 

 they have a common expression, but no further. Degeneration may be 

 defined as the somatic expression of a phylogenetic degradation and loss 

 of genetic factors. 



The degenerative results in the ostrich further reveal that the 

 factors for different classes of somatic structures tend to have different 

 rates of persistence in the germ plasm. Superficial skin structures, like 

 feathers, scales and claws, are rather late in gaining ontogenetic ex- 

 pression compared with more fundamental structures, like limbs, the 

 skeleton, nervous system, and sense-organs. Likewise in degeneration 

 the factors for the former readily disappear from the germ plasm, while 

 those for the more fundamental parts, as exemplified by the digits, 

 persist much longer, and disappear with extreme slowness. As a general 

 relationship we may say that superficial or skin structures are late in 

 phylogeny as also in ontogeny, while fundamental structures are early in 

 phylogeny and in ontogeny; in degeneration the genetic factors for 

 structures late in phylogeny and ontogeny early disappear, those for 

 structures early in phylogeny and ontogeny are late in disappearing. 



12—2 



